Cross Reference: /src/sys/arch/i386/include

History log of /src/sys/arch/i386/include
Revision	Date	Author	Comments
1.53	30-Nov-2024	christos	Create a new header lwp_private.h to contain _lwp_getprivate_fast, _lwp_gettcb_fast, _lwp_settcb and remove them from mcontext.h, so that: 1. we don't need special hacks to hide them 2. we can include <lwp.h> where needed to get the necessary prototypes without redefining them locally.
1.52	04-Nov-2024	christos	Undo previous lwp.h change.
1.51	03-Nov-2024	christos	Split __lwp_getprivate_fast and __lwp_*tcb from mcontext.h into a separate lwp.h file.
1.50	13-Sep-2019	manu	branches: 1.50.32; Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h
1.49	12-Jul-2018	maxv	branches: 1.49.4; Remove the kernel PMC code. Sent yesterday on tech-kern@. This change: * Removes "options PERFCTRS", the associated includes, and the associated ifdefs. In doing so, it removes several XXXSMPs in the MI code, which is good. * Removes the PMC code of ARM XSCALE. * Removes all the pmc.h files. They were all empty, except for ARM XSCALE. * Reorders the x86 PMC code not to rely on the legacy pmc.h file. The definitions are put in sysarch.h. * Removes the kern/sys_pmc.c file, and along with it, the sys_pmc_control and sys_pmc_get_info syscalls. They are marked as OBSOL in kern, netbsd32 and rump. * Removes the pmc_evid_t and pmc_ctr_t types. * Removes all the associated man pages. The sets are marked as obsolete.
1.48	12-Aug-2017	maxv	branches: 1.48.2; 1.48.4; Remove vm86. Pass 4.
1.47	09-Aug-2017	maxv	Remove ibcs2_machdep.h on i386, and don't install it.
1.46	08-Aug-2017	maxv	Move freebsd_machdep.h into sys/compat/freebsd, and don't install it. Now, the compat_freebsd files are all contained in sys/compat/freebsd.
1.45	29-Jul-2017	maxv	Unlink svr4_machdep.h.
1.44	27-Feb-2016	tls	branches: 1.44.10; Add cpu_rng, a framework for simple on-CPU random number generators.
1.43	23-Jul-2014	alnsn	branches: 1.43.4; Rename sljitarch.h to sljit_machdep.h.
1.42	12-Feb-2014	dsl	branches: 1.42.2; Change i386 to use x86/fpu.c instead of i386/isa/npx.c This changes the trap10 and trap13 code to call directly into fpu.c, removing all the code for T_ARITHTRAP, T_XMM and T_FPUNDA from i386/trap.c Not all of the code thate appeared to handle fpu traps was ever called! Most of the changes just replace the include of machine/npx.h with x86/fpu.h (or remove it entirely).
1.41	05-Nov-2012	alnsn	branches: 1.41.2; Build sljit test when MKSLJIT != no and set MKSLJIT to yes on amd64 and i386.
1.40	08-Aug-2012	drochner	branches: 1.40.2; on x86, <machine/cpufunc.h> only pulls in <x86/cpufunc.h>. The latter is not installed to userland and noone missed it, so the former ones can not be useful either. Don't install them.
1.39	17-Jul-2011	joerg	branches: 1.39.2; Retire varargs.h support. Move machine/stdarg.h logic into MI sys/stdarg.h and expect compiler to provide proper builtins, defaulting to the GCC interface. lint still has a special fallback. Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and derive va_list as required by standards.
1.38	17-Jul-2011	dyoung	On i386, good-bye <machine/bus.h>.
1.37	31-Jul-2010	joerg	Add support for fenv.h interface for i386 and amd64. Submitted by Stathis Kamperis as part of GSoC 2010 and ported from FreeBSD.
1.36	20-Dec-2007	ad	branches: 1.36.10; 1.36.24; 1.36.30; 1.36.32; - Make __cpu_simple_lock and similar real functions and patch at runtime. - Remove old x86 atomic ops. - Drop text alignment back to 16 on i386 (really, this time). - Minor cleanup.
1.35	14-Nov-2007	ad	branches: 1.35.2; 1.35.6; - Remove I486_CPU, I586_CPU, I686_CPU options. They buy us nothing and clutter the code significantly. - Remove pccons.
1.34	09-Feb-2007	ad	branches: 1.34.6; 1.34.12; 1.34.22; 1.34.24; 1.34.28; 1.34.30; Merge newlock2 to head.
1.33	26-Jul-2006	drochner	branches: 1.33.4; don't install <machine/db_machdep.h>, this is kernel only
1.32	04-Feb-2006	jmmv	branches: 1.32.2; Revert yesterday's change that attempted to fix the detection of the boot device when using a Multiboot boot loader. It couldn't work because these boot loaders do not pass a checksum of the disk so matchbiosdisk() cannot really find any matches. I should have gone to sleep before commiting... Found by xtraeme@.
1.31	03-Feb-2006	jmmv	branches: 1.31.2; When booting an i386 kernel with Multiboot, properly detect the boot device by looking it up in the x86_alldisks table (instead of trying to match it to 'wd*' manually). In order to do this, move the cpu_rootconf function from x86 common code to amd64 and i386 specific one. This way, i386 can do an extra step (call the appropriate Multiboot code) in the appropriate place (after x86_matchbiosdisks and before findroot()).
1.30	03-Feb-2006	jmmv	Implement support for 'The Multiboot Specification' so that i386 kernels can be booted directly from Multiboot-compliant boot loaders (e.g. GRUB). See the added multiboot(8) manual page for more information. No objections in tech-kern@; only positive comments.
1.29	11-Dec-2005	christos	branches: 1.29.2; 1.29.4; merge ktrace-lwp.
1.28	03-May-2005	tron	branches: 1.28.2; "loadfile_machdep.h" should not be installed as pointed out by Izumi Tsutsui. This needs to be fixed in the build process.
1.27	02-May-2005	tron	Install "loadfile_machdep.h" to fix build of "libsa".
1.26	08-May-2004	kleink	Factor out W{CHAR,INT}_{MAX,MIN} into their own header file.
1.25	15-May-2003	wiz	branches: 1.25.2; Don't install machine/mouse.h on i386 and prep; there are no users for the old busmouse interface any longer, we use wsmouse now.
1.24	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.23	26-Nov-2002	lukem	Remove KDIR=, since SYS_INCLUDE=symlinks and KDIR are not supported any more.
1.22	23-Nov-2002	lha	install intrdefs.h so it can be used by intr.h (and libkvm)
1.21	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.20	06-Sep-2002	gehenna	Merge the gehenna-devsw branch into the trunk. This merge changes the device switch tables from static array to dynamically generated by config(8). - All device switches is defined as a constant structure in device drivers. - The new grammer ``device-major'' is introduced to ``files''. device-major <prefix> char <num> [block <num>] [<rules>] - All device major numbers must be listed up in port dependent majors.<arch> by using this grammer. - Added the new naming convention. The name of the device switch must be <prefix>_[bc]devsw for auto-generation of device switch tables. - The backward compatibility of loading block/character device switch by LKM framework is broken. This is necessary to convert from block/character device major to device name in runtime and vice versa. - The restriction to assign device major by LKM is completely removed. We don't need to reserve LKM entries for dynamic loading of device switch. - In compile time, device major numbers list is packed into the kernel and the LKM framework will refer it to assign device major number dynamically.
1.19	12-Sep-2001	jmc	branches: 1.19.2; 1.19.8; 1.19.12; Add mtrr.h to INCS list to get installed on a make includes
1.18	15-Apr-2001	kleink	branches: 1.18.4; 1.18.6; Add definitions of C99 integer format conversion macros. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.17	15-Apr-2001	kleink	Add definitions of C99 specified-width integer type limits. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.16	14-Apr-2001	kleink	Add definitions of C99 integer constant macros. Tidy Makefiles up a little.
1.15	14-Apr-2001	kleink	Add definitions of C99 minimum-width and greatest-width integer types. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.14	05-Nov-2000	thorpej	branches: 1.14.4; Machine pmc.h
1.13	26-Jun-2000	kleink	Add <machine/int_types.h>, which provides namespace-pure definitions of exact-width integer types.
1.12	29-Apr-2000	thorpej	Require that each each MACHINE/MACHINE_ARCH supply a lock.h. This file contains the values __SIMPLELOCK_LOCKED and __SIMPLELOCK_UNLOCKED, which replace the old SIMPLELOCK_LOCKED and SIMPLELOCK_UNLOCKED. These files are also required to supply inline functions __cpu_simple_lock(), __cpu_simple_lock_try(), and __cpu_simple_unlock() if locking is to be supported on that platform (i.e. if MULTIPROCESSOR is defined in the _KERNEL case). Change these functions to take an int * (&alp->lock_data) rather than the struct simplelock * itself. These changes make it possible for userland to use the locking primitives by including <machine/lock.h>.
1.11	17-Mar-2000	tron	Install "endian_machdep.h" required for "sys/endian.h".
1.10	23-Dec-1999	kleink	C99: Define a NAN macro in <math.h> which evaulates to a constant expression of a single-precision quiet NaN; only to be defined on platforms that do support this value.
1.9	25-Oct-1999	drochner	Don't let loadfile_machdep.h make it to /usr/include - the i386 stand Makefiles already set up the compile environment so that machine/loadfile_machdep.h can be found.
1.8	17-Jun-1999	thorpej	branches: 1.8.2; 1.8.4; 1.8.6; PMAP_NEW is no longer optional on the i386; the old pmap's page table allocation strategy no longer works at all. Move pmap.new.* to pmap.. To read the revision history of PMAP_NEW up until this merge, use cvs rlog of the old pmap.new. files.
1.7	28-Apr-1999	christos	MD defines for the MI loadfile
1.6	10-Mar-1999	mycroft	branches: 1.6.4; Add an ieee.h.
1.5	16-Jan-1999	tron	Add missing backslashes between continued lines.
1.4	16-Jan-1999	bouyer	Oups, need byte_swap.h too. Pointed out by Robert V. Baron
1.3	15-Jan-1999	bouyer	Move the bswap functions from libutil to libc (this bups the minor of libc and the major of libutil). For little-endian architectures merge the bnswap() assembly versions with nto* and hton* using symbols aliasing. Use symbol renaming for the bswap function in this case to avoid namespace pollution. Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian machines, common code for inline macros go in machine/byte_swap.h Sync libkern with libc. Adjust #include in kernel sources for machine/bswap.h.
1.2	01-Oct-1998	erh	linux_machdep.h has moved.
1.1	12-Jun-1998	cgd	Rework the way kernel include files are installed. In the new method, as with user-land programs, include files are installed by each directory in the tree that has includes to install. (This allows more flexibility as to what gets installed, makes 'partial installs' easier, and gives us more options as to which machines' includes get installed at any given time.) The old SYS_INCLUDES={symlinks,copies} behaviours are _both_ still supported, though at least one bug in the 'symlinks' case is fixed by this change. Include files can't be build before installation, so directories that have includes as targets (e.g. dev/pci) have to move those targets into a different Makefile.
1.6.4.1	21-Jun-1999	thorpej	Sync w/ -current.
1.8.6.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.8.4.1	15-Nov-1999	fvdl	Sync with -current
1.8.2.3	21-Apr-2001	bouyer	Sync with HEAD
1.8.2.2	22-Nov-2000	bouyer	Sync with HEAD.
1.8.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.14.4.6	11-Dec-2002	thorpej	Sync with HEAD.
1.14.4.5	18-Oct-2002	nathanw	Catch up to -current.
1.14.4.4	17-Sep-2002	nathanw	Catch up to -current.
1.14.4.3	21-Sep-2001	nathanw	Catch up to -current.
1.14.4.2	21-Jun-2001	nathanw	Catch up to -current.
1.14.4.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.18.6.1	01-Oct-2001	fvdl	Catch up with -current.
1.18.4.2	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.18.4.1	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.19.12.1	19-May-2002	gehenna	Remove port-dependent conf.h.
1.19.8.1	17-Mar-2002	thorpej	Add first-cut mutex and rwlock implementation bits for the x86. XXX Does not actually handle i386 CPUs yet, only i486 and later.
1.19.2.2	11-Dec-2001	sommerfeld	Install atomic.h for benefit of userland
1.19.2.1	12-Sep-2001	sommerfeld	file Makefile was added on branch sommerfeld_i386mp_1 on 2001-12-11 23:51:25 +0000
1.25.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.25.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.25.2.1	03-Aug-2004	skrll	Sync with HEAD
1.28.2.5	21-Jan-2008	yamt	sync with head
1.28.2.4	15-Nov-2007	yamt	sync with head.
1.28.2.3	26-Feb-2007	yamt	sync with head.
1.28.2.2	30-Dec-2006	yamt	sync with head.
1.28.2.1	21-Jun-2006	yamt	sync with head.
1.29.4.1	09-Sep-2006	rpaulo	sync with head
1.29.2.1	18-Feb-2006	yamt	sync with head.
1.31.2.1	22-Apr-2006	simonb	Sync with head.
1.32.2.1	11-Aug-2006	yamt	sync with head
1.33.4.1	24-Oct-2006	ad	Compile fixes
1.34.30.2	27-Dec-2007	mjf	Sync with HEAD.
1.34.30.1	19-Nov-2007	mjf	Sync with HEAD.
1.34.28.1	18-Nov-2007	bouyer	Sync with HEAD
1.34.24.1	09-Jan-2008	matt	sync with HEAD
1.34.22.1	21-Nov-2007	joerg	Sync with HEAD.
1.34.12.1	18-Apr-2007	thorpej	Convert i386 and amd64 to the new atomic ops API.
1.34.6.1	03-Dec-2007	ad	Sync with HEAD.
1.35.6.1	02-Jan-2008	bouyer	Sync with HEAD
1.35.2.1	26-Dec-2007	ad	Sync with head.
1.36.32.1	05-Mar-2011	rmind	sync with head
1.36.30.1	17-Aug-2010	uebayasi	Sync with HEAD.
1.36.24.2	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.36.24.1	24-Oct-2010	jym	Sync with HEAD
1.36.10.1	11-Aug-2010	yamt	sync with head.
1.39.2.3	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.39.2.2	16-Jan-2013	yamt	sync with (a bit old) head
1.39.2.1	30-Oct-2012	yamt	sync with head
1.40.2.3	03-Dec-2017	jdolecek	update from HEAD
1.40.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.40.2.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.41.2.1	18-May-2014	rmind	sync with head
1.42.2.1	10-Aug-2014	tls	Rebase.
1.43.4.2	28-Aug-2017	skrll	Sync with HEAD
1.43.4.1	19-Mar-2016	skrll	Sync with HEAD
1.44.10.1	18-Sep-2019	martin	Pull up following revision(s) [adapted, via patch] (requested by manu in ticket #1382): sys/arch/i386/include/multiboot2.h: revision 1.1 sys/arch/i386/stand/lib/exec.c: revision 1.74 sys/arch/i386/stand/efiboot/efiboot.c: revision 1.11 sys/arch/i386/stand/efiboot/bootia32/multiboot32.S: revision 1.1 sys/arch/i386/stand/lib/biosdisk.c: revision 1.52 sys/arch/i386/stand/efiboot/bootx64/efibootx64.c: revision 1.5 distrib/sets/lists/comp/md.amd64: revision 1.273 sys/arch/i386/stand/efiboot/eficons.c: revision 1.7 sys/arch/i386/stand/efiboot/bootia32/efibootia32.c: revision 1.5 distrib/sets/lists/comp/md.i386: revision 1.191 sys/arch/i386/stand/lib/libi386.h: revision 1.45 sys/arch/i386/stand/lib/bootinfo_memmap.c: revision 1.6 sys/arch/i386/stand/lib/pread.c: revision 1.8 sys/arch/i386/stand/efiboot/efiboot.h: revision 1.10 sys/arch/i386/stand/efiboot/bootia32/Makefile: revision 1.3 sys/arch/i386/stand/lib/Makefile: revision 1.47 sys/arch/i386/stand/efiboot/bootx64/Makefile: revision 1.2 sys/arch/i386/stand/lib/multiboot.S: revision 1.3 sys/arch/i386/stand/efiboot/bootx64/multiboot64.S: revision 1.1 sys/arch/i386/stand/efiboot/efimemory.c: revision 1.9 sys/arch/i386/stand/efiboot/boot.c: revision 1.16 sys/arch/i386/include/Makefile: revision 1.50 sys/arch/i386/stand/efiboot/Makefile.efiboot: revision 1.16 share/man/man8/man8.x86/boot.8: revision 1.20 sys/arch/i386/stand/boot/Makefile.boot: revision 1.73 sys/arch/i386/stand/pxeboot/Makefile: revision 1.26 sys/arch/i386/stand/lib/exec_multiboot1.c: revision 1.1 sys/arch/i386/stand/netboot/Makefile.netboot: revision 1.11 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.1 sys/arch/i386/stand/lib/biosdisk.h: revision 1.12 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.2 sys/arch/i386/stand/dosboot/Makefile: revision 1.32 sys/external/bsd/gnu-efi/dist/inc/efiapi.h (apply patch) Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h Remove debug define. It remained there unseen because it was misspelled! Remove obsoeolete BUGS note that UEFI bootloader does not support multiboot Add new multiboot2.h header.
1.48.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.48.4.1	10-Jun-2019	christos	Sync with HEAD
1.48.2.1	28-Jul-2018	pgoyette	Sync with HEAD
1.49.4.1	17-Sep-2019	martin	Pull up following revision(s) (requested by manu in ticket #203): sys/arch/i386/include/multiboot2.h: revision 1.1 sys/arch/i386/stand/lib/exec.c: revision 1.74 sys/arch/i386/stand/efiboot/efiboot.c: revision 1.11 sys/arch/i386/stand/efiboot/bootia32/multiboot32.S: revision 1.1 sys/arch/i386/stand/lib/biosdisk.c: revision 1.52 sys/arch/i386/stand/efiboot/bootx64/efibootx64.c: revision 1.5 sys/arch/i386/stand/efiboot/eficons.c: revision 1.7 sys/arch/i386/stand/efiboot/bootia32/efibootia32.c: revision 1.5 sys/arch/i386/stand/lib/libi386.h: revision 1.45 sys/arch/i386/stand/lib/bootinfo_memmap.c: revision 1.6 sys/arch/i386/stand/lib/pread.c: revision 1.8 sys/arch/i386/stand/efiboot/efiboot.h: revision 1.10 sys/arch/i386/stand/efiboot/bootia32/Makefile: revision 1.3 sys/arch/i386/stand/lib/Makefile: revision 1.47 sys/arch/i386/stand/efiboot/bootx64/Makefile: revision 1.2 sys/arch/i386/stand/lib/multiboot.S: revision 1.3 sys/arch/i386/stand/efiboot/bootx64/multiboot64.S: revision 1.1 sys/arch/i386/stand/efiboot/efimemory.c: revision 1.9 sys/arch/i386/stand/efiboot/boot.c: revision 1.16 sys/arch/i386/include/Makefile: revision 1.50 sys/arch/i386/stand/efiboot/Makefile.efiboot: revision 1.16 share/man/man8/man8.x86/boot.8: revision 1.20 sys/arch/i386/stand/boot/Makefile.boot: revision 1.73 sys/arch/i386/stand/pxeboot/Makefile: revision 1.26 sys/arch/i386/stand/lib/exec_multiboot1.c: revision 1.1 sys/arch/i386/stand/netboot/Makefile.netboot: revision 1.11 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.1 sys/arch/i386/stand/lib/biosdisk.h: revision 1.12 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.2 sys/arch/i386/stand/dosboot/Makefile: revision 1.32 Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h Remove debug define. It remained there unseen because it was misspelled! Remove obsoeolete BUGS note that UEFI bootloader does not support multiboot
1.50.32.1	02-Aug-2025	perseant	Sync with HEAD
1.3	19-Oct-2014	joerg	Disable SSE and AVX for kernel modules too.
1.2	24-May-2014	jakllsch	branches: 1.2.2; Now that GCC 4.8 on i486 has stolen the frame pointer, do like amd64 and add -fno-omit-frame-pointer to kernel DEFCOPTS and modules CFLAGS.
1.1	25-Nov-2009	tron	branches: 1.1.4; 1.1.8; 1.1.22; 1.1.26; 1.1.40; Enable SSP (Stack Smash Protection) in x86 kernels by default (except in i386 *TINY kernels). The NetBSD/i386 "ALL" kernel is unconditionally compiled with SSP enabled. Change approved by the core team.
1.1.40.1	10-Aug-2014	tls	Rebase.
1.1.26.2	03-Dec-2017	jdolecek	update from HEAD
1.1.26.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.1.22.1	10-Nov-2014	msaitoh	Pull up following revision(s) (requested by joerg in ticket #1172): sys/arch/amd64/include/Makefile.inc: revision 1.5 sys/arch/i386/include/Makefile.inc: revision 1.3 via patch Disable SSE and AVX for kernel modules too.
1.1.8.2	24-Oct-2010	jym	Sync with HEAD
1.1.8.1	25-Nov-2009	jym	file Makefile.inc was added on branch jym-xensuspend on 2010-10-24 22:48:02 +0000
1.1.4.2	11-Mar-2010	yamt	sync with head
1.1.4.1	25-Nov-2009	yamt	file Makefile.inc was added on branch yamt-nfs-mp on 2010-03-11 15:02:29 +0000
1.2.2.1	19-Oct-2014	martin	Pull up following revision(s) (requested by joerg in ticket #152): sys/arch/amd64/include/Makefile.inc: revision 1.5 sys/arch/i386/include/Makefile.inc: revision 1.3 Disable SSE and AVX for kernel modules too.
1.11	24-May-2008	jmcneill	MI implementation of AcpiAcquireGlobalLock and AcpiReleaseGlobalLock.
1.10	24-May-2008	jmcneill	Replace asm implementation of ACPI global lock functions with C equivalents.
1.9	11-Apr-2008	jmcneill	branches: 1.9.2; 1.9.4; 1.9.6; Revert previous.
1.8	11-Apr-2008	jmcneill	Remove a nasty hack that slipped in during the ACPICA update that should never have made it in-tree. How ACPI has ever worked since 4.99.42 is beyond me.
1.7	09-Dec-2007	jmcneill	branches: 1.7.10; Merge jmcneill-pm branch.
1.6	24-Dec-2005	perry	branches: 1.6.48; 1.6.50; 1.6.60; 1.6.62; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.5	11-Dec-2005	christos	merge ktrace-lwp.
1.4	13-Aug-2003	yamt	branches: 1.4.16; - leave a compiler which register is used for an asm argument. - use correct asm constraints. - eliminate usage of dummy variables.
1.3	13-Aug-2003	yamt	use correct asm operand indexes. (previously 'GLptr' was used as mask.) this long standing bug seems to be uncovered by my previous change(rev.1.2) on some machines. PR/22457 from Matthias Scheler, and tested by him.
1.2	09-Aug-2003	yamt	in ACPI_ACQUIRE_GLOBAL_LOCK, don't assume that the reserved bits of the global lock are all zero. it's actually needed for, at least, IBM Thinkpad X21 (2662-64J).
1.1	11-May-2003	fvdl	branches: 1.1.2; Add a new include file with some ACPI MD macros, to move i386 specific stuff out of acnetbsd.h
1.1.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.1.2.1	03-Aug-2004	skrll	Sync with HEAD
1.4.16.2	21-Jan-2008	yamt	sync with head
1.4.16.1	21-Jun-2006	yamt	sync with head.
1.6.62.1	11-Dec-2007	yamt	sync with head.
1.6.60.1	26-Dec-2007	ad	Sync with head.
1.6.50.1	09-Jan-2008	matt	sync with HEAD
1.6.48.1	02-Oct-2007	jmcneill	Update to ACPI-CA 20070320
1.7.10.1	02-Jun-2008	mjf	Sync with HEAD.
1.9.6.2	23-Jun-2008	wrstuden	Remove files removed on branch. Updating using patch has its drawbacks. :-)
1.9.6.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.9.4.1	04-May-2009	yamt	sync with head.
1.9.2.1	04-Jun-2008	yamt	sync with head
1.7	11-May-2003	fvdl	Redirect to x86/.
1.6	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.5	18-Jul-2002	kanaoka	branches: 1.5.2; - Disable Interrput before execute AcpiEnterSleepState(). - Add acpi_md_OsDisableInterrupt() for this.
1.4	18-Jun-2002	tshiozak	add MD support for ACPI sleep/wakeup.
1.3	15-Jun-2002	thorpej	Update for acpica-unix-20020612. From Takayoshi Kochi.
1.2	28-Sep-2001	thorpej	branches: 1.2.2; 1.2.4; 1.2.10; 1.2.18; Bah, fix typo that crept in (tank you vi(1)).
1.1	28-Sep-2001	thorpej	Machine-dependent ACPI CA Osd routines for i386.
1.2.18.3	21-Jul-2002	gehenna	catch up with -current.
1.2.18.2	16-Jul-2002	gehenna	catch up with -current.
1.2.18.1	14-Jul-2002	gehenna	catch up with -current.
1.2.10.4	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.2.10.3	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.2.10.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.2.10.1	28-Sep-2001	thorpej	file acpi_machdep.h was added on branch kqueue on 2002-01-10 19:44:49 +0000
1.2.4.5	07-Jan-2003	thorpej	Sync with HEAD.
1.2.4.4	01-Aug-2002	nathanw	Catch up to -current.
1.2.4.3	20-Jun-2002	nathanw	Catch up to -current.
1.2.4.2	08-Oct-2001	nathanw	Catch up to -current.
1.2.4.1	28-Sep-2001	nathanw	file acpi_machdep.h was added on branch nathanw_sa on 2001-10-08 20:10:46 +0000
1.2.2.2	01-Oct-2001	fvdl	Catch up with -current.
1.2.2.1	28-Sep-2001	fvdl	file acpi_machdep.h was added on branch thorpej-devvp on 2001-10-01 12:39:58 +0000
1.5.2.2	18-Jul-2002	kanaoka	- Disable Interrput before execute AcpiEnterSleepState(). - Add acpi_md_OsDisableInterrupt() for this.
1.5.2.1	18-Jul-2002	kanaoka	file acpi_machdep.h was added on branch sommerfeld_i386mp_1 on 2002-07-18 12:05:13 +0000
1.28	07-May-2019	kamil	Switch all users (except ia64) of custom machine/ansi.h to common_ansi.h Deduplicate the code among ports and poll definitions of types directly from a compiler. This fixes miscompilation of certain programs that instruct compilers to generate code for different types. This bug has been detected with -fshort-wchar in EFI firmware. Proposed and discussed on a mailing list (twice). Itanium uses custom !ELF fallback switch, temporarily leave it as it is.
1.27	17-Jun-2015	martin	branches: 1.27.18; Make clock_t unsigned int everywhere. Ok: matt@, mrg@
1.26	09-Dec-2014	pooka	On i386, prefer __WCHAR/WINT_TYPE__ where available instead of hardcoded _BSD_WCHAR_T_ value. This mimics some other ports, and more importantly fixes compilation with i386 compilers which define __WCHAR_TYPE__ as "long int" instead of the "int" we had in there. That superficial mismatch resulted in errors such as: lib/libc/gen/vis.c:109:1: error: array of inappropriate type initialized from string constant static const wchar_t char_shell[] = L"'`\";&<>()\|{}]\\$!^~"; No change to NetBSD (our gcc does define __WCHAR_TYPE__, but it is "int" ... as expected ... since the NetBSD build worked ;)
1.25	17-Jul-2011	joerg	branches: 1.25.12; 1.25.28; 1.25.30; Retire varargs.h support. Move machine/stdarg.h logic into MI sys/stdarg.h and expect compiler to provide proper builtins, defaulting to the GCC interface. lint still has a special fallback. Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and derive va_list as required by standards.
1.24	27-Mar-2010	tnozaki	1. {wctype,wctrans,mbstate}_t: switch MD to MI like other libc implementation (such as BSD and glibc2). 2. don't typedef void wc{type,trans}_t, suggested by soda@-san. it may pass through compiler type check, it's harmful. so i introduce dummy struct __tag_wc{type,trans}_t(iconv_t already does). no ABI change was made.
1.23	11-Jan-2009	christos	branches: 1.23.2; 1.23.4; 1.23.6; merge christos-time_t
1.22	26-Oct-2008	mrg	branches: 1.22.2; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.21	17-Oct-2007	garbled	branches: 1.21.16; 1.21.18; 1.21.22; 1.21.28; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.20	03-Sep-2007	drochner	clean up some definitions around rune_t which are not needed anymore
1.19	04-Oct-2006	tnozaki	branches: 1.19.8; 1.19.16; 1.19.22; 1.19.26; 1.19.28; fix gcc -Werror -Wmissing-braces problem mbstate_t(this is opaque object)'s initializer should be ``{ 0 }'', so changed 1st field of union from character array to integer.
1.18	11-Dec-2005	christos	branches: 1.18.20; 1.18.22; merge ktrace-lwp.
1.17	07-Aug-2003	agc	branches: 1.17.16; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.16	02-Mar-2003	tshiozak	branches: 1.16.2; add some ISO C 1995 I18N functions and types: btowc, wctrans, towctrans, wcscoll, wcsxfrm, wctype_t and wctrans_t.
1.15	03-Jan-2001	takemura	branches: 1.15.2; replace 'long long' with int64_t to compile stand alone program with compiler other than GCC.
1.14	26-Dec-2000	itojun	make mbstate_t bigger (32 -> 128 bytes). XXX if you have libc after citrus locale import, please recompile libc, and your applications that use mbstate_t (rather rare). really sorry for the mess.
1.13	21-Dec-2000	itojun	populate _BSD_MBSTATE_T_. add warning regarding to rune_t.
1.12	27-Jun-2000	kleink	G/c _BSD_INTPTR_T_ and _BSD_UINTPTR_T_.
1.11	05-May-2000	thorpej	branches: 1.11.2; Use __GNUC_PREREQ__().
1.10	03-May-2000	thorpej	Make varargs and stdarg work with GCC 2.96 and later.
1.9	27-Apr-1998	kleink	branches: 1.9.14; Provide definitions for intptr_t and uintptr_t, signed resp. unsigned integral types large enough to hold any pointer.
1.8	23-Nov-1997	kleink	Add _BSD_SUSECONDS_T_ and _BSD_USECONDS_T_; do some space vs. tab formatting cleanup
1.7	15-Nov-1996	jtc	Define _BSD_CLOCKID_T_ and _BSD_TIMER_T_
1.6	16-Mar-1996	jtc	Add _BSD_WINT_T_ definition so we can handle wint_t type added in NA1.
1.5	27-Oct-1994	cgd	new RCS ID format.
1.4	21-May-1994	cgd	new way of naming things
1.3	20-Dec-1993	mycroft	branches: 1.3.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.1	14-Nov-1993	mycroft	Add _SSIZE_T_.
1.3.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.3.2.1	20-Dec-1993	mycroft	file ansi.h was added on branch magnum on 1993-12-20 05:25:03 +0000
1.9.14.2	05-Jan-2001	bouyer	Sync with HEAD
1.9.14.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.11.2.1	28-May-2000	minoura	Citrus Project XPG4DL, an implementation of I18N (locale) framework, is imported.
1.15.2.2	03-Jan-2001	takemura	replace 'long long' with int64_t to compile stand alone program with compiler other than GCC.
1.15.2.1	03-Jan-2001	takemura	file ansi.h was added on branch sommerfeld_i386mp_1 on 2001-01-03 10:09:00 +0000
1.16.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.16.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.16.2.1	03-Aug-2004	skrll	Sync with HEAD
1.17.16.2	27-Oct-2007	yamt	sync with head.
1.17.16.1	30-Dec-2006	yamt	sync with head.
1.18.22.1	22-Oct-2006	yamt	sync with head
1.18.20.1	18-Nov-2006	ad	Sync with head.
1.19.28.1	06-Nov-2007	matt	sync with HEAD
1.19.26.1	02-Oct-2007	joerg	Sync with HEAD.
1.19.22.1	10-Sep-2007	skrll	Sync with HEAD.
1.19.16.1	03-Oct-2007	garbled	Sync with HEAD
1.19.8.1	09-Oct-2007	ad	Sync with head.
1.21.28.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.21.22.2	11-Aug-2010	yamt	sync with head.
1.21.22.1	04-May-2009	yamt	sync with head.
1.21.18.4	01-Nov-2008	christos	catch up with changes in head.
1.21.18.3	01-Nov-2008	christos	Sync with head.
1.21.18.2	30-Mar-2008	christos	time_t is now __int64_t
1.21.18.1	29-Mar-2008	christos	Welcome to the time_t=long long dev_t=uint64_t branch.
1.21.16.1	17-Jan-2009	mjf	Sync with HEAD.
1.22.2.1	19-Jan-2009	skrll	Sync with HEAD.
1.23.6.1	30-May-2010	rmind	sync with head
1.23.4.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.23.2.2	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.23.2.1	24-Oct-2010	jym	Sync with HEAD
1.25.30.2	22-Sep-2015	skrll	Sync with HEAD
1.25.30.1	06-Apr-2015	skrll	Sync with HEAD
1.25.28.1	16-Jul-2015	riz	Pull up following revision(s) (requested by martin in ticket #846): sys/arch/mips/include/ansi.h: revision 1.29 sys/arch/sh3/include/ansi.h: revision 1.16 sys/arch/sparc64/include/ansi.h: revision 1.18 sys/arch/m68k/include/ansi.h: revision 1.24 sys/arch/powerpc/include/ansi.h: revision 1.30 sys/arch/hppa/include/ansi.h: revision 1.14 sys/arch/i386/include/ansi.h: revision 1.27 sys/arch/alpha/include/ansi.h: revision 1.25 sys/arch/usermode/include/ansi.h: revision 1.5 sys/arch/sparc/include/ansi.h: revision 1.24 Make _BSD_CLOCK_T_ unsigned int so it's the same for IPL32 and LP64 environments. We don't really have a powerpc64 native userland and the mips64 native userland is IPL32 so this shouldn't affect anything. Make clock_t unsigned Make clock_t unsigned int everywhere. Ok: matt@, mrg@
1.25.12.1	03-Dec-2017	jdolecek	update from HEAD
1.27.18.1	10-Jun-2019	christos	Sync with HEAD
1.8	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.7	10-Dec-2002	thorpej	Rename __LDPGSZ to AOUT_LDPGSZ, to accurately reflect what it is.
1.6	27-Oct-1994	cgd	branches: 1.6.42; new RCS ID format.
1.5	28-Jan-1994	jtc	Fix spelling error in Copyright notice
1.4	18-Oct-1993	pk	Put definition of __LDPGSZ and `struct relocation_info' where they belong: in <machine/exec.h>.
1.3	18-Jun-1993	cgd	branches: 1.3.4; convert magic numbers to network byte order, and attendent changes
1.2	03-Jun-1993	cgd	add prototype for a function to be called from execve to check a_mid type
1.1	02-Jun-1993	cgd	machine-dependent exec stuff
1.3.4.2	18-Oct-1993	mycroft	Merge changes from trunk.
1.3.4.1	15-Oct-1993	mycroft	Define __LDPGSZ.
1.6.42.1	11-Dec-2002	thorpej	Sync with HEAD.
1.4	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.3	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file apicvar.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.3	27-Feb-2000	sommerfeld	Multiple-include protection
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	stuff shared between local and i/o apic drivers.
1.2.2.3	07-Jan-2003	thorpej	Sync with HEAD.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file apicvar.h was added on branch nathanw_sa on 2002-10-18 02:37:53 +0000
1.26	01-Oct-2012	dsl	Remove some definitions that were part of the APM bos interface.
1.25	28-Apr-2008	martin	branches: 1.25.34; 1.25.44; Remove clause 3 and 4 from TNF licenses
1.24	08-Jul-2006	christos	branches: 1.24.58; 1.24.60; 1.24.62; remove functions we don't use and adjust to the new prototype.
1.23	11-Dec-2005	christos	branches: 1.23.4; 1.23.8; 1.23.16; merge ktrace-lwp.
1.22	01-Jun-2005	drochner	branches: 1.22.2; kill the "busname" attach arg for acpi, apm, pnpbios and vesabios, and pass NULL "print" function pointers to their config_found functions because they shouldn't be called anyway
1.21	31-May-2005	drochner	inspired by the last commit -- apm_attach_args aren't necessary at all, kill them being here fix cast-qual problems in debug code
1.20	31-May-2005	christos	PR/30387: Nicolas Joly: -current i386 GENERIC_LAPTOP kernel compilation failure
1.19	25-Dec-2003	jmc	Wrap some areas with #ifndef _LOCORE so assembly files can pick up some defs from here. Needed for bioscall fixes in PR#14946
1.18	27-Oct-2003	junyoung	Nuke __P().
1.17	02-Apr-2003	thorpej	branches: 1.17.2; Use PAGE_SIZE rather than NBPG.
1.16	14-Oct-2002	takemura	Moved MI APM definitions into dev/apm.
1.15	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.14	10-Sep-2001	perry	Make it possible to query each battery individually by making the APM_IOC_GETPOWER ioctl read/write. Setting the batteryid in the passed structure returns the data for just that battery. The old ioctl remains for binary compatibility but has been renamed. Itojun already did the hard work -- I just added a couple of lines. reviewed by: thorpej
1.13	22-Jul-2001	itojun	branches: 1.13.2; support multiple batteries (currently only # of batteris is visible - need to tweak ioctl API more). PR 10545.
1.12	21-May-2001	perry	branches: 1.12.2; remove needless externs in front of function prototypes
1.11	24-Mar-2001	christos	Some bios's when apm is disabled [or even some functions of it are disabled], they keep giving error 0x30b. Unfortunately this keeps happening every second making the console unusable. Keep track of the last error that occurred and if the last 10 times apm was called the same error was returned, give up and exit the apm thread.
1.10	12-Jan-2001	simonb	branches: 1.10.2; Enable the APM_IOC_DEV_CTL ioctl. Works at least for turning off & on a LCD screen on a Dell Inspiron 5000e (and probably many others).
1.9	17-Aug-1999	drochner	branches: 1.9.2; 1.9.10; add some v1.2 bits, comment out the APM_IOC_DEV_CTL ioctl which I didn't find a use for
1.8	03-Oct-1998	jtk	copyright has been assigned to TNF, update notice
1.7	31-Aug-1998	jtk	add 32-bit register support and beginnings of APM V1.2 support
1.6	15-Oct-1997	jtk	put in place a proper bioscall.h (sorry about that!) finish up incomplete job of moving structure used by bioscall() to <machine/bioscall.h>, with associated changes in include file strategy, genassym stuff, etc.
1.5	22-Nov-1996	jtk	branches: 1.5.10; add pre-probe step so that autoconfiguration does not print 'apm0 at mainbus0 not configured' message if configured but not present in the BIOS
1.4	10-Sep-1996	jtk	clean up missing prototype, extra variable
1.3	08-Sep-1996	jtk	change APM to use new almost-generalized BIOS trampoline: /* * void bioscall(int function, struct apmregs regs): call the BIOS interrupt "function" from real mode with * registers as specified in "regs" * (for the flags, though, only these flags are passed to the BIOS; * the remainder come from the flags register at the time of the call: * (PSL_C\|PSL_PF\|PSL_AF\|PSL_Z\|PSL_N\|PSL_D\|PSL_V) * * Fills in regs with registers as returned by BIOS. / Thanks to Charles Hannum for complaining and inspiring me to hack this together.
1.2	30-Aug-1996	jtk	make all constants use APM_ prefix, and some other minor cleanup so that this can be the sole APM include file
1.1	25-Aug-1996	jtk	add Advanced Power Management driver Includes bootstrap support from FreeBSD (Tatsumi HOSOKAWA and <ukai>)
1.5.10.1	15-Oct-1997	thorpej	Sync w/ trunk.
1.9.10.5	22-Sep-2001	sommerfeld	Merge branch with current, post ubcperf, MTRR, etc., etc., with the recent fifo_putpages fix to the trunk, GENERIC.MPDEBUG config boots multiuser on my BP6.
1.9.10.4	03-Sep-2001	sommerfeld	Merge with -current once more.
1.9.10.3	23-May-2001	sommerfeld	Merge branch with tonight's current. boots multiuser.
1.9.10.2	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.9.10.1	23-Jan-2001	thorpej	Sync with the trunk.
1.9.2.2	27-Mar-2001	bouyer	Sync with HEAD.
1.9.2.1	18-Jan-2001	bouyer	Sync with head (for UBC+NFS fixes, mostly).
1.10.2.5	18-Oct-2002	nathanw	Catch up to -current.
1.10.2.4	21-Sep-2001	nathanw	Catch up to -current.
1.10.2.3	24-Aug-2001	nathanw	Catch up with -current.
1.10.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.10.2.1	09-Apr-2001	nathanw	Catch up with -current.
1.12.2.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.12.2.2	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.12.2.1	03-Aug-2001	lukem	update to -current
1.13.2.1	01-Oct-2001	fvdl	Catch up with -current.
1.17.2.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.17.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.17.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.17.2.1	03-Aug-2004	skrll	Sync with HEAD
1.22.2.1	30-Dec-2006	yamt	sync with head.
1.23.16.1	13-Jul-2006	gdamore	Merge from HEAD.
1.23.8.1	11-Aug-2006	yamt	sync with head
1.23.4.1	09-Sep-2006	rpaulo	sync with head
1.24.62.1	16-May-2008	yamt	sync with head.
1.24.60.1	18-May-2008	yamt	sync with head.
1.24.58.1	02-Jun-2008	mjf	Sync with HEAD.
1.25.44.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.25.34.1	30-Oct-2012	yamt	sync with head
1.47	05-Jan-2025	riastradh	i386/asm.h: Missed a spot in previous concatenated-.asciz change. PR toolchain/58960: Missing support for _NETBSD_REVISIONID on various ports
1.46	05-Jan-2025	riastradh	x86 machine/asm.h: Avoid juxtaposition for concatenation. clang asm doesn't seem to like it. Instead of `.asciz "foo" "bar"', do `.ascii "foo"; .asciz "bar"'. PR toolchain/58960: Missing support for _NETBSD_REVISIONID on various ports
1.45	09-Jun-2024	riastradh	branches: 1.45.2; i386/asm.h: Respect _NETBSD_REVISIONID.
1.44	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.43	17-Apr-2020	joerg	Mark the .ident section as mergable string section to avoid redundant entries.
1.42	15-May-2016	chs	branches: 1.42.18; 1.42.28; define ALTENTRY(), needed by dtrace.
1.41	12-Sep-2013	joerg	branches: 1.41.6; Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small vs big PIC mode. Retire -DPIC and -DBIGPIC.
1.40	16-Jun-2011	joerg	branches: 1.40.2; 1.40.12; 1.40.16; Use minimal alignment for functions in standalone code.
1.39	20-Dec-2010	joerg	branches: 1.39.6; Consistently use .gnu.warning with .pushsectio and .popsection on all architectures instead of obsolete STABS frames for linker warnings.
1.38	03-May-2008	yamt	branches: 1.38.14; 1.38.22; add macros to designate the end of asm functions. ok by Andrew Doran.
1.37	20-Dec-2007	ad	branches: 1.37.6; 1.37.8; 1.37.10; - Make __cpu_simple_lock and similar real functions and patch at runtime. - Remove old x86 atomic ops. - Drop text alignment back to 16 on i386 (really, this time). - Minor cleanup.
1.36	12-Dec-2007	yamt	restore section after RCSID.
1.35	14-Nov-2007	ad	branches: 1.35.2; 1.35.4; 1.35.6; Drop _ALIGN_TEXT back to 16. 32 was a bit silly.
1.34	17-Oct-2007	garbled	branches: 1.34.2; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.33	05-Aug-2007	ad	branches: 1.33.2; 1.33.6; Keep alignment small for _STANDALONE.
1.32	05-Aug-2007	ad	- Align assembly routines on 32-byte boundaries. - Don't bother doing anything special for CPUVAR() when !MULTIPROCESSOR.
1.31	09-Feb-2007	ad	branches: 1.31.6; 1.31.14; 1.31.16; 1.31.18; 1.31.22; Merge newlock2 to head.
1.30	20-Jan-2006	christos	branches: 1.30.18; Add a STRONG_ALIAS macro
1.29	11-Dec-2005	christos	branches: 1.29.2; merge ktrace-lwp.
1.28	13-May-2004	yamt	branches: 1.28.12; fix a compile error in the case DDB && !GPROF. the problem pointed by Jaka Jejcic on current-users@ and by several others privately.
1.27	13-May-2004	yamt	call mcount in Xspllower.
1.26	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.25	02-May-2003	yamt	branches: 1.25.2; set symbol to be a function using .type directive in IDTVEC macro so that ddb backtrace can pick them up after recent ksyms changes. suggested by Matt Thomas on tech-kern. ok'ed by Frank van der Linden.
1.24	03-Mar-2003	fvdl	Make CVAR asm reference macro CVAROFF, taking an offset.
1.23	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.22	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.21	31-May-2002	thorpej	branches: 1.21.2; Move the IDTVEC() macro to <machine/asm.h>, and make it work with an ISO C preprocessor.
1.20	13-Jan-2002	drochner	branches: 1.20.10; for ELF, put RCSID string into .ident section, add __KERNEL_RCSID macro which is undefined in case of NO_KERNEL_RCSIDS
1.19	14-Dec-2000	mycroft	branches: 1.19.4; 1.19.6; Remove ALTENTRY(). It's dangerous and should not be used.
1.18	23-Jun-2000	kleink	Add a WEAK_ALIAS() macro.
1.17	09-Nov-1999	marc	branches: 1.17.6; charles was on drugs. There's no such thing as #elifdef.
1.16	07-Nov-1999	mycroft	branches: 1.16.2; Fix WARN_REFERENCES() here for the ELF case.
1.15	27-Sep-1999	kleink	branches: 1.15.2; 1.15.4; 1.15.6; For ELF, rename the profiling entry from mcount to __mcount to avoid clashing with the user's namespace, also within file scope; for binary compatibility, define a weak alias for the previous name to cover our tracks.
1.14	23-Aug-1999	kleink	Unlike in an i386 a.out assembler, where in an .align n directive n is meant to be the logarithm to base 2 of the alignment, in an ELF environment n is the actual alignment boundary; thus, adjust the directives accordingly. Albeit the wonderful i386 architecture doesn't mind the smaller alignment in an obvious way, it is likely to have resulted in some performance penalty during the a.out->ELF transition.
1.13	23-Dec-1998	christos	C_LABEL has no _ if __ELF__
1.12	02-Dec-1998	thorpej	Argh, fix the ANSI CPP version of the a.out WARN_REFERENCES().
1.11	02-Dec-1998	thorpej	Implement WARN_REFERENCES().
1.10	05-Jul-1997	thorpej	Define the ALTENTRY() macro here.
1.9	04-Dec-1996	fvdl	Change asm.h to make it usable for kernel profiling as well. Make mcount() a prototype in profile.h
1.8	30-Nov-1996	jtc	PROF -> GPROF
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	02-Sep-1994	jtc	Misc changes to make this file more like the asm.h headers used by other architectures: Renamed _C_FUNC() to _C_LABEL() Renamed _ASM_FUNC() to _ASM_LABEL() Merged _BEGIN_ENTRY with _ENTRY Renamed _END_ENTRY to _PROF_PROLOGUE When assembling with profiling enabled, set up stack frame before calling mcount() in _PROF_PROLOGUE. Also omit emitting .long 0's, as they are no longer used by the profiling code.
1.5	12-Mar-1994	jtc	branches: 1.5.2; Added RCSID macro, for RCS ID's (what else?).
1.4	07-Jan-1994	pk	branches: 1.4.2; Add .type directive to ENRTY(). Move PIC_ macros over here from SYS.h
1.3	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.2	08-Sep-1993	jtc	branches: 1.2.2; STDC -> __STDC__
1.1	16-Jun-1993	mycroft	Move asm.h from /usr/include/sys to /usr/include/machine.
1.2.2.1	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.4.2.2	07-Jan-1994	pk	Add .type directive to ENRTY(). Move PIC_ macros over here from SYS.h
1.4.2.1	07-Jan-1994	pk	file asm.h was added on branch magnum on 1994-01-07 00:46:21 +0000
1.5.2.1	06-Oct-1994	mycroft	Update from trunk.
1.15.6.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.15.4.1	15-Nov-1999	fvdl	Sync with -current
1.15.2.2	05-Jan-2001	bouyer	Sync with HEAD
1.15.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.16.2.2	07-Nov-1999	mycroft	Fix WARN_REFERENCES() here for the ELF case.
1.16.2.1	07-Nov-1999	mycroft	file asm.h was added on branch comdex-fall-1999 on 1999-11-07 07:20:27 +0000
1.17.6.1	25-Jul-2000	kleink	Pull up rev. 1.18 (approved by thorpej): For ELF, add a WEAK_ALIAS() macro.
1.19.6.2	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.19.6.1	11-Feb-2002	jdolecek	Sync w/ -current.
1.19.4.3	11-Dec-2002	thorpej	Sync with HEAD.
1.19.4.2	20-Jun-2002	nathanw	Catch up to -current.
1.19.4.1	28-Feb-2002	nathanw	Catch up to -current.
1.20.10.1	14-Jul-2002	gehenna	catch up with -current.
1.21.2.2	31-May-2002	thorpej	Move the IDTVEC() macro to <machine/asm.h>, and make it work with an ISO C preprocessor.
1.21.2.1	31-May-2002	thorpej	file asm.h was added on branch sommerfeld_i386mp_1 on 2002-05-31 18:07:32 +0000
1.25.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.25.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.25.2.1	03-Aug-2004	skrll	Sync with HEAD
1.28.12.5	21-Jan-2008	yamt	sync with head
1.28.12.4	15-Nov-2007	yamt	sync with head.
1.28.12.3	03-Sep-2007	yamt	sync with head.
1.28.12.2	26-Feb-2007	yamt	sync with head.
1.28.12.1	21-Jun-2006	yamt	sync with head.
1.29.2.1	01-Feb-2006	yamt	sync with head.
1.30.18.3	06-Feb-2007	ad	Remove now unused SPLLOWER() macro.
1.30.18.2	26-Jan-2007	ad	Add a macro to define a global label.
1.30.18.1	10-Sep-2006	ad	Add updated locking primatives.
1.31.22.2	21-Nov-2007	joerg	Sync with HEAD.
1.31.22.1	09-Aug-2007	jmcneill	Sync with HEAD.
1.31.18.1	15-Aug-2007	skrll	Sync with HEAD.
1.31.16.1	07-Aug-2007	matt	Sync with HEAD.
1.31.14.1	03-Oct-2007	garbled	Sync with HEAD
1.31.6.4	03-Dec-2007	ad	Sync with HEAD.
1.31.6.3	09-Oct-2007	ad	Sync with head.
1.31.6.2	20-Aug-2007	ad	Sync with HEAD.
1.31.6.1	29-Jul-2007	ad	- When zeroing/copying pages, use SSE2 movtni to avoid polluting the cache. - By default, align assembly routines on 32-byte starting boundaries. - There are now 8 interrupt priority levels, half of which are softints. Update intrdefs.h to match. - Always clear/set spinlock words - removes lots of ifdefs. - Remove the horrible ci_self150 hack that I introduced. - Overhaul how TLB shootdown is performed. Inspired by a similar change in OpenBSD but implemented quite differently. This should be a lot faster but I have not benchmarked it yet.
1.33.6.1	18-Nov-2007	bouyer	Sync with HEAD
1.33.2.2	09-Jan-2008	matt	sync with HEAD
1.33.2.1	06-Nov-2007	matt	sync with HEAD
1.34.2.2	27-Dec-2007	mjf	Sync with HEAD.
1.34.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.35.6.2	02-Jan-2008	bouyer	Sync with HEAD
1.35.6.1	13-Dec-2007	bouyer	Sync with HEAD
1.35.4.1	13-Dec-2007	yamt	sync with head.
1.35.2.1	26-Dec-2007	ad	Sync with head.
1.37.10.1	16-May-2008	yamt	sync with head.
1.37.8.1	18-May-2008	yamt	sync with head.
1.37.6.1	02-Jun-2008	mjf	Sync with HEAD.
1.38.22.1	05-Mar-2011	rmind	sync with head
1.38.14.2	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.38.14.1	10-Jan-2011	jym	Sync with HEAD
1.39.6.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.40.16.1	18-May-2014	rmind	sync with head
1.40.12.2	03-Dec-2017	jdolecek	update from HEAD
1.40.12.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.40.2.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.41.6.1	29-May-2016	skrll	Sync with HEAD
1.42.28.1	20-Apr-2020	bouyer	Sync with HEAD
1.42.18.1	21-Apr-2020	martin	Sync with HEAD
1.45.2.1	02-Aug-2025	perseant	Sync with HEAD
1.14	20-Dec-2007	ad	- Make __cpu_simple_lock and similar real functions and patch at runtime. - Remove old x86 atomic ops. - Drop text alignment back to 16 on i386 (really, this time). - Minor cleanup.
1.13	17-Oct-2007	garbled	branches: 1.13.2; 1.13.4; 1.13.8; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.12	03-Oct-2007	ad	Make the atomics inline unless !__GNUC__.
1.11	26-Sep-2007	ad	Make build in userspace again.
1.10	26-Sep-2007	ad	x86 changes for pcc and LKMs. - Replace most inline assembly with proper functions. As a side effect this reduces the size of amd64 GENERIC by about 120kB, and i386 by a smaller amount. Nearly all of the inlines did something slow, or something that does not need to be fast. - Make curcpu() and curlwp functions proper, unless __GNUC__ && _KERNEL. In that case make them inlines. Makes curlwp LKM and preemption safe. - Make bus_space and bus_dma more LKM friendly. - Share a few more files between the ports. - Other minor changes.
1.9	24-Mar-2007	christos	branches: 1.9.2; 1.9.4; 1.9.12; 1.9.14; 1.9.16; KNF, properly protect.
1.8	09-Feb-2007	ad	branches: 1.8.2; 1.8.6; 1.8.8; 1.8.10; Merge newlock2 to head.
1.7	28-Dec-2005	perry	branches: 1.7.20; inline -> __inline
1.6	26-Dec-2005	perry	u_intN_t -> uintN_t
1.5	24-Dec-2005	perry	__asm__ -> __asm __const__ -> const __inline__ -> inline __volatile__ -> volatile
1.4	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.3	26-Feb-2003	fvdl	branches: 1.3.18; Redirect some includes to x86/foo.h
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file atomic.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.6	03-May-2002	sommerfeld	Lint.
1.1.2.5	11-Dec-2001	sommerfeld	Tweak to let this allow userland to build
1.1.2.4	31-Dec-2000	sommerfeld	Tweak i386_atomic_clearbits_l based on suggestions from mycroft
1.1.2.3	18-Aug-2000	sommerfeld	setbits/clearbits now take volatile u_int32_t *
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	Primitives for a few atomic operations.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file atomic.h was added on branch nathanw_sa on 2002-10-18 02:37:53 +0000
1.3.18.5	21-Jan-2008	yamt	sync with head
1.3.18.4	27-Oct-2007	yamt	sync with head.
1.3.18.3	03-Sep-2007	yamt	sync with head.
1.3.18.2	26-Feb-2007	yamt	sync with head.
1.3.18.1	21-Jun-2006	yamt	sync with head.
1.7.20.2	27-Jan-2007	ad	Work around gcc trying to use a register only available on x86_64. XXX
1.7.20.1	29-Dec-2006	ad	Checkpoint work in progress.
1.8.10.1	29-Mar-2007	reinoud	Pullup to -current
1.8.8.1	11-Jul-2007	mjf	Sync with head.
1.8.6.2	09-Oct-2007	ad	Sync with head.
1.8.6.1	10-Apr-2007	ad	Sync with head.
1.8.2.1	15-Apr-2007	yamt	sync with head.
1.9.16.1	06-Oct-2007	yamt	sync with head.
1.9.14.2	23-Mar-2008	matt	sync with HEAD
1.9.14.1	06-Nov-2007	matt	sync with HEAD
1.9.12.2	04-Oct-2007	joerg	Sync with HEAD.
1.9.12.1	02-Oct-2007	joerg	Sync with HEAD.
1.9.4.1	03-Oct-2007	garbled	Sync with HEAD
1.9.2.1	18-Apr-2007	thorpej	Convert i386 and amd64 to the new atomic ops API.
1.13.8.1	02-Jan-2008	bouyer	Sync with HEAD
1.13.4.1	26-Dec-2007	ad	Sync with head.
1.13.2.1	27-Dec-2007	mjf	Sync with HEAD.
1.3	18-Oct-2011	dyoung	Define some optional routines that will help device_register() to register ISA & PCI devices. Add stub implementations of the routines.
1.2	04-Feb-2006	jmmv	Revert yesterday's change that attempted to fix the detection of the boot device when using a Multiboot boot loader. It couldn't work because these boot loaders do not pass a checksum of the disk so matchbiosdisk() cannot really find any matches. I should have gone to sleep before commiting... Found by xtraeme@.
1.1	03-Feb-2006	jmmv	branches: 1.1.2; When booting an i386 kernel with Multiboot, properly detect the boot device by looking it up in the x86_alldisks table (instead of trying to match it to 'wd*' manually). In order to do this, move the cpu_rootconf function from x86 common code to amd64 and i386 specific one. This way, i386 can do an extra step (call the appropriate Multiboot code) in the appropriate place (after x86_matchbiosdisks and before findroot()).
1.1.2.1	22-Apr-2006	simonb	Sync with head.
1.2	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.1	17-Nov-1999	thorpej	branches: 1.1.2; 1.1.10; Basic interface to BIOS32. Separated from UCHIYAMA Yasushi's PCI BIOS patches and cleaned up a bit.
1.1.10.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.1.10.1	17-Nov-1999	bouyer	file bios32.h was added on branch thorpej_scsipi on 2000-11-20 20:09:25 +0000
1.1.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.11	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.10	11-Dec-2005	christos	branches: 1.10.74; 1.10.76; 1.10.78; merge ktrace-lwp.
1.9	27-Oct-2003	junyoung	Nuke __P().
1.8	02-Apr-2003	thorpej	branches: 1.8.2; Use PAGE_SIZE rather than NBPG.
1.7	02-May-2001	jdolecek	branches: 1.7.2; Add support for getting %es value from the bios call.
1.6	12-Jul-2000	thorpej	branches: 1.6.4; Clean up bios_reg a little, and add support for accessing the byte variants of the registers (e.g. %ah, %al, etc.).
1.5	03-Oct-1998	jtk	branches: 1.5.12; copyright has been assigned to TNF, update notice
1.4	01-Oct-1998	jtk	add prototype for bioscall()
1.3	31-Aug-1998	jtk	add 32-bit register support to BIOSCALL interface
1.2	15-Oct-1997	jtk	branches: 1.2.2; put in place a proper bioscall.h (sorry about that!) finish up incomplete job of moving structure used by bioscall() to <machine/bioscall.h>, with associated changes in include file strategy, genassym stuff, etc.
1.1	14-Oct-1997	christos	Someone [who will remain nameless] forgot this...
1.2.2.3	15-Oct-1997	thorpej	Sync w/ trunk.
1.2.2.2	15-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.2.2.1	15-Oct-1997	thorpej	file bioscall.h was added on branch marc-pcmcia on 1997-10-15 05:27:38 +0000
1.5.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.6.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.7.2.2	02-May-2001	jdolecek	Add support for getting %es value from the bios call.
1.7.2.1	02-May-2001	jdolecek	file bioscall.h was added on branch sommerfeld_i386mp_1 on 2001-05-02 13:12:47 +0000
1.8.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.8.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.8.2.1	03-Aug-2004	skrll	Sync with HEAD
1.10.78.1	16-May-2008	yamt	sync with head.
1.10.76.1	18-May-2008	yamt	sync with head.
1.10.74.1	02-Jun-2008	mjf	Sync with HEAD.
1.14	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.13	23-Feb-2003	fvdl	For x86_64, split the bootinfo memory entry words to avoid growth of the structure due to alignment.
1.12	28-Dec-2002	kristerw	The correct way to decorate objects with attributes is to use __attribute__ instead of the synonyms understood by gcc.
1.11	30-Nov-2002	fvdl	Make biosgeom struct packed, so that it works out on x86_64.
1.10	29-Nov-2002	fvdl	Think before committing.. previous wasn't actually needed, the values are guaranteed to fit in 32 bits when loading the kernel into physmem, and it avoids bootloader incompatibility.
1.9	29-Nov-2002	fvdl	Use long for ssym and esym in btinfo_symtab, so that it works out for both i386 and x86_64.
1.8	02-Jun-2000	thorpej	branches: 1.8.4; 1.8.6; Actually define the ACPI memory map entry (bi_memmap_entry).
1.7	12-Mar-1999	fvdl	branches: 1.7.8; 1.7.18; Clean up the BIOS disk matching code a bit (better naming, one structure and sysctl to export to userland). Also, only use total number of sectors given in the extended parameters if the physical chs geometry is marked invalid. Hopefully fixes a problem where BIOSs would not correctly fill in this field.
1.6	10-Mar-1999	fvdl	Create a list of native disks too, and make it retrievable. It contains all matching BIOS disks per entry, so that we have complete match info. Enable the matching code.
1.5	08-Mar-1999	drochner	add a definition to pass the "memory map", ie the outcome of INT15/0xe820, to the kernel
1.4	08-Mar-1999	fvdl	Add some fields to the already defined, but unused, bootinfo structure for BIOS geometry. Define CPU_BIOS_GEOM sysctl value in cpu.h.
1.3	28-Jan-1999	christos	Add btinfo_symtab, to pass symbol table information. Actually the current ddb_init interface sucks, since there is no magic number for a.out and it applies heuristics. It would be nice if the boot blocks passed more info.
1.2	27-Jan-1999	thorpej	Use the MI <sys/disklabel_mbr.h>, not our own machine-dependent definitions.
1.1	20-Sep-1997	drochner	branches: 1.1.2; Data structures passed from the boot loader to the kernel.
1.1.2.2	22-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.1.2.1	20-Sep-1997	thorpej	file bootinfo.h was added on branch marc-pcmcia on 1997-09-22 06:31:09 +0000
1.7.18.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.7.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.8.6.2	29-Dec-2002	thorpej	Sync with HEAD.
1.8.6.1	11-Dec-2002	thorpej	Sync with HEAD.
1.8.4.2	02-Jun-2000	thorpej	Actually define the ACPI memory map entry (bi_memmap_entry).
1.8.4.1	02-Jun-2000	thorpej	file bootinfo.h was added on branch sommerfeld_i386mp_1 on 2000-06-02 18:27:48 +0000
1.2	27-Oct-1994	cgd	new RCS ID format.
1.1	09-Jan-1994	cgd	branches: 1.1.2; LBL "/dev/audio" include file. probably will need to be hacked if "the right things" are done to bsd_audio.c
1.1.2.2	09-Jan-1994	cgd	LBL "/dev/audio" include file. probably will need to be hacked if "the right things" are done to bsd_audio.c
1.1.2.1	09-Jan-1994	cgd	file bsd_audioio.h was added on branch magnum on 1994-01-09 19:30:18 +0000
1.4	26-Oct-2008	mrg	- use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.3	31-Jan-2006	dsl	branches: 1.3.72; 1.3.76; 1.3.82; Change sys/arch/xxx/include/bswap.h to #include machine/byte_swap.h then sys/bswap.h in order to pick up the MD inline routines and the constant folding definitions in the right order. Code can include either sys/bswap.h or machine/bswap.h with the same effect.
1.2	21-Aug-1999	simonb	branches: 1.2.54; 1.2.66; Include <sys/bswap.h> for function prototypes. i386, pc532 and vax still include <machine/byte_swap.h> and define macros for some of the bswap*() functions.
1.1	15-Jan-1999	bouyer	Move the bswap functions from libutil to libc (this bups the minor of libc and the major of libutil). For little-endian architectures merge the bnswap() assembly versions with nto* and hton* using symbols aliasing. Use symbol renaming for the bswap function in this case to avoid namespace pollution. Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian machines, common code for inline macros go in machine/byte_swap.h Sync libkern with libc. Adjust #include in kernel sources for machine/bswap.h.
1.2.66.1	01-Feb-2006	yamt	sync with head.
1.2.54.1	21-Jun-2006	yamt	sync with head.
1.3.82.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.3.76.1	04-May-2009	yamt	sync with head.
1.3.72.1	17-Jan-2009	mjf	Sync with HEAD.
1.42	17-Jul-2011	dyoung	On i386, good-bye <machine/bus.h>.
1.41	26-Feb-2003	fvdl	branches: 1.41.122; Redirect some includes to x86/foo.h
1.40	28-Jan-2003	kent	Introduce BUS_DMA_NOCACHE, and bus_dmamem_map() of i386 supports it.
1.39	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.38	10-Nov-2001	perry	trivially rename some macro parameters to avoid lint warnings.
1.37	23-Sep-2001	jdolecek	branches: 1.37.2; fix typo
1.36	04-Sep-2001	thorpej	branches: 1.36.2; Implement bus_space_mmap().
1.35	19-Jul-2001	thorpej	Add BUS_DMA_READ and BUS_DMA_WRITE flags, that hint the back-end at dmamap load time that the mapping will be used for a unidirectional transfer of the specified direction.
1.34	30-Apr-2001	lukem	branches: 1.34.2; remove some lint, including ansifying some inlines
1.33	25-Apr-2001	thorpej	Move _i386_memio_unmap() into bus_machdep.c; the ACPI code wants it.
1.32	10-Apr-2001	fvdl	Fix bus_space_write_multi_* in the memory case; the asm constraints were wrong.
1.31	07-Mar-2001	thorpej	Add the BUS_DMA_STREAMING flag.
1.30	31-Jul-2000	briggs	branches: 1.30.2; 1.30.4; Define bus_space_*_stream_N functions as their non-stream counterparts.
1.29	26-Jun-2000	simonb	Change the kernel mmap interface so that the offset to map is an "off_t" and the return value is a "paddr_t" to allow mappings at offsets past 2^31 bytes. Somewhat inspired by FreeBSD, which only changed the offset to a "vm_offset_t". Includes updates for the i386, pc532 and sh3 mmmmap from Jason Thorpe.
1.28	17-Apr-2000	drochner	branches: 1.28.6; implement bus_space_vaddr()
1.27	15-Mar-2000	drochner	explicitely #include <sys/systm.h> #if BUS_SPACE_DEBUG to get a printf() prototype
1.26	13-Mar-2000	kristerw	Make the inline asm conform to what gcc thinks they should look like (per discussions in http://gcc.gnu.org/ml/gcc-patches/1998-11/msg00004.html) Approved by fvdl.
1.25	25-Jan-2000	drochner	define a "BUS_SPACE_MAP_PREFETCHABLE" flag which basically means that device accesses are idempotent (but should not be cached by the CPU)
1.24	18-Jun-1999	cgd	branches: 1.24.2; nuke current definitions provided by __BUS_SPACE_COMPAT_OLDDEFS. I'll do what I can to make new ones soon. 8-)
1.23	23-Mar-1999	drochner	branches: 1.23.4; Add a macro to check for sufficient pointer alignment in bus.h context, "BUS_SPACE_ALIGNED_POINTER()". Equal to the param.h "ALIGNED_POINTER()" normally, but obeys additional requirements of the bus_space_xxx_n() macros. (BUS_SPACE_DEBUG)
1.22	03-Oct-1998	thorpej	Add a private bounce threshold to the DMA map, initialized in a bus-dependent fashion.
1.21	30-Aug-1998	cgd	don't define __BUS_SPACE_COMPAT_OLDDEFS by default. As stated in bus_space(9), if drivers want it (they shouldn't; easy to convert) they can define it right before including bus.h. There's been a release since the interfaces were (slightly) changed, and no code in the source tree uses the old interfaces as far as I can tell.
1.20	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.19	09-Jun-1998	thorpej	Don't enable BUS_SPACE_DEBUG automatically if DEBUG is defined.
1.18	07-Jun-1998	thorpej	Add debugging support to the bus_space macros. If BUS_SPACE_DEBUG (or DEBUG) is defined, the bus_space macros will check to ensure that the bus address and the target buffer (if applicable) are aligned properly for the size of the type being used. If they are not, a message will be displayed on the console. While strict alignment is not strictly necessary on the x86, ensuring proper alignment can aid performance, and help make drivers more portable to architectures (like the Alpha and StrongARM) which _do_ require strict alignment.
1.17	03-Jun-1998	thorpej	Add a `bounce threshold' to the i386 DMA tag.
1.16	26-Apr-1998	thorpej	Garbage-collect the DMA tag's "_cookie" member; it's not used for anything.
1.15	04-Feb-1998	thorpej	Add offset and length parameters to bus_dmamap_sync(), used for specifiying partial syncs of a DMA mapping.
1.14	04-Feb-1998	thorpej	Change the last argument of bus_dmamap_sync() from bus_dmasync_op_t to int, and allow more than one synchronization operation to be specified in a single call. Dissallow mixing of PRE and POST operations.
1.13	04-Feb-1998	thorpej	Add dm_mapsize to bus_dmamap_t and rename BUS_DMAMEM_NOSYNC to BUS_DMA_COHERENT.
1.12	01-Oct-1997	fvdl	Provide workaround (by Charles Hannum) for error in register allocation that could occur in bus_space_write_multi*
1.11	15-Aug-1997	cgd	Sync with bus_space(9): * make map and alloc take 'flags' rather than 'cacheable,' for more flexibility. * rename BUS_BARRIER_* to BUS_SPACE_BARRIER_, for consistency. rename bus_space_copy_* to bus_space_copy_region_* and make their defns match the updated spec. Backward compatibility is provided by defining __BUS_SPACE_COMPAT_OLDDEFS, which is currently defined by default.
1.10	12-Aug-1997	perry	fix spelling error, pr-3974 from Dave Huang
1.9	10-Jul-1997	cgd	branches: 1.9.2; rename i386 bus_space_{map,unmap,subregion,alloc,free} implementation functions (which only work on memory and i/o space) to i386_memio_, and make bus_space_ in bus.h be #defines which invoke them. This makes life easier for people who need to define the all of the bus_space functions so that they work on spaces other than memory and I/O space. Also, add an _i386_memio_map function which is like i386_memio_map but doesn't do the extent map checking or allocation. _i386_memio_map and i386_memio_* are for use only by machine-dependent code.
1.8	06-Jun-1997	thorpej	Pull thorpej-bus-dma branch into mainline.
1.7	02-Dec-1996	thorpej	branches: 1.7.8; Oops, bus_space_set_{multi,region}_N() is supposed to have an `offset' argument. Add one to the `prototype' in the comment which preceeds `not implemented' :-)
1.6	10-Nov-1996	thorpej	Two things pointed out by Chris Demetriou <cgd@cs.cmu.edu>: - The boundary argument to bus_space_alloc() should be a bus_size_t, not a bus_addr_t. - The buffer arguments in the "multiple write" methods should have const qualifiers. And one from me: - Make bus_space_barrier() eat up the arguments passed to it so that the compiler doesn't needlessly whine.
1.5	21-Oct-1996	thorpej	New bus.h implementation/interface: - No more distinction between i/o-mapped and memory-mapped devices. It's all "bus space" now, and space tags differentiate the space with finer grain than the bus chipset tag. - Add memory barrier methods. - Implement space alloc/free methods. - Implement region read/write methods (like memcpy to/from bus space). This interface provides a better abstraction for dealing with machine-independent chipset drivers.
1.4	23-Jun-1996	thorpej	New implementation of bus_io_{,un}map() for the i386. Details: - A fixed extent map (statically allocated descriptor storage) is created in init386(), just before the call to consinit(). The fixed descriptor storage has enough room for 8 region entires, which is plenty for early initialization, but doesn't chew up that much memory. This extent map (ioport_ex) manages the i386 i/o port space (0x0 - 0xffff). - Just before the call to configure() in cpu_startup(), a flag is set which notifies the bus_io functions that it is safe to use malloc() to allocate descriptor storage, in the event that more than 8 regions are needed. - bus_io_map() attempts to allocate the specified region from ioport_ex. If the allocation succeeds, the io handle is filled in. If the allocation fails, it is implied that something else is already using that io space, and an error condition is returned. - bus_io_unmap() frees a region previously allocated from ioport_ex in bus_io_map(). If the free fails, a warning is printed on the conole. These changes implement "port accounting". This is required for proper autoconfiguration on the i386 port, and makes dealing with, among other things, PCMCIA io mappings _much_ easier.
1.3	03-May-1996	christos	Avoid unused variable warnings, by using the first argument to the bus functions.
1.2	05-Apr-1996	thorpej	Add bus_io_{read,write}_multi_{1,2,4,8}() to the <machine/bus.h> interface, and implement them (except for the 8-byte versions) on the i386.
1.1	08-Mar-1996	cgd	bus memory and I/O space access functions/macros, to be used by machine-independent code for more sane access to bus resources. New functions will be added to this set, in the future, as appropriate, but this is a good starting set. Defines: bus_{io,mem}_{map,unmap} bus_{io,mem}_{read,write}_{1,2,4,8} functions, and several types to go with them.
1.7.8.4	06-Jun-1997	thorpej	Right, that's a bonk on the head! Zero-length arrays aren't legal in C. Make the bus_dmamap_t variable-length array size 1, and compensate for the size screw in _bus_dmamap_create(). Pointed out by Chris Demetriou.
1.7.8.3	25-May-1997	thorpej	Implement the missing methods in the i386 bus space implementation: bus_space_set_multi, bus_space_set_region, bus_space_copy
1.7.8.2	17-May-1997	thorpej	Add two arguments to bus_dmamem_alloc(): - alignment - Each segment in the allocation will be aligned to this power-of-two boundary. - boundary - No segment in the allocation may cross this power-of-two boundary. If no boundary is necessary, specify 0.
1.7.8.1	13-May-1997	thorpej	Define types, constants, and access methods required by the bus dma API specification.
1.9.2.2	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.9.2.1	23-Aug-1997	thorpej	Update marc-pcmcia branch from trunk.
1.23.4.1	21-Jun-1999	thorpej	Sync w/ -current.
1.24.2.3	21-Apr-2001	bouyer	Sync with HEAD
1.24.2.2	12-Mar-2001	bouyer	Sync with HEAD.
1.24.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.28.6.2	06-Aug-2000	briggs	Pull up revision 1.30 -- approved by thorpej -- Define bus_space_*_stream_N functions as their non-stream counterparts.
1.28.6.1	30-Jun-2000	simonb	Pull up mmap paddr_t/off_t changes from trunk.
1.30.4.6	14-Nov-2001	nathanw	Catch up to -current.
1.30.4.5	26-Sep-2001	nathanw	Catch up to -current. Again.
1.30.4.4	21-Sep-2001	nathanw	Catch up to -current.
1.30.4.3	24-Aug-2001	nathanw	Catch up with -current.
1.30.4.2	21-Jun-2001	nathanw	Catch up to -current.
1.30.4.1	09-Apr-2001	nathanw	Catch up with -current.
1.30.2.8	29-Dec-2001	sommerfeld	Yet another mergeup. New work: - Frank van der Linden's tlb shootdown fix. Mainline functionality merged: - IrDA - Kernel RCSID's - Transmeta CPU support - ACPI - XMM register access through procfs
1.30.2.7	24-Sep-2001	sommerfeld	Synch again with current, largely to test updates to my merge tools which drop an entry in MP-UPDATING so there's a deterministic way to discover the time of the last branch synch..
1.30.2.6	06-Sep-2001	sommerfeld	here we go again
1.30.2.5	20-Jul-2001	sommerfeld	merge up with -current
1.30.2.4	30-Apr-2001	sommerfeld	Catch up on some files missed the last time 'round
1.30.2.3	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.30.2.2	10-Mar-2001	he	Pull up revision 1.31, approved by sommerfeld: Add the BUS_DMA_STREAMING flag.
1.30.2.1	31-Jul-2000	he	file bus.h was added on branch sommerfeld_i386mp_1 on 2001-03-10 19:41:46 +0000
1.34.2.4	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.34.2.3	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.34.2.2	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.34.2.1	03-Aug-2001	lukem	update to -current
1.36.2.1	01-Oct-2001	fvdl	Catch up with -current.
1.37.2.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.41.122.1	27-Aug-2011	jym	Add/remove files, like in HEAD.
1.1	01-Jul-2011	dyoung	branches: 1.1.2; Per discussion at <http://mail-index.netbsd.org/tech-kern/2010/04/02/msg007941.html>, divide each machine's bus.h into bus_defs.h (constants & data types) and bus_funcs.h (macro implementations of bus_space(9) routines and MD prototypes). Note that some bus_space(9) routines' implementation will move to .c files from inline subroutines or macros in .h files. I've only made the split for machine architectures where there is PCI. All of the non-PCI-having architectures will require a similar split. These #include files are not referenced by any (committed) Makefiles or header files, yet. Changes to Makefiles, to <sys/bus.h>, and to some more machine-dependent files will dribble in before I throw the switch.
1.1.2.2	27-Aug-2011	jym	Add/remove files, like in HEAD.
1.1.2.1	01-Jul-2011	jym	file bus_defs.h was added on branch jym-xensuspend on 2011-08-27 15:59:48 +0000
1.1	01-Jul-2011	dyoung	branches: 1.1.2; Per discussion at <http://mail-index.netbsd.org/tech-kern/2010/04/02/msg007941.html>, divide each machine's bus.h into bus_defs.h (constants & data types) and bus_funcs.h (macro implementations of bus_space(9) routines and MD prototypes). Note that some bus_space(9) routines' implementation will move to .c files from inline subroutines or macros in .h files. I've only made the split for machine architectures where there is PCI. All of the non-PCI-having architectures will require a similar split. These #include files are not referenced by any (committed) Makefiles or header files, yet. Changes to Makefiles, to <sys/bus.h>, and to some more machine-dependent files will dribble in before I throw the switch.
1.1.2.2	27-Aug-2011	jym	Add/remove files, like in HEAD.
1.1.2.1	01-Jul-2011	jym	file bus_funcs.h was added on branch jym-xensuspend on 2011-08-27 15:59:48 +0000
1.2	11-Dec-2005	christos	merge ktrace-lwp.
1.1	16-Apr-2005	yamt	branches: 1.1.2; 1.1.4; 1.1.12; add files which i forgot to add with arch/x86/x86/bus_dma.c rev.1.21.
1.1.12.2	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.1.12.1	16-Apr-2005	skrll	file bus_private.h was added on branch ktrace-lwp on 2005-11-10 13:56:53 +0000
1.1.4.2	29-Apr-2005	kent	sync with -current
1.1.4.1	16-Apr-2005	kent	file bus_private.h was added on branch kent-audio2 on 2005-04-29 11:28:13 +0000
1.1.2.2	21-Apr-2005	tron	Pull up revision 1.1 (requested by yamt in ticket #175): add files which i forgot to add with arch/x86/x86/bus_dma.c rev.1.21.
1.1.2.1	16-Apr-2005	tron	file bus_private.h was added on branch netbsd-3 on 2005-04-21 18:43:01 +0000
1.17	10-Aug-2020	rin	Clean up _LKM --> _MODULE leftovers. Note that _KERNEL is always defined for modules.
1.16	23-Jul-2020	skrll	Trailing whitespace
1.15	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.14	14-Nov-2007	ad	branches: 1.14.14; 1.14.16; 1.14.18; - Remove I486_CPU, I586_CPU, I686_CPU options. They buy us nothing and clutter the code significantly. - Remove pccons.
1.13	17-Oct-2007	garbled	branches: 1.13.2; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.12	26-Sep-2007	ad	branches: 1.12.2; x86 changes for pcc and LKMs. - Replace most inline assembly with proper functions. As a side effect this reduces the size of amd64 GENERIC by about 120kB, and i386 by a smaller amount. Nearly all of the inlines did something slow, or something that does not need to be fast. - Make curcpu() and curlwp functions proper, unless __GNUC__ && _KERNEL. In that case make them inlines. Makes curlwp LKM and preemption safe. - Make bus_space and bus_dma more LKM friendly. - Share a few more files between the ports. - Other minor changes.
1.11	24-Jan-2007	hubertf	branches: 1.11.6; 1.11.14; 1.11.22; 1.11.24; 1.11.26; Remove duplicate #includes, patch contributed in private mail by Slava Semushin <slava.semushin@gmail.com>. To verify that no nasty side effects of duplicate includes (or their removal) have an effect here, I've compiled an i386/ALL kernel with and without the patch, and the only difference in the resulting .o files was in shifted line numbers in some assert() calls. The comparison of the .o files was based on the output of "objdump -D". Thanks to martin@ for the input on testing.
1.10	30-Jan-2006	dsl	branches: 1.10.18; Move all the stuff that detects bswapxx(constant) into the MI sys/bswap.h Put the minimum to define the required inline assembler or C into the MD files. NB: there may be some fallout from this!
1.9	28-Dec-2005	perry	branches: 1.9.2; inline -> __inline
1.8	26-Dec-2005	perry	u_intN_t -> uintN_t
1.7	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.6	29-Nov-2001	lukem	branches: 1.6.2; 1.6.34; minor knf
1.5	03-Nov-2001	lukem	explicitly pull in sys/types.h
1.4	02-Nov-2001	lukem	replace __byte_swap_long_variable and __byte_swap_word_variable #define ({ }) with static __inline { }
1.3	30-May-2001	mrg	branches: 1.3.4; 1.3.8; use _KERNEL_OPT
1.2	16-Mar-2000	mycroft	branches: 1.2.8; This code has nothing to do with UCB.
1.1	15-Jan-1999	bouyer	branches: 1.1.8; Move the bswap functions from libutil to libc (this bups the minor of libc and the major of libutil). For little-endian architectures merge the bnswap() assembly versions with nto* and hton* using symbols aliasing. Use symbol renaming for the bswap function in this case to avoid namespace pollution. Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian machines, common code for inline macros go in machine/byte_swap.h Sync libkern with libc. Adjust #include in kernel sources for machine/bswap.h.
1.1.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.2.8.3	08-Jan-2002	nathanw	Catch up to -current.
1.2.8.2	14-Nov-2001	nathanw	Catch up to -current.
1.2.8.1	21-Jun-2001	nathanw	Catch up to -current.
1.3.8.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.3.4.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.6.34.4	15-Nov-2007	yamt	sync with head.
1.6.34.3	27-Oct-2007	yamt	sync with head.
1.6.34.2	26-Feb-2007	yamt	sync with head.
1.6.34.1	21-Jun-2006	yamt	sync with head.
1.6.2.2	29-Nov-2001	lukem	minor knf
1.6.2.1	29-Nov-2001	lukem	file byte_swap.h was added on branch sommerfeld_i386mp_1 on 2001-11-29 02:46:56 +0000
1.9.2.1	01-Feb-2006	yamt	sync with head.
1.10.18.1	01-Feb-2007	ad	Sync with head.
1.11.26.1	06-Oct-2007	yamt	sync with head.
1.11.24.2	09-Jan-2008	matt	sync with HEAD
1.11.24.1	06-Nov-2007	matt	sync with HEAD
1.11.22.2	21-Nov-2007	joerg	Sync with HEAD.
1.11.22.1	02-Oct-2007	joerg	Sync with HEAD.
1.11.14.1	03-Oct-2007	garbled	Sync with HEAD
1.11.6.2	03-Dec-2007	ad	Sync with HEAD.
1.11.6.1	09-Oct-2007	ad	Sync with head.
1.12.2.1	18-Nov-2007	bouyer	Sync with HEAD
1.13.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.14.18.1	16-May-2008	yamt	sync with head.
1.14.16.1	18-May-2008	yamt	sync with head.
1.14.14.1	02-Jun-2008	mjf	Sync with HEAD.
1.9	20-Jan-2012	joerg	Change CMSG_SPACE and CMSG_LEN to provide Integer Constant Expressions again. This was changed in sys/socket.h r1.51 to work around fallout from the IPv6 aux data migration. It broke the historic ABI on some platforms. This commit restores compatibility for netbsd32 code on such platforms and provides a template for future changes to the CMSG_* alignment. Revert PCC/Clang workarounds in postfix and tmux.
1.8	16-Jun-2011	joerg	branches: 1.8.2; 1.8.6; Add an optional MD calling convention flag for use in libsa when space optimisation is critical. Use this on i386 to switch to register passing calling convention for the file system entry points and most assembler call backs that have to preserve at least 3 registers.
1.7	26-Oct-2008	mrg	branches: 1.7.8; 1.7.26; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.6	20-Mar-1999	thorpej	branches: 1.6.146; 1.6.150; 1.6.156; Garbage-collect.
1.5	11-Jan-1999	christos	Add ELF crap.
1.4	04-Nov-1997	thorpej	Bug fixes and cleanup from Chris Demetriou <cgd@pa.dec.com>: - fix _C_LABEL so that it actually works. - make __RENAME use _C_LABEL. - fix __RENAME so that it expects an unquoted argument. - fix __indr_reference and __warn_references so that they supply their own final semicolon. - define __warn_references to nothing if not GNU C (required by the way it's used). The __warn_references semicolon change has to be made so that __warn_references can be defined into nothing. (A ; all by itself isn't a great idea.) The __indr_reference change was made for consistency.
1.3	22-Oct-1997	thorpej	Implement __RENAME() in <machine/cdefs.h>
1.2	23-Mar-1995	jtc	branches: 1.2.18; Changed name of __weak_reference() to __indr_reference(). They really are indirect references, and I want to add a real __weak_reference() macro to <machine/cdefs.h> soon.
1.1	19-Jan-1995	jtc	This file, which will be included by <sys/cdefs.h>, will contain macros such as __warn_references() and __weak_reference() which are actually machine dependant. This will make it easier for ports that are being bootstraped with ELF and ECOFF based toolchains. This change also introduces a new macro, _C_LABEL(x). _C_LABEL expands its argument, an identifier, to a character string of the identifier name as it is represented in an object file. For most ports, _C_LABEL(x) will expand to "_x", for ELF based ports _C_LABEL(x) will expand to "x".
1.2.18.2	04-Nov-1997	thorpej	Pull up from trunk: bug fixes and cleaups.
1.2.18.1	22-Oct-1997	thorpej	Pull up from trunk: Implement __RENAME() in <machine/cdefs.h>
1.6.156.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.6.150.1	04-May-2009	yamt	sync with head.
1.6.146.1	17-Jan-2009	mjf	Sync with HEAD.
1.7.26.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.7.8.1	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.8.6.1	18-Feb-2012	mrg	merge to -current.
1.8.2.1	17-Apr-2012	yamt	sync with head
1.2	07-Jun-2006	kardel	branches: 1.2.4; 1.2.10; convert to timecounters (from branch simonb-timecounters)
1.1	04-Feb-2006	simonb	branches: 1.1.2; 1.1.4; 1.1.6; 1.1.12; Prototypes for i386-related clock functions.
1.1.12.1	19-Jun-2006	chap	Sync with head.
1.1.6.1	26-Jun-2006	yamt	sync with head.
1.1.4.2	18-Feb-2006	yamt	sync with head.
1.1.4.1	04-Feb-2006	yamt	file clock.h was added on branch yamt-uio_vmspace on 2006-02-18 15:38:37 +0000
1.1.2.3	27-May-2006	kardel	move declaration into TSC specific file
1.1.2.2	05-Feb-2006	simonb	Pull up from HEAD - file mistakenly added to HEAD instead of timecounters branch.
1.1.2.1	04-Feb-2006	simonb	file clock.h was added on branch simonb-timecounters on 2006-02-05 02:34:02 +0000
1.2.10.2	09-Sep-2006	rpaulo	sync with head
1.2.10.1	07-Jun-2006	rpaulo	file clock.h was added on branch rpaulo-netinet-merge-pcb on 2006-09-09 02:40:14 +0000
1.2.4.2	21-Jun-2006	yamt	sync with head.
1.2.4.1	07-Jun-2006	yamt	file clock.h was added on branch yamt-lazymbuf on 2006-06-21 14:52:30 +0000
1.2	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.1	04-Jun-2000	mycroft	branches: 1.1.4; 1.1.6; 1.1.8; 1.1.134; 1.1.136; 1.1.138; Move a bunch of the COFF definitions into exec_coff.h, so they can be used by other code. XXX Not tested on SH3 yet.
1.1.138.1	16-May-2008	yamt	sync with head.
1.1.136.1	18-May-2008	yamt	sync with head.
1.1.134.1	02-Jun-2008	mjf	Sync with HEAD.
1.1.8.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.1.8.1	04-Jun-2000	bouyer	file coff_machdep.h was added on branch thorpej_scsipi on 2000-11-20 20:09:26 +0000
1.1.6.2	04-Jun-2000	mycroft	Move a bunch of the COFF definitions into exec_coff.h, so they can be used by other code. XXX Not tested on SH3 yet.
1.1.6.1	04-Jun-2000	mycroft	file coff_machdep.h was added on branch sommerfeld_i386mp_1 on 2000-06-04 16:24:00 +0000
1.1.4.2	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.1.4.1	04-Jun-2000	minoura	file coff_machdep.h was added on branch minoura-xpg4dl on 2000-06-22 17:00:31 +0000
1.11	06-Sep-2002	gehenna	Merge the gehenna-devsw branch into the trunk. This merge changes the device switch tables from static array to dynamically generated by config(8). - All device switches is defined as a constant structure in device drivers. - The new grammer ``device-major'' is introduced to ``files''. device-major <prefix> char <num> [block <num>] [<rules>] - All device major numbers must be listed up in port dependent majors.<arch> by using this grammer. - Added the new naming convention. The name of the device switch must be <prefix>_[bc]devsw for auto-generation of device switch tables. - The backward compatibility of loading block/character device switch by LKM framework is broken. This is necessary to convert from block/character device major to device name in runtime and vice versa. - The restriction to assign device major by LKM is completely removed. We don't need to reserve LKM entries for dynamic loading of device switch. - In compile time, device major numbers list is packed into the kernel and the LKM framework will refer it to assign device major number dynamically.
1.10	18-Apr-2002	wiz	branches: 1.10.2; 1.10.4; Remove opms(4) and its device, /dev/pms0, from the i386 port, because it has been obsoleted by pms(4). Reviewed by fvdl and christos.
1.9	28-Feb-2002	simonb	Use "#define<tab>".
1.8	27-Feb-2002	christos	- define other DEV_ constants that the local port uses. - delete cdev_decl(mm) since <sys/conf.h> does it.
1.6	21-Mar-2001	lukem	branches: 1.6.4; now that duplicated stuff has been moved to sys/conf.h, remove unnecessary defs for: lpt, joy, pc
1.5	05-Sep-1998	christos	branches: 1.5.12; 1.5.26; Assign copyright to TNF.
1.4	15-Apr-1998	drochner	remove unneeded stuff
1.3	07-Sep-1996	mycroft	Implement poll(2).
1.2	05-May-1996	christos	Add the joystick
1.1	03-May-1996	christos	Fwd declarations for MD drivers.
1.5.26.5	17-Sep-2002	nathanw	Catch up to -current.
1.5.26.4	20-Jun-2002	nathanw	Catch up to -current.
1.5.26.3	01-Apr-2002	nathanw	Catch up to -current. (CVS: It's not just a program. It's an adventure!)
1.5.26.2	28-Feb-2002	nathanw	Catch up to -current.
1.5.26.1	09-Apr-2001	nathanw	Catch up with -current.
1.5.12.1	27-Mar-2001	bouyer	Sync with HEAD.
1.6.4.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.6.4.2	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.6.4.1	16-Mar-2002	jdolecek	Catch up with -current.
1.10.4.1	19-May-2002	gehenna	Remove port-dependent conf.h.
1.10.2.2	18-Apr-2002	wiz	Remove opms(4) and its device, /dev/pms0, from the i386 port, because it has been obsoleted by pms(4). Reviewed by fvdl and christos.
1.10.2.1	18-Apr-2002	wiz	file conf.h was added on branch sommerfeld_i386mp_1 on 2002-04-18 12:54:16 +0000
1.185	04-Sep-2023	mrg	x86: avoid annoying GCC 12 bounds check in curcpu() and curlwp(). these functions read %gs and return an pointer at an offset from this value (the current cpu, or lwp pointers), and GCC is complaining that they're accessing a array cpu_info[0] (ie, zero length, no storage.) several attempts to workaround it have failed, and because of the asm volatile nature of this, it seems very unlikely a compiler would take this and do something wrong with it.
1.184	09-Apr-2023	riastradh	i386: Make curlwp and curcpu() flushable. The only effect of the `volatile' qualifier on an asm block with outputs is to force the instructions to appear in the generated code, even if the outputs end up being unused. Since these instructions have no (architectural) side effects -- provided %fs is set correctly, which must be the case here -- there's no need for the volatile qualifier, so nix it.
1.183	02-Nov-2021	ryo	In order to prevent _mcount() from being recursively called when built with COPTS=-O0, sprinkle `__always_inline' to make _mcount() be generated as a single function.
1.182	21-Nov-2019	ad	mi_userret(): take care of calling preempt(), set spc_curpriority directly, and remove MD code that does the same.
1.181	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.180	18-Nov-2018	cherry	On Xen, copy just the bits we need from the trapframe for hardclock(9) and statclock(9). Current, the macros that use the trapframe are: CLKF_USERMODE() CLKF_PC() CLKF_INTR() Of these, CLKF_INTR() already ignores the frame and uses the ci_idepth variable to do its job. Convert the two remaining ones to do this, but only for XEN.
1.179	17-Sep-2017	maxv	branches: 1.179.2; 1.179.4; Remove the second argument from USERMODE and KERNELMODE, it is unused now that we don't have vm86 anymore.
1.178	30-Dec-2011	cherry	branches: 1.178.6; Move the per-cpu l3 page allocation code to a separate MD function. Avoids code duplication for xen PAE
1.177	26-Jul-2011	yamt	branches: 1.177.2; 1.177.6; whitespace
1.176	29-Dec-2008	pooka	branches: 1.176.2; 1.176.12; _LKM -> _MODULE
1.175	22-May-2008	ad	branches: 1.175.6; Mark x86_curlwp() with __attribute__ ((const)), so gcc can CSE it and know that it does not clobber global data.
1.174	11-May-2008	ad	Wrap stuff in #ifdef _KERNEL
1.173	11-May-2008	ad	Share cpu.h between the x86 ports.
1.172	11-May-2008	ad	Simplify x86 identcpu code, and share between i386/amd64.
1.171	10-May-2008	ad	Improve x86 tsc handling: - Ditch the cross-CPU calibration stuff. It didn't work properly, and it's near impossible to synchronize the CPUs in a running system, because bus traffic will interfere with any calibration attempt, messing up the timings. - Only enable the TSC on CPUs where we are sure it does not drift. If we are On a known good CPU, give the TSC high timecounter quality, making it the default. - When booting CPUs, detect TSC skew and account for it. Most Intel MP systems have synchronized counters, but that need not be true if the system has a complicated bus structure. As far as I know, AMD systems do not have synchronized TSCs and so we need to handle skew. - While an AP is waiting to be set running, try and make the TSC drift by entering a reduced power state. If we detect drift, ensure that the TSC does not get a high timecounter quality. This should not happen and is only for safety. - Make cpu_counter() stuff LKM safe.
1.170	09-May-2008	joerg	Make cpu_idle a macro calling a function pointer on x86. Select the Xen idle routine for Xen, mwait if supported by the CPU and it is not AMD and halt otherwise. As reported by Christoph Egger, AMD Barcelona keeps the CPU in C0 state with MWAIT, contrary to HLT, which uses C1 and therefore much less power.
1.169	30-Apr-2008	ad	branches: 1.169.2; Avoid unneeded AST faults.
1.168	28-Apr-2008	ad	Add support for kernel preeemption to the i386 and amd64 ports. Notes: - I have seen one isolated panic in the x86 pmap, but otherwise i386 seems stable with preemption enabled. - amd64 is missing the FPU handling changes and it's not yet safe to enable it there. - The usual level for kern.sched.kpreempt_pri will be 128 once enabled by default. For testing, setting it to 0 helps to shake out bugs.
1.167	24-Apr-2008	ad	branches: 1.167.2; - Give ci_want_resched a single cache line, and align. This is for monitor/ mwait. At least one errata sheet from Intel notes that a single line should be used. - Align cc_microtime.
1.166	21-Apr-2008	cegger	Access Xen's vcpu info structure per-CPU. Tested on i386 and amd64 (both dom0 and domU) by me. Xen2 tested (both dom0 and domU) by bouyer. OK bouyer
1.165	27-Feb-2008	xtraeme	branches: 1.165.2; Add back the CPU_TMLR_* definitions for binary sysctl compatibility with NetBSD pre-2.0, and use them in the longrun driver again. Requested by cube@.
1.164	27-Feb-2008	xtraeme	Remove CTL_MACHDEP_NAMES, it's not used anywhere. Ok by martin@.
1.163	27-Feb-2008	xtraeme	- Fixup the machdep.tm_longrun_* sysctl nodes creation. - The CTL_TMLR_ static definitions for CTL_MACHDEP are not needed anymore.
1.162	26-Feb-2008	xtraeme	Split off the Transmeta Crusoe Longrun code into its own file, that way identcpu.c and machdep.c are not cluttered with foreign code. The driver is built by default as before, but the sysctl subtree will only be created if longrun is detected and not always as the old code did. This matches what the FreeBSD code does. Ok by christos@.
1.161	10-Feb-2008	ad	branches: 1.161.2; 1.161.6; Align cc_microtime and struct cpu_info to 64b.
1.160	16-Jan-2008	ad	Remove options MATH_EMULATE.
1.159	15-Jan-2008	joerg	Introduce optional cpu_offline_md to execute MD actions at the end of cpu_offline. Use this on amd64/i386 to force a FPU save. As this was triggered by npxsave_cpu/fpusave_cpu not working for a different CPU, remove the cpu_info argument and adjust npxsave_/fpusave_ to use bool for the save. OK ad@
1.158	11-Jan-2008	bouyer	Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.157	05-Jan-2008	yamt	move IOMAP_INVALOFF from cpu.h to tss.h
1.156	05-Jan-2008	yamt	g/c ci_idle_pcb_paddr
1.155	04-Jan-2008	yamt	i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.154	01-Jan-2008	yamt	try to detect processor resource sharing topologies. ie. package/core/smt IDs.
1.153	25-Dec-2007	perry	Convert many of the uses of __attribute__ to equivalent __packed, __unused and __dead macros from cdefs.h
1.152	18-Dec-2007	joerg	Add new IPI for saving CPU state explicitly, share high-level part of ACPI wakeup code and teach it how to start the APs again. As a side effect the CPU_START interface allows choosing between different bootstrap codes more easily now.
1.151	03-Dec-2007	joerg	branches: 1.151.4; Add a CPU local timer based on the LAPIC. This is consistently faster than TSC, but doesn't suffer from SpeedStep as TSC does. The default quality is higher than HPET for UP, but -100 for MULTIPROCESSOR as it needs CPU local state which doesn't exist yet.
1.150	29-Oct-2007	ad	branches: 1.150.2; 1.150.4; Mark cpu_info::ci_tlbstate volatile to ensure that the compiler doesn't reorder accesses to it. It's updated from the TLB IPI handlers and we don't block those, so the order in which things are read/updated is important.
1.149	26-Oct-2007	joerg	Match delay/DELAY on x86 with delay(9). It takes an unsigned int as argument. Use this and replace the inline assembly (mul + div using the 64bit intermediate result) with normal 32bit multiplication and division. The compiler can turn the division into a multiplication and shift, making it even cheaper then the original assembly. For extreme long delays, just use 64bit arithmetic.
1.148	18-Oct-2007	yamt	merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.147	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.146	26-Sep-2007	ad	branches: 1.146.2; x86 changes for pcc and LKMs. - Replace most inline assembly with proper functions. As a side effect this reduces the size of amd64 GENERIC by about 120kB, and i386 by a smaller amount. Nearly all of the inlines did something slow, or something that does not need to be fast. - Make curcpu() and curlwp functions proper, unless __GNUC__ && _KERNEL. In that case make them inlines. Makes curlwp LKM and preemption safe. - Make bus_space and bus_dma more LKM friendly. - Share a few more files between the ports. - Other minor changes.
1.145	25-Sep-2007	ad	ci_astpending is no more.
1.144	29-Aug-2007	ad	branches: 1.144.2; Merge most x86 changes from the vmlocking branch, except the threaded soft interrupt stuff. This is mostly comprised of changes to the pmap modules to work on multiprocessor systems without kernel_lock, and changes to speed up tlb shootdowns.
1.143	09-Jul-2007	ad	branches: 1.143.4; 1.143.8; 1.143.10; Merge some of the less invasive changes from the vmlocking branch: - kthread, callout, devsw API changes - select()/poll() improvements - miscellaneous MT safety improvements
1.142	03-Jun-2007	xtraeme	Make the Enhanced Speedstep driver available for i386 and amd64. To use it on EM64T CPUs supporting the EST CPUID feature. Note that some CPUs still don't work with this driver, like Xeon or Pentium 4. Move the p[34]_get_bus_clock functions into its own file, intel_busclock.c and remove this code from i386/identcpu.c. Tested on i386 by myself and amd64 by Tonerre.
1.141	17-May-2007	yamt	merge yamt-idlelwp branch. asked by core@. some ports still needs work. from doc/BRANCHES: idle lwp, and some changes depending on it. 1. separate context switching and thread scheduling. (cf. gmcgarry_ctxsw) 2. implement idle lwp. 3. clean up related MD/MI interfaces. 4. make scheduler(s) modular.
1.140	16-Apr-2007	ad	branches: 1.140.2; Share the sysarch stuff between the x86 ports. PR kern/36046.
1.139	21-Mar-2007	xtraeme	- Remove ci_msr_rvalue, it's not useful anymore as yamt@ pointed out. - Remove completely debug from msr_ipifuncs, now it's known to work.
1.138	20-Mar-2007	xtraeme	MSR read and write IPI handlers for x86. A MSR will be read or written in all CPUs available in the system. This adds another member to struct cpu_info, ci_msr_rvalue; it will contain the value of the MSR in a previous operation. Tested with clockmod in UP and SMP by me, tested with est in SMP by Daniel Carosone and Michael Van Elst. Ok'ed by Andrew Doran and Matthew R. Green.
1.137	18-Mar-2007	xtraeme	There's no need to run est_init or k8_powernow_init on each CPU. Just run it once (in the first cpu probed) with the RUN_ONCE(9) framework. Change the argument of est_init and k8_powernow_init to void, we don't need cpu_info * anymore. Suggested by tls@ and mrg@.
1.136	12-Mar-2007	ad	branches: 1.136.2; 1.136.4; Include sys/simplelock.h, not sys/lock.h.
1.135	05-Mar-2007	drochner	branches: 1.135.2; clean up how cpus and ioapics are attached at the mainbus: Seperate "cpubus" and "ioapicbus" -- while they share a common "address space" (the apic id), the kernel doesn't use this fact. There are different data passed to cpus and apics, which caused some ugly polymorphism. This also saves the special "submatch" functions needed to distingush cpus and ioapics for autoconf. (And it makes that "apid" locators wired in the kernel configuration are honored now; this allows one to dumb down an mp box to singleprocessor by userconfig.) Print "apid" locators in the buses "print" function "as everyone does", so the per-port cpu drivers don't need to do it. Being here, constify "struct cpu_functions" and g/c the unused MP_PICMODE flag.
1.134	04-Mar-2007	christos	Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.133	17-Feb-2007	daniel	branches: 1.133.2; Add an opencrypto provider for the AES xcrypt instructions found on VIA C5P and later cores (also known as 'ACE', which is part of the VIA PadLock security engine). Ported from OpenBSD. Reviewed on tech-crypto and port-i386, no objections to commiting this.
1.132	16-Feb-2007	ad	Remove spllowersoftclock() and CLKF_BASEPRI(), and always dispatch callouts via a soft interrupt. In the near future, softclock will be run from process context.
1.131	09-Feb-2007	ad	Merge newlock2 to head.
1.130	08-Dec-2006	yamt	- pass intrframe by-pointer, not by-value. - make i386 and xen use per-cpu interrupt stack. xen part is reviewed by Manuel Bouyer.
1.129	03-Sep-2006	xtraeme	branches: 1.129.2; 1.129.4; 1.129.6; Conditionalize *_get_bus_clock() functions inside of #ifdef ENHANCED_SPEEDSTEP to fix build when ENHANCED_SPEEDSTEP is not defined. Reported by Geoff Wing.
1.128	03-Sep-2006	xtraeme	Update the enhanced speedstep driver and sync the code with OpenBSD: est.c: * Use a quintuplet (vendor, MHz_hi, mV_hi, MHz_lo, mV_lo } to match CPUs more correctly than parsing the brand string. * Add support for a bunch of models. * Create a fake table on the fly if the CPU is unknown (there's no table for it) with the current/highest/lowest frequency. specialreg.h: * Add some MSRs needed to get the bus clock value. identcpu.c: * Add functions specific to Pentium III, Pentium M and Pentium 4 to get the bus clock value. Note that the new fake table code from Simon Burge is not included on this commit. Ok'ed by simonb and dogcow.
1.127	23-Aug-2006	xtraeme	- Move k7_powernow_* prototypes from i386/include/cpu.h to x86/include/powernow.h - Protect k[78]_powernow_init() functions with #ifdef POWERNOW_K[78] to make it build without these options. This fixes the problem reported by hubertf.
1.126	08-Aug-2006	xtraeme	- Add k7_powernow_destroy() prototype into cpu.h. - Call k7_powernow_destroy() unloading the module.
1.125	06-Aug-2006	xtraeme	branches: 1.125.2; * Sync powernow_k7.c driver with OpenBSD, it seems to work much better finding PST tables. * Use functions defined in powernow_common.c to detect if the CPU can run powernow. * Update the LKM with current code.
1.124	07-Jun-2006	kardel	convert to timecounters (from branch simonb-timecounters)
1.123	16-Feb-2006	perry	branches: 1.123.2; 1.123.8; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.122	03-Feb-2006	xtraeme	branches: 1.122.2; Fix "hard reboot" problem reported by Rhialto weeks ago on current-users@. The problem was that pnowk7_init() was called too early in the boot process, at this point the required calls were not available. Thanks to Rhialto for testing and cube/christos for comments.
1.121	31-Dec-2005	xtraeme	branches: 1.121.2; AMD PowerNow K7 driver written by Martin Vegiard via PR port-i386/26239. Enabled by default on GENERIC and GENERIC_LAPTOP. Imported 1 year later... but it's here finally.
1.120	26-Dec-2005	perry	branches: 1.120.2; u_intN_t -> uintN_t
1.119	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.118	11-Aug-2005	cube	Change all archs that did: #define clockframe somethingelse to: struct clockframe { struct somethingelse cf_se; }; and change access macros accordingly. That means that, at least for that very issue, things will not go ka-boomy if you don't have the actual definition of struct clockframe before including systm.h.
1.117	21-Feb-2005	he	branches: 1.117.4; Probe and print the Intel Extended Feature Bits, as documented in the CPUID instruction description in the "Intel Extended Memory 64 Technology Software Developer's Guide, Volume 1 of 2" available at ftp://download.intel.com/technology/64bitextensions/30083402.pdf This presently consists of the SYSCALL/SYSRET and the EM64T features. CPUs with the EM64T feature available should be able to run amd64 code. Reviewed by fvdl
1.116	22-Sep-2004	yamt	branches: 1.116.4; 1.116.6; move some per-cpu data definitions to MI place so that they can be modified without touching all ports. discussed on tech-kern@.
1.115	16-May-2004	yamt	curcpu: fix a typo. (cpuinfo -> cpu_info) no functional difference.
1.114	30-Apr-2004	lukem	Integrate support for Enhanced SpeedStep Technology (est) that Michael Eriksson posted to port-i386 on 20031102, with various modifications by me to work in the new sysctl(9) framework. The code is enabled with 'options ENHANCED_SPEEDSTEP', and if the CPU supports EST the following sysctl(8) nodes appear (with the values that a Dell Inspiron 8600 + WUXGA with a 1.4GHz Pentium M CPU supports): machdep.est.cpu_brand = Intel(R) Pentium(R) M processor 1400MHz machdep.est.frequency.target = 1400 machdep.est.frequency.current = 1400 machdep.est.frequency.available = 1400 1200 1000 800 600 If EST support isn't available, the "machdep.est" sysctl sub-MIB is not created. Once we have a more general "CPU frequency" control API we can migrate this code to using that. Thanks to Michael Erikkson for providing this code!
1.113	20-Feb-2004	yamt	defer pmap switching until it's really needed to avoid frequent loading of cr3 register, which involves tlb flush. with some fixes/improvements from Stephan Uphoff and Bang Jun-Young.
1.112	04-Jan-2004	jdolecek	Process and lwp exit path always executes lwp_exit2() now and cpu_exit() is empty besides calling switch_exit(). So, rename switch_exit() to cpu_exit() and modify the routine to call lwp_exit2() direct. This saves couple cycles on the exit path.
1.111	30-Dec-2003	pk	Replace the traditional buffer memory management -- based on fixed per buffer virtual memory reservation and a private pool of memory pages -- by a scheme based on memory pools. This allows better utilization of memory because buffers can now be allocated with a granularity finer than the system's native page size (useful for filesystems with e.g. 1k or 2k fragment sizes). It also avoids fragmentation of virtual to physical memory mappings (due to the former fixed virtual address reservation) resulting in better utilization of MMU resources on some platforms. Finally, the scheme is more flexible by allowing run-time decisions on the amount of memory to be used for buffers. On the other hand, the effectiveness of the LRU queue for buffer recycling may be somewhat reduced compared to the traditional method since, due to the nature of the pool based memory allocation, the actual least recently used buffer may release its memory to a pool different from the one needed by a newly allocated buffer. However, this effect will kick in only if the system is under memory pressure.
1.110	30-Dec-2003	yamt	g/c unused x86_ipisend.
1.109	27-Oct-2003	junyoung	Nuke __P().
1.108	10-Oct-2003	simonb	Take previous even further - leave only CTL_MACHDEP and related definitions and structures outside the #ifdef _KERNEL block.
1.107	10-Oct-2003	yamt	revert the previous (thanks, lukem) and wrap curcpu, cpu_info and friends with #ifdef _KERNEL completely.
1.106	10-Oct-2003	lukem	Userland needs <stddef.h> for offsetof(), not <lib/libkern/libkern.h> ... (Arguably a chunk more of this file could be protected with _KERNEL. That's a separate project.)
1.105	09-Oct-2003	yamt	avoid a magic number in curcpu().
1.104	06-Sep-2003	christos	SA_SIGINFO changes.
1.103	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.102	26-Jun-2003	drochner	branches: 1.102.2; -make cpu_feature unsigned, just like ci_feature_flags -clean up duplicate declaration of cpu_info_primary
1.101	20-May-2003	kristerw	Use __asm instead of asm, to make lint happy.
1.100	25-Apr-2003	fvdl	Share some common cache info cpuid code between i386 and x86_64.
1.99	18-Mar-2003	fvdl	Add some underscores in curcpu() to avoid a -Wshadow.
1.98	01-Mar-2003	fvdl	Move the CPU identification out of machdep.c into a seperate file.
1.97	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.96	05-Feb-2003	nakayama	Share i386/tsc_microtime.c with alpha and sparc64 as kern_microtime.c. (approved by martin)
1.95	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.94	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.93	16-Dec-2002	jdolecek	use the __HAVE_CPU_MAXPROC hook to limit kern.maxproc to number of available GDT slots add code in init386() to force maxproc to be maximum cpu_maxproc() this fixes port-i386/1635 by John Kohl
1.92	06-Dec-2002	junyoung	Don't pollute struct cpu_info with throwaway variables.
1.91	06-Dec-2002	junyoung	Display the extended feature flags with non-Intel processors rather than the standard flags. See also PR#19163. Before: cpu0: AMD Athlon XP 1800+ (686-class), 1532.11 MHz cpu0: features 383f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR> cpu0: features 383f9ff<PGE,MCA,CMOV,FGPAT,PSE36,MMX> cpu0: features 383f9ff<FXSR,SSE> After: cpu0: AMD Athlon XP 1800+ (686-class), 1532.11 MHz cpu0: features c3cbf9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR> cpu0: features c3cbf9ff<PGE,MCA,CMOV,PAT,PSE36,MPC,MMXX,MMX> cpu0: features c3cbf9ff<FXSR,SSE,3DNOW2,3DNOW> While I'm here, amd_cpuid_cpu_cacheinfo() is an info function rather than a probe function.
1.90	28-Nov-2002	fvdl	Add ci_apicid field.
1.89	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.88	02-Nov-2002	perry	/CONTCOND/ a while (0) macro
1.87	06-Oct-2002	fvdl	Remove ci_lapic_ints from struct cpu_info again, it isn't needed anymore.
1.86	06-Oct-2002	fvdl	Keep size of struct cpu_info independent of DIAGNOSTIC/LOCKDEBUG (was done in rev. 1.81, got lost in the MP merge).
1.85	06-Oct-2002	fvdl	Add per-CPU local apic redir table (2 pointers).
1.84	06-Oct-2002	fvdl	cpu_swapin now exists, so remove the empty define.
1.83	05-Oct-2002	fvdl	Add mask for TLB IPI rendezvous, and stackpointers + TSS structures for TSS gates per CPU. XXX struct cpu_info is getting to be a bit large
1.82	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.81	09-Jul-2002	jdolecek	always include ci_s*_locks in struct cpu_info, so that the size doesn't depend on DIAGNOSTIC/LOCKDEBUG settings
1.80	12-May-2002	matt	branches: 1.80.2; 1.80.4; Eliminate commons.
1.79	08-May-2002	mycroft	Wire CLKF_BASEPRI() to 0 on this platform. (See tech-kern.)
1.78	31-Jan-2002	christos	Fix comments.
1.77	27-Dec-2001	christos	add a new function to the cpu struct's that can be called to print extra cpu information after initialization.
1.76	17-Nov-2001	christos	Add transmeta crusoe cpu support from toshi.
1.75	02-Aug-2001	thorpej	- Rename cpu_use_fxsave to i386_use_fxsave. - If we detect SSE/SSE2 support in the CPU, enable SSE exceptions and set i386_has_{sse,sse2} as appropriate. - Expose i386_use_fxsave and i386_has_{sse,sse2} through sysctl as machdep.{osfsxr,sse,sse2}.
1.74	02-Aug-2001	thorpej	Add support for saving/restoring SSE/SSE2 state using FXSAVE/FXRSTOR. Reviewed by Frank.
1.73	01-Aug-2001	thorpej	Add several more Intel cache info entries, and fetch the CFLUSH line size if we have the CFLUSH insn.
1.72	14-Jun-2001	thorpej	branches: 1.72.2; Don't need to prototype child_return() here, it's in <sys/proc.h>.
1.71	30-May-2001	mrg	use _KERNEL_OPT
1.70	03-May-2001	thorpej	Rearrange the cache info fetching code some more, and add support for fetching cache info for AMD processors.
1.69	02-May-2001	thorpej	- Keep cache/tlb info in the cpu_info structure. - Add "associativity" to the cache_info structure. - Add a (cpu_cacheinfo)() function pointer, like we have a (cpu_setup)() function pointer. Cache info in the `cpuid' is vendor-specific.
1.68	14-Feb-2001	nathanw	branches: 1.68.2; Remove prototype for long-defunct i386_user_cleanup(). Remove duplicate prototype for i386_{set,get}_ldt() from sys_machdep.c. Change i386_iopl() and i386_{set,get}_{ldt,ioperm}() to take a second argument of "void " instead of "char ", for consistency with other syscalls.
1.67	10-Dec-2000	mycroft	Introduce PROC_PC(), which is used to get a process's user PC. If this is defined, call addupc_intr() directly from statclock() in the system time case, using the same P_OWEUPC path if the copyin/copyout fails. Use this in i386 to remove profiling code from the normal userret() path.
1.66	16-Nov-2000	jdolecek	make i386_nocpuid_cpus[] and i386_cpuid_cpus[] const - this pushes another ~1.4KB to text section
1.65	25-Aug-2000	thorpej	Make need_resched() take a "struct cpu_info *" argument. This causes gives a primitive form of processor affinity. Its use in roundrobin() still needs some work.
1.64	04-Jun-2000	mycroft	branches: 1.64.2; Implement CLKF_INTR(), to count interrupt time separately.
1.63	26-May-2000	thorpej	branches: 1.63.2; First sweep at scheduler state cleanup. Collect MI scheduler state into global and per-CPU scheduler state: - Global state: sched_qs (run queues), sched_whichqs (bitmap of non-empty run queues), sched_slpque (sleep queues). NOTE: These may collectively move into a struct schedstate at some point in the future. - Per-CPU state, struct schedstate_percpu: spc_runtime (time process on this CPU started running), spc_flags (replaces struct proc's p_schedflags), and spc_curpriority (usrpri of processes on this CPU). - Every platform must now supply a struct cpu_info and a curcpu() macro. Simplify existing cpu_info declarations where appropriate. - All references to per-CPU scheduler state now made through curcpu(). NOTE: this will likely be adjusted in the future after further changes to struct proc are made. Tested on i386 and Alpha. Changes are mostly mechanical, but apologies in advance if it doesn't compile on a particular platform.
1.62	21-Apr-2000	thorpej	- Declare biosbasemem and biosextmem in <machine/cpu.h>, don't extern them everywhere they're used. - Avoid integer overflow when converting bios{base,ext}mem (in units of kilobytes) to bytes.
1.61	29-Mar-2000	simonb	Remove redundant decl of consinit() - it's in <sys/systm.h>. Remove duplicate definition of delay() - we only need one in this file.
1.60	28-Mar-2000	thorpej	Extern cpu_id and cpu_vendor[] here.
1.59	26-Jan-2000	drochner	branches: 1.59.2; next step in getting machdep.c maintainable - put bus_space related stuff into a separate file
1.58	21-Dec-1999	drochner	machdep.c is becoming an unmaintainable mess put console initialisation into a separate file
1.57	06-Nov-1999	enami	branches: 1.57.2; - Export initrtclock(). - Call it from apm_resume(). This fixes timer problem on some laptop after hybernation (PR#4808). Reviewd by: fvdl@netbsd.org
1.56	06-Oct-1999	fvdl	branches: 1.56.2; 1.56.4; Define CPU_FPU_PRESENT sysctl value.
1.55	10-Aug-1999	thorpej	branches: 1.55.2; Define cpu_number() as discussed on tech-smp.
1.54	20-Jul-1999	thorpej	Implement cpu_wait(), and move the call to tss_free() from switch_exit() to cpu_wait(); tss_free() may block, and thus requires a valid context.
1.53	01-Apr-1999	thorpej	branches: 1.53.2; 1.53.4; Do proc0's TSS and LDT initialization at the end of configure() (i.e. once the FPU has been attached).
1.52	12-Mar-1999	fvdl	Clean up the BIOS disk matching code a bit (better naming, one structure and sysctl to export to userland). Also, only use total number of sectors given in the extended parameters if the physical chs geometry is marked invalid. Hopefully fixes a problem where BIOSs would not correctly fill in this field.
1.51	10-Mar-1999	fvdl	Create a list of native disks too, and make it retrievable. It contains all matching BIOS disks per entry, so that we have complete match info. Enable the matching code.
1.50	08-Mar-1999	fvdl	Add some fields to the already defined, but unused, bootinfo structure for BIOS geometry. Define CPU_BIOS_GEOM sysctl value in cpu.h.
1.49	11-Nov-1998	thorpej	Changes to support fork_kthread(): - cpu_set_kpc() now takes void *arg third argument, passed to the entry point. - cpu_fork() allows parent to be non-curproc iff parent is proc0. When forking non-curproc, assume its state has already been saved. - Adjust various pieces of machine-dependent code to account of all of this.
1.48	07-Oct-1998	thorpej	Erg, frame -> trapframe in last. (thanks to john kohl)
1.47	07-Oct-1998	thorpej	Oops, back out a buglet I didn't intend to commit in the last change.
1.46	06-Oct-1998	thorpej	Move the code that does the deferred (post-autoconfiguration) buffer cache allocation to its own function, and call it at the end of configure().
1.45	06-Oct-1998	thorpej	configure() prototype is in <sys/device.h>
1.44	22-Jan-1998	thorpej	Generate dependencies on the USER_LDT option.
1.43	21-Jan-1998	thorpej	Generate dependencies on the MATH_EMULATE option.
1.42	15-Jan-1998	thorpej	Generate dependencies on the VM86 option.
1.41	27-Nov-1997	bouyer	Add a cpu_setup() function to "struct cpu_nocpuid_nameclass" and "struct cpu_cpuid_family", so that we can have a cpu_setup function per known cpu type. For now use it only for cyrix 6x86 to enable suspend-on-halt and implement cyrix's workaround to the "coma bug".
1.40	20-Sep-1997	drochner	branches: 1.40.2; Define "machdep.booted_kernel" for sysctl.
1.39	04-Feb-1997	perry	branches: 1.39.8; Nuke some options GENERIC residue.
1.38	03-Dec-1996	fvdl	Added types/definitions for changed CPU recognition code.
1.37	18-Nov-1996	fvdl	Changes for NKPDE calculation from cgd. - New variables: biosextmem, biosbasemem, nkpde - Above can be set by using the options BIOSEXTMEM, BIOSBASEMEM, and NKPDE respectively (EXTMEM_SIZE is now called BIOSEXTMEM). When preset this way, they won't be filled in / calculated. - Readable by sysctl using machdep.nkpde, machdep.biosbasemem and machdep.biosextmem. - nkpde is calculated as: min(NKPDE_MAX, NKPDE_BASE + (biosextmem >> 10) * NKPDE_SCALE) Where NKPDE_MAX is 31, NKPDE_BASE is 4, NKPDE_SCALE = 1.
1.36	11-Oct-1996	christos	printf -> kprintf, sprintf -> ksprintf
1.35	05-May-1996	christos	Change the prototype of fillw from (int, caddr_t, size_t) to (short, void *, size_t).
1.34	03-May-1996	christos	Add prototypes from other files.
1.33	29-Mar-1996	mycroft	Prototype delay().
1.32	24-Dec-1995	mycroft	Enable paging of the user area.
1.31	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.30	10-Oct-1995	mycroft	Add USERMODE() and KERNELMODE() macros, and use them.
1.29	28-Jun-1995	cgd	remove unused cpu_exec() definitions. moved "broken swap" markers, for ports that still need it, to types.h.
1.28	05-May-1995	cgd	define BROKEN_SWAP and/or cpu_swapout as appropriate.
1.27	22-Apr-1995	christos	- added sunos_machdep.c for sun3, atari, amiga and mac68k. - changed machdep.c and trap.c to use struct emul. - remove ep_setup references. - added struct emul to all emulations.
1.26	21-Apr-1995	mycroft	Alias delay() to DELAY() for mi code.
1.25	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.24	27-Oct-1994	cgd	new RCS ID format.
1.23	14-Jun-1994	mycroft	profile_tick() is no longer needed.
1.22	27-May-1994	mycroft	Add cpu_set_init_frame().
1.21	23-May-1994	cgd	add cpu_swapin
1.20	19-May-1994	cgd	make cpu_wait a macro, and define cpu_coredump
1.19	07-May-1994	cgd	cpu-dependent sysctl structs
1.18	06-May-1994	mycroft	Correct comment.
1.17	05-May-1994	cgd	lots of changes: prototype migration, move lots of variables, definitions, and structure elements around. kill some unnecessary type and macro definitions. standardize clock handling. More changes than you'd want.
1.16	04-May-1994	cgd	Rename a lot of process flags.
1.15	07-Apr-1994	mycroft	Implement dynamic IRQ configuration and IRQ sharing. Inline spl*() calls. Reorganize and clean up the relevant code.
1.14	03-Apr-1994	mycroft	This is #included multiple times.
1.13	08-Mar-1994	mycroft	Add `com' softintr.
1.12	17-Dec-1993	mycroft	branches: 1.12.2; From magnum branch: Remove Jolitz's netisr kluge. Make sure cpl == 0 really means base priority. Other minor cleanup.
1.11	11-Sep-1993	jtc	Remove NEED_* defines, as they are now satisfied by libkern.
1.10	02-Sep-1993	cgd	branches: 1.10.2; get rid of bogus #define of resettodr(); now setting time actually works!
1.9	16-Jun-1993	jtc	Move my assembly versions of strlen and ffs into the kernel
1.8	03-Jun-1993	cgd	add prototype for a function to be called from execve to check a_mid type
1.7	02-Jun-1993	cgd	set things up so that the signal trampoline code is on the stack, like it normally is in BSD systems. still has a bit of hair...
1.6	21-May-1993	cgd	add "cpu classes" in addition to cpu names, put cpu names in table, along with mapping to classes. this is for upgraded cpu-id code coming "soon"
1.5	21-May-1993	cgd	move cpu types out of assym.s and into cputypes.h and have that file included by cpu.h and by locore.s
1.4	20-May-1993	cgd	hack on the uname "machine name" stuff for hopefully the last time. now it uses MACHINE, as defined in param.h
1.3	20-May-1993	cgd	clean up for new cpu stuff
1.2	20-May-1993	cgd	make uname stuff in init_main machine independent
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.10.2.9	01-Feb-1994	mycroft	Revert clockframe and profiling stuff.
1.10.2.8	11-Jan-1994	mycroft	SIR_GENERIC --> SIR_AST, to match main branch.
1.10.2.7	29-Oct-1993	mycroft	Set sir bits correctly.
1.10.2.6	18-Oct-1993	mycroft	Add a more `normal' way of dealing with software interrupts.
1.10.2.5	18-Oct-1993	mycroft	Add need_proftick(), and some minor cleanup.
1.10.2.4	14-Oct-1993	mycroft	Remove aston().
1.10.2.3	13-Oct-1993	mycroft	Remove defunct profile_tick(). Put pending AST flag back in netisr.
1.10.2.2	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.10.2.1	14-Sep-1993	mycroft	New i386 code.
1.12.2.2	17-Dec-1993	mycroft	From magnum branch: Remove Jolitz's netisr kluge. Make sure cpl == 0 really means base priority. Other minor cleanup.
1.12.2.1	17-Dec-1993	mycroft	file cpu.h was added on branch magnum on 1993-12-17 00:10:50 +0000
1.39.8.1	22-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.40.2.1	28-Nov-1997	mellon	Pull rev 1.41 up from trunk (bouyer)
1.53.4.1	02-Aug-1999	thorpej	Update from trunk.
1.53.2.1	27-Jun-2000	he	Pull up revision 1.57 (via patch, requested by mjl): Reinitialize clock after hibernation, something which seems to be required on some laptops. Fixes PR#4808.
1.55.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.56.4.1	15-Nov-1999	fvdl	Sync with -current
1.56.2.4	12-Mar-2001	bouyer	Sync with HEAD.
1.56.2.3	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.56.2.2	22-Nov-2000	bouyer	Sync with HEAD.
1.56.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.57.2.2	06-Nov-1999	enami	- Export initrtclock(). - Call it from apm_resume(). This fixes timer problem on some laptop after hybernation (PR#4808). Reviewd by: fvdl@netbsd.org
1.57.2.1	06-Nov-1999	enami	file cpu.h was added on branch comdex-fall-1999 on 1999-11-06 01:14:01 +0000
1.59.2.32	18-May-2002	sommerfeld	Catch up with -current. (commons elimination, changes to MI pci framework)
1.59.2.31	27-Apr-2002	sommerfeld	Add options TRAPLOG: When enabled, each cpu gets a ring buffer; on every trap entry, we add an entry to the trap log using the new locore "TLOG" macro. Somewhat like a flight recorder, this may give some post-mortem clues as to what was going on as we were spiralling in... XXX currently depends on various intel MSR's to record the low order 32 bits of the cycle counter as well as last-branch from and to addresses.
1.59.2.30	24-Feb-2002	sommerfeld	Resynch with mainline.
1.59.2.29	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.59.2.28	29-Dec-2001	sommerfeld	Yet another mergeup. New work: - Frank van der Linden's tlb shootdown fix. Mainline functionality merged: - IrDA - Kernel RCSID's - Transmeta CPU support - ACPI - XMM register access through procfs
1.59.2.27	03-Sep-2001	sommerfeld	Merge with -current once more.
1.59.2.26	02-Jul-2001	fvdl	Since cpuid_level value -1 is used to indicate "not available", it should be signed.
1.59.2.25	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.59.2.24	26-May-2001	sommerfeld	Implement MP-synchronzed microtime(), based on Jason Thorpe's port of Dave Mills's cycle-counter-based microtime. XXX We need to figure out how to deal with processors where the cycle counter speed varies. We don't have a whole lot of alternatives; we could only use this on multiprocessors, which are less likely to throttle the clock to save power. The local apic timer runs at a fixed frequency, but it's count-down-to-zero, not the count-up-forever type of timer which the Mills code expects. This also does not solve the problem of microtime on 486-based multiprocessors..
1.59.2.23	07-May-2001	sommerfeld	Rework cache/tlb parsing, etc., One More Time. Compact tables, rework associativity.
1.59.2.22	07-May-2001	sommerfeld	"Merge" in revised cache probing code (including amd cache size probing). Separate feature probing from output formatting, and call uvm_recolor() at an appropriate moment.
1.59.2.21	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.59.2.20	10-Jan-2001	sommerfeld	Significant rewrite of slave CPU attach and initialization, centering on identifycpu() and spreading outward. Separate feature probes (cpuid invocation), which runs on the cpu being attached, from the display of this information, which runs on the boot cpu; remove some of the probe code from locore as it doesn't need to run that early. We now spin up secondary cpu's during cpu_attach, have them do a cpu feature probe, and then have them spin waiting for a "go" signal. We no longer pay attention to the cpu signature and cpuid features reported by the MP BIOS since those are known to be truncated for at least some bioses. We now do npx initialization (including FDIV bug detection) on all cpus. XXX Change some of the cyrix bug workarounds to fit the new identifycpu() system; this is untested.
1.59.2.19	08-Jan-2001	sommerfeld	Mostly fix cpu identification: - move globals which contain per-cpu state into cpu_info. - fix cache size probing to work off of cpu_info rather than globals - print out itlb/dtlb sizes - add a few more cache identifiers from the latest version of AP-485 cpu timing, cache probing, and processor serial number lookup still happens only on the main cpu; this needs to be restructured; mark what code needs to move into per-cpu probe code. While we're here, eradicate "cpu_info_store" in favor of "cpu_info_primary".
1.59.2.18	07-Jan-2001	sommerfeld	Correct merge botches. Post-merge kernel now boots multiuser.
1.59.2.17	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.59.2.16	04-Jan-2001	thorpej	Make sure that there's at least one "cpu_info" slot, so that we don't trash random memory in the non-MULTIPROCESSOR case.
1.59.2.15	04-Jan-2001	thorpej	Always statically allocated the primary CPU's cpu_info, and make the cpu_info list available in the non-MULTIPROCESSOR case, as well.
1.59.2.14	18-Nov-2000	sommerfeld	Add some minimal support for iterating over cpus
1.59.2.13	23-Sep-2000	sommerfeld	Preliminary support for cloning MTRR values between CPU's at boot time. XXX no API to set MTRR values yet.
1.59.2.12	25-Aug-2000	sommerfeld	Get this to at least build again in the !MULTIPROCESSOR case..
1.59.2.11	25-Aug-2000	sommerfeld	The need_resched() function now takes a (cpu_info *) parameter. Get asts out of <machine/intr.h> Untangle astpending and want_resched variables somewhat.
1.59.2.10	18-Aug-2000	sommerfeld	misc changes: - CPUF_PAUSE (indicating CPU is paused in debugger). - ci_flags is now a u_int32_t. - npx interface changes.
1.59.2.9	12-Aug-2000	sommerfeld	Get ready for context switch changes: - initialize idle pcbs earlier, before the first tsleep (instead of as part of booting the secondary cpu's), since we'll actually use them when idling in the near future. - clean up cpu_hatch() - fixup curproc's cpu pointer when switching to "real" cpu_info (XXX). - turn on local apic clock in cpu_hatch() - load the right gdt descriptor in cpu_hatch() - return from cpu_hatch() instead of spinning there; current mptramp code will spin; future code will jump to idle loop.
1.59.2.8	26-Jun-2000	sommerfeld	Start making npx.c MP-safe; Interface to npx now specifies the process and/or cpu to act on. Implement an IPI for MP lazy FP save. Tested and working when only one CPU is running; untested in the MP case since we can't run user processes on multiple CPU's yet. Along for the ride: - Change i386_send_ipi to take a "struct cpu_info *" rather than a cpu number. - Turn off the "give me the brain" test IPI's.
1.59.2.7	25-Jun-2000	sommerfeld	Add CPUF_PRIMARY, CPU_IS_PRIMARY()
1.59.2.6	25-Jun-2000	sommerfeld	Merge up to just-post-1.5 -current
1.59.2.5	22-Apr-2000	sommerfeld	Merge up to -current.
1.59.2.4	17-Apr-2000	sommerfeld	Catch up sommerfeld_i386mp_1 branch with -current.
1.59.2.3	24-Feb-2000	sommerfeld	Patch from Matt Debergalis to un-break bootblock build
1.59.2.2	21-Feb-2000	sommerfeld	MP: attempt to deal with the case where the BSP is at apic id 1 instead of apic id 0.
1.59.2.1	20-Feb-2000	sommerfeld	MP: include cpu_info structure, which picks up what used to be in several global variables. XXX cpu_info is bigger than it should be..
1.63.2.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.64.2.1	25-Apr-2001	he	Pull up revision 1.66 (requested by jdolecek): Add ``const'' to cpu tables. Properly identify and distinguish newer Intel CPUs using Brand ID information.
1.68.2.18	07-Jan-2003	thorpej	Sync with HEAD.
1.68.2.17	03-Jan-2003	thorpej	Merge switch_exit() and switch_lwp_exit().
1.68.2.16	19-Dec-2002	thorpej	Sync with HEAD.
1.68.2.15	11-Dec-2002	thorpej	Sync with HEAD.
1.68.2.14	25-Nov-2002	nathanw	Make ASTs per-proc rather than per-cpu; per-proc makes more sense when a process that needs a trap can be running on multiple CPUs.
1.68.2.13	11-Nov-2002	nathanw	Catch up to -current
1.68.2.12	18-Oct-2002	nathanw	cpu_info->fp_curlwp is a struct lwp, not a struct proc. npxsave_proc() -> npxsave_lwp().
1.68.2.11	18-Oct-2002	nathanw	Define curlwp, not curproc.
1.68.2.10	18-Oct-2002	nathanw	Catch up to -current.
1.68.2.9	01-Aug-2002	nathanw	Catch up to -current.
1.68.2.8	20-Jun-2002	nathanw	Catch up to -current.
1.68.2.7	28-Feb-2002	nathanw	Catch up to -current.
1.68.2.6	23-Feb-2002	gmcgarry	Protect cpu_proc_fork() with _KERNEL
1.68.2.5	08-Jan-2002	nathanw	Catch up to -current.
1.68.2.4	08-Dec-2001	thorpej	Add a cpu_proc_fork(), called from uvm_proc_fork(), which takes care of machine-dependent handling a fork() time (this is different from forking the actual context in an LWP world). #define it away on platforms which do not need it. Problem noted by Gregory McGarry.
1.68.2.3	24-Aug-2001	nathanw	Catch up with -current.
1.68.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.68.2.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.72.2.6	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.72.2.5	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.72.2.4	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.72.2.3	11-Feb-2002	jdolecek	Sync w/ -current.
1.72.2.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.72.2.1	03-Aug-2001	lukem	update to -current
1.80.4.1	07-Sep-2003	tron	Apply patch (requested by drochner in ticket 1344): Make output of CPU features look nicer on new Pentium IV CPUs.
1.80.2.1	16-Jul-2002	gehenna	catch up with -current.
1.102.2.6	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.102.2.5	04-Mar-2005	skrll	Sync with HEAD. Hi Perry!
1.102.2.4	24-Sep-2004	skrll	Sync with HEAD.
1.102.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.102.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.102.2.1	03-Aug-2004	skrll	Sync with HEAD
1.116.6.1	19-Mar-2005	yamt	sync with head. xen and whitespace. xen part is not finished.
1.116.4.1	29-Apr-2005	kent	sync with -current
1.117.4.11	17-Mar-2008	yamt	sync with head.
1.117.4.10	27-Feb-2008	yamt	sync with head.
1.117.4.9	11-Feb-2008	yamt	sync with head.
1.117.4.8	21-Jan-2008	yamt	sync with head
1.117.4.7	07-Dec-2007	yamt	sync with head
1.117.4.6	15-Nov-2007	yamt	sync with head.
1.117.4.5	27-Oct-2007	yamt	sync with head.
1.117.4.4	03-Sep-2007	yamt	sync with head.
1.117.4.3	26-Feb-2007	yamt	sync with head.
1.117.4.2	30-Dec-2006	yamt	sync with head.
1.117.4.1	21-Jun-2006	yamt	sync with head.
1.120.2.2	18-Feb-2006	yamt	sync with head.
1.120.2.1	15-Jan-2006	yamt	sync with head.
1.121.2.1	09-Sep-2006	rpaulo	sync with head
1.122.2.3	22-Apr-2006	simonb	Sync with head.
1.122.2.2	28-Feb-2006	kardel	initrtclock() now gets the frequency as parameter
1.122.2.1	04-Feb-2006	simonb	Remove port-specific microtime() function support.
1.123.8.1	19-Jun-2006	chap	Sync with head.
1.123.2.3	03-Sep-2006	yamt	sync with head.
1.123.2.2	11-Aug-2006	yamt	sync with head
1.123.2.1	26-Jun-2006	yamt	sync with head.
1.125.2.4	07-Sep-2006	tron	Pull up following revision(s) (requested by xtraeme in ticket #114): sys/arch/i386/include/cpu.h: revision 1.129 sys/arch/i386/i386/identcpu.c: revision 1.40 Conditionalize *_get_bus_clock() functions inside of #ifdef ENHANCED_SPEEDSTEP to fix build when ENHANCED_SPEEDSTEP is not defined. Reported by Geoff Wing.
1.125.2.3	06-Sep-2006	riz	Pull up following revision(s) (requested by xtraeme in ticket #111): sys/arch/x86/include/specialreg.h: revision 1.11 sys/arch/i386/i386/identcpu.c: revision 1.39 sys/arch/i386/include/cpu.h: revision 1.128 sys/arch/i386/i386/est.c: revision 1.26 Update the enhanced speedstep driver and sync the code with OpenBSD: est.c: * Use a quintuplet (vendor, MHz_hi, mV_hi, MHz_lo, mV_lo } to match CPUs more correctly than parsing the brand string. * Add support for a bunch of models. * Create a fake table on the fly if the CPU is unknown (there's no table for it) with the current/highest/lowest frequency. specialreg.h: * Add some MSRs needed to get the bus clock value. identcpu.c: * Add functions specific to Pentium III, Pentium M and Pentium 4 to get the bus clock value. Note that the new fake table code from Simon Burge is not included on this commit. Ok'ed by simonb and dogcow.
1.125.2.2	27-Aug-2006	tron	Pull up following revision(s) (requested by xtraeme in ticket #57): sys/arch/i386/i386/identcpu.c: revision 1.37 sys/arch/x86/include/powernow.h: revision 1.4 sys/arch/i386/include/cpu.h: revision 1.127 - Move k7_powernow_* prototypes from i386/include/cpu.h to x86/include/powernow.h - Protect k[78]_powernow_init() functions with #ifdef POWERNOW_K[78] to make it build without these options. This fixes the problem reported by hubertf.
1.125.2.1	11-Aug-2006	riz	Pull up following revision(s) (requested by xtraeme in ticket #10): sys/arch/i386/include/cpu.h: revision 1.126 sys/lkm/arch/i386/powernow/lkminit_powernow.c: revision 1.5 - Add k7_powernow_destroy() prototype into cpu.h. - Call k7_powernow_destroy() unloading the module.
1.129.6.2	12-Sep-2007	msaitoh	Pull up following patches (requested by xtraeme in ticket #809) share/man/man4/options.4 patch sys/arch/i386/conf/files.i386 patch sys/arch/i386/i386/est.c delete sys/arch/i386/i386/identcpu.c patch sys/arch/i386/include/cpu.h patch sys/arch/x86/conf/files.x86 patch sys/arch/x86/include/cpuvar.h patch sys/arch/x86/x86/est.c new file sys/arch/x86/x86/intel_busclock.c new file sys/arch/amd64/amd64/identcpu.c patch sys/arch/amd64/conf/GENERIC patch Add support for the VIA C7-M and Eden processors in the Enhanced Speedstep driver. amd64: The Enhanced Speedstep driver is now able to work on EM64T CPUs running in 64bit mode.
1.129.6.1	20-Apr-2007	bouyer	branches: 1.129.6.1.2; Pull up following revision(s) (requested by mlelstv in ticket #575): sys/arch/i386/i386/est.c sync with 1.37 sys/arch/i386/i386/ipifuncs.c sync with 1.16 sys/arch/x86/include/cpu_msr.h sync with 1.4 sys/arch/x86/include/intrdefs.h sync with 1.8 sys/arch/x86/include/powernow.h sync with 1.9 sys/arch/x86/x86/powernow_k8.c sync with 1.20 sys/arch/x86/x86/msr_ipifuncs.c sync with 1.8 sys/arch/amd64/amd64/ipifuncs.c sync with 1.9 sys/arch/i386/i386/identcpu.c patch sys/arch/i386/i386/machdep.c patch sys/arch/i386/include/cpu.h patch sys/arch/x86/conf/files.x86 patch sys/arch/x86/x86/x86_machdep.c patch sys/arch/amd64/amd64/machdep.c patch Add MSR write IPI handler for x86. Use it and the RUN_ONCE framework to make est and powernow drivers work properly with SMP.
1.129.6.1.2.1	23-Sep-2007	wrstuden	Sync with somewhat-recent netbsd-4.
1.129.4.1	10-Dec-2006	yamt	sync with head.
1.129.2.5	27-Jan-2007	ad	If running on a PPro or later, at boot patch in versions of spllower() and similar that use cmpxchg8b instead of cli/sti. Cuts the clock cycles for splx() by a factor of ~6 on the P4, and ~3 on the PIII when bracketed by serializing instructions (and hopefully more when not).
1.129.2.4	12-Jan-2007	ad	Sync with head.
1.129.2.3	11-Jan-2007	ad	Checkpoint work in progress.
1.129.2.2	17-Nov-2006	ad	Checkpoint work in progress.
1.129.2.1	20-Oct-2006	ad	- Make ASTs per-LWP. - The signal stack and signal mask will be per-LWP shortly. need_resched(), need_proftick(), signotify(): - Prefix with cpu_ - Make per-LWP. - Send an IPI if the LWP is on another CPU.
1.133.2.8	07-May-2007	yamt	sync with head.
1.133.2.7	24-Mar-2007	yamt	sync with head.
1.133.2.6	17-Mar-2007	rmind	Backport lock.h split into the simplelock.h and other #include changes from HEAD. This fixes the problems with circular includes.
1.133.2.5	12-Mar-2007	rmind	Sync with HEAD.
1.133.2.4	03-Mar-2007	yamt	curpcb: wrap the definition with parenthesises.
1.133.2.3	23-Feb-2007	yamt	- introduce sys/cpu.h which has cpu_idle and cpu_need_resched. - use it where appropriate. - while i'm here, remove several unnecessary #include.
1.133.2.2	17-Feb-2007	yamt	- separate context switching and thread scheduling. - introduce idle lwp. - change some related MD/MI interfaces and implement i386 version.
1.133.2.1	17-Feb-2007	yamt	file cpu.h was added on branch yamt-idlelwp on 2007-02-17 10:30:49 +0000
1.135.2.14	03-Dec-2007	ad	Sync with HEAD.
1.135.2.13	23-Oct-2007	ad	Sync with head.
1.135.2.12	10-Oct-2007	ad	Share cpu_intr_p() between amd64/i386.
1.135.2.11	09-Oct-2007	ad	Sync with head.
1.135.2.10	21-Aug-2007	ad	- Add ci_cpumask and use in preference to (1 << ci_cpuid). Some processors are bad at shifting by non-constant amounts. - pmap_tlb_shootdown() must be called with preemption disabled. - Defer waiting for tlb shootdown to complete after kenter_pa()/kremove(). Spin in pmap_update() and in pmap_detactivate(), in case the LWP has been preempted - When preparing to send a multicast shootdown IPI, avoid scanning the list of CPUs if the pmap's bitmask indicates that it's only active locally.
1.135.2.9	29-Jul-2007	ad	- When zeroing/copying pages, use SSE2 movtni to avoid polluting the cache. - By default, align assembly routines on 32-byte starting boundaries. - There are now 8 interrupt priority levels, half of which are softints. Update intrdefs.h to match. - Always clear/set spinlock words - removes lots of ifdefs. - Remove the horrible ci_self150 hack that I introduced. - Overhaul how TLB shootdown is performed. Inspired by a similar change in OpenBSD but implemented quite differently. This should be a lot faster but I have not benchmarked it yet.
1.135.2.8	17-Jun-2007	ad	- Increase the number of thread priorities from 128 to 256. How the space is set up is to be revisited. - Implement soft interrupts as kernel threads. A generic implementation is provided, with hooks for fast-path MD code that can run the interrupt threads over the top of other threads executing in the kernel. - Split vnode::v_flag into three fields, depending on how the flag is locked (by the interlock, by the vnode lock, by the file system). - Miscellaneous locking fixes and improvements.
1.135.2.7	09-Jun-2007	ad	Sync with head.
1.135.2.6	27-May-2007	ad	Sync with head.
1.135.2.5	29-Apr-2007	ad	Replace another simplelock.
1.135.2.4	28-Apr-2007	ad	- Make the PV cache per-CPU and eliminate the global pvalloc_lock. Cuts contention on the freelist locks by a factor of ~50 on a dual core box. - Mirror a patch for amd64 posted by oster@, and make the TLB shootdown freelist per-CPU. Increase the number of TLB shootdown jobs from 16 to 32 per CPU.
1.135.2.3	10-Apr-2007	ad	Sync with head.
1.135.2.2	05-Apr-2007	ad	Fix the pmap locking.
1.135.2.1	13-Mar-2007	ad	Sync with head.
1.136.4.1	29-Mar-2007	reinoud	Pullup to -current
1.136.2.1	11-Jul-2007	mjf	Sync with head.
1.140.2.3	03-Oct-2007	garbled	Sync with HEAD
1.140.2.2	26-Jun-2007	garbled	Sync with HEAD.
1.140.2.1	22-May-2007	matt	Update to HEAD.
1.143.10.3	23-Mar-2008	matt	sync with HEAD
1.143.10.2	09-Jan-2008	matt	sync with HEAD
1.143.10.1	06-Nov-2007	matt	sync with HEAD
1.143.8.6	09-Dec-2007	jmcneill	Sync with HEAD.
1.143.8.5	29-Oct-2007	joerg	Sync with HEAD.
1.143.8.4	28-Oct-2007	joerg	Sync with HEAD.
1.143.8.3	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.143.8.2	02-Oct-2007	joerg	Sync with HEAD.
1.143.8.1	03-Sep-2007	jmcneill	Sync with HEAD.
1.143.4.1	03-Sep-2007	skrll	Sync with HEAD.
1.144.2.3	07-Oct-2007	yamt	g/c nkpde.
1.144.2.2	06-Oct-2007	yamt	sync with head.
1.144.2.1	23-Sep-2007	yamt	- sync with amd64. - pmap_alloc_level: remove fast call special case. - tweak some printf to work for both of LP64 and LP32. - remove machdep.nkpde sysctl. - remove automatic adjustment of nkpde for now. will revisit later.
1.146.2.2	13-Nov-2007	bouyer	Sync with HEAD
1.146.2.1	25-Oct-2007	bouyer	Sync with HEAD.
1.150.4.4	26-Dec-2007	ad	Sync with head.
1.150.4.3	26-Dec-2007	ad	Back out the dumppcb changes for the time being.
1.150.4.2	12-Dec-2007	ad	- Dump basic trap and register info before entering the debugger, in case ddb screws the machine up further. - Fill dumppcb asap so that gdb has less chance to get confused.
1.150.4.1	08-Dec-2007	ad	Sync with head.
1.150.2.3	18-Feb-2008	mjf	Sync with HEAD.
1.150.2.2	27-Dec-2007	mjf	Sync with HEAD.
1.150.2.1	08-Dec-2007	mjf	Sync with HEAD.
1.151.4.4	19-Jan-2008	bouyer	Sync with HEAD
1.151.4.3	08-Jan-2008	bouyer	Sync with HEAD
1.151.4.2	06-Jan-2008	bouyer	Merge needed changes to genassym.cf and locore.S for xeni386 back to arch/i386. Switch xeni386 to use the arch/i386 cpu.h.
1.151.4.1	02-Jan-2008	bouyer	Sync with HEAD
1.161.6.3	17-Jan-2009	mjf	Sync with HEAD.
1.161.6.2	02-Jun-2008	mjf	Sync with HEAD.
1.161.6.1	03-Apr-2008	mjf	Sync with HEAD.
1.161.2.1	24-Mar-2008	keiichi	sync with head.
1.165.2.2	04-Jun-2008	yamt	sync with head
1.165.2.1	18-May-2008	yamt	sync with head.
1.167.2.2	04-May-2009	yamt	sync with head.
1.167.2.1	16-May-2008	yamt	sync with head.
1.169.2.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.175.6.1	19-Jan-2009	skrll	Sync with HEAD.
1.176.12.1	17-Aug-2011	cherry	Pullup relevant changes from -current
1.176.2.1	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.177.6.1	18-Feb-2012	mrg	merge to -current.
1.177.2.1	17-Apr-2012	yamt	sync with head
1.178.6.1	03-Dec-2017	jdolecek	update from HEAD
1.179.4.2	08-Apr-2020	martin	Merge changes from current as of 20200406
1.179.4.1	10-Jun-2019	christos	Sync with HEAD
1.179.2.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.9	10-May-2008	ad	Merge cpu_counter.h.
1.8	28-Apr-2008	martin	branches: 1.8.2; Remove clause 3 and 4 from TNF licenses
1.7	14-Nov-2007	ad	branches: 1.7.14; 1.7.16; 1.7.18; - Remove I486_CPU, I586_CPU, I686_CPU options. They buy us nothing and clutter the code significantly. - Remove pccons.
1.6	17-Oct-2007	garbled	branches: 1.6.2; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.5	07-Jul-2007	tsutsui	branches: 1.5.8; 1.5.10; 1.5.14; Move x86 common cpu_counter functions into <x86/cpu_counter.h>.
1.4	16-Feb-2006	perry	branches: 1.4.24; 1.4.26; 1.4.32; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.3	24-Dec-2005	perry	branches: 1.3.2; 1.3.4; 1.3.6; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.2	25-Jun-2005	fair	branches: 1.2.2; document more precisely the Cyrix TSC lossage: if you execute "hlt" when in powersave mode, the TSC stops counting!
1.1	05-Feb-2003	nakayama	branches: 1.1.2; 1.1.16; Share i386/tsc_microtime.c with alpha and sparc64 as kern_microtime.c. (approved by martin)
1.1.16.1	12-Jul-2005	tron	Pull up revision 1.2 (requested by fair in ticket #554): document more precisely the Cyrix TSC lossage: if you execute "hlt" when in powersave mode, the TSC stops counting!
1.1.2.1	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.2.2.2	15-Nov-2007	yamt	sync with head.
1.2.2.1	03-Sep-2007	yamt	sync with head.
1.3.6.1	22-Apr-2006	simonb	Sync with head.
1.3.4.1	09-Sep-2006	rpaulo	sync with head
1.3.2.1	18-Feb-2006	yamt	sync with head.
1.4.32.1	03-Oct-2007	garbled	Sync with HEAD
1.4.26.1	11-Jul-2007	mjf	Sync with head.
1.4.24.2	03-Dec-2007	ad	Sync with HEAD.
1.4.24.1	15-Jul-2007	ad	Sync with head.
1.5.14.1	18-Nov-2007	bouyer	Sync with HEAD
1.5.10.2	09-Jan-2008	matt	sync with HEAD
1.5.10.1	06-Nov-2007	matt	sync with HEAD
1.5.8.1	21-Nov-2007	joerg	Sync with HEAD.
1.6.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.7.18.1	16-May-2008	yamt	sync with head.
1.7.16.1	18-May-2008	yamt	sync with head.
1.7.14.1	02-Jun-2008	mjf	Sync with HEAD.
1.8.2.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.1	27-Feb-2016	tls	branches: 1.1.2; 1.1.18; Add cpu_rng, a framework for simple on-CPU random number generators.
1.1.18.2	03-Dec-2017	jdolecek	update from HEAD
1.1.18.1	27-Feb-2016	jdolecek	file cpu_rng.h was added on branch tls-maxphys on 2017-12-03 11:36:17 +0000
1.1.2.2	19-Mar-2016	skrll	Sync with HEAD
1.1.2.1	27-Feb-2016	skrll	file cpu_rng.h was added on branch nick-nhusb on 2016-03-19 11:30:00 +0000
1.40	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.39	26-Sep-2007	ad	x86 changes for pcc and LKMs. - Replace most inline assembly with proper functions. As a side effect this reduces the size of amd64 GENERIC by about 120kB, and i386 by a smaller amount. Nearly all of the inlines did something slow, or something that does not need to be fast. - Make curcpu() and curlwp functions proper, unless __GNUC__ && _KERNEL. In that case make them inlines. Makes curlwp LKM and preemption safe. - Make bus_space and bus_dma more LKM friendly. - Share a few more files between the ports. - Other minor changes.
1.38	04-Mar-2007	christos	branches: 1.38.2; 1.38.10; 1.38.18; 1.38.20; 1.38.22; Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.37	09-Feb-2007	ad	branches: 1.37.2; Merge newlock2 to head.
1.36	14-Jan-2007	ad	.. but only if _KERNEL is defined.
1.35	14-Jan-2007	ad	On second thought, implement x86_pause() as a regular function. The small delay from the call is useful for spinlock backoff.
1.34	01-Jan-2007	ad	Report on and where possible, try to work around some of the known errata for Athlon 64 and Opteron processors. Tested briefly by cube@ and elad@.
1.33	26-Aug-2006	ad	branches: 1.33.2; Add x86_sfence(), x86_mfence().
1.32	19-Aug-2006	dsl	Change the 'asm' pattern for lidt so that gcc4 knows it references memory and doesn't optimise out the assignments to the descriptor memory area. i386 kernels now build with -Os (and -O3) applied to machdep.c
1.31	28-Dec-2005	perry	branches: 1.31.4; 1.31.8; 1.31.18; inline -> __inline
1.30	26-Dec-2005	perry	u_intN_t -> uintN_t
1.29	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.28	14-Jan-2004	yamt	branches: 1.28.14; 1.28.16; issue memory read barrier for BUS_DMASYNC_POSTREAD operation. PR/21665 from Stephan Uphoff.
1.27	27-Oct-2003	junyoung	Nuke __P().
1.26	26-Jun-2003	drochner	branches: 1.26.2; make cpu_feature unsigned, just like ci_feature_flags
1.25	08-May-2003	fvdl	Move x86_pause() out of ifdef _KERNEL.
1.24	08-May-2003	fvdl	Add x86_pause() inline function, containing the "pause" instruction for i386, and nothing for amd64. Sprinkle it in various spinloops, as recommended by Intel.
1.23	04-Mar-2003	fvdl	Define read_psl and write_psl as aliases for read_eflags and write_eflags.
1.22	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.21	03-Jul-2002	yamt	add KSTACK_CHECK_DR0.
1.20	31-Jul-2001	thorpej	branches: 1.20.2; 1.20.14; Set up function pointers for copyin/copyout in preparation for adding optimized versions for various CPU classes/models. Split the 386 version of copyout into a separate routine, and add a 486 version that doesn't have the class/page-writeability check.
1.19	28-Mar-2000	thorpej	branches: 1.19.8; 1.19.10; The assembler understands rdmsr, wrmsr, rdtsc, and rdpmc, so there's no need to use .byte.
1.18	28-Mar-2000	thorpej	Add functions to read the TSC and Performance Counters.
1.17	24-Mar-2000	thorpej	Add some MSR-related calls. From FreeBSD.
1.16	15-Aug-1998	mycroft	branches: 1.16.12; Assign my copyrights to TNF.
1.15	15-Aug-1998	mycroft	Make copyright notices with my name consistent.
1.14	26-Feb-1998	perry	define kernel only inline static functions only #ifdef _KERNEL
1.13	25-Feb-1998	perry	add /* LINTLIBRARY */ to supress lots of "unused function" warnings from lint.
1.12	05-Feb-1998	drochner	provide pushfl/popfl to C code
1.11	09-Sep-1997	mycroft	Remove some vestiges of BDB.
1.10	05-Sep-1997	chuck	add invlpg, and rcr4/lcr4
1.9	05-Jul-1997	thorpej	branches: 1.9.2; Add a breakpoint() inline, used by DDB and KGDB.
1.8	27-Oct-1994	cgd	new RCS ID format.
1.7	09-Oct-1994	mycroft	#include cpufunc.h from pmap.h.
1.6	09-Oct-1994	mycroft	Eliminate register restrictions where possible.
1.5	28-Jan-1994	jtc	branches: 1.5.2; 1.5.4; Fix spelling error in Copyright notice
1.4	03-Jan-1994	mycroft	Inline all the [lr]cr[023]() `functions' in cpufunc.h.
1.3	20-Dec-1993	mycroft	Remove some things which should not be here.
1.2	02-Aug-1993	mycroft	branches: 1.2.2; Add RCS identifiers, remove some completely useless RCS logs and patchkit headers, and a few other insignificant changes.
1.1	06-Jun-1993	cgd	incorporate Bruce Evans' interrupt changes, as hacked by Rod grimes, for patchkit patch 158. it appears to work well.
1.2.2.3	11-Oct-1993	mycroft	Move {disable,enable}_intr() to psl.h, and inline all of {in,out}{,s}[bwl] in pio.h.
1.2.2.2	10-Oct-1993	mycroft	Change an `unsigned int' to `u_int'.
1.2.2.1	10-Oct-1993	mycroft	Add inline versions of lidt(), lldt(), ltr(), tlbflush(), lcr0(), rcro(), rcr2(), lcr3(), and rcr3().
1.5.4.1	11-Oct-1994	mycroft	Update from trunk.
1.5.2.2	01-Feb-1994	mycroft	Remove *_intr().
1.5.2.1	28-Jan-1994	mycroft	file cpufunc.h was added on branch magnum on 1994-02-01 18:39:24 +0000
1.9.2.2	16-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.9.2.1	06-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.16.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.19.10.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.19.10.2	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.19.10.1	03-Aug-2001	lukem	update to -current
1.19.8.3	18-Oct-2002	nathanw	Catch up to -current.
1.19.8.2	01-Aug-2002	nathanw	Catch up to -current.
1.19.8.1	24-Aug-2001	nathanw	Catch up with -current.
1.20.14.1	16-Jul-2002	gehenna	catch up with -current.
1.20.2.3	19-Aug-2002	sommerfeld	merged with -current as of today. rearranged gdt init to deal with KVM86 changes; however, KVM86 has no chance of working yet; good thing it's not part of GENERIC
1.20.2.2	22-Sep-2001	sommerfeld	Move tlbflushg to cpufunc.h
1.20.2.1	31-Jul-2001	sommerfeld	file cpufunc.h was added on branch sommerfeld_i386mp_1 on 2001-09-22 23:07:34 +0000
1.26.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.26.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.26.2.1	03-Aug-2004	skrll	Sync with HEAD
1.28.16.5	27-Oct-2007	yamt	sync with head.
1.28.16.4	03-Sep-2007	yamt	sync with head.
1.28.16.3	26-Feb-2007	yamt	sync with head.
1.28.16.2	30-Dec-2006	yamt	sync with head.
1.28.16.1	21-Jun-2006	yamt	sync with head.
1.28.14.1	28-Sep-2008	jdc	Pull up revisions: sys/arch/amd64/include/cpufunc.h patch sys/arch/i386/include/cpufunc.h patch sys/arch/x86/x86/bus_dma.c 1.45 via patch requested by bouyer in ticket 1945.
1.31.18.1	27-Aug-2006	riz	Pull up following revision(s) (requested by dsl in ticket #67): sys/arch/i386/include/cpufunc.h: revision 1.32 Change the 'asm' pattern for lidt so that gcc4 knows it references memory and doesn't optimise out the assignments to the descriptor memory area. i386 kernels now build with -Os (and -O3) applied to machdep.c
1.31.8.1	03-Sep-2006	yamt	sync with head.
1.31.4.1	09-Sep-2006	rpaulo	sync with head
1.33.2.4	01-Feb-2007	ad	Sync with head.
1.33.2.3	27-Jan-2007	ad	If running on a PPro or later, at boot patch in versions of spllower() and similar that use cmpxchg8b instead of cli/sti. Cuts the clock cycles for splx() by a factor of ~6 on the P4, and ~3 on the PIII when bracketed by serializing instructions (and hopefully more when not).
1.33.2.2	25-Jan-2007	ad	Add x86_flush(): flush pipelines by doing a far return.
1.33.2.1	12-Jan-2007	ad	Sync with head.
1.37.2.1	12-Mar-2007	rmind	Sync with HEAD.
1.38.22.1	06-Oct-2007	yamt	sync with head.
1.38.20.1	06-Nov-2007	matt	sync with HEAD
1.38.18.1	02-Oct-2007	joerg	Sync with HEAD.
1.38.10.1	03-Oct-2007	garbled	Sync with HEAD
1.38.2.1	09-Oct-2007	ad	Sync with head.
1.15	01-Jan-2007	ad	Report on and where possible, try to work around some of the known errata for Athlon 64 and Opteron processors. Tested briefly by cube@ and elad@.
1.14	16-Jul-2005	grant	branches: 1.14.26; fix typo in comment
1.13	17-Nov-2001	christos	branches: 1.13.2; 1.13.18; 1.13.34; Add transmeta crusoe cpu support from toshi.
1.12	29-Sep-2000	fvdl	branches: 1.12.4; 1.12.6; Bump CPU_MAXFAMILY to 7.
1.11	15-Oct-1998	bad	branches: 1.11.12; Add CPUVENDOR_IDT and CPU_C6.
1.10	18-Oct-1997	mikel	branches: 1.10.2; cosmetic cleanup (newlines, comments)
1.9	17-Oct-1997	bouyer	Add detection of cyrix/ibm 6x86 cpu: these CPU pass the 486 test, but fail the cyrix486 one. Add a test which succeed for all cyrix CPUs (from FreeBSD).
1.8	03-Dec-1996	fvdl	Added types/definitions for changed CPU recognition code.
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	24-May-1994	deraadt	cyrix CPU_486DLC
1.5	28-Jan-1994	jtc	branches: 1.5.2; Fix spelling error in Copyright notice
1.4	19-Dec-1993	mycroft	Forgot to decrement these.
1.3	19-Dec-1993	mycroft	Clean up trap handling and macroize interrupt entry and exit code. From magnum branch.
1.2	21-May-1993	cgd	branches: 1.2.4; add "cpu classes" in addition to cpu names, put cpu names in table, along with mapping to classes. this is for upgraded cpu-id code coming "soon"
1.1	21-May-1993	cgd	move cpu types out of assym.s and into cputypes.h and have that file included by cpu.h and by locore.s
1.2.4.1	08-Nov-1993	mycroft	Garbage collect some constants, and the entire 286 nonsense. (Like we even get to start() on a 286? NOT!)
1.5.2.2	28-Jan-1994	jtc	Fix spelling error in Copyright notice
1.5.2.1	28-Jan-1994	jtc	file cputypes.h was added on branch magnum on 1994-01-28 23:44:12 +0000
1.10.2.1	16-Oct-1998	cgd	pull up rev 1.11 from trunk (bad)
1.11.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.12.6.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.12.4.1	08-Jan-2002	nathanw	Catch up to -current.
1.13.34.2	26-Feb-2007	yamt	sync with head.
1.13.34.1	21-Jun-2006	yamt	sync with head.
1.13.18.1	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.13.2.2	17-Nov-2001	christos	Add transmeta crusoe cpu support from toshi.
1.13.2.1	17-Nov-2001	christos	file cputypes.h was added on branch sommerfeld_i386mp_1 on 2001-11-17 08:21:44 +0000
1.14.26.1	12-Jan-2007	ad	Sync with head.
1.5	01-Mar-2003	fvdl	Redirect to x86/cpuvar.h
1.4	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.3	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file cpuvar.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.6	10-Jan-2001	sommerfeld	Significant rewrite of slave CPU attach and initialization, centering on identifycpu() and spreading outward. Separate feature probes (cpuid invocation), which runs on the cpu being attached, from the display of this information, which runs on the boot cpu; remove some of the probe code from locore as it doesn't need to run that early. We now spin up secondary cpu's during cpu_attach, have them do a cpu feature probe, and then have them spin waiting for a "go" signal. We no longer pay attention to the cpu signature and cpuid features reported by the MP BIOS since those are known to be truncated for at least some bioses. We now do npx initialization (including FDIV bug detection) on all cpus. XXX Change some of the cyrix bug workarounds to fit the new identifycpu() system; this is untested.
1.1.2.5	23-Sep-2000	sommerfeld	Preliminary support for cloning MTRR values between CPU's at boot time. XXX no API to set MTRR values yet.
1.1.2.4	18-Aug-2000	sommerfeld	Add cleanup hook to return the NVRAM reset byte to its original value.
1.1.2.3	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.2	21-Feb-2000	sommerfeld	MP: attempt to deal with the case where the BSP is at apic id 1 instead of apic id 0.
1.1.2.1	20-Feb-2000	sommerfeld	(some) per-CPU state.
1.2.2.3	07-Jan-2003	thorpej	Sync with HEAD.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file cpuvar.h was added on branch nathanw_sa on 2002-10-18 02:37:54 +0000
1.6	26-Apr-2011	joerg	Remove Darwin, MACH and Mach-O support.
1.5	28-Apr-2008	martin	branches: 1.5.14; 1.5.22; 1.5.28; Remove clause 3 and 4 from TNF licenses
1.4	25-Jun-2005	christos	branches: 1.4.82; 1.4.84; 1.4.86; move stack somewhere where it is mappable, and add defines for i386 commpage.
1.3	04-Jul-2004	christos	Add USRSTACK definitions
1.2	05-Dec-2003	christos	Add a dummy darwin_slock
1.1	12-Dec-2002	christos	branches: 1.1.6; 1.1.8; This new file is wanted by compat_darwin.
1.1.8.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.1.8.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.8.2	18-Sep-2004	skrll	Sync with HEAD.
1.1.8.1	03-Aug-2004	skrll	Sync with HEAD
1.1.6.2	19-Dec-2002	thorpej	Sync with HEAD.
1.1.6.1	12-Dec-2002	thorpej	file darwin_machdep.h was added on branch nathanw_sa on 2002-12-19 00:33:51 +0000
1.4.86.1	16-May-2008	yamt	sync with head.
1.4.84.1	18-May-2008	yamt	sync with head.
1.4.82.1	02-Jun-2008	mjf	Sync with HEAD.
1.5.28.1	06-Jun-2011	jruoho	Sync with HEAD.
1.5.22.1	31-May-2011	rmind	sync with head
1.5.14.1	02-May-2011	jym	Sync with head.
1.31	06-Nov-2017	christos	Cleanup and clarify the ELFSIZE mess: We now have 2 variables automatically set in elf_machdep.h: ARCH_ELFSIZE: the size for userland binaries KERN_ELFSIZE: the size for the kernel binaries DB_ELFSIZE has been deleted and KERN_ELFSIZE should have always the same values DB_ELFSIZE used to have. In sys/exec_elf.h, if ELFSIZE is not set, it is set to KERN_ELFSIZE for the kernel and ARCH_ELFSIZE for userland. These defaults should eliminate the need for most manual ELFSIZE setting.
1.30	26-May-2011	joerg	branches: 1.30.14; Introduce DDB_EXPR_FMT and replace the logic around DB_EXPR_T_IS_QUAD.
1.29	11-Apr-2011	mrg	obsolete DB_AOUT_SYMBOLS. however, we need to leave most of the code in db_sym.[ch] as it is used by the elf version of crash(8). i will be cleaning up the db_sym.c code in a follow up commit to avoid having dead code compiled.
1.28	10-Apr-2011	christos	Merge db_trace for x86. From: Vladimir Kirillov proger at wilab dot org dot ua
1.27	07-Mar-2009	ad	branches: 1.27.4; 1.27.6; Make ddb compile and work in userspace. Mostly this is comprised of three types of changes: - Add a few new methods to replace stuff like p_find(), CPU_INFO_FOREACH. - Use db_read_bytes() instead of accessing kernel structures directly, and similar changes. - Add ifdef _KERNEL where the above hasn't been done, and an XXX comment.
1.26	21-Feb-2007	thorpej	branches: 1.26.46; 1.26.54; 1.26.60; Replace the Mach-derived boolean_t type with the C99 bool type. A future commit will replace use of TRUE and FALSE with true and false.
1.25	01-Apr-2006	cherry	branches: 1.25.14; closes: PR kern/32359 modifies machine/db_machdep.h: BKPT_SET(inst) to BKPT_SET(inst, addr) for all archs ie; passess the breakpoint address as well. Patch from cherry@mahiti.org
1.24	27-Oct-2003	junyoung	branches: 1.24.16; 1.24.30; 1.24.32; 1.24.34; 1.24.36; 1.24.38; Nuke __P().
1.23	11-Aug-2003	atatat	Make sure to wrap include opt_foo.h in _KERNEL_OPT.
1.22	23-Jun-2003	martin	branches: 1.22.2; Make sure to include opt_foo.h if a defflag option FOO is used.
1.21	29-Apr-2003	scw	Add a BKPT_ADDR() macro which gives MD code a chance to munge a breakpoint address before it's used. Currently a no-op on all but sh5. This is useful on sh5, for example, to mask off the instruction type encoding in the bottom two address bits, and makes it possible to do "db> break $rXX" instead of manually munging the address.
1.20	30-Mar-2003	kristerw	Remove unneeded cast that prevents PC_REGS from being used as lvalue.
1.19	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.18	12-May-2002	matt	Eliminate commons.
1.17	17-Jun-2001	sommerfeld	branches: 1.17.2; Add %fs/%gs to trap frame and save/restore them on trap/interrupt/syscall entry from userspace. Remove special-case "by hand" validation of fs/gs register values as well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multiprocessor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT. Discussed in advance with Frank van der Linden.
1.16	07-Jun-2001	lukem	work around lint issue (inspired by similar work in sparc port)
1.15	29-Jun-2000	mrg	branches: 1.15.2; 1.15.4; remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>
1.14	04-Dec-1998	thorpej	branches: 1.14.10; 1.14.22; Support both a.out and Elf32 symbols in DDB.
1.13	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.12	26-Jun-1997	thorpej	#define DB_AOUT_SYMBOLS
1.11	06-Feb-1997	gwr	FIXUP_PC_AFTER_BREAK now takes an arg of type db_regs_t *
1.10	03-Feb-1997	mycroft	db_expr_t became a long.
1.9	03-May-1996	christos	Comment out unused routines; add prototype for kdb_trap
1.8	27-Oct-1994	cgd	new RCS ID format.
1.7	09-Oct-1994	mycroft	Fix typo.
1.6	09-Oct-1994	mycroft	Eliminate i386_saved_state.
1.5	03-Apr-1994	mycroft	branches: 1.5.2; Clean up #includes.
1.4	03-Apr-1994	mycroft	Clean up #includes.
1.3	19-Dec-1993	mycroft	branches: 1.3.2; Cleanup from magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.4	14-Nov-1993	mycroft	EFL_TF --> PSL_T
1.2.4.3	13-Nov-1993	mycroft	In watchpoint test, check type == T_TRCTRAP, just to be safe.
1.2.4.2	13-Nov-1993	mycroft	Test for watchpoint trap by inspecting the saved image of %dr6.
1.2.4.1	16-Oct-1993	mycroft	VM_MIN_KERNEL_ADDRESS --> VM_MAX_ADDRESS
1.3.2.2	19-Dec-1993	mycroft	Cleanup from magnum branch.
1.3.2.1	19-Dec-1993	mycroft	file db_machdep.h was added on branch magnum on 1993-12-19 03:41:52 +0000
1.5.2.1	11-Oct-1994	mycroft	Update from trunk.
1.14.22.1	17-Jun-2001	he	Pull up revision 1.17 (requested by sommerfeld): Add %fs/%gs to trap frame and save/restore them on trap/interupt/ syscall entry from userspace. Remove special-case ``by hand'' validation of %fs/%gs register values s well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multi- processor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT.
1.14.10.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.15.4.3	18-Oct-2002	nathanw	Catch up to -current.
1.15.4.2	20-Jun-2002	nathanw	Catch up to -current.
1.15.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.15.2.5	18-May-2002	sommerfeld	Catch up with -current. (commons elimination, changes to MI pci framework)
1.15.2.4	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.15.2.3	21-Aug-2000	sommerfeld	Add prototype for cpu_debug_dump
1.15.2.2	18-Aug-2000	sommerfeld	Preliminary multi-CPU support. Stop other cpu's on entry to DDB, resume on exit. Add "mach cpu" command to display the list of CPU's, and "mach cpu N" to view the registers of cpu N.
1.15.2.1	29-Jun-2000	sommerfeld	file db_machdep.h was added on branch sommerfeld_i386mp_1 on 2000-08-18 13:28:27 +0000
1.17.2.2	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.17.2.1	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.22.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.22.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.22.2.1	03-Aug-2004	skrll	Sync with HEAD
1.24.38.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.24.36.1	19-Apr-2006	elad	sync with head - hopefully this will work
1.24.34.1	11-Apr-2006	yamt	sync with head
1.24.32.1	22-Apr-2006	simonb	Sync with head.
1.24.30.1	09-Sep-2006	rpaulo	sync with head
1.24.16.2	26-Feb-2007	yamt	sync with head.
1.24.16.1	21-Jun-2006	yamt	sync with head.
1.25.14.1	27-Feb-2007	yamt	- sync with head. - move sched_changepri back to kern_synch.c as it doesn't know PPQ anymore.
1.26.60.4	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.26.60.3	02-May-2011	jym	Sync with head.
1.26.60.2	01-Nov-2009	jym	Sync with HEAD.
1.26.60.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.26.54.1	28-Apr-2009	skrll	Sync with HEAD.
1.26.46.1	04-May-2009	yamt	sync with head.
1.27.6.1	06-Jun-2011	jruoho	Sync with HEAD.
1.27.4.2	31-May-2011	rmind	sync with head
1.27.4.1	21-Apr-2011	rmind	sync with head
1.30.14.1	03-Dec-2017	jdolecek	update from HEAD
1.16	30-Aug-2011	bouyer	Add getlabelusesmbr(), as proposed in http://mail-index.netbsd.org/tech-userlevel/2011/08/25/msg005404.html This is used by disk tools such as disklabel(8) to dynamically decide is the undelyling platform uses a disklabel-in-mbr-partition or not (instead of using a compile-time list of ports). getlabelusesmbr() reads the sysctl kern.labelusesmbr, takes its value from the machdep #define LABELUSESMBR. For evbmips, make LABELUSESMBR 1 if the platform uses pmon as bootloader, and 0 (the previous value) otherwise.
1.15	23-Nov-2009	pooka	If cpu_disklabel includes struct dkbad, define __HAVE_DISKLABEL_DKBAD. This allows use of subr_disk_mbr on all archs. Default to it for the rump disk component. No functional change for regular kernels. (The other option would've been to include dkbad in disklabels everywhere, but arguably this approach has less possible side-effects, especially given that wedges and related magic will take over the world any second now).
1.14	26-Oct-2008	mrg	branches: 1.14.8; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.13	12-Jun-2005	dyoung	branches: 1.13.82; 1.13.86; 1.13.92; Make disklabel(8) and fdisk(8) into "host tools " last step: build and install ${TOOLDIR}/bin/${MACHINE_GNU_PLATFORM}-disklabel, ${TOOLDIR}/bin/${MACHINE_GNU_PLATFORM}-fdisk by "reaching over" to the sources in ${NETBSDSRCDIR}/sbin/{disklabel fdisk}/. To avoid clashes with a build-host's header files, especially on *BSD, the host-tools versions of fdisk and disklabel search for #includes such as disklabel.h, disklabel_acorn.h, disklabel_gpt.h, and bootinfo.h in a new #includes namespace, nbinclude/. That is, they #include <nbinclude/sys/disklabel.h>, <nbinclude/machine/disklabel.h>, <nbinclude/sparc64/disklabel.h>, instead of <sys/disklabel.h> and such. I have also updated the system headers to #include from nbinclude/-space when HAVE_NBTOOL_CONFIG_H is #defined.
1.12	08-Oct-2003	lukem	Overhaul MBR handling (part 1): <sys/bootblock.h>: * Added definitions for the Master Boot Record (MBR) used by a variety of systems (primarily i386), including the format of the BIOS Parameter Block (BPB). This information was cribbed from a variety of sources including <sys/disklabel_mbr.h> which this is a superset of. As part of this, some data structure elements and #defines were renamed to be more "namespace friendly" and consistent with other bootblocks and MBR documentation. Update all uses of the old names to the new names. <sys/disklabel_mbr.h>: * Deprecated in favor of <sys/bootblock.h> (the latter is more "host tool" friendly). amd64 & i386: * Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to be consistent with the naming convention of the msdosfs tools. * Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1 and it's confusing to have two functionally equivalent bootblocks, especially given that "ufs" has multiple meanings (it could be a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems). * Rework pbr.S (the first sector of bootxx_): + Ensure that BPB (bytes 11..89) and the partition table (bytes 446..509) do not contain code. + Add support for booting from FAT partitions if BOOT_FROM_FAT is defined. (Only set for bootxx_msdos). + Remove "dummy" partition 3; if people want to installboot(8) these to the start of the disk they can use fdisk(8) to create a real MBR partition table... + Compile with TERSE_ERROR so it fits because of the above. Whilst this is less user friendly, I feel it's important to have a valid partition table and BPB in the MBR/PBR. Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent with other platforms. * Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that we can boot off FAT partitions. * Crank version of /usr/mdec/boot to 3.1, and fix some of the other entries in the version file. installboot(8) (i386): * Read the existing MBR of the filesystem and retain the BIOS Parameter Block (BPB) in bytes 11..89 and the MBR partition table in bytes 446..509. (Previously installboot(8) would trash those two sections of the MBR.) mbrlabel(8): * Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code to map the MBR partition type to the NetBSD disklabel type. Test built "make release" for i386, and new bootblocks verified to work (even off FAT!).
1.11	07-Jul-2003	dsl	Remove 'dosparts' it hasn't been used for a while (if ever).
1.10	10-May-2003	thorpej	branches: 1.10.2; Remove redundant bounds_check_with_label() prototype.
1.9	02-Jan-2001	tron	branches: 1.9.2; Use highest bit of minor as an extra bit for the partition number to bump the maximum partition number on NetBSD-i386 to 16. (Approved by fvdl)
1.8	27-Jan-1999	thorpej	branches: 1.8.8; 1.8.20; Use the MI <sys/disklabel_mbr.h>, not our own machine-dependent definitions.
1.7	07-Jul-1998	thorpej	Use fixed-size types.
1.6	22-Feb-1998	drochner	-correct misleading name and comment for MBR partition type 14 -add FAT32 and EXT2FS
1.5	19-Feb-1998	drochner	-define new own partition ID for NetBSD (0xa9) -protect structure definitions from inclusion in assembler files
1.4	01-Oct-1996	christos	branches: 1.4.14; Fix PR/2782: Remove definition of dos_partitions[].
1.3	09-Mar-1996	ghudson	To enable people to mount DOS partitions on disks with no NetBSD partition, add the MBR partitions to the default (faked-up) disklabel used by NetBSD if it can't find a real one. If the type of the MBR partition is one of the common DOS ones, mark the partition as having an MSDOS filesystem.
1.2	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.1	14-Oct-1994	cgd	move arch dependence to archs. clean up a bit. deal with weird MAXPARTITONS.
1.4.14.3	23-Nov-1998	cgd	pull up rev 1.5 from trunk (jonathan)
1.4.14.2	23-Nov-1998	cgd	fix a spacing bogon in 1.4.14.1. (space in patch requested by fvdl, tab on branch, making subsequent pullup harder.)
1.4.14.1	09-Oct-1998	cgd	pull up support for DOS FAT32 and Linux ext2fs partitions in the default disklabel, via patch. (fvdl)
1.8.20.1	01-May-2001	he	Pull up revision 1.9 (requested by fvdl): Increase the number of BSD disklabel partitions on i386 to 16.
1.8.8.1	05-Jan-2001	bouyer	Sync with HEAD
1.9.2.2	02-Jan-2001	tron	Use highest bit of minor as an extra bit for the partition number to bump the maximum partition number on NetBSD-i386 to 16. (Approved by fvdl)
1.9.2.1	02-Jan-2001	tron	file disklabel.h was added on branch sommerfeld_i386mp_1 on 2001-01-02 23:06:55 +0000
1.10.2.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.10.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.10.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.10.2.1	03-Aug-2004	skrll	Sync with HEAD
1.13.92.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.13.86.2	11-Mar-2010	yamt	sync with head
1.13.86.1	04-May-2009	yamt	sync with head.
1.13.82.1	17-Jan-2009	mjf	Sync with HEAD.
1.14.8.1	24-Oct-2010	jym	Sync with HEAD
1.4	17-Jun-1993	mycroft	Clean up deleted files.
1.3	17-Jun-1993	mycroft	Make ioctl definitions ANSI-compliant.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.1	24-Sep-2022	riastradh	x86: Support EFI runtime services. This creates a special pmap, efi_runtime_pmap, which avoids setting PTE_U but allows mappings to lie in what would normally be user VM -- this way we don't fall afoul of SMAP/SMEP when executing EFI runtime services from CPL 0. SVS does not apply to the EFI runtime pmap. The mechanism is intended to work with either physical addressing or virtual addressing; currently the bootloader does physical addressing but in principle it could be modified to do virtual addressing instead, if it allocated virtual pages, assigned them in the memory map, and issued RT->SetVirtualAddressMap. Not sure pmap_activate_sync and pmap_deactivate_sync are correct, need more review from an x86 wizard. If this causes fallout, it can be disabled temporarily without reverting anything by just making efi_runtime_init return immediately without doing anything, or by removing options EFI_RUNTIME. amd64-only for now pending type fixes and testing on i386.
1.3	19-Dec-1993	mycroft	Clean up deleted files.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.14	27-Apr-2015	knakahara	fix i386 build for MSI/MSI-X support code.
1.13	29-Mar-2014	christos	branches: 1.13.6; make pci_intr_string and eisa_intr_string take a buffer and a length instead of relying in local static storage.
1.12	17-Nov-2009	dyoung	branches: 1.12.12; 1.12.22; 1.12.26; Delete unused functions eisa_mem_alloc() and eisa_mem_free().
1.11	27-Jun-2008	cegger	branches: 1.11.10; struct device * -> device_t
1.10	26-Feb-2003	fvdl	branches: 1.10.104; 1.10.108; 1.10.110; 1.10.112; Redirect some includes to x86/foo.h
1.9	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.8	11-Aug-2000	thorpej	branches: 1.8.2; 1.8.4; Add functions to read EISA configuration data for MEM, IRQ, DMA, and IO. XXX Just error stubs on the i386 right now -- someone needs to write XXX EISA BIOS code for i386.
1.7	04-Jun-2000	cgd	branches: 1.7.2; Implement the more flexiable `evcnt' interface as discussed (briefly) on tech-kern and now documented in evcnt(9).
1.6	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.5	19-Mar-1999	cgd	branches: 1.5.8; 1.5.16; 1.5.18; Moved from ../eisa/eisa_machdep.h,v
1.4	06-Jun-1997	thorpej	Pull thorpej-bus-dma branch into mainline.
1.3	21-Oct-1996	thorpej	branches: 1.3.8; Implement eisa_mem_{alloc,free}(), a method for allocating/freeing EISA bus physical address space. This i386 implementation allocates system physical address space after the end of RAM.
1.2	09-Apr-1996	cgd	make the EISA code look a bit more like the PCI code, in the way the bus is configured, interrupts are handled, etc. define eisa_chipset_tag_t and eisa_intr_handle_t types, and the following functions: eisa_attach_hook() eisa_maxslots() eisa_intr_map() eisa_intr_string() eisa_intr_establish() eisa_intr_disestablish() to do the right things for the i386.
1.1	04-Mar-1996	cgd	Attach isa, eisa, and pci to mainbus now, rather than root. The i386 now has a single device tree (rather than a forest). Also, attach EISA only if it's present, and attach in order: PCI, EISA, ISA (most specific to least specific).
1.3.8.1	18-May-1997	thorpej	Add an external declaration for the eisa dma tag.
1.5.18.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.5.16.3	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.5.16.2	25-Jun-2000	sommerfeld	Merge up to just-post-1.5 -current
1.5.16.1	29-Feb-2000	sommerfeld	mpbiosreg.h
1.5.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.7.2.1	11-Aug-2000	thorpej	Pullup from trunk: Add EISA configuration readining infrastructure.
1.8.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.8.2.1	18-Oct-2002	nathanw	Catch up to -current.
1.10.112.1	27-Jun-2008	simonb	Sync with head.
1.10.110.1	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.10.108.2	11-Mar-2010	yamt	sync with head
1.10.108.1	04-May-2009	yamt	sync with head.
1.10.104.1	29-Jun-2008	mjf	Sync with HEAD.
1.11.10.1	24-Oct-2010	jym	Sync with HEAD
1.12.26.1	18-May-2014	rmind	sync with head
1.12.22.2	03-Dec-2017	jdolecek	update from HEAD
1.12.22.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.12.12.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.13.6.1	06-Jun-2015	skrll	Sync with HEAD
1.14	11-Feb-2025	jkoshy	Add an alternate spelling for an i386 relocation type.
1.13	06-Nov-2017	christos	branches: 1.13.40; Cleanup and clarify the ELFSIZE mess: We now have 2 variables automatically set in elf_machdep.h: ARCH_ELFSIZE: the size for userland binaries KERN_ELFSIZE: the size for the kernel binaries DB_ELFSIZE has been deleted and KERN_ELFSIZE should have always the same values DB_ELFSIZE used to have. In sys/exec_elf.h, if ELFSIZE is not set, it is set to KERN_ELFSIZE for the kernel and ARCH_ELFSIZE for userland. These defaults should eliminate the need for most manual ELFSIZE setting.
1.12	02-Feb-2016	christos	one more.
1.11	02-Feb-2016	christos	more relocations.
1.10	30-May-2009	skrll	branches: 1.10.22; 1.10.40; Add TLS relocation definitions.
1.9	09-Dec-2001	thorpej	branches: 1.9.2; 1.9.124; 1.9.138; Add support for dumping ELF-cormat core files.
1.8	02-Apr-2000	minoura	branches: 1.8.8; 1.8.10; Move dl* function definitions to libc on ELF. Based on the patch supplied by Takuya Shiozaki <tshiozak@astec.co.jp>. See http://mail-index.netbsd.org/tech-userlevel/2000/02/23/0000.html.
1.7	25-Oct-1999	kleink	Update to match new SVR4-style definition names in <sys/exec_elf.h>.
1.6	25-Feb-1999	christos	branches: 1.6.8; 1.6.10; 1.6.12; Don't need to stick the interpreter at 0 anymore, it relocates itself.
1.5	07-Feb-1999	christos	s/JUMP/JMP/ like every other arch.
1.4	11-Jan-1999	christos	We cannot relocate our interpreter (yet).
1.3	12-Jul-1998	thorpej	Add i386 relocations.
1.2	11-Oct-1997	christos	Add ENDIANNESS constant
1.1	26-Sep-1996	cgd	branches: 1.1.10; add and use a machine-dependent header, which currently defines some macros to use to remove #ifdefs from the machine ID case check. Eventually, these headers will contain other information, e.g. machine-dependent relocation information, etc.
1.1.10.1	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.6.12.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.6.10.1	15-Nov-1999	fvdl	Sync with -current
1.6.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.8.10.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.8.8.1	08-Jan-2002	nathanw	Catch up to -current.
1.9.138.2	01-Nov-2009	jym	Sync with HEAD.
1.9.138.1	31-May-2009	jym	Sync with HEAD.
1.9.124.1	20-Jun-2009	yamt	sync with head
1.9.2.2	09-Dec-2001	thorpej	Add support for dumping ELF-cormat core files.
1.9.2.1	09-Dec-2001	thorpej	file elf_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-12-09 23:05:58 +0000
1.10.40.1	19-Mar-2016	skrll	Sync with HEAD
1.10.22.1	03-Dec-2017	jdolecek	update from HEAD
1.13.40.1	02-Aug-2025	perseant	Sync with HEAD
1.29	17-Mar-2000	mycroft	branches: 1.29.2; In the `MY THAT'S GROSS' department... Eliminate the recursive include of machine/endian.h from sys/endian.h.
1.28	16-Mar-2000	mycroft	Foolish consistency. Mainly, always use underscores and sys/endian.h.
1.27	21-Aug-1999	simonb	branches: 1.27.2; Include <sys/endian.h> after defining whether where are little- or big-endian. i386, pc532 and vax still include <machine/byte_swap.h> and define macros for the {n,h}to{h,n}*() functions. mips also defines some endian-independent assembly-code aliases for unaligned memory accesses.
1.26	24-Jan-1999	mycroft	Mark [hn]to[nh][ls]() with __const__, so they are subject to CSE.
1.25	15-Jan-1999	bouyer	Move the bswap functions from libutil to libc (this bups the minor of libc and the major of libutil). For little-endian architectures merge the bnswap() assembly versions with nto* and hton* using symbols aliasing. Use symbol renaming for the bswap function in this case to avoid namespace pollution. Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian machines, common code for inline macros go in machine/byte_swap.h Sync libkern with libc. Adjust #include in kernel sources for machine/bswap.h.
1.24	16-Dec-1998	kleink	Support for compound statements enclosed in parenthes to appear as expressions is a GNU C extension; mark as such them with __extension__ to suppress portability warnings. Addresses kern/3562 and misc/6185, as suggested by Dave Sainty <dave@dtsp.co.nz>.
1.23	15-Aug-1998	mycroft	Assign my copyrights to TNF.
1.22	15-Jan-1998	thorpej	Generate dependencies on the CPU type options (I386_CPU, I486_CPU, ...)
1.21	09-Oct-1997	bouyer	Add byte-swapping functions (bswap16, bswap32, bswap64) to libkern. Only assembly version for i386 bswap16 and bswap32 for now (bswap64 uses bswap32). Contribution of assembly versions of these are welcome. Add byte-swapping of ext2fs metadata for big-endian systems. Tested on i386 and sparc.
1.20	17-Jul-1997	thorpej	branches: 1.20.2; Give the __GNUC__ versions of ntohl() and friends the proper (in_addr_t or in_port_t) signature.
1.19	11-Oct-1996	christos	add in_addr_t and in_port_t and use them.
1.18	08-May-1996	jtc	Revert last change. Changing types from unsigned {long,short} to u_int{32,16}_t causes problems in a lot of userlevel code because those types aren't defined when machine/ endian.h is included. And IMO including sys/types.h in machine/endian.h is clearly wrong. XPG4.2 defines new types in_addr_t and in_port_t for the {h,n}to{n,h}{l,s} functions. But we don't have time to figure out the implementation issues given the current release schedule.
1.17	03-May-1996	christos	use u_int32_t instead of unsigned long and u_int16_t instead of short in the {h,n}to{n,h}{l,s} inline routines and macros.
1.16	01-Jun-1995	mycroft	Allow [nh]to[hn][ls]() to do constant folding.
1.15	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.14	27-Oct-1994	cgd	new RCS ID format.
1.13	09-Oct-1994	mycroft	Make multiple-inclusion protection consistent.
1.12	10-Sep-1994	jtc	asm -> __asm (PR #463).
1.11	15-Jul-1994	mycroft	Since we're no longer using xchg instructions, there's no reason to prefer a q reg over and other. Also, fix some types in the case of _POSIX_SOURCE.
1.10	16-Mar-1994	jtc	branches: 1.10.2; Inline assembly [hn]to[nh][sl] functions used xchgb %h1,%b1 instead of the faster (on the i486 & i586) rorw $8, %w1. The inline assembly for GCC 1.X already used rorw. Using rorw is one byte longer, but we wouldn't be inlining at all if we weren't optimizing for speed.
1.9	12-Feb-1994	cgd	fix from Thomas Eberhardt <thomas@mathematik.uni-Bremen.de>: byte swap macros used a variable "X" internally, which would get screwed up if user programs #defined "X". change uses of X to __x.
1.8	20-Dec-1993	mycroft	branches: 1.8.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.7	17-Sep-1993	cgd	get rid of the #ifndef KERNEL around the include of sys/cdefs.h, a la 4.4, because it seriously breaks things otherwise, given inclusion of endian.h in types.h
1.6	16-Sep-1993	mycroft	Add an #ifndef _POSIX_SOURCE...
1.5	16-Sep-1993	mycroft	Add _QUAD_HIGHWORD and _QUAD_LOWWORD for Torek's quad stuff.
1.4	16-Aug-1993	jtc	branches: 1.4.2; Declare ntohl, ntohs, htonl, and htons functions. Use GCC's inline assembly iff we are using GCC. Use bswap instruction if using gcc and defined(KERNEL) && ((defined(I486_CPU) \|\| defined(I586_CPU)) && !defined(I386_CPU))
1.3	22-May-1993	cgd	add rcsids to everything and clean up headers
1.2	21-Mar-1993	cgd	after 0.2.2 "stable" patches applied
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.4.2.1	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.8.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.8.2.1	20-Dec-1993	mycroft	file endian.h was added on branch magnum on 1993-12-20 05:25:12 +0000
1.10.2.3	11-Oct-1994	mycroft	Update from trunk.
1.10.2.2	12-Sep-1994	cgd	from trunk
1.10.2.1	15-Jul-1994	cgd	from trunk
1.20.2.1	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.27.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.29.2.2	17-Mar-2000	mycroft	In the `MY THAT'S GROSS' department... Eliminate the recursive include of machine/endian.h from sys/endian.h.
1.29.2.1	17-Mar-2000	mycroft	file endian.h was added on branch sommerfeld_i386mp_1 on 2000-03-17 00:09:21 +0000
1.3	30-Jan-2006	dsl	Move the definitions of ntohl() and friends into sys/endian.h where they are defined in terms of bswap32() and bswap16(). This makes the definition be in the same place for all systems regardless of creed^Wendianness.
1.2	10-Jun-2004	kleink	branches: 1.2.12; 1.2.22; Reflect <sys/endian.h> rev. 1.4: make htonl() et al. arguments and results uint{16,32}_t. Noted by Ian Zagorskih.
1.1	17-Mar-2000	mycroft	branches: 1.1.2; 1.1.8; 1.1.32; 1.1.34; In the `MY THAT'S GROSS' department... Eliminate the recursive include of machine/endian.h from sys/endian.h.
1.1.34.1	14-Jun-2004	tron	Pull up revision 1.2 (requested by kleink in ticket #467): Reflect <sys/endian.h> rev. 1.4: make htonl() et al. arguments and results uint{16,32}_t. Noted by Ian Zagorskih.
1.1.32.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.32.2	18-Sep-2004	skrll	Sync with HEAD.
1.1.32.1	03-Aug-2004	skrll	Sync with HEAD
1.1.8.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.1.8.1	17-Mar-2000	bouyer	file endian_machdep.h was added on branch thorpej_scsipi on 2000-11-20 20:09:26 +0000
1.1.2.2	17-Mar-2000	mycroft	In the `MY THAT'S GROSS' department... Eliminate the recursive include of machine/endian.h from sys/endian.h.
1.1.2.1	17-Mar-2000	mycroft	file endian_machdep.h was added on branch sommerfeld_i386mp_1 on 2000-03-17 00:09:21 +0000
1.2.22.1	01-Feb-2006	yamt	sync with head.
1.2.12.1	21-Jun-2006	yamt	sync with head.
1.7	08-Oct-1996	cgd	moved to aout_machdep.h (via repository copy)
1.6	27-Oct-1994	cgd	new RCS ID format.
1.5	28-Jan-1994	jtc	branches: 1.5.2; Fix spelling error in Copyright notice
1.4	18-Oct-1993	pk	Put definition of __LDPGSZ and `struct relocation_info' where they belong: in <machine/exec.h>.
1.3	18-Jun-1993	cgd	branches: 1.3.4; convert magic numbers to network byte order, and attendent changes
1.2	03-Jun-1993	cgd	add prototype for a function to be called from execve to check a_mid type
1.1	02-Jun-1993	cgd	machine-dependent exec stuff
1.3.4.2	18-Oct-1993	mycroft	Merge changes from trunk.
1.3.4.1	15-Oct-1993	mycroft	Define __LDPGSZ.
1.5.2.2	28-Jan-1994	jtc	Fix spelling error in Copyright notice
1.5.2.1	28-Jan-1994	jtc	file exec.h was added on branch magnum on 1994-01-28 23:44:16 +0000
1.2	12-Feb-2014	dsl	Add definitions of the default control words directly to this file instead of pulling the kernel definition of the fpu (etc) into userspace programs. I've included machine/fenv.h into x86/cpu.c to ensure the duplicated definitions stay in step. The default control words are now the hardware defaults. XXX: Anyone care to explain the differences between the i386 and amd64 versions of this file?
1.1	31-Jul-2010	joerg	branches: 1.1.2; 1.1.4; 1.1.6; 1.1.12; 1.1.16; 1.1.26; 1.1.30; Add support for fenv.h interface for i386 and amd64. Submitted by Stathis Kamperis as part of GSoC 2010 and ported from FreeBSD.
1.1.30.1	18-May-2014	rmind	sync with head
1.1.26.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.1.16.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.1.12.2	05-Mar-2011	rmind	sync with head
1.1.12.1	31-Jul-2010	rmind	file fenv.h was added on branch rmind-uvmplock on 2011-03-05 20:50:41 +0000
1.1.6.2	24-Oct-2010	jym	Sync with HEAD
1.1.6.1	31-Jul-2010	jym	file fenv.h was added on branch jym-xensuspend on 2010-10-24 22:48:02 +0000
1.1.4.2	17-Aug-2010	uebayasi	Sync with HEAD.
1.1.4.1	31-Jul-2010	uebayasi	file fenv.h was added on branch uebayasi-xip on 2010-08-17 06:44:37 +0000
1.1.2.2	11-Aug-2010	yamt	sync with head.
1.1.2.1	31-Jul-2010	yamt	file fenv.h was added on branch yamt-nfs-mp on 2010-08-11 22:52:12 +0000
1.12	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.11	21-May-2001	perry	branches: 1.11.2; remove needless externs in front of function prototypes
1.10	18-Feb-1998	mycroft	branches: 1.10.28; Include correct information for long doubles (i387 extended precision).
1.9	03-Feb-1998	mycroft	Prototype __flt_rounds() consistently.
1.8	20-Jun-1995	jtc	branches: 1.8.18; Wrap with #ifndef _XXX_FLOAT_H_/#define _XXX_FLOAT_H_/ ... /#endif.
1.7	20-Jun-1995	jtc	#include <sys/cdefs.h>. Wrap __flt_rounds() declaration with __BEGIN_DECLS/__END_DECLS.
1.6	07-Apr-1995	jtc	Changed FLT_ROUNDS from constant to a call to __flt_rounds(), so that the current rounding mode is accurately reported.
1.5	27-Oct-1994	cgd	new RCS ID format.
1.4	09-Jul-1993	cgd	update for better FP routines, from AT&T & elsewhere
1.3	22-May-1993	cgd	add rcsids to everything and clean up headers
1.2	21-Mar-1993	cgd	after 0.2.2 "stable" patches applied
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.8.18.1	08-May-1998	mycroft	Sync with trunk, per request of mycroft.
1.10.28.1	21-Jun-2001	nathanw	Catch up to -current.
1.11.2.2	21-May-2001	perry	remove needless externs in front of function prototypes
1.11.2.1	21-May-2001	perry	file float.h was added on branch sommerfeld_i386mp_1 on 2001-05-21 04:47:36 +0000
1.41	26-Dec-2021	andvar	fix various typos, mainly in comments.
1.40	14-Feb-2019	cherry	Welcome XENPVHVM mode. It is UP only, has xbd(4) and xennet(4) as PV drivers. The console is com0 at isa and the native portion is very rudimentary AT architecture, so is probably suboptimal to run without PV support.
1.39	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.38	19-Nov-2018	kre	Hide differences between i386 and amd64 interrupt frames so XEN does not need to know there is one. Hopefully unbreak i386 build.
1.37	12-Aug-2017	maxv	branches: 1.37.2; 1.37.4; Remove the vm86 fields from the trap frame. It seems to me that we could now remove the '-16' when initializing pcb_esp0.
1.36	14-Jun-2017	chs	add an lwp_trapframe() interface to return an LWP's user trapframe. needed by dtrace.
1.35	19-Feb-2012	rmind	branches: 1.35.2; 1.35.16; Remove COMPAT_SA / KERN_SA. Welcome to 6.99.3! Approved by core@.
1.34	02-Jun-2011	dsl	branches: 1.34.2; 1.34.6; Replace the trapframe fields for segment registers with two uint16_t fields - one of which is an explicit pad.
1.33	14-Nov-2008	ad	branches: 1.33.4; 1.33.8; 1.33.10; 1.33.14; Wrap sigframe_sigcontext in _KERNEL
1.32	14-Nov-2008	ad	Remove some compat ifdefs that aren't really needed.
1.31	15-Oct-2008	wrstuden	branches: 1.31.2; Merge wrstuden-revivesa into HEAD.
1.30	28-Apr-2008	martin	branches: 1.30.2; 1.30.6; Remove clause 3 and 4 from TNF licenses
1.29	17-Oct-2007	garbled	branches: 1.29.16; 1.29.18; 1.29.20; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.28	17-May-2007	yamt	branches: 1.28.10; merge yamt-idlelwp branch. asked by core@. some ports still needs work. from doc/BRANCHES: idle lwp, and some changes depending on it. 1. separate context switching and thread scheduling. (cf. gmcgarry_ctxsw) 2. implement idle lwp. 3. clean up related MD/MI interfaces. 4. make scheduler(s) modular.
1.27	09-Feb-2007	ad	branches: 1.27.2; 1.27.6; 1.27.8; 1.27.14; Merge newlock2 to head.
1.26	26-Dec-2005	perry	branches: 1.26.20; u_intN_t -> uintN_t
1.25	25-Sep-2003	christos	branches: 1.25.16; constify sendsig/trapsignal
1.24	10-Sep-2003	christos	expose sendsig_sigcontext for compat_16
1.23	10-Sep-2003	christos	#ifdef struct sigframe_sigcontext and struct sigcontext, so that only code that really needs them, get them. XXX1: We need to fix the vm86 api. XXX2: We need to fix the ibcs2 signal delivery mechanism.
1.22	06-Sep-2003	christos	SA_SIGINFO changes.
1.21	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.20	17-Jan-2003	thorpej	branches: 1.20.2; Merge the nathanw_sa branch.
1.19	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.18	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.17	23-Jun-2002	thorpej	During native signal delivery (and iBCS2, since it uses the native sigcode), arrange to have the signal handler invoked directly, using the trampoline only for the return path, saving a call insn. Play some other stack trickery in the trampoline to turn 2 pushl's into one movl.
1.16	17-Jun-2001	sommerfeld	branches: 1.16.2; 1.16.14; Add %fs/%gs to trap frame and save/restore them on trap/interrupt/syscall entry from userspace. Remove special-case "by hand" validation of fs/gs register values as well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multiprocessor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT. Discussed in advance with Frank van der Linden.
1.15	08-Feb-2001	christos	branches: 1.15.2; appease lint: use u_int32_t and name the bit fields.
1.14	11-Dec-2000	tron	Protect against multiple inclusion.
1.13	15-Aug-1998	mycroft	branches: 1.13.12; 1.13.20; 1.13.24; Assign my copyrights to TNF.
1.12	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.11	01-May-1995	mycroft	Add VM86 segment registers to trapframe.
1.10	27-Oct-1994	cgd	new RCS ID format.
1.9	09-Oct-1994	mycroft	Eliminate tf_isp.
1.8	07-Apr-1994	mycroft	branches: 1.8.2; Implement dynamic IRQ configuration and IRQ sharing. Inline spl*() calls. Reorganize and clean up the relevant code.
1.7	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.6	21-Sep-1993	brezak	Include sys/signal.h instead of machine/signal.h
1.5	19-Sep-1993	brezak	Define signal frame here.
1.4	16-Sep-1993	brezak	Changes to use a trap frame for syscalls.
1.3	27-Jun-1993	andrew	branches: 1.3.4; Clobbered '/*' within comment, to keep gcc -Wall happy.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.3.4.4	11-Jan-1994	mycroft	Put the trapframe back in the old order; the reason for changing it is now gone.
1.3.4.3	13-Oct-1993	mycroft	Remove if_vec from intrframe, as it no longer exists.
1.3.4.2	09-Oct-1993	mycroft	Rearrange trap and interrupt frames so segment registers are pushed before pushal.
1.3.4.1	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.8.2.1	11-Oct-1994	mycroft	Update from trunk.
1.13.24.2	30-Jul-2001	he	Pull up revision 1.14 (requested by he): Protect against multiple inclusion.
1.13.24.1	17-Jun-2001	he	Pull up revision 1.16 (requested by sommerfeld): Add %fs/%gs to trap frame and save/restore them on trap/interupt/ syscall entry from userspace. Remove special-case ``by hand'' validation of %fs/%gs register values s well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multi- processor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT.
1.13.20.5	25-Jun-2002	sommerfeld	Resynch with -current.
1.13.20.4	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.13.20.3	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.13.20.2	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.13.20.1	24-Aug-2000	sommerfeld	Since cpu_switch() is now always entered and left at IPL_HIGH, don't bother saving CPL as part of the switch. Adjust "struct switchframe" in frame.h and cpu_fork() in vm_machdep.c to compensate.
1.13.12.2	11-Feb-2001	bouyer	Sync with HEAD.
1.13.12.1	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.15.2.9	11-Dec-2002	thorpej	Sync with HEAD.
1.15.2.8	18-Oct-2002	nathanw	Catch up to -current.
1.15.2.7	01-Aug-2002	nathanw	Catch up to -current.
1.15.2.6	19-Jul-2002	nathanw	Invoke upcalls directly, instead of going through a trampoline. (Since upcalls shouldn't return at all, even the return part of the trampoline is unnecessary).
1.15.2.5	02-Jul-2002	nathanw	Whitespace.
1.15.2.4	17-Nov-2001	nathanw	Adapt to new upcall signature and possible argument copying.
1.15.2.3	24-Aug-2001	nathanw	A few files and lwp/proc conversions I missed in the last big update. GENERIC runs again.
1.15.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.15.2.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.16.14.1	16-Jul-2002	gehenna	catch up with -current.
1.16.2.2	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.16.2.1	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.20.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.20.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.20.2.1	03-Aug-2004	skrll	Sync with HEAD
1.25.16.3	03-Sep-2007	yamt	sync with head.
1.25.16.2	26-Feb-2007	yamt	sync with head.
1.25.16.1	21-Jun-2006	yamt	sync with head.
1.26.20.2	30-Jan-2007	ad	Remove support for SA. Ok core@.
1.26.20.1	17-Nov-2006	ad	Checkpoint work in progress.
1.27.14.1	22-May-2007	matt	Update to HEAD.
1.27.8.1	11-Jul-2007	mjf	Sync with head.
1.27.6.1	27-May-2007	ad	Sync with head.
1.27.2.1	17-Feb-2007	yamt	- separate context switching and thread scheduling. - introduce idle lwp. - change some related MD/MI interfaces and implement i386 version.
1.28.10.1	06-Nov-2007	matt	sync with HEAD
1.29.20.2	04-May-2009	yamt	sync with head.
1.29.20.1	16-May-2008	yamt	sync with head.
1.29.18.1	18-May-2008	yamt	sync with head.
1.29.16.2	17-Jan-2009	mjf	Sync with HEAD.
1.29.16.1	02-Jun-2008	mjf	Sync with HEAD.
1.30.6.2	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.30.6.1	19-Oct-2008	haad	Sync with HEAD.
1.30.2.3	22-Jun-2008	wrstuden	Re-add cpu_upcall() and page fault code. i386 kernels now compile. They don't boot, but that seems to be a consequence of current from the day this branch was started.
1.30.2.2	14-May-2008	wrstuden	Per discussion with ad, remove most of the #include <sys/sa.h> lines as they were including sa.h just for the type(s) needed for syscallargs.h. Instead, create a new file, sys/satypes.h, which contains just the types needed for syscallargs.h. Yes, there's only one now, but that may change and it's probably more likely to change if it'd be difficult to handle. :-) Per discussion with matt at n dot o, add an include of satypes.h to sigtypes.h. Upcall handlers are kinda signal handlers, and signalling is the header file that's already included for syscallargs.h that closest matches SA. This shaves about 3000 lines off of the diff of the branch relative to the base. That also represents about 18% of the total before this checkin. I think this reduction is very good thing.
1.30.2.1	10-May-2008	wrstuden	Initial checkin of re-adding SA. Everything except kern_sa.c compiles in GENERIC for i386. This is still a work-in-progress, but this checkin covers most of the mechanical work (changing signalling to be able to accomidate SA's process-wide signalling and re-adding includes of sys/sa.h and savar.h). Subsequent changes will be much more interesting. Also, kern_sa.c has received partial cleanup. There's still more to do, though.
1.31.2.1	19-Jan-2009	skrll	Sync with HEAD.
1.33.14.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.33.10.1	06-Jun-2011	jruoho	Sync with HEAD.
1.33.8.1	12-Jun-2011	rmind	sync with head
1.33.4.1	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.34.6.1	24-Feb-2012	mrg	sync to -current.
1.34.2.1	17-Apr-2012	yamt	sync with head
1.35.16.1	28-Aug-2017	skrll	Sync with HEAD
1.35.2.1	03-Dec-2017	jdolecek	update from HEAD
1.37.4.1	10-Jun-2019	christos	Sync with HEAD
1.37.2.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.35	30-Jul-2022	riastradh	x86: Eliminate mfence hotpatch for membar_sync. The more-compatible LOCK ADD $0,-N(%rsp) turns out to be cheaper than MFENCE anyway. Let's save some space and maintenance and rip out the hotpatching for it.
1.34	09-Apr-2022	riastradh	x86: Every load is a load-acquire, so membar_consumer is a noop. lfence is only needed for MD logic, such as operations on I/O memory rather than normal cacheable memory, or special instructions like RDTSC -- never for MI synchronization between threads/CPUs. No need for hot-patching to do lfence here. (The x86_lfence function might reasonably be patched on i386 to do lfence for MD logic, but it isn't now and this doesn't change that.)
1.33	01-May-2020	maxv	Switch the rest of i386 to the x86_hotpatch mechanism.
1.32	01-May-2020	maxv	Use the hotpatch framework when patching _atomic_cas_64.
1.31	26-Apr-2020	maxv	Use the hotpatch framework for LFENCE/MFENCE.
1.30	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.29	12-Oct-2019	maxv	branches: 1.29.6; Rewrite the FPU code on x86. This greatly simplifies the logic and removes the dependency on IPL_HIGH. NVMM is updated accordingly. Posted on port-amd64 a week ago. Bump the kernel version to 9.99.16.
1.28	14-Feb-2019	cherry	Welcome XENPVHVM mode. It is UP only, has xbd(4) and xennet(4) as PV drivers. The console is com0 at isa and the native portion is very rudimentary AT architecture, so is probably suboptimal to run without PV support.
1.27	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.26	17-Jun-2018	maxv	branches: 1.26.2; i586 and below don't have this 3-byte nop, so use three 1-byte nops, reported by Nathanial Sloss
1.25	27-Jan-2018	maxv	branches: 1.25.2; Declare INTR_RECURSE_HWFRAME, same as amd64.
1.24	27-Jan-2018	maxv	Add SMAP support for i386.
1.23	27-Jan-2018	maxv	Remove DO_DEFERRED_SWITCH and DO_DEFERRED_SWITCH_RETRY, unused.
1.22	27-Jan-2018	maxv	Use .pushsection (like amd64), and align INTRENTRY.
1.21	07-Jan-2018	maxv	Switch x86_retpatch[] -> HOTPATCH().
1.20	07-Jan-2018	maxv	Switch x86_lockpatch[] -> HOTPATCH().
1.19	07-Jan-2018	maxv	Implement a real hotpatch feature. Define a HOTPATCH() macro, that puts a label and additional information in the new .rodata.hotpatch kernel section. In patch.c, scan the section and patch what needs to be. Now it is possible to hotpatch the content of a macro. SMAP is switched to use this new system; this saves a call+ret in each kernel entry/exit point. Many other operating systems do the same.
1.18	17-Sep-2017	maxv	Declare INTRFASTEXIT as a function, like amd64; will be expanded soon.
1.17	17-Sep-2017	maxv	Remove TRAPLOG from i386. Nowadays there are better instrumentation tools, in both software and hardware.
1.16	14-Jun-2017	maxv	style
1.15	26-Jul-2011	yamt	branches: 1.15.12; 1.15.30; 1.15.46; comment
1.14	07-Jul-2008	gmcgarry	branches: 1.14.2; 1.14.10; Selector registers are 16-bit and binutils 2.18 insists that only 16-bit accesses are permitted on them. Therefore, change movl to movw. No change to machine code generated.
1.13	21-Apr-2008	cegger	branches: 1.13.2; 1.13.4; 1.13.6; Access Xen's vcpu info structure per-CPU. Tested on i386 and amd64 (both dom0 and domU) by me. Xen2 tested (both dom0 and domU) by bouyer. OK bouyer
1.12	29-Feb-2008	yamt	branches: 1.12.2; don't bother to check curlwp==NULL.
1.11	11-Jan-2008	bouyer	branches: 1.11.2; 1.11.6; Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.10	29-Nov-2007	yamt	branches: 1.10.6; instead of setting pcb_onfault on every calls of copyin and friends, make the fault handler investigate program counter of faulting code. inspired from linux.
1.9	14-Nov-2007	ad	Clear the direction flag on entry to the kernel.
1.8	18-Oct-2007	yamt	branches: 1.8.2; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.7	09-Feb-2007	ad	branches: 1.7.6; 1.7.22; 1.7.24; 1.7.26; 1.7.28; Merge newlock2 to head.
1.6	08-Dec-2006	yamt	- pass intrframe by-pointer, not by-value. - make i386 and xen use per-cpu interrupt stack. xen part is reviewed by Manuel Bouyer.
1.5	12-May-2004	yamt	branches: 1.5.12; 1.5.40; 1.5.42; add a comment to note that INTRFASTEXIT and trap() should be in sync.
1.4	20-Feb-2004	yamt	defer pmap switching until it's really needed to avoid frequent loading of cr3 register, which involves tlb flush. with some fixes/improvements from Stephan Uphoff and Bang Jun-Young.
1.3	04-Oct-2003	junyoung	Rearrange instructions for better use of pipeline. This shaves off 0.5% of time on interrupt/trap entry on my Athlon 1800.
1.2	17-Jan-2003	thorpej	branches: 1.2.2; Merge the nathanw_sa branch.
1.1	22-Nov-2002	fvdl	branches: 1.1.2; Asm macros for interrupt/trap entry/exit, used in multiple .S files.
1.1.2.2	11-Dec-2002	thorpej	Sync with HEAD.
1.1.2.1	22-Nov-2002	thorpej	file frameasm.h was added on branch nathanw_sa on 2002-12-11 06:01:00 +0000
1.2.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.2.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.2.2.1	03-Aug-2004	skrll	Sync with HEAD
1.5.42.1	10-Dec-2006	yamt	sync with head.
1.5.40.2	12-Jan-2007	ad	Sync with head.
1.5.40.1	20-Oct-2006	ad	Make ASTs per-LWP.
1.5.12.7	17-Mar-2008	yamt	sync with head.
1.5.12.6	21-Jan-2008	yamt	sync with head
1.5.12.5	07-Dec-2007	yamt	sync with head
1.5.12.4	15-Nov-2007	yamt	sync with head.
1.5.12.3	27-Oct-2007	yamt	sync with head.
1.5.12.2	26-Feb-2007	yamt	sync with head.
1.5.12.1	30-Dec-2006	yamt	sync with head.
1.7.28.2	18-Nov-2007	bouyer	Sync with HEAD
1.7.28.1	25-Oct-2007	bouyer	Sync with HEAD.
1.7.26.1	29-Sep-2007	yamt	DO_DEFERRED_SWITCH, CHECK_DEFERRED_SWITCH: remove an unused argument.
1.7.24.3	23-Mar-2008	matt	sync with HEAD
1.7.24.2	09-Jan-2008	matt	sync with HEAD
1.7.24.1	06-Nov-2007	matt	sync with HEAD
1.7.22.3	03-Dec-2007	joerg	Sync with HEAD.
1.7.22.2	14-Nov-2007	joerg	Sync with HEAD.
1.7.22.1	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.7.6.4	03-Dec-2007	ad	Sync with HEAD.
1.7.6.3	03-Dec-2007	ad	Sync with HEAD.
1.7.6.2	30-Oct-2007	ad	CHECK_ASTPENDING: no need to check for curlwp == NULL.
1.7.6.1	23-Oct-2007	ad	Sync with head.
1.8.2.3	18-Feb-2008	mjf	Sync with HEAD.
1.8.2.2	08-Dec-2007	mjf	Sync with HEAD.
1.8.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.10.6.1	09-Jan-2008	bouyer	Move Xen bits to i386/include/frameasm.h and add CLI/STI macros for i386. Make native i386 kernels build again.
1.11.6.3	28-Sep-2008	mjf	Sync with HEAD.
1.11.6.2	02-Jun-2008	mjf	Sync with HEAD.
1.11.6.1	03-Apr-2008	mjf	Sync with HEAD.
1.11.2.1	24-Mar-2008	keiichi	sync with head.
1.12.2.1	18-May-2008	yamt	sync with head.
1.13.6.1	18-Jul-2008	simonb	Sync with head.
1.13.4.1	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.13.2.1	04-May-2009	yamt	sync with head.
1.14.10.1	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.14.2.2	07-Jul-2008	gmcgarry	Selector registers are 16-bit and binutils 2.18 insists that only 16-bit accesses are permitted on them. Therefore, change movl to movw. No change to machine code generated.
1.14.2.1	07-Jul-2008	gmcgarry	file frameasm.h was added on branch haad-dm on 2008-07-07 13:01:17 +0000
1.15.46.1	01-Aug-2017	snj	Pull up following revision(s) (requested by maxv in ticket #164): distrib/sets/lists/base/md.amd64: revision 1.269 distrib/sets/lists/debug/md.amd64: revision 1.97 sys/arch/amd64/conf/GENERIC: revision 1.460 sys/arch/amd64/conf/files.amd64: revision 1.89 sys/arch/i386/conf/GENERIC: revision 1.1157 sys/arch/i386/conf/files.i386: revision 1.379 sys/arch/i386/i386/i386_trap.S: revision 1.7-1.8 sys/arch/i386/include/frameasm.h: revision 1.16 sys/arch/x86/include/sysarch.h: revision 1.12 sys/arch/x86/x86/pmc.c: revision 1.8-1.10 sys/arch/x86/x86/sys_machdep.c: revision 1.36 sys/arch/xen/conf/files.compat: revision 1.26 sys/secmodel/suser/secmodel_suser.c: revision 1.43 sys/sys/kauth.h: revision 1.74 usr.bin/pmc/Makefile: revision 1.5 usr.bin/pmc/pmc.1: revision 1.12-1.13 usr.bin/pmc/pmc.c: revision 1.24-1.25 style -- style -- Disable interrupts for T_NMI (inline calltrap). Note that there's still a way to evade the NMI mode here, if a segment register faults in INTRFASTEXIT; but we don't care. I didn't test this change, but it seems fine enough. -- Make the PMC syscalls privileged. -- Check argc, and add a message. -- include opt_pmc.h -- Build the pmc tool on amd64. -- Properly handle overflows, and take them into account in userland. -- Update. -- Enable PMCs by default. -- Sort sections. Fix macro usage.
1.15.30.1	28-Aug-2017	skrll	Sync with HEAD
1.15.12.1	03-Dec-2017	jdolecek	update from HEAD
1.25.2.1	25-Jun-2018	pgoyette	Sync with HEAD
1.26.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.26.2.1	10-Jun-2019	christos	Sync with HEAD
1.29.6.1	11-Apr-2020	bouyer	Include ci_isources[] for XenPV too. Adjust spllower() to XenPV needs, and switch XenPV to the native spllower(). Remove xen_spllower().
1.15	08-Aug-2017	maxv	Move freebsd_machdep.h into sys/compat/freebsd, and don't install it. Now, the compat_freebsd files are all contained in sys/compat/freebsd.
1.14	29-Jul-2017	maxv	Remove unused.
1.13	12-Feb-2014	dsl	branches: 1.13.6; Change i386 to use x86/fpu.c instead of i386/isa/npx.c This changes the trap10 and trap13 code to call directly into fpu.c, removing all the code for T_ARITHTRAP, T_XMM and T_FPUNDA from i386/trap.c Not all of the code thate appeared to handle fpu traps was ever called! Most of the changes just replace the include of machine/npx.h with x86/fpu.h (or remove it entirely).
1.12	19-Jan-2014	dsl	Rip out the code that might once have supported the freebsd ptrace system calls. Removes some very dodgy conversions of FP register layouts. Most of it has been disabled since 2007 when I removed the stackgap.
1.11	09-Dec-2013	dsl	Include machine/npx.h here. The only file that needs 'struct freebsd_save87' includes it already.
1.10	08-Dec-2013	dsl	Rename the fields of 'struct envxmm' (itself badly named) so that they are different from those used for the x87 fpu. Use a union to describe the 64bit 'address' field so that it correctly spilt for 32bit binaries. The freebsd emulation defined its own 'struct freebsd_env87' that the code just assumed matched the netbsd one. Since this is a hardware structure, just use the netbsd one. This shouldn't change the binaries except that 'en_fos' used to be incorrectly a 32bit field.
1.9	14-Sep-2005	he	branches: 1.9.118; 1.9.128; 1.9.134; Need <compat/sys/sigtypes.h> here for sigset13_t.
1.8	27-Oct-2003	junyoung	branches: 1.8.16; Nuke __P().
1.7	06-Sep-2003	christos	SA_SIGINFO changes.
1.6	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.5	04-Jul-2002	thorpej	branches: 1.5.2; 1.5.8; Add kernel support for having userland provide the signal trampoline: * struct sigacts gets a new sigact_sigdesc structure, which has the sigaction and the trampoline/version. Version 0 means "legacy kernel provided trampoline". Other versions are coordinated with machine- dependent code in libc. * sigaction1() grows two more arguments -- the trampoline pointer and the trampoline version. * A new __sigaction_sigtramp() system call is provided to register a trampoline along with a signal handler. * The handler is no longer passed to sensig() functions. Instead, sendsig() looks up the handler by peeking in the sigacts for the process getting the signal (since it has to look in there for the trampoline anyway). * Native sendsig() functions now select the appropriate trampoline and its arguments based on the trampoline version in the sigacts. Changes to libc to use the new facility will be checked in later. Kernel version not bumped; we will ride the 1.6C bump made recently.
1.4	27-Oct-2001	jdolecek	branches: 1.4.10; Update freebsd_sigcontext to what FreeBSD currently uses, and save/restore %fs/%gs as appropriate. XXX Note that the new sigcontext uses the new sigset_t instead of old int sc_mask. The new FreeBSD sigcontext doesn't contain any backward compatibility sc_mask, so basically old FreeBSD application making use of sigcontext are hosed. This can't be fixed in NetBSD compat code.
1.3	18-Dec-2000	thorpej	branches: 1.3.4; 1.3.6; 1.3.10; Prototype freebsd_syscall_intern() in freebsd_machdep.h so that freebsd_exec.c has access to it.
1.2	11-Sep-1998	mycroft	branches: 1.2.12; Substantial signal handling changes: * Increase the size of sigset_t to accomodate 128 signals -- adding new versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() to handle the changed arguments. * Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() into separate functions, and call them from all the emulations rather than hard-coding everything. (Avoids uses the stackgap crap for these system calls.) * Add a new flag (p_checksig) to indicate that a process may have signals pending and userret() needs to do the full (slow) check. * Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE. * Correct emulation bugs with restoring SS_ONSTACK. * Make the signal mask in the sigcontext always use the emulated mask format. * Store signals internally in sigaction structures, rather than maintaining a bunch of little sigsets for each SA_* bit. * Keep track of where we put the signal trampoline, rather than figuring it out in _sendsig(). Issue a warning when a non-emulated sigaction bit is observed. * Add missing emulated signals, and a native SIGPWR (currently not used). * Implement the `not reset when caught' semantics for relevant signals. Note: Only code touched by the i386 port has been modified. Other ports and emulations need to be updated.
1.1	10-Oct-1995	mycroft	FreeBSD binary compatibility module, from Noriyuki Soda.
1.2.12.1	05-Jan-2001	bouyer	Sync with HEAD
1.3.10.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.3.6.2	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.3.6.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.3.4.2	01-Aug-2002	nathanw	Catch up to -current.
1.3.4.1	14-Nov-2001	nathanw	Catch up to -current.
1.4.10.1	16-Jul-2002	gehenna	catch up with -current.
1.5.8.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.5.8.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.5.8.2	18-Sep-2004	skrll	Sync with HEAD.
1.5.8.1	03-Aug-2004	skrll	Sync with HEAD
1.5.2.2	04-Jul-2002	thorpej	Add kernel support for having userland provide the signal trampoline: * struct sigacts gets a new sigact_sigdesc structure, which has the sigaction and the trampoline/version. Version 0 means "legacy kernel provided trampoline". Other versions are coordinated with machine- dependent code in libc. * sigaction1() grows two more arguments -- the trampoline pointer and the trampoline version. * A new __sigaction_sigtramp() system call is provided to register a trampoline along with a signal handler. * The handler is no longer passed to sensig() functions. Instead, sendsig() looks up the handler by peeking in the sigacts for the process getting the signal (since it has to look in there for the trampoline anyway). * Native sendsig() functions now select the appropriate trampoline and its arguments based on the trampoline version in the sigacts. Changes to libc to use the new facility will be checked in later. Kernel version not bumped; we will ride the 1.6C bump made recently.
1.5.2.1	04-Jul-2002	thorpej	file freebsd_machdep.h was added on branch sommerfeld_i386mp_1 on 2002-07-04 23:32:06 +0000
1.8.16.1	21-Jun-2006	yamt	sync with head.
1.9.134.1	18-May-2014	rmind	sync with head
1.9.128.2	03-Dec-2017	jdolecek	update from HEAD
1.9.128.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.9.118.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.13.6.1	28-Aug-2017	skrll	Sync with HEAD
1.20	30-Apr-2021	christos	Merge the x86 gdt function and constant definitions
1.19	30-Apr-2021	christos	Bump MAX_USERLDT_SIZE to the max size (wastes some memory). wine needs more than PAGE_SIZE and fails spuriously. XXX: Note the duplicate definition hacks. Should really create <x86/gdt.h>, put the just the constants there and unify them. This would also avoid the hack in: src/tests/lib/libi386/t_user_ldt.c#46
1.18	25-Apr-2020	bouyer	branches: 1.18.6; Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.17	24-Apr-2020	maxv	Give the ldt a fixed size of one page (512 slots), and drop the variable- sized mechanism that was too complex. This fixes a race between USER_LDT and SVS: during context switches, the way SVS installs the new ldt relies on the ldt pointer AND the ldt size, but both cannot be accessed atomically at the same time.
1.16	02-Jul-2017	maxv	branches: 1.16.16; Define MINGDTSIZ/MAXGDTSIZ in bytes, not in number of slots; same as amd64.
1.15	08-Feb-2017	maxv	Remove gdt_reload_cpu. GDTR takes a VA as base, and in our x86 implementation this VA is per-cpu and does not change; there is therefore no need to remotely reload GDTR.
1.14	20-Aug-2016	maxv	branches: 1.14.2; Localify.
1.13	28-Apr-2008	martin	branches: 1.13.44; 1.13.64; 1.13.68; Remove clause 3 and 4 from TNF licenses
1.12	04-Jan-2008	yamt	branches: 1.12.6; 1.12.8; 1.12.10; i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.11	16-Apr-2007	ad	branches: 1.11.12; 1.11.18; 1.11.24; Fix crappy locking around LDT updates. i386 part of PR kern/36046.
1.10	27-Oct-2003	junyoung	branches: 1.10.16; 1.10.54; 1.10.58; 1.10.60; Nuke __P().
1.9	01-Oct-2002	fvdl	branches: 1.9.6; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.8	16-Aug-2000	thorpej	branches: 1.8.2; 1.8.4; Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.7	12-May-1999	thorpej	branches: 1.7.2; 1.7.10; 1.7.14; Move the user-set LDT out of the PCB and into the pmap. Applications which set the LDT and share VM space (e.g. new versions of WINE) expect the LDT to be logically coupled to the address space. Use the new pmap_fork() interface to copy non-shared user-set LDTs when the address space is forked.
1.6	13-Nov-1997	mycroft	branches: 1.6.10; Some general cleanup: * Call gdt_init() earlier on. * Don't bother to check for TSS and LDT descriptors in verr_gdt(). * Nuke the sti special case for trace and breakpoint traps. * Move some variable declarations into more appropriate locations.
1.5	09-Oct-1997	jtc	branches: 1.5.2; Fix tipo inherited from old version of TNF copyright template.
1.4	25-Aug-1996	jtk	branches: 1.4.10; add Advanced Power Management driver Includes bootstrap support from FreeBSD (Tatsumi HOSOKAWA and <ukai>)
1.3	27-Feb-1996	jtc	Copyright assigned to the NetBSD Foundation
1.2	18-Feb-1996	mycroft	Add the obligatory copyright notice.
1.1	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.4.10.1	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.5.2.1	13-Nov-1997	mellon	Pull rev 1.6 up from trunk (mycroft)
1.6.10.1	21-Jun-1999	thorpej	Sync w/ -current.
1.7.14.1	16-Aug-2000	thorpej	Update from trunk: Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.7.10.6	19-Aug-2002	sommerfeld	merged with -current as of today. rearranged gdt init to deal with KVM86 changes; however, KVM86 has no chance of working yet; good thing it's not part of GENERIC
1.7.10.5	24-Feb-2002	sommerfeld	Initialize gdt on boot cpu earlier in boot instead of in gdt_init(). Put apm back in GENERIC and APM_NO_IDLE in GENERIC.MP. Subsystems which install gdt entries during autoconf no longer explode. (Do not expect full apm functionality on multiprocessors).
1.7.10.4	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.7.10.3	07-Jan-2001	sommerfeld	Correct merge botches. Post-merge kernel now boots multiuser.
1.7.10.2	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.7.10.1	12-Aug-2000	sommerfeld	Split creation/allocation of the memory containing the GDT from loading of the GDT descriptor into the processor; the former happens once, the latter, once per CPU.
1.7.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.8.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.8.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.8.2.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.9.6.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.9.6.2	18-Sep-2004	skrll	Sync with HEAD.
1.9.6.1	03-Aug-2004	skrll	Sync with HEAD
1.10.60.1	11-Jul-2007	mjf	Sync with head.
1.10.58.1	27-May-2007	ad	Sync with head.
1.10.54.1	07-May-2007	yamt	sync with head.
1.10.16.2	21-Jan-2008	yamt	sync with head
1.10.16.1	03-Sep-2007	yamt	sync with head.
1.11.24.1	08-Jan-2008	bouyer	Sync with HEAD
1.11.18.1	18-Feb-2008	mjf	Sync with HEAD.
1.11.12.1	09-Jan-2008	matt	sync with HEAD
1.12.10.1	16-May-2008	yamt	sync with head.
1.12.8.1	18-May-2008	yamt	sync with head.
1.12.6.1	02-Jun-2008	mjf	Sync with HEAD.
1.13.68.1	20-Mar-2017	pgoyette	Sync with HEAD
1.13.64.2	28-Aug-2017	skrll	Sync with HEAD
1.13.64.1	05-Oct-2016	skrll	Sync with HEAD
1.13.44.1	03-Dec-2017	jdolecek	update from HEAD
1.14.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.16.16.1	25-Apr-2020	bouyer	Sync with bouyer-xenpvh-base2 (HEAD)
1.18.6.1	13-May-2021	thorpej	Sync with HEAD.
1.11	13-Nov-2017	nakayama	Don't write a 1 to the read only RIRR bit in the IOAPIC redirection register to fix "tlp0: filter setup and transmit timeout" observed on Hyper-V VMs with the Legacy Network Adapter. From OpenBSD via PR kern/49323: https://marc.info/?l=openbsd-cvs&m=146718035432599&w=2 \| Modified files: \| sys/arch/amd64/amd64: ioapic.c \| sys/arch/amd64/include: i82093reg.h \| \| Log message: \| Don't write a 1 to the RIRR bit in the IOAPIC redirection register. This bit \| is R/O, and although it should not matter what value is written there, \| Hyper-V's emulated IOAPIC interprets a write of 1 in some unexpected way and \| subsequently blocks interrupt delivery. This primarily manifests itself as \| de(4) timeouts when using Hyper-V VMs with the "Legacy Network Adapter" \| interface. Tested both amd64 and i386 on Client Hyper-V on Windows 10.
1.10	23-May-2017	nonaka	branches: 1.10.2; x86: Add preliminary x2APIC support. x2APIC is used only when x2APIC is enabled in BIOS/UEFI. LAPIC ID is not supported above 256.
1.9	25-Nov-2016	maxv	Move the virtual address of the LAPIC page out of the data segment on amd64 and i386. The old design was error-prone, and it didn't allow us to map the data segment with large pages. Now, the VA is allocated dynamically in the pmap bootstrap code, and entered manually later. We go from using &local_apic to using *local_apic_va, and we therefore need one more level of indirection in the asm code. Discussed on tech-kern.
1.8	03-Jul-2008	drochner	branches: 1.8.40; 1.8.58; 1.8.60; 1.8.62; 1.8.64; 1.8.68; Remove "struct device" from "struct pic", where it was only real for ioapics and faked up for others. Add it to "struct ioapic_softc" for now, until device/softc get split. This required all typecasts between "struct pic" and "struct ioapic_softc" to be replaced, I hope I got them all. functionally tested on i386, compile-tested on xen, untested on amd64
1.7	08-Dec-2006	yamt	branches: 1.7.44; 1.7.48; 1.7.50; 1.7.52; - pass intrframe by-pointer, not by-value. - make i386 and xen use per-cpu interrupt stack. xen part is reviewed by Manuel Bouyer.
1.6	11-May-2003	fvdl	branches: 1.6.18; 1.6.48; 1.6.50; The i82093 spec is unclear about it, but my assumption that the selected ioapic register would stick across a read+write appears to be invalid for some ioapics. So, select it again for the mask write.
1.5	04-May-2003	fvdl	Block level-triggered interrupts at the ioapic if they are deferred. Avoids interrupt storms seen on some systems. Many thanks to Stoned Elipot for testing.
1.4	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.3	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file i82093reg.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	MP: Registers and state definitions for I/O APIC, which routes interrupts from I/O devices onto the APIC bus.
1.2.2.3	11-Dec-2002	thorpej	Sync with HEAD.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file i82093reg.h was added on branch nathanw_sa on 2002-10-18 02:37:55 +0000
1.6.50.1	10-Dec-2006	yamt	sync with head.
1.6.48.1	12-Jan-2007	ad	Sync with head.
1.6.18.1	30-Dec-2006	yamt	sync with head.
1.7.52.1	03-Jul-2008	simonb	Sync with head.
1.7.50.1	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.7.48.1	04-May-2009	yamt	sync with head.
1.7.44.1	28-Sep-2008	mjf	Sync with HEAD.
1.8.68.1	03-Jan-2018	snj	Pull up following revision(s) (requested by nakayama in ticket #1527): sys/arch/amd64/include/i82093reg.h: revision 1.9 sys/arch/i386/include/i82093reg.h: revision 1.11 sys/arch/x86/x86/ioapic.c: revision 1.54 Don't write a 1 to the read only RIRR bit in the IOAPIC redirection register to fix "tlp0: filter setup and transmit timeout" observed on Hyper-V VMs with the Legacy Network Adapter. From OpenBSD via PR kern/49323: https://marc.info/?l=openbsd-cvs&m=146718035432599&w=2
1.8.64.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.8.62.1	03-Jan-2018	snj	Pull up following revision(s) (requested by nakayama in ticket #1527): sys/arch/amd64/include/i82093reg.h: revision 1.9 sys/arch/i386/include/i82093reg.h: revision 1.11 sys/arch/x86/x86/ioapic.c: revision 1.54 Don't write a 1 to the read only RIRR bit in the IOAPIC redirection register to fix "tlp0: filter setup and transmit timeout" observed on Hyper-V VMs with the Legacy Network Adapter. From OpenBSD via PR kern/49323: https://marc.info/?l=openbsd-cvs&m=146718035432599&w=2
1.8.60.2	28-Aug-2017	skrll	Sync with HEAD
1.8.60.1	05-Dec-2016	skrll	Sync with HEAD
1.8.58.1	03-Jan-2018	snj	Pull up following revision(s) (requested by nakayama in ticket #1527): sys/arch/amd64/include/i82093reg.h: revision 1.9 sys/arch/i386/include/i82093reg.h: revision 1.11 sys/arch/x86/x86/ioapic.c: revision 1.54 Don't write a 1 to the read only RIRR bit in the IOAPIC redirection register to fix "tlp0: filter setup and transmit timeout" observed on Hyper-V VMs with the Legacy Network Adapter. From OpenBSD via PR kern/49323: https://marc.info/?l=openbsd-cvs&m=146718035432599&w=2
1.8.40.1	03-Dec-2017	jdolecek	update from HEAD
1.10.2.1	21-Nov-2017	martin	Pull up following revision(s) (requested by nakayama in ticket #359): sys/arch/amd64/include/i82093reg.h: revision 1.9 sys/arch/x86/x86/ioapic.c: revision 1.54 sys/arch/i386/include/i82093reg.h: revision 1.11 Don't write a 1 to the read only RIRR bit in the IOAPIC redirection register to fix "tlp0: filter setup and transmit timeout" observed on Hyper-V VMs with the Legacy Network Adapter. From OpenBSD via PR kern/49323: https://marc.info/?l=openbsd-cvs&m=146718035432599&w=2 Modified files: sys/arch/amd64/amd64: ioapic.c sys/arch/amd64/include: i82093reg.h Log message: Don't write a 1 to the RIRR bit in the IOAPIC redirection register. This bit is R/O, and although it should not matter what value is written there, Hyper-V's emulated IOAPIC interprets a write of 1 in some unexpected way and subsequently blocks interrupt delivery. This primarily manifests itself as de(4) timeouts when using Hyper-V VMs with the "Legacy Network Adapter" interface. Tested both amd64 and i386 on Client Hyper-V on Windows 10.
1.7	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.6	13-Jan-2003	fvdl	Declare nioapics as well.
1.5	12-Jan-2003	fvdl	Export the ioapic softc linked list, for the benefit of the mpbios code.
1.4	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.3	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file i82093var.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.5	19-Nov-2000	sommerfeld	Notice if multiple handlers with different IPL's share a single interrupt pin. For now, warn if this is the case; in the future, we'll handle this correctly.
1.1.2.4	21-Sep-2000	sommerfeld	Deal more sanely with ioapic softc lookup in an attempt to deal with systems which use weird id's for the ioapics. - Use a linked list rather a fixed-size array indexed by ioapic id. On all MP systems I've found so far, we only have one, and we only need to look for them by id when establishing/disestablishing interrupt handlers; this is not a performance-critical path.. - similarly, use ioapic_find() in mpbios.c rather than reaching into ioapic.c's data. While we're here, support single-apic systems which have interrupts wired to "all ioapics".
1.1.2.3	27-Feb-2000	sommerfeld	multiple-include protection. flush bogons. add kludge to let ioapic know who the boot cpu is.
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	MP: Registers and state definitions for I/O APIC, which routes interrupts from I/O devices onto the APIC bus.
1.2.2.5	15-Jan-2003	thorpej	Sync with HEAD.
1.2.2.4	07-Jan-2003	thorpej	Sync with HEAD.
1.2.2.3	11-Dec-2002	thorpej	Sync with HEAD.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file i82093var.h was added on branch nathanw_sa on 2002-10-18 02:37:55 +0000
1.3	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file i82489reg.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.1	20-Feb-2000	sommerfeld	MP: registers and variable definitions for the Local APIC, which sits between the APIC bus and the processor, and does a bunch of useful things in an MP system: - accepts APIC bus interrupt requests and passes the interrupt vector to the processor - manages the processor's interrupt priority. - initiates priority-based software interrupts directed at this or other processors it also includes a local countdown timer and a few other functions.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file i82489reg.h was added on branch nathanw_sa on 2002-10-18 02:37:55 +0000
1.5	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.4	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.3	01-Nov-2002	fvdl	Make splsched == splclock, and splserial > splsched. Unfortunately, we have to retain the 'pick unlocked interrupt handler' hack for the MP case for now.
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file i82489var.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.5	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.1.2.4	30-Apr-2001	sommerfeld	Do softints purely in software. Using the local apic for softints is slow; exactly why isn't clear, but ripping out the code and using pure-software softints speeds up ping -f packet rates by more than 5%.
1.1.2.3	12-Aug-2000	sommerfeld	Move lapic_initclocks() prototype to header file.
1.1.2.2	21-Feb-2000	sommerfeld	MP: attempt to deal with the case where the BSP is at apic id 1 instead of apic id 0.
1.1.2.1	20-Feb-2000	sommerfeld	MP: registers and variable definitions for the Local APIC, which sits between the APIC bus and the processor, and does a bunch of useful things in an MP system: - accepts APIC bus interrupt requests and passes the interrupt vector to the processor - manages the processor's interrupt priority. - initiates priority-based software interrupts directed at this or other processors it also includes a local countdown timer and a few other functions.
1.2.2.4	11-Dec-2002	thorpej	Sync with HEAD.
1.2.2.3	11-Nov-2002	nathanw	Catch up to -current
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file i82489var.h was added on branch nathanw_sa on 2002-10-18 02:37:56 +0000
1.2	02-Mar-2003	fvdl	Redirect to x86/
1.1	22-Nov-2002	fvdl	branches: 1.1.2; New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.1.2.2	11-Dec-2002	thorpej	Sync with HEAD.
1.1.2.1	22-Nov-2002	thorpej	file i8259.h was added on branch nathanw_sa on 2002-12-11 06:01:00 +0000
1.18	09-Aug-2017	maxv	Remove ibcs2_machdep.h on i386, and don't install it.
1.17	10-Dec-2009	matt	branches: 1.17.22; 1.17.40; Change u_long to vaddr_t/vsize_t in exec code where appropriate (mostly involves setregs and vmcmds). Should result in no code differences.
1.16	28-Apr-2008	martin	branches: 1.16.14; Remove clause 3 and 4 from TNF licenses
1.15	20-Dec-2007	dsl	branches: 1.15.6; 1.15.8; 1.15.10; Convert all the system call entry points from: int foo(struct lwp l, void v, register_t retval) to: int foo(struct lwp l, const struct foo_args uap, register_t retval) Fixup compat code to not write into 'uap' and (in some cases) to actually pass a correctly formatted 'uap' structure with the right name to the next routine. A few 'compat' routines that just call standard ones have been deleted. All the 'compat' code compiles (along with the kernels required to test build it). 98% done by automated scripts.
1.14	27-Oct-2003	junyoung	branches: 1.14.16; 1.14.78; 1.14.88; 1.14.92; Nuke __P().
1.13	06-Sep-2003	christos	SA_SIGINFO changes.
1.12	17-Jan-2003	thorpej	branches: 1.12.2; Merge the nathanw_sa branch.
1.11	04-Jul-2002	thorpej	branches: 1.11.2; Add kernel support for having userland provide the signal trampoline: * struct sigacts gets a new sigact_sigdesc structure, which has the sigaction and the trampoline/version. Version 0 means "legacy kernel provided trampoline". Other versions are coordinated with machine- dependent code in libc. * sigaction1() grows two more arguments -- the trampoline pointer and the trampoline version. * A new __sigaction_sigtramp() system call is provided to register a trampoline along with a signal handler. * The handler is no longer passed to sensig() functions. Instead, sendsig() looks up the handler by peeking in the sigacts for the process getting the signal (since it has to look in there for the trampoline anyway). * Native sendsig() functions now select the appropriate trampoline and its arguments based on the trampoline version in the sigacts. Changes to libc to use the new facility will be checked in later. Kernel version not bumped; we will ride the 1.6C bump made recently.
1.10	11-Dec-2000	mycroft	branches: 1.10.4; 1.10.6; 1.10.18; Introduce 2 new flags in types.h: * __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by e_syscall_intern, which is called at key places in the kernel. This can be used to set a MD syscall handler pointer. This obsoletes and replaces the _HAS_SEPARATED_SYSCALL flags. __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in struct emul are omitted.
1.9	09-Dec-2000	jdolecek	prototype ibcs2_syscall() in compat/ibcs2/ibcs2_exec.c
1.8	02-Dec-2000	jdolecek	add prototype for ibcs2_syscall()
1.7	02-Dec-2000	jdolecek	if IBCS2_MACHDEP_HAS_SEPARATED_SYSCALL is defined, put ibcs2_syscall to emul_ibcs2 define IBCS2_MACHDEP_HAS_SEPARATED_SYSCALL in i386 ibcs2_machdep.h
1.6	21-Jun-2000	matt	Move COFF_LDPGSZ to ibcs2_machdep.h. Fix nmagic loader to deal with subpage mappings.
1.5	10-Jan-2000	matt	branches: 1.5.6; Add some i386 defines.
1.4	11-Sep-1998	mycroft	branches: 1.4.12; Substantial signal handling changes: * Increase the size of sigset_t to accomodate 128 signals -- adding new versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() to handle the changed arguments. * Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() into separate functions, and call them from all the emulations rather than hard-coding everything. (Avoids uses the stackgap crap for these system calls.) * Add a new flag (p_checksig) to indicate that a process may have signals pending and userret() needs to do the full (slow) check. * Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE. * Correct emulation bugs with restoring SS_ONSTACK. * Make the signal mask in the sigcontext always use the emulated mask format. * Store signals internally in sigaction structures, rather than maintaining a bunch of little sigsets for each SA_* bit. * Keep track of where we put the signal trampoline, rather than figuring it out in _sendsig(). Issue a warning when a non-emulated sigaction bit is observed. * Add missing emulated signals, and a native SIGPWR (currently not used). * Implement the `not reset when caught' semantics for relevant signals. Note: Only code touched by the i386 port has been modified. Other ports and emulations need to be updated.
1.3	05-Mar-1998	scottb	Added prototype for ibcs2_sendsig.
1.2	09-Oct-1997	jtc	branches: 1.2.2; Fix tipo inherited from old version of TNF copyright template.
1.1	11-Sep-1997	mycroft	branches: 1.1.2; Add an ibcs2_setregs().
1.1.2.3	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.1.2.2	16-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.1.2.1	11-Sep-1997	thorpej	file ibcs2_machdep.h was added on branch marc-pcmcia on 1997-09-16 03:48:40 +0000
1.2.2.1	05-May-1998	mycroft	Pull up 1.3, per request of scottb.
1.4.12.3	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.4.12.2	08-Dec-2000	bouyer	Sync with HEAD.
1.4.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.5.6.1	22-Jun-2000	matt	Move COFF_LDPGSZ to machine dependent includes.
1.10.18.1	16-Jul-2002	gehenna	catch up with -current.
1.10.6.1	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.10.4.2	01-Aug-2002	nathanw	Catch up to -current.
1.10.4.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.11.2.2	04-Jul-2002	thorpej	Add kernel support for having userland provide the signal trampoline: * struct sigacts gets a new sigact_sigdesc structure, which has the sigaction and the trampoline/version. Version 0 means "legacy kernel provided trampoline". Other versions are coordinated with machine- dependent code in libc. * sigaction1() grows two more arguments -- the trampoline pointer and the trampoline version. * A new __sigaction_sigtramp() system call is provided to register a trampoline along with a signal handler. * The handler is no longer passed to sensig() functions. Instead, sendsig() looks up the handler by peeking in the sigacts for the process getting the signal (since it has to look in there for the trampoline anyway). * Native sendsig() functions now select the appropriate trampoline and its arguments based on the trampoline version in the sigacts. Changes to libc to use the new facility will be checked in later. Kernel version not bumped; we will ride the 1.6C bump made recently.
1.11.2.1	04-Jul-2002	thorpej	file ibcs2_machdep.h was added on branch sommerfeld_i386mp_1 on 2002-07-04 23:32:06 +0000
1.12.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.12.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.12.2.1	03-Aug-2004	skrll	Sync with HEAD
1.14.92.1	02-Jan-2008	bouyer	Sync with HEAD
1.14.88.1	26-Dec-2007	ad	Sync with head.
1.14.78.1	09-Jan-2008	matt	sync with HEAD
1.14.16.1	21-Jan-2008	yamt	sync with head
1.15.10.2	11-Mar-2010	yamt	sync with head
1.15.10.1	16-May-2008	yamt	sync with head.
1.15.8.1	18-May-2008	yamt	sync with head.
1.15.6.1	02-Jun-2008	mjf	Sync with HEAD.
1.16.14.1	24-Oct-2010	jym	Sync with HEAD
1.17.40.1	28-Aug-2017	skrll	Sync with HEAD
1.17.22.1	03-Dec-2017	jdolecek	update from HEAD
1.2	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.1	10-Mar-1999	mycroft	Add an ieee.h.
1.3	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.2	05-Jan-1998	perry	RCSID Police.
1.1	09-Apr-1995	mycroft	Move fp_except and fp_rnd definitions to machine/include.
1.4	25-Jul-2014	joerg	Add generic versions of machine/int_*.h for compilers providing appropiate macros for all necessary types.
1.3	29-May-2010	tnozaki	branches: 1.3.18; 1.3.32; fix wrong integer promotion rule(removed U suffix from UINT{8,16}_C). see ISO/IEC 9899:1999 7.18.4.3.
1.2	28-Apr-2008	martin	branches: 1.2.14; 1.2.20; 1.2.22; Remove clause 3 and 4 from TNF licenses
1.1	14-Apr-2001	kleink	branches: 1.1.2; 1.1.4; 1.1.6; 1.1.8; 1.1.132; 1.1.134; 1.1.136; Add definitions of C99 integer constant macros.
1.1.136.2	11-Aug-2010	yamt	sync with head.
1.1.136.1	16-May-2008	yamt	sync with head.
1.1.134.1	18-May-2008	yamt	sync with head.
1.1.132.1	02-Jun-2008	mjf	Sync with HEAD.
1.1.8.2	21-Jun-2001	nathanw	Catch up to -current.
1.1.8.1	14-Apr-2001	nathanw	file int_const.h was added on branch nathanw_sa on 2001-06-21 19:25:48 +0000
1.1.6.2	14-Apr-2001	kleink	Add definitions of C99 integer constant macros.
1.1.6.1	14-Apr-2001	kleink	file int_const.h was added on branch sommerfeld_i386mp_1 on 2001-04-14 22:38:39 +0000
1.1.4.1	21-Apr-2001	thorpej	Add files to sommerfeld_i386mp_1 branch that have recently appeared on the trunk.
1.1.2.2	21-Apr-2001	bouyer	Sync with HEAD
1.1.2.1	14-Apr-2001	bouyer	file int_const.h was added on branch thorpej_scsipi on 2001-04-21 17:53:52 +0000
1.2.22.1	30-May-2010	rmind	sync with head
1.2.20.1	17-Aug-2010	uebayasi	Sync with HEAD.
1.2.14.1	24-Oct-2010	jym	Sync with HEAD
1.3.32.1	10-Aug-2014	tls	Rebase.
1.3.18.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.9	28-Jul-2014	joerg	GCC sets up u?int_fast8_t to be int, so be consistent with it.
1.8	25-Jul-2014	joerg	Add generic versions of machine/int_*.h for compilers providing appropiate macros for all necessary types.
1.7	28-Apr-2008	martin	branches: 1.7.44; 1.7.60; Remove clause 3 and 4 from TNF licenses
1.6	22-May-2004	kleink	branches: 1.6.96; 1.6.98; 1.6.100; Bring back {u,}int_fast8_t back to 8 bits, as I had misunderstood Charles.
1.5	21-May-2004	kleink	Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum- width implementation was a rather poor choice. Per discussion with Charles Hannum. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.4	30-May-2003	kleink	branches: 1.4.2; 1.4.4; {u,}intptr_t are ints again.
1.3	19-Sep-2001	simonb	branches: 1.3.2; {,u}intptr_t are longs; adjust printf/scanf formats.
1.2	26-Apr-2001	kleink	branches: 1.2.4; 1.2.6; 1.2.8; Add definitions for C99 fastest minimum-width integer types.
1.1	15-Apr-2001	kleink	branches: 1.1.2; 1.1.4; Add definitions of C99 integer format conversion macros. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.1.4.1	21-Apr-2001	thorpej	Add files to sommerfeld_i386mp_1 branch that have recently appeared on the trunk.
1.1.2.2	21-Apr-2001	bouyer	Sync with HEAD
1.1.2.1	15-Apr-2001	bouyer	file int_fmtio.h was added on branch thorpej_scsipi on 2001-04-21 17:53:52 +0000
1.2.8.1	01-Oct-2001	fvdl	Catch up with -current.
1.2.6.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.2.4.3	21-Sep-2001	nathanw	Catch up to -current.
1.2.4.2	21-Jun-2001	nathanw	Catch up to -current.
1.2.4.1	26-Apr-2001	nathanw	file int_fmtio.h was added on branch nathanw_sa on 2001-06-21 19:25:48 +0000
1.3.2.2	19-Sep-2001	simonb	{,u}intptr_t are longs; adjust printf/scanf formats.
1.3.2.1	19-Sep-2001	simonb	file int_fmtio.h was added on branch sommerfeld_i386mp_1 on 2001-09-19 05:23:44 +0000
1.4.4.2	22-May-2004	he	Pull up revision 1.6 (requested by kleink in ticket #376): Change {u,}int_fast16_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it. Reverses previous commit for 8-bit types.
1.4.4.1	22-May-2004	he	Pull up revision 1.5 (requested by kleink in ticket #376): Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.4.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.4.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.4.2.1	03-Aug-2004	skrll	Sync with HEAD
1.6.100.1	16-May-2008	yamt	sync with head.
1.6.98.1	18-May-2008	yamt	sync with head.
1.6.96.1	02-Jun-2008	mjf	Sync with HEAD.
1.7.60.1	10-Aug-2014	tls	Rebase.
1.7.44.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.9	25-Jul-2014	joerg	Add generic versions of machine/int_*.h for compilers providing appropiate macros for all necessary types.
1.8	28-Apr-2008	martin	branches: 1.8.44; 1.8.60; Remove clause 3 and 4 from TNF licenses
1.7	17-Oct-2007	garbled	branches: 1.7.16; 1.7.18; 1.7.20; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.6	31-Aug-2007	drochner	Fix definitions of UCHAR_MAX/USHRT_MAX and related types. C99 requires that these definitions promote to (signed/unsigned) integer the same way as the types the definition is for. And since unsigned char/short fit into an "int" on all our archs and thus promote to signed int, the definitions must not be unsigned. Fixes PR lib/31306 by Neil Booth.
1.5	22-May-2004	kleink	branches: 1.5.12; 1.5.52; 1.5.60; 1.5.66; 1.5.70; 1.5.72; Bring back {u,}int_fast8_t back to 8 bits, as I had misunderstood Charles.
1.4	21-May-2004	kleink	Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum- width implementation was a rather poor choice. Per discussion with Charles Hannum. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.3	08-May-2004	kleink	Factor out W{CHAR,INT}_{MAX,MIN} into their own header file.
1.2	26-Apr-2001	kleink	branches: 1.2.2; 1.2.4; 1.2.26; 1.2.28; Add definitions for C99 fastest minimum-width integer types.
1.1	15-Apr-2001	kleink	branches: 1.1.2; 1.1.4; Add definitions of C99 specified-width integer type limits. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.1.4.1	21-Apr-2001	thorpej	Add files to sommerfeld_i386mp_1 branch that have recently appeared on the trunk.
1.1.2.2	21-Apr-2001	bouyer	Sync with HEAD
1.1.2.1	15-Apr-2001	bouyer	file int_limits.h was added on branch thorpej_scsipi on 2001-04-21 17:53:52 +0000
1.2.28.2	22-May-2004	he	Pull up revision 1.5 (requested by kleink in ticket #376): Change {u,}int_fast16_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it. Reverses preivous commit with respect to 8-bit types.
1.2.28.1	22-May-2004	he	Pull up revision 1.4 (requested by kleink in ticket #376): Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.2.26.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.2.26.2	18-Sep-2004	skrll	Sync with HEAD.
1.2.26.1	03-Aug-2004	skrll	Sync with HEAD
1.2.4.2	21-Jun-2001	nathanw	Catch up to -current.
1.2.4.1	26-Apr-2001	nathanw	file int_limits.h was added on branch nathanw_sa on 2001-06-21 19:25:49 +0000
1.2.2.2	26-Apr-2001	kleink	Add definitions for C99 fastest minimum-width integer types.
1.2.2.1	26-Apr-2001	kleink	file int_limits.h was added on branch sommerfeld_i386mp_1 on 2001-04-26 16:25:22 +0000
1.5.72.1	06-Nov-2007	matt	sync with HEAD
1.5.70.1	03-Sep-2007	jmcneill	Sync with HEAD.
1.5.66.1	03-Sep-2007	skrll	Sync with HEAD.
1.5.60.1	03-Oct-2007	garbled	Sync with HEAD
1.5.52.1	09-Oct-2007	ad	Sync with head.
1.5.12.1	03-Sep-2007	yamt	sync with head.
1.7.20.1	16-May-2008	yamt	sync with head.
1.7.18.1	18-May-2008	yamt	sync with head.
1.7.16.1	02-Jun-2008	mjf	Sync with HEAD.
1.8.60.1	10-Aug-2014	tls	Rebase.
1.8.44.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.8	28-Jul-2014	joerg	GCC sets up u?int_fast8_t to be int, so be consistent with it.
1.7	25-Jul-2014	joerg	Add generic versions of machine/int_*.h for compilers providing appropiate macros for all necessary types.
1.6	28-Apr-2008	martin	branches: 1.6.44; 1.6.60; Remove clause 3 and 4 from TNF licenses
1.5	24-Dec-2005	perry	branches: 1.5.74; 1.5.76; 1.5.78; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.4	22-May-2004	kleink	branches: 1.4.12; Bring back {u,}int_fast8_t back to 8 bits, as I had misunderstood Charles.
1.3	21-May-2004	kleink	Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum- width implementation was a rather poor choice. Per discussion with Charles Hannum. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.2	26-Apr-2001	kleink	branches: 1.2.2; 1.2.4; 1.2.26; 1.2.28; Add definitions for C99 fastest minimum-width integer types.
1.1	14-Apr-2001	kleink	branches: 1.1.2; 1.1.4; Add definitions of C99 minimum-width and greatest-width integer types. XXX Fastest minimum-width integer types haven't been decided upon yet.
1.1.4.1	21-Apr-2001	thorpej	Add files to sommerfeld_i386mp_1 branch that have recently appeared on the trunk.
1.1.2.2	21-Apr-2001	bouyer	Sync with HEAD
1.1.2.1	14-Apr-2001	bouyer	file int_mwgwtypes.h was added on branch thorpej_scsipi on 2001-04-21 17:53:52 +0000
1.2.28.2	22-May-2004	he	Pull up revision 1.4 (requested by kleink in ticket #376): Change {u,}int_fast16_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it. Reverses preivous commit with respect to 8-bit types.
1.2.28.1	22-May-2004	he	Pull up revision 1.3 (requested by kleink in ticket #376): Change {u,}int_fast{8,16}_t to 32-bit types, as the previous minimum-width implementation was a rather poor choice. Note: While this is technically an ABI change I believe it is a change that we can afford at this time (and to be pulled up to 2.0). The types are not widely used yet, and a survey of pkgsrc has not shown uses that would be adversely affected by it.
1.2.26.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.2.26.2	18-Sep-2004	skrll	Sync with HEAD.
1.2.26.1	03-Aug-2004	skrll	Sync with HEAD
1.2.4.2	21-Jun-2001	nathanw	Catch up to -current.
1.2.4.1	26-Apr-2001	nathanw	file int_mwgwtypes.h was added on branch nathanw_sa on 2001-06-21 19:25:50 +0000
1.2.2.2	26-Apr-2001	kleink	Add definitions for C99 fastest minimum-width integer types.
1.2.2.1	26-Apr-2001	kleink	file int_mwgwtypes.h was added on branch sommerfeld_i386mp_1 on 2001-04-26 16:25:22 +0000
1.4.12.1	21-Jun-2006	yamt	sync with head.
1.5.78.1	16-May-2008	yamt	sync with head.
1.5.76.1	18-May-2008	yamt	sync with head.
1.5.74.1	02-Jun-2008	mjf	Sync with HEAD.
1.6.60.1	10-Aug-2014	tls	Rebase.
1.6.44.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.11	25-Jul-2014	joerg	Add generic versions of machine/int_*.h for compilers providing appropiate macros for all necessary types.
1.10	24-Dec-2005	perry	branches: 1.10.122; 1.10.138; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.9	25-May-2005	kleink	branches: 1.9.2; Include <sys/cdefs.h> for __signed; related to lib/30072.
1.8	07-Aug-2003	agc	branches: 1.8.6; 1.8.14; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.7	29-May-2003	christos	branches: 1.7.2; Back to unsigned int and int for {u,}intptr_t, otherwise java has no chance...
1.6	28-Apr-2001	kleink	branches: 1.6.2; * Move definitions of exact-width integer types from <machine/types.h> to <sys/types.h> and <sys/stdint.h>. * Add a new C99 <stdint.h> header, which provides integer types of explicit width, related limits and integer constant macros. * Extend <inttypes.h> to provide <stdint.h> definitions and format macros for printf() and scanf(). * Add C99 strtoimax() and strtoumax() functions. * Use the latter within scanf(). * Add C99 %j, %t and %z printf()/scanf() conversions for intmax_t, pointer-type and size_t arguments.
1.5	12-Apr-2001	kleink	Replace the 'unsigned __COMPILER_INT64__' construct with a new name, __COMPILER_UINT64__, to be supplied - if such a case is made, it shouldn't be assumed that the unsigned type-specifier may be applied to it.
1.4	03-Jan-2001	takemura	branches: 1.4.4; replace 'long long' with int64_t to compile stand alone program with compiler other than GCC.
1.3	16-Jul-2000	christos	branches: 1.3.4; change intptr_t and uintptr_t to long and unsigned long respectively. This is a noop on 32 bit machines, but it makes lint quiet and it is more correct.
1.2	27-Jun-2000	kleink	Resolve some formatting nits; add __intptr_t and __uintptr_t.
1.1	26-Jun-2000	kleink	Add <machine/int_types.h>, which provides namespace-pure definitions of exact-width integer types.
1.3.4.4	21-Apr-2001	bouyer	Sync with HEAD
1.3.4.3	05-Jan-2001	bouyer	Sync with HEAD
1.3.4.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.3.4.1	16-Jul-2000	bouyer	file int_types.h was added on branch thorpej_scsipi on 2000-11-20 20:09:27 +0000
1.4.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.6.2.2	28-Apr-2001	kleink	* Move definitions of exact-width integer types from <machine/types.h> to <sys/types.h> and <sys/stdint.h>. * Add a new C99 <stdint.h> header, which provides integer types of explicit width, related limits and integer constant macros. * Extend <inttypes.h> to provide <stdint.h> definitions and format macros for printf() and scanf(). * Add C99 strtoimax() and strtoumax() functions. * Use the latter within scanf(). * Add C99 %j, %t and %z printf()/scanf() conversions for intmax_t, pointer-type and size_t arguments.
1.6.2.1	28-Apr-2001	kleink	file int_types.h was added on branch sommerfeld_i386mp_1 on 2001-04-28 15:41:32 +0000
1.7.2.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.7.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.7.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.7.2.1	03-Aug-2004	skrll	Sync with HEAD
1.8.14.1	28-May-2005	tron	Pull up revision 1.9 (requested by klein in ticket #346): Include <sys/cdefs.h> for __signed; related to lib/30072.
1.8.6.1	29-May-2005	riz	Pull up revision 1.9 (requested by kleink in ticket #1555): Include <sys/cdefs.h> for __signed; related to lib/30072.
1.9.2.1	21-Jun-2006	yamt	sync with head.
1.10.138.1	10-Aug-2014	tls	Rebase.
1.10.122.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.37	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.36	16-Dec-2002	fvdl	Remove leftover declarations used for debugging purposes only.
1.35	11-Dec-2002	fvdl	Since a CPU may spin waiting for another CPU to handle an IPI, an IPI must be the very first thing to be handled in splx(), if it had been blocked before. Change things around a bit to guarantee this.
1.34	01-Dec-2002	fvdl	For softints, there's no intrframe on the stack, so don't try to use it. Use a seperate lock function (in taking the same lock) instead.
1.33	27-Nov-2002	fvdl	Specify the interrupt frame as the argument to i386_int(un)lock.
1.32	23-Nov-2002	fvdl	Remove "ioapic.h" include, and always include the intr_find_mpmapping prototype; it's not worth the extra bytes to avoid including this.
1.31	22-Nov-2002	fvdl	Removed unused function. From kristerw.
1.30	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.29	01-Nov-2002	fvdl	Make splsched == splclock, and splserial > splsched. Unfortunately, we have to retain the 'pick unlocked interrupt handler' hack for the MP case for now.
1.28	05-Oct-2002	fvdl	Adapt i386_send_ipi prototype. Make spllock < splipi (IPI paths changed to use __cpu_simple_lock because of LOCKDEBUG conflict).
1.27	01-Oct-2002	matt	ANSI'fy the inline functions.
1.26	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.25	12-May-2002	matt	Eliminate commons.
1.24	18-Dec-2001	yamt	branches: 1.24.4; use #define<TAB> style. pointed by enami.
1.23	18-Dec-2001	fvdl	Prevent gcc from reordering things such that instructions may fall out of a range which is protected by spl/splx. Originally proposed by YAMAMOTO Takashi (yamt@netbsd.org) on tech-kern. This change mirrors the one done by Bill Sommerfeld on the i386 mp branch.
1.22	16-Jul-2001	thorpej	In i386_softintr_lock(), use splserial() rather than splhigh(), because of splhigh() braindamage on the i386 port. Fixes port-i386/13038 and port-i386/12985.
1.21	21-May-2001	perry	branches: 1.21.2; remove needless externs in front of function prototypes
1.20	30-Apr-2001	lukem	remove some lint, including ansifying some inlines
1.19	14-Apr-2001	thorpej	Generic soft interrupt implementation for NetBSD/i386. This could probably be tuned somewhat, but this is a stop-gap measure to hold us until Bill Sommerfeld's new interrupt code comes in from the MP branch.
1.18	13-Apr-2001	thorpej	Remove the use of splimp() from the NetBSD kernel. splnet() and only splnet() is allowed for the protection of data structures used by network devices.
1.17	14-Jan-2001	thorpej	branches: 1.17.2; Make sure everybody has an splvm() and equate it with splimp() (splimp() is the historical name for this interrupt level, and the historical name is going to go away in the near future).
1.16	22-Aug-2000	thorpej	Add spllock(). See spl(9) for details.
1.15	21-Aug-2000	thorpej	Make sure we provide splsched() as described in spl(9).
1.14	08-Jun-2000	cgd	make spl-lowering functions return void.
1.13	04-Jun-2000	mycroft	Implement CLKF_INTR(), to count interrupt time separately.
1.12	05-Aug-1999	thorpej	branches: 1.12.2; 1.12.10; 1.12.12; Change the semantics of splsoftclock() to be like other spl() functions, that is priority is rasied. Add a new spllowersoftclock() to provide the atomic drop-to-softclock semantics that the old splsoftclock() provided, and update calls accordingly. This fixes a problem with using the "rnd" pseudo-device from within interrupt context to extract random data (e.g. from within the softnet interrupt) where doing so would incorrectly unblock interrupts (causing all sorts of lossage). XXX 4 platforms do not have priority-raising capability: newsmips, sparc, XXX sparc64, and VAX. This platforms still have this bug until their XXX spl() functions are fixed.
1.11	26-Jan-1999	christos	Fix this properly, using gcc assembly syntax (thanks ws!)
1.10	26-Jan-1999	christos	ifdef __ELF__ the softintr inline since we cannot use _C_LABEL to produce a name. XXX: Is there a better way?
1.9	15-Aug-1998	mycroft	Assign my copyrights to TNF.
1.8	18-Jul-1998	is	Switching dev/ic/lpt.c to use spllpt() instead of spltty(). It doesn't use tty structures, and on some machines (namely the DraCo internal lpt, and some multi-i/o boards for Amigas and DraCos), tying spltty to the pretty high printer interupt level would hurt serial performance. On all affected ports but Amiga, spllpt() has been defined in machine/intr.h to be spltty(), thus preserving old behaviour. Portmasters are encouraged to change is, if they feel something else is better (e.g., one of its own were possible).
1.7	21-Mar-1997	mycroft	Add IPL_AUDIO.
1.6	28-Feb-1997	mycroft	Import interrupt priority changes from com patches: * Make it a strict hierarchy. (It was close anyway). * Add `serial' and rename `softtty' to `softserial'. * Make soft interrupts a bit less special-case.
1.5	13-May-1996	mycroft	branches: 1.5.8; Make imask[] larger, to match other changes.
1.4	13-May-1996	mycroft	Add multiple-inclusion protection.
1.3	13-May-1996	mycroft	Move spl*() and soft interrupt definitions into intr.h. For now, include it from psl.h. XXX Will change after release.
1.2	12-May-1996	mycroft	Implement IPL_HIGH.
1.1	12-May-1996	mycroft	Move IPL_* and IST_* constants into intr.h.
1.5.8.1	12-Mar-1997	is	Merge in Changes from The Trunk.
1.12.12.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.12.10.21	25-Jun-2002	fvdl	Add a spinlock around softintr queue management. Use a lock prefix to protect manipulation of the 'ipending' variable. Fixes 'softclock stops being called' AKA 'processes entering nanosleep do not return' problem.
1.12.10.20	18-May-2002	sommerfeld	Catch up with -current. (commons elimination, changes to MI pci framework)
1.12.10.19	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.12.10.18	29-Dec-2001	sommerfeld	Yet another mergeup. New work: - Frank van der Linden's tlb shootdown fix. Mainline functionality merged: - IrDA - Kernel RCSID's - Transmeta CPU support - ACPI - XMM register access through procfs
1.12.10.17	18-Dec-2001	sommerfeld	Recent versions of gcc may reorder instructions around inline functions, breaking the semantics of splraise() and spllower(); to counteract this, insert a "barrier" (empty asm) as a sequence point. Reportedly fixes a stability problems on AMD dual-processor systems. the code-gen barrier primitive will likely move to <sys/cdefs.h> once we figure out what to call it; in the mean time define it locally.
1.12.10.16	19-Jul-2001	sommerfeld	catch up with -current
1.12.10.15	26-May-2001	sommerfeld	Implement MP-synchronzed microtime(), based on Jason Thorpe's port of Dave Mills's cycle-counter-based microtime. XXX We need to figure out how to deal with processors where the cycle counter speed varies. We don't have a whole lot of alternatives; we could only use this on multiprocessors, which are less likely to throttle the clock to save power. The local apic timer runs at a fixed frequency, but it's count-down-to-zero, not the count-up-forever type of timer which the Mills code expects. This also does not solve the problem of microtime on 486-based multiprocessors..
1.12.10.14	23-May-2001	sommerfeld	Merge branch with tonight's current. boots multiuser.
1.12.10.13	30-Apr-2001	sommerfeld	Do softints purely in software. Using the local apic for softints is slow; exactly why isn't clear, but ripping out the code and using pure-software softints speeds up ping -f packet rates by more than 5%.
1.12.10.12	30-Apr-2001	sommerfeld	Catch up on some files missed the last time 'round
1.12.10.11	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.12.10.10	14-Jan-2001	thorpej	Add splvm().
1.12.10.9	31-Dec-2000	thorpej	Fill in the guts of i386_broadcast_ipi() and implement i386_multicast_ipi(). Reviewed by Bill Sommerfeld.
1.12.10.8	23-Sep-2000	sommerfeld	Preliminary support for cloning MTRR values between CPU's at boot time. XXX no API to set MTRR values yet.
1.12.10.7	25-Aug-2000	sommerfeld	Get this to at least build again in the !MULTIPROCESSOR case..
1.12.10.6	25-Aug-2000	sommerfeld	The need_resched() function now takes a (cpu_info *) parameter. Get asts out of <machine/intr.h> Untangle astpending and want_resched variables somewhat.
1.12.10.5	18-Aug-2000	sommerfeld	Add splipi(), IPL_IPI, and rearrange IPI's a bit. Also, add SPL_ASSERT_ATMOST
1.12.10.4	12-Aug-2000	sommerfeld	Add spllock() (temporary) and splsched()
1.12.10.3	26-Jun-2000	sommerfeld	Start making npx.c MP-safe; Interface to npx now specifies the process and/or cpu to act on. Implement an IPI for MP lazy FP save. Tested and working when only one CPU is running; untested in the MP case since we can't run user processes on multiple CPU's yet. Along for the ride: - Change i386_send_ipi to take a "struct cpu_info *" rather than a cpu number. - Turn off the "give me the brain" test IPI's.
1.12.10.2	25-Jun-2000	sommerfeld	Merge up to just-post-1.5 -current
1.12.10.1	20-Feb-2000	sommerfeld	MP: MP-safe and APIC-friendly spl*/splx() implementation. Keep spl state as a level, rather than interrupt mask, since on multiple APIC systems there may be more interrupt lines than can fit in a single 32-bit word. Interrupt levels are now compatible with the APIC priority scheme (high order 4 bits of the IDT vector number).
1.12.2.3	21-Apr-2001	bouyer	Sync with HEAD
1.12.2.2	18-Jan-2001	bouyer	Sync with head (for UBC+NFS fixes, mostly).
1.12.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.17.2.9	19-Dec-2002	thorpej	Sync with HEAD.
1.17.2.8	11-Dec-2002	thorpej	Sync with HEAD.
1.17.2.7	11-Dec-2002	thorpej	Sync with HEAD.
1.17.2.6	11-Nov-2002	nathanw	Catch up to -current
1.17.2.5	18-Oct-2002	nathanw	Catch up to -current.
1.17.2.4	20-Jun-2002	nathanw	Catch up to -current.
1.17.2.3	08-Jan-2002	nathanw	Catch up to -current.
1.17.2.2	24-Aug-2001	nathanw	Catch up with -current.
1.17.2.1	21-Jun-2001	nathanw	Catch up to -current.
1.21.2.4	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.21.2.3	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.21.2.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.21.2.1	03-Aug-2001	lukem	update to -current
1.24.4.2	17-Mar-2002	thorpej	IPL_IMP -> IPL_VM
1.24.4.1	17-Mar-2002	thorpej	Add splraiseipl() for use by kern_mutex.c
1.6	30-Apr-2020	bouyer	Don't #include xen/intrdefs.h is !XEN. Should fix third-party module builds (e.g. virtualbox)
1.5	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.4	10-Aug-2011	cherry	branches: 1.4.64; Include Xen specific definitions.
1.3	26-Feb-2003	fvdl	branches: 1.3.122; 1.3.140; Redirect some includes to x86/foo.h
1.2	11-Dec-2002	fvdl	Since a CPU may spin waiting for another CPU to handle an IPI, an IPI must be the very first thing to be handled in splx(), if it had been blocked before. Change things around a bit to guarantee this.
1.1	22-Nov-2002	fvdl	branches: 1.1.2; Interrupt defines that needed to be in a seperate file to avoid cyclic includes.
1.1.2.3	11-Dec-2002	thorpej	Sync with HEAD.
1.1.2.2	11-Dec-2002	thorpej	Sync with HEAD.
1.1.2.1	22-Nov-2002	thorpej	file intrdefs.h was added on branch nathanw_sa on 2002-12-11 06:01:01 +0000
1.3.140.1	31-Jul-2011	cherry	grow MP support for i386. boots to single user
1.3.122.1	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.4.64.1	16-Apr-2020	bouyer	Avoid overflow of ci_ipi_events[] in the PVHVM case (it's size is XEN_NIPIS but we use x86 IPIs): size XEN_NIPIS only for PV, and CTASSERT that XEN_NIPIS <= X86_NIPI if we ever use Xen IPIs for PVHVM.
1.2	16-Mar-2018	maxv	Remove ipkdb from i386. Also remove unused references in amd64. I already talked about doing that six months ago on port-i386@. Back then it was as general cleanup, but now, with SVS etc, we do actually have good reasons for simplifying the entry points. Ok kamil@. (christos@ was in the conversation too)
1.1	22-Mar-2000	ws	branches: 1.1.2; 1.1.8; 1.1.44; Make IPKDB working again. Add support for i386 debugging and pci-based ne2000 boards.
1.1.44.1	22-Mar-2018	pgoyette	Synch with HEAD, resolve conflicts
1.1.8.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.1.8.1	22-Mar-2000	bouyer	file ipkdb.h was added on branch thorpej_scsipi on 2000-11-20 20:09:27 +0000
1.1.2.2	22-Mar-2000	ws	Make IPKDB working again. Add support for i386 debugging and pci-based ne2000 boards.
1.1.2.1	22-Mar-2000	ws	file ipkdb.h was added on branch sommerfeld_i386mp_1 on 2000-03-22 20:58:29 +0000
1.22	27-Feb-2003	fvdl	Adapt for move to x86/include.
1.21	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.20	22-Nov-2002	fvdl	New interrupt code. The basic idea behind it is to hide the differences in interrupt controllers in struct pic, and try to keep as much common code as possible. At the lowest (asm) level, this is done with CPP macros. The main structure is now struct intrsource, describing an established interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic). For quick masking, there may be a maximum of 32 sources per CPU. Sources can be assigned to any CPU in the MP case, though currently they all go to the boot CPU.
1.19	18-Jun-2002	tshiozak	branches: 1.19.2; add MD support for ACPI sleep/wakeup.
1.18	15-Nov-2000	thorpej	branches: 1.18.4; 1.18.6; 1.18.18; Garbage-collect some more stuff.
1.17	15-Nov-2000	enami	Make isa_machdep.c compile again.
1.16	04-Jun-2000	cgd	Implement the more flexiable `evcnt' interface as discussed (briefly) on tech-kern and now documented in evcnt(9).
1.15	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.14	07-Feb-2000	thorpej	branches: 1.14.4; Add a new function call to the ISA DMA API: isa_dmamaxsize(), which returns the maximum transfer size for the specified DMA channel. Make all clients of ISA DMA use this call to determine their maximum transfer size.
1.13	19-Mar-1999	cgd	branches: 1.13.8; Moved from arch/i386/isa/isa_machdep.h,v
1.12	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.11	09-Jun-1998	thorpej	Add isa_dma{freeze,thaw}(), used to temporarily stop and then restart all ISA DMA. Needed by e.g. the SmartCard reader for Sharks.
1.10	09-Jun-1998	thorpej	Change the ISA DMA API to take an isa_chipset_tag_t rather than a struct device * corresponding to the ISA bus device. The ISA DMA controller driver functions have been renamed and now take a struct isa_dma_state *, and are called indirectly by machine-dependent code which provides the DMA state. These changes allow e.g. `ofisa' (the OpenFirmware configuration mechanism for the ISA bus, used by e.g. Sharks) to use the MI ISA DMA controller code.
1.9	03-Jun-1998	thorpej	Implement bounce buffers for mbufs.
1.8	14-Oct-1997	thorpej	Pull down from marc-pcmcia branch: Implement new ISA interface functions: * isa_intr_alloc() - allocate an ISA IRQ * isa_mem_alloc() - allocate ISA memory space * isa_mem_free() - free ISA memory space
1.7	06-Jun-1997	thorpej	branches: 1.7.4; Pull thorpej-bus-dma branch into mainline.
1.6	03-May-1996	christos	branches: 1.6.10; - Add missing prototypes. - Fix gcc warnings
1.5	11-Apr-1996	cgd	prototype MD functions for use by MI ISA code (isa_attach_hook(), isa_intr_{,dis}establish()) here. Make this file safely includable by <dev/isa/isavar.h> and by all of the places in the i386 that are currently including it, which means multiple inclusion protection and a few otherwise-extraneous 'forward' structure declarations. isa_intr_establish() and isa_intr_disestablish() now take as their first argument a machine-dependent cookie of type isa_chipset_tag_t, which is also defined here.
1.4	04-May-1995	cgd	use the machine-independent definitions for the mc146818. wrap the various uses of the chip (including the uses of the NVRAM on it) around the new defintions, structures, and functions specified in the header.
1.3	21-Apr-1995	mycroft	Clean up hole conversion functions a little.
1.2	21-Apr-1995	mycroft	Remove relocation of Crtat from locore.
1.1	17-Apr-1995	cgd	clean up several ISA device interfaces: autoconfiguration, header inclusion, and interrupt configuration. more work still needs to be done, but it's getting better...
1.6.10.2	18-May-1997	thorpej	Add an external declaration for the isa dma tag.
1.6.10.1	13-May-1997	thorpej	Implement ISA-specific bus dma methods, and provide the bus dma tag to children via isa_attach_hook(). These bus dma methods are mostly wrappers around the generic i386 bus dma methods, but include support for DMA bounce buffers where required.
1.7.4.3	29-Aug-1997	thorpej	Make isa_mem_alloc() and isa_mem_free() functions, not macros, so that IOM_* constants don't have to be in scope in order to call these. Also, fix a fencepost in isa_mem_alloc().
1.7.4.2	11-Aug-1997	thorpej	Make isa_intr_alloc() return non-zero on IRQ allocation failure, rather than panicing.
1.7.4.1	30-Jul-1997	marc	added pcmcia infrastructure and a few devices
1.13.8.2	22-Nov-2000	bouyer	Sync with HEAD.
1.13.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.14.4.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.18.18.1	16-Jul-2002	gehenna	catch up with -current.
1.18.6.1	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.18.4.2	11-Dec-2002	thorpej	Sync with HEAD.
1.18.4.1	20-Jun-2002	nathanw	Catch up to -current.
1.19.2.2	18-Jun-2002	tshiozak	add MD support for ACPI sleep/wakeup.
1.19.2.1	18-Jun-2002	tshiozak	file isa_machdep.h was added on branch sommerfeld_i386mp_1 on 2002-06-18 07:56:15 +0000
1.6	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.5	01-Jun-2000	cgd	branches: 1.5.4; 1.5.130; 1.5.132; 1.5.134; kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.4	19-Mar-1999	cgd	branches: 1.4.8; 1.4.18; Moved from arch/i386/isa/isapnp_machdep.h,v
1.3	05-Sep-1998	christos	Assign copyright to TNF.
1.2	04-Oct-1997	thorpej	Copyright assigned to The NetBSD Foundation.
1.1	16-Jan-1997	christos	branches: 1.1.2; 1.1.10; i386 plug and play glue
1.1.10.1	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.1.2.1	18-Jan-1997	thorpej	Update from trunk.
1.4.18.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.4.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.5.134.1	16-May-2008	yamt	sync with head.
1.5.132.1	18-May-2008	yamt	sync with head.
1.5.130.1	02-Jun-2008	mjf	Sync with HEAD.
1.5.4.2	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.5.4.1	01-Jun-2000	cgd	file isapnp_machdep.h was added on branch sommerfeld_i386mp_1 on 2000-06-01 00:04:52 +0000
1.2	02-Jul-2004	drochner	just include <sys/joystick.h>
1.1	27-Mar-1996	perry	branches: 1.1.66; add joystick driver from Matthieu Herrb (pr #2267)
1.1.66.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.66.2	18-Sep-2004	skrll	Sync with HEAD.
1.1.66.1	03-Aug-2004	skrll	Sync with HEAD
1.7	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.6	25-Apr-2020	maxv	Switch to the new PTE naming. The old naming is now unused, remove it.
1.5	05-Oct-2010	jym	branches: 1.5.70; Import PAE support for kvm(3): - add kvm_i386pae.c (used for PAE memory translations), and update Makefile for libkvm build. - in pdppaddr: pass a flag to indicate PAE mode. Use a bit ignored by the MMU. Mask address with PG_FRAME to avoid side effects. Tested with vmstat(1)/netstat(1) to debug core files of PAE and !PAE kernels. Older kernel dumps will default to native i386 (!PAE) mode. XXX Currently, savecore(8) will fail to dump a PAE kernel in a !PAE environment (and reciprocally). So you need to sync and reboot with a kernel of the same mode as the one that crashed. Once the dump is successful, this does not matter anymore.
1.4	12-Jan-2008	ad	branches: 1.4.10; 1.4.24; 1.4.30; 1.4.32; - Split crashdump code out into its own file. - Remove NO_SPARSE_DUMP. - Minor KNF, sprinkle static.
1.3	26-Dec-2005	perry	branches: 1.3.50; 1.3.56; 1.3.64; u_intN_t -> uintN_t
1.2	26-Aug-2004	junyoung	branches: 1.2.12; As suggested on tech-kern@ days ago: * Get rid of PTmap, PTD, PTDpde, APTmap, APTD, and APTDpde from locore.S. * Rename PTDpaddr to PDPpaddr, ptdpaddr in struct cpu_kcore_hdr to pdppaddr for consistency.
1.1	18-Feb-1998	thorpej	branches: 1.1.50; Switch the i386 port to the new kernel crash dump format.
1.1.50.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.50.2	18-Sep-2004	skrll	Sync with HEAD.
1.1.50.1	03-Sep-2004	skrll	Sync with HEAD
1.2.12.2	21-Jan-2008	yamt	sync with head
1.2.12.1	21-Jun-2006	yamt	sync with head.
1.3.64.1	19-Jan-2008	bouyer	Sync with HEAD
1.3.56.1	18-Feb-2008	mjf	Sync with HEAD.
1.3.50.1	23-Mar-2008	matt	sync with HEAD
1.4.32.1	05-Mar-2011	rmind	sync with head
1.4.30.1	22-Oct-2010	uebayasi	Sync with HEAD (-D20101022).
1.4.24.1	24-Oct-2010	jym	Sync with HEAD
1.4.10.1	09-Oct-2010	yamt	sync with head
1.5.70.1	25-Apr-2020	bouyer	Sync with bouyer-xenpvh-base2 (HEAD)
1.7	03-Oct-2012	dsl	Remove all references to KVM86. It was only ever used by APMBIOS - and then only if an option was selected. Probably didn't work well at all!
1.6	26-Oct-2008	mrg	branches: 1.6.28; 1.6.38; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.5	26-Dec-2005	perry	branches: 1.5.74; 1.5.78; 1.5.84; u_intN_t -> uintN_t
1.4	02-Oct-2002	drochner	branches: 1.4.22; add kvm86_incall again which got lost in the SMP merge
1.3	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.2	10-Jul-2002	drochner	branches: 1.2.2; 1.2.4; 1.2.6; 1.2.8; add a function to read from the BIOS vm86 (most of it is already mapped to KVM, but it appears easier to me to do it this way instead of a lot of range checks)
1.1	07-Jul-2002	drochner	This is a (still experimental) framework to run BIOS code in a virtual 8086 machine. Ifff it works, it is much easier and more elegant than going to real real mode: -simpler code -no need for "identity" memory mappings -easy passing of buffers for bulk data to functions -some more control There is no interrupt support ATM, and it lacks a function to access random virtual memory of the VM. MP issues to consider.
1.2.8.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.2.8.2	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.2.8.1	10-Jul-2002	jdolecek	file kvm86.h was added on branch kqueue on 2002-09-06 08:36:27 +0000
1.2.6.2	19-Aug-2002	sommerfeld	merged with -current as of today. rearranged gdt init to deal with KVM86 changes; however, KVM86 has no chance of working yet; good thing it's not part of GENERIC
1.2.6.1	10-Jul-2002	sommerfeld	file kvm86.h was added on branch sommerfeld_i386mp_1 on 2002-08-19 01:22:36 +0000
1.2.4.2	01-Aug-2002	nathanw	Catch up to -current.
1.2.4.1	10-Jul-2002	nathanw	file kvm86.h was added on branch nathanw_sa on 2002-08-01 02:42:10 +0000
1.2.2.2	16-Jul-2002	gehenna	catch up with -current.
1.2.2.1	10-Jul-2002	gehenna	file kvm86.h was added on branch gehenna-devsw on 2002-07-16 08:29:51 +0000
1.4.22.1	21-Jun-2006	yamt	sync with head.
1.5.84.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.5.78.1	04-May-2009	yamt	sync with head.
1.5.74.1	17-Jan-2009	mjf	Sync with HEAD.
1.6.38.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.6.28.1	30-Oct-2012	yamt	sync with head
1.26	21-Jan-2019	dholland	Fix wrong scoping of {U,}LLONG_MAX. More cases, not just amd64. PR 53298 from Roberto E. Vargas Caballero.
1.25	21-Apr-2014	matt	branches: 1.25.26; 1.25.28; Since all our compilers support __DBL_* and __FLT_*, use them to define {DBL,FLT}_{DIG,MIN,MAX}
1.24	11-Apr-2013	christos	branches: 1.24.4; 1.24.8; add missing SSIZE_MIN
1.23	28-Mar-2012	christos	branches: 1.23.2; - Normalize inclusion protection (remove) - Move CHAR_{MIN,MAX} to a common file. - Fix broken comments
1.22	07-Jun-2010	tnozaki	branches: 1.22.8; 1.22.12; 1. MB_LEN_MAX switch MD to MI. 2. unfortunately hppa's MB_LEN_MAX is defined incorrectly 6 instead of 32 so we have to add more setlocale(3) __RENAME func, __setlocale50. 3. move setlocale1.c and setlocale32.c to lib/libc/compat/locale/* prepareing for next libc major crunk. 4. bump libc minor version.
1.21	26-Oct-2008	mrg	branches: 1.21.8; 1.21.14; 1.21.16; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.20	17-Oct-2007	garbled	branches: 1.20.16; 1.20.20; 1.20.26; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.19	31-Aug-2007	drochner	Fix definitions of UCHAR_MAX/USHRT_MAX and related types. C99 requires that these definitions promote to (signed/unsigned) integer the same way as the types the definition is for. And since unsigned char/short fit into an "int" on all our archs and thus promote to signed int, the definitions must not be unsigned. Fixes PR lib/31306 by Neil Booth.
1.18	23-Sep-2006	elad	branches: 1.18.8; 1.18.16; 1.18.22; 1.18.26; 1.18.28; PR/30760: David A. Holland: obsolete comment in i386/limits.h Patch applied, thanks!
1.17	07-Aug-2003	agc	branches: 1.17.16; 1.17.46; 1.17.48; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.16	28-Apr-2003	bjh21	branches: 1.16.2; Add a new feature-test macro, _NETBSD_SOURCE. If this is defined by the application, all NetBSD interfaces are made visible, even if some other feature-test macro (like _POSIX_C_SOURCE) is defined. <sys/featuretest.h> defined _NETBSD_SOURCE if none of _ANSI_SOURCE, _POSIX_C_SOURCE and _XOPEN_SOURCE is defined, so as to preserve existing behaviour. This has two major advantages: + Programs that require non-POSIX facilities but define _POSIX_C_SOURCE can trivially be overruled by putting -D_NETBSD_SOURCE in their CFLAGS. + It makes most of the #ifs simpler, in that they're all now ORs of the various macros, rather than having checks for (!defined(_ANSI_SOURCE) \|\| !defined(_POSIX_C_SOURCE) \|\| !defined(_XOPEN_SOURCE)) all over the place. I've tried not to change the semantics of the headers in any case where _NETBSD_SOURCE wasn't defined, but there were some places where the current semantics were clearly mad, and retaining them was harder than correcting them. In particular, I've mostly normalised things so that _ANSI_SOURCE gets you the smallest set of stuff, then _POSIX_C_SOURCE, _XOPEN_SOURCE and _NETBSD_SOURCE in that order. Tested by building for vax, encouraged by thorpej, and uncontested in tech-userlevel for a week.
1.15	08-Aug-2000	tshiozak	branches: 1.15.2; Preparation for the future introduction of multibyte locale. - MB_LEN_MAX is increased to 32. - To ensure binary compatibility for old executables under multibyte locale, versioned setlocale is added. - __mb_len_cur definision is added in setlocale.c and enable it in stdlib.h . It is also important for multibyte locale stuffs, but I just forgot.
1.14	07-Mar-2000	kleink	branches: 1.14.4; 1.14.6; Define ISO C99 (unsigned) long long (min, max) symbols.
1.13	06-Aug-1998	kleink	branches: 1.13.12; _POSIX_SOURCE -> _POSIX_C_SOURCE
1.12	09-Jan-1998	perry	multiple include protect machine/limits.h, fixes pr 4473 (from Mika Nystrom)
1.11	21-Dec-1995	mycroft	Minor cleanup.
1.10	20-Dec-1995	mycroft	Use the U suffix to insure unsigned values are treated as such.
1.9	27-Oct-1994	cgd	new RCS ID format.
1.8	05-Oct-1994	jtc	Add constants required by XPG3
1.7	09-May-1994	cgd	branches: 1.7.2; new defn's
1.6	20-Dec-1993	mycroft	branches: 1.6.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.5	24-Sep-1993	jtc	Do not provide quad constants if compiling in a strict ansi or a strict posix environment.
1.4	16-Sep-1993	mycroft	Add constants for quad support.
1.3	18-Aug-1993	jtc	branches: 1.3.2; Removed CLK_TCK, it's value is not machine dependant. It's supposed to be defined in <time.h>, not <limits.h> anyway.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.3.2.2	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.3.2.1	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.6.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.6.2.1	20-Dec-1993	mycroft	file limits.h was added on branch magnum on 1993-12-20 05:25:20 +0000
1.7.2.1	11-Oct-1994	mycroft	Update from trunk.
1.13.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.14.6.1	09-Aug-2000	tshiozak	pull up the following changes (approved by thorpej): > cvs rdiff -r1.9 -r1.10 basesrc/include/locale.h > cvs rdiff -r1.45 -r1.46 basesrc/include/stdlib.h > cvs rdiff -r1.16 -r1.17 basesrc/lib/libc/locale/Makefile.inc > cvs rdiff -r1.17 -r1.18 basesrc/lib/libc/locale/setlocale.c > cvs rdiff -r0 -r1.2 basesrc/lib/libc/locale/setlocale_sb.c > cvs rdiff -r1.6 -r1.7 syssrc/sys/arch/alpha/include/limits.h > cvs rdiff -r1.1 -r1.2 syssrc/sys/arch/arm26/include/limits.h > cvs rdiff -r1.7 -r1.8 syssrc/sys/arch/arm32/include/limits.h > cvs rdiff -r1.14 -r1.15 syssrc/sys/arch/i386/include/limits.h > cvs rdiff -r1.12 -r1.13 syssrc/sys/arch/m68k/include/limits.h > cvs rdiff -r1.13 -r1.14 syssrc/sys/arch/mips/include/limits.h > cvs rdiff -r1.10 -r1.11 syssrc/sys/arch/pc532/include/limits.h > cvs rdiff -r1.6 -r1.7 syssrc/sys/arch/powerpc/include/limits.h > cvs rdiff -r1.2 -r1.3 syssrc/sys/arch/sh3/include/limits.h > cvs rdiff -r1.11 -r1.12 syssrc/sys/arch/sparc/include/limits.h > cvs rdiff -r1.7 -r1.8 syssrc/sys/arch/sparc64/include/limits.h > cvs rdiff -r1.9 -r1.10 syssrc/sys/arch/vax/include/limits.h > > Outline: > > Preparation for the future introduction of multibyte locale. > - MB_LEN_MAX is increased to 32. > - To ensure binary compatibility for old executables > under multibyte locale, versioned setlocale is added. > - __mb_len_cur definision is added in setlocale.c > and enable it in stdlib.h . > It is also important for multibyte locale stuffs, > but I just forgot.
1.14.4.1	28-May-2000	minoura	Citrus Project XPG4DL, an implementation of I18N (locale) framework, is imported.
1.15.2.2	08-Aug-2000	tshiozak	Preparation for the future introduction of multibyte locale. - MB_LEN_MAX is increased to 32. - To ensure binary compatibility for old executables under multibyte locale, versioned setlocale is added. - __mb_len_cur definision is added in setlocale.c and enable it in stdlib.h . It is also important for multibyte locale stuffs, but I just forgot.
1.15.2.1	08-Aug-2000	tshiozak	file limits.h was added on branch sommerfeld_i386mp_1 on 2000-08-08 22:31:14 +0000
1.16.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.16.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.16.2.1	03-Aug-2004	skrll	Sync with HEAD
1.17.48.1	22-Oct-2006	yamt	sync with head
1.17.46.1	18-Nov-2006	ad	Sync with head.
1.17.16.2	03-Sep-2007	yamt	sync with head.
1.17.16.1	30-Dec-2006	yamt	sync with head.
1.18.28.1	06-Nov-2007	matt	sync with HEAD
1.18.26.1	03-Sep-2007	jmcneill	Sync with HEAD.
1.18.22.1	03-Sep-2007	skrll	Sync with HEAD.
1.18.16.1	03-Oct-2007	garbled	Sync with HEAD
1.18.8.1	09-Oct-2007	ad	Sync with head.
1.20.26.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.20.20.2	11-Aug-2010	yamt	sync with head.
1.20.20.1	04-May-2009	yamt	sync with head.
1.20.16.1	17-Jan-2009	mjf	Sync with HEAD.
1.21.16.1	03-Jul-2010	rmind	sync with head
1.21.14.1	17-Aug-2010	uebayasi	Sync with HEAD.
1.21.8.1	24-Oct-2010	jym	Sync with HEAD
1.22.12.1	05-Apr-2012	mrg	sync to latest -current.
1.22.8.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.22.8.1	17-Apr-2012	yamt	sync with head
1.23.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.23.2.1	23-Jun-2013	tls	resync from head
1.24.8.1	10-Aug-2014	tls	Rebase.
1.24.4.1	18-May-2014	rmind	sync with head
1.25.28.1	10-Jun-2019	christos	Sync with HEAD
1.25.26.1	26-Jan-2019	pgoyette	Sync with HEAD
1.8	01-Oct-1998	erh	Moved to compat/linux/i386/linux_machdep.h.
1.7	11-Sep-1998	mycroft	Substantial signal handling changes: * Increase the size of sigset_t to accomodate 128 signals -- adding new versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() to handle the changed arguments. * Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() into separate functions, and call them from all the emulations rather than hard-coding everything. (Avoids uses the stackgap crap for these system calls.) * Add a new flag (p_checksig) to indicate that a process may have signals pending and userret() needs to do the full (slow) check. * Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE. * Correct emulation bugs with restoring SS_ONSTACK. * Make the signal mask in the sigcontext always use the emulated mask format. * Store signals internally in sigaction structures, rather than maintaining a bunch of little sigsets for each SA_* bit. * Keep track of where we put the signal trampoline, rather than figuring it out in _sendsig(). Issue a warning when a non-emulated sigaction bit is observed. * Add missing emulated signals, and a native SIGPWR (currently not used). * Implement the `not reset when caught' semantics for relevant signals. Note: Only code touched by the i386 port has been modified. Other ports and emulations need to be updated.
1.6	11-Jun-1998	drochner	Since wscons can support VT_xxx ioctl()s, enable them if NWSDISPLAY > 0. Implement KDGKBTYPE and KDGKBENT, this will make X servers happy.
1.5	03-May-1996	christos	Add missing prototypes.
1.4	27-Aug-1995	fvdl	A couple of definitions added for previous changes.
1.3	07-May-1995	mycroft	Improve Linux signal emulation, and emulate the modify_ldt() call.
1.2	22-Apr-1995	christos	- added sunos_machdep.c for sun3, atari, amiga and mac68k. - changed machdep.c and trap.c to use struct emul. - remove ep_setup references. - added struct emul to all emulations.
1.1	07-Apr-1995	fvdl	Arch dependant defs for Linux. Just signal stackframe definition for now.
1.9	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.8	01-Oct-2007	ad	Now that the bootblocks are the same, share loadfile_machdep.h between amd64 and i386.
1.7	27-Sep-2007	ad	- Make i386 bootblocks boot amd64 kernels, and change to say "NetBSD/x86..." - Don't bother booting a.out kernels any more.
1.6	25-Jan-2006	christos	branches: 1.6.28; 1.6.36; 1.6.46; 1.6.48; 1.6.50; free -> dealloc unsigned -> size_t for alloc/dealloc
1.5	27-Oct-2003	junyoung	branches: 1.5.16; 1.5.28; Nuke __P().
1.4	31-Oct-2001	thorpej	branches: 1.4.2; 1.4.18; Split elf32 vs. elf64 handling in loadfile(), including support for each separately and explicitly. BOOT_ELF is now BOOT_ELF32 and BOOT_ELF64, and ELFSIZE should no longer be defined in loadfile_machdep.h.
1.3	30-Nov-1999	drochner	branches: 1.3.8; 1.3.10; 1.3.14; Allow loading of kernels larger than (16-1)M. XXX This is a stopgap fix which can be pulled up to 1.4.x. It only replaces the arbitrary 16M boundary by an arbitrary 128M boundary. A clean solution would need changes to the mi loadfile.c parts.
1.2	28-Apr-1999	christos	branches: 1.2.2; 1.2.4; 1.2.10; Fix tyop
1.1	28-Apr-1999	christos	MD defines for the MI loadfile
1.2.10.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.2.4.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.2.2.1	21-Jun-1999	thorpej	Sync w/ -current.
1.3.14.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.3.10.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.3.8.1	14-Nov-2001	nathanw	Catch up to -current.
1.4.18.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.4.18.2	18-Sep-2004	skrll	Sync with HEAD.
1.4.18.1	03-Aug-2004	skrll	Sync with HEAD
1.4.2.2	31-Oct-2001	thorpej	Split elf32 vs. elf64 handling in loadfile(), including support for each separately and explicitly. BOOT_ELF is now BOOT_ELF32 and BOOT_ELF64, and ELFSIZE should no longer be defined in loadfile_machdep.h.
1.4.2.1	31-Oct-2001	thorpej	file loadfile_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-10-31 17:20:47 +0000
1.5.28.1	01-Feb-2006	yamt	sync with head.
1.5.16.2	27-Oct-2007	yamt	sync with head.
1.5.16.1	21-Jun-2006	yamt	sync with head.
1.6.50.1	06-Oct-2007	yamt	sync with head.
1.6.48.1	06-Nov-2007	matt	sync with HEAD
1.6.46.1	02-Oct-2007	joerg	Sync with HEAD.
1.6.36.1	03-Oct-2007	garbled	Sync with HEAD
1.6.28.1	09-Oct-2007	ad	Sync with head.
1.9	27-Feb-2003	fvdl	Adapt for move to x86/include.
1.8	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.7	30-Jan-2003	thorpej	Test _KERNEL_OPT before including opt_*.h files.
1.6	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.5	30-Apr-2001	lukem	branches: 1.5.2; 1.5.10; remove some lint, including ansifying some inlines
1.4	02-May-2000	thorpej	branches: 1.4.6; 1.4.8; Let each platform typedef the new __cpu_simple_lock_t, which should be the most efficient type used for the atomic operations in the simplelock structure, and should also be __volatile.
1.3	29-Apr-2000	thorpej	Implement __cpu_simple_{,un}lock{,try}(), modeled after the Alpha implementation.
1.2	29-Apr-2000	thorpej	Require that each each MACHINE/MACHINE_ARCH supply a lock.h. This file contains the values __SIMPLELOCK_LOCKED and __SIMPLELOCK_UNLOCKED, which replace the old SIMPLELOCK_LOCKED and SIMPLELOCK_UNLOCKED. These files are also required to supply inline functions __cpu_simple_lock(), __cpu_simple_lock_try(), and __cpu_simple_unlock() if locking is to be supported on that platform (i.e. if MULTIPROCESSOR is defined in the _KERNEL case). Change these functions to take an int * (&alp->lock_data) rather than the struct simplelock * itself. These changes make it possible for userland to use the locking primitives by including <machine/lock.h>.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; file lock.h was initially added on branch sommerfeld_i386mp_1.
1.1.2.7	05-May-2002	sommerfeld	Make C++-safe
1.1.2.6	29-Dec-2001	sommerfeld	Add barriers, just in case
1.1.2.5	30-Apr-2001	sommerfeld	Catch up on some files missed the last time 'round
1.1.2.4	07-Jan-2001	sommerfeld	For the sake of my sanity, make this match the code in lock_machdep.c a bit more closely.
1.1.2.3	31-Aug-2000	sommerfeld	Protect include of opt_lockdebug.h
1.1.2.2	03-May-2000	sommerfeld	Only do cpu_simple locks out-of-line if LOCKDEBUG is defined
1.1.2.1	20-Feb-2000	sommerfeld	Declarations of of cpu_simple_lock functions. (currently not inlined, for debuggability)
1.4.8.2	18-Oct-2002	nathanw	Catch up to -current.
1.4.8.1	21-Jun-2001	nathanw	Catch up to -current.
1.4.6.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.4.6.1	02-May-2000	bouyer	file lock.h was added on branch thorpej_scsipi on 2000-11-20 20:09:29 +0000
1.5.10.3	20-Mar-2002	thorpej	* Add a memory clobber to the atomic operations in the lock-acquire path, so that the compiler will treat them as a memory access barrier. * Spin with a normal memory load until the lock is released and then retry the compare-and-swap in __cpu_simple_lock().
1.5.10.2	20-Mar-2002	thorpej	Use the new P4 "pause" insn in the body of the __cpu_simple_lock() loop.
1.5.10.1	19-Mar-2002	thorpej	Change __cpu_simple_lock_t to a single byte.
1.5.2.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.2	04-Nov-2024	christos	Undo previous lwp.h change.
1.1	03-Nov-2024	christos	Split __lwp_getprivate_fast and __lwp_*tcb from mcontext.h into a separate lwp.h file.
1.1	30-Nov-2024	christos	branches: 1.1.4; Create a new header lwp_private.h to contain _lwp_getprivate_fast, _lwp_gettcb_fast, _lwp_settcb and remove them from mcontext.h, so that: 1. we don't need special hacks to hide them 2. we can include <lwp.h> where needed to get the necessary prototypes without redefining them locally.
1.1.4.2	02-Aug-2025	perseant	Sync with HEAD
1.1.4.1	30-Nov-2024	perseant	file lwp_private.h was added on branch perseant-exfatfs on 2025-08-02 05:55:44 +0000
1.3	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.2	25-Jun-2005	christos	branches: 1.2.82; 1.2.84; 1.2.86; the siginfo trap codes are mi.
1.1	04-Dec-2003	christos	branches: 1.1.4; add stub file so that the kernel compiles.
1.1.4.5	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.1.4.4	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.4.3	18-Sep-2004	skrll	Sync with HEAD.
1.1.4.2	03-Aug-2004	skrll	Sync with HEAD
1.1.4.1	04-Dec-2003	skrll	file mach_machdep.h was added on branch ktrace-lwp on 2004-08-03 10:36:04 +0000
1.2.86.1	16-May-2008	yamt	sync with head.
1.2.84.1	18-May-2008	yamt	sync with head.
1.2.82.1	02-Jun-2008	mjf	Sync with HEAD.
1.5	26-Apr-2011	joerg	Remove Darwin, MACH and Mach-O support.
1.4	28-Apr-2008	martin	branches: 1.4.14; 1.4.22; 1.4.28; Remove clause 3 and 4 from TNF licenses
1.3	29-Oct-2002	manu	branches: 1.3.108; 1.3.110; 1.3.112; Introduce an array of supported CPU types by a given arch for Mach-O
1.2	14-Jul-2001	christos	branches: 1.2.2; 1.2.4; 1.2.6; add MACHO_MACHDEP_CASES
1.1	14-Jul-2001	christos	new i386 files for macho and mach.
1.2.6.3	11-Nov-2002	nathanw	Catch up to -current
1.2.6.2	24-Aug-2001	nathanw	Catch up with -current.
1.2.6.1	14-Jul-2001	nathanw	file macho_machdep.h was added on branch nathanw_sa on 2001-08-24 00:08:34 +0000
1.2.4.2	03-Aug-2001	lukem	update to -current
1.2.4.1	14-Jul-2001	lukem	file macho_machdep.h was added on branch kqueue on 2001-08-03 04:11:45 +0000
1.2.2.2	14-Jul-2001	christos	add MACHO_MACHDEP_CASES
1.2.2.1	14-Jul-2001	christos	file macho_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-07-14 03:05:52 +0000
1.3.112.1	16-May-2008	yamt	sync with head.
1.3.110.1	18-May-2008	yamt	sync with head.
1.3.108.1	02-Jun-2008	mjf	Sync with HEAD.
1.4.28.1	06-Jun-2011	jruoho	Sync with HEAD.
1.4.22.1	31-May-2011	rmind	sync with head
1.4.14.1	02-May-2011	jym	Sync with head.
1.5	22-Oct-2003	kleink	Use a common <machine/math.h> for amd64 and i386.
1.4	19-Feb-2002	simonb	branches: 1.4.2; 1.4.18; Clean up some rampant code duplication wrt ieee number handling: - Add alignment-safe double and float unions. - Use the above for the __infinity and __nan constants on all architectures that use the standard ieee754 representation of those constants. - Add a single copy of various ieee754 math functions (frexp, isinf, isnan, ldexp and modf) that had numerous duplicates among the arch-specific directories. - Use the above functions on all architectures where the generic C versions where used. Architectures that had local assembly routines are untouched (for those functions only).
1.3	05-Feb-2000	kleink	branches: 1.3.8; 1.3.10; 1.3.12; Improve namespace test macros a bit.
1.2	04-Jan-2000	kleink	const -> __const and include <sys/cdefs.h> earlier; fixes PR lib/9052 by Takahiro Kambe.
1.1	23-Dec-1999	kleink	C99: Define a NAN macro in <math.h> which evaulates to a constant expression of a single-precision quiet NaN; only to be defined on platforms that do support this value.
1.3.12.1	16-Mar-2002	jdolecek	Catch up with -current.
1.3.10.1	28-Feb-2002	nathanw	Catch up to -current.
1.3.8.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.3.8.1	05-Feb-2000	bouyer	file math.h was added on branch thorpej_scsipi on 2000-11-20 20:09:29 +0000
1.4.18.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.4.18.2	18-Sep-2004	skrll	Sync with HEAD.
1.4.18.1	03-Aug-2004	skrll	Sync with HEAD
1.4.2.2	19-Feb-2002	simonb	Clean up some rampant code duplication wrt ieee number handling: - Add alignment-safe double and float unions. - Use the above for the __infinity and __nan constants on all architectures that use the standard ieee754 representation of those constants. - Add a single copy of various ieee754 math functions (frexp, isinf, isnan, ldexp and modf) that had numerous duplicates among the arch-specific directories. - Use the above functions on all architectures where the generic C versions where used. Architectures that had local assembly routines are untouched (for those functions only).
1.4.2.1	19-Feb-2002	simonb	file math.h was added on branch sommerfeld_i386mp_1 on 2002-02-19 13:08:15 +0000
1.15	01-Jul-2011	dyoung	#include <sys/bus.h> instead of <machine/bus.h>.
1.14	03-Apr-2011	dyoung	Clean up excessive #ifdef'age of NMI trap handling for amd64/i386/xen. Handle NMI in all Xen kernels.
1.13	04-May-2009	cegger	branches: 1.13.4; 1.13.6; struct device * -> device_t No functional changes intended.
1.12	11-Apr-2007	garbled	branches: 1.12.38; 1.12.52; add the define of MCA_MAX_SLOTS here
1.11	04-Mar-2007	christos	branches: 1.11.2; 1.11.4; Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.10	26-Dec-2005	perry	branches: 1.10.26; u_intN_t -> uintN_t
1.9	26-Feb-2003	fvdl	branches: 1.9.18; Redirect some includes to x86/foo.h
1.8	02-Dec-2001	jdolecek	branches: 1.8.2; Add couple hooks, needed primarily for NCR 53C90 MCA card driver: * add flag to explicitly specify if the DMA should be done as 16bit or 8bit * add flag to specify the DMA should happen via I/O port * add new function mca_dma_set_ioport(), to set I/O port to be used for the DMA operation Also clarify copyright (welcome to 2001 :), and couple other minor nits
1.7	23-Nov-2001	jdolecek	add mca_dmamap_create() prototype for MI code
1.6	02-May-2001	jdolecek	branches: 1.6.4; The system configuration block structure doesn't need to be public
1.5	22-Apr-2001	jdolecek	Introduce machdep mca_disk_[un]busy(), which turn on/off disk LED. Call as appropriate from the ESDI disk driver
1.4	09-Jul-2000	jdolecek	branches: 1.4.4; 1.4.6; move struct sys_config from mca_machdep.c to mca_machdep.h fix copyright (overlooked in original commit) use bitmask_snprintf() instead of printf("...%b") (though that part is commented out anyway ATM)
1.3	04-Jun-2000	cgd	Implement the more flexiable `evcnt' interface as discussed (briefly) on tech-kern and now documented in evcnt(9).
1.2	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.1	11-May-2000	jdolecek	branches: 1.1.2; Add i386-specific part of MicroChannel Architecture bus support, as found in some older IBM PS/2 machines. This code is based upon work by Scott D. Telford, with some minor bits in arch/i386/mca/mca_machdep.c taken from FreeBSD. XXX this is still very experimental and development version; use at your XXX own risk
1.1.2.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.4.6.2	08-Jan-2002	nathanw	Catch up to -current.
1.4.6.1	21-Jun-2001	nathanw	Catch up to -current.
1.4.4.3	23-Apr-2001	bouyer	Sync with HEAD.
1.4.4.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.4.4.1	09-Jul-2000	bouyer	file mca_machdep.h was added on branch thorpej_scsipi on 2000-11-20 20:09:29 +0000
1.6.4.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.8.2.2	02-Dec-2001	jdolecek	Add couple hooks, needed primarily for NCR 53C90 MCA card driver: * add flag to explicitly specify if the DMA should be done as 16bit or 8bit * add flag to specify the DMA should happen via I/O port * add new function mca_dma_set_ioport(), to set I/O port to be used for the DMA operation Also clarify copyright (welcome to 2001 :), and couple other minor nits
1.8.2.1	02-Dec-2001	jdolecek	file mca_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-12-02 17:02:34 +0000
1.9.18.2	03-Sep-2007	yamt	sync with head.
1.9.18.1	21-Jun-2006	yamt	sync with head.
1.10.26.2	15-Apr-2007	yamt	sync with head.
1.10.26.1	12-Mar-2007	rmind	Sync with HEAD.
1.11.4.1	11-Jul-2007	mjf	Sync with head.
1.11.2.1	27-May-2007	ad	Sync with head.
1.12.52.4	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.12.52.3	02-May-2011	jym	Sync with head.
1.12.52.2	01-Nov-2009	jym	Sync with HEAD.
1.12.52.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.12.38.1	16-May-2009	yamt	sync with head
1.13.6.1	06-Jun-2011	jruoho	Sync with HEAD.
1.13.4.1	21-Apr-2011	rmind	sync with head
1.19	30-Nov-2024	christos	Create a new header lwp_private.h to contain _lwp_getprivate_fast, _lwp_gettcb_fast, _lwp_settcb and remove them from mcontext.h, so that: 1. we don't need special hacks to hide them 2. we can include <lwp.h> where needed to get the necessary prototypes without redefining them locally.
1.18	04-Nov-2024	christos	Undo previous lwp.h change.
1.17	03-Nov-2024	christos	Split __lwp_getprivate_fast and __lwp_*tcb from mcontext.h into a separate lwp.h file.
1.16	18-May-2024	thorpej	branches: 1.16.2; Clean up the <sys/ucontext.h> <-> <machine/mcontext.h> interface up a little: - Define _UC_MD_BIT* constants for the available machine-dependent bits, and use those constants to define the machine-dependent bits as well as the machine-independent bits that have machine-dependent values. - Explicitly generate an error if _UC_TLSBASE, _UC_SETSTACK, or _UC_CLRSTACK are not defined by <machine/mcontext.h>.
1.15	27-Dec-2019	kamil	Harmonize the namespace of fast TLS base pointer getter functions Protect __lwp_getprivate_fast() with _RTLD_SOURCE, _LIBC_SOURCE and __LIBPTHREAD_SOURCE__. Include in this namespace <sys/tcl.h> and use __BEGIN_DECLS/__END_DECLS for the sake of consistency.
1.14	15-Feb-2018	kamil	branches: 1.14.4; Introduce _UC_MACHINE_FP() as a macro _UC_MACHINE_FP() is a helper macro to extract from mcontext a frame pointer. Don't rely on this interface as a compiler might strip frame pointer or optimize it making this interface unreliable. For hppa assume a small frame context, for larger frames FP might be located in a different register (4 instead of 3). For ia64 there is no strict frame pointer, and registers might rotate. Reuse 79 following: ./gcc/config/ia64/ia64.h:#define HARD_FRAME_POINTER_REGNUM LOC_REG (79) Once ia64 will mature, this should be revisited. A macro can encapsulate a real function for extracting Frame Pointer on more complex CPUs / ABIs. For the remaining CPUs, reuse standard register as defined in appropriate ABI. The direct users of this macro are LLVM and GCC with Sanitizers. Proposed on tech-userlevel@. Sponsored by <The NetBSD Foundation>
1.13	12-Aug-2017	maxv	Remove vm86. Pass 4.
1.12	15-Feb-2014	dsl	branches: 1.12.6; 1.12.22; Load and save the fpu registers (for copies to/from userspace) using helper functions in arch/x86/x86/fpu.c They (hopefully) ensure that we write to the entire buffer and don't load values that might cause faults in kernel. Also zero out the 'pad' field of the i386 mcontext fp area that I think once contained the registers of any Weitek fpu. Dunno why it wasn't pasrt of the union. Some of these copies could be removed if the code directly copied the save area to/from userspace addresses.
1.11	04-Feb-2014	dsl	Remove the fields from 'struct save87' and 'struct fxsave' that were used to save the control word and tag word on fp faults. Nothing I could find looked at them before I removed the code that wrote them. It might appear that these fields got exported to userspace (via mcontext), but that code carefully avoided those fields. I've also deleted some very historic fields from the union inside __fpregset_t this doesn't affect its size. The __fpregset_t fields are now the same size as the ones in the pcb. Which is good because they get cast to each other in a few places.
1.10	25-Feb-2011	joerg	branches: 1.10.4; 1.10.14; 1.10.18; Be nicer to software that insists on -ansi and use __inline.
1.9	24-Feb-2011	joerg	Allow storing and receiving the LWP private pointer via ucontext_t on all platforms except VAX and IA64. Add fast access via register for AMD64, i386 and SH3 ports. Use this fast access in libpthread to replace the stack based pthread_self(). Implement skeleton support for Alpha, HPPA, PowerPC, SPARC and SPARC64, but leave it disabled. Ports that support this feature provide __HAVE____LWP_GETPRIVATE_FAST in machine/types.h and a corresponding __lwp_getprivate_fast in machine/mcontext.h. This material is based upon work partially supported by The NetBSD Foundation under a contract with Joerg Sonnenberger.
1.8	23-Feb-2011	joerg	Fix ucontext32_t on AMD64. Add optional compile time assertions for ucontext_t and ucontext32_t to ensure that they don't change. Provide the constants for AMD64, ARM, i386, and M68K.
1.7	28-Apr-2008	martin	branches: 1.7.14; 1.7.22; 1.7.28; 1.7.30; Remove clause 3 and 4 from TNF licenses
1.6	08-Oct-2003	thorpej	branches: 1.6.102; 1.6.104; 1.6.106; Add some accessor macros for the ucontext: * _UC_MACHINE_PC() - access the program counter * _UC_MACHINE_INTRV() - access the integer return value register * _UC_MACHINE_SET_PC() - set the program counter (this requires special handling on some platforms).
1.5	06-Sep-2003	christos	SA_SIGINFO changes.
1.4	25-Aug-2003	tron	Protect inclusion of "opt_vm86.h" with "#ifdef _KERNEL_OPT" as appropriate.
1.3	25-Aug-2003	drochner	fix _UC_MACHINE_SP() for the vm86 case: we have to add the SS offset
1.2	17-Jan-2003	thorpej	branches: 1.2.2; Merge the nathanw_sa branch.
1.1	05-Mar-2001	nathanw	branches: 1.1.2; file mcontext.h was initially added on branch nathanw_sa.
1.1.2.5	19-Dec-2002	thorpej	Expose _UC_UCONTEXT_ALIGN.
1.1.2.4	06-Sep-2002	nathanw	Define an alignment constraint, _UC_UCONTEXT_ALIGN; put it and _UC_MACHINE_SP() inside #ifdef _KERNEL.
1.1.2.3	14-Aug-2002	nathanw	Add space for the fxsave/XMM save area to mcontext_t; since the fxsave format includes the x87 register state (although in a different bit pattern), make it overlap the x87 save area, and add a flag to indicate that this format is in use. This makes mcontext_t 264 bytes larger, for a total of 720 bytes.
1.1.2.2	28-Dec-2001	nathanw	Add a macro, _UC_MACHINE_SP(), to fetch the user stack pointer from a ucontext_t.
1.1.2.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.2.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.2.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.2.2.1	03-Aug-2004	skrll	Sync with HEAD
1.6.106.1	16-May-2008	yamt	sync with head.
1.6.104.1	18-May-2008	yamt	sync with head.
1.6.102.1	02-Jun-2008	mjf	Sync with HEAD.
1.7.30.1	05-Mar-2011	bouyer	Sync with HEAD
1.7.28.1	06-Jun-2011	jruoho	Sync with HEAD.
1.7.22.1	05-Mar-2011	rmind	sync with head
1.7.14.1	28-Mar-2011	jym	Sync with HEAD. TODO before merge: - shortcut for suspend code in sysmon, when powerd(8) is not running. Borrow ``xs_watch'' thread context? - bug hunting in xbd + xennet resume. Rings are currently thrashed upon resume, so current implementation force flush them on suspend. It's not really needed.
1.10.18.1	18-May-2014	rmind	sync with head
1.10.14.2	03-Dec-2017	jdolecek	update from HEAD
1.10.14.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.10.4.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.12.22.3	21-Mar-2018	martin	Pull up the following, requested by kamil in ticket #552: external/gpl3/gcc{.old}/dist/libsanitizer/asan/asan_linux.cc 1.4 sys/arch/aarch64/include/mcontext.h 1.2 sys/arch/alpha/include/mcontext.h 1.9 sys/arch/amd64/include/mcontext.h 1.19 sys/arch/arm/include/mcontext.h 1.19 sys/arch/hppa/include/mcontext.h 1.9 sys/arch/i386/include/mcontext.h 1.14 sys/arch/ia64/include/mcontext.h 1.6 sys/arch/m68k/include/mcontext.h 1.10 sys/arch/mips/include/mcontext.h 1.22 sys/arch/or1k/include/mcontext.h 1.2 sys/arch/powerpc/include/mcontext.h 1.18 sys/arch/riscv/include/mcontext.h 1.5 sys/arch/sh3/include/mcontext.h 1.11 sys/arch/sparc/include/mcontext.h 1.14-1.17 sys/arch/sparc64/include/mcontext.h 1.10 sys/arch/vax/include/mcontext.h 1.9 tests/lib/libc/sys/Makefile 1.50 tests/lib/libc/sys/t_ucontext.c 1.2-1.5 sys/arch/hppa/include/mcontext.h 1.10 sys/arch/ia64/include/mcontext.h 1.7 - Introduce _UC_MACHINE_FP(). _UC_MACHINE_FP() is a helper macro to extract from mcontext a frame pointer. - Add new tests in lib/libc/sys/t_ucontext: * ucontext_sp (testing _UC_MACHINE_SP) * ucontext_fp (testing _UC_MACHINE_FP) * ucontext_pc (testing _UC_MACHINE_PC) * ucontext_intrv (testing _UC_MACHINE_INTRV) Add a dummy implementation of _UC_MACHINE_INTRV() for ia64. Implement _UC_MACHINE_INTRV() for hppa. Make the t_ucontext.c test more portable. We now have _UC_MACHINE_FP.
1.12.22.2	26-Feb-2018	snj	revert ticket 552, which broke the build
1.12.22.1	25-Feb-2018	snj	Pull up following revision(s) (requested by kamil in ticket #552): sys/arch/aarch64/include/mcontext.h: 1.2 sys/arch/alpha/include/mcontext.h: 1.9 sys/arch/amd64/include/mcontext.h: 1.19 sys/arch/arm/include/mcontext.h: 1.19 sys/arch/hppa/include/mcontext.h: 1.9 sys/arch/i386/include/mcontext.h: 1.14 sys/arch/ia64/include/mcontext.h: 1.6 sys/arch/m68k/include/mcontext.h: 1.10 sys/arch/mips/include/mcontext.h: 1.22 sys/arch/or1k/include/mcontext.h: 1.2 sys/arch/powerpc/include/mcontext.h: 1.18 sys/arch/riscv/include/mcontext.h: 1.5 sys/arch/sh3/include/mcontext.h: 1.11 sys/arch/sparc/include/mcontext.h: 1.14-1.17 sys/arch/sparc64/include/mcontext.h: 1.10 sys/arch/vax/include/mcontext.h: 1.9 tests/lib/libc/sys/Makefile: 1.50 tests/lib/libc/sys/t_ucontext.c: 1.2 Introduce _UC_MACHINE_FP() as a macro _UC_MACHINE_FP() is a helper macro to extract from mcontext a frame pointer. Don't rely on this interface as a compiler might strip frame pointer or optimize it making this interface unreliable. For hppa assume a small frame context, for larger frames FP might be located in a different register (4 instead of 3). For ia64 there is no strict frame pointer, and registers might rotate. Reuse 79 following: ./gcc/config/ia64/ia64.h:#define HARD_FRAME_POINTER_REGNUM LOC_REG (79) Once ia64 will mature, this should be revisited. A macro can encapsulate a real function for extracting Frame Pointer on more complex CPUs / ABIs. For the remaining CPUs, reuse standard register as defined in appropriate ABI. The direct users of this macro are LLVM and GCC with Sanitizers. Proposed on tech-userlevel@. Sponsored by <The NetBSD Foundation> -- Improve _UC_MACHINE_FP() for SPARC/SPARC64 Introduce a static inline function _uc_machine_fp() that contains improved caluclation of a frame pointer. Algorithm: uptr stk_ptr; # if defined (__arch64__) stk_ptr = (uptr ) (sp + 2047); # else stk_ptr = (uptr ) sp; # endif bp = stk_ptr[15]; Noted by <mrg> -- Make _UC_MACHINE_FP() compile again and fix it so that it does not add the offset twice. -- fix _UC_MACHINE32_FP() -- use 32 bit pointer value so that [15] is the right offset. do this by using __greg32_t, which is only in the sparc64 version, and these are only useful there, so move them. -- Add new tests in lib/libc/sys/t_ucontext New tests: - ucontext_sp - ucontext_fp - ucontext_pc - ucontext_intrv They test respectively: - _UC_MACHINE_SP - _UC_MACHINE_FP - _UC_MACHINE_PC - _UC_MACHINE_INTRV These tests attempt to access and print the values from ucontext, without interpreting the values. This is a follow up of the _UC_MACHINE_FP() introduction. These tests use PRIxREGISTER, and require to be built with -D_KERNTYPES. Sponsored by <The NetBSD Foundation>
1.12.6.1	28-Aug-2017	skrll	Sync with HEAD
1.14.4.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.16.2.1	02-Aug-2025	perseant	Sync with HEAD
1.5	15-May-2003	wiz	Don't install machine/mouse.h on i386 and prep; there are no users for the old busmouse interface any longer, we use wsmouse now.
1.4	27-Oct-1994	cgd	new RCS ID format.
1.3	20-Dec-1993	mycroft	branches: 1.3.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.2	02-Aug-1993	mycroft	branches: 1.2.2; Add RCS identifiers, remove some completely useless RCS logs and patchkit headers, and a few other insignificant changes.
1.1	13-Jun-1993	andrew	Microsoft/Logitech busmouse ioctls.
1.2.2.1	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.3.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.3.2.1	20-Dec-1993	mycroft	file mouse.h was added on branch magnum on 1993-12-20 05:25:25 +0000
1.3	11-May-2003	fvdl	Redirect to x86/
1.2	10-Jan-2003	fvdl	Pass acpi_softc down to mpacpi functions.
1.1	07-Jan-2003	fvdl	branches: 1.1.2; Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.1.2.3	15-Jan-2003	thorpej	Sync with HEAD.
1.1.2.2	07-Jan-2003	thorpej	Sync with HEAD.
1.1.2.1	07-Jan-2003	thorpej	file mpacpi.h was added on branch nathanw_sa on 2003-01-07 21:11:47 +0000
1.4	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.3	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file mpbiosreg.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.4	23-Sep-2000	sommerfeld	Eliminate magic number -- use MPS_ALL_APICS instead of 0xff
1.1.2.3	29-Feb-2000	sommerfeld	Multiple include protection.
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	"register" definitions for the Intel MP BIOS (spec rev 1.4). These data structures are defined by the BIOS and are read by our MP BIOS driver to figure out the configuration of a multiprocessor PC.
1.2.2.3	07-Jan-2003	thorpej	Sync with HEAD.
1.2.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.2.2.1	01-Oct-2002	nathanw	file mpbiosreg.h was added on branch nathanw_sa on 2002-10-18 02:37:56 +0000
1.6	04-Mar-2003	fvdl	Enable the define for EISA MPBIOS support.
1.5	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.4	07-Jan-2003	fvdl	Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.3	06-Oct-2002	fvdl	branches: 1.3.2; Add cpu_id field to mp_intr_map structure. Declare mp_nintr.
1.2	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	20-Feb-2000	sommerfeld	branches: 1.1.2; 1.1.4; file mpbiosvar.h was initially added on branch sommerfeld_i386mp_1.
1.1.4.1	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.2.5	25-Jun-2002	sommerfeld	Resynch with -current.
1.1.2.4	21-Aug-2000	sommerfeld	Give systems reporting EISA interrupts in the mpbios tables a prayer of working.
1.1.2.3	29-Feb-2000	sommerfeld	mpbiosreg.h
1.1.2.2	21-Feb-2000	sommerfeld	RCS ID patrol
1.1.2.1	20-Feb-2000	sommerfeld	MP BIOS support, including some of the interrupt-routing data structures built after parsing the MP BIOS information.
1.3.2.3	07-Jan-2003	thorpej	Sync with HEAD.
1.3.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.3.2.1	06-Oct-2002	nathanw	file mpbiosvar.h was added on branch nathanw_sa on 2002-10-18 02:37:56 +0000
1.2	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.1	07-Jan-2003	fvdl	branches: 1.1.2; Add various external declarations to accomodate ACPI MP probing as an alternate method to MPBIOS probing. Move some stuff common to the two methods into mpconfig.h
1.1.2.2	07-Jan-2003	thorpej	Sync with HEAD.
1.1.2.1	07-Jan-2003	thorpej	file mpconfig.h was added on branch nathanw_sa on 2003-01-07 21:11:49 +0000
1.3	16-Jan-1994	mycroft	Clean up deleted files.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.8	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.7	19-Feb-2003	fvdl	Revert previous. This file is also used on x86_64, you see.
1.6	09-Feb-2003	jdolecek	make k6_mtrr_funcs, i686_mtrr_funcs and mtrr_funcs const remove extern definition for i686_mtrr_funcs from include/mtrr.h, it's not used outside mtrr_i686.c
1.5	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.4	19-Sep-2001	thorpej	branches: 1.4.2; Add code to frob the MTRR-like registers on the AMD K6-2 and AMD K6-III.
1.3	19-Sep-2001	thorpej	Define UWCCR bits for the AMD K6 (these are its MTRR-like registers).
1.2	10-Sep-2001	fvdl	MTRR include file, based on the one by Bill Sommerfeld from the i386 mp branch.
1.1	23-Sep-2000	sommerfeld	branches: 1.1.2; 1.1.4; 1.1.6; file mtrr.h was initially added on branch sommerfeld_i386mp_1.
1.1.6.1	01-Oct-2001	fvdl	Catch up with -current.
1.1.4.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.1.4.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.1.4.1	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.1.2.2	22-Sep-2001	sommerfeld	Merge branch with current, post ubcperf, MTRR, etc., etc., with the recent fifo_putpages fix to the trunk, GENERIC.MPDEBUG config boots multiuser on my BP6.
1.1.2.1	23-Sep-2000	sommerfeld	Preliminary support for cloning MTRR values between CPU's at boot time. XXX no API to set MTRR values yet.
1.4.2.2	21-Sep-2001	nathanw	Catch up to -current.
1.4.2.1	19-Sep-2001	nathanw	file mtrr.h was added on branch nathanw_sa on 2001-09-21 22:35:09 +0000
1.11	18-Oct-2019	manu	Multiboot2 kernel support for i386 That implementation works either with BIOS or UEFI bootstrap This requires the following kernel changes: Add UEFI boot services and I/O method protoypes src/sys/arch/x86/include/efi.h 1.8 - 1.9 Fix EFI system table mapping in virtual space src/sys/arch/x86/x86/efi.c 1.19 - 1.20 Make sure no bioscall is issued when booting off UEFI system src/sys/arch/i386/i386/machdep.c 1.821 - 1.822 src/sys/arch/i386/pci/piixpcib.c 1.22 - 1.23 And the following bootstrap changes: Add kernel symbols for multiboot1 src/sys/arch/i386/stand/lib/exec_multiboot1.c 1.2 - 1.3 src/sys/arch/i386/stand/lib/libi386.h 1.45 - 1.47 Fix kernel symbols for multiboot2 src/sys/arch/i386/stand/lib/exec_multiboot2.c 1.2 - 1.3
1.10	13-Nov-2018	mlelstv	Framebuffer definitions for multiboot protocol.
1.9	13-Apr-2018	khorben	branches: 1.9.2; Correct discrepancy with the Multiboot specification The VBE mode was missing, and the types of the subsequent VBE members were also wrong. Nothing in NetBSD's base seems to be using this, and therefore nothing is expected to break as a result of this fix, or any binary to change for that matter. The latest specification (as of today) can be found at: https://www.gnu.org/software/grub/manual/multiboot/multiboot.html This was already reported in misc/52366, and addresses part of it. Patch sent to port-i386@. Build-tested on NetBSD/amd64.
1.8	22-Feb-2009	ahoka	branches: 1.8.62; Add a missing opt_multiboot.h into sys/arch/i386/i386/machdep.c. Also do the missed rename of multiboot_ksyms_init to multiboot_ksyms_addsyms_elf to compile with MULTIBOOT set. This caused a minor and a more serious bug in the past: - dmesg did not contain the information about the loader - /dev/ksyms did not work when the kernel was booted with a multiboot bootloader (grub for example) Ok by jmcneill, joerg.
1.7	11-Oct-2008	joerg	branches: 1.7.2; 1.7.8; Add "multiboot" command to boot multiboot compliant kernels like Xen. Patch provided by Robert Swindell with fixes for the command line parsing and addition of passing module options from me. The kernel now always gets the full string for modules like for the command line, everything before the first space/tab is the path name of the module.
1.6	28-Apr-2008	martin	branches: 1.6.6; Remove clause 3 and 4 from TNF licenses
1.5	21-Feb-2007	thorpej	branches: 1.5.42; 1.5.44; 1.5.46; Replace the Mach-derived boolean_t type with the C99 bool type. A future commit will replace use of TRUE and FALSE with true and false.
1.4	25-Oct-2006	jmmv	branches: 1.4.4; Remove the usage of Multiboot's "a.out kludge" to tell the boot loader to reserve some more space for the BSS section than the binary says. This trick was used to leave room after the kernel's image to copy the symbol table following the format required by ksyms_init. (It was also used to workaround a bug in the physical address fields of the binary, but this has been long fixed.) Yes, the MULTIBOOT_SYMTAB_SPACE option goes away; yay! Instead, copy the required data after the kernel in a way that avoids having to reserve space and use the new ksyms_init_explicit function to avoid the need to construct a minimal ELF image. Fixes ksyms when using an "unpatched" GRUB (one that does not contain the fix to honour the "a.out kludge" for ELF images, even when present) -- i.e. ddb and lkms. As a side effect, the new code is much clearer to read and digest. Closes PR port-i386/32865.
1.3	04-Feb-2006	jmmv	branches: 1.3.2; 1.3.14; 1.3.18; 1.3.20; 1.3.22; Revert yesterday's change that attempted to fix the detection of the boot device when using a Multiboot boot loader. It couldn't work because these boot loaders do not pass a checksum of the disk so matchbiosdisk() cannot really find any matches. I should have gone to sleep before commiting... Found by xtraeme@.
1.2	03-Feb-2006	jmmv	branches: 1.2.2; When booting an i386 kernel with Multiboot, properly detect the boot device by looking it up in the x86_alldisks table (instead of trying to match it to 'wd*' manually). In order to do this, move the cpu_rootconf function from x86 common code to amd64 and i386 specific one. This way, i386 can do an extra step (call the appropriate Multiboot code) in the appropriate place (after x86_matchbiosdisks and before findroot()).
1.1	03-Feb-2006	jmmv	Implement support for 'The Multiboot Specification' so that i386 kernels can be booted directly from Multiboot-compliant boot loaders (e.g. GRUB). See the added multiboot(8) manual page for more information. No objections in tech-kern@; only positive comments.
1.2.2.1	22-Apr-2006	simonb	Sync with head.
1.3.22.1	10-Dec-2006	yamt	sync with head.
1.3.20.2	09-Sep-2006	rpaulo	sync with head
1.3.20.1	04-Feb-2006	rpaulo	file multiboot.h was added on branch rpaulo-netinet-merge-pcb on 2006-09-09 02:40:14 +0000
1.3.18.1	18-Nov-2006	ad	Sync with head.
1.3.14.4	26-Feb-2007	yamt	sync with head.
1.3.14.3	30-Dec-2006	yamt	sync with head.
1.3.14.2	21-Jun-2006	yamt	sync with head.
1.3.14.1	04-Feb-2006	yamt	file multiboot.h was added on branch yamt-lazymbuf on 2006-06-21 14:52:30 +0000
1.3.2.2	18-Feb-2006	yamt	sync with head.
1.3.2.1	04-Feb-2006	yamt	file multiboot.h was added on branch yamt-uio_vmspace on 2006-02-18 15:38:37 +0000
1.4.4.1	27-Feb-2007	yamt	- sync with head. - move sched_changepri back to kern_synch.c as it doesn't know PPQ anymore.
1.5.46.2	04-May-2009	yamt	sync with head.
1.5.46.1	16-May-2008	yamt	sync with head.
1.5.44.1	18-May-2008	yamt	sync with head.
1.5.42.2	17-Jan-2009	mjf	Sync with HEAD.
1.5.42.1	02-Jun-2008	mjf	Sync with HEAD.
1.6.6.1	19-Oct-2008	haad	Sync with HEAD.
1.7.8.2	01-Nov-2009	jym	Sync with HEAD.
1.7.8.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.7.2.1	03-Mar-2009	skrll	Sync with HEAD.
1.8.62.2	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.8.62.1	16-Apr-2018	pgoyette	Sync with HEAD, resolve some conflicts
1.9.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.9.2.1	10-Jun-2019	christos	Sync with HEAD
1.1	13-Sep-2019	manu	branches: 1.1.2; 1.1.4; 1.1.12; Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h
1.1.12.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.1.12.1	13-Sep-2019	martin	file multiboot2.h was added on branch phil-wifi on 2020-04-13 08:03:52 +0000
1.1.4.2	18-Sep-2019	martin	Pull up following revision(s) [adapted, via patch] (requested by manu in ticket #1382): sys/arch/i386/include/multiboot2.h: revision 1.1 sys/arch/i386/stand/lib/exec.c: revision 1.74 sys/arch/i386/stand/efiboot/efiboot.c: revision 1.11 sys/arch/i386/stand/efiboot/bootia32/multiboot32.S: revision 1.1 sys/arch/i386/stand/lib/biosdisk.c: revision 1.52 sys/arch/i386/stand/efiboot/bootx64/efibootx64.c: revision 1.5 distrib/sets/lists/comp/md.amd64: revision 1.273 sys/arch/i386/stand/efiboot/eficons.c: revision 1.7 sys/arch/i386/stand/efiboot/bootia32/efibootia32.c: revision 1.5 distrib/sets/lists/comp/md.i386: revision 1.191 sys/arch/i386/stand/lib/libi386.h: revision 1.45 sys/arch/i386/stand/lib/bootinfo_memmap.c: revision 1.6 sys/arch/i386/stand/lib/pread.c: revision 1.8 sys/arch/i386/stand/efiboot/efiboot.h: revision 1.10 sys/arch/i386/stand/efiboot/bootia32/Makefile: revision 1.3 sys/arch/i386/stand/lib/Makefile: revision 1.47 sys/arch/i386/stand/efiboot/bootx64/Makefile: revision 1.2 sys/arch/i386/stand/lib/multiboot.S: revision 1.3 sys/arch/i386/stand/efiboot/bootx64/multiboot64.S: revision 1.1 sys/arch/i386/stand/efiboot/efimemory.c: revision 1.9 sys/arch/i386/stand/efiboot/boot.c: revision 1.16 sys/arch/i386/include/Makefile: revision 1.50 sys/arch/i386/stand/efiboot/Makefile.efiboot: revision 1.16 share/man/man8/man8.x86/boot.8: revision 1.20 sys/arch/i386/stand/boot/Makefile.boot: revision 1.73 sys/arch/i386/stand/pxeboot/Makefile: revision 1.26 sys/arch/i386/stand/lib/exec_multiboot1.c: revision 1.1 sys/arch/i386/stand/netboot/Makefile.netboot: revision 1.11 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.1 sys/arch/i386/stand/lib/biosdisk.h: revision 1.12 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.2 sys/arch/i386/stand/dosboot/Makefile: revision 1.32 sys/external/bsd/gnu-efi/dist/inc/efiapi.h (apply patch) Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h Remove debug define. It remained there unseen because it was misspelled! Remove obsoeolete BUGS note that UEFI bootloader does not support multiboot Add new multiboot2.h header.
1.1.4.1	13-Sep-2019	martin	file multiboot2.h was added on branch netbsd-8 on 2019-09-18 17:30:05 +0000
1.1.2.2	17-Sep-2019	martin	Pull up following revision(s) (requested by manu in ticket #203): sys/arch/i386/include/multiboot2.h: revision 1.1 sys/arch/i386/stand/lib/exec.c: revision 1.74 sys/arch/i386/stand/efiboot/efiboot.c: revision 1.11 sys/arch/i386/stand/efiboot/bootia32/multiboot32.S: revision 1.1 sys/arch/i386/stand/lib/biosdisk.c: revision 1.52 sys/arch/i386/stand/efiboot/bootx64/efibootx64.c: revision 1.5 sys/arch/i386/stand/efiboot/eficons.c: revision 1.7 sys/arch/i386/stand/efiboot/bootia32/efibootia32.c: revision 1.5 sys/arch/i386/stand/lib/libi386.h: revision 1.45 sys/arch/i386/stand/lib/bootinfo_memmap.c: revision 1.6 sys/arch/i386/stand/lib/pread.c: revision 1.8 sys/arch/i386/stand/efiboot/efiboot.h: revision 1.10 sys/arch/i386/stand/efiboot/bootia32/Makefile: revision 1.3 sys/arch/i386/stand/lib/Makefile: revision 1.47 sys/arch/i386/stand/efiboot/bootx64/Makefile: revision 1.2 sys/arch/i386/stand/lib/multiboot.S: revision 1.3 sys/arch/i386/stand/efiboot/bootx64/multiboot64.S: revision 1.1 sys/arch/i386/stand/efiboot/efimemory.c: revision 1.9 sys/arch/i386/stand/efiboot/boot.c: revision 1.16 sys/arch/i386/include/Makefile: revision 1.50 sys/arch/i386/stand/efiboot/Makefile.efiboot: revision 1.16 share/man/man8/man8.x86/boot.8: revision 1.20 sys/arch/i386/stand/boot/Makefile.boot: revision 1.73 sys/arch/i386/stand/pxeboot/Makefile: revision 1.26 sys/arch/i386/stand/lib/exec_multiboot1.c: revision 1.1 sys/arch/i386/stand/netboot/Makefile.netboot: revision 1.11 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.1 sys/arch/i386/stand/lib/biosdisk.h: revision 1.12 sys/arch/i386/stand/lib/exec_multiboot2.c: revision 1.2 sys/arch/i386/stand/dosboot/Makefile: revision 1.32 Add multiboot 2 support to x86 bootloaders multiboot 2 is required to boot Xen on an EFI system. This also require a kernel patch for properly discovering the ACPI RSDP, which is available after 20190912, in src/sys/arch/x86/acpi/acpi_machdep.c 1.26-1.28 There are a few missing bit in this multiboot 2 implementation (which are unused by Xen): - Header tags Address, Freambuffer, and Relocatable are ignored - Tags APM and Network are not provided - Tags ACPI old and ACP new are only provided for ACPI boot - Tag boot device does not provides the subpart (BSD disklabel partition) Notes: - multiboot2 is disabled in dosboot, otherwise the binary gets too big and build fails. - in src/sys/arch/i386/stand/efiboot, consinit() is renamed as efi_consinit() to avoid prototype conflicts in src/sys/sys/systm.h Remove debug define. It remained there unseen because it was misspelled! Remove obsoeolete BUGS note that UEFI bootloader does not support multiboot
1.1.2.1	13-Sep-2019	martin	file multiboot2.h was added on branch netbsd-9 on 2019-09-17 19:31:59 +0000
1.2	09-Feb-2007	ad	branches: 1.2.4; Merge newlock2 to head.
1.1	10-Sep-2006	ad	branches: 1.1.2; file mutex.h was initially added on branch newlock2.
1.1.2.4	29-Dec-2006	ad	Checkpoint work in progress.
1.1.2.3	24-Oct-2006	ad	Compile fixes
1.1.2.2	20-Oct-2006	ad	- Don't need locked bus cycles on release from C code. - Save an integer ID in the lock structures for LOCKDEBUG code.
1.1.2.1	10-Sep-2006	ad	Add updated locking primatives.
1.2.4.2	26-Feb-2007	yamt	sync with head.
1.2.4.1	09-Feb-2007	yamt	file mutex.h was added on branch yamt-lazymbuf on 2007-02-26 09:07:02 +0000
1.1	17-Mar-2002	thorpej	branches: 1.1.2; file mutex_impl.h was initially added on branch newlock.
1.1.2.5	22-Mar-2002	thorpej	Wrap the stuff that manipulates the guts of a mutex in __MUTEX_PRIVATE.
1.1.2.4	22-Mar-2002	thorpej	Add debug info structures if MUTEX_DEBUG or RWLOCK_DEBUG is defined. XXX Revisit, so that the structure doesn't change when debugging is turned on; do it out-of-band?
1.1.2.3	19-Mar-2002	thorpej	Shuffle around the mutex structure and change some types, so that now the mutex structure is 1/2 the size it was before (8 bytes vs 16).
1.1.2.2	18-Mar-2002	thorpej	Remove an untrue comment.
1.1.2.1	17-Mar-2002	thorpej	Add first-cut mutex and rwlock implementation bits for the x86. XXX Does not actually handle i386 CPUs yet, only i486 and later.
1.36	12-Feb-2014	dsl	Change i386 to use x86/fpu.c instead of i386/isa/npx.c This changes the trap10 and trap13 code to call directly into fpu.c, removing all the code for T_ARITHTRAP, T_XMM and T_FPUNDA from i386/trap.c Not all of the code thate appeared to handle fpu traps was ever called! Most of the changes just replace the include of machine/npx.h with x86/fpu.h (or remove it entirely).
1.35	07-Feb-2014	dsl	Convert the amd64 build to use x86/cpu_extended_state.h so that the fpu definitions match those of i386. Mostly just structure and field renames, in addition: 1) process_xmm_to_s87() and process_s87_to_xmm() moved into x86/convert_xmm_s87.c so they can be used by amd64's netbsd32 code. 2) The linux signal code simplified to use a structure copy for ths fxsave data - it matches the hardware definition and won't change.
1.34	07-Feb-2014	dsl	Move all the hardware register layout for the x86 cpus into a header that can also be used by amd64. Add in skeleton definitions for XSAVE and AVX. Update some comments to match reality.
1.33	04-Feb-2014	dsl	Remove the fields from 'struct save87' and 'struct fxsave' that were used to save the control word and tag word on fp faults. Nothing I could find looked at them before I removed the code that wrote them. It might appear that these fields got exported to userspace (via mcontext), but that code carefully avoided those fields. I've also deleted some very historic fields from the union inside __fpregset_t this doesn't affect its size. The __fpregset_t fields are now the same size as the ones in the pcb. Which is good because they get cast to each other in a few places.
1.32	26-Jan-2014	dsl	Remove support for 'external' floating point units and the MS-DOS compatible method of handling floating point exceptions. Make kernel support for teh fpu non-optional (486SX should still work). Only 386 cpus support external fpu, and i386 support was removed years ago. This means that the npx code no longer uses port 0xf0 or interupt 13. All the "npx at isa" lines go from the configs, arch/i386/isa/npx.c is now mandatory for all i386 kernels. I've renamed npxinit() to fpuinit() and npxinit_cpu() to fpuinit_cpu() to match the very similar amd64 functions. The fpu of the boot cpu is now initialised by a direct call from cpu_configure(), this enables FP emulation for a 486SX. (for amd64 the cr0 values are set in locore.S and similar). This fixes a long-standing bug in linux_setregs() - which did not save the fpu regsiters if they were active. I've test booted a single cpu i386 kernel (using anita). amd64 builds - none of teh changes should affect it. The i386 XEN kernels build, but I'm not sure where they set cr0, and it might have got lost!
1.31	25-Jan-2014	dsl	Change the way the x87 fp register is defined so that the copies between fsave and fxsave layouts can be done by structure assignments.
1.30	25-Jan-2014	christos	help lint.
1.29	19-Jan-2014	dsl	Add a couple of __CTASSERT() for the sizes of the fp save structures.
1.28	19-Jan-2014	dsl	Flatten the floating point register structures bu moving the 'environment' registers into the main strcuture. There are x87 instructions that only affect the environment, but that is no real excuse for all the sub-structures. Rename fields so that there are far fewer collisions between the two sets of field names.
1.27	08-Dec-2013	dsl	Rename the fields of 'struct envxmm' (itself badly named) so that they are different from those used for the x87 fpu. Use a union to describe the 64bit 'address' field so that it correctly spilt for 32bit binaries. The freebsd emulation defined its own 'struct freebsd_env87' that the code just assumed matched the netbsd one. Since this is a hardware structure, just use the netbsd one. This shouldn't change the binaries except that 'en_fos' used to be incorrectly a 32bit field.
1.26	11-Nov-2013	joerg	NetBSD 6.99.26: Switch i386 and amd64 to the x87 default control word as initial value for new processes. This means that long double computations get the expected 63bit mantissa. Binaries tagged as compiled for 6.99.25 and older get the old value. Add a simple test case to ensure that double and long double computation are working correctly.
1.25	31-Jul-2010	joerg	branches: 1.25.8; 1.25.18; 1.25.22; Add support for fenv.h interface for i386 and amd64. Submitted by Stathis Kamperis as part of GSoC 2010 and ported from FreeBSD.
1.24	25-Jul-2010	christos	map npx control/status codes into siginfo codes.
1.23	25-Sep-2009	mlelstv	branches: 1.23.2; 1.23.4; The FPU Tag word is a 16bit register, in FPU (387) mode it defines 2-bit tags for each FPU data register, in MMX mode it defines 1-bit tags for each data register. The single bit tags are stored in the lower 8 bits and thus in the first byte of the save frame. See amd64/include/fpu.h and the IA-32 Software Developer's manual Vol 2A.
1.22	02-May-2006	drochner	branches: 1.22.12; 1.22.66; 1.22.76; 1.22.80; 1.22.84; -explicitely align "struct savexmm" so that we don't get surprised if we play with the PCB -remove historical "struct emcsts"; there is no support in context switching code for ages (if there has ever been)
1.21	10-Oct-2005	christos	branches: 1.21.10; 1.21.12; 1.21.14; 1.21.16; 1.21.18; Add defines for the control and status bits of the 80387.
1.20	07-Aug-2003	agc	branches: 1.20.2; 1.20.6; 1.20.14; 1.20.16; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.19	01-Oct-2002	fvdl	branches: 1.19.6; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.18	03-Aug-2001	thorpej	Upon further reading of the manual, don't save the MXCSR-at-last-exception. Its status bits are sticky, and unaffacted by FNINIT.
1.17	03-Aug-2001	thorpej	In setregs(), initialize the process's MXCSR to the reset-default value as documented in the IA-32 Instruction Set Reference (in the description of the LDMXCSR insn).
1.16	03-Aug-2001	thorpej	Remember the MXCSR at last-exception the way we do the FPU SW/TW.
1.15	02-Aug-2001	thorpej	Add support for saving/restoring SSE/SSE2 state using FXSAVE/FXRSTOR. Reviewed by Frank.
1.14	26-Jan-1999	christos	branches: 1.14.16; 1.14.22; 1.14.24; add prototypes for the assembly functions that moved to locore.s
1.13	24-Jan-1998	mycroft	Add some comments about what the control words do.
1.12	24-Jan-1998	mycroft	Store the initial NPX control word appropriate for the emulation in the saved FPU state, and use it when initializing the FPU on demand.
1.11	27-Oct-1994	cgd	branches: 1.11.20; new RCS ID format.
1.10	22-Sep-1994	jtc	Updated bde's comments wrt. i387 control word settings so that they describe NetBSD, rather than 386BSD 0.1. Removed 386BSD and BDE control word constants, since we don't need and will never use them. However, The iBCS control word constant is kept, since we might want to use it in the iBCS binary compatibility code.
1.9	25-Jan-1994	jtc	branches: 1.9.2; 1.9.4; FPU control word of 0x137f doesn't work well in "real" world applications so I've changed it to 0x127f. 64 bit precision causes errors in paranoia and our math library, changing to 53 bit precision solves these problems. I've been told that long double support in Gcc 2.5 requires 64 bit mode, so a comprehensive FP strategy (compiler, library, FPU, emulator) will have to be addressed in the future.
1.8	17-Jan-1994	jtc	My previous checkin didn't really change the FPU Exception Mask -- it was trapped inside an #ifdef -- this commit changes it as I had originally intended.
1.7	08-Jan-1994	jtc	Change default NetBSD NPX control word to 0x137f. This masks all FP exceptions (like every other UNIX system), and makes it possible for gcc 2.5's long double support to work.
1.6	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.5	07-Dec-1993	deraadt	__386BSD__ -> __NetBSD__
1.4	29-Jul-1993	jtc	branches: 1.4.2; Change "#endif FOO" to "#endif /* FOO */", to keep gcc -Wall, lint, etc. from complaining about a trivial issue.
1.3	22-May-1993	cgd	add rcsids to everything and clean up headers
1.2	09-May-1993	deraadt	npx patches from Bruce Evans. patchkit 10002
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.4.2.1	10-Dec-1993	cgd	386bsd -> NetBSD
1.9.4.1	06-Oct-1994	mycroft	Update from trunk.
1.9.2.2	25-Jan-1994	jtc	FPU control word of 0x137f doesn't work well in "real" world applications so I've changed it to 0x127f. 64 bit precision causes errors in paranoia and our math library, changing to 53 bit precision solves these problems. I've been told that long double support in Gcc 2.5 requires 64 bit mode, so a comprehensive FP strategy (compiler, library, FPU, emulator) will have to be addressed in the future.
1.9.2.1	25-Jan-1994	jtc	file npx.h was added on branch magnum on 1994-01-25 19:58:31 +0000
1.11.20.1	29-Jan-1998	mellon	Pull up 1.12 and 1.13 (mycroft)
1.14.24.2	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.14.24.1	03-Aug-2001	lukem	update to -current
1.14.22.2	18-Oct-2002	nathanw	Catch up to -current.
1.14.22.1	24-Aug-2001	nathanw	Catch up with -current.
1.14.16.2	03-Sep-2001	sommerfeld	Merge with -current once more.
1.14.16.1	10-Jan-2001	sommerfeld	Significant rewrite of slave CPU attach and initialization, centering on identifycpu() and spreading outward. Separate feature probes (cpuid invocation), which runs on the cpu being attached, from the display of this information, which runs on the boot cpu; remove some of the probe code from locore as it doesn't need to run that early. We now spin up secondary cpu's during cpu_attach, have them do a cpu feature probe, and then have them spin waiting for a "go" signal. We no longer pay attention to the cpu signature and cpuid features reported by the MP BIOS since those are known to be truncated for at least some bioses. We now do npx initialization (including FDIV bug detection) on all cpus. XXX Change some of the cyrix bug workarounds to fit the new identifycpu() system; this is untested.
1.19.6.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.19.6.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.19.6.2	18-Sep-2004	skrll	Sync with HEAD.
1.19.6.1	03-Aug-2004	skrll	Sync with HEAD
1.20.16.1	21-Jun-2006	yamt	sync with head.
1.20.14.1	11-Oct-2005	tron	Pull up following revision(s) (requested by christos in ticket #880): sys/arch/i386/include/npx.h: revision 1.21 Add defines for the control and status bits of the 80387.
1.20.6.1	11-Oct-2005	riz	Pull up following revision(s) (requested by christos in ticket #5923): sys/arch/i386/include/npx.h: revision 1.21 Add defines for the control and status bits of the 80387.
1.20.2.1	11-Oct-2005	riz	Pull up following revision(s) (requested by christos in ticket #5923): sys/arch/i386/include/npx.h: revision 1.21 Add defines for the control and status bits of the 80387.
1.21.18.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.21.16.1	11-May-2006	elad	sync with head
1.21.14.1	24-May-2006	yamt	sync with head.
1.21.12.1	01-Jun-2006	kardel	Sync with head.
1.21.10.1	09-Sep-2006	rpaulo	sync with head
1.22.84.1	21-Apr-2010	matt	sync to netbsd-5
1.22.80.2	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.22.80.1	01-Nov-2009	jym	Sync with HEAD.
1.22.76.1	03-Oct-2009	snj	Pull up following revision(s) (requested by mlelstv in ticket #1050): sys/arch/i386/include/npx.h: revision 1.23 The FPU Tag word is a 16bit register, in FPU (387) mode it defines 2-bit tags for each FPU data register, in MMX mode it defines 1-bit tags for each data register. The single bit tags are stored in the lower 8 bits and thus in the first byte of the save frame. See amd64/include/fpu.h and the IA-32 Software Developer's manual Vol 2A.
1.22.66.2	11-Aug-2010	yamt	sync with head.
1.22.66.1	11-Mar-2010	yamt	sync with head
1.22.12.1	18-Oct-2009	bouyer	Pull up following revision(s) (requested by mlelstv in ticket #1362): sys/arch/i386/include/npx.h: revision 1.23 The FPU Tag word is a 16bit register, in FPU (387) mode it defines 2-bit tags for each FPU data register, in MMX mode it defines 1-bit tags for each data register. The single bit tags are stored in the lower 8 bits and thus in the first byte of the save frame. See amd64/include/fpu.h and the IA-32 Software Developer's manual Vol 2A.
1.23.4.1	05-Mar-2011	rmind	sync with head
1.23.2.1	17-Aug-2010	uebayasi	Sync with HEAD.
1.25.22.1	18-May-2014	rmind	sync with head
1.25.18.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.25.8.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.89	20-Apr-2025	riastradh	t_signal_and_sp: Add i386 support. i386 too is confused by misaligned sigaltstack or esp on interrupt. PR kern/59327: user stack pointer is not aligned properly
1.88	31-May-2021	simonb	branches: 1.88.18; Include "opt_param.h" (ifdef _KERNEL_OPT) everywhere that MSGBUFSIZE is referenced since some sources include <machine/param.h>.
1.87	05-Mar-2020	riastradh	branches: 1.87.8; Revert "Include opt_diagnostic.h for DIAGNOSTIC." This did not do what I thought it did. opt_diagnostic.h is only for the unused _DIAGNOSTIC, which seems like an abortive attempt to incrementally convert DIAGNOSTIC to an opt_*.h option rather than a command-line option.
1.86	05-Mar-2020	riastradh	Include opt_diagnostic.h for DIAGNOSTIC. ...at least, in header files, which may not have already included libkern.h.
1.85	20-Aug-2019	riastradh	New macro ALIGNED_POINTER_LOAD. To be used with ALIGNED_POINTER(p,t) instead of writing (const t )p directly. This way, on machines without strict alignment, we can use memcpy to pacify sanitizers, while getting the same compiled code in the end with a single (say) MOV instruction.
1.84	07-Jan-2019	jdolecek	branches: 1.84.4; move DEV_BSIZE, DEV_BSHIFT out of MD param.h, they are same on all ports also move BLKDEV_IOSIZE, MAXPHYS, but allow override since some ports have different value (powerpc uses NBPG for BLKDEV_IOSIZE, sun2/sun3 have lower MAXPHYS)
1.83	19-Feb-2018	sborrill	branches: 1.83.2; 1.83.4; Double size of MSGBUFSIZE as existing value is not big enough to hold boot dmesg on modern server-class hardware with lots of CPUs, etc.
1.82	29-Jul-2017	maxv	Remove unused.
1.81	14-Jun-2017	maxv	Define MAXPHYSMEM globally.
1.80	20-Jan-2017	maya	branches: 1.80.6; increase max io mem on amd64. some devices need it.
1.79	26-Aug-2016	mlelstv	branches: 1.79.2; The 64MB was intended.
1.78	27-Oct-2015	mrg	branches: 1.78.2; make sure MSGBUFSIZE can't expand strangely by using parens.
1.77	20-Apr-2012	rmind	branches: 1.77.2; 1.77.14; 1.77.16; - Convert x86 MD code, mainly pmap(9) e.g. TLB shootdown code, to use kcpuset(9) and thus replace hardcoded CPU bitmasks. This removes the limitation of maximum CPUs. - Support up to 256 CPUs on amd64 architecture by default. Bug fixes, improvements, completion of Xen part and testing on 64-core AMD Opteron(tm) Processor 6282 SE (also, as Xen HVM domU with 128 CPUs) by Manuel Bouyer.
1.76	10-Feb-2012	para	branches: 1.76.2; proper sizing of kmem_arena on different ports PR port-i386/45946: Kernel locks up in VMEM system
1.75	04-Feb-2012	para	improve sizing of kmem_arena now that more allocations are made from it don't enforce limits if not required ok: riz@
1.74	24-Jan-2012	christos	Use and define ALIGN() ALIGN_POINTER() and STACK_ALIGN() consistently, and avoid definining them in 10 different places if not needed.
1.73	20-Jan-2012	joerg	Change CMSG_SPACE and CMSG_LEN to provide Integer Constant Expressions again. This was changed in sys/socket.h r1.51 to work around fallout from the IPv6 aux data migration. It broke the historic ABI on some platforms. This commit restores compatibility for netbsd32 code on such platforms and provides a template for future changes to the CMSG_* alignment. Revert PCC/Clang workarounds in postfix and tmux.
1.72	08-Feb-2010	joerg	branches: 1.72.10; 1.72.14; Remove separate mb_map. The nmbclusters is computed at boot time based on the amount of physical memory and limited by NMBCLUSTERS if present. Architectures without direct mapping also limit it based on the kmem_map size, which is used as backing store. On i386 and ARM, the maximum KVA used for mbuf clusters is limited to 64MB by default. The old default limits and limits based on GATEWAY have been removed. key_registered_sb_max is hard-wired to a value derived from 2048 clusters.
1.71	10-Mar-2009	bouyer	branches: 1.71.2; More i386PAE fixes: - x86_round_page, x86_trunc_page, x86_btop and x86_ptob macros are used with physical addresses; cast to paddr_t instead of u_long. Issue pointed out by jym@ - machine_to_phys_mapping[] is a long. This is fine as it holds page frame numbers (and this fits in a 32bit int as physical addresses are only 36bits), but cast to paddr_t before << PAGE_SHIFT - xen_start_info.store_mfn is a long; cast it to paddr_t before << PAGE_SHIFT. should fix issue pointed out by cegger@
1.70	20-Dec-2008	ad	branches: 1.70.2; Revert part of previous.
1.69	20-Dec-2008	ad	opt_noredzone.h is gone.
1.68	20-Dec-2008	ad	- Kill NOREDZONE. - Make the redzone conditional on DIAGNOSTIC. - Give amd64 an additional page for the uarea. 2 is not enough.
1.67	18-Oct-2007	yamt	branches: 1.67.16; 1.67.20; 1.67.28; 1.67.30; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.66	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.65	13-Oct-2007	joerg	branches: 1.65.2; Bump default size of the message buffer from 16 KB to 32 KB. This is large enough that boot -v on most systems fits into the message buffer, which makes it easier for debugging.
1.64	03-Oct-2007	ad	Back out the ALIGNBYTES change. There are structures used in userspace that depend on the constant.
1.63	26-Sep-2007	ad	x86 changes for pcc and LKMs. - Replace most inline assembly with proper functions. As a side effect this reduces the size of amd64 GENERIC by about 120kB, and i386 by a smaller amount. Nearly all of the inlines did something slow, or something that does not need to be fast. - Make curcpu() and curlwp functions proper, unless __GNUC__ && _KERNEL. In that case make them inlines. Makes curlwp LKM and preemption safe. - Make bus_space and bus_dma more LKM friendly. - Share a few more files between the ports. - Other minor changes.
1.62	08-Dec-2006	yamt	branches: 1.62.6; 1.62.14; 1.62.22; 1.62.24; 1.62.26; - pass intrframe by-pointer, not by-value. - make i386 and xen use per-cpu interrupt stack. xen part is reviewed by Manuel Bouyer.
1.61	28-Aug-2006	yamt	branches: 1.61.2; 1.61.4; - remove unused bdbtofsb. - move the following macros from MD headers to sys/param.h. ctod dtoc ctob btoc dbtob btodb
1.60	04-Jun-2006	perry	Wrap the UPAGES definition with an #ifndef so that (in emergencies) it can be put into a kernel config file. Stack overflows should be fixed, not patched around, but this makes it easier to test if one is causing a panic.
1.59	22-May-2006	yamt	i386 and xen: move struct user to the highest address in uarea and reduce UPAGES from 4 to 3 in the case of !NOREDZONE.
1.58	07-Nov-2005	yamt	branches: 1.58.6; 1.58.8; 1.58.10; 1.58.14; 1.58.16; some assym cleanup. - move copyin and friends from locore.S to their own file, copy.S. share it between i386 and xen. - defparam KERNBASE and kill KERNBASE_LOCORE hack. - add more symbols to assym.h and use it where appropriate.
1.57	08-Aug-2003	christos	branches: 1.57.16; cleanup the mess with KERNBASE and KERNBASE_LOCORE.
1.56	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.55	18-Jul-2003	tls	Allow redefiniton of KERNBASE. Because the type of this constant is different for locore.S and the rest of the kernel, if you redefine it from the default value, you must also define KERNBASE_LOCORE -- this is ugly, but so is having the same constant with two different types depending where in the kernel it's used...
1.54	26-Feb-2003	fvdl	branches: 1.54.2; Redirect some includes to x86/foo.h
1.53	02-Nov-2002	perry	1) egg, meet face. face, meet egg. 2) re-UL KERNBASE, but this time, protect it with #ifndef _LOCORE so we don't get a big kaboom in .S files.
1.52	02-Nov-2002	kristerw	Revert previous. The assembler bombs on UL integer suffix.
1.51	02-Nov-2002	perry	define KERNBASE with a UL -- eliminates 830 or so lines of lint.
1.50	06-Oct-2002	fvdl	If NOREDZONE is defined, keep UPAGES at 2 as before.
1.49	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.48	26-Feb-2002	simonb	branches: 1.48.10; Purge CLSIZE, CLSIZELOG2 and MCLOFSET. Be consistant in the way that MSIZE, MCLSHIFT, MCLBYTES and NMBCLUSTERS are defined. Remove old VM constants from cesfic port. Bump MSIZE to 256 on mipsco (the only one that wasn't already 256).
1.47	03-Aug-2001	groo	Remove trailing "U" on KERNBASE introduced in last commit. Breaks assembly of locore.
1.46	03-Aug-2001	thorpej	Define KERNTEXTOFF in terms of KERNBASE.
1.45	30-May-2001	mrg	branches: 1.45.2; use _KERNEL_OPT
1.44	02-Apr-2001	fvdl	Override NFS read and write size to default to 32k for the i386 here (can be overridden from a config file).
1.43	19-Sep-2000	fvdl	branches: 1.43.2; Bring the default value of NMBCLUSTERS into the modern age.
1.42	13-Aug-2000	christos	cast to u_long before casting to u_int to appease lint.
1.41	30-Jun-2000	itojun	raise MSIZE from 128 to 256. - for sizeof(void *) == 8 arch, this is mandatory. MHLEN is too small already (less than 80) and there are chances for unwanted packet loss due to m_pullup restriction. - for other cases, the change should avoid allocating clusters in most cases (even when you have IPv4 IPsec tunnel, or IPv6 with moderate amount of extension header) portmasters: if your arch chokes with the change (high memory usage or whatever), please backout the change for your arch.
1.40	11-Feb-2000	thorpej	branches: 1.40.2; 1.40.6; Update for the NKMEMPAGES changes.
1.39	11-Feb-2000	thorpej	Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1G of virtual address space, leaving userland with 3G, and update comments to match the new reality. We knew we were going to have to bite this bullet eventually, and there are a couple of outstanding PRs related to this issue (9389 and 9313). Complete solution to those PRs is going to involve some sort of run-time decision on how large kmem_map should be, as well as changing some data structure allocation strategies in UVM. However, this change will at least allow the PR submitter to simply throw resources at the problem.
1.38	04-Dec-1999	ragge	CL* discarding.
1.37	10-Mar-1999	sommerfe	branches: 1.37.2; 1.37.8; 1.37.14; Increase default MSGBUFSIZE due to more and chattier drivers in -current
1.36	01-Jul-1998	mjacob	allow a config file to redefine MAXPHYS
1.35	29-Apr-1998	thorpej	Pull in opt_gateway.h as appropriate.
1.34	20-Jan-1998	drochner	remove unused and misleading KERNSIZE definition
1.33	20-Oct-1997	fvdl	Change various constants that depend on the kernel VM base address. Lower it to 0xf0000000 to give it more breathing space.
1.32	19-Sep-1997	leo	Implement the kernel part of pr-1891. This allows for a more flexible sized msgbuf. Note that old 'dmesg' and 'syslogd' binaries will continue running, though old 'dmesg' binaries will output a few bytes of junk at the start of the buffer, and will miss a few bytes at the end of the buffer.
1.31	26-Feb-1997	jonathan	branches: 1.31.4; Protect MCLSHIFT definition so users can customize MCLBYTES.
1.30	24-Feb-1997	fvdl	Define ALIGNED_POINTER
1.29	04-Mar-1996	cgd	branches: 1.29.8; add _MACHINE and _MACHINE_ARCH, which are like MACHINE and MACHINE_ARCH, execpt without quotes. meant to be __CONCAT()ted for easy #includes of machine-dependent headers for MI code (e.g. for the MI ISA/EISA/PCI/TC bus code).
1.28	01-Mar-1996	cgd	Clean up tabbing/spacing in defns of MACHINE, MACHINE_ARCH, and MID_MACHINE.
1.27	01-Feb-1996	mycroft	LOCORE -> _LOCORE
1.26	26-Jun-1995	cgd	clean up definitions of ctod(), dtoc(), ctob(), btoc(), btodb(), and dbtob(). remove unused definitions of ctos(), stoc(), and dtob().
1.25	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.24	18-Mar-1995	cgd	don't be so quick to cast macro arguments to unsigned; if the arguments are really off_t's (e.g. to btodb), then you can lose information. This was the "> 4G file systems don't work" bug; physio uses btodb, which was broken.
1.23	27-Oct-1994	cgd	new RCS ID format.
1.22	14-Oct-1994	cgd	move arch dependence to archs. clean up a bit. deal with weird MAXPARTITONS.
1.21	09-Oct-1994	mycroft	Make the conversion macros more consistent.
1.20	24-Aug-1994	mycroft	Add USPACE.
1.19	15-Aug-1994	mycroft	Eliminate struct pte and struct pde.
1.18	08-Apr-1994	mycroft	branches: 1.18.2; Only #include other stuff #ifdef KERNEL.
1.17	03-Apr-1994	mycroft	Make sure we get psl.h.
1.16	09-Mar-1994	mycroft	Back out previous change for now.
1.15	08-Mar-1994	mycroft	Inline spl*() calls.
1.14	22-Feb-1994	hpeyerl	Move some machine dependant disklabel stuff in here.
1.13	05-Jan-1994	mycroft	branches: 1.13.2; Move kernel to 0xf8000000.
1.12	22-Dec-1993	cgd	add NPTEPG back in. gdb wants it. Thanks to Rich Murphey for pointing it out.
1.11	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.10	04-Dec-1993	cgd	add MACHINE_ARCH as i386
1.9	09-Oct-1993	cgd	patch from rod grimes: Fix definitions to be in terms of a minimum number of constants by using the SHIFT value as the basis for the other values, for example instead of: #define NBPG 4096 #define PGSHIFT 12 use: #define PGSHIFT 12 #define NBPG (1 << PGSHIFT)
1.8	03-Aug-1993	mycroft	branches: 1.8.2; Add ALIGNBYTES for new fts().
1.7	05-Jul-1993	deraadt	No need for the DELAY() macro. in userland. It just polutes the namespace.
1.6	18-Jun-1993	cgd	convert magic numbers to network byte order, and attendent changes
1.5	05-Jun-1993	cgd	change MCLSHIFT to 11, so mbuf cluster size is bigger than MTU of network interfaces, for performance. to use old size, use "options MCLSHIFT=10" in your kernel config.
1.4	22-May-1993	cgd	add rcsids to everything and clean up headers
1.3	22-Apr-1993	mycroft	Make mbuf sizes tunable from config file (from 386BSD path 134).
1.2	01-Apr-1993	cgd	changed value of NKMEMCLUSTERS, to help avoid kmem_map too small problem
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.8.2.4	04-Dec-1993	cgd	add MACHINE_ARCH
1.8.2.3	26-Oct-1993	mycroft	Reduce NKMEMCLUSTERS back to 6MB. We don't need 16.
1.8.2.2	15-Oct-1993	mycroft	Major cleanup of include files and constants.
1.8.2.1	09-Oct-1993	mycroft	Merge changes from trunk.
1.13.2.2	05-Jan-1994	mycroft	Move kernel to 0xf8000000.
1.13.2.1	05-Jan-1994	mycroft	file param.h was added on branch magnum on 1994-01-05 16:02:37 +0000
1.18.2.2	11-Oct-1994	mycroft	Update from trunk.
1.18.2.1	15-Aug-1994	mycroft	update from trunk
1.29.8.1	12-Mar-1997	is	Merge in Changes from The Trunk.
1.31.4.1	22-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.37.14.2	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.37.14.1	21-Dec-1999	wrstuden	Initial commit of recent changes to make DEV_BSIZE go away. Runs on i386, needs work on other arch's. Main kernel routines should be fine, but a number of the stand programs need help. cd, fd, ccd, wd, and sd have been updated. sd has been tested with non-512 byte block devices. vnd, raidframe, and lfs need work. Non 2**n block support is automatic for LKM's and conditional for kernels on "options NON_PO2_BLOCKS".
1.37.8.2	21-Apr-2001	bouyer	Sync with HEAD
1.37.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.37.2.1	19-Feb-2000	he	Apply patch (requested by thorpej): Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1GB of virtual address space, leaving userland with 3GB, and update comments to match the new reality.
1.40.6.3	25-Apr-2001	he	Pull up revision 1.43 (requested by itojun): Bump NMBCLUSTERS to 1024 or 2048 in GATEWAY case (was 256/512).
1.40.6.2	06-Apr-2001	he	Pull up revision 1.44 (requested by fvdl): Adjust default NFS operation size back to 8KB on all systems except on i386.
1.40.6.1	23-Jul-2000	itojun	pullup from main trunc (approved by releng-1-5) raise MSIZE from 128 to 256. - for sizeof(void *) == 8 arch, this is mandatory. MHLEN is too small already (less than 80) and there are chances for unwanted packet loss due to m_pullup restriction. - for other cases, the change should avoid allocating clusters in most cases (even when you have IPv4 IPsec tunnel, or IPv6 with moderate amount of extension header) portmasters: if your arch chokes with the change (high memory usage or whatever), please backout the change for your arch. 1.1 -> 1.2 syssrc/sys/arch/amigappc/include/param.h 1.13 -> 1.14 syssrc/sys/arch/arc/include/param.h 1.1 -> 1.2 syssrc/sys/arch/arm26/include/param.h 1.17 -> 1.18 syssrc/sys/arch/arm32/include/param.h 1.6 -> 1.7 syssrc/sys/arch/bebox/include/param.h 1.4 -> 1.5 syssrc/sys/arch/cobalt/include/param.h 1.4 -> 1.5 syssrc/sys/arch/evbsh3/include/param.h 1.2 -> 1.3 syssrc/sys/arch/hpcmips/include/param.h 1.40 -> 1.41 syssrc/sys/arch/i386/include/param.h 1.7 -> 1.8 syssrc/sys/arch/m68k/include/param.h 1.9 -> 1.10 syssrc/sys/arch/macppc/include/param.h 1.4 -> 1.5 syssrc/sys/arch/mmeye/include/param.h 1.6 -> 1.7 syssrc/sys/arch/newsmips/include/param.h 1.7 -> 1.8 syssrc/sys/arch/ofppc/include/param.h 1.23 -> 1.24 syssrc/sys/arch/pc532/include/param.h 1.32 -> 1.33 syssrc/sys/arch/pmax/include/param.h 1.7 -> 1.8 syssrc/sys/arch/powerpc/include/param.h 1.1 -> 1.2 syssrc/sys/arch/prep/include/param.h 1.1 -> 1.2 syssrc/sys/arch/sgimips/include/param.h 1.3 -> 1.4 syssrc/sys/arch/sh3/include/param.h 1.44 -> 1.45 syssrc/sys/arch/sparc/include/param.h 1.21 -> 1.22 syssrc/sys/arch/sparc64/include/param.h 1.43 -> 1.44 syssrc/sys/arch/vax/include/param.h
1.40.2.7	27-Apr-2002	sommerfeld	Merge with current as of today
1.40.2.6	03-Sep-2001	sommerfeld	Merge with -current once more.
1.40.2.5	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.40.2.4	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.40.2.3	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.40.2.2	07-Aug-2000	sommerfeld	Merge sommerfeld_i386mp_1 branch forward to 20000806 sources
1.40.2.1	20-Feb-2000	sommerfeld	Bump MSGBUFSIZE since mp_verbose is pretty verbose.
1.43.2.6	11-Nov-2002	nathanw	Catch up to -current
1.43.2.5	18-Oct-2002	nathanw	Catch up to -current.
1.43.2.4	28-Feb-2002	nathanw	Catch up to -current.
1.43.2.3	24-Aug-2001	nathanw	Catch up with -current.
1.43.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.43.2.1	09-Apr-2001	nathanw	Catch up with -current.
1.45.2.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.45.2.2	16-Mar-2002	jdolecek	Catch up with -current.
1.45.2.1	03-Aug-2001	lukem	update to -current
1.48.10.1	13-Aug-2003	grant	Pull up revision 1.55 via patch (requested by tls in ticket #1387): Allow redefiniton of KERNBASE. Because the type of this constant is different for locore.S and the rest of the kernel, if you redefine it from the default value, you must also define KERNBASE_LOCORE -- this is ugly, but so is having the same constant with two different types depending where in the kernel it's used...
1.54.2.4	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.54.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.54.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.54.2.1	03-Aug-2004	skrll	Sync with HEAD
1.57.16.3	27-Oct-2007	yamt	sync with head.
1.57.16.2	30-Dec-2006	yamt	sync with head.
1.57.16.1	21-Jun-2006	yamt	sync with head.
1.58.16.1	19-Jun-2006	chap	Sync with head.
1.58.14.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.58.10.3	03-Sep-2006	yamt	sync with head.
1.58.10.2	26-Jun-2006	yamt	sync with head.
1.58.10.1	24-May-2006	yamt	sync with head.
1.58.8.2	07-Jun-2006	kardel	Sync with head.
1.58.8.1	01-Jun-2006	kardel	Sync with head.
1.58.6.1	09-Sep-2006	rpaulo	sync with head
1.61.4.1	10-Dec-2006	yamt	sync with head.
1.61.2.1	12-Jan-2007	ad	Sync with head.
1.62.26.3	14-Oct-2007	yamt	sync with head.
1.62.26.2	08-Oct-2007	yamt	remove an #ifdef _LOCORE.
1.62.26.1	23-Sep-2007	yamt	- sync with amd64. - pmap_alloc_level: remove fast call special case. - tweak some printf to work for both of LP64 and LP32. - remove machdep.nkpde sysctl. - remove automatic adjustment of nkpde for now. will revisit later.
1.62.24.1	06-Nov-2007	matt	sync with HEAD
1.62.22.3	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.62.22.2	04-Oct-2007	joerg	Sync with HEAD.
1.62.22.1	02-Oct-2007	joerg	Sync with HEAD.
1.62.14.2	16-Oct-2007	garbled	Sync with HEAD
1.62.14.1	03-Oct-2007	garbled	Sync with HEAD
1.62.6.1	23-Oct-2007	ad	Sync with head.
1.65.2.1	25-Oct-2007	bouyer	Sync with HEAD.
1.67.30.4	30-Sep-2009	snj	Pull up following revision(s) (requested by bouyer in ticket #1040): sys/arch/i386/include/param.h: revision 1.71 sys/arch/i386/i386/db_memrw.c: revision 1.25 sys/arch/xen/include/xenpmap.h: revision 1.22 sys/arch/xen/xen/xenevt.c: revision 1.31 More i386PAE fixes: - x86_round_page, x86_trunc_page, x86_btop and x86_ptob macros are used with physical addresses; cast to paddr_t instead of u_long. Issue pointed out by jym@ - machine_to_phys_mapping[] is a long. This is fine as it holds page frame numbers (and this fits in a 32bit int as physical addresses are only 36bits), but cast to paddr_t before << PAGE_SHIFT - xen_start_info.store_mfn is a long; cast it to paddr_t before << PAGE_SHIFT. should fix issue pointed out by cegger@
1.67.30.3	16-Feb-2009	snj	branches: 1.67.30.3.4; Pull up following revision(s) (requested by ad in ticket #355): sys/arch/i386/include/param.h: revision 1.70 Revert part of previous.
1.67.30.2	16-Feb-2009	snj	Pull up following revision(s) (requested by ad in ticket #355): sys/arch/i386/include/param.h: revision 1.69 opt_noredzone.h is gone.
1.67.30.1	16-Feb-2009	snj	Pull up following revision(s) (requested by ad in ticket #355): sys/arch/i386/conf/GENERIC_TINY: revision 1.116 sys/arch/i386/conf/INSTALL_TINY: revision 1.122 sys/arch/i386/conf/files.i386: revision 1.341 sys/arch/i386/i386/vm_machdep.c: revision 1.142 sys/arch/i386/include/param.h: revision 1.68 sys/arch/xen/conf/files.xen: revision 1.91 - Kill NOREDZONE. - Make the redzone conditional on DIAGNOSTIC. - Give amd64 an additional page for the uarea. 2 is not enough.
1.67.30.3.4.1	21-Apr-2010	matt	sync to netbsd-5
1.67.28.2	28-Apr-2009	skrll	Sync with HEAD.
1.67.28.1	19-Jan-2009	skrll	Sync with HEAD.
1.67.20.2	11-Mar-2010	yamt	sync with head
1.67.20.1	04-May-2009	yamt	sync with head.
1.67.16.1	17-Jan-2009	mjf	Sync with HEAD.
1.70.2.3	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.70.2.2	01-Nov-2009	jym	Sync with HEAD.
1.70.2.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.71.2.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.72.14.2	29-Apr-2012	mrg	sync to latest -current.
1.72.14.1	18-Feb-2012	mrg	merge to -current.
1.72.10.2	23-May-2012	yamt	sync with head.
1.72.10.1	17-Apr-2012	yamt	sync with head
1.76.2.1	09-May-2012	riz	Pull up following revision(s) (requested by rmind in ticket #202): sys/arch/x86/include/cpuvar.h: revision 1.46 sys/arch/xen/include/xenpmap.h: revision 1.34 sys/arch/i386/include/param.h: revision 1.77 sys/arch/x86/x86/pmap_tlb.c: revision 1.5 sys/arch/x86/x86/pmap_tlb.c: revision 1.6 sys/arch/i386/i386/genassym.cf: revision 1.92 sys/arch/xen/x86/cpu.c: revision 1.91 sys/arch/x86/x86/pmap.c: revision 1.177 sys/arch/xen/x86/xen_pmap.c: revision 1.21 sys/arch/x86/acpi/acpi_wakeup.c: revision 1.31 sys/kern/subr_kcpuset.c: revision 1.5 sys/arch/amd64/include/param.h: revision 1.18 sys/sys/kcpuset.h: revision 1.5 sys/arch/x86/x86/mtrr_i686.c: revision 1.26 sys/arch/x86/x86/mtrr_i686.c: revision 1.27 sys/arch/xen/x86/x86_xpmap.c: revision 1.43 sys/arch/x86/x86/cpu.c: revision 1.98 sys/arch/amd64/amd64/mptramp.S: revision 1.14 sys/kern/sys_sched.c: revision 1.42 sys/arch/amd64/amd64/genassym.cf: revision 1.50 sys/arch/i386/i386/mptramp.S: revision 1.24 sys/arch/x86/include/pmap.h: revision 1.52 sys/arch/x86/include/cpu.h: revision 1.50 - Convert x86 MD code, mainly pmap(9) e.g. TLB shootdown code, to use kcpuset(9) and thus replace hardcoded CPU bitmasks. This removes the limitation of maximum CPUs. - Support up to 256 CPUs on amd64 architecture by default. Bug fixes, improvements, completion of Xen part and testing on 64-core AMD Opteron(tm) Processor 6282 SE (also, as Xen HVM domU with 128 CPUs) by Manuel Bouyer. - pmap_tlb_shootdown: do not overwrite tp_cpumask with pm_cpus, but merge like pm_kernel_cpus. Remove unecessary intersection with kcpuset_running. Do not reset tp_userpmap if pmap_kernel(). - Remove pmap_tlb_mailbox_t wrapping, which is pointless after recent changes. - pmap_tlb_invalidate, pmap_tlb_intr: constify for packet structure. i686_mtrr_init_first: handle the case when there are no variable-size MTRR registers available (i686_mtrr_vcnt == 0).
1.77.16.4	28-Aug-2017	skrll	Sync with HEAD
1.77.16.3	05-Feb-2017	skrll	Sync with HEAD
1.77.16.2	05-Oct-2016	skrll	Sync with HEAD
1.77.16.1	27-Dec-2015	skrll	Sync with HEAD (as of 26th Dec)
1.77.14.1	26-Mar-2017	snj	Pull up following revision(s) (requested by maya in ticket #1375): sys/arch/amd64/include/param.h: revision 1.20 sys/arch/i386/include/param.h: revision 1.80 sys/arch/x86/x86/bus_space.c: revision 1.39 increase max io mem on amd64. some devices need it.
1.77.2.1	03-Dec-2017	jdolecek	update from HEAD
1.78.2.1	20-Mar-2017	pgoyette	Sync with HEAD
1.79.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.80.6.1	11-Apr-2018	martin	Pull up following revision(s) (requested by sborrill in ticket #736): sys/arch/i386/include/param.h: revision 1.83 sys/arch/amd64/include/param.h: revision 1.24 Double size of MSGBUFSIZE as existing value is not big enough to hold boot dmesg on modern server-class hardware with lots of CPUs, etc.
1.83.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.83.4.1	10-Jun-2019	christos	Sync with HEAD
1.83.2.1	18-Jan-2019	pgoyette	Synch with HEAD
1.84.4.1	08-Dec-2019	martin	Pull up following revision(s) (requested by riastradh in ticket #505): common/lib/libc/hash/murmurhash/murmurhash.c: revision 1.7 common/lib/libc/hash/murmurhash/murmurhash.c: revision 1.8 sys/sys/param.h: revision 1.610 sys/arch/amd64/include/param.h: revision 1.31 sys/arch/i386/include/param.h: revision 1.85 New macro ALIGNED_POINTER_LOAD. To be used with ALIGNED_POINTER(p,t) instead of writing (const t )p directly. This way, on machines without strict alignment, we can use memcpy to pacify sanitizers, while getting the same compiled code in the end with a single (say) MOV instruction. Fix byte order bug in murmurhash and pacify sanitizers. add now required includes for memcpy prototypes analogue to other hash functions (fix the build)
1.87.8.1	17-Jun-2021	thorpej	Sync w/ HEAD.
1.88.18.1	02-Aug-2025	perseant	Sync with HEAD
1.61	24-Apr-2025	kre	offsetof() needs <stddef.h> (<sys/stddef.h>) Include <sys/stddef.h> when offsetof() is to be used. First step in fixing x86 builds.
1.60	24-Apr-2025	riastradh	amd64: Allocate FPU save state outside pcb if it's too large. We have seen x86_fpu_save_size values (CPUID[EAX=0x0d, ECX=0].ECX) as large as 11008 bytes, notably with Intel AMX TILEDATA's 8192-byte state. We only do this for user threads, and only on machines where it's necessary, to avoid incurring much overhead. There is still a tiny bit of overhead when saving and restoring the FPU state by using a pointer indirection instead of arithmetic indirection for access to struct pcb::pcb_savefpu, but this is probably a drop in the bucket compared to the memory traffic incurred by the FPU state save/restore anyway. For now, these paths are mostly disabled on i386. We could enable them but it will require either rewriting cpu_uarea_alloc/free for i386, or adopting a guard page like amd64 does, which might be costly and so should be undertaken only with some thought and care. And since Intel AMX instructions only work in 64-bit mode, it's not likely to be useful on i386. PR port-amd64/57661: Crash when booting on Xeon Silver 4416+ in KVM/Qemu These changes, as a side effect, may fix: PR kern/57258: kthread_fpu_enter/exit problem by making sure to allocate an FPU save space that is large enough to guarantee fpu_kern_enter/leave work safely, instead of just using a union savefpu object on the stack (which, at 576 bytes, may be too small on some machines, particularly with AVX512 requiring ~2.5K). (But we'll have to do some extra work with kthread_fpu_enter/exit_md -- if we try doing them again on x86 -- to actually allocate the separate pcb on these machines!)
1.59	12-Oct-2019	maxv	branches: 1.59.32; Rewrite the FPU code on x86. This greatly simplifies the logic and removes the dependency on IPL_HIGH. NVMM is updated accordingly. Posted on port-amd64 a week ago. Bump the kernel version to 9.99.16.
1.58	26-Jul-2018	maxv	Rework dbregs, to switch the registers during context switches, and not on each user->kernel transition via userret. Reloads of DR6/DR7 are expensive on both native and xen.
1.57	31-Oct-2017	maxv	branches: 1.57.2; 1.57.4; Don't embed our own values in the reserved fields of the XSAVE area, it really is a bad idea. Move them into the PCB.
1.56	12-Aug-2017	maxv	Remove the vm86 fields from the pcb.
1.55	23-Feb-2017	kamil	Introduce PT_GETDBREGS and PT_SETDBREGS in ptrace(2) on i386 and amd64 This interface is modeled after FreeBSD API with the usage. This replaced previous watchpoint API. The previous one was introduced recently in NetBSD-current and remove its spurs without any backward-compatibility. Design choices for Debug Register accessors: - exec() (TRAP_EXEC event) must remove debug registers from LWP - debug registers are only per-LWP, not per-process globally - debug registers must not be inherited after (v)forking a process - debug registers must not be inherited after forking a thread - a debugger is responsible to set global watchpoints/breakpoints with the debug registers, to achieve this PTRACE_LWP_CREATE/PTRACE_LWP_EXIT event monitoring function is designed to be used - debug register traps must generate SIGTRAP with si_code TRAP_DBREG - debugger is responsible to retrieve debug register state to distinguish the exact debug register trap (DR6 is Status Register on x86) - kernel must not remove debug register traps after triggering a trap event a debugger is responsible to detach this trap with appropriate PT_SETDBREGS call (DR7 is Control Register on x86) - debug registers must not be exposed in mcontext - userland must not be allowed to set a trap on the kernel Implementation notes on i386 and amd64: - the initial state of debug register is retrieved on boot and this value is stored in a local copy (initdbregs), this value is used to initialize dbreg context after PT_GETDBREGS - struct dbregs is stored in pcb as a pointer and by default not initialized - reserved registers (DR4-DR5, DR9-DR15) are ignored Further ideas: - restrict this interface with securelevel Tested on real hardware i386 (Intel Pentium IV) and amd64 (Intel i7). This commit enables 390 debug register ATF tests in kernel/arch/x86. All tests are passing. This commit does not cover netbsd32 compat code. Currently other interface PT_GET_SIGINFO/PT_SET_SIGINFO is required in netbsd32 compat code in order to validate reliably PT_GETDBREGS/PT_SETDBREGS. This implementation does not cover FreeBSD specific defines in their <x86/reg.h>: DBREG_DR7_LOCAL_ENABLE, DBREG_DR7_GLOBAL_ENABLE, DBREG_DR7_LEN_1 etc. These values tend to be reinvented by each tracer on its own. GNU Debugger (GDB) works with NetBSD debug registers after adding this patch: --- gdb/amd64bsd-nat.c.orig 2016-02-10 03:19:39.000000000 +0000 +++ gdb/amd64bsd-nat.c @@ -167,6 +167,10 @@ amd64bsd_target (void) #ifdef HAVE_PT_GETDBREGS +#ifndef DBREG_DRX +#define DBREG_DRX(d,x) ((d)->dr[(x)]) +#endif + static unsigned long amd64bsd_dr_get (ptid_t ptid, int regnum) { Another reason to stop introducing unpopular defines covering machine specific register macros is that these value varies across generations of the same CPU family. GDB demo: (gdb) c Continuing. Watchpoint 2: traceme Old value = 0 New value = 16 main (argc=1, argv=0x7f7fff79fe30) at test.c:8 8 printf("traceme=%d\n", traceme); (Currently the GDB interface is not reliable due to NetBSD support bugs) Sponsored by <The NetBSD Foundation>
1.54	21-Apr-2014	christos	branches: 1.54.4; 1.54.8; 1.54.12; disable assert for lint.
1.53	20-Feb-2014	dsl	branches: 1.53.2; Move the amd64 and i386 pcb to the bottom of the uarea, and move the kernel stack to the top. Change the pcb layouts so that fpu save area is at the end and is 64byte aligned ready for xsave (saving the ymm registers). Welcome to 6.99.32
1.52	12-Feb-2014	dsl	Change i386 to use x86/fpu.c instead of i386/isa/npx.c This changes the trap10 and trap13 code to call directly into fpu.c, removing all the code for T_ARITHTRAP, T_XMM and T_FPUNDA from i386/trap.c Not all of the code thate appeared to handle fpu traps was ever called! Most of the changes just replace the include of machine/npx.h with x86/fpu.h (or remove it entirely).
1.51	19-Jan-2014	dsl	Remove the unused 'struct md_coredump'.
1.50	01-Dec-2013	christos	revert fpu/pcu changes until we figure out what's wrong; they cause random freezes
1.49	23-Oct-2013	drochner	Use the MI "pcu" framework for bookkeeping of npx/fpu states on x86. This reduces the amount of MD code enormously, and makes it easier to implement support for newer CPU features which require more fpu state, or for fpu usage by the kernel. For access to FPU state across CPUs, an xcall kthread is used now rather than a dedicated IPI. No user visible changes intended.
1.48	23-Apr-2010	joerg	branches: 1.48.8; 1.48.18; 1.48.22; Use struct segment_descriptor for pcb_fsd and pcb_gsd instead of int[2].
1.47	21-Mar-2009	ad	branches: 1.47.2; 1.47.4; PR port-i386/40143 Viewing an mpeg transport stream with mplayer causes crash Fix numerous problems: 1. LDT updates are not atomic. 2. Number of processes running with private LDTs and/or I/O bitmaps is not capped. System with high maxprocs can be paniced. 3. LDTR can be leaked over context switch. 4. GDT slot allocations can race, giving the same LDT slot to two procs. 5. Incomplete interrupt/trap frames can be stacked. 6. In some rare cases segment faults are not handled correctly.
1.46	26-Oct-2008	mrg	branches: 1.46.2; 1.46.4; 1.46.8; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.45	28-Apr-2008	martin	branches: 1.45.6; Remove clause 3 and 4 from TNF licenses
1.44	05-Jan-2008	yamt	branches: 1.44.6; 1.44.8; 1.44.10; remove a stale comment.
1.43	04-Jan-2008	yamt	fix a bug in the previous. (per-cpu tss change) keep iopl in pcb so that it won't lost.
1.42	04-Jan-2008	yamt	i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.41	10-Nov-2007	ad	branches: 1.41.6; fsbase/gsbase: - Fix a few bugs with it, in particular fork/exec handling. - Store the descriptors in the PCB, not in the LWP.
1.40	04-Mar-2007	christos	branches: 1.40.2; 1.40.18; 1.40.20; 1.40.24; 1.40.26; Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.39	02-Mar-2007	ad	Define pcb_esp0.
1.38	02-May-2006	drochner	branches: 1.38.14; -explicitely align "struct savexmm" so that we don't get surprised if we play with the PCB -remove historical "struct emcsts"; there is no support in context switching code for ages (if there has ever been)
1.37	20-Feb-2004	yamt	branches: 1.37.16; 1.37.30; 1.37.32; 1.37.34; 1.37.36; 1.37.38; defer pmap switching until it's really needed to avoid frequent loading of cr3 register, which involves tlb flush. with some fixes/improvements from Stephan Uphoff and Bang Jun-Young.
1.36	09-Nov-2003	tsutsui	Use #if defined(_KERNEL_OPT) to protect #include "opt_xxx.h" from building LKM etc. Suggested by mrg.
1.35	24-Aug-2003	chs	add support for non-executable mappings (where the hardware allows this) and make the stack and heap non-executable by default. the changes fall into two basic catagories: - pmap and trap-handler changes. these are all MD: = alpha: we already track per-page execute permission with the (software) PG_EXEC bit, so just have the trap handler pay attention to it. = i386: use a new GDT segment for %cs for processes that have no executable mappings above a certain threshold (currently the bottom of the stack). track per-page execute permission with the last unused PTE bit. = powerpc/ibm4xx: just use the hardware exec bit. = powerpc/oea: we already track per-page exec bits, but the hardware only implements non-exec mappings at the segment level. so track the number of executable mappings in each segment and turn on the no-exec segment bit iff the count is 0. adjust the trap handler to deal. = sparc (sun4m): fix our use of the hardware protection bits. fix the trap handler to recognize text faults. = sparc64: split the existing unified TSB into data and instruction TSBs, and only load TTEs into the appropriate TSB(s) for the permissions. fix the trap handler to check for execute permission. = not yet implemented: amd64, hppa, sh5 - changes in all the emulations that put a signal trampoline on the stack. instead, we now put the trampoline into a uvm_aobj and map that into the process separately. originally from openbsd, adapted for netbsd by me.
1.34	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.33	08-Oct-2002	fvdl	branches: 1.33.6; Delete pcb_flags.
1.32	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.31	12-May-2002	matt	Eliminate commons.
1.30	16-Feb-2002	christos	save cr2 for the benefit of linux emulation, and normalize use of pcb in trap(), i.e. always initialize *pcb and use it instead of doing it sometimes.
1.29	02-Aug-2001	thorpej	Add support for saving/restoring SSE/SSE2 state using FXSAVE/FXRSTOR. Reviewed by Frank.
1.28	17-Jun-2001	sommerfeld	branches: 1.28.2; Add %fs/%gs to trap frame and save/restore them on trap/interrupt/syscall entry from userspace. Remove special-case "by hand" validation of fs/gs register values as well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multiprocessor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT. Discussed in advance with Frank van der Linden.
1.27	16-Aug-2000	thorpej	branches: 1.27.2; Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.26	12-Sep-1999	chs	branches: 1.26.2; 1.26.10; 1.26.14; eliminate the PMAP_NEW option by making it required for all ports. ports which previously had no support for PMAP_NEW now implement the pmap_k* interfaces as wrappers around the non-k versions.
1.25	12-May-1999	thorpej	Move the user-set LDT out of the PCB and into the pmap. Applications which set the LDT and share VM space (e.g. new versions of WINE) expect the LDT to be logically coupled to the address space. Use the new pmap_fork() interface to copy non-shared user-set LDTs when the address space is forked.
1.24	15-Aug-1998	mycroft	branches: 1.24.8; Assign my copyrights to TNF.
1.23	10-Feb-1998	mrg	- add defopt's for UVM, UVMHIST and PMAP_NEW. - remove unnecessary UVMHIST_DECL's.
1.22	06-Feb-1998	mrg	add the i386 MD portions for UVM.
1.21	08-Jan-1996	mycroft	VM86 support, by John Kohl, touched up a bit by me.
1.20	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.19	06-Aug-1995	mycroft	Move the `used fpu' flag into mdproc, so it can be referenced when a process is swapped out. Implement process_{read,write}_fpregs.
1.18	03-May-1995	mycroft	Add PCB_USEDFPU.
1.17	01-May-1995	mycroft	Remove historical garbage.
1.16	01-May-1995	mycroft	Add pcb_cr0.
1.15	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.14	08-Nov-1994	mycroft	pcb.h requires segments.h now.
1.13	05-Nov-1994	mycroft	Only use ssdtosd() when creating, expanding, or copying the LDT. Keep a copy of the result in the PCB so it can be loaded quickly on context switches.
1.12	27-Oct-1994	cgd	new RCS ID format.
1.11	21-May-1994	cgd	struct md_coredump
1.10	21-Dec-1993	mycroft	branches: 1.10.2; Remove two unused fields.
1.9	14-Dec-1993	mycroft	Fold in most of the magnum vm changes.
1.8	19-Sep-1993	brezak	Add per-process LDT slot.
1.7	16-Sep-1993	brezak	Changes to use a trap frame for syscalls.
1.6	27-Jun-1993	andrew	branches: 1.6.4; Re-entrancy protection.
1.5	02-Jun-1993	cgd	get rid of last bit of signal-trampoline-in-pcb hackery
1.4	02-Jun-1993	cgd	set things up so that the signal trampoline code is on the stack, like it normally is in BSD systems. still has a bit of hair...
1.3	22-May-1993	cgd	add rcsids to everything and clean up headers
1.2	09-May-1993	deraadt	npx patches from Bruce Evans. patchkit 10002
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.6.4.3	11-Nov-1993	mycroft	pcb_cmap2 could not justify its existance.
1.6.4.2	09-Oct-1993	mycroft	Make pcb_iml an int.
1.6.4.1	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.10.2.2	21-Dec-1993	mycroft	Remove two unused fields.
1.10.2.1	21-Dec-1993	mycroft	file pcb.h was added on branch magnum on 1993-12-21 09:47:46 +0000
1.24.8.1	21-Jun-1999	thorpej	Sync w/ -current.
1.26.14.2	17-Jun-2001	he	Pull up revision 1.28 (requested by sommerfeld): Add %fs/%gs to trap frame and save/restore them on trap/interupt/ syscall entry from userspace. Remove special-case ``by hand'' validation of %fs/%gs register values s well as special handling of them in various signal handling paths. Now, like %ds and %es, they are validated by the hardware on return to userland. This paves the way for the use of %fs for per-cpu data on multi- processor systems, and fixes an otherwise difficult-to-fix interaction between threads/clone(2) and USER_LDT.
1.26.14.1	16-Aug-2000	thorpej	Update from trunk: Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.26.10.7	24-Feb-2002	sommerfeld	Resynch with mainline.
1.26.10.6	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.26.10.5	03-Sep-2001	sommerfeld	Merge with -current once more.
1.26.10.4	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.26.10.3	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.26.10.2	26-Jun-2000	sommerfeld	Start making npx.c MP-safe; Interface to npx now specifies the process and/or cpu to act on. Implement an IPI for MP lazy FP save. Tested and working when only one CPU is running; untested in the MP case since we can't run user processes on multiple CPU's yet. Along for the ride: - Change i386_send_ipi to take a "struct cpu_info *" rather than a cpu number. - Turn off the "give me the brain" test IPI's.
1.26.10.1	20-Feb-2000	sommerfeld	On MULTIPROCESSOR kernels, delete the "curpcb" global variable since it's per-CPU (it lives in struct cpu_info, defined in cpu.h)
1.26.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.27.2.5	18-Oct-2002	nathanw	Catch up to -current.
1.27.2.4	20-Jun-2002	nathanw	Catch up to -current.
1.27.2.3	28-Feb-2002	nathanw	Catch up to -current.
1.27.2.2	24-Aug-2001	nathanw	Catch up with -current.
1.27.2.1	21-Jun-2001	nathanw	Catch up to -current.
1.28.2.4	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.28.2.3	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.28.2.2	16-Mar-2002	jdolecek	Catch up with -current.
1.28.2.1	03-Aug-2001	lukem	update to -current
1.33.6.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.33.6.2	18-Sep-2004	skrll	Sync with HEAD.
1.33.6.1	03-Aug-2004	skrll	Sync with HEAD
1.37.38.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.37.36.1	11-May-2006	elad	sync with head
1.37.34.1	24-May-2006	yamt	sync with head.
1.37.32.1	01-Jun-2006	kardel	Sync with head.
1.37.30.1	09-Sep-2006	rpaulo	sync with head
1.37.16.4	21-Jan-2008	yamt	sync with head
1.37.16.3	15-Nov-2007	yamt	sync with head.
1.37.16.2	03-Sep-2007	yamt	sync with head.
1.37.16.1	21-Jun-2006	yamt	sync with head.
1.38.14.1	12-Mar-2007	rmind	Sync with HEAD.
1.40.26.2	18-Feb-2008	mjf	Sync with HEAD.
1.40.26.1	19-Nov-2007	mjf	Sync with HEAD.
1.40.24.1	13-Nov-2007	bouyer	Sync with HEAD
1.40.20.1	09-Jan-2008	matt	sync with HEAD
1.40.18.1	11-Nov-2007	joerg	Sync with HEAD.
1.40.2.1	03-Dec-2007	ad	Sync with HEAD.
1.41.6.1	08-Jan-2008	bouyer	Sync with HEAD
1.44.10.3	11-Aug-2010	yamt	sync with head.
1.44.10.2	04-May-2009	yamt	sync with head.
1.44.10.1	16-May-2008	yamt	sync with head.
1.44.8.1	18-May-2008	yamt	sync with head.
1.44.6.2	17-Jan-2009	mjf	Sync with HEAD.
1.44.6.1	02-Jun-2008	mjf	Sync with HEAD.
1.45.6.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.46.8.3	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.46.8.2	01-Nov-2009	jym	Sync with HEAD.
1.46.8.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.46.4.1	04-Apr-2009	snj	Pull up following revision(s) (requested by ad in ticket #656): sys/arch/amd64/amd64/gdt.c: revision 1.21 via patch sys/arch/amd64/amd64/machdep.c: revision 1.129 via patch sys/arch/i386/i386/gdt.c: revision 1.47 via patch sys/arch/i386/i386/kvm86.c: revision 1.17 via patch sys/arch/i386/i386/locore.S: revision 1.85 via patch sys/arch/i386/i386/machdep.c: revision 1.666 via patch sys/arch/i386/i386/vector.S: revision 1.45 via patch sys/arch/i386/include/pcb.h: revision 1.47 via patch sys/arch/x86/include/pmap.h: revision 1.22 via patch sys/arch/x86/include/sysarch.h: revision 1.8 via patch sys/arch/x86/x86/pmap.c: revision 1.80 via patch sys/arch/x86/x86/sys_machdep.c: revision 1.17 via patch sys/compat/linux/arch/i386/linux_machdep.c: revision 1.143 via patch sys/kern/init_main.c: revision 1.384 via patch PR port-i386/40143 Viewing an mpeg transport stream with mplayer causes crash Fix numerous problems: 1. LDT updates are not atomic. 2. Number of processes running with private LDTs and/or I/O bitmaps is not capped. System with high maxprocs can be paniced. 3. LDTR can be leaked over context switch. 4. GDT slot allocations can race, giving the same LDT slot to two procs. 5. Incomplete interrupt/trap frames can be stacked. 6. In some rare cases segment faults are not handled correctly.
1.46.2.1	28-Apr-2009	skrll	Sync with HEAD.
1.47.4.1	30-May-2010	rmind	sync with head
1.47.2.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.48.22.1	18-May-2014	rmind	sync with head
1.48.18.2	03-Dec-2017	jdolecek	update from HEAD
1.48.18.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.48.8.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.53.2.1	10-Aug-2014	tls	Rebase.
1.54.12.1	21-Apr-2017	bouyer	Sync with HEAD
1.54.8.1	20-Mar-2017	pgoyette	Sync with HEAD
1.54.4.1	28-Aug-2017	skrll	Sync with HEAD
1.57.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.57.4.1	10-Jun-2019	christos	Sync with HEAD
1.57.2.1	28-Jul-2018	pgoyette	Sync with HEAD
1.59.32.1	02-Aug-2025	perseant	Sync with HEAD
1.10	14-Nov-2007	ad	Remove pccons.
1.9	27-Oct-2003	junyoung	branches: 1.9.16; 1.9.58; 1.9.76; 1.9.78; 1.9.82; 1.9.84; Nuke __P().
1.8	06-Feb-1999	drochner	branches: 1.8.44; sort the dependencies between terminal devices a bit and add consistency checks to catch conflicting devices at compile time
1.7	22-Mar-1998	drochner	Export pcconskbd_cnattach() - it is used if pccons attaches to the MI keyboard controller driver.
1.6	23-Aug-1997	drochner	Put all console initialization into 1 exported function (pccnattach()). Delay setting of cn_tab->cn_dev until autoconfiguration attach to get the minor number right. Delete unused pccnprobe() and pccninit().
1.5	14-Aug-1997	drochner	-Export variables needed for system console initialization. -Because *cnputc()'s second argument is an "int" in cons.h, correct pccnputc() accordingly.
1.4	02-Feb-1996	mycroft	branches: 1.4.12; Fix #includes.
1.3	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.2	27-Oct-1994	cgd	new RCS ID format.
1.1	28-Sep-1993	andrew	branches: 1.1.2; Exports the pccons ioctls.
1.1.2.2	28-Sep-1993	andrew	Exports the pccons ioctls.
1.1.2.1	28-Sep-1993	andrew	file pccons.h was added on branch magnum on 1993-09-28 03:26:18 +0000
1.4.12.2	01-Sep-1997	thorpej	Update marc-pcmcia-branch from trunk (missed this the last time).
1.4.12.1	23-Aug-1997	thorpej	Update marc-pcmcia branch from trunk.
1.8.44.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.8.44.2	18-Sep-2004	skrll	Sync with HEAD.
1.8.44.1	03-Aug-2004	skrll	Sync with HEAD
1.9.84.1	19-Nov-2007	mjf	Sync with HEAD.
1.9.82.1	18-Nov-2007	bouyer	Sync with HEAD
1.9.78.1	23-Mar-2008	matt	sync with HEAD
1.9.76.1	21-Nov-2007	joerg	Sync with HEAD.
1.9.58.1	03-Dec-2007	ad	Sync with HEAD.
1.9.16.1	15-Nov-2007	yamt	sync with head.
1.19	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.18	15-May-2002	thorpej	branches: 1.18.4; Let machine-dependent code specify how to enumerate the bus. Currently, everyone uses pci_enumerate_bus_generic().
1.17	08-Jun-2001	simonb	branches: 1.17.4; For ports that wire up pciide in compatibility mode, have them define __HAVE_PCIIDE_MACHDEP_COMPAT_INTR_ESTABLISH in pci_machdep.h and pciide_map_compat_intr() only calls pciide_machdep_compat_intr_establish() if that preprocessor define exists. Ports that don't need to do this no longer need to supply a dummy function.
1.16	28-Dec-2000	sommerfeld	branches: 1.16.4; Change pci_intr_map to get interrupt source information from a "struct pci_attach_args *" instead of from four separate parameters which in all cases were extracted from the same "struct pci_attach_args". This both simplifies the driver api, and allows for alternate PCI interrupt mapping schemes, such as one using the tables described in the Intel Multiprocessor Spec which describe interrupt wirings for devices behind pci-pci bridges based on the device's location rather the bridge's location. Tested on alpha and i386; welcome to 1.5Q
1.15	18-Jul-2000	soda	add "#define I386_PCI_INTERRUPT_LINE_NO_CONNECTION 0xff"
1.14	04-Jun-2000	cgd	branches: 1.14.2; Implement the more flexiable `evcnt' interface as discussed (briefly) on tech-kern and now documented in evcnt(9).
1.13	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.12	19-Mar-1999	cgd	branches: 1.12.8; 1.12.18; Moved from arch/i386/pci/pci_machdep.h,v
1.11	15-Aug-1998	mycroft	Make copyright notices with my name consistent.
1.10	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.9	04-May-1998	thorpej	Add a function, pci_bus_flags(), that determines which flags (for memory and i/o enabled) should be passed to the primary PCI bus's autconfiguration node. Use this function to detect broken PCI-Host bridges, which have problems with memory-mapped access. Add the SIS 85C496 to this list.
1.8	26-Aug-1997	thorpej	Notify drivers that we "prefer" I/O mapped space, per discussion with Chris Demetriou and Matt Thomas, some two or three months ago. (This hint is used by the latest "de" driver.)
1.7	06-Jun-1997	thorpej	branches: 1.7.4; Pull thorpej-bus-dma branch into mainline.
1.6	17-Dec-1996	thorpej	branches: 1.6.8; Implement pci_decompose_tag().
1.5	27-Mar-1996	cgd	modify these to provide a new, better-specified PCI interface (soon to be documented on mailing lists; eventually in section 9 manual pages), most importantly: (1) support interrupt pin swizzling on non-i386 systems with PCI-PCI bridges (per PPB spec; done, but meaningless, on i386). (2) provide pci_{io,mem}_find(), to determine what I/O or memory space is described by a given PCI configuration space mapping register. (3) provide pci_intr_map(), pci_intr_string(), and pci_intr_{,dis}establish() to manipulate and print info about PCI interrupts. (4) deprecate the pci_map_* functions, and provide them only as compatibility interfaces (in pci_compat.c) which will eventually go away, implemented as wrappers around the functions described above. (5) make pci functions take as an argument a machine-dependent cookie, to allow more flexibility in implementation.
1.4	14-Mar-1996	cgd	(1) remove the #ifdef i386 from pci.c, and provide a machine-dependent hook (pci_md_attach_hook()) to do any machine-dependent attachment gunk, e.g. on the i386 printing out the configuration mode (if bus 0) (2) don't pass max device number for a given bus in, use PCI_MAX_DEVICE_NUMBER, which can be defined on a per-machine basis. (defaults to 32. on i386, it's 32 if pci conf mode == 1, 16 if 2.)
1.3	17-Apr-1995	cgd	clean up several ISA device interfaces: autoconfiguration, header inclusion, and interrupt configuration. more work still needs to be done, but it's getting better...
1.2	27-Oct-1994	cgd	new RCS ID format.
1.1	09-Aug-1994	mycroft	Add PCI autoconfiguration support.
1.6.8.1	18-May-1997	thorpej	Add an external declaration for the pci dma tag.
1.7.4.1	27-Aug-1997	thorpej	Update marc-pcmcia branch from trunk.
1.12.18.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.12.8.2	05-Jan-2001	bouyer	Sync with HEAD
1.12.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.14.2.1	10-Aug-2000	soda	Pull up to netbsd-1-5 branch Approved by: thorpej - define PCI_INTERRUPT_PIN_MAX and use it instead of magic number. - define I386_PCI_INTERRUPT_LINE_NO_CONNECTION and use it instead of magic number. Revisions pulled up: > cvs rdiff -r1.28 -r1.29 syssrc/sys/dev/pci/pcireg.h > cvs rdiff -r1.14 -r1.15 syssrc/sys/arch/i386/include/pci_machdep.h > cvs rdiff -r1.38 -r1.39 syssrc/sys/arch/i386/pci/pci_machdep.c > cvs rdiff -r1.1 -r1.2 syssrc/sys/arch/i386/pci/opti82c558.c \ > syssrc/sys/arch/i386/pci/sis85c503.c \ > syssrc/sys/arch/i386/pci/via82c586.c
1.16.4.2	20-Jun-2002	nathanw	Catch up to -current.
1.16.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.17.4.1	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.18.4.2	15-May-2002	thorpej	Let machine-dependent code specify how to enumerate the bus. Currently, everyone uses pci_enumerate_bus_generic().
1.18.4.1	15-May-2002	thorpej	file pci_machdep.h was added on branch sommerfeld_i386mp_1 on 2002-05-15 19:23:55 +0000
1.2	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.1	22-Nov-2002	fvdl	branches: 1.1.2; Define a generic PIC structure, used by the interrupt code.
1.1.2.2	11-Dec-2002	thorpej	Sync with HEAD.
1.1.2.1	22-Nov-2002	thorpej	file pic.h was added on branch nathanw_sa on 2002-12-11 06:01:01 +0000
1.20	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.19	28-Dec-2001	thorpej	branches: 1.19.2; Fix some register constraints; specifically, the "port" operand, if not an immediate, must be in a 16-bit register.
1.18	13-Nov-2001	perry	Make the type of ports in the in()/out() routines u_long instead of int. Massively reduced kernel lint, causes no harm. Discussed extensively with christos and thorpej.
1.17	30-Apr-2001	lukem	branches: 1.17.4; remove some lint, including ansifying some inlines
1.16	26-Apr-2001	thorpej	Const poison.
1.15	13-Mar-2000	kristerw	branches: 1.15.8; Make the inline asm conform to what gcc thinks they should look like (per discussions in http://gcc.gnu.org/ml/gcc-patches/1998-11/msg00004.html) Approved by fvdl.
1.14	15-Aug-1998	mycroft	branches: 1.14.12; Assign my copyrights to TNF.
1.13	08-Mar-1996	cgd	make this idempotent. bus.h needs to include it, and other things which may include something that includes bus.h may include it as well.
1.12	20-Dec-1995	mycroft	Use explicit integer sizes. Adjust constraints for GCC 2.7.2.
1.11	22-Apr-1995	mycroft	branches: 1.11.2; Clean this up a bit: * Abstract the conditional for immediate port numbers, and make it always false if not optimizing. * Add some whitespace to the string functions to make them more readable.
1.10	20-Nov-1994	mycroft	Use immediate constants for port numbers < 0x100.
1.9	18-Nov-1994	mycroft	Convert port, IRQ, and DRQ numbers to ints.
1.8	27-Oct-1994	cgd	new RCS ID format.
1.7	28-Jan-1994	jtc	branches: 1.7.2; Fix spelling error in Copyright notice
1.6	31-Dec-1993	mycroft	Make sure we get the correct `rep' prefix per the Intel manual (though really GAS should figure this out itself).
1.5	30-Dec-1993	deraadt	fix __asm() clobber list for some instructions. fix from Lennart Augustsson <augustss@cs.chalmers.se>
1.4	20-Dec-1993	mycroft	New, more complete version.
1.3	06-Jun-1993	cgd	branches: 1.3.4; incorporate Bruce Evans' interrupt changes, as hacked by Rod grimes, for patchkit patch 158. it appears to work well.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	after 0.2.2 "stable" patches applied
1.3.4.3	19-Dec-1993	mycroft	Clear the direction flag before string insns.
1.3.4.2	29-Oct-1993	mycroft	Fix definitions of outs[bwl](), and note that none of the string functions bashes dx.
1.3.4.1	11-Oct-1993	mycroft	Move {disable,enable}_intr() to psl.h, and inline all of {in,out}{,s}[bwl] in pio.h.
1.7.2.2	28-Jan-1994	jtc	Fix spelling error in Copyright notice
1.7.2.1	28-Jan-1994	jtc	file pio.h was added on branch magnum on 1994-01-28 23:44:19 +0000
1.11.2.1	01-Mar-1996	scottr	Fix register constraints so that kernels can be compiled with gcc-2.7.2
1.14.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.15.8.3	08-Jan-2002	nathanw	Catch up to -current.
1.15.8.2	14-Nov-2001	nathanw	Catch up to -current.
1.15.8.1	21-Jun-2001	nathanw	Catch up to -current.
1.17.4.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.19.2.2	28-Dec-2001	thorpej	Fix some register constraints; specifically, the "port" operand, if not an immediate, must be in a 16-bit register.
1.19.2.1	28-Dec-2001	thorpej	file pio.h was added on branch sommerfeld_i386mp_1 on 2001-12-28 06:44:13 +0000
1.129	20-Aug-2022	riastradh	x86: Move definition of struct pmap to pmap_private.h. This makes pmap_resident_count and pmap_wired_count out-of-line functions instead of inline. No functional change intended otherwise.
1.128	20-Aug-2022	riastradh	x86: Split most of pmap.h into pmap_private.h or vmparam.h. This way pmap.h only contains the MD definition of the MI pmap(9) API, which loads of things in the kernel rely on, so changing x86 pmap internals no longer requires recompiling the entire kernel every time. Callers needing these internals must now use machine/pmap_private.h. Note: This is not x86/pmap_private.h because it contains three parts: 1. CPU-specific (different for i386/amd64) definitions used by... 2. common definitions, including Xenisms like xpmap_ptetomach, further used by... 3. more CPU-specific inlines for pmap_pte_* operations So {amd64,i386}/pmap_private.h defines 1, includes x86/pmap_private.h for 2, and then defines 3. Maybe we should split that out into a new pmap_pte.h to reduce this trouble. No functional change intended, other than that some .c files must include machine/pmap_private.h when previously uvm/uvm_pmap.h polluted the namespace with pmap internals. Note: This migrates part of i386/pmap.h into i386/vmparam.h -- specifically the parts that are needed for several constants defined in vmparam.h: VM_MAXUSER_ADDRESS VM_MAX_ADDRESS VM_MAX_KERNEL_ADDRESS VM_MIN_KERNEL_ADDRESS Since i386 needs PDP_SIZE in vmparam.h, I added it there on amd64 too, just to keep things parallel.
1.127	20-Aug-2022	riastradh	x86: Move struct vm_page_md to common x86/pmap.h.
1.126	19-Jul-2020	maxv	don't include opt_user_ldt.h when it is not needed
1.125	15-May-2020	ad	Revert previous after thinking about it. It was wrong, don't need to use an atomic to clear a PTE or set initial version unless the circumstances call for it.
1.124	17-Mar-2020	ad	Always set PTEs using atomics. There are too many assumptions to go wrong.
1.123	09-Mar-2019	maxv	New software PTE bits.
1.122	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.121	19-Nov-2018	maxv	Introduce pl_pi, will be used soon.
1.120	19-Nov-2018	maxv	Rename 'mask' -> 'frame', we will use the real 'mask' soon.
1.119	25-Jul-2018	maxv	Remove NPTECL, unused.
1.118	19-Sep-2016	maya	branches: 1.118.14; 1.118.16; move function prototype to x86, so it is available to amd64 too
1.117	21-Apr-2014	christos	branches: 1.117.4; 1.117.8; forward decl for struct pcb.
1.116	18-Mar-2013	para	branches: 1.116.6; 1.116.10; i386 pmap.h got a forward declarition for vm_map from systm.h insert a forward declaration
1.115	19-Feb-2012	cherry	branches: 1.115.2; Removing remaining references to the alternate PTE space. Modify documentation appropriately
1.114	19-Jan-2012	bouyer	pmap_pte_set() is not supposed to be atomic, so only raise IPL, no need to take pte_lock
1.113	09-Jan-2012	cherry	Make cross-cpu pte access MP safe. XXX: review cases of use of pmap_set_pte() vs direct use of xpq_queue_pte_update()
1.112	06-Nov-2011	cherry	branches: 1.112.4; [merging from cherry-xenmp] make pmap_kernel() shadow PMD per-cpu and MP aware.
1.111	06-Nov-2011	cherry	[merging from cherry-xenmp] Make the xen MMU op queue locking api private. Implement per-cpu queues.
1.110	13-Aug-2011	cherry	branches: 1.110.2; Add locking around ops to the hypervisor MMU "queue".
1.109	01-Feb-2011	chuck	branches: 1.109.2; udpate license clauses on my code to match the new-style BSD licenses. based on diff that rmind@ sent me. no functional change with this commit.
1.108	14-Nov-2010	uebayasi	branches: 1.108.2; 1.108.4; Move struct vm_page_md definition from vmparam.h to pmap.h, because it's used only by pmap. vmparam.h has definitions for wider audience. All GENERIC kernels build tested, except ia64. powerpc/include/booke/vmparam.h has one too, but it has no pmap.h, so it's left as is.
1.107	24-Jul-2010	jym	Welcome PAE inside i386 current. This patch is inspired by work previously done by Jeremy Morse, ported by me to -current, merged with the work previously done for port-xen, together with additionals fixes and improvements. PAE option is disabled by default in GENERIC (but will be enabled in ALL in the next few days). In quick, PAE switches the CPU to a mode where physical addresses become 36 bits (64 GiB). Virtual address space remains at 32 bits (4 GiB). To cope with the increased size of the physical address, they are manipulated as 64 bits variables by kernel and MMU. When supported by the CPU, it also allows the use of the NX/XD bit that provides no-execution right enforcement on a per physical page basis. Notes: - reworked locore.S - introduce cpu_load_pmap(), used to switch pmap for the curcpu. Due to the different handling of pmap mappings with PAE vs !PAE, Xen vs native, details are hidden within this function. This helps calling it from assembly, as some features, like BIOS calls, switch to pmap_kernel before mapping trampoline code in low memory. - some changes in bioscall and kvm86_call, to reflect the above. - the L3 is "pinned" per-CPU, and is only manipulated by a reduced set of functions within pmap. To track the L3, I added two elements to struct cpu_info, namely ci_l3_pdirpa (PA of the L3), and ci_l3_pdir (the L3 VA). Rest of the code considers that it runs "just like" a normal i386, except that the L2 is 4 pages long (PTP_LEVELS is still 2). - similar to the ci_pae_l3_pdir{,pa} variables, amd64's xen_current_user_pgd becomes an element of cpu_info (slowly paving the way for MP world). - bootinfo_source struct declaration is modified, to cope with paddr_t size change with PAE (it is not correct to assume that bs_addr is a paddr_t when compiled with PAE - it should remain 32 bits). bs_addrs is now a void * array (in bootloader's code under i386/stand/, the bs_addrs is a physaddr_t, which is an unsigned long). - fixes in multiboot code (same reason as bootinfo): paddr_t size change. I used Elf32_* types, use RELOC() where necessary, and move the memcpy() functions out of the if/else if (I do not expect sym and str tables to overlap with ELF). - 64 bits atomic functions for pmap - all pmap_pdirpa access are now done through the pmap_pdirpa macro. It hides the L3/L2 stuff from PAE, as well as the pm_pdirpa change in struct pmap (it now becomes a PDP_SIZE array, with or without PAE). - manipulation of recursive mappings ( PDIR_SLOT_{,A}PTEs ) is done via loops on PDP_SIZE. See also http://mail-index.netbsd.org/port-i386/2010/07/17/msg002062.html No objection raised on port-i386@ and port-xen@R for about a week. XXX kvm(3) will be fixed in another patch to properly handle both PAE and !PAE kernel dumps (VA => PA macros are slightly different, and need proper 64 bits PA support in kvm_i386). XXX Mixing PAE and !PAE modules may lead to unwanted/unexpected results. This cannot be solved easily, and needs lots of thinking before being declared safe (paddr_t/bus_addr_t size handling, PD/PT macros abstractions).
1.106	15-Jul-2010	jym	Purely cosmetic: move L2_SLOT_KERNBASE out of the PAE #ifdef/#endif block.
1.105	01-Mar-2010	jym	branches: 1.105.2; Use PDP_SIZE for NTOPLEVEL_PDES (number of top level PDEs) instead of #ifdef'ing PAE.
1.104	09-Feb-2010	jym	Fix typos in comments.
1.103	26-Oct-2008	mrg	branches: 1.103.8; 1.103.14; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.102	23-Jan-2008	bouyer	branches: 1.102.6; 1.102.10; 1.102.16; Merge the bouyer-xeni386 branch. This brings in PAE support to NetBSD xeni386 (domU only). PAE support is enabled by 'options PAE', see the new XEN3PAE_DOMU and INSTALL_XEN3PAE_DOMU kernel config files. See the comments in arch/i386/include/{pte.h,pmap.h} to see how it works. In short, we still handle it as a 2-level MMU, with the second level page directory being 4 pages in size. pmap switching is done by switching the L2 pages in the L3 entries, instead of loading %cr3. This is almost required by Xen, which handle the last L2 page (the one mapping 0xc0000000 - 0xffffffff) in a very special way. But this approach should also work for native PAE support if ever supported (in fact, the pmap should almost suport native PAE, what's missing is bootstrap code in locore.S).
1.101	20-Jan-2008	bouyer	Make first argument of Xen's pmap_pte_cas() volatile, fix a warning building pmap.c.
1.100	13-Jan-2008	yamt	add pmap_pte_cas.
1.99	12-Jan-2008	bouyer	Xen doesn't need NKPTP_MIN/NKPTP_MAX any more.
1.98	11-Jan-2008	bouyer	Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.97	28-Nov-2007	ad	branches: 1.97.6; Remove remaining CPUCLASS_386 tests.
1.96	28-Nov-2007	ad	Use the new atomic ops.
1.95	22-Nov-2007	bouyer	Pull up the bouyer-xenamd64 branch to HEAD. This brings in amd64 support to NetBSD/Xen, both Dom0 and DomU.
1.94	18-Oct-2007	yamt	branches: 1.94.2; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.93	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.92	17-Oct-2007	ad	Match amd64: -#if defined(_KERNEL) \|\| defined(_LKM) +#if defined(_KERNEL)
1.91	26-Sep-2007	ad	branches: 1.91.2; Only include machine/cpufunc.h if _KERNEL.
1.90	29-Aug-2007	ad	branches: 1.90.2; Merge most x86 changes from the vmlocking branch, except the threaded soft interrupt stuff. This is mostly comprised of changes to the pmap modules to work on multiprocessor systems without kernel_lock, and changes to speed up tlb shootdowns.
1.89	21-Feb-2007	thorpej	branches: 1.89.4; 1.89.12; 1.89.16; 1.89.20; 1.89.22; Replace the Mach-derived boolean_t type with the C99 bool type. A future commit will replace use of TRUE and FALSE with true and false.
1.88	16-Nov-2006	christos	branches: 1.88.4; __unused removal on arguments; approved by core.
1.87	12-Oct-2006	christos	- sprinkle __unused on function decls. - fix a couple of unused bugs - no more -Wno-unused for i386
1.86	16-Feb-2006	perry	branches: 1.86.14; 1.86.16; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.85	26-Dec-2005	perry	branches: 1.85.2; 1.85.4; 1.85.6; u_intN_t -> uintN_t
1.84	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.83	08-Aug-2005	junyoung	- Fix typo in comment. - Remove trailing spaces.
1.82	26-Aug-2004	junyoung	branches: 1.82.12; As suggested on tech-kern@ days ago: * Get rid of PTmap, PTD, PTDpde, APTmap, APTD, and APTDpde from locore.S. * Rename PTDpaddr to PDPpaddr, ptdpaddr in struct cpu_kcore_hdr to pdppaddr for consistency.
1.81	22-Aug-2004	junyoung	Fix various typos in comments including ones that weren't noticed for years.
1.80	05-Aug-2004	mycroft	Set NKPTP default to 0 (it's calculated in locore). Also reduce NKPTP_MIN to 2 -- though, with our current rounding of the data section, we will never take advantage of that.
1.79	20-Feb-2004	yamt	defer pmap switching until it's really needed to avoid frequent loading of cr3 register, which involves tlb flush. with some fixes/improvements from Stephan Uphoff and Bang Jun-Young.
1.78	27-Oct-2003	junyoung	Nuke __P().
1.77	23-Oct-2003	chs	rename pv_next to pv_node. use SPLAY_RIGHT() instead of expanding it inline.
1.76	23-Oct-2003	provos	changed linked list in pmap_remove_pv to a splay tree; approved: fvdl@
1.75	24-Aug-2003	chs	add support for non-executable mappings (where the hardware allows this) and make the stack and heap non-executable by default. the changes fall into two basic catagories: - pmap and trap-handler changes. these are all MD: = alpha: we already track per-page execute permission with the (software) PG_EXEC bit, so just have the trap handler pay attention to it. = i386: use a new GDT segment for %cs for processes that have no executable mappings above a certain threshold (currently the bottom of the stack). track per-page execute permission with the last unused PTE bit. = powerpc/ibm4xx: just use the hardware exec bit. = powerpc/oea: we already track per-page exec bits, but the hardware only implements non-exec mappings at the segment level. so track the number of executable mappings in each segment and turn on the no-exec segment bit iff the count is 0. adjust the trap handler to deal. = sparc (sun4m): fix our use of the hardware protection bits. fix the trap handler to recognize text faults. = sparc64: split the existing unified TSB into data and instruction TSBs, and only load TTEs into the appropriate TSB(s) for the permissions. fix the trap handler to check for execute permission. = not yet implemented: amd64, hppa, sh5 - changes in all the emulations that put a signal trampoline on the stack. instead, we now put the trampoline into a uvm_aobj and map that into the process separately. originally from openbsd, adapted for netbsd by me.
1.74	22-Jul-2003	yamt	move pv_head and page attributes into md part of vm_page structure instead of having separate arrays for them. (i.e. use __HAVE_VM_PAGE_MD instead of __HAVE_PMAP_PHYSSEG.)
1.73	09-Apr-2003	thorpej	branches: 1.73.2; Add the ability for pool caches to cache the physical address of objects. Clients of the pool_cache API must consistently use the "paddr" variants or not, otherwise behavior is undefined. Enable this on Alpha, ARM, MIPS, and x86. Other platforms must define POOL_VTOPHYS() in the appropriate manner in order to enable the feature. Part 1 of a series of simple patches contributed by Wasabi Systems to improve network performance.
1.72	09-Apr-2003	thorpej	Remove something accidentally included in last commit.
1.71	02-Apr-2003	thorpej	Use PAGE_SIZE rather than NBPG.
1.70	03-Mar-2003	fvdl	define pmap_cpu_has_pg_n and pmap_cpu_has_invlpg to make code sharing easier.
1.69	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.68	29-Jan-2003	drochner	give some VM which was reserved for the UAREA but not used anymore back to the user and adjust some comments (which were not accurate anyway since NOREDZONE) binary compatibility note: changing VM_MAXUSER_ADDRESS might influence some sanity check in kvm_proc, where arguments on the stack are dealt with, but it was a variable anyway and noone did care...
1.67	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.66	02-Nov-2002	perry	/ARGSUSED/ one function -- eliminates 414 lint complaints
1.65	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.64	22-Sep-2002	chs	it really helps to get the stub right before cutting + pasting it 27 times. alas, I did not. doh.
1.63	22-Sep-2002	chs	add pmap_remove_all() hook (empty on most platforms so far).
1.62	23-Dec-2001	thorpej	Use ANSI declarations for static inline functions, and give them the __unused__ attribute.
1.61	07-Nov-2001	yamt	correct comment.
1.60	10-Sep-2001	chris	branches: 1.60.2; Update pmap_update to now take the updated pmap as an argument. This will allow improvements to the pmaps so that they can more easily defer expensive operations, eg tlb/cache flush, til the last possible moment. Currently this is a no-op on most platforms, so they should see no difference. Reviewed by Jason.
1.59	12-Aug-2001	chs	branches: 1.59.2; update some comments.
1.58	30-May-2001	mrg	branches: 1.58.2; use _KERNEL_OPT
1.57	26-May-2001	chs	replace {simple_,}lock{_data,}_t with struct {simple,}lock {,*}.
1.56	29-Apr-2001	thorpej	The idle loop page zero'er no longer needs to do uncached access now that we have page coloring.
1.55	22-Apr-2001	thorpej	Undo a misguided previous change to the pmap_update() API.
1.54	22-Apr-2001	thorpej	Give pmap_update() an argument (a pmap_t) so that it knows which pmap it should be updating.
1.53	04-Jan-2001	thorpej	branches: 1.53.2; Just garbage-collect the #if 0'd pmap_transfer() stuff.
1.52	01-Jan-2001	thorpej	Don't equate pmap_update() with tlbflush(), but instead call tlbflush() when that's what we mean (currently everywhere, except in the one place in MI code where it is called). The whole pmap_update() thing needs to be reexamined, but this helps to clarify things a little with the currently-defined semantics of that function.
1.51	06-Dec-2000	chs	in kvtopte(), return the mapping of the pde in the recursive page-table mapping space rather than the mapping in the kernel address space.
1.50	28-Sep-2000	is	We do wired page accounting. Tell UVM about it.
1.49	21-Sep-2000	thorpej	Make PMAP_PAGEIDLEZERO() return a boolean value. FALSE indidcates that the page being zero'd was not completed and that page zeroing should be aborted. This may be used by machine-dependent code doing slow page access to reduce the latency of running a process that has become runnable while in the middle of doing a slow page zero.
1.48	07-Sep-2000	thorpej	Add optional support for mapping the kernel with large (4MB) pages. Right now, only does kernel text, and wastes a little memory, but there is some noticeable performance improvement even with just text.
1.47	06-Sep-2000	thorpej	vtopte() and kvtopte() now check for kernel vs. non-kernel addresses, respectively.
1.46	06-Sep-2000	thorpej	Remove some vto*() macros that won't stand a chance of working with PSE or PAE, and reimplement vtphys() in terms of pmap_extract().
1.45	15-Jun-2000	mycroft	branches: 1.45.2; Disable pmap_copy() for now.
1.44	24-Apr-2000	thorpej	branches: 1.44.2; Glue in uvm_pageidlezero(). While here, improve some panic messages in pmap.c, as requested by Jonathan Stone.
1.43	11-Feb-2000	thorpej	branches: 1.43.2; Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1G of virtual address space, leaving userland with 3G, and update comments to match the new reality. We knew we were going to have to bite this bullet eventually, and there are a couple of outstanding PRs related to this issue (9389 and 9313). Complete solution to those PRs is going to involve some sort of run-time decision on how large kmem_map should be, as well as changing some data structure allocation strategies in UVM. However, this change will at least allow the PR submitter to simply throw resources at the problem.
1.42	28-Jul-1999	thorpej	branches: 1.42.2; Don't inline pmap_kenter_pa(). It doesn't buy us much to do so, and it's nice to have it show up in stack traces.
1.41	18-Jul-1999	chs	KNF.
1.40	17-Jun-1999	thorpej	PMAP_NEW is no longer optional on the i386; the old pmap's page table allocation strategy no longer works at all. Move pmap.new.* to pmap.. To read the revision history of PMAP_NEW up until this merge, use cvs rlog of the old pmap.new. files.
1.39	12-May-1999	thorpej	Move the user-set LDT out of the PCB and into the pmap. Applications which set the LDT and share VM space (e.g. new versions of WINE) expect the LDT to be logically coupled to the address space. Use the new pmap_fork() interface to copy non-shared user-set LDTs when the address space is forked.
1.38	15-Aug-1998	mycroft	branches: 1.38.6; 1.38.8; Assign my copyrights to TNF.
1.37	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.36	18-Feb-1998	thorpej	Fixup an inconsistency in the declaration of PTDpaddr, pointed out by Chris Demetriou.
1.35	18-Feb-1998	cgd	Move pmap_map() function definition to MD headers, as appropriate. It's an internal function, and the VM system shouldn't try to prototype it. (Note that some ports _don't_ prototype it.)
1.34	10-Feb-1998	mrg	- add defopt's for UVM, UVMHIST and PMAP_NEW. - remove unnecessary UVMHIST_DECL's.
1.33	06-Feb-1998	mrg	add the i386 MD portions for UVM.
1.32	13-Jan-1998	mrg	this is the i386 MD portion of the MACHINE_NEW_NONCONTIG code, as written by chuck cranor. thanks to mycroft for helping me find the one little line of code i accidentally deleted while merging it. this is not enabled by default. `options MACHINE_NEW_NONCONTIG' will use this code. eventually, this should go into <machine/vmparam.h> insteaed of MACHINE_NONCONTIG.
1.31	06-Jan-1998	thorpej	Garbage-collect pm_pdchanged; it's not used by anything.
1.30	03-Jan-1998	thorpej	Now that all ports have pmap_activate(), and it has an identical interface, prototype it in <vm/pmap.h>
1.29	07-Dec-1997	fvdl	Define constants for the start and end VA of user page tables.
1.28	04-Nov-1997	thorpej	Make pmap_{,de}activate() take a pointer to a proc.
1.27	20-Oct-1997	fvdl	branches: 1.27.2; Change various constants that depend on the kernel VM base address. Lower it to 0xf0000000 to give it more breathing space.
1.26	12-Jun-1997	thorpej	Nuke __VM_PMAP_HACK.
1.25	16-May-1997	gwr	Add #define __VM_PMAP_HACK as a temporary measure.
1.24	18-Nov-1996	fvdl	Changes for NKPDE calculation from cgd. - New variables: biosextmem, biosbasemem, nkpde - Above can be set by using the options BIOSEXTMEM, BIOSBASEMEM, and NKPDE respectively (EXTMEM_SIZE is now called BIOSEXTMEM). When preset this way, they won't be filled in / calculated. - Readable by sysctl using machdep.nkpde, machdep.biosbasemem and machdep.biosextmem. - nkpde is calculated as: min(NKPDE_MAX, NKPDE_BASE + (biosextmem >> 10) * NKPDE_SCALE) Where NKPDE_MAX is 31, NKPDE_BASE is 4, NKPDE_SCALE = 1.
1.23	03-May-1996	christos	Add missing prototypes.
1.22	12-Feb-1996	christos	Move pmap_page_index declaration to <vm/pmap.h>
1.21	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.20	26-Jun-1995	cgd	IdlePTD has been gone for a while. PTDpaddr now exists.
1.19	11-May-1995	jtc	KERNEL -> _KERNEL
1.18	10-Apr-1995	mycroft	Bring back pmap_kernel(), for now always inlined as a pointer to kernel_pmap_store.
1.17	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.16	31-Oct-1994	andrew	Increase NKPDE to cope with 64MB RAM.
1.15	27-Oct-1994	cgd	new RCS ID format.
1.14	09-Oct-1994	mycroft	#include cpufunc.h from pmap.h.
1.13	09-Oct-1994	mycroft	Define pmap_update() using tlbflush().
1.12	09-Sep-1994	mycroft	Make kernel_pmap a macro that points directly to kernel_pmap_store.
1.11	07-Sep-1994	mycroft	Various changes: * Inline some functions. * Eliminate redundant calls to pmap_page_index(). * Reduce the number of TLB flushes in accordance with the i386 SSWG. * Inline pmap_remove() in pmap_remove_all(). (Most if it is unnecessary.) * Allocate pv_entry's in pages, and keep a list of free entries in each page, and a list of pages with free entries. (Trying to reduce malloc() overhead, and improve locality of reference.) * Remove a couple of macros that aren't really useful. * Remove pmap_kernel() completely.
1.10	15-Aug-1994	mycroft	Eliminate struct pte and struct pde.
1.9	02-Jun-1994	mycroft	branches: 1.9.2; Increase default NKPDE.
1.8	23-May-1994	cgd	cast to quiet warning
1.7	05-Jan-1994	mycroft	branches: 1.7.2; Move kernel to 0xf8000000.
1.6	14-Dec-1993	mycroft	Fold in most of the magnum vm changes.
1.5	30-Aug-1993	brezak	branches: 1.5.2; Add prototype for pmap_bootstrap().
1.4	27-Aug-1993	brezak	i386 Machine dependant changes to handle non-contiguous memory.
1.3	29-Jul-1993	jtc	Change "#endif FOO" to "#endif /* FOO */", to keep gcc -Wall, lint, etc. from complaining about a trivial issue.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.5.2.4	13-Dec-1993	mycroft	Stylistic changes.
1.5.2.3	26-Oct-1993	mycroft	Nuke I386_PAGE_SIZE and I386_PDR_SIZE. Make the kernel page directory 7 pages.
1.5.2.2	15-Oct-1993	mycroft	Major cleanup of include files and constants.
1.5.2.1	10-Oct-1993	mycroft	load_cr3 --> lcr3
1.7.2.2	05-Jan-1994	mycroft	Move kernel to 0xf8000000.
1.7.2.1	05-Jan-1994	mycroft	file pmap.h was added on branch magnum on 1994-01-05 16:02:39 +0000
1.9.2.2	11-Oct-1994	mycroft	Update from trunk.
1.9.2.1	15-Aug-1994	mycroft	update from trunk
1.27.2.1	09-Dec-1997	thorpej	Pull up rev 1.29 (fvdl)
1.38.8.2	02-Aug-1999	thorpej	Update from trunk.
1.38.8.1	21-Jun-1999	thorpej	Sync w/ -current.
1.38.6.1	19-Feb-2000	he	Apply patch (requested by thorpej): Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1GB of virtual address space, leaving userland with 3GB, and update comments to match the new reality.
1.42.2.3	05-Jan-2001	bouyer	Sync with HEAD
1.42.2.2	08-Dec-2000	bouyer	Sync with HEAD.
1.42.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.43.2.16	29-Dec-2001	sommerfeld	Yet another mergeup. New work: - Frank van der Linden's tlb shootdown fix. Mainline functionality merged: - IrDA - Kernel RCSID's - Transmeta CPU support - ACPI - XMM register access through procfs
1.43.2.15	22-Sep-2001	sommerfeld	Merge branch with current, post ubcperf, MTRR, etc., etc., with the recent fifo_putpages fix to the trunk, GENERIC.MPDEBUG config boots multiuser on my BP6.
1.43.2.14	03-Sep-2001	sommerfeld	Merge with -current once more.
1.43.2.13	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.43.2.12	26-May-2001	sommerfeld	Merge up with -current
1.43.2.11	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.43.2.10	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.43.2.9	03-Jan-2001	thorpej	Clean up the TLB invalidation logic. Remove the last vestiges of the old "remove record" stuff, and just use the shootdown path, even for uniprocessor kernels. Change the shootdown path to defer actually peforming work (or sending the IPI to signal other processors to begin work) until explicitly triggered by a call to pmap_tlb_shootnow(). Also remove all of the PTP and PV entry stealing code -- it's complicated, not often called, and not really useful anymore now that UVM itself can handle pmap_enter() failures.
1.43.2.8	02-Jan-2001	thorpej	Mirror a change made on the trunk: Don't equate pmap_update() with tlbflush(), but instead call tlbflush() when that's what we mean (currently everywhere, except in the one place in MI code where it is called). The whole pmap_update() thing needs to be reexamined, but this helps to clarify things a little with the currently-defined semantics of that function.
1.43.2.7	31-Dec-2000	thorpej	- Keep track of which CPUs are using a given pmap, and use this information to flush the APTE space of other CPUs as necessary. - Use atomic operations in the appropriate places to avoid race conditions between the kernel and the CPU microcode which could lose modified/referenced information. Reviewed by Bill Sommerfeld.
1.43.2.6	23-Sep-2000	sommerfeld	Pending more complete pmap merge, deal with PMAP_PAGEIDLEZERO change
1.43.2.5	06-Sep-2000	sommerfeld	If MULTIPROCESSOR, allocate per-cpu VA space and special PTE for page zero, page copy, and ptp mapping. Eliminate the locks which were used to protect those special PTE/VA pairs since they're no longer necessary.
1.43.2.4	18-Aug-2000	sommerfeld	Do tlb shootdown work at splipi(). Pass cpu_info to tlb shootdown ipi entrypoint.
1.43.2.3	25-Jun-2000	sommerfeld	Merge up to just-post-1.5 -current
1.43.2.2	03-May-2000	sommerfeld	Merge with -current
1.43.2.1	20-Feb-2000	sommerfeld	MP: preliminary TLB shootdown support. Note that the pmap previously had its own idea of a deferred invalidation queue; use the TLB shootdown queue mechanism for this as well.
1.44.2.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.45.2.1	28-Sep-2000	is	Pullup from -current, requested by is, approved by thorpej: sys/arch/arm32/include/pmap.h 1.18 sys/arch/atari/include/pmap.h 1.21 sys/arch/i386/include/pmap.h 1.50 sys/arch/pc532/include/pmap.h 1.29 sys/arch/sh3/include/pmap.h 1.6 These are architectures that do implemented wired page accounting; advertize it to UVM, so that mlock() from non-root processes works.
1.53.2.8	11-Nov-2002	nathanw	Catch up to -current
1.53.2.7	18-Oct-2002	nathanw	Catch up to -current.
1.53.2.6	08-Jan-2002	nathanw	Catch up to -current.
1.53.2.5	14-Nov-2001	nathanw	Catch up to -current.
1.53.2.4	21-Sep-2001	nathanw	Catch up to -current.
1.53.2.3	24-Aug-2001	nathanw	Catch up with -current.
1.53.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.53.2.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.58.2.4	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.58.2.3	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.58.2.2	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.58.2.1	25-Aug-2001	thorpej	Merge Aug 24 -current into the kqueue branch.
1.59.2.1	01-Oct-2001	fvdl	Catch up with -current.
1.60.2.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.73.2.7	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.73.2.6	21-Sep-2004	skrll	Fix the sync with head I botched.
1.73.2.5	18-Sep-2004	skrll	Sync with HEAD.
1.73.2.4	03-Sep-2004	skrll	Sync with HEAD
1.73.2.3	25-Aug-2004	skrll	Sync with HEAD.
1.73.2.2	12-Aug-2004	skrll	Sync with HEAD.
1.73.2.1	03-Aug-2004	skrll	Sync with HEAD
1.82.12.7	04-Feb-2008	yamt	sync with head.
1.82.12.6	21-Jan-2008	yamt	sync with head
1.82.12.5	07-Dec-2007	yamt	sync with head
1.82.12.4	27-Oct-2007	yamt	sync with head.
1.82.12.3	03-Sep-2007	yamt	sync with head.
1.82.12.2	26-Feb-2007	yamt	sync with head.
1.82.12.1	21-Jun-2006	yamt	sync with head.
1.85.6.1	22-Apr-2006	simonb	Sync with head.
1.85.4.1	09-Sep-2006	rpaulo	sync with head
1.85.2.1	18-Feb-2006	yamt	sync with head.
1.86.16.2	10-Dec-2006	yamt	sync with head.
1.86.16.1	22-Oct-2006	yamt	sync with head
1.86.14.1	18-Nov-2006	ad	Sync with head.
1.88.4.1	27-Feb-2007	yamt	- sync with head. - move sched_changepri back to kern_synch.c as it doesn't know PPQ anymore.
1.89.22.3	23-Mar-2008	matt	sync with HEAD
1.89.22.2	09-Jan-2008	matt	sync with HEAD
1.89.22.1	06-Nov-2007	matt	sync with HEAD
1.89.20.6	03-Dec-2007	joerg	Sync with HEAD.
1.89.20.5	27-Nov-2007	joerg	Sync with HEAD. amd64 Xen support needs testing.
1.89.20.4	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.89.20.3	02-Oct-2007	joerg	Sync with HEAD.
1.89.20.2	10-Sep-2007	joerg	Introduce pmap_init_tmp_pgtbl to build a temporary copy of the kernel side page mapping and an identity mapping low page for use in real mode. Switch MP bootstrap and i386 ACPI wakeup code to use it.
1.89.20.1	03-Sep-2007	jmcneill	Sync with HEAD.
1.89.16.1	03-Sep-2007	skrll	Sync with HEAD.
1.89.12.2	17-Oct-2007	garbled	Sync with HEAD
1.89.12.1	03-Oct-2007	garbled	Sync with HEAD
1.89.4.8	03-Dec-2007	ad	Sync with HEAD.
1.89.4.7	23-Oct-2007	ad	Sync with head.
1.89.4.6	09-Oct-2007	ad	Sync with head.
1.89.4.5	23-Aug-2007	ad	Add pmap_pte_set, pmap_pte_setbits, pmap_pte_clearbits where missing.
1.89.4.4	21-Aug-2007	ad	- Add ci_cpumask and use in preference to (1 << ci_cpuid). Some processors are bad at shifting by non-constant amounts. - pmap_tlb_shootdown() must be called with preemption disabled. - Defer waiting for tlb shootdown to complete after kenter_pa()/kremove(). Spin in pmap_update() and in pmap_detactivate(), in case the LWP has been preempted - When preparing to send a multicast shootdown IPI, avoid scanning the list of CPUs if the pmap's bitmask indicates that it's only active locally.
1.89.4.3	18-Aug-2007	yamt	- fix TLB problems in recursive PTE mappings, which causes various crashes. as they are used by the kernel itself, they need to be synced regardless of lazy tlb flushing mode. - pmap_tlb_shootdown: fix the case of sva != -1 && eva == 0. ok'ed by Andrew Doran.
1.89.4.2	29-Jul-2007	ad	- When zeroing/copying pages, use SSE2 movtni to avoid polluting the cache. - By default, align assembly routines on 32-byte starting boundaries. - There are now 8 interrupt priority levels, half of which are softints. Update intrdefs.h to match. - Always clear/set spinlock words - removes lots of ifdefs. - Remove the horrible ci_self150 hack that I introduced. - Overhaul how TLB shootdown is performed. Inspired by a similar change in OpenBSD but implemented quite differently. This should be a lot faster but I have not benchmarked it yet.
1.89.4.1	28-Apr-2007	ad	- Make the PV cache per-CPU and eliminate the global pvalloc_lock. Cuts contention on the freelist locks by a factor of ~50 on a dual core box. - Mirror a patch for amd64 posted by oster@, and make the TLB shootdown freelist per-CPU. Increase the number of TLB shootdown jobs from 16 to 32 per CPU.
1.90.2.13	18-Oct-2007	yamt	sync with head.
1.90.2.12	14-Oct-2007	yamt	forward decl of trapframe.
1.90.2.11	08-Oct-2007	yamt	merge some parts of x86 pmap.h.
1.90.2.10	07-Oct-2007	yamt	tweak assertions to reduce diffs between i386 and amd64.
1.90.2.9	07-Oct-2007	yamt	remove unnecessary casts.
1.90.2.8	07-Oct-2007	yamt	sync comments and whitespaces.
1.90.2.7	07-Oct-2007	yamt	g/c nkpde.
1.90.2.6	07-Oct-2007	yamt	whitespace.
1.90.2.5	07-Oct-2007	yamt	remove unused definitions.
1.90.2.4	06-Oct-2007	yamt	sync with head.
1.90.2.3	04-Oct-2007	yamt	remove LARGEPAGES option. always use large pages if available.
1.90.2.2	24-Sep-2007	yamt	remove hardcoded numbers of PTPs used for kernel image and bootstrap tables.
1.90.2.1	23-Sep-2007	yamt	- sync with amd64. - pmap_alloc_level: remove fast call special case. - tweak some printf to work for both of LP64 and LP32. - remove machdep.nkpde sysctl. - remove automatic adjustment of nkpde for now. will revisit later.
1.91.2.3	26-Oct-2007	bouyer	Make amd64, i386 and xen kernels build and work again.
1.91.2.2	25-Oct-2007	bouyer	Finish sync with HEAD. Especially use the new x86 pmap for xenamd64. For this: - rename pmap_pte_set() to pmap_pte_testset() - make pmap_pte_set() a function or macro for non-atomic PTE write - define and use pmap_pa2pte()/pmap_pte2pa() to read/write PTE entries - define pmap_pte_flush() which is a nop in x86 case, and flush the MMUops queue in the Xen case
1.91.2.1	25-Oct-2007	bouyer	Sync with HEAD.
1.94.2.2	18-Feb-2008	mjf	Sync with HEAD.
1.94.2.1	08-Dec-2007	mjf	Sync with HEAD.
1.97.6.9	20-Jan-2008	bouyer	Sync with HEAD: make first argument of pmap_pte_cas() volatile.
1.97.6.8	19-Jan-2008	bouyer	Make things build again after sync with HEAD
1.97.6.7	19-Jan-2008	bouyer	Sync with HEAD
1.97.6.6	18-Jan-2008	bouyer	Fix APDP handling. A XEN i386PAE kernel now boots multiuser
1.97.6.5	17-Jan-2008	bouyer	- Fix L2_SLOT_APTE value (not sure how I got this value but it was definitively wrong) - Use global variable for the PAE L3 page adresses, so that pmap.c can get it from the bootstrap code - Extent the size of our virtual PDP from 3 to 4 pages, so that pmap->pm_pdir[] is contigous for the whole VA range. The last page is a shadow of the kernel's real PDP (L3[3]). - make pm_pdirpa an array of 4 paddr_t if using PAE. introduce a pmap_pdirpa macro to get the physical address of a given PD entry. - fix pmap_map_pte The kernel now boots single-user. fsck will cause a kernel fault in pmap_pdes_invalid() on exit.
1.97.6.4	15-Jan-2008	bouyer	Snapshot of work in progress: an Xen i386PAE kernel boots and start init on a amd64 dom0, but panics when init forks. This code needs a lot of cleanup, and the pmap handling is minimal to allow init to start. It's a proof of concept of how PAE on Xen can work. For PAE guest, the Xen MMU handling differs in some significant way from the i386 or amd64 Xen. The L3 page has only 4 entries, the last one mapping 0xc0000000->0xffffffff (which happens to be our kenrel VM range, that's cool). The L2 page pointed to by this last entry is handled specially by Xen because it contains some Xen private mapping, including a recursive mapping. So this page can only be pointed to by exactly one L3 entry, and nothing else (it can't be part of a recursive mapping for example). In addition, it would waste too much VA space to do recursive mapping at the L3 level. We do pmap switching at the L# level, instead of doing it though %cr3. %cr3 is static, as is L3[3] which contains only kenrel mappings. pmap_load() does pmap switching though the first 3 entries for L3. PTE mapping is done though 4 contigous L2 entries; the last one pointing to a shadow of L3[3]. This way we can consider we have a 2-level VM system, but with the L2 being 4 pages in size instead of one. The plx_i() macros can be used with it to access the PTE without changes. This can be reused as is for native PAE support (without the L3[3] shadow which wouldn't be needed here)
1.97.6.3	13-Jan-2008	bouyer	Update system macros for PAE. Because the L3 page directory has only 4 entries, and because of the way Xen handles it, it's better to have it has static as possible. So we claim to have only 2 levels, with level 2 being a "super" PD of 4 pages. pmap switches will be handled by changing the L3 entries instead of %cr3. Only the first 3 entries will have to be changed, the last one mapping KVA space only.
1.97.6.2	13-Jan-2008	bouyer	Work in progress on xeni386 PAE support: Make xeni386 build with a 64bit paddr_t. For this vaddr_t vs paddr_t vs pointers usages had to be clarified. If 'options PAE' is present in a Xen3 kernel, switch paddr_t, pd_entry_t and pt_entry_t to 64bits, and add the PAE entry in the __xen_guest ELF section.
1.97.6.1	11-Dec-2007	bouyer	Switch i386 to x86/x86/pmap.c
1.102.16.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.102.10.3	11-Aug-2010	yamt	sync with head.
1.102.10.2	11-Mar-2010	yamt	sync with head
1.102.10.1	04-May-2009	yamt	sync with head.
1.102.6.1	17-Jan-2009	mjf	Sync with HEAD.
1.103.14.3	16-Nov-2010	uebayasi	Sync with HEAD.
1.103.14.2	17-Aug-2010	uebayasi	Sync with HEAD.
1.103.14.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.103.8.4	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.103.8.3	28-Mar-2011	jym	Sync with HEAD. TODO before merge: - shortcut for suspend code in sysmon, when powerd(8) is not running. Borrow ``xs_watch'' thread context? - bug hunting in xbd + xennet resume. Rings are currently thrashed upon resume, so current implementation force flush them on suspend. It's not really needed.
1.103.8.2	10-Jan-2011	jym	Sync with HEAD
1.103.8.1	24-Oct-2010	jym	Sync with HEAD
1.105.2.1	05-Mar-2011	rmind	sync with head
1.108.4.1	08-Feb-2011	bouyer	Sync with HEAD
1.108.2.1	06-Jun-2011	jruoho	Sync with HEAD.
1.109.2.3	20-Sep-2011	cherry	Remove the "xpq lock", since we have per-cpu mmu queues now. This may need further testing. Also add some preliminary locking around queue-ops in the network backend driver
1.109.2.2	20-Aug-2011	cherry	PAE MP support (preliminary), amd64 per-cpu L4 model redesigned, i386 pmap_pa_start/end fixup
1.109.2.1	31-Jul-2011	cherry	grow MP support for i386. boots to single user
1.110.2.3	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.110.2.2	17-Apr-2012	yamt	sync with head
1.110.2.1	10-Nov-2011	yamt	sync with head
1.112.4.2	24-Feb-2012	mrg	sync to -current.
1.112.4.1	18-Feb-2012	mrg	merge to -current.
1.115.2.3	03-Dec-2017	jdolecek	update from HEAD
1.115.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.115.2.1	23-Jun-2013	tls	resync from head
1.116.10.1	10-Aug-2014	tls	Rebase.
1.116.6.1	18-May-2014	rmind	sync with head
1.117.8.1	04-Nov-2016	pgoyette	Sync with HEAD
1.117.4.1	05-Oct-2016	skrll	Sync with HEAD
1.118.16.2	08-Apr-2020	martin	Merge changes from current as of 20200406
1.118.16.1	10-Jun-2019	christos	Sync with HEAD
1.118.14.2	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.118.14.1	28-Jul-2018	pgoyette	Sync with HEAD
1.10	17-Jun-1999	thorpej	PMAP_NEW is no longer optional on the i386; the old pmap's page table allocation strategy no longer works at all. Move pmap.new.* to pmap.. To read the revision history of PMAP_NEW up until this merge, use cvs rlog of the old pmap.new. files.
1.9	12-May-1999	thorpej	Move the user-set LDT out of the PCB and into the pmap. Applications which set the LDT and share VM space (e.g. new versions of WINE) expect the LDT to be logically coupled to the address space. Use the new pmap_fork() interface to copy non-shared user-set LDTs when the address space is forked.
1.8	05-May-1999	chs	update resident_count and wired_count properly in pmap_kenter_*() and pmap_kremove().
1.7	13-Aug-1998	thorpej	branches: 1.7.6; vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.6	05-May-1998	chuck	correct some addresses in the comment that i missed when KERNBASE changed a while back (from Matthias Drochner)
1.5	02-Mar-1998	fvdl	Don't make simple_lock decls dependent on NCPU > 1, be consistent with the rest of the kernel.
1.4	18-Feb-1998	thorpej	Fixup an inconsistency in the declaration of PTDpaddr, pointed out by Chris Demetriou.
1.3	18-Feb-1998	cgd	Move pmap_map() function definition to MD headers, as appropriate. It's an internal function, and the VM system shouldn't try to prototype it. (Note that some ports _don't_ prototype it.)
1.2	06-Feb-1998	thorpej	RCS ID police.
1.1	06-Feb-1998	mrg	oops, forgot to add these new files
1.7.6.2	19-Feb-2000	he	Apply patch (requested by thorpej): Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1GB of virtual address space, leaving userland with 3GB, and update comments to match the new reality.
1.7.6.1	05-May-1999	perry	branches: 1.7.6.1.2; pullup 1.7->1.8 (chuq)
1.7.6.1.2.1	21-Jun-1999	thorpej	Sync w/ -current.
1.4	21-Aug-2022	riastradh	x86 Move VA_SIGN_POS/NEG to machine/pte.h. It's used by pl[1-4]_pi, also defined in machine/pte.h, and used in libkvm without pmap_private.h.
1.3	20-Aug-2022	riastradh	{amd64,i386}/pmap_private.h: Fix minor whitespace issues.
1.2	20-Aug-2022	riastradh	x86: Move definition of struct pmap to pmap_private.h. This makes pmap_resident_count and pmap_wired_count out-of-line functions instead of inline. No functional change intended otherwise.
1.1	20-Aug-2022	riastradh	x86: Split most of pmap.h into pmap_private.h or vmparam.h. This way pmap.h only contains the MD definition of the MI pmap(9) API, which loads of things in the kernel rely on, so changing x86 pmap internals no longer requires recompiling the entire kernel every time. Callers needing these internals must now use machine/pmap_private.h. Note: This is not x86/pmap_private.h because it contains three parts: 1. CPU-specific (different for i386/amd64) definitions used by... 2. common definitions, including Xenisms like xpmap_ptetomach, further used by... 3. more CPU-specific inlines for pmap_pte_* operations So {amd64,i386}/pmap_private.h defines 1, includes x86/pmap_private.h for 2, and then defines 3. Maybe we should split that out into a new pmap_pte.h to reduce this trouble. No functional change intended, other than that some .c files must include machine/pmap_private.h when previously uvm/uvm_pmap.h polluted the namespace with pmap internals. Note: This migrates part of i386/pmap.h into i386/vmparam.h -- specifically the parts that are needed for several constants defined in vmparam.h: VM_MAXUSER_ADDRESS VM_MAX_ADDRESS VM_MAX_KERNEL_ADDRESS VM_MIN_KERNEL_ADDRESS Since i386 needs PDP_SIZE in vmparam.h, I added it there on amd64 too, just to keep things parallel.
1.12	12-Jul-2018	maxv	Remove the kernel PMC code. Sent yesterday on tech-kern@. This change: * Removes "options PERFCTRS", the associated includes, and the associated ifdefs. In doing so, it removes several XXXSMPs in the MI code, which is good. * Removes the PMC code of ARM XSCALE. * Removes all the pmc.h files. They were all empty, except for ARM XSCALE. * Reorders the x86 PMC code not to rely on the legacy pmc.h file. The definitions are put in sysarch.h. * Removes the kern/sys_pmc.c file, and along with it, the sys_pmc_control and sys_pmc_get_info syscalls. They are marked as OBSOL in kern, netbsd32 and rump. * Removes the pmc_evid_t and pmc_ctr_t types. * Removes all the associated man pages. The sets are marked as obsolete.
1.11	10-Mar-2017	maxv	branches: 1.11.12; 1.11.14; unused
1.10	18-Feb-2017	maxv	There is currently an ugly mix between the PERFCTRS subsystem (MI), and i386's own PMC interface (MD). Stop using PERFCTRS and use PMC instead. While here remove some unused flags, which are wrong on the latest CPUs anyway.
1.9	17-Feb-2017	maxv	Support PMCs on multi-processor systems. Still several things to fix, but at least it works a little. Will be improved and moved into x86/ soon.
1.8	16-Apr-2007	ad	branches: 1.8.80; 1.8.100; 1.8.104; 1.8.108; Share the sysarch stuff between the x86 ports. PR kern/36046.
1.7	17-Jan-2003	thorpej	branches: 1.7.18; 1.7.56; 1.7.60; 1.7.62; Merge the nathanw_sa branch.
1.6	09-Aug-2002	thorpej	branches: 1.6.2; PMC_TYPE_I586 -> PMC_CLASS_I586 PMC_TYPE_I686 -> PMC_CLASS_I686 PMC_TYPE_K7 -> PMC_CLASS_K7 To reflect terminilogy used in pmc(3).
1.5	08-Aug-2002	thorpej	Add macros to encode the "unit number" in the event selector.
1.4	08-Aug-2002	thorpej	Define new-style PMC types for i586, i686, and K7 performance counters.
1.3	07-Aug-2002	briggs	Stubs for new pmc(9) interface.
1.2	07-Aug-2002	briggs	Forward declare structures that are used here.
1.1	13-Sep-2000	thorpej	branches: 1.1.2; 1.1.6; 1.1.8; 1.1.20; Add support for using the 586- and 686-class performance counters.
1.1.20.1	31-Aug-2002	gehenna	catch up with -current.
1.1.8.1	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.1.6.2	13-Aug-2002	nathanw	Catch up to -current.
1.1.6.1	21-Jun-2002	gmcgarry	LWPify
1.1.2.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.1.2.1	13-Sep-2000	bouyer	file pmc.h was added on branch thorpej_scsipi on 2000-11-20 20:09:30 +0000
1.6.2.2	09-Aug-2002	thorpej	PMC_TYPE_I586 -> PMC_CLASS_I586 PMC_TYPE_I686 -> PMC_CLASS_I686 PMC_TYPE_K7 -> PMC_CLASS_K7 To reflect terminilogy used in pmc(3).
1.6.2.1	09-Aug-2002	thorpej	file pmc.h was added on branch sommerfeld_i386mp_1 on 2002-08-09 05:28:09 +0000
1.7.62.1	11-Jul-2007	mjf	Sync with head.
1.7.60.1	27-May-2007	ad	Sync with head.
1.7.56.1	07-May-2007	yamt	sync with head.
1.7.18.1	03-Sep-2007	yamt	sync with head.
1.8.108.1	21-Apr-2017	bouyer	Sync with HEAD
1.8.104.1	20-Mar-2017	pgoyette	Sync with HEAD
1.8.100.1	28-Aug-2017	skrll	Sync with HEAD
1.8.80.1	03-Dec-2017	jdolecek	update from HEAD
1.11.14.1	10-Jun-2019	christos	Sync with HEAD
1.11.12.1	28-Jul-2018	pgoyette	Sync with HEAD
1.48	13-Jun-2020	ad	Print a rate limited warning if the TSC timecounter goes backwards from the viewpoint of any single LWP.
1.47	13-Jan-2020	ad	Remove now unused mdlwp fields md_gc_pmap and md_gc_ptp.
1.46	12-Oct-2019	maxv	branches: 1.46.2; Rewrite the FPU code on x86. This greatly simplifies the logic and removes the dependency on IPL_HIGH. NVMM is updated accordingly. Posted on port-amd64 a week ago. Bump the kernel version to 9.99.16.
1.45	25-Feb-2017	kamil	branches: 1.45.14; Garbage collect unneeded inclusion of <x86/dbregs.h> in <machine/proc.h> This is left over after introduction of Debug Register accessors. This interface replaced older watchpoint API. Sponsored by <The NetBSD Foundation>
1.44	23-Feb-2017	kamil	Introduce PT_GETDBREGS and PT_SETDBREGS in ptrace(2) on i386 and amd64 This interface is modeled after FreeBSD API with the usage. This replaced previous watchpoint API. The previous one was introduced recently in NetBSD-current and remove its spurs without any backward-compatibility. Design choices for Debug Register accessors: - exec() (TRAP_EXEC event) must remove debug registers from LWP - debug registers are only per-LWP, not per-process globally - debug registers must not be inherited after (v)forking a process - debug registers must not be inherited after forking a thread - a debugger is responsible to set global watchpoints/breakpoints with the debug registers, to achieve this PTRACE_LWP_CREATE/PTRACE_LWP_EXIT event monitoring function is designed to be used - debug register traps must generate SIGTRAP with si_code TRAP_DBREG - debugger is responsible to retrieve debug register state to distinguish the exact debug register trap (DR6 is Status Register on x86) - kernel must not remove debug register traps after triggering a trap event a debugger is responsible to detach this trap with appropriate PT_SETDBREGS call (DR7 is Control Register on x86) - debug registers must not be exposed in mcontext - userland must not be allowed to set a trap on the kernel Implementation notes on i386 and amd64: - the initial state of debug register is retrieved on boot and this value is stored in a local copy (initdbregs), this value is used to initialize dbreg context after PT_GETDBREGS - struct dbregs is stored in pcb as a pointer and by default not initialized - reserved registers (DR4-DR5, DR9-DR15) are ignored Further ideas: - restrict this interface with securelevel Tested on real hardware i386 (Intel Pentium IV) and amd64 (Intel i7). This commit enables 390 debug register ATF tests in kernel/arch/x86. All tests are passing. This commit does not cover netbsd32 compat code. Currently other interface PT_GET_SIGINFO/PT_SET_SIGINFO is required in netbsd32 compat code in order to validate reliably PT_GETDBREGS/PT_SETDBREGS. This implementation does not cover FreeBSD specific defines in their <x86/reg.h>: DBREG_DR7_LOCAL_ENABLE, DBREG_DR7_GLOBAL_ENABLE, DBREG_DR7_LEN_1 etc. These values tend to be reinvented by each tracer on its own. GNU Debugger (GDB) works with NetBSD debug registers after adding this patch: --- gdb/amd64bsd-nat.c.orig 2016-02-10 03:19:39.000000000 +0000 +++ gdb/amd64bsd-nat.c @@ -167,6 +167,10 @@ amd64bsd_target (void) #ifdef HAVE_PT_GETDBREGS +#ifndef DBREG_DRX +#define DBREG_DRX(d,x) ((d)->dr[(x)]) +#endif + static unsigned long amd64bsd_dr_get (ptid_t ptid, int regnum) { Another reason to stop introducing unpopular defines covering machine specific register macros is that these value varies across generations of the same CPU family. GDB demo: (gdb) c Continuing. Watchpoint 2: traceme Old value = 0 New value = 16 main (argc=1, argv=0x7f7fff79fe30) at test.c:8 8 printf("traceme=%d\n", traceme); (Currently the GDB interface is not reliable due to NetBSD support bugs) Sponsored by <The NetBSD Foundation>
1.43	15-Dec-2016	kamil	branches: 1.43.2; Add support for hardware assisted watchpoints/breakpoints API in ptrace(2) Add new ptrace(2) calls: - PT_COUNT_WATCHPOINTS - count the number of available hardware watchpoints - PT_READ_WATCHPOINT - read struct ptrace_watchpoint from the kernel state - PT_WRITE_WATCHPOINT - write new struct ptrace_watchpoint state, this includes enabling and disabling watchpoints The ptrace_watchpoint structure contains MI and MD parts: typedef struct ptrace_watchpoint { int pw_index; /* HW Watchpoint ID (count from 0) / lwpid_t pw_lwpid; / LWP described / struct mdpw pw_md; / MD fields / } ptrace_watchpoint_t; For example amd64 defines MD as follows: struct mdpw { void md_address; int md_condition; int md_length; }; These calls are protected with the __HAVE_PTRACE_WATCHPOINTS guard. Tested on amd64, initial support added for i386 and XEN. Sponsored by <The NetBSD Foundation>
1.42	20-Feb-2014	dsl	branches: 1.42.6; 1.42.10; Move the amd64 and i386 pcb to the bottom of the uarea, and move the kernel stack to the top. Change the pcb layouts so that fpu save area is at the end and is 64byte aligned ready for xsave (saving the ymm registers). Welcome to 6.99.32
1.41	15-Feb-2014	dsl	Remove all references to MDL_USEDFPU and deferred fpu initialisation. The cost of zeroing the save area on exec is minimal. This stops the FP registers of a random process being used the first time an lwp uses the fpu. sendsig_siginfo() and get_mcontext() now unconditionally copy the FP registers. I'll remove the double-copy for signal handlers soon. get_mcontext() might have been leaking kernel memory to userspace - and may still do so if i386_use_fxsave is false (short copies).
1.40	01-Dec-2013	christos	revert fpu/pcu changes until we figure out what's wrong; they cause random freezes
1.39	23-Oct-2013	drochner	Use the MI "pcu" framework for bookkeeping of npx/fpu states on x86. This reduces the amount of MD code enormously, and makes it easier to implement support for newer CPU features which require more fpu state, or for fpu usage by the kernel. For access to FPU state across CPUs, an xcall kthread is used now rather than a dedicated IPI. No user visible changes intended.
1.38	14-Jan-2011	rmind	branches: 1.38.8; 1.38.18; 1.38.22; Retire struct user, remove sys/user.h inclusions. Note sys/user.h header as obsolete. Remove USER_TO_UAREA/UAREA_TO_USER macros. Various #include fixes and review by matt@.
1.37	20-Dec-2008	ad	branches: 1.37.2; 1.37.6; Remove stray opt_redzone.h
1.36	05-Jun-2008	ad	branches: 1.36.6; 1.36.8; pmap_remove_all() for x86. Also, always defer freeing ptps to pmap_update(). There may be a better way to do this, but for now this is simple and avoids potential bugs. Proposed on tech-kern and discussed with chs@.
1.35	04-Jan-2008	yamt	branches: 1.35.6; 1.35.8; 1.35.10; 1.35.12; i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.34	16-Nov-2007	skrll	branches: 1.34.6; s/proc/lwp/ in comment
1.33	10-Nov-2007	ad	fsbase/gsbase: - Fix a few bugs with it, in particular fork/exec handling. - Store the descriptors in the PCB, not in the LWP.
1.32	10-Nov-2007	ad	- When computing the TSC frequency, call i8254_delay() and not DELAY(). - Use atomics to adjust the pmap reference count, instead of taking locks. - Implement I386_{SET,GET}_{FS,GS}BASE, allowing %fs and %gs to be used as per-thread registers. This is compatible with FreeBSD. - Run patches after we have attached CPUs, since we then know if the system is uniprocessor or not. Eliminates a lot of #ifdef MULTIPROCESSOR and makes running MP kernels on UP systems cheaper. - Patch out many of the 'lock' prefixes to nops if uniprocessor. - Do a wbinvd after patching to ensure that the trace/instruction cache is up to date.
1.31	04-Mar-2007	christos	branches: 1.31.2; 1.31.18; 1.31.20; 1.31.24; 1.31.26; Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.30	09-Feb-2007	ad	branches: 1.30.2; Merge newlock2 to head.
1.29	23-Sep-2006	yamt	fix a typo in 1.28. (ccaddr_t -> caddr_t)
1.28	22-May-2006	yamt	branches: 1.28.6; 1.28.8; i386 and xen: move struct user to the highest address in uarea and reduce UPAGES from 4 to 3 in the case of !NOREDZONE.
1.27	24-Dec-2005	perry	branches: 1.27.4; 1.27.6; 1.27.8; 1.27.12; 1.27.14; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.26	21-Feb-2004	junyoung	branches: 1.26.16; - Locate comment in right place. - Fix wrt protection against multiple inclusion.
1.25	28-Jan-2004	yamt	rename i386 mdlwp flags from MDP_ to MDL_ to avoid confusion with mdproc flags which also have MDP_ prefixes.
1.24	14-Jan-2004	yamt	remove a duplicated definition of MDP_USEDFPU.
1.23	27-Oct-2003	junyoung	Nuke __P().
1.22	20-Aug-2003	fvdl	Pass pointers to frames from assembly, do not use the 'frame on stack as argument passed by value' trick, as gcc 3.3.x makes (valid) assumptions about the stack that will not be true. Costs 2 instructions per trap/syscall on i386, 4 per interrupt for MP. One instruction per trap/syscall on amd64, 2 per interrupt for MP. I expect gcc 3.3.1 to make up for this by better optimization (it'd better..) While here, make amd64 compile again by using subr_mbr_disk.c
1.21	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.20	23-Jun-2003	martin	branches: 1.20.2; Protect kernel opt_*.h include by #ifdef _KERNEL_OPT
1.19	23-Jun-2003	martin	Make sure to include opt_foo.h if a defflag option FOO is used.
1.18	02-Apr-2003	thorpej	Use PAGE_SIZE rather than NBPG.
1.17	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.16	12-Oct-2002	yamt	re-do my previous commit correctly
1.15	11-Oct-2002	yamt	override KSTACK_LOWEST_ADDR and KSTACK_SIZE for redzone.
1.14	10-Sep-2001	fvdl	branches: 1.14.2; Add flag to indicate that a process set some mtrrs that need to be cleaned up automatically on exit.
1.13	11-Dec-2000	tron	branches: 1.13.4; 1.13.6; 1.13.8; Include "machine/frame.h" to avoid lint errors while building "libc".
1.12	11-Dec-2000	mycroft	Introduce 2 new flags in types.h: * __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by e_syscall_intern, which is called at key places in the kernel. This can be used to set a MD syscall handler pointer. This obsoletes and replaces the _HAS_SEPARATED_SYSCALL flags. __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in struct emul are omitted.
1.11	16-Aug-2000	thorpej	Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.10	06-Aug-1995	mycroft	branches: 1.10.32; 1.10.44; Move the `used fpu' flag into mdproc, so it can be referenced when a process is swapped out. Implement process_{read,write}_fpregs.
1.9	01-May-1995	mycroft	Make md_regs a pointer to a trapframe.
1.8	01-May-1995	mycroft	Rename ibcs2_sigflags to __spare, but leave the space for now.
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	15-Aug-1994	mycroft	Add flags for iBCS2 compat.
1.5	05-May-1994	cgd	branches: 1.5.2; lots of changes: prototype migration, move lots of variables, definitions, and structure elements around. kill some unnecessary type and macro definitions. standardize clock handling. More changes than you'd want.
1.4	21-Dec-1993	mycroft	branches: 1.4.2; Remove two unused fields.
1.3	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.1	14-Nov-1993	mycroft	md_flags is not used.
1.4.2.2	21-Dec-1993	mycroft	Remove two unused fields.
1.4.2.1	21-Dec-1993	mycroft	file proc.h was added on branch magnum on 1993-12-21 09:47:58 +0000
1.5.2.1	15-Aug-1994	mycroft	update from trunk
1.10.44.1	16-Aug-2000	thorpej	Update from trunk: Move the TSS selector out of the PCB and put it into mdproc. We need to access this when we have the proclist locked for reading, and thus cannot store it in the PCB (which may be swapped out). As part of this, call pmap_activate() from cpu_switch() to switch to the new address space, and refresh the PCB's copy of the LDT selector from the pmap structure (see above paragraph). We need to do this for MP support anyhow. Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via gdt_compact() reported by Nathan Williams.
1.10.32.2	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.10.32.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.13.8.1	01-Oct-2001	fvdl	Catch up with -current.
1.13.6.1	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.13.4.4	25-Nov-2002	nathanw	Make ASTs per-proc rather than per-cpu; per-proc makes more sense when a process that needs a trap can be running on multiple CPUs.
1.13.4.3	18-Oct-2002	nathanw	Catch up to -current.
1.13.4.2	21-Sep-2001	nathanw	Catch up to -current.
1.13.4.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.14.2.2	10-Sep-2001	fvdl	Add flag to indicate that a process set some mtrrs that need to be cleaned up automatically on exit.
1.14.2.1	10-Sep-2001	fvdl	file proc.h was added on branch sommerfeld_i386mp_1 on 2001-09-10 10:11:22 +0000
1.20.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.20.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.20.2.1	03-Aug-2004	skrll	Sync with HEAD
1.26.16.6	21-Jan-2008	yamt	sync with head
1.26.16.5	07-Dec-2007	yamt	sync with head
1.26.16.4	03-Sep-2007	yamt	sync with head.
1.26.16.3	26-Feb-2007	yamt	sync with head.
1.26.16.2	30-Dec-2006	yamt	sync with head.
1.26.16.1	21-Jun-2006	yamt	sync with head.
1.27.14.1	19-Jun-2006	chap	Sync with head.
1.27.12.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.27.8.1	24-May-2006	yamt	sync with head.
1.27.6.1	01-Jun-2006	kardel	Sync with head.
1.27.4.1	09-Sep-2006	rpaulo	sync with head
1.28.8.1	22-Oct-2006	yamt	sync with head
1.28.6.2	18-Nov-2006	ad	Sync with head.
1.28.6.1	20-Oct-2006	ad	Make ASTs per-LWP.
1.30.2.1	12-Mar-2007	rmind	Sync with HEAD.
1.31.26.2	18-Feb-2008	mjf	Sync with HEAD.
1.31.26.1	19-Nov-2007	mjf	Sync with HEAD.
1.31.24.1	18-Nov-2007	bouyer	Sync with HEAD
1.31.20.1	09-Jan-2008	matt	sync with HEAD
1.31.18.2	21-Nov-2007	joerg	Sync with HEAD.
1.31.18.1	11-Nov-2007	joerg	Sync with HEAD.
1.31.2.1	03-Dec-2007	ad	Sync with HEAD.
1.34.6.1	08-Jan-2008	bouyer	Sync with HEAD
1.35.12.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.35.10.1	04-May-2009	yamt	sync with head.
1.35.8.1	17-Jun-2008	yamt	sync with head.
1.35.6.2	17-Jan-2009	mjf	Sync with HEAD.
1.35.6.1	29-Jun-2008	mjf	Sync with HEAD.
1.36.8.1	16-Feb-2009	snj	Pull up following revision(s) (requested by ad in ticket #355): sys/arch/i386/i386/vm_machdep.c: revision 1.143 sys/arch/i386/include/proc.h: revision 1.37 Remove stray opt_redzone.h
1.36.6.1	19-Jan-2009	skrll	Sync with HEAD.
1.37.6.1	05-Mar-2011	rmind	sync with head
1.37.2.1	28-Mar-2011	jym	Sync with HEAD. TODO before merge: - shortcut for suspend code in sysmon, when powerd(8) is not running. Borrow ``xs_watch'' thread context? - bug hunting in xbd + xennet resume. Rings are currently thrashed upon resume, so current implementation force flush them on suspend. It's not really needed.
1.38.22.1	18-May-2014	rmind	sync with head
1.38.18.2	03-Dec-2017	jdolecek	update from HEAD
1.38.18.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.38.8.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.42.10.2	20-Mar-2017	pgoyette	Sync with HEAD
1.42.10.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.42.6.2	28-Aug-2017	skrll	Sync with HEAD
1.42.6.1	05-Feb-2017	skrll	Sync with HEAD
1.43.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.45.14.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.45.14.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.46.2.1	17-Jan-2020	ad	Sync with head.
1.38	02-Nov-2021	ryo	In order to prevent _mcount() from being recursively called when built with COPTS=-O0, sprinkle `__always_inline' to make _mcount() be generated as a single function.
1.37	08-Jun-2017	christos	use compiler builtins to compute addresses instead %ebp assembly arithmetic. (from joerg)
1.36	04-Jun-2017	christos	- prevent optimization from omitting the frame pointer (and re-using it); we use it in our inline assembly. - convert to c a little assembly to make the code more understandable.
1.35	31-May-2017	christos	branches: 1.35.2; prevent coredumps by skipping everything early if we are not counting.
1.34	10-Jan-2016	ryo	__mcount_lock is moved to MI from MD. because it is needed for all MULTIPROCESSOR arch, but it is exists only in i386 and amd64. ok christos@, on tech-kern@
1.33	20-Dec-2007	ad	branches: 1.33.54; 1.33.74; - Make __cpu_simple_lock and similar real functions and patch at runtime. - Remove old x86 atomic ops. - Drop text alignment back to 16 on i386 (really, this time). - Minor cleanup.
1.32	17-Oct-2007	garbled	branches: 1.32.4; 1.32.8; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.31	26-Sep-2007	ad	Inline interrupt disable for _mcount, so the real function calls don't screw with profiling.
1.30	26-Sep-2007	ad	Make build in userspace again.
1.29	26-Sep-2007	xtraeme	Fix profiling kernels: read_psl -> x86_read_psl write_psl -> x86_write_psl disable_intr -> x86_disable_intr
1.28	16-Apr-2007	christos	branches: 1.28.2; 1.28.10; 1.28.12; 1.28.14; As noted by uwe, we need to also save %edx because it can be used with the regparm calling convention. XXX: The c stub should be replaced with an assembly one.
1.27	16-Apr-2007	christos	save and restore %eax, %ecx, otherwise profiling does not work properly. From Anon Ymous. Example program that fails: int main(int argc, char *argv[]) { return argc; }
1.26	09-Feb-2007	ad	branches: 1.26.2; 1.26.6; 1.26.8; 1.26.12; Merge newlock2 to head.
1.25	16-Feb-2006	perry	branches: 1.25.14; 1.25.18; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.24	24-Dec-2005	perry	branches: 1.24.2; 1.24.4; 1.24.6; __asm__ -> __asm __const__ -> const __inline__ -> inline __volatile__ -> volatile
1.23	24-Dec-2005	perry	Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.22	02-Oct-2005	chs	put mcount.c in the normal object list, so that "make depend" works. use a gcc attribute to prevent it from being instrumented.
1.21	25-Feb-2005	chs	branches: 1.21.4; fix a typoe in previous.
1.20	21-Feb-2005	chs	allow profiling of MP kernels, add a spinlock around the body of mcount().
1.19	24-May-2004	yamt	branches: 1.19.4; 1.19.6; instead of blindly set IF in MCOUNT_EXIT, save eflags in MCOUNT_ENTER and restore it.
1.18	27-Oct-2003	junyoung	Nuke __P().
1.17	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.16	30-Nov-2001	enami	branches: 1.16.2; 1.16.18; Declare some asm statement with output as volatile. Without this, gcc with -march=pentiumpro produce bad code.
1.15	23-Aug-2001	chs	use cli/sti instead of splhigh/splx for MCOUNT_ENTER/MCOUNT_EXIT. this makes clearer how much time mcount() is really taking.
1.14	22-Jan-2000	mycroft	branches: 1.14.8; 1.14.10; Delint.
1.13	28-Sep-1999	enami	branches: 1.13.2; Add missing backslash.
1.12	27-Sep-1999	kleink	For ELF, rename the profiling entry from mcount to __mcount to avoid clashing with the user's namespace, also within file scope; for binary compatibility, define a weak alias for the previous name to cover our tracks.
1.11	13-Nov-1998	christos	cast _mcount args to u_long, to avoid lint warnings.
1.10	05-Nov-1997	thorpej	asm -> __asm__
1.9	07-Dec-1996	fvdl	branches: 1.9.14; Remove _MCOUNT_DECL prototype once more.
1.8	05-Dec-1996	fvdl	Add _MCOUNT_DECL prototype.
1.7	04-Dec-1996	fvdl	Change asm.h to make it usable for kernel profiling as well. Make mcount() a prototype in profile.h
1.6	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.5	27-Oct-1994	cgd	new RCS ID format.
1.4	13-May-1994	mycroft	Duh.
1.3	13-May-1994	mycroft	Need some more macros not in the 4.4-Lite version.
1.2	13-May-1994	mycroft	Do this better.
1.1	13-May-1994	mycroft	Needed for new profiling code.
1.9.14.1	05-Nov-1997	thorpej	Pull up from trunk: asm -> __asm__
1.13.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.14.10.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.14.10.1	25-Aug-2001	thorpej	Merge Aug 24 -current into the kqueue branch.
1.14.8.2	08-Jan-2002	nathanw	Catch up to -current.
1.14.8.1	21-Sep-2001	nathanw	Catch up to -current.
1.16.18.5	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.16.18.4	04-Mar-2005	skrll	Sync with HEAD. Hi Perry!
1.16.18.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.16.18.2	18-Sep-2004	skrll	Sync with HEAD.
1.16.18.1	03-Aug-2004	skrll	Sync with HEAD
1.16.2.2	30-Nov-2001	enami	Declare some asm statement with output as volatile. Without this, gcc with -march=pentiumpro produce bad code.
1.16.2.1	30-Nov-2001	enami	file profile.h was added on branch sommerfeld_i386mp_1 on 2001-11-30 06:53:57 +0000
1.19.6.1	19-Mar-2005	yamt	sync with head. xen and whitespace. xen part is not finished.
1.19.4.1	29-Apr-2005	kent	sync with -current
1.21.4.5	21-Jan-2008	yamt	sync with head
1.21.4.4	27-Oct-2007	yamt	sync with head.
1.21.4.3	03-Sep-2007	yamt	sync with head.
1.21.4.2	26-Feb-2007	yamt	sync with head.
1.21.4.1	21-Jun-2006	yamt	sync with head.
1.24.6.1	22-Apr-2006	simonb	Sync with head.
1.24.4.1	09-Sep-2006	rpaulo	sync with head
1.24.2.1	18-Feb-2006	yamt	sync with head.
1.25.18.1	20-Apr-2007	bouyer	Pull up following revision(s) (requested by christos in ticket #579): sys/arch/i386/include/profile.h: revisions 1.27 - 1.28 save and restore %eax, %ecx, otherwise profiling does not work properly.
1.25.14.1	06-Feb-2007	ad	mcount(): fix entry so LOCKDEBUG+GPROF can be used together. Previously it would recurse until eventually the machine triple faulted.
1.26.12.1	18-Apr-2007	thorpej	Convert i386 and amd64 to the new atomic ops API.
1.26.8.1	11-Jul-2007	mjf	Sync with head.
1.26.6.2	09-Oct-2007	ad	Sync with head.
1.26.6.1	27-May-2007	ad	Sync with head.
1.26.2.1	07-May-2007	yamt	sync with head.
1.28.14.1	06-Oct-2007	yamt	sync with head.
1.28.12.2	09-Jan-2008	matt	sync with HEAD
1.28.12.1	06-Nov-2007	matt	sync with HEAD
1.28.10.1	02-Oct-2007	joerg	Sync with HEAD.
1.28.2.1	03-Oct-2007	garbled	Sync with HEAD
1.32.8.1	02-Jan-2008	bouyer	Sync with HEAD
1.32.4.1	26-Dec-2007	ad	Sync with head.
1.33.74.2	28-Aug-2017	skrll	Sync with HEAD
1.33.74.1	19-Mar-2016	skrll	Sync with HEAD
1.33.54.1	03-Dec-2017	jdolecek	update from HEAD
1.35.2.1	22-Mar-2018	martin	Pull up following revision(s) (requested by christos in ticket #653): sys/arch/i386/include/profile.h: revision 1.36 sys/arch/i386/include/profile.h: revision 1.37 - prevent optimization from omitting the frame pointer (and re-using it); we use it in our inline assembly. - convert to c a little assembly to make the code more understandable. use compiler builtins to compute addresses instead %ebp assembly arithmetic. (from joerg)
1.35	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.34	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.33	30-May-2001	mrg	branches: 1.33.2; use _KERNEL_OPT
1.32	15-Jan-1998	thorpej	branches: 1.32.28; Generate dependencies on the VM86 option.
1.31	19-Mar-1997	mycroft	Do not permit the user to set the nested task flag.
1.30	13-May-1996	mycroft	branches: 1.30.4; Move spl*() and soft interrupt definitions into intr.h. For now, include it from psl.h. XXX Will change after release.
1.29	12-May-1996	mycroft	Move IPL_* and IST_* constants into intr.h.
1.28	03-May-1996	christos	Add missing prototypes.
1.27	01-Feb-1996	mycroft	LOCORE -> _LOCORE
1.26	07-Jan-1996	mycroft	Hand-code softintr() to deal with GCC braindamage.
1.25	07-Jan-1996	mycroft	Deal with GCC's dead code elimination being suboptimal.
1.24	07-Jan-1996	mycroft	Modify splraise() to allow better optimization.
1.23	07-Jan-1996	mycroft	Make cpl, ipending, and astpending volatile.
1.22	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.21	13-Aug-1995	mycroft	Replace splnet() with splsoftnet(). Add splnet().
1.20	04-Jul-1995	mycroft	Make hardware interrupts always block all software interrupts.
1.19	04-Jun-1995	mycroft	Increase size of imask[]. Sigh.
1.18	04-Jun-1995	mycroft	Add a separate IPL_IMP, used by splimp().
1.17	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.16	15-Jan-1995	mycroft	Remove PSL_USERCLR, and add a PSL_USERSTATIC. Change the bit descriptions to more closely match the 386 manual. (Probably the mnemonics should be changed, too.)
1.15	03-Jan-1995	mycroft	Add interrupt sharing types.
1.14	06-Nov-1994	mycroft	As pointed out by Bruce, splsoftclock() needs to process pending interrupts when lowering the priority. Do this by simply calling splx(). For the benefit of NFS (for now), modify splx() to return a value so that splsoftclock() does. This work is optimized out for every use of splx() that ignores the value, so it only affects relevant code. Also, clean up this file a little.
1.13	27-Oct-1994	cgd	new RCS ID format.
1.12	19-Sep-1994	mycroft	Prototype spllower().
1.11	13-Aug-1994	mycroft	Don't allow VM86 mode, for now.
1.10	05-May-1994	mycroft	branches: 1.10.2; Mostly stylistic change.
1.9	05-May-1994	cgd	lots of changes: prototype migration, move lots of variables, definitions, and structure elements around. kill some unnecessary type and macro definitions. standardize clock handling. More changes than you'd want.
1.8	07-Apr-1994	mycroft	Implement dynamic IRQ configuration and IRQ sharing. Inline spl*() calls. Reorganize and clean up the relevant code.
1.7	09-Mar-1994	mycroft	Back out previous change for now.
1.6	08-Mar-1994	mycroft	Inline spl*() calls.
1.5	09-Jan-1994	mycroft	Use PSL_MBO and PSL_MBZ in the more traditional way.
1.4	22-Dec-1993	cgd	add (possibly iffy) definitions of PSL_USERSET and PSL_USERCLR, "standard" PSL masks.
1.3	15-Dec-1993	mycroft	Cleanup of startup and signal handling code from magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.9	01-Feb-1994	mycroft	Sigh. There is code which includes this twice.
1.2.4.8	11-Jan-1994	mycroft	PSL_MB[OZ] changes from main branch.
1.2.4.7	08-Nov-1993	mycroft	Garbage collect some constants, and the entire 286 nonsense. (Like we even get to start() on a 286? NOT!)
1.2.4.6	11-Oct-1993	mycroft	Move {disable,enable}_intr() to psl.h, and inline all of {in,out}{,s}[bwl] in pio.h.
1.2.4.5	09-Oct-1993	mycroft	Add splast(), and correct definitions of splsoftclock() and splstatclock().
1.2.4.4	06-Oct-1993	mycroft	Don't define spl stuff is not in kernel or in locore.s.
1.2.4.3	30-Sep-1993	mycroft	Temporary kluge.
1.2.4.2	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.2.4.1	14-Sep-1993	mycroft	New i386 code.
1.10.2.2	06-Oct-1994	mycroft	Update from trunk.
1.10.2.1	13-Aug-1994	mycroft	update from trunk
1.30.4.1	19-Mar-1997	mycroft	Pull up trap handling changes.
1.32.28.2	21-Jun-2001	nathanw	Catch up to -current.
1.32.28.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.33.2.2	30-May-2001	mrg	use _KERNEL_OPT
1.33.2.1	30-May-2001	mrg	file psl.h was added on branch sommerfeld_i386mp_1 on 2001-05-30 12:28:46 +0000
1.36	21-Aug-2022	riastradh	x86 Move VA_SIGN_POS/NEG to machine/pte.h. It's used by pl[1-4]_pi, also defined in machine/pte.h, and used in libkvm without pmap_private.h.
1.35	20-Aug-2022	riastradh	x86: Forbid using x86/pte.h directly; use machine/pte.h. machine/pte.h already used outside sys/arch, so let's make it the primary thing and make sure to use x86/pte.h only as a subroutine.
1.34	20-Aug-2022	riastradh	amd64/pte.h, i386/pte.h: Need sys/stdint.h for uintN_t.
1.33	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.32	25-Apr-2020	maxv	Switch to the new PTE naming. The old naming is now unused, remove it.
1.31	09-Mar-2019	maxv	branches: 1.31.10; Start replacing the x86 PTE bits.
1.30	07-Mar-2019	maxv	Introduce a new set of PTE bits, with a different naming convention. PG_V -> PTE_P /* Present / PG_RW -> PTE_W / Write / PG_u -> PTE_U / User / PG_WT -> PTE_PWT / Write-Through / PG_N -> PTE_PCD / Cache-Disable / PG_U -> PTE_A / Accessed / PG_M -> PTE_D / Dirty / PG_PAT -> PTE_PAT / PAT on 4KB Pages / PG_PS -> PTE_PS / Large Page Size / PG_G -> PTE_G / Global Translation / PG_AVAIL1 -> PTE_AVL1 / Ignored by Hardware / PG_AVAIL2 -> PTE_AVL2 / Ignored by Hardware / PG_AVAIL3 -> PTE_AVL3 / Ignored by Hardware / PG_LGPAT -> PTE_LGPAT / PAT on Large Pages / PG_NX -> PTE_NX / No Execute */ Until now we were using "PG_BIT". The "BIT" part of the naming did not follow the x86 naming convention in the spec, and was very confusing. We don't want the "PG_" part of it either, because UVM has similar flags (ie PG_BUSY).
1.29	07-Mar-2019	maxv	Drop PG_RO, PG_KR and PG_PROT, they are useless and create confusion.
1.28	07-Mar-2019	maxv	Style, and remove useless comments.
1.27	01-Feb-2011	chuck	branches: 1.27.56; udpate license clauses on my code to match the new-style BSD licenses. based on diff that rmind@ sent me. no functional change with this commit.
1.26	02-Oct-2010	jym	branches: 1.26.2; 1.26.4; Update the comment for PG_PS when used with PAE.
1.25	14-Sep-2010	jym	Add ULL suffix to PG_NX when it is defined.
1.24	06-Jul-2010	cegger	Turn PMAP_NOCACHE into MI flag. Add MI flags PMAP_WRITE_COMBINE, PMAP_WRITE_BACK, PMAP_NOCACHE_OVR. Update pmap(9) manpage. hppa: Remove MD PMAP_NOCACHE flag as it exists as MI flag mips: Rename MD PMAP_NOCACHE to PGC_NOCACHE. x86: Implement new MI flags using Page-Attribute Tables. x86: Implement BUS_SPACE_MAP_PREFETCHABLE. Patch presented on tech-kern@: http://mail-index.netbsd.org/tech-kern/2010/06/30/msg008458.html No comments on this last version.
1.23	04-May-2010	jym	Enable the NX bit feature for Xen i386pae and amd64 kernels. Tested with Xen 3.1 and Xen 3.3, dom0 and domU, by bouyer@ and jym@. Ok bouyer@.
1.22	06-Apr-2010	jld	Fix typos in PAE comment.
1.21	26-Feb-2010	jym	branches: 1.21.2; Fixes regarding paddr_t/pd_entry_t types in MD x86 code, exposed by PAE: - NBPD_* macros are set to the types that better match their architecture (UL for i386 and amd64, ULL for i386 PAE) - will revisit when paddr_t is set to 64 bits for i386 non-PAE. - type fixes in printf/printk messages (Use PRIxPADDR when printing paddr_t values, instead of %lx - paddr_t/pd_entry_t being 64 bits with PAE) - remove casts that are no more needed now that Xen2 support has been dropped Some fixes are from jmorse@ patches for PAE. Compile + tested for i386 GENERIC and XEN3 kernels. Only compile tested for amd64. Reviewed by bouyer@. See also http://mail-index.netbsd.org/tech-kern/2010/02/22/msg007373.html
1.20	17-Nov-2009	dyoung	branches: 1.20.2; Fix spelling in comments, s/extention/extension/.
1.19	15-Apr-2009	cegger	No Execute bit is available in PAE-paging mode. However to use it, NXE bit in EFER register must be set.
1.18	21-Dec-2008	ad	branches: 1.18.2; Fix a comment.
1.17	23-Jan-2008	bouyer	branches: 1.17.6; 1.17.10; 1.17.18; Merge the bouyer-xeni386 branch. This brings in PAE support to NetBSD xeni386 (domU only). PAE support is enabled by 'options PAE', see the new XEN3PAE_DOMU and INSTALL_XEN3PAE_DOMU kernel config files. See the comments in arch/i386/include/{pte.h,pmap.h} to see how it works. In short, we still handle it as a 2-level MMU, with the second level page directory being 4 pages in size. pmap switching is done by switching the L2 pages in the L3 entries, instead of loading %cr3. This is almost required by Xen, which handle the last L2 page (the one mapping 0xc0000000 - 0xffffffff) in a very special way. But this approach should also work for native PAE support if ever supported (in fact, the pmap should almost suport native PAE, what's missing is bootstrap code in locore.S).
1.16	18-Oct-2007	yamt	branches: 1.16.2; 1.16.8; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.15	26-Dec-2005	perry	branches: 1.15.30; 1.15.48; 1.15.50; 1.15.52; 1.15.54; u_intN_t -> uintN_t
1.14	24-Aug-2003	chs	branches: 1.14.16; add support for non-executable mappings (where the hardware allows this) and make the stack and heap non-executable by default. the changes fall into two basic catagories: - pmap and trap-handler changes. these are all MD: = alpha: we already track per-page execute permission with the (software) PG_EXEC bit, so just have the trap handler pay attention to it. = i386: use a new GDT segment for %cs for processes that have no executable mappings above a certain threshold (currently the bottom of the stack). track per-page execute permission with the last unused PTE bit. = powerpc/ibm4xx: just use the hardware exec bit. = powerpc/oea: we already track per-page exec bits, but the hardware only implements non-exec mappings at the segment level. so track the number of executable mappings in each segment and turn on the no-exec segment bit iff the count is 0. adjust the trap handler to deal. = sparc (sun4m): fix our use of the hardware protection bits. fix the trap handler to recognize text faults. = sparc64: split the existing unified TSB into data and instruction TSBs, and only load TTEs into the appropriate TSB(s) for the permissions. fix the trap handler to check for execute permission. = not yet implemented: amd64, hppa, sh5 - changes in all the emulations that put a signal trampoline on the stack. instead, we now put the trampoline into a uvm_aobj and map that into the process separately. originally from openbsd, adapted for netbsd by me.
1.13	02-Apr-2003	thorpej	branches: 1.13.2; Use PAGE_SIZE rather than NBPG.
1.12	05-Sep-2000	thorpej	branches: 1.12.2; Define PG_LGFRAME, which is the page frame mask for large (4MB) pages.
1.11	06-Feb-1998	thorpej	branches: 1.11.14; Allow userland access to pd_entry_t and pt_entry_t, like the old pte.h used to do. Also, RCS ID police.
1.10	06-Feb-1998	mrg	add the i386 MD portions for UVM.
1.9	01-Feb-1996	mycroft	LOCORE -> _LOCORE
1.8	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	09-Oct-1994	mycroft	Nuke dirty().
1.5	15-Aug-1994	mycroft	Eliminate struct pte and struct pde.
1.4	14-Dec-1993	mycroft	branches: 1.4.2; 1.4.4; Fold in most of the magnum vm changes.
1.3	27-Jun-1993	andrew	branches: 1.3.4; Re-entrancy protection.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.3.4.5	13-Dec-1993	mycroft	Stylistic changes.
1.3.4.4	09-Nov-1993	mycroft	Rearrange bits to match hardware.
1.3.4.3	09-Nov-1993	mycroft	Change comment on PG_W to reflect that it is in fact used.
1.3.4.2	08-Nov-1993	mycroft	Garbage collect some constants, and the entire 286 nonsense. (Like we even get to start() on a 286? NOT!)
1.3.4.1	15-Oct-1993	mycroft	Major cleanup of include files and constants.
1.4.4.2	11-Oct-1994	mycroft	Update from trunk.
1.4.4.1	15-Aug-1994	mycroft	update from trunk
1.4.2.2	14-Dec-1993	mycroft	Fold in most of the magnum vm changes.
1.4.2.1	14-Dec-1993	mycroft	file pte.h was added on branch magnum on 1993-12-14 05:31:41 +0000
1.11.14.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.12.2.2	05-Sep-2000	thorpej	Define PG_LGFRAME, which is the page frame mask for large (4MB) pages.
1.12.2.1	05-Sep-2000	thorpej	file pte.h was added on branch sommerfeld_i386mp_1 on 2000-09-05 21:52:17 +0000
1.13.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.13.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.13.2.1	03-Aug-2004	skrll	Sync with HEAD
1.14.16.3	04-Feb-2008	yamt	sync with head.
1.14.16.2	27-Oct-2007	yamt	sync with head.
1.14.16.1	21-Jun-2006	yamt	sync with head.
1.15.54.1	25-Oct-2007	bouyer	Sync with HEAD.
1.15.52.1	23-Sep-2007	yamt	- sync with amd64. - pmap_alloc_level: remove fast call special case. - tweak some printf to work for both of LP64 and LP32. - remove machdep.nkpde sysctl. - remove automatic adjustment of nkpde for now. will revisit later.
1.15.50.2	23-Mar-2008	matt	sync with HEAD
1.15.50.1	06-Nov-2007	matt	sync with HEAD
1.15.48.1	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.15.30.1	23-Oct-2007	ad	Sync with head.
1.16.8.3	17-Jan-2008	bouyer	- Fix L2_SLOT_APTE value (not sure how I got this value but it was definitively wrong) - Use global variable for the PAE L3 page adresses, so that pmap.c can get it from the bootstrap code - Extent the size of our virtual PDP from 3 to 4 pages, so that pmap->pm_pdir[] is contigous for the whole VA range. The last page is a shadow of the kernel's real PDP (L3[3]). - make pm_pdirpa an array of 4 paddr_t if using PAE. introduce a pmap_pdirpa macro to get the physical address of a given PD entry. - fix pmap_map_pte The kernel now boots single-user. fsck will cause a kernel fault in pmap_pdes_invalid() on exit.
1.16.8.2	13-Jan-2008	bouyer	Update system macros for PAE. Because the L3 page directory has only 4 entries, and because of the way Xen handles it, it's better to have it has static as possible. So we claim to have only 2 levels, with level 2 being a "super" PD of 4 pages. pmap switches will be handled by changing the L3 entries instead of %cr3. Only the first 3 entries will have to be changed, the last one mapping KVA space only.
1.16.8.1	13-Jan-2008	bouyer	Work in progress on xeni386 PAE support: Make xeni386 build with a 64bit paddr_t. For this vaddr_t vs paddr_t vs pointers usages had to be clarified. If 'options PAE' is present in a Xen3 kernel, switch paddr_t, pd_entry_t and pt_entry_t to 64bits, and add the PAE entry in the __xen_guest ELF section.
1.16.2.1	18-Feb-2008	mjf	Sync with HEAD.
1.17.18.2	28-Apr-2009	skrll	Sync with HEAD.
1.17.18.1	19-Jan-2009	skrll	Sync with HEAD.
1.17.10.4	09-Oct-2010	yamt	sync with head
1.17.10.3	11-Aug-2010	yamt	sync with head.
1.17.10.2	11-Mar-2010	yamt	sync with head
1.17.10.1	04-May-2009	yamt	sync with head.
1.17.6.1	17-Jan-2009	mjf	Sync with HEAD.
1.18.2.4	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.18.2.3	28-Mar-2011	jym	Sync with HEAD. TODO before merge: - shortcut for suspend code in sysmon, when powerd(8) is not running. Borrow ``xs_watch'' thread context? - bug hunting in xbd + xennet resume. Rings are currently thrashed upon resume, so current implementation force flush them on suspend. It's not really needed.
1.18.2.2	01-Nov-2009	jym	Sync with HEAD.
1.18.2.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.20.2.3	22-Oct-2010	uebayasi	Sync with HEAD (-D20101022).
1.20.2.2	17-Aug-2010	uebayasi	Sync with HEAD.
1.20.2.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.21.2.2	05-Mar-2011	rmind	sync with head
1.21.2.1	30-May-2010	rmind	sync with head
1.26.4.1	08-Feb-2011	bouyer	Sync with HEAD
1.26.2.1	06-Jun-2011	jruoho	Sync with HEAD.
1.27.56.1	10-Jun-2019	christos	Sync with HEAD
1.31.10.1	25-Apr-2020	bouyer	Sync with bouyer-xenpvh-base2 (HEAD)
1.28	12-Aug-2025	gutteridge	i386/include/ptrace.h: move __HAVE_PROCFS_MACHDEP under _KERNEL For consistency and readability, guard with _KERNEL and move to be next to the PROCFS_MACHDEP_NODE_TYPES definition, as is done already with other ports like powerpc. As discussed in PR port-i386/59402. NFCI.
1.27	20-Nov-2023	simonb	Note some large xstate stack objects what Somebody(tm) should look at when they find some round tuits.
1.26	30-May-2020	maxv	Introduce PTRACE_REGS_ALIGN, and on x86, enforce a 16-byte alignment, due to fpregs having fxsave which requires 16-byte alignment. Reported-by: syzbot+f44d47e617ebf7fda081@syzkaller.appspotmail.com
1.25	09-Jan-2020	kamil	Add missing compat define of PT32_GETXSTATE Fixes i386 build. Reviewed by <mgorny>
1.24	08-Jan-2020	mgorny	Include XSTATE note in x86 core dumps Introduce a simple COREDUMP_MACHDEP_LWP_NOTES logic to provide machdep API for injecting per-LWP notes into coredumps, and use it to append PT_GETXSTATE note. Since the XSTATE block uses the same format on i386 and amd64, the code does not have to conditionalize between 32-bit and 64-bit ELF format on that. However, it does need to distinguish between 32-bit and 64-bit PT_* values. In order to do that, it reuses PT32_* constant already present for ptrace(), and adds a matching PT64_GETXSTATE to satisfy the cpp logic.
1.23	26-Jun-2019	mgorny	branches: 1.23.4; Implement PT_GETXSTATE and PT_SETXSTATE Introduce two new ptrace() requests: PT_GETXSTATE and PT_SETXSTATE, that provide access to the extended (and extensible) set of FPU registers on amd64 and i386. At the moment, this covers AVX (YMM) and AVX-512 (ZMM, opmask) registers. It can be easily extended to cover further register types without breaking backwards compatibility. PT_GETXSTATE issues the XSAVE instruction with all kernel-supported extended components enabled. The data is copied into 'struct xstate' (which -- unlike the XSAVE area itself -- has stable format and offsets). PT_SETXSTATE issues the XRSTOR instruction to restore the register values from user-provided 'struct xstate'. The function replaces only the specific XSAVE components that are listed in 'xs_rfbm' field, making it possible to issue partial updates. Both syscalls take a 'struct iovec' pointer rather than a direct argument. This requires the caller to explicitly specify the buffer size. As a result, existing code will continue to work correctly when the structure is extended (performing partial reads/updates).
1.22	18-Jun-2019	kamil	Introduce PTRACE_REG_FP() a helper macro to retrieve the frame pointer The macro is dummy for ia64 (the FP register is unknown and can change freely) and sparc/sparc64 (not stored in struct reg).
1.21	17-Jun-2019	kamil	Define PTRACE_ILLEGAL_ASM for i386
1.20	22-May-2018	kamil	branches: 1.20.2; Define PTRACE_BREAKPOINT_ASM for i386 in the MD part of <sys/ptrace.h> x86 uses int3 for software breakpoint on i386 and amd64. Sponsored by <The NetBSD Foundation>
1.19	08-Apr-2017	kamil	branches: 1.19.12; Define PT_SETSTEP and PT_CLEARSTEP for i386 (ptrace(2)) Sponsored by <The NetBSD Foundation>
1.18	23-Feb-2017	kamil	Introduce PT_GETDBREGS and PT_SETDBREGS in ptrace(2) on i386 and amd64 This interface is modeled after FreeBSD API with the usage. This replaced previous watchpoint API. The previous one was introduced recently in NetBSD-current and remove its spurs without any backward-compatibility. Design choices for Debug Register accessors: - exec() (TRAP_EXEC event) must remove debug registers from LWP - debug registers are only per-LWP, not per-process globally - debug registers must not be inherited after (v)forking a process - debug registers must not be inherited after forking a thread - a debugger is responsible to set global watchpoints/breakpoints with the debug registers, to achieve this PTRACE_LWP_CREATE/PTRACE_LWP_EXIT event monitoring function is designed to be used - debug register traps must generate SIGTRAP with si_code TRAP_DBREG - debugger is responsible to retrieve debug register state to distinguish the exact debug register trap (DR6 is Status Register on x86) - kernel must not remove debug register traps after triggering a trap event a debugger is responsible to detach this trap with appropriate PT_SETDBREGS call (DR7 is Control Register on x86) - debug registers must not be exposed in mcontext - userland must not be allowed to set a trap on the kernel Implementation notes on i386 and amd64: - the initial state of debug register is retrieved on boot and this value is stored in a local copy (initdbregs), this value is used to initialize dbreg context after PT_GETDBREGS - struct dbregs is stored in pcb as a pointer and by default not initialized - reserved registers (DR4-DR5, DR9-DR15) are ignored Further ideas: - restrict this interface with securelevel Tested on real hardware i386 (Intel Pentium IV) and amd64 (Intel i7). This commit enables 390 debug register ATF tests in kernel/arch/x86. All tests are passing. This commit does not cover netbsd32 compat code. Currently other interface PT_GET_SIGINFO/PT_SET_SIGINFO is required in netbsd32 compat code in order to validate reliably PT_GETDBREGS/PT_SETDBREGS. This implementation does not cover FreeBSD specific defines in their <x86/reg.h>: DBREG_DR7_LOCAL_ENABLE, DBREG_DR7_GLOBAL_ENABLE, DBREG_DR7_LEN_1 etc. These values tend to be reinvented by each tracer on its own. GNU Debugger (GDB) works with NetBSD debug registers after adding this patch: --- gdb/amd64bsd-nat.c.orig 2016-02-10 03:19:39.000000000 +0000 +++ gdb/amd64bsd-nat.c @@ -167,6 +167,10 @@ amd64bsd_target (void) #ifdef HAVE_PT_GETDBREGS +#ifndef DBREG_DRX +#define DBREG_DRX(d,x) ((d)->dr[(x)]) +#endif + static unsigned long amd64bsd_dr_get (ptid_t ptid, int regnum) { Another reason to stop introducing unpopular defines covering machine specific register macros is that these value varies across generations of the same CPU family. GDB demo: (gdb) c Continuing. Watchpoint 2: traceme Old value = 0 New value = 16 main (argc=1, argv=0x7f7fff79fe30) at test.c:8 8 printf("traceme=%d\n", traceme); (Currently the GDB interface is not reliable due to NetBSD support bugs) Sponsored by <The NetBSD Foundation>
1.17	16-Jan-2017	kamil	Refactor ptrace_watchpoint structure to allow extensions Add new field pw_type in the ptrace_watchpoint structure. amd64 and i386 offer the current set of watchpoints as PTRACE_PW_TYPE_DBREGS. On other archs than x86, there are readily available different types of hardware assisted watchpoints like for code-only or data-only registers on ARM. Also in future there is an option to implement MMU-based watchpoints and future per-port or per-cpu extensions. Next step is to alter this interface on x86 to generate SIGTRAP with si_code TRAP_HWWTRAP with additional information on occurred event: - which watchpoint fired, - additional watchpoint-type specific information, like on amd64 with PTRACE_PW_TYPE_DBREGS.: * only watchpoint fired * watchpoint fired and single step occurred Adjust ATF tests for the pw_type change. Sponsored by <The NetBSD Foundation>
1.16	15-Dec-2016	kamil	branches: 1.16.2; Add support for hardware assisted watchpoints/breakpoints API in ptrace(2) Add new ptrace(2) calls: - PT_COUNT_WATCHPOINTS - count the number of available hardware watchpoints - PT_READ_WATCHPOINT - read struct ptrace_watchpoint from the kernel state - PT_WRITE_WATCHPOINT - write new struct ptrace_watchpoint state, this includes enabling and disabling watchpoints The ptrace_watchpoint structure contains MI and MD parts: typedef struct ptrace_watchpoint { int pw_index; /* HW Watchpoint ID (count from 0) / lwpid_t pw_lwpid; / LWP described / struct mdpw pw_md; / MD fields / } ptrace_watchpoint_t; For example amd64 defines MD as follows: struct mdpw { void md_address; int md_condition; int md_length; }; These calls are protected with the __HAVE_PTRACE_WATCHPOINTS guard. Tested on amd64, initial support added for i386 and XEN. Sponsored by <The NetBSD Foundation>
1.15	25-Sep-2015	christos	branches: 1.15.2; For processors that have memory breakpoints, add macros for them to help libproc
1.14	15-Sep-2015	christos	Provide access to pc/sp/syscall-return registers like we have for mcontext
1.13	05-Mar-2006	christos	branches: 1.13.2; 1.13.116; 1.13.136; provide strings for the machdep ptrace calls.
1.12	11-Dec-2005	christos	branches: 1.12.4; 1.12.6; merge ktrace-lwp.
1.11	29-Jun-2003	fvdl	branches: 1.11.2; 1.11.18; Back out the lwp/ktrace changes. They contained a lot of colateral damage, and need to be examined and discussed more.
1.10	28-Jun-2003	darrenr	Pass lwp pointers throughtout the kernel, as required, so that the lwpid can be inserted into ktrace records. The general change has been to replace "struct proc " with "struct lwp " in various function prototypes, pass the lwp through and use l_proc to get the process pointer when needed. Bump the kernel rev up to 1.6V
1.9	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.8	09-May-2002	thorpej	branches: 1.8.4; Mirror cleanup wrt. procfs-centric naming made to the MI ptrace code.
1.7	05-Dec-2001	thorpej	* Allow machine-dependent code to specify hooks for ptrace(2) (__HAVE_PTRACE_MACHDEP) and procfs (__HAVE_PROCFS_MACHDEP). These changes will allow platforms like x86 (XMM) and PowerPC (AltiVec) to export extended register sets in a sane manner. * Use __HAVE_PTRACE_MACHDEP to export x86 XMM registers (standard FP + SSE/SSE2) using PT_{GET,SET}XMMREGS (in the machdep ptrace request space). * Use __HAVE_PROCFS_MACHDEP to export x86 XMM registers via /proc/N/xmmregs in procfs.
1.6	06-Aug-1995	mycroft	branches: 1.6.46; 1.6.48; Move the `used fpu' flag into mdproc, so it can be referenced when a process is swapped out. Implement process_{read,write}_fpregs.
1.5	27-Oct-1994	cgd	new RCS ID format.
1.4	28-Jan-1994	jtc	branches: 1.4.2; Fix spelling error in Copyright notice
1.3	28-Jan-1994	cgd	getting and setting FP regs aren't appropriate yet
1.2	09-Jan-1994	cgd	define PT_STEP, and clean up some
1.1	08-Jan-1994	cgd	code reorg
1.4.2.2	28-Jan-1994	jtc	Fix spelling error in Copyright notice
1.4.2.1	28-Jan-1994	jtc	file ptrace.h was added on branch magnum on 1994-01-28 23:44:22 +0000
1.6.48.2	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.6.48.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.6.46.5	15-Oct-2002	nathanw	Make _validfoo() routines go back to taking a proc.
1.6.46.4	21-Jun-2002	gmcgarry	LWPify
1.6.46.3	20-Jun-2002	nathanw	Catch up to -current.
1.6.46.2	09-Jan-2002	nathanw	Adapt xmm handling to LWPs.
1.6.46.1	08-Jan-2002	nathanw	Catch up to -current.
1.8.4.2	09-May-2002	thorpej	Mirror cleanup wrt. procfs-centric naming made to the MI ptrace code.
1.8.4.1	09-May-2002	thorpej	file ptrace.h was added on branch sommerfeld_i386mp_1 on 2002-05-09 16:28:13 +0000
1.11.18.1	21-Jun-2006	yamt	sync with head.
1.11.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.11.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.11.2.1	02-Jul-2003	darrenr	Apply the aborted ktrace-lwp changes to a specific branch. This is just for others to review, I'm concerned that patch fuziness may have resulted in some errant code being generated but I'll look at that later by comparing the diff from the base to the branch with the file I attempt to apply to it. This will, at the very least, put the changes in a better context for others to review them and attempt to tinker with removing passing of 'struct lwp' through the kernel.
1.12.6.1	22-Apr-2006	simonb	Sync with head.
1.12.4.1	09-Sep-2006	rpaulo	sync with head
1.13.136.4	28-Aug-2017	skrll	Sync with HEAD
1.13.136.3	05-Feb-2017	skrll	Sync with HEAD
1.13.136.2	27-Dec-2015	skrll	Sync with HEAD (as of 26th Dec)
1.13.136.1	22-Sep-2015	skrll	Sync with HEAD
1.13.116.1	03-Dec-2017	jdolecek	update from HEAD
1.13.2.2	05-Mar-2006	christos	provide strings for the machdep ptrace calls.
1.13.2.1	05-Mar-2006	christos	file ptrace.h was added on branch yamt-pdpolicy on 2006-03-05 07:17:22 +0000
1.15.2.3	26-Apr-2017	pgoyette	Sync with HEAD
1.15.2.2	20-Mar-2017	pgoyette	Sync with HEAD
1.15.2.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.16.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.19.12.1	25-Jun-2018	pgoyette	Sync with HEAD
1.20.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.20.2.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.23.4.1	17-Jan-2020	ad	Sync with head.
1.9	01-Mar-2010	jym	Change rbus_min_start_hint() semantic for i386. "ram" is now psize_t (instead of size_t) to avoid possible overflow when system may have more than 4GB of memory (like PAE). The behavior of rbus_min_start_hint() remains the same. While here, fix printf's format strings (paddr_t => PRIxPADDR). Use ctob() and cast physmem to psize_t to avoid losing bits above 4GB. Comes from PAE patch from Jeremy Morse; adaptation by me. Compile tested for GENERIC only. No regression expected.
1.8	15-Dec-2009	snj	branches: 1.8.2; Move to 2-clause license. Approved by HAYAKAWA Koichi (copyright holder).
1.7	12-Jan-2007	gdt	branches: 1.7.48; 1.7.62; Add a heuristic to set rbus_min_start based on total RAM. The new behavior is to choose 0.5 GB for <= 192 MB, 1 GB normally, and 2 GB for >= 1 GB. This should make the defaults work additionally old Thinkpad 600Es, and also on notebooks with lots of RAM (e.g. T60 with 2GB). ok christos@
1.6	26-Feb-2003	fvdl	branches: 1.6.18; 1.6.48; 1.6.54; Redirect some includes to x86/foo.h
1.5	25-Apr-2001	thorpej	branches: 1.5.2; Move _i386_memio_unmap() into bus_machdep.c; the ACPI code wants it.
1.4	06-Jun-2000	haya	branches: 1.4.6; Remove unnecessary function arguments.
1.3	01-Jun-2000	cgd	kill __P in these files. (I had to look at them anyway with an eye for adding some protos... and adding them with __P seems wrong, but mixing __P and not __P in the same file seems wrong too, so...)
1.2	15-Oct-1999	haya	branches: 1.2.2; 1.2.8; 1.2.12; add NetBSD ID tag.
1.1	15-Oct-1999	haya	This is the first check-in of CardBus driver. CardBus driver contains CardBus bus stub, YENTA PCI-CardBus bridge (cbb), 3Com 3C575TX driver (ex) and Intel fxp driver. TODO: o Conform to the KNF more strictly. o Be unified with pcmcia code as much as possible. o Add more drivers for CardBus card, such as APA-1480 or USB card. The affected files are listed below. sys/arch/i386/conf/files.i386 sys/arch/macppc/conf/files.macppc sys/conf/files sys/dev/ic/elinkxl.c sys/dev/ic/elinkxlvar.h sys/dev/ic/i82365.c sys/dev/ic/i82365var.h sys/dev/isa/i82365_isasubr.c sys/dev/pci/files.pci sys/dev/pcmcia/pcmcia.c sys/dev/pcmcia/pcmciachip.h The added files are listed below. sys/arch/i386/conf/CARDBUS sys/arch/i386/include/rbus_machdep.h sys/arch/i386/i386/rbus_machdep.c sys/arch/macppc/include/rbus_machdep.h sys/arch/macppc/macppc/rbus_machdep.c sys/dev/cardbus/if_ex_cardbus.c sys/dev/cardbus/Makefile.cardbusdevs sys/dev/cardbus/cardbus.c sys/dev/cardbus/cardbus_map.c sys/dev/cardbus/cardbusdevs sys/dev/cardbus/cardbusdevs.h sys/dev/cardbus/cardbusdevs_data.h sys/dev/cardbus/cardbusvar.h sys/dev/cardbus/cardslot.c sys/dev/cardbus/cardslotvar.h sys/dev/cardbus/devlist2h.awk sys/dev/cardbus/files.cardbus sys/dev/cardbus/if_fxp_cardbus.c sys/dev/cardbus/pccardcis.h sys/dev/cardbus/rbus.c sys/dev/cardbus/rbus.h sys/dev/pci/pccbb.c sys/dev/pci/pccbbreg.h sys/dev/pci/pccbbvar.h
1.2.12.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.2.8.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.2.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.4.6.1	21-Jun-2001	nathanw	Catch up to -current.
1.5.2.2	25-Apr-2001	thorpej	Move _i386_memio_unmap() into bus_machdep.c; the ACPI code wants it.
1.5.2.1	25-Apr-2001	thorpej	file rbus_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-04-25 03:31:24 +0000
1.6.54.1	28-Mar-2007	jdc	Pull up revision 1.7 (requested by gdt in ticket #521) Add a heuristic to set rbus_min_start based on total RAM. The new behavior is to choose 0.5 GB for <= 192 MB, 1 GB normally, and 2 GB for >= 1 GB. This should make the defaults work additionally old Thinkpad 600Es, and also on notebooks with lots of RAM (e.g. T60 with 2GB). ok christos@
1.6.48.1	01-Feb-2007	ad	Sync with head.
1.6.18.1	26-Feb-2007	yamt	sync with head.
1.7.62.1	24-Oct-2010	jym	Sync with HEAD
1.7.48.1	11-Mar-2010	yamt	sync with head
1.8.2.1	30-Apr-2010	uebayasi	Sync with HEAD.
1.22	18-May-2019	christos	lint gets the wrong size.
1.21	10-May-2019	mgorny	Fill 'struct fpreg' & 'struct xmmregs' in The 'struct fpreg' data type used by PT_GETFPREGS, and 'struct xmmregs' used by PT_GETXMMREGS are currently opaque. Define them to contain correct data structs instead, the same way they are defined for amd64. For 'struct fpreg' this means 'struct save87', and for 'struct xmmregs' this means 'struct fxsave'. This makes it more transparent for consumers how the data is formatted, and allows using it without need for explicit casts. Reviewed by <kamil>.
1.20	23-Feb-2017	kamil	branches: 1.20.14; Introduce PT_GETDBREGS and PT_SETDBREGS in ptrace(2) on i386 and amd64 This interface is modeled after FreeBSD API with the usage. This replaced previous watchpoint API. The previous one was introduced recently in NetBSD-current and remove its spurs without any backward-compatibility. Design choices for Debug Register accessors: - exec() (TRAP_EXEC event) must remove debug registers from LWP - debug registers are only per-LWP, not per-process globally - debug registers must not be inherited after (v)forking a process - debug registers must not be inherited after forking a thread - a debugger is responsible to set global watchpoints/breakpoints with the debug registers, to achieve this PTRACE_LWP_CREATE/PTRACE_LWP_EXIT event monitoring function is designed to be used - debug register traps must generate SIGTRAP with si_code TRAP_DBREG - debugger is responsible to retrieve debug register state to distinguish the exact debug register trap (DR6 is Status Register on x86) - kernel must not remove debug register traps after triggering a trap event a debugger is responsible to detach this trap with appropriate PT_SETDBREGS call (DR7 is Control Register on x86) - debug registers must not be exposed in mcontext - userland must not be allowed to set a trap on the kernel Implementation notes on i386 and amd64: - the initial state of debug register is retrieved on boot and this value is stored in a local copy (initdbregs), this value is used to initialize dbreg context after PT_GETDBREGS - struct dbregs is stored in pcb as a pointer and by default not initialized - reserved registers (DR4-DR5, DR9-DR15) are ignored Further ideas: - restrict this interface with securelevel Tested on real hardware i386 (Intel Pentium IV) and amd64 (Intel i7). This commit enables 390 debug register ATF tests in kernel/arch/x86. All tests are passing. This commit does not cover netbsd32 compat code. Currently other interface PT_GET_SIGINFO/PT_SET_SIGINFO is required in netbsd32 compat code in order to validate reliably PT_GETDBREGS/PT_SETDBREGS. This implementation does not cover FreeBSD specific defines in their <x86/reg.h>: DBREG_DR7_LOCAL_ENABLE, DBREG_DR7_GLOBAL_ENABLE, DBREG_DR7_LEN_1 etc. These values tend to be reinvented by each tracer on its own. GNU Debugger (GDB) works with NetBSD debug registers after adding this patch: --- gdb/amd64bsd-nat.c.orig 2016-02-10 03:19:39.000000000 +0000 +++ gdb/amd64bsd-nat.c @@ -167,6 +167,10 @@ amd64bsd_target (void) #ifdef HAVE_PT_GETDBREGS +#ifndef DBREG_DRX +#define DBREG_DRX(d,x) ((d)->dr[(x)]) +#endif + static unsigned long amd64bsd_dr_get (ptid_t ptid, int regnum) { Another reason to stop introducing unpopular defines covering machine specific register macros is that these value varies across generations of the same CPU family. GDB demo: (gdb) c Continuing. Watchpoint 2: traceme Old value = 0 New value = 16 main (argc=1, argv=0x7f7fff79fe30) at test.c:8 8 printf("traceme=%d\n", traceme); (Currently the GDB interface is not reliable due to NetBSD support bugs) Sponsored by <The NetBSD Foundation>
1.19	16-Jan-2008	ad	branches: 1.19.54; 1.19.74; 1.19.78; 1.19.82; Remove options MATH_EMULATE.
1.18	07-Aug-2003	agc	branches: 1.18.16; 1.18.78; 1.18.84; 1.18.92; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.17	05-Dec-2001	thorpej	branches: 1.17.2; 1.17.18; * Allow machine-dependent code to specify hooks for ptrace(2) (__HAVE_PTRACE_MACHDEP) and procfs (__HAVE_PROCFS_MACHDEP). These changes will allow platforms like x86 (XMM) and PowerPC (AltiVec) to export extended register sets in a sane manner. * Use __HAVE_PTRACE_MACHDEP to export x86 XMM registers (standard FP + SSE/SSE2) using PT_{GET,SET}XMMREGS (in the machdep ptrace request space). * Use __HAVE_PROCFS_MACHDEP to export x86 XMM registers via /proc/N/xmmregs in procfs.
1.16	17-Jul-2001	fvdl	Redefine evil tXXX register offset defines into the trapframe structure to use offsetof. These should be completely nuked. Fixes math_emulate lossage.
1.15	16-Oct-1997	mycroft	branches: 1.15.28; 1.15.30; 1.15.32; Update a comment dating from 4.3/VAX.
1.14	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.13	06-Aug-1995	mycroft	Move the `used fpu' flag into mdproc, so it can be referenced when a process is swapped out. Implement process_{read,write}_fpregs.
1.12	26-Jan-1995	mycroft	Rearrange struct reg to match what the rest of the world expects.
1.11	27-Oct-1994	cgd	new RCS ID format.
1.10	09-Oct-1994	mycroft	Make reg, frame, and sigcontext more similar.
1.9	09-Oct-1994	mycroft	Eliminate tf_isp.
1.8	22-Dec-1993	cgd	branches: 1.8.2; 1.8.4; should be 'struct reg' not 'struct regs' -- it's in reg.h, after all! also, delete r_fs and r_gs; looking at the code and comments describing/using this struct, they don't belong here (though if support for them is ever put in here, they should be, but for i486-en only... how do you do that 'reasonably,' though? interface issues...)
1.7	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.6	19-Sep-1993	brezak	Add tISP offset
1.5	16-Sep-1993	brezak	Changes to use a trap frame for syscalls.
1.4	05-Sep-1993	sef	branches: 1.4.2; Yet more of the ptrace() reorg; now ptrace_setregs() and ptrace_getregs() are present, along with PT_GETREGS and PT_SETREGS ptrace commands.
1.3	24-May-1993	cgd	move sipcreg into i386/include/reg.h where it belongs, right next to ipcreg.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.4.2.3	14-Nov-1993	mycroft	Eliminate tEDI, tESI, etc.
1.4.2.2	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.4.2.1	24-Sep-1993	mycroft	Changes from trunk. cpu.h: Add dummy (for now) CLKF_INTR(). #include psl.h. endian.h: Add multiple-inclusion protection. #define _QUAD_{LOW,HIGH}WORD. #include sys/cdefs for __P(). Don't define byte order stuff if _POSIX_SOURCE. Remove big endian shit. limits.h: Add {,U}QUAD_{MIN,MAX}. psl.h: Get PSL_MB[ZO] right. Make cpl volatile just in case. Add alias for spl0(). splnone() and splx() return int for now.
1.8.4.1	11-Oct-1994	mycroft	Update from trunk.
1.8.2.2	22-Dec-1993	cgd	should be 'struct reg' not 'struct regs' -- it's in reg.h, after all! also, delete r_fs and r_gs; looking at the code and comments describing/using this struct, they don't belong here (though if support for them is ever put in here, they should be, but for i486-en only... how do you do that 'reasonably,' though? interface issues...)
1.8.2.1	22-Dec-1993	cgd	file reg.h was added on branch magnum on 1993-12-22 12:57:25 +0000
1.15.32.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.15.32.1	03-Aug-2001	lukem	update to -current
1.15.30.2	08-Jan-2002	nathanw	Catch up to -current.
1.15.30.1	24-Aug-2001	nathanw	Catch up with -current.
1.15.28.1	29-Jul-2001	he	Pull up revision 1.16 (requested by fvdl): Redefine register offsets using offsetof(). Fixes math emulation lossage.
1.17.18.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.17.18.2	18-Sep-2004	skrll	Sync with HEAD.
1.17.18.1	03-Aug-2004	skrll	Sync with HEAD
1.17.2.2	05-Dec-2001	thorpej	* Allow machine-dependent code to specify hooks for ptrace(2) (__HAVE_PTRACE_MACHDEP) and procfs (__HAVE_PROCFS_MACHDEP). These changes will allow platforms like x86 (XMM) and PowerPC (AltiVec) to export extended register sets in a sane manner. * Use __HAVE_PTRACE_MACHDEP to export x86 XMM registers (standard FP + SSE/SSE2) using PT_{GET,SET}XMMREGS (in the machdep ptrace request space). * Use __HAVE_PROCFS_MACHDEP to export x86 XMM registers via /proc/N/xmmregs in procfs.
1.17.2.1	05-Dec-2001	thorpej	file reg.h was added on branch sommerfeld_i386mp_1 on 2001-12-05 00:58:07 +0000
1.18.92.1	19-Jan-2008	bouyer	Sync with HEAD
1.18.84.1	18-Feb-2008	mjf	Sync with HEAD.
1.18.78.1	23-Mar-2008	matt	sync with HEAD
1.18.16.1	21-Jan-2008	yamt	sync with head
1.19.82.1	21-Apr-2017	bouyer	Sync with HEAD
1.19.78.1	20-Mar-2017	pgoyette	Sync with HEAD
1.19.74.1	28-Aug-2017	skrll	Sync with HEAD
1.19.54.1	03-Dec-2017	jdolecek	update from HEAD
1.20.14.1	10-Jun-2019	christos	Sync with HEAD
1.1	03-Nov-2009	dyoung	branches: 1.1.4; 1.1.8; Add return_address(9) for reading the Nth return address from the call stack.
1.1.8.2	24-Oct-2010	jym	Sync with HEAD
1.1.8.1	03-Nov-2009	jym	file return.h was added on branch jym-xensuspend on 2010-10-24 22:48:02 +0000
1.1.4.2	11-Mar-2010	yamt	sync with head
1.1.4.1	03-Nov-2009	yamt	file return.h was added on branch yamt-nfs-mp on 2010-03-11 15:02:29 +0000
1.5	05-Feb-2003	nakayama	Replace machine/rnd.h with more appropriate name to share it with cycle counter based microtime in kern/kern_microtime.c.
1.4	06-Dec-2001	sommerfeld	branches: 1.4.2; Use #if defined(FOO_CPU) instead of #if FOO_CPU just to be picky. Spotted by Thomas Klausner.
1.3	02-Jul-2000	sommerfeld	branches: 1.3.4; 1.3.6; 1.3.8; Return the full width of the counter here.
1.2	06-Jun-2000	sommerfeld	branches: 1.2.2; 1.2.4; Replace "timestamp" with "counter" in previous change, to avoid any possible confusion that it has anything to do with accurate measurement of time. New names: __HAVE_CPU_COUNTER cpu_counter() cpu_hascounter()
1.1	05-Jun-2000	sommerfeld	Let rnd(4) know how to read the (pentium and higher) cycle counter: If __HAVE_CPU_TIMESTAMP is defined, <machine/rnd.h> exists and defines two functions: 1) u_int32_t cpu_timestamp(void); returns the 32 low order bits of a reasonably high frequency counter. 2) int cpu_havetimestamp(void); returns non-zero if cpu_timestamp() actually works. The timestamp counter should run at a frequency greater than 1 MHz (otherwise, microtime would be sufficient); other than that, the exact frequency and origin of the counter are unspecified.
1.2.4.2	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.2.4.1	06-Jun-2000	minoura	file rnd.h was added on branch minoura-xpg4dl on 2000-06-22 17:00:34 +0000
1.2.2.1	02-Jul-2000	sommerfeld	Pull up fix from current: let cpu_counter() be wider than 32 bits on platforms for which is the case; clip to 32 bits in rnd.c Approved by thorpej
1.3.8.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.3.6.1	08-Jan-2002	nathanw	Catch up to -current.
1.3.4.2	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.3.4.1	02-Jul-2000	bouyer	file rnd.h was added on branch thorpej_scsipi on 2000-11-20 20:09:30 +0000
1.4.2.2	06-Dec-2001	sommerfeld	Use #if defined(FOO_CPU) instead of #if FOO_CPU just to be picky. Spotted by Thomas Klausner.
1.4.2.1	06-Dec-2001	sommerfeld	file rnd.h was added on branch sommerfeld_i386mp_1 on 2001-12-06 01:56:20 +0000
1.2	09-Feb-2007	ad	branches: 1.2.4; 1.2.144; Merge newlock2 to head.
1.1	10-Sep-2006	ad	branches: 1.1.2; file rwlock.h was initially added on branch newlock2.
1.1.2.3	29-Dec-2006	ad	Checkpoint work in progress.
1.1.2.2	24-Oct-2006	ad	Compile fixes
1.1.2.1	10-Sep-2006	ad	Add updated locking primatives.
1.2.144.2	29-Feb-2020	ad	Sync with head.
1.2.144.1	19-Jan-2020	ad	empty these; remove later.
1.2.4.2	26-Feb-2007	yamt	sync with head.
1.2.4.1	09-Feb-2007	yamt	file rwlock.h was added on branch yamt-lazymbuf on 2007-02-26 09:07:02 +0000
1.1	17-Mar-2002	thorpej	branches: 1.1.2; file rwlock_impl.h was initially added on branch newlock.
1.1.2.2	22-Mar-2002	thorpej	Add debug info structures if MUTEX_DEBUG or RWLOCK_DEBUG is defined. XXX Revisit, so that the structure doesn't change when debugging is turned on; do it out-of-band?
1.1.2.1	17-Mar-2002	thorpej	Add first-cut mutex and rwlock implementation bits for the x86. XXX Does not actually handle i386 CPUs yet, only i486 and later.
1.70	18-May-2022	andvar	fix typos in word "access" and it derivatives, mainly in comments.
1.69	14-Jul-2020	yamaguchi	Introduce per-cpu IDTs This is realized by following modifications: - Add IDT pages and its allocation maps for each cpu in "struct cpu_info" - Load per-cpu IDTs at cpu_init_idt(struct cpu_info*) - Copy the IDT entries for cpu0 to other CPUs at attach - These are, for example, exceptions, db, system calls, etc. And, added a kernel option named PCPU_IDT to enable the feature.
1.68	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.67	23-Sep-2018	cherry	Fix for i386, functionality intended in: http://mail-index.netbsd.org/source-changes/2018/09/23/msg099357.html This should fix the build for both GENERIC and XEN3PAE_DOM0 This has not been boot tested on native or xen3pae Notes: pmap_changeprot_local() seems to be x86_64 only. I was a bit surprised by this initially, but I suspect that the table protections are enforced via ring0/ring1 fencing rather than page protections the gdt registration code in i386 is still messy. I will leave it as is for now - to avoid a rabbit hole.
1.66	23-Sep-2018	cherry	Summary: Remove older and now conflicting type declaration for global variable 'idt'. This should fix the i386 build failure now.
1.65	23-Sep-2018	cherry	Make XEN use the same api as native, for idt vector allocation and registration. lidt() placed in xenfunc() on maxv@ suggestion. There should be no functional change due to this commit. Tested on amd64 native and XEN.
1.64	31-Dec-2017	maxv	branches: 1.64.2; 1.64.4; Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.63	04-Nov-2017	cherry	In XEN PV, the idt vector table is not required to be altered at runtime, since only entries for exceptions/traps are registered with the hypervisor and interrupts are managed via a completely different mechanism. This change uses the idt_vec_reserve() mechanism nevertheless, modifying it slightly to only do namespace management in XEN, while on native it will continue to do idt entry init as before. Rationale: Consistent API usage and potential future merging of XEN/non-XEN code. There are no functional changes in this commit.
1.62	17-Sep-2017	maxv	Remove the second argument from USERMODE and KERNELMODE, it is unused now that we don't have vm86 anymore.
1.61	12-Aug-2017	maxv	Remove vm86. Pass 4.
1.60	02-Jul-2017	maxv	Use a bitmap-based allocator for i386, same as amd64. Several functions are now identical - or nearly identical - on both sides. I couldn't test this change on xen, because I'm having some unrelated issues with my VM and I've spent enough time not understanding what's wrong with it.
1.59	08-Feb-2017	maxv	branches: 1.59.6; Localify, add a comment and merge some others.
1.58	05-Feb-2017	maxv	Rename ldt->ldtstore and gdt->gdtstore on i386. It reduces the diff with amd64, and makes it easier to track down these variables on nxr - 'ldt' and 'gdt' being common keywords.
1.57	02-Sep-2016	maxv	branches: 1.57.2; KNF, and give the structure sizes.
1.56	27-Aug-2016	maxv	Remove idt_init.
1.55	24-Jan-2016	christos	branches: 1.55.2; expose __vaddr_t for segments.h since it is a userland api.
1.54	26-Apr-2011	joerg	branches: 1.54.10; 1.54.14; 1.54.16; 1.54.24; 1.54.30; 1.54.32; 1.54.34; 1.54.38; Remove Darwin, MACH and Mach-O support.
1.53	17-May-2009	bouyer	branches: 1.53.4; 1.53.6; on Xen the GDT has to be updated though HYPERVISOR_update_descriptor(). Export i386/i386/gdt.c:update_descriptor() and use it in x86_set_sdbase(), as a direct write to the GDT will cause a kernel trap. Fix PR port-xen/41401.
1.52	21-Mar-2009	ad	A couple more comments.
1.51	21-Mar-2009	ad	Add a comment.
1.50	26-Oct-2008	christos	branches: 1.50.2; 1.50.4; 1.50.8; use a macro to get to d_type.
1.49	19-Apr-2008	cegger	branches: 1.49.2; 1.49.8; idt_* are not implemented for Xen. So don't provide the prototypes for Xen.
1.48	11-Jan-2008	bouyer	branches: 1.48.6; 1.48.8; Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.47	04-Jan-2008	yamt	i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.46	26-Dec-2007	yamt	- share idt entry allocation code among x86. - introduce a function to reserve an idt entry and use it instead of manipulating idt_allocmap directly. - rename idt to xen_idt for amd64 xen. add missing #ifdef XEN.
1.45	25-Dec-2007	perry	Convert many of the uses of __attribute__ to equivalent __packed, __unused and __dead macros from cdefs.h
1.44	10-Nov-2007	ad	branches: 1.44.2; 1.44.6; - When computing the TSC frequency, call i8254_delay() and not DELAY(). - Use atomics to adjust the pmap reference count, instead of taking locks. - Implement I386_{SET,GET}_{FS,GS}BASE, allowing %fs and %gs to be used as per-thread registers. This is compatible with FreeBSD. - Run patches after we have attached CPUs, since we then know if the system is uniprocessor or not. Eliminates a lot of #ifdef MULTIPROCESSOR and makes running MP kernels on UP systems cheaper. - Patch out many of the 'lock' prefixes to nops if uniprocessor. - Do a wbinvd after patching to ensure that the trace/instruction cache is up to date.
1.43	18-Oct-2007	yamt	branches: 1.43.2; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.42	02-Mar-2007	ad	branches: 1.42.2; 1.42.18; 1.42.20; 1.42.22; 1.42.24; Shuffle the descriptor slots slightly to make the layout match what sysenter/sysexit require. As far as I can tell this will not introduce any incompatibility.
1.41	05-Mar-2004	junyoung	branches: 1.41.16; 1.41.54; Drop trailing spaces.
1.40	13-Feb-2004	wiz	Uppercase CPU, plural is CPUs.
1.39	27-Oct-2003	junyoung	Nuke __P().
1.38	24-Aug-2003	chs	add support for non-executable mappings (where the hardware allows this) and make the stack and heap non-executable by default. the changes fall into two basic catagories: - pmap and trap-handler changes. these are all MD: = alpha: we already track per-page execute permission with the (software) PG_EXEC bit, so just have the trap handler pay attention to it. = i386: use a new GDT segment for %cs for processes that have no executable mappings above a certain threshold (currently the bottom of the stack). track per-page execute permission with the last unused PTE bit. = powerpc/ibm4xx: just use the hardware exec bit. = powerpc/oea: we already track per-page exec bits, but the hardware only implements non-exec mappings at the segment level. so track the number of executable mappings in each segment and turn on the no-exec segment bit iff the count is 0. adjust the trap handler to deal. = sparc (sun4m): fix our use of the hardware protection bits. fix the trap handler to recognize text faults. = sparc64: split the existing unified TSB into data and instruction TSBs, and only load TTEs into the appropriate TSB(s) for the permissions. fix the trap handler to check for execute permission. = not yet implemented: amd64, hppa, sh5 - changes in all the emulations that put a signal trampoline on the stack. instead, we now put the trampoline into a uvm_aobj and map that into the process separately. originally from openbsd, adapted for netbsd by me.
1.37	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.36	03-Mar-2003	fvdl	branches: 1.36.2; The IDT only contains gate descriptors, so define it as such.
1.35	05-Oct-2002	fvdl	Change setgate() prototype. Define GDT entries for DDB and double fault TSS.
1.34	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.33	23-Jan-2002	rafal	Re-arrange GDT entries somewhat so that: (1) I can reserve the 8th entry for BIOS-bug-avoidance [fixes part of my own port-i386/11299], and (2) lets you have APM and COMPAT_MACHO in the kernel at the same time. Also comments on the special 8th entry and what else changing some of these values affects. OK'ed by Frank.
1.32	14-Jul-2001	christos	Use global descriptor 7 for mach traps. Unfortunately this is already used by apm 16 bit code segments so we cannot have both at the same time.
1.31	30-May-2001	mrg	branches: 1.31.2; use _KERNEL_OPT
1.30	12-Nov-1999	drochner	branches: 1.30.2; 1.30.8; add GDT segments for pnpbios use
1.29	03-Oct-1998	christos	branches: 1.29.12; 1.29.14; 1.29.18; Add an LDT for Solaris-2.6 syscalls [from Frank]
1.28	23-Jan-1998	mycroft	Some rearrangement of the startup code: * Move some of the memory bootstrapping and the DDB startup earlier. * Always put the IDT and the initial GDT and LDT in a separate page. * Various minor changes.
1.27	15-Jan-1998	thorpej	Generate dependencies on the VM86 option.
1.26	13-Nov-1997	mycroft	Some general cleanup: * Call gdt_init() earlier on. * Don't bother to check for TSS and LDT descriptors in verr_gdt(). * Nuke the sti special case for trace and breakpoint traps. * Move some variable declarations into more appropriate locations.
1.25	08-Sep-1996	jtk	branches: 1.25.14; change APM to use new almost-generalized BIOS trampoline: /* * void bioscall(int function, struct apmregs regs): call the BIOS interrupt "function" from real mode with * registers as specified in "regs" * (for the flags, though, only these flags are passed to the BIOS; * the remainder come from the flags register at the time of the call: * (PSL_C\|PSL_PF\|PSL_AF\|PSL_Z\|PSL_N\|PSL_D\|PSL_V) * * Fills in regs with registers as returned by BIOS. / Thanks to Charles Hannum for complaining and inspiring me to hack this together.
1.24	25-Aug-1996	jtk	add Advanced Power Management driver Includes bootstrap support from FreeBSD (Tatsumi HOSOKAWA and <ukai>)
1.23	01-Feb-1996	mycroft	LOCORE -> _LOCORE
1.22	20-Dec-1995	mycroft	Use #pragma only for 2.0 <= GCC < 2.7.
1.21	20-Dec-1995	mycroft	Use alternate structure packing method added in GCC 2.7.
1.20	12-Oct-1995	mycroft	branches: 1.20.2; Move the initialization of proc0's TSS and LDT to the end of cpu_startup(), and switch to dynamically allocating the TSS.
1.19	11-Oct-1995	mycroft	Go back to statically allocating the GDT entries for proc0.
1.18	11-Oct-1995	mycroft	Remove reference to dead variable.
1.17	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.16	10-Oct-1995	mycroft	Add USERMODE() and KERNELMODE() macros, and use them.
1.15	09-Oct-1995	mycroft	Eliminate soft segment descriptors; instead use a functional interface for creating segment descriptors. Add the duplicate system call gate used by BSD/OS 2.0 executables. Prototype setsegment() and setgate(), and fix a bogon caught by this.
1.14	06-May-1995	mycroft	Add a copy of the user code and data descriptors in the GDT.
1.13	04-May-1995	mycroft	Wrap NGDT and NLDT in parens.
1.12	01-May-1995	mycroft	Make this includable from locore. Remove unused GDT and LDT slots.
1.11	01-May-1995	mycroft	Expose idt[] here.
1.10	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.9	08-Nov-1994	mycroft	pcb.h requires segments.h now.
1.8	05-Nov-1994	mycroft	Fully prototype ssdtosd() and sdtossd().
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	03-Apr-1994	mycroft	Clean this up a little.
1.5	20-Dec-1993	mycroft	branches: 1.5.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.4	26-Oct-1993	brezak	Cleanup LDT/GDT defines
1.3	22-May-1993	cgd	branches: 1.3.4; add rcsids to everything and clean up headers
1.2	03-Apr-1993	cgd	added changes from Steven Reiz <sreiz@aie.nl> (based on those by Poul-Henning Kamp <phk@data.fls.dk>) to get the kernel to compile properly when gcc2.* is cc. (should still work when gcc1.39 is in use.)
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.3.4.1	26-Oct-1993	mycroft	Merge changes from trunk.
1.5.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.5.2.1	20-Dec-1993	mycroft	file segments.h was added on branch magnum on 1993-12-20 05:25:44 +0000
1.20.2.2	12-Oct-1995	mycroft	Move the initialization of proc0's TSS and LDT to the end of cpu_startup(), and switch to dynamically allocating the TSS.
1.20.2.1	12-Oct-1995	mycroft	file segments.h was added on branch netbsd-1-1 on 1995-10-12 17:57:02 +0000
1.25.14.2	17-Oct-1998	cgd	pull up rev 1.29 from trunk (christos)
1.25.14.1	13-Nov-1997	mellon	Pull rev 1.26 up from trunk (mycroft)
1.29.18.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.29.14.1	15-Nov-1999	fvdl	Sync with -current
1.29.12.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.30.8.4	18-Oct-2002	nathanw	Catch up to -current.
1.30.8.3	28-Feb-2002	nathanw	Catch up to -current.
1.30.8.2	24-Aug-2001	nathanw	Catch up with -current.
1.30.8.1	21-Jun-2001	nathanw	Catch up to -current.
1.30.2.5	28-Jan-2002	sommerfeld	Yet Another mergeup with -current.
1.30.2.4	29-Dec-2001	sommerfeld	Set up per-cpu GDT. Use %fs for per-cpu data access.
1.30.2.3	19-Jul-2001	sommerfeld	catch up with -current
1.30.2.2	18-Jun-2001	sommerfeld	checkpoint merge-in-progress from mainline.
1.30.2.1	20-Feb-2000	sommerfeld	Include functions for allocation and freeing of IDT slots.
1.31.2.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.31.2.2	11-Feb-2002	jdolecek	Sync w/ -current.
1.31.2.1	03-Aug-2001	lukem	update to -current
1.36.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.36.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.36.2.1	03-Aug-2004	skrll	Sync with HEAD
1.41.54.1	12-Mar-2007	rmind	Sync with HEAD.
1.41.16.4	21-Jan-2008	yamt	sync with head
1.41.16.3	15-Nov-2007	yamt	sync with head.
1.41.16.2	27-Oct-2007	yamt	sync with head.
1.41.16.1	03-Sep-2007	yamt	sync with head.
1.42.24.2	13-Nov-2007	bouyer	Sync with HEAD
1.42.24.1	25-Oct-2007	bouyer	Sync with HEAD.
1.42.22.1	18-Oct-2007	yamt	reduce #ifdef.
1.42.20.3	23-Mar-2008	matt	sync with HEAD
1.42.20.2	09-Jan-2008	matt	sync with HEAD
1.42.20.1	06-Nov-2007	matt	sync with HEAD
1.42.18.2	11-Nov-2007	joerg	Sync with HEAD.
1.42.18.1	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.42.2.2	03-Dec-2007	ad	Sync with HEAD.
1.42.2.1	23-Oct-2007	ad	Sync with head.
1.43.2.2	18-Feb-2008	mjf	Sync with HEAD.
1.43.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.44.6.4	09-Jan-2008	bouyer	Merge Xen support back to i386/i386/vector.S
1.44.6.3	08-Jan-2008	bouyer	Sync with HEAD
1.44.6.2	07-Jan-2008	bouyer	Move Xen support to i386/i386/machdep.c. two less files in xen/ :)
1.44.6.1	02-Jan-2008	bouyer	Sync with HEAD
1.44.2.1	26-Dec-2007	ad	Sync with head.
1.48.8.1	18-May-2008	yamt	sync with head.
1.48.6.2	17-Jan-2009	mjf	Sync with HEAD.
1.48.6.1	02-Jun-2008	mjf	Sync with HEAD.
1.49.8.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.49.2.2	20-Jun-2009	yamt	sync with head
1.49.2.1	04-May-2009	yamt	sync with head.
1.50.8.4	02-May-2011	jym	Sync with head.
1.50.8.3	01-Nov-2009	jym	Sync with HEAD.
1.50.8.2	31-May-2009	jym	Sync with HEAD.
1.50.8.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.50.4.2	21-Mar-2012	jdc	Apply patch (requested by bouyer in ticket #1738). Do not special-case XEN and always use the proper selectors for %fs and %gs in buildcontext() and setregs(). The consequence was that signal handlers would have the wrong %fs/%gs. Found by running atf tests under Xen/i386.
1.50.4.1	18-Nov-2008	snj	Apply patch (requested by bouyer in ticket #65): Add i386-specific COMPAT_30_PTHREAD option (enabled where COMPAT_30 is enabled), which restore binary compatibility with netbsd-3 libpthread.
1.50.2.1	28-Apr-2009	skrll	Sync with HEAD.
1.53.6.1	06-Jun-2011	jruoho	Sync with HEAD.
1.53.4.1	31-May-2011	rmind	sync with head
1.54.38.1	22-Jan-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1550): sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch sys/arch/amd64/include/segments.h: revision 1.34 via patch sys/arch/i386/i386/machdep.c: revision 1.800 via patch sys/arch/i386/include/segments.h: revision 1.64 via patch sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.54.34.1	22-Jan-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1550): sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch sys/arch/amd64/include/segments.h: revision 1.34 via patch sys/arch/i386/i386/machdep.c: revision 1.800 via patch sys/arch/i386/include/segments.h: revision 1.64 via patch sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.54.32.3	28-Aug-2017	skrll	Sync with HEAD
1.54.32.2	05-Oct-2016	skrll	Sync with HEAD
1.54.32.1	19-Mar-2016	skrll	Sync with HEAD
1.54.30.1	22-Jan-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1550): sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch sys/arch/amd64/include/segments.h: revision 1.34 via patch sys/arch/i386/i386/machdep.c: revision 1.800 via patch sys/arch/i386/include/segments.h: revision 1.64 via patch sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.54.24.1	19-Feb-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1517): sys/arch/amd64/amd64/machdep.c: 1.280 via patch sys/arch/amd64/include/segments.h: 1.34 via patch sys/arch/i386/i386/machdep.c: 1.800 sys/arch/i386/include/segments.h: 1.64 sys/arch/x86/x86/vm_machdep.c: 1.30 Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.54.16.1	19-Feb-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1517): sys/arch/amd64/amd64/machdep.c: 1.280 via patch sys/arch/amd64/include/segments.h: 1.34 via patch sys/arch/i386/i386/machdep.c: 1.800 sys/arch/i386/include/segments.h: 1.64 sys/arch/x86/x86/vm_machdep.c: 1.30 Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.54.14.1	03-Dec-2017	jdolecek	update from HEAD
1.54.10.1	19-Feb-2018	snj	Pull up following revision(s) (requested by maxv in ticket #1517): sys/arch/amd64/amd64/machdep.c: 1.280 via patch sys/arch/amd64/include/segments.h: 1.34 via patch sys/arch/i386/i386/machdep.c: 1.800 sys/arch/i386/include/segments.h: 1.64 sys/arch/x86/x86/vm_machdep.c: 1.30 Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.55.2.1	20-Mar-2017	pgoyette	Sync with HEAD
1.57.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.59.6.1	01-Jan-2018	snj	Pull up following revision(s) (requested by maxv in ticket #477): sys/arch/amd64/amd64/machdep.c: revision 1.280 sys/arch/amd64/include/segments.h: revision 1.34 sys/arch/i386/i386/machdep.c: revision 1.800 sys/arch/i386/include/segments.h: revision 1.64 via patch sys/arch/x86/x86/vm_machdep.c: revision 1.30 Fix a huge privilege separation vulnerability in Xen-amd64. On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL equals SEL_UPL. While Xen can make a distinction between usermode and kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes to read and write to the CPU ports. It is easy, then, to completely escalate privileges; by reprogramming the PIC, by reading the ATA disks, by intercepting the keyboard interrupts (keylogger), etc. Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use the ports but not userland. I didn't test this change on i386, but it seems fine enough.
1.64.4.1	10-Jun-2019	christos	Sync with HEAD
1.64.2.1	30-Sep-2018	pgoyette	Ssync with HEAD
1.2	14-Sep-1998	thorpej	Need 13 longs for jmp_buf; signal mask is 128 bits now. XXX Actually, libc appears to use only 7 of the previous 10, so increasing the size isn't actually necessary! But there was a gap at the end before, so we'll keep it.
1.1	20-Dec-1994	cgd	make the definition of _JBLEN mach. dep. header-dependent.
1.31	29-Oct-2021	thorpej	Define __HAVE_STRUCT_SIGCONTEXT regardless of its current visibility.
1.30	26-Oct-2021	christos	Merge all MD __sigaction14_sigtramp.c copies into one: - sparc and sparc64 were not using version 0 sigcontext when there were no arguments in the signal version. This was probably a bug. - vax is using +1 the version numbers of the other archs. - Only hppa was defining __LIBC12_SOURCE__ so it was getting a working sigcontext before. all the other ports that supported sigcontext had the compat code disabled. [pointed out by thorpej, thanks!] If we want to remove sigcontext support from userland at least now there is less work to do so.
1.29	19-Nov-2008	ad	Make the emulations, exec formats, coredump, NFS, and the NFS server into modules. By and large this commit: - shuffles header files and ifdefs - splits code out where necessary to be modular - adds module glue for each of the components - adds/replaces hooks for things that can be installed at runtime
1.28	14-Nov-2008	ad	Remove some compat ifdefs that aren't really needed.
1.27	20-Jan-2005	drochner	branches: 1.27.92; 1.27.96; 1.27.102; 1.27.104; restrict visibility of struct sigcontext* as much as possible; include the needed opt_compat_*.h for kernel code
1.26	10-May-2004	drochner	branches: 1.26.4; SIGTRAMP_VALID() should not pollute the user namespace
1.25	26-Mar-2004	drochner	nothing cares about __HAVE_SIGINFO anymore, so nuke it
1.24	27-Sep-2003	christos	add SIGTRAMP_VALID and gc old sigcontext <-> ucontext macros.
1.23	12-Sep-2003	christos	make sigcontext visible all the time instead of including opt_*.h from a large number of files.
1.22	11-Sep-2003	christos	fix typo.
1.21	11-Sep-2003	christos	Bring back sigcontext to userland by popular demand.
1.20	11-Sep-2003	christos	XXX: vm86 for sigcontext comment is gone.
1.19	10-Sep-2003	christos	#ifdef struct sigframe_sigcontext and struct sigcontext, so that only code that really needs them, get them. XXX1: We need to fix the vm86 api. XXX2: We need to fix the ibcs2 signal delivery mechanism.
1.18	06-Sep-2003	christos	SA_SIGINFO changes.
1.17	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.16	28-Apr-2003	bjh21	branches: 1.16.2; Add a new feature-test macro, _NETBSD_SOURCE. If this is defined by the application, all NetBSD interfaces are made visible, even if some other feature-test macro (like _POSIX_C_SOURCE) is defined. <sys/featuretest.h> defined _NETBSD_SOURCE if none of _ANSI_SOURCE, _POSIX_C_SOURCE and _XOPEN_SOURCE is defined, so as to preserve existing behaviour. This has two major advantages: + Programs that require non-POSIX facilities but define _POSIX_C_SOURCE can trivially be overruled by putting -D_NETBSD_SOURCE in their CFLAGS. + It makes most of the #ifs simpler, in that they're all now ORs of the various macros, rather than having checks for (!defined(_ANSI_SOURCE) \|\| !defined(_POSIX_C_SOURCE) \|\| !defined(_XOPEN_SOURCE)) all over the place. I've tried not to change the semantics of the headers in any case where _NETBSD_SOURCE wasn't defined, but there were some places where the current semantics were clearly mad, and retaining them was harder than correcting them. In particular, I've mostly normalised things so that _ANSI_SOURCE gets you the smallest set of stuff, then _POSIX_C_SOURCE, _XOPEN_SOURCE and _NETBSD_SOURCE in that order. Tested by building for vax, encouraged by thorpej, and uncontested in tech-userlevel for a week.
1.15	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.14	14-Sep-1998	thorpej	branches: 1.14.26; oops, missed one.
1.13	14-Sep-1998	thorpej	sigset13_t -> int.
1.12	13-Sep-1998	thorpej	__LIBCxx_SOURCE__ refers to the library major number, not the NetBSD release number.
1.11	13-Sep-1998	thorpej	Fix thinko in last thinko fix. We always have to store an old style signal mask since a 1.3 binary may attempt to invoke sigreturn(2) directly for an alternate exit from the signal handler. If we don't do this, it will get a garbage signal mask if it tries to do that.
1.10	12-Sep-1998	mycroft	Fix thinko in previous.
1.9	12-Sep-1998	mycroft	Version sys_sigreturn, to avoid breaking programs that use it explicitly.
1.8	11-Sep-1998	mycroft	Substantial signal handling changes: * Increase the size of sigset_t to accomodate 128 signals -- adding new versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() to handle the changed arguments. * Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() into separate functions, and call them from all the emulations rather than hard-coding everything. (Avoids uses the stackgap crap for these system calls.) * Add a new flag (p_checksig) to indicate that a process may have signals pending and userret() needs to do the full (slow) check. * Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE. * Correct emulation bugs with restoring SS_ONSTACK. * Make the signal mask in the sigcontext always use the emulated mask format. * Store signals internally in sigaction structures, rather than maintaining a bunch of little sigsets for each SA_* bit. * Keep track of where we put the signal trampoline, rather than figuring it out in _sendsig(). Issue a warning when a non-emulated sigaction bit is observed. * Add missing emulated signals, and a native SIGPWR (currently not used). * Implement the `not reset when caught' semantics for relevant signals. Note: Only code touched by the i386 port has been modified. Other ports and emulations need to be updated.
1.7	25-May-1998	kleink	If any of _ANSI_SOURCE, _POSIX_C_SOURCE or _XOPEN_SOURCE are defined, don't provide any identifiers other than sig_atomic_t.
1.6	08-Jan-1996	mycroft	VM86 support, by John Kohl, touched up a bit by me.
1.5	01-May-1995	mycroft	Add sc_gs and sc_fs. Move sc_onstack and sc_mask to end.
1.4	10-Jan-1995	jtc	Only define sig_atomic_t when _ANSI_SOURCE is defined.
1.3	27-Oct-1994	cgd	new RCS ID format.
1.2	09-Oct-1994	mycroft	Make reg, frame, and sigcontext more similar.
1.1	19-Sep-1993	brezak	branches: 1.1.2; 1.1.4; Machine dependant signal.h. Defines sigcontext and sig_atomic_t for i386 port.
1.1.4.1	11-Oct-1994	mycroft	Update from trunk.
1.1.2.2	19-Sep-1993	brezak	Machine dependant signal.h. Defines sigcontext and sig_atomic_t for i386 port.
1.1.2.1	19-Sep-1993	brezak	file signal.h was added on branch magnum on 1993-09-19 18:57:49 +0000
1.14.26.1	16-Jan-2003	thorpej	* Include <sys/sigtypes.h> rather than <sys/signal.h> in <sys/ucontext.h>. * Define _UCONTEXT_TO_SIGCONTEXT() and _SIGCONTEXT_TO_UCONTEXT() macros for converting a ucontext -> sigcontext and back again. These macros in turn use machine-dependent macros _MCONTEXT_TO_SIGCONTEXT() and _SIGCONTEXT_TO_MCONTEXT() provided by <machine/signal.h>. The conversion process is not 100% accurate, but should be close enough. Also note that the mcontext conversion may not be enough for all platforms (m68k is a good example of this). These macros should be used only if you really know what you're doing.
1.16.2.4	24-Jan-2005	skrll	Sync with HEAD.
1.16.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.16.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.16.2.1	03-Aug-2004	skrll	Sync with HEAD
1.26.4.1	29-Apr-2005	kent	sync with -current
1.27.104.1	19-Jan-2009	skrll	Sync with HEAD.
1.27.102.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.27.96.1	04-May-2009	yamt	sync with head.
1.27.92.1	17-Jan-2009	mjf	Sync with HEAD.
1.1	23-Jul-2014	alnsn	branches: 1.1.2; 1.1.6; Rename sljitarch.h to sljit_machdep.h.
1.1.6.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.1.6.1	23-Jul-2014	tls	file sljit_machdep.h was added on branch tls-maxphys on 2014-08-20 00:03:06 +0000
1.1.2.2	10-Aug-2014	tls	Rebase.
1.1.2.1	23-Jul-2014	tls	file sljit_machdep.h was added on branch tls-earlyentropy on 2014-08-10 06:53:59 +0000
1.4	23-Jul-2014	alnsn	Rename sljitarch.h to sljit_machdep.h.
1.3	17-Nov-2013	alnsn	branches: 1.3.2; Always define SLJIT_CACHE_FLUSH(), start include guards with '_' and use _LP64 guard.
1.2	18-Nov-2012	alnsn	branches: 1.2.2; 1.2.4; Disable generation of fastcall functions on i386.
1.1	28-Oct-2012	alnsn	branches: 1.1.2; Build bpfjit on i386.
1.1.2.4	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.1.2.3	16-Jan-2013	yamt	sync with (a bit old) head
1.1.2.2	30-Oct-2012	yamt	sync with head
1.1.2.1	28-Oct-2012	yamt	file sljitarch.h was added on branch yamt-pagecache on 2012-10-30 17:19:50 +0000
1.2.4.1	18-May-2014	rmind	sync with head
1.2.2.3	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.2.2.2	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.2.2.1	18-Nov-2012	tls	file sljitarch.h was added on branch tls-maxphys on 2012-11-20 03:01:28 +0000
1.3.2.1	10-Aug-2014	tls	Rebase.
1.29	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.28	05-Jan-2003	sommerfeld	Add symbolic names for MSR's used by SYSENTER/EXIT. (I was experimenting a while back, might as well commit this one file..)
1.27	11-Dec-2002	joda	cpuid feature bit 30 is set on IA-64 (in x86 mode)
1.26	06-Dec-2002	junyoung	Display the extended feature flags with non-Intel processors rather than the standard flags. See also PR#19163. Before: cpu0: AMD Athlon XP 1800+ (686-class), 1532.11 MHz cpu0: features 383f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR> cpu0: features 383f9ff<PGE,MCA,CMOV,FGPAT,PSE36,MMX> cpu0: features 383f9ff<FXSR,SSE> After: cpu0: AMD Athlon XP 1800+ (686-class), 1532.11 MHz cpu0: features c3cbf9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR> cpu0: features c3cbf9ff<PGE,MCA,CMOV,PAT,PSE36,MPC,MMXX,MMX> cpu0: features c3cbf9ff<FXSR,SSE,3DNOW2,3DNOW> While I'm here, amd_cpuid_cpu_cacheinfo() is an info function rather than a probe function.
1.25	05-Dec-2002	junyoung	- FGPAT -> PAT, as documented in Intel manual. - Feature flag 31 is "Signal Break on FERR" (SBF).
1.24	25-Oct-2002	fvdl	Add macros for cpuid family/step, etc shifting. Moved here from machdep.c.
1.23	07-Jun-2002	gmcgarry	branches: 1.23.2; Add Athlon performance counters.
1.22	07-Jun-2002	gmcgarry	Add i586 performance counters.
1.21	20-Mar-2002	joda	branches: 1.21.6; add HTT cpu flag, and further split message
1.20	19-Sep-2001	thorpej	Define the AMD K6 cache/write-combinding control register MSR.
1.19	01-Aug-2001	thorpej	branches: 1.19.4; Add some more CPUID feature bits.
1.18	22-Jul-2001	wiz	seperate -> separate
1.17	19-Jun-2001	wiz	branches: 1.17.4; `accessible' only has one `a'.
1.16	19-Jun-2001	fvdl	Add fxsr CR4 bits.
1.15	21-Dec-2000	enami	branches: 1.15.4; Print cpu features line only when corresponding part of bits are set.
1.14	20-Sep-2000	fvdl	Fix typo.
1.13	13-Sep-2000	thorpej	Define some 586-class CESR MSR bits.
1.12	13-Sep-2000	thorpej	Add 686-class performance counter events.
1.11	27-Mar-2000	thorpej	Add/correct some MSRs, from Intel Architecture Software Developer's Manual, Volume 3 (System Programming).
1.10	24-Mar-2000	thorpej	Add some Model Specific Register definitions. From FreeBSD.
1.9	13-Dec-1999	sommerfeld	Add new CPU feature flags (up through recent Pentium III and Celerons)
1.8	05-Sep-1997	chuck	branches: 1.8.18; 1.8.24; add bits for %cr4 and cpu_feature [from freebsd and www.sandpile.org]
1.7	27-Oct-1994	cgd	branches: 1.7.16; new RCS ID format.
1.6	03-Aug-1994	deraadt	document cyrix configuration registers
1.5	24-May-1994	mycroft	branches: 1.5.2; Add some more constants.
1.4	20-Dec-1993	mycroft	branches: 1.4.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.3	22-May-1993	cgd	branches: 1.3.4; add rcsids to everything and clean up headers
1.2	09-May-1993	deraadt	npx patches from Bruce Evans. patchkit 10002
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.3.4.1	08-Nov-1993	mycroft	Garbage collect some constants, and the entire 286 nonsense. (Like we even get to start() on a 286? NOT!)
1.4.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.4.2.1	20-Dec-1993	mycroft	file specialreg.h was added on branch magnum on 1993-12-20 05:25:47 +0000
1.5.2.1	03-Aug-1994	mycroft	update from trunk
1.7.16.1	06-Sep-1997	thorpej	Update marc-pcmcia branch from trunk.
1.8.24.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.8.18.2	05-Jan-2001	bouyer	Sync with HEAD
1.8.18.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.15.4.9	07-Jan-2003	thorpej	Sync with HEAD.
1.15.4.8	11-Dec-2002	thorpej	Sync with HEAD.
1.15.4.7	11-Dec-2002	thorpej	Sync with HEAD.
1.15.4.6	11-Nov-2002	nathanw	Catch up to -current
1.15.4.5	20-Jun-2002	nathanw	Catch up to -current.
1.15.4.4	01-Apr-2002	nathanw	Catch up to -current. (CVS: It's not just a program. It's an adventure!)
1.15.4.3	21-Sep-2001	nathanw	Catch up to -current.
1.15.4.2	24-Aug-2001	nathanw	Catch up with -current.
1.15.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.17.4.3	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.17.4.2	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.17.4.1	03-Aug-2001	lukem	update to -current
1.19.4.1	01-Oct-2001	fvdl	Catch up with -current.
1.21.6.1	14-Jul-2002	gehenna	catch up with -current.
1.23.2.2	07-Jun-2002	gmcgarry	Add Athlon performance counters.
1.23.2.1	07-Jun-2002	gmcgarry	file specialreg.h was added on branch sommerfeld_i386mp_1 on 2002-06-07 04:03:50 +0000
1.4	15-Apr-1998	drochner	"spkr" is now mi. Keep this file for compatibility.
1.3	27-Oct-1994	cgd	new RCS ID format.
1.2	11-Jan-1994	mycroft	branches: 1.2.2; Gratuitous formatting changes.
1.1	16-Jun-1993	brezak	branches: 1.1.4; Move spkr.h to include/machine so users can use it.
1.1.4.2	14-Nov-1993	mycroft	Missed something...
1.1.4.1	14-Nov-1993	mycroft	Clean up a bit.
1.2.2.2	11-Jan-1994	mycroft	Gratuitous formatting changes.
1.2.2.1	11-Jan-1994	mycroft	file spkr.h was added on branch magnum on 1994-01-11 13:30:41 +0000
1.24	17-Jul-2011	joerg	Retire varargs.h support. Move machine/stdarg.h logic into MI sys/stdarg.h and expect compiler to provide proper builtins, defaulting to the GCC interface. lint still has a special fallback. Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and derive va_list as required by standards.
1.23	18-Aug-2009	drochner	use __builtin_va_start rather than __builtin_stdarg_start for gcc>=4.1, I don't know why it was made that inconsistent at that time, but the in-tree gcc-4.1 works with that, and gcc-4.4.1 removes builtin support for old style varargs.
1.22	21-Jun-2008	gmcgarry	branches: 1.22.10; Add stdargs support for pcc.
1.21	07-Aug-2003	agc	branches: 1.21.102; 1.21.106; 1.21.108; 1.21.110; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.20	28-Apr-2003	bjh21	branches: 1.20.2; Add a new feature-test macro, _NETBSD_SOURCE. If this is defined by the application, all NetBSD interfaces are made visible, even if some other feature-test macro (like _POSIX_C_SOURCE) is defined. <sys/featuretest.h> defined _NETBSD_SOURCE if none of _ANSI_SOURCE, _POSIX_C_SOURCE and _XOPEN_SOURCE is defined, so as to preserve existing behaviour. This has two major advantages: + Programs that require non-POSIX facilities but define _POSIX_C_SOURCE can trivially be overruled by putting -D_NETBSD_SOURCE in their CFLAGS. + It makes most of the #ifs simpler, in that they're all now ORs of the various macros, rather than having checks for (!defined(_ANSI_SOURCE) \|\| !defined(_POSIX_C_SOURCE) \|\| !defined(_XOPEN_SOURCE)) all over the place. I've tried not to change the semantics of the headers in any case where _NETBSD_SOURCE wasn't defined, but there were some places where the current semantics were clearly mad, and retaining them was harder than correcting them. In particular, I've mostly normalised things so that _ANSI_SOURCE gets you the smallest set of stuff, then _POSIX_C_SOURCE, _XOPEN_SOURCE and _NETBSD_SOURCE in that order. Tested by building for vax, encouraged by thorpej, and uncontested in tech-userlevel for a week.
1.19	05-May-2000	thorpej	branches: 1.19.6; Use __GNUC_PREREQ__().
1.18	03-May-2000	thorpej	Make varargs and stdarg work with GCC 2.96 and later.
1.17	03-Feb-2000	kleink	Add a C99-style va_copy macro.
1.16	03-May-1999	christos	branches: 1.16.2; Define __builtin_*() for lint
1.15	22-Jan-1999	mycroft	branches: 1.15.4; Use __builtin_next_arg(). Fixed PR 6862.
1.14	26-Sep-1998	christos	delint
1.13	27-Jul-1998	mycroft	Delint.
1.12	25-Dec-1995	mycroft	branches: 1.12.16; Stylistic changes.
1.11	25-Dec-1995	mycroft	Stylistic changes.
1.10	25-Dec-1995	mycroft	Update for GCC 2.7, and fix bugs.
1.9	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.8	28-Jan-1995	jtc	ANSI says that <stdarg.h>'s va_end macro must expand to a void expression. For consistancy, I'm changing <varargs.h> too.
1.7	28-Dec-1994	cgd	do the right thing with strange-sized varargs objects.
1.6	27-Oct-1994	cgd	new RCS ID format.
1.5	15-Oct-1994	cgd	make <stdarg.h> a symlink, and clean up ports' stdarg.h and varargs.h files.
1.4	21-May-1994	cgd	update from lite
1.3	20-Dec-1993	mycroft	branches: 1.3.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.1	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.3.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.3.2.1	20-Dec-1993	mycroft	file stdarg.h was added on branch magnum on 1993-12-20 05:25:49 +0000
1.12.16.1	01-Feb-1999	cgd	pull up revs 1.13-1.15 from trunk (PR#6862). (mycroft)
1.15.4.1	21-Jun-1999	thorpej	Sync w/ -current.
1.16.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.19.6.2	05-May-2000	thorpej	Use __GNUC_PREREQ__().
1.19.6.1	05-May-2000	thorpej	file stdarg.h was added on branch sommerfeld_i386mp_1 on 2000-05-05 00:21:49 +0000
1.20.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.20.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.20.2.1	03-Aug-2004	skrll	Sync with HEAD
1.21.110.1	27-Jun-2008	simonb	Sync with head.
1.21.108.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.21.106.2	19-Aug-2009	yamt	sync with head.
1.21.106.1	04-May-2009	yamt	sync with head.
1.21.102.1	29-Jun-2008	mjf	Sync with HEAD.
1.22.10.2	27-Aug-2011	jym	Add/remove files, like in HEAD.
1.22.10.1	01-Nov-2009	jym	Sync with HEAD.
1.17	01-Aug-2017	maxv	Remove svr4_machdep.h right away, no one should include it.
1.16	28-Apr-2008	martin	branches: 1.16.44; 1.16.64; Remove clause 3 and 4 from TNF licenses
1.15	27-Oct-2003	junyoung	branches: 1.15.102; 1.15.104; 1.15.106; Nuke __P().
1.14	21-Dec-2001	jdolecek	branches: 1.14.2; 1.14.18; rename (incorrect) SVR4_TRAP_CLOCK_SETTIME to SVR4_TRAP_GETHRESTIME - this is what Solaris8 calls it implement SVR4_TRAP_GETHRESTIME, though obviously only with microsecond accuracy now clock_gettime(CLOCK_REALTIME, tp) works on Solaris8/i386
1.13	31-Oct-2001	jdolecek	Add defines for the rest of fasttraps as defined on Solaris8/i386. For documentation purposes only (though CLOCK_SETTIME a.k.a T_GETHRESTIME shouldn't be too hard to implement).
1.12	11-Dec-2000	mycroft	branches: 1.12.4; 1.12.6; 1.12.10; Introduce 2 new flags in types.h: * __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by e_syscall_intern, which is called at key places in the kernel. This can be used to set a MD syscall handler pointer. This obsoletes and replaces the _HAS_SEPARATED_SYSCALL flags. __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in struct emul are omitted.
1.11	09-Dec-2000	jdolecek	prototype svr4_syscall() (only) in compat/svr4/svr4_exec.c
1.10	09-Dec-2000	mycroft	Add a separate svr4_syscall().
1.9	21-Jan-1999	christos	branches: 1.9.8; Move prototypes to mi code; add MD constant for padding
1.8	13-Dec-1998	christos	Add defines for the fast trap vectors.
1.7	11-Sep-1998	mycroft	Substantial signal handling changes: * Increase the size of sigset_t to accomodate 128 signals -- adding new versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() to handle the changed arguments. * Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(), sys_sigpending() and sys_sigsuspend() into separate functions, and call them from all the emulations rather than hard-coding everything. (Avoids uses the stackgap crap for these system calls.) * Add a new flag (p_checksig) to indicate that a process may have signals pending and userret() needs to do the full (slow) check. * Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE. * Correct emulation bugs with restoring SS_ONSTACK. * Make the signal mask in the sigcontext always use the emulated mask format. * Store signals internally in sigaction structures, rather than maintaining a bunch of little sigsets for each SA_* bit. * Keep track of where we put the signal trampoline, rather than figuring it out in _sendsig(). Issue a warning when a non-emulated sigaction bit is observed. * Add missing emulated signals, and a native SIGPWR (currently not used). * Implement the `not reset when caught' semantics for relevant signals. Note: Only code touched by the i386 port has been modified. Other ports and emulations need to be updated.
1.6	05-Sep-1998	christos	Assign copyright to TNF.
1.5	31-Mar-1995	christos	Reorganization of the signal frame and ucontext to make it shareable with the sparc code. Added trap defines.
1.4	01-Feb-1995	christos	Added defines for svr4_sysarch()
1.3	25-Jan-1995	christos	Modify the floating point part of our context to reflect reality.
1.2	09-Jan-1995	christos	Added missing forward declaration
1.1	08-Jan-1995	christos	New file for signal handling and {g,s}etcontext().
1.9.8.1	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.12.10.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.12.6.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.12.4.2	08-Jan-2002	nathanw	Catch up to -current.
1.12.4.1	14-Nov-2001	nathanw	Catch up to -current.
1.14.18.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.14.18.2	18-Sep-2004	skrll	Sync with HEAD.
1.14.18.1	03-Aug-2004	skrll	Sync with HEAD
1.14.2.2	21-Dec-2001	jdolecek	rename (incorrect) SVR4_TRAP_CLOCK_SETTIME to SVR4_TRAP_GETHRESTIME - this is what Solaris8 calls it implement SVR4_TRAP_GETHRESTIME, though obviously only with microsecond accuracy now clock_gettime(CLOCK_REALTIME, tp) works on Solaris8/i386
1.14.2.1	21-Dec-2001	jdolecek	file svr4_machdep.h was added on branch sommerfeld_i386mp_1 on 2001-12-21 07:02:24 +0000
1.15.106.1	16-May-2008	yamt	sync with head.
1.15.104.1	18-May-2008	yamt	sync with head.
1.15.102.1	02-Jun-2008	mjf	Sync with HEAD.
1.16.64.1	28-Aug-2017	skrll	Sync with HEAD
1.16.44.1	03-Dec-2017	jdolecek	update from HEAD
1.18	16-Apr-2007	ad	Share the sysarch stuff between the x86 ports. PR kern/36046.
1.17	12-Apr-2006	drochner	branches: 1.17.14; 1.17.18; 1.17.20; second step to retire the useless i386_set_ioperm/i386_get_ioperm calls: remove declarations from public headers (it still needs to be built for binary compatibility; one might consider establishing a separate "compat" department as libc does, but these little wrapper functions aren't worth the effort yet)
1.16	26-Dec-2005	perry	branches: 1.16.4; 1.16.6; 1.16.8; 1.16.10; 1.16.12; u_intN_t -> uintN_t
1.15	27-Oct-2003	junyoung	branches: 1.15.16; Nuke __P().
1.14	11-Sep-2003	christos	make vm86 use __gregset_t
1.13	11-Sep-2003	kleink	__{BEGIN,END}_DECLS-wrap prototypes.
1.12	07-Jun-2002	gmcgarry	branches: 1.12.2; 1.12.8; Support Athlon performance counters.
1.11	10-Sep-2001	fvdl	branches: 1.11.12; Add definitions for mtrr syscalls.
1.10	13-Sep-2000	thorpej	branches: 1.10.4; 1.10.6; 1.10.8; Add support for using the 586- and 686-class performance counters.
1.9	25-Feb-1998	perry	branches: 1.9.14; change second parm of sysarch() from char * to void *
1.8	08-Jan-1996	mycroft	VM86 support, by John Kohl, touched up a bit by me.
1.7	11-Oct-1995	mycroft	Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.6	05-Jun-1995	pk	Prototype sysarch(2) (Thorsten Lockert, PR#1096)
1.5	06-May-1995	mycroft	Define i386_[gs]et_ldt_args here.
1.4	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.3	27-Oct-1994	cgd	new RCS ID format.
1.2	20-Dec-1993	mycroft	branches: 1.2.2; Cleanup and garbage collection; nothing significant. From magnum branch.
1.1	19-Sep-1993	brezak	branches: 1.1.2; Defines architecure dependant syscalls.
1.1.2.1	14-Nov-1993	mycroft	Do the multiple-inclusion thang.
1.2.2.2	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.2.2.1	20-Dec-1993	mycroft	file sysarch.h was added on branch magnum on 1993-12-20 05:25:53 +0000
1.9.14.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.10.8.1	01-Oct-2001	fvdl	Catch up with -current.
1.10.6.2	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.10.6.1	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.10.4.2	20-Jun-2002	nathanw	Catch up to -current.
1.10.4.1	21-Sep-2001	nathanw	Catch up to -current.
1.11.12.1	14-Jul-2002	gehenna	catch up with -current.
1.12.8.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.12.8.2	18-Sep-2004	skrll	Sync with HEAD.
1.12.8.1	03-Aug-2004	skrll	Sync with HEAD
1.12.2.2	07-Jun-2002	gmcgarry	Support Athlon performance counters.
1.12.2.1	07-Jun-2002	gmcgarry	file sysarch.h was added on branch sommerfeld_i386mp_1 on 2002-06-07 04:04:48 +0000
1.15.16.2	03-Sep-2007	yamt	sync with head.
1.15.16.1	21-Jun-2006	yamt	sync with head.
1.16.12.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.16.10.1	19-Apr-2006	elad	sync with head - hopefully this will work
1.16.8.1	24-May-2006	yamt	sync with head.
1.16.6.1	22-Apr-2006	simonb	Sync with head.
1.16.4.1	09-Sep-2006	rpaulo	sync with head
1.17.20.1	11-Jul-2007	mjf	Sync with head.
1.17.18.1	27-May-2007	ad	Sync with head.
1.17.14.1	07-May-2007	yamt	sync with head.
1.5	17-Sep-2017	maxv	Remove tlog.h - unused now. Note that it is not installed.
1.4	13-Feb-2004	wiz	branches: 1.4.150; Uppercase CPU, plural is CPUs.
1.3	11-Oct-2002	fvdl	branches: 1.3.2; 1.3.8; Add NetBSD RCS Id.
1.2	01-Oct-2002	fvdl	branches: 1.2.2; Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.1	27-Apr-2002	sommerfeld	branches: 1.1.2; file tlog.h was initially added on branch sommerfeld_i386mp_1.
1.1.2.1	27-Apr-2002	sommerfeld	Add options TRAPLOG: When enabled, each cpu gets a ring buffer; on every trap entry, we add an entry to the trap log using the new locore "TLOG" macro. Somewhat like a flight recorder, this may give some post-mortem clues as to what was going on as we were spiralling in... XXX currently depends on various intel MSR's to record the low order 32 bits of the cycle counter as well as last-branch from and to addresses.
1.2.2.2	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.2.2.1	01-Oct-2002	jdolecek	file tlog.h was added on branch kqueue on 2002-10-10 18:33:30 +0000
1.3.8.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.3.8.2	18-Sep-2004	skrll	Sync with HEAD.
1.3.8.1	03-Aug-2004	skrll	Sync with HEAD
1.3.2.2	18-Oct-2002	nathanw	Catch up to -current.
1.3.2.1	11-Oct-2002	nathanw	file tlog.h was added on branch nathanw_sa on 2002-10-18 02:37:58 +0000
1.4.150.1	03-Dec-2017	jdolecek	update from HEAD
1.7	26-Feb-2003	fvdl	Redirect some includes to x86/foo.h
1.6	05-Oct-2002	fvdl	Define some masks to be able to pass information about a trap being a TSS trap down to things like DDB. XXX a bit of a hack.
1.5	17-Jun-2002	fvdl	branches: 1.5.2; Add XMM trap type.
1.4	27-Oct-1994	cgd	branches: 1.4.48; 1.4.50; 1.4.62; new RCS ID format.
1.3	19-Dec-1993	mycroft	branches: 1.3.2; Clean up trap handling and macroize interrupt entry and exit code. From magnum branch.
1.2	22-May-1993	cgd	branches: 1.2.4; add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.2.4.1	13-Nov-1993	mycroft	Garbage collect unused traps.
1.3.2.2	19-Dec-1993	mycroft	Clean up trap handling and macroize interrupt entry and exit code. From magnum branch.
1.3.2.1	19-Dec-1993	mycroft	file trap.h was added on branch magnum on 1993-12-19 06:58:28 +0000
1.4.62.1	16-Jul-2002	gehenna	catch up with -current.
1.4.50.1	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.4.48.2	18-Oct-2002	nathanw	Catch up to -current.
1.4.48.1	20-Jun-2002	nathanw	Catch up to -current.
1.5.2.2	17-Jun-2002	fvdl	Add XMM trap type.
1.5.2.1	17-Jun-2002	fvdl	file trap.h was added on branch sommerfeld_i386mp_1 on 2002-06-17 21:38:21 +0000
1.12	07-Jul-2018	kamil	Correct unportable signed integer left shift in i386/amd64 tss code Change the type of IOMAP_INVALOFF to unsigned int. sys/arch/amd64/amd64/machdep.c:518:42, left shift of 65535 by 16 places cannot be represented in type 'int' Detected with Kernel Undefined Behavior Sanitizer.
1.11	04-Jan-2018	maxv	branches: 1.11.2; 1.11.4; Declare IOMAP_VALIDOFF, not to use ci_tss pointers.
1.10	05-Jan-2008	yamt	move IOMAP_INVALOFF from cpu.h to tss.h
1.9	04-Jan-2008	yamt	i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.8	07-Aug-2003	agc	branches: 1.8.16; 1.8.78; 1.8.84; 1.8.92; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.7	05-Oct-2002	fvdl	branches: 1.7.6; Protect against multiple inclusion.
1.6	11-Oct-1995	mycroft	branches: 1.6.46; Various changes from John Kohl and me: Map kernel stacks only at unique addresses. Use one TSS per process. Add sysarch calls for modifying IOPL and the I/O permission bitmap. Add a compacting GDT entry allocator, for TSS and LDT selectors. Enable modifying %fs and %gs with PT_SETREGS. Sanitize various bits of code.
1.5	01-May-1995	mycroft	Remove historical garbage.
1.4	27-Oct-1994	cgd	new RCS ID format.
1.3	22-May-1993	cgd	add rcsids to everything and clean up headers
1.2	07-Apr-1993	deraadt	/* ... /* ... */ problem in an include file
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.6.46.1	18-Oct-2002	nathanw	Catch up to -current.
1.7.6.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.7.6.2	18-Sep-2004	skrll	Sync with HEAD.
1.7.6.1	03-Aug-2004	skrll	Sync with HEAD
1.8.92.1	08-Jan-2008	bouyer	Sync with HEAD
1.8.84.1	18-Feb-2008	mjf	Sync with HEAD.
1.8.78.1	09-Jan-2008	matt	sync with HEAD
1.8.16.1	21-Jan-2008	yamt	sync with head
1.11.4.1	10-Jun-2019	christos	Sync with HEAD
1.11.2.1	28-Jul-2018	pgoyette	Sync with HEAD
1.95	08-May-2025	imil	Rename BOOTCYCLETIME kernel option and subsequent files to BOOT_DURATION
1.94	06-May-2025	imil	Add BOOTCYCLETIME option to print kernel boot time Introduce a new kernel option, BOOTCYCLETIME, which will print the time taken for the kernel to boot on (for now) amd64 and i386 architectures.
1.93	01-Apr-2021	simonb	branches: 1.93.22; Whitespace: #define<tab>
1.92	01-Aug-2020	jdolecek	branches: 1.92.2; 1.92.4; move __HAVE_PCI_MSI_MSIX to <x86/pci_machdep_common.h>
1.91	04-May-2020	jdolecek	add support for using MSI for XenPV Dom0 use PHYSDEVOP_map_pirq to get the pirq/gsi for MSI/MSI-X, switch also INTx to use it instead of PHYSDEVOP_alloc_irq_vector MSI confirmed working with single-vector MSI for wm(4), ahcisata(4), bge(4) XXX added some provision for MSI-X, but it doesn't actually work (no interrupts delivered), needs some further investigation; disable MSI-X for XENPV via flag in x86/pci/pci_machdep.c
1.90	23-Sep-2019	kamil	Disable __NO_STRICT_ALIGNMENT on amd64/i386 for UBSan builds This change allows to pick code paths in the kernel that are tuned for alignment sensitive (and stricted in C meaning) code paths. In particular the IPv6 code uses this heavily and skips whenever possible the process of aligning of networking data. With this modification all ATF tests are executed on amd64 without triggering any UBSan reports in dmesg. In theory __NO_STRICT_ALIGNMENT could be tuned for vax and m68k, however these machines are still unsupported in LLVM sanitizers and syzkaller. sys/netinet6/scope6.c:404:6, member access within misaligned address 0xfffffaea81276086 for type 'struct in6_addr' which requires 4 byte alignment Reported-by: syzbot+a86f58d17685317b3df9@syzkaller.appspotmail.com sys/net/rtsock_shared.c:629:41, member access within misaligned address 0xffffddb5db3ff04c for type 'struct rt_msghdr50' which requires 8 byte alignment Reported-by: syzbot+0a3a022bc9d2b8880c16@syzkaller.appspotmail.com
1.89	06-Apr-2019	thorpej	Overhaul the API used to fetch and store individual memory cells in userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(), subyte(), suword(), etc.) are retired and replaced with new ufetch(9) and ustore(9) APIs that can return proper error codes, etc. and are implemented consistently across all platforms. The interrupt-safe variants are no longer supported (and several of the existing attempts at fuswintr(), etc. were buggy and not actually interrupt-safe). Also augmement the ucas(9) API, making it consistently available on all plaforms, supporting uniprocessor and multiprocessor systems, even those that do not have CAS or LL/SC primitives. Welcome to NetBSD 8.99.37.
1.88	12-Jul-2018	maxv	Remove the kernel PMC code. Sent yesterday on tech-kern@. This change: * Removes "options PERFCTRS", the associated includes, and the associated ifdefs. In doing so, it removes several XXXSMPs in the MI code, which is good. * Removes the PMC code of ARM XSCALE. * Removes all the pmc.h files. They were all empty, except for ARM XSCALE. * Reorders the x86 PMC code not to rely on the legacy pmc.h file. The definitions are put in sysarch.h. * Removes the kern/sys_pmc.c file, and along with it, the sys_pmc_control and sys_pmc_get_info syscalls. They are marked as OBSOL in kern, netbsd32 and rump. * Removes the pmc_evid_t and pmc_ctr_t types. * Removes all the associated man pages. The sets are marked as obsolete.
1.87	27-Feb-2016	tls	branches: 1.87.16; 1.87.18; Add cpu_rng, a framework for simple on-CPU random number generators.
1.86	24-Jan-2016	christos	expose __vaddr_t for segments.h since it is a userland api.
1.85	23-Jan-2016	christos	expose the kernel types for standalone code.
1.84	23-Jan-2016	christos	Hide {p,v}{addr,size}_t and register_t (and a couple more types that are machine-specific) from userland unless _KERNEL/_KMEMUSER and a new _KERNTYPES variables is defined. The _KERNTYPES should be fixed for many subsystems that should not be using it (rump)...
1.83	27-Aug-2015	pooka	Fix PTHREAD_FOO_INITIALIZER for C++ by not using volatile in the relevant pthread types in C++ builds, attempt 2. The problem with attempt 1 was making assumptions of what the MD __cpu_simple_lock_t (declared volatile) looks like. To get a same type except non-volatile, we change the MD type to __cpu_simple_lock_nv_t and typedef __cpu_simple_lock_t as a volatile __cpu_simple_lock_nv_t. IMO, __cpu_simple_lock_t should not be volatile at all, but changing it now is too risky. Fixes at least Rumprun w/ gcc 5.1/5.2. Furthermore, the mpd application (and possibly others) will no longer require NetBSD-specific patches. Tested: build.sh for i386, Rumprun for x86_64 w/ gcc 5.2. Based on the patch from Christos in lib/49989.
1.82	21-Aug-2015	pooka	Make it possible to explicitly disable MSI/MSIX with NO_PCI_MSI_MSIX. Some platforms, e.g. linux uio-pci-generic, do not support MSI at all. XXX: does MSI being defined intentionally depend on _KERNEL_OPT on amd64 but not i386?
1.81	27-Apr-2015	knakahara	add x86 MD MSI/MSI-X support code.
1.80	22-Apr-2015	pooka	Revise rhetoric for why __HAVE_ATOMIC64_OPS is defined with _KERNEL even though we don't know for sure that the target i386 CPU will support them.
1.79	24-Apr-2014	christos	branches: 1.79.4; Explain what is going on with the 64 bit atomic ops, and restore them for pure kernel use.
1.78	22-Apr-2014	christos	we don't really have 64 bit atomic ops.
1.77	03-Apr-2014	christos	branches: 1.77.2; we have cpu_bootconf()
1.76	01-Dec-2013	christos	revert fpu/pcu changes until we figure out what's wrong; they cause random freezes
1.75	23-Oct-2013	drochner	Use the MI "pcu" framework for bookkeeping of npx/fpu states on x86. This reduces the amount of MD code enormously, and makes it easier to implement support for newer CPU features which require more fpu state, or for fpu usage by the kernel. For access to FPU state across CPUs, an xcall kthread is used now rather than a dedicated IPI. No user visible changes intended.
1.74	06-Jul-2011	dyoung	branches: 1.74.2; 1.74.12; 1.74.16; Implement bus_space_tag_create() and _destroy(). Factor bus_space_reserve(), bus_space_release(), et cetera out of bus_space_alloc(), bus_space_map(), bus_space_free(), bus_space_unmap(), et cetera. For i386 and amd64, activate the use of <machine/bus_defs.h> and <machine/bus_funcs.h> by #defining __HAVE_NEW_STYLE_BUS_H in their respective types.h. While I'm here, remove unnecessary __HAVE_DEVICE_REGISTER #defines.
1.73	12-Jun-2011	rmind	Welcome to 5.99.53! Merge rmind-uvmplock branch: - Reorganize locking in UVM and provide extra serialisation for pmap(9). New lock order: [vmpage-owner-lock] -> pmap-lock. - Simplify locking in some pmap(9) modules by removing P->V locking. - Use lock object on vmobjlock (and thus vnode_t::v_interlock) to share the locks amongst UVM objects where necessary (tmpfs, layerfs, unionfs). - Rewrite and optimise x86 TLB shootdown code, make it simpler and cleaner. Add TLBSTATS option for x86 to collect statistics about TLB shootdowns. - Unify /dev/mem et al in MI code and provide required locking (removes kernel-lock on some ports). Also, avoid cache-aliasing issues. Thanks to Andrew Doran and Joerg Sonnenberger, as their initial patches formed the core changes of this branch.
1.72	12-Mar-2011	joerg	branches: 1.72.2; Add TLS support for AMD64, i386 and SH3. This material is based upon work partially supported by The NetBSD Foundation under a contract with Joerg Sonnenberger.
1.71	24-Feb-2011	joerg	Allow storing and receiving the LWP private pointer via ucontext_t on all platforms except VAX and IA64. Add fast access via register for AMD64, i386 and SH3 ports. Use this fast access in libpthread to replace the stack based pthread_self(). Implement skeleton support for Alpha, HPPA, PowerPC, SPARC and SPARC64, but leave it disabled. Ports that support this feature provide __HAVE____LWP_GETPRIVATE_FAST in machine/types.h and a corresponding __lwp_getprivate_fast in machine/mcontext.h. This material is based upon work partially supported by The NetBSD Foundation under a contract with Joerg Sonnenberger.
1.70	22-Dec-2010	christos	branches: 1.70.2; 1.70.4; Make __HAVE_CPU_DATA_FIRST true
1.69	14-Dec-2010	haad	Revert change made in revision 1.66 by ad@ this is not true and 64bit atomic ops should be enabled in libc by default.
1.68	21-Sep-2010	jym	Promote paddr_t to 64 bits for i386 userland (non PAE and PAE). paddr_t has limited interest in userland, and should only be used by specific code like kvm(3) (for VA => PA translations). To support non PAE and PAE kernels simultaneously in kvm(3), promote paddr_t to 64 bits. No objection on current-users@. See also: http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html Kernel remains untouched by this change. In-kernel 64 bits for paddr_t is for a future commit (Yes, I do not forget about that :) )
1.67	11-Dec-2009	matt	branches: 1.67.2; 1.67.4; Add PRIx{P,V}{ADDR,SIZE}, PRIu{P,V}SIZE, and PRIxREGISTER{,32,64} for all (except where they will be added via merge). These should be used to print {p,v}{addr,size}_t and register*_t as appropriate.
1.66	28-Jun-2009	ad	Define __HAVE_ATOMIC64_OPS only in kernel, since it doesn't work in userspace on the i486.
1.65	19-Apr-2009	ad	cpuctl: - Add interrupt shielding (direct hardware interrupts away from the specified CPUs). Not documented just yet but will be soon. - Redo /dev/cpu time_t compat so no kernel changes are needed. x86: - Make intr_establish, intr_disestablish safe to use when !cold. - Distribute hardware interrupts among the CPUs, instead of directing everything to the boot CPU. - Add MD code for interrupt sheilding. This works in most cases but there is a bug where delivery is not accepted by an LAPIC after redistribution. It also needs re-balancing to make things fair after interrupts are turned back on for a CPU.
1.64	05-Apr-2009	tsutsui	Use #define<tab> consistently.
1.63	05-Apr-2009	tsutsui	Remove __HAVE_UFS2_BOOT since it belongs to sysinst for now. "Feel free to change it" by ad@.
1.62	04-Apr-2009	ad	+__HAVE_UFS2_BOOT
1.61	29-Mar-2009	ad	_lwp_setprivate: provide the value to MD code if a hook is present. This will be used to support TLS. The MD method must match the ELF TLS spec for that CPU architecture (if there is a spec). At this time it is only implemented for i386, where it means setting the per-thread base address for %gs. Please implement this for your platform!
1.60	26-Oct-2008	mrg	branches: 1.60.2; 1.60.8; - use _I386_FOO_H for multi-include protection - use <i386/foo.h> in a couple of places
1.59	21-Feb-2008	ad	branches: 1.59.2; 1.59.6; 1.59.12; #define __HAVE_ATOMIC_AS_MEMBAR, indicating that: - atomic_cas_ni() does an implicit membar_exit() - all other atomic operations do an implicit membar_sync() While this might seem kind of arbitrary it's the basis for some important optimizations.
1.58	23-Jan-2008	bouyer	branches: 1.58.2; Merge the bouyer-xeni386 branch. This brings in PAE support to NetBSD xeni386 (domU only). PAE support is enabled by 'options PAE', see the new XEN3PAE_DOMU and INSTALL_XEN3PAE_DOMU kernel config files. See the comments in arch/i386/include/{pte.h,pmap.h} to see how it works. In short, we still handle it as a 2-level MMU, with the second level page directory being 4 pages in size. pmap switching is done by switching the L2 pages in the L3 entries, instead of loading %cr3. This is almost required by Xen, which handle the last L2 page (the one mapping 0xc0000000 - 0xffffffff) in a very special way. But this approach should also work for native PAE support if ever supported (in fact, the pmap should almost suport native PAE, what's missing is bootstrap code in locore.S).
1.57	20-Jan-2008	joerg	Now that __HAVE_TIMECOUNTER and __HAVE_GENERIC_TODR are invariants, remove the conditionals and the code associated with the undef case.
1.56	15-Jan-2008	joerg	Introduce optional cpu_offline_md to execute MD actions at the end of cpu_offline. Use this on amd64/i386 to force a FPU save. As this was triggered by npxsave_cpu/fpusave_cpu not working for a different CPU, remove the cpu_info argument and adjust npxsave_/fpusave_ to use bool for the save. OK ad@
1.55	11-Jan-2008	bouyer	Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.54	04-Jan-2008	yamt	i386: - make tss per-cpu. this considerably speeds up context switch for, at least, pentium4, where ltr instruction seems very slow. i386, xen: - kill cpu_maxproc. kvm86: - adapt to per-cpu tss. - cleanup and simplify. - move kvm86_mp_lock to more meaningful place. - disable preemption during a call.
1.53	20-Dec-2007	ad	64-bit atomic ops for i386.
1.52	17-Oct-2007	garbled	branches: 1.52.2; 1.52.4; 1.52.8; Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.51	14-Jul-2007	ad	branches: 1.51.10; Generic soft interrupts are mandatory.
1.50	09-Feb-2007	ad	branches: 1.50.6; 1.50.14; Merge newlock2 to head.
1.49	03-Sep-2006	bjh21	branches: 1.49.2; Nothing in the kernel now tests __HAVE_NWSCONS, so stop defining it everywhere.
1.48	03-Sep-2006	gdamore	Convert i386 to use common TODR code. Not tested, but Christos says he'll test it for me right now. :-)
1.47	07-Jun-2006	kardel	convert to timecounters (from branch simonb-timecounters)
1.46	24-Dec-2005	perry	branches: 1.46.4; 1.46.6; 1.46.8; 1.46.14; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.45	18-Jan-2004	martin	branches: 1.45.16; Do not export __HAVE_RAS to userland. Applications are supposed to try rasctl() and detect failure with EOPNOTSUPP.
1.44	26-Sep-2003	nathanw	Move __cpu_simple_lock_t and __SIMPLELOCK_{UN,}LOCKED to machine/types.h so that they can be used in a namespace-friendly way.
1.43	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.42	28-Apr-2003	bjh21	branches: 1.42.2; Add a new feature-test macro, _NETBSD_SOURCE. If this is defined by the application, all NetBSD interfaces are made visible, even if some other feature-test macro (like _POSIX_C_SOURCE) is defined. <sys/featuretest.h> defined _NETBSD_SOURCE if none of _ANSI_SOURCE, _POSIX_C_SOURCE and _XOPEN_SOURCE is defined, so as to preserve existing behaviour. This has two major advantages: + Programs that require non-POSIX facilities but define _POSIX_C_SOURCE can trivially be overruled by putting -D_NETBSD_SOURCE in their CFLAGS. + It makes most of the #ifs simpler, in that they're all now ORs of the various macros, rather than having checks for (!defined(_ANSI_SOURCE) \|\| !defined(_POSIX_C_SOURCE) \|\| !defined(_XOPEN_SOURCE)) all over the place. I've tried not to change the semantics of the headers in any case where _NETBSD_SOURCE wasn't defined, but there were some places where the current semantics were clearly mad, and retaining them was harder than correcting them. In particular, I've mostly normalised things so that _ANSI_SOURCE gets you the smallest set of stuff, then _POSIX_C_SOURCE, _XOPEN_SOURCE and _NETBSD_SOURCE in that order. Tested by building for vax, encouraged by thorpej, and uncontested in tech-userlevel for a week.
1.41	17-Dec-2002	gmcgarry	Move to MI runqueue functions.
1.40	16-Dec-2002	jdolecek	use the __HAVE_CPU_MAXPROC hook to limit kern.maxproc to number of available GDT slots add code in init386() to force maxproc to be maximum cpu_maxproc() this fixes port-i386/1635 by John Kohl
1.39	01-Oct-2002	fvdl	Merge Bill Sommerfeld's i386 MP branch. This code has some known caveats, but works quite well in a lot of MP cases, and all UP cases that I have tested. Parts of this will hopefully be reworked in the not-too-distant future.
1.38	23-Sep-2002	gmcgarry	Back-out previous.
1.37	22-Sep-2002	simonb	Use "#define\t" instead of "#define ".
1.36	22-Sep-2002	gmcgarry	MD component of the new cpu_switch() interface. Introduces nextrunqueue(), cpu_idle() and new cpu_switch() to replace the old cpu_switch() which did the lot. Runs leaner without overly blocking interrupts. Includes cleanup of the RAS code to make use of callee-saved registers. Benchmarks on DX4 @ 100MHz reveal a slight performance improvement but probably not statistically signficant. More TBD to verify this. Changes passed a pounding on Athlon @ 1GHz too.
1.35	22-Sep-2002	gmcgarry	Add __HAVE_MD_RUNQUEUE flag for MD code to override MI run queue primitives.
1.34	28-Aug-2002	gmcgarry	RAS support for i386.
1.33	07-Aug-2002	tron	Redo last change. "pmc_evid_t" and "pmc_ctr_t" need to be visible for userland because they are used in "sys/pmc.h".
1.32	07-Aug-2002	tron	Embed "pmc_evid_t" and "pmc_ctr_t" into "if defined(_KERNEL)" and use "__uint64_t" from "int_types.h". This fixes "lint" errors while building "libc".
1.31	07-Aug-2002	briggs	define pmc_evid_t, pmc_ctr_t.
1.30	28-Feb-2002	simonb	branches: 1.30.8; Use "#define<tab>".
1.29	28-Apr-2001	kleink	branches: 1.29.2; * Move definitions of exact-width integer types from <machine/types.h> to <sys/types.h> and <sys/stdint.h>. * Add a new C99 <stdint.h> header, which provides integer types of explicit width, related limits and integer constant macros. * Extend <inttypes.h> to provide <stdint.h> definitions and format macros for printf() and scanf(). * Add C99 strtoimax() and strtoumax() functions. * Use the latter within scanf(). * Add C99 %j, %t and %z printf()/scanf() conversions for intmax_t, pointer-type and size_t arguments.
1.28	14-Apr-2001	thorpej	Generic soft interrupt implementation for NetBSD/i386. This could probably be tuned somewhat, but this is a stop-gap measure to hold us until Bill Sommerfeld's new interrupt code comes in from the MP branch.
1.27	07-Jan-2001	fvdl	branches: 1.27.2; Define __HAVE_OLD_DISKLABEL in <machine/types.h> for i386 (where the other __HAVE_* defines are). Conditionalize definition of old disklabel struct and ODIOC* definitions on it, allowing other ports to bump the number of partitions in the label if they want to (see comment in sys/disklabel.h).
1.26	03-Jan-2001	takemura	replace 'long long' with int64_t to compile stand alone program with compiler other than GCC.
1.25	11-Dec-2000	mycroft	Introduce 2 new flags in types.h: * __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by e_syscall_intern, which is called at key places in the kernel. This can be used to set a MD syscall handler pointer. This obsoletes and replaces the _HAS_SEPARATED_SYSCALL flags. __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in struct emul are omitted.
1.24	06-Jun-2000	sommerfeld	branches: 1.24.2; 1.24.4; Replace "timestamp" with "counter" in previous change, to avoid any possible confusion that it has anything to do with accurate measurement of time. New names: __HAVE_CPU_COUNTER cpu_counter() cpu_hascounter()
1.23	05-Jun-2000	sommerfeld	Let rnd(4) know how to read the (pentium and higher) cycle counter: If __HAVE_CPU_TIMESTAMP is defined, <machine/rnd.h> exists and defines two functions: 1) u_int32_t cpu_timestamp(void); returns the 32 low order bits of a reasonably high frequency counter. 2) int cpu_havetimestamp(void); returns non-zero if cpu_timestamp() actually works. The timestamp counter should run at a frequency greater than 1 MHz (otherwise, microtime would be sufficient); other than that, the exact frequency and origin of the counter are unspecified.
1.22	05-Feb-2000	cgd	branches: 1.22.4; clean up (group, sort) __HAVE_* defines
1.21	01-Feb-2000	danw	#define __HAVE_DEVICE_REGISTER on ports that have it, and check for that, rather than a list of architecture defines, in config_attach
1.20	05-Feb-1999	thorpej	branches: 1.20.8; The x86 does not have strict alignment requirements, so define __NO_STRICT_ALIGNMENT here.
1.19	13-Aug-1998	thorpej	Garbage-collect vm_offset_t and vm_size_t.
1.18	13-Aug-1998	eeh	Merge paddr_t changes into the main branch.
1.17	14-Jun-1998	kleink	branches: 1.17.2; GC the unused `physadr' type, which was not able to hold a complete physical address on 2 architectures anyhow. Also, move the definition of the `label_t' type inside _KERNEL protection, since it is specific to the in-kernel setjmp()/longjmp() implementations.
1.16	22-Mar-1998	drochner	tell drivers that this port knows about the new wscons code (and generates the NWSCONS etc include files)
1.15	22-Mar-1998	drochner	switch to non-BROKEN_INDIRECT_CONFIG
1.14	05-Nov-1997	thorpej	Mark uses of long long with /* LONGLONG */ for lint. From Chris Demetriou <cgd@pa.dec.com>.
1.13	05-Dec-1996	cgd	branches: 1.13.14; First step inn removing config_scan() and the hacks that gave devices on indirect-config busses a (permanent) softc that they could share between 'match' and 'attach' routines: Define __BROKEN_INDIRECT_CONFIG so that old autoconfiguration interfaces are used, until drivers are converted to use the new interfaces (actually, converted back to use the _older_ interfaces) which prohibit indirect configuration devices from receiving a softc in their match routine that they can share with their attach routine.
1.12	24-Dec-1995	mycroft	Enable paging of the user area.
1.11	06-Jul-1995	cgd	add <sys/cdefs.h> inclusions. namsspace-protect physadr, label_t def'ns against _POSIX_SOURCE and _ANSI_SOURCE.
1.10	28-Jun-1995	cgd	remove unused cpu_exec() definitions. moved "broken swap" markers, for ports that still need it, to types.h.
1.9	20-Nov-1994	christos	Included <sys/cdefs.h> since this file is now using __signed. It broke non gnu compilers.
1.8	27-Oct-1994	cgd	new RCS ID format.
1.7	20-Oct-1994	cgd	update for new syscall args description mechanism
1.6	20-Jul-1994	cgd	define __BIT_TYPES_DEFINED__ for compatibility with things like BIND and nvi
1.5	24-May-1994	cgd	branches: 1.5.2; use __signed instead of signed, per Lite
1.4	23-May-1994	cgd	can't use u_long
1.3	14-Mar-1994	cgd	add basic integral types (a la sparc port) that new nvi wants.
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.5.2.1	20-Jul-1994	cgd	update from trunk.
1.13.14.1	05-Nov-1997	thorpej	Update from trunk: Mark usese of long long with /* LONGLONG */ for lint.
1.17.2.2	12-Aug-1998	eeh	Protect XOPEN and POSIX code from vm_offset_t, paddr_t, vaddr_t, vm_size_t, psize_t, and vsize_t.
1.17.2.1	30-Jul-1998	eeh	Split vm_offset_t and vm_size_t into paddr_t, psize_t, vaddr_t, and vsize_t.
1.20.8.5	21-Apr-2001	bouyer	Sync with HEAD
1.20.8.4	18-Jan-2001	bouyer	Sync with head (for UBC+NFS fixes, mostly).
1.20.8.3	05-Jan-2001	bouyer	Sync with HEAD
1.20.8.2	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.20.8.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.22.4.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.24.4.7	31-Aug-2002	sommerfeld	SMP branch resync. Mostly RAS support.
1.24.4.6	19-Aug-2002	sommerfeld	merged with -current as of today. rearranged gdt init to deal with KVM86 changes; however, KVM86 has no chance of working yet; good thing it's not part of GENERIC
1.24.4.5	27-Apr-2002	sommerfeld	Merge with current as of today
1.24.4.4	30-Apr-2001	sommerfeld	Merge sommerfeld_i386mp_1 branch with -current as of today. GENERIC.MP kernel boots and runs multiuser on my ABIT BP6
1.24.4.3	07-Jan-2001	sommerfeld	Snapshot of merge-in-progress with -current. [Not expected to build]. Catch up with the last N months worth of changes to -current.
1.24.4.2	07-Jan-2001	sommerfeld	Partial merge with -current, largely to pick up __HAVE_OLD_DISKLABEL.
1.24.4.1	06-Jun-2000	sommerfeld	file types.h was added on branch sommerfeld_i386mp_1 on 2001-01-07 18:46:22 +0000
1.24.2.1	01-May-2001	he	Pull up revision 1.27 (via patch, requested by fvdl): Increase the number of BSD disklabel partitions on i386 to 16.
1.27.2.6	19-Dec-2002	thorpej	Sync with HEAD.
1.27.2.5	18-Oct-2002	nathanw	Catch up to -current.
1.27.2.4	17-Sep-2002	nathanw	Catch up to -current.
1.27.2.3	13-Aug-2002	nathanw	Catch up to -current.
1.27.2.2	01-Apr-2002	nathanw	Catch up to -current. (CVS: It's not just a program. It's an adventure!)
1.27.2.1	21-Jun-2001	nathanw	Catch up to -current.
1.29.2.3	10-Oct-2002	jdolecek	sync kqueue with -current; this includes merge of gehenna-devsw branch, merge of i386 MP branch, and part of autoconf rototil work
1.29.2.2	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.29.2.1	16-Mar-2002	jdolecek	Catch up with -current.
1.30.8.1	31-Aug-2002	gehenna	catch up with -current.
1.42.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.42.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.42.2.1	03-Aug-2004	skrll	Sync with HEAD
1.45.16.9	27-Feb-2008	yamt	sync with head.
1.45.16.8	11-Feb-2008	yamt	remove __HAVE_LAZY_MBUF for now.
1.45.16.7	04-Feb-2008	yamt	sync with head.
1.45.16.6	21-Jan-2008	yamt	sync with head
1.45.16.5	03-Sep-2007	yamt	sync with head.
1.45.16.4	26-Feb-2007	yamt	sync with head.
1.45.16.3	30-Dec-2006	yamt	sync with head.
1.45.16.2	21-Jun-2006	yamt	sync with head.
1.45.16.1	07-Jul-2005	yamt	define __HAVE_LAZY_MBUF for i386 and amd64.
1.46.14.1	19-Jun-2006	chap	Sync with head.
1.46.8.3	14-Sep-2006	yamt	sync with head.
1.46.8.2	03-Sep-2006	yamt	sync with head.
1.46.8.1	26-Jun-2006	yamt	sync with head.
1.46.6.1	04-Feb-2006	simonb	Define __HAVE_TIMECOUNTER.
1.46.4.1	09-Sep-2006	rpaulo	sync with head
1.49.2.1	29-Dec-2006	ad	Checkpoint work in progress.
1.50.14.1	03-Oct-2007	garbled	Sync with HEAD
1.50.6.1	15-Jul-2007	ad	Sync with head.
1.51.10.3	23-Mar-2008	matt	sync with HEAD
1.51.10.2	09-Jan-2008	matt	sync with HEAD
1.51.10.1	06-Nov-2007	matt	sync with HEAD
1.52.8.7	23-Jan-2008	bouyer	Sync with HEAD.
1.52.8.6	19-Jan-2008	bouyer	Sync with HEAD
1.52.8.5	13-Jan-2008	bouyer	Work in progress on xeni386 PAE support: Make xeni386 build with a 64bit paddr_t. For this vaddr_t vs paddr_t vs pointers usages had to be clarified. If 'options PAE' is present in a Xen3 kernel, switch paddr_t, pd_entry_t and pt_entry_t to 64bits, and add the PAE entry in the __xen_guest ELF section.
1.52.8.4	11-Jan-2008	bouyer	Xen has generic todr too
1.52.8.3	10-Jan-2008	bouyer	Change xeni386 build to match xenamd64: - machine/ points to xen/include/i386 - remove includes that were identical to the i386/include/ ones; the build will find them though the xen-ma/machine link. - include xen files using xen/ not machine/
1.52.8.2	08-Jan-2008	bouyer	Sync with HEAD
1.52.8.1	02-Jan-2008	bouyer	Sync with HEAD
1.52.4.1	26-Dec-2007	ad	Sync with head.
1.52.2.1	18-Feb-2008	mjf	Sync with HEAD.
1.58.2.1	24-Mar-2008	keiichi	sync with head.
1.59.12.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.59.6.4	09-Oct-2010	yamt	sync with head
1.59.6.3	11-Mar-2010	yamt	sync with head
1.59.6.2	18-Jul-2009	yamt	sync with head.
1.59.6.1	04-May-2009	yamt	sync with head.
1.59.2.2	17-Jan-2009	mjf	Sync with HEAD.
1.59.2.1	21-Feb-2008	mjf	file types.h was added on branch mjf-devfs2 on 2009-01-17 13:28:04 +0000
1.60.8.6	27-Aug-2011	jym	Sync with HEAD. Most notably: uvm/pmap work done by rmind@, and MP Xen work of cherry@. No regression observed on suspend/restore.
1.60.8.5	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.60.8.4	28-Mar-2011	jym	Sync with HEAD. TODO before merge: - shortcut for suspend code in sysmon, when powerd(8) is not running. Borrow ``xs_watch'' thread context? - bug hunting in xbd + xennet resume. Rings are currently thrashed upon resume, so current implementation force flush them on suspend. It's not really needed.
1.60.8.3	01-Nov-2009	jym	Sync with HEAD.
1.60.8.2	23-Jul-2009	jym	Sync with HEAD.
1.60.8.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.60.2.1	28-Apr-2009	skrll	Sync with HEAD.
1.67.4.3	21-Apr-2011	rmind	sync with head
1.67.4.2	05-Mar-2011	rmind	sync with head
1.67.4.1	18-Mar-2010	rmind	Unify /dev/{mem,kmem,zero,null} implementations in MI code. Based on patch from Joerg Sonnenberger, proposed on tech-kern@, in February 2008. Work and depression still in progress.
1.67.2.1	22-Oct-2010	uebayasi	Sync with HEAD (-D20101022).
1.70.4.1	05-Mar-2011	bouyer	Sync with HEAD
1.70.2.1	06-Jun-2011	jruoho	Sync with HEAD.
1.72.2.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.74.16.1	18-May-2014	rmind	sync with head
1.74.12.2	03-Dec-2017	jdolecek	update from HEAD
1.74.12.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.74.2.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.77.2.1	10-Aug-2014	tls	Rebase.
1.79.4.3	19-Mar-2016	skrll	Sync with HEAD
1.79.4.2	22-Sep-2015	skrll	Sync with HEAD
1.79.4.1	06-Jun-2015	skrll	Sync with HEAD
1.87.18.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.87.18.1	10-Jun-2019	christos	Sync with HEAD
1.87.16.1	28-Jul-2018	pgoyette	Sync with HEAD
1.92.4.1	03-Apr-2021	thorpej	Sync with HEAD.
1.92.2.1	03-Apr-2021	thorpej	Sync with HEAD.
1.93.22.1	02-Aug-2025	perseant	Sync with HEAD
1.15	26-Jul-2018	maxv	Rework dbregs, to switch the registers during context switches, and not on each user->kernel transition via userret. Reloads of DR6/DR7 are expensive on both native and xen.
1.14	23-Feb-2017	kamil	branches: 1.14.12; 1.14.14; Introduce PT_GETDBREGS and PT_SETDBREGS in ptrace(2) on i386 and amd64 This interface is modeled after FreeBSD API with the usage. This replaced previous watchpoint API. The previous one was introduced recently in NetBSD-current and remove its spurs without any backward-compatibility. Design choices for Debug Register accessors: - exec() (TRAP_EXEC event) must remove debug registers from LWP - debug registers are only per-LWP, not per-process globally - debug registers must not be inherited after (v)forking a process - debug registers must not be inherited after forking a thread - a debugger is responsible to set global watchpoints/breakpoints with the debug registers, to achieve this PTRACE_LWP_CREATE/PTRACE_LWP_EXIT event monitoring function is designed to be used - debug register traps must generate SIGTRAP with si_code TRAP_DBREG - debugger is responsible to retrieve debug register state to distinguish the exact debug register trap (DR6 is Status Register on x86) - kernel must not remove debug register traps after triggering a trap event a debugger is responsible to detach this trap with appropriate PT_SETDBREGS call (DR7 is Control Register on x86) - debug registers must not be exposed in mcontext - userland must not be allowed to set a trap on the kernel Implementation notes on i386 and amd64: - the initial state of debug register is retrieved on boot and this value is stored in a local copy (initdbregs), this value is used to initialize dbreg context after PT_GETDBREGS - struct dbregs is stored in pcb as a pointer and by default not initialized - reserved registers (DR4-DR5, DR9-DR15) are ignored Further ideas: - restrict this interface with securelevel Tested on real hardware i386 (Intel Pentium IV) and amd64 (Intel i7). This commit enables 390 debug register ATF tests in kernel/arch/x86. All tests are passing. This commit does not cover netbsd32 compat code. Currently other interface PT_GET_SIGINFO/PT_SET_SIGINFO is required in netbsd32 compat code in order to validate reliably PT_GETDBREGS/PT_SETDBREGS. This implementation does not cover FreeBSD specific defines in their <x86/reg.h>: DBREG_DR7_LOCAL_ENABLE, DBREG_DR7_GLOBAL_ENABLE, DBREG_DR7_LEN_1 etc. These values tend to be reinvented by each tracer on its own. GNU Debugger (GDB) works with NetBSD debug registers after adding this patch: --- gdb/amd64bsd-nat.c.orig 2016-02-10 03:19:39.000000000 +0000 +++ gdb/amd64bsd-nat.c @@ -167,6 +167,10 @@ amd64bsd_target (void) #ifdef HAVE_PT_GETDBREGS +#ifndef DBREG_DRX +#define DBREG_DRX(d,x) ((d)->dr[(x)]) +#endif + static unsigned long amd64bsd_dr_get (ptid_t ptid, int regnum) { Another reason to stop introducing unpopular defines covering machine specific register macros is that these value varies across generations of the same CPU family. GDB demo: (gdb) c Continuing. Watchpoint 2: traceme Old value = 0 New value = 16 main (argc=1, argv=0x7f7fff79fe30) at test.c:8 8 printf("traceme=%d\n", traceme); (Currently the GDB interface is not reliable due to NetBSD support bugs) Sponsored by <The NetBSD Foundation>
1.13	16-Jan-2017	kamil	Allow to mix single-step with hardware assisted watchpoints on i386 This case needs new handling in trap recognition. Sponsored by <The NetBSD Foundation>
1.12	15-Dec-2016	kamil	branches: 1.12.2; Add support for hardware assisted watchpoints/breakpoints API in ptrace(2) Add new ptrace(2) calls: - PT_COUNT_WATCHPOINTS - count the number of available hardware watchpoints - PT_READ_WATCHPOINT - read struct ptrace_watchpoint from the kernel state - PT_WRITE_WATCHPOINT - write new struct ptrace_watchpoint state, this includes enabling and disabling watchpoints The ptrace_watchpoint structure contains MI and MD parts: typedef struct ptrace_watchpoint { int pw_index; /* HW Watchpoint ID (count from 0) / lwpid_t pw_lwpid; / LWP described / struct mdpw pw_md; / MD fields / } ptrace_watchpoint_t; For example amd64 defines MD as follows: struct mdpw { void md_address; int md_condition; int md_length; }; These calls are protected with the __HAVE_PTRACE_WATCHPOINTS guard. Tested on amd64, initial support added for i386 and XEN. Sponsored by <The NetBSD Foundation>
1.11	28-Apr-2008	martin	branches: 1.11.44; 1.11.64; 1.11.68; Remove clause 3 and 4 from TNF licenses
1.10	09-Feb-2007	ad	branches: 1.10.44; 1.10.46; 1.10.48; Merge newlock2 to head.
1.9	15-Apr-2006	simonb	branches: 1.9.8; Remove "register", ANSIfy.
1.8	16-Feb-2006	perry	branches: 1.8.2; 1.8.4; 1.8.6; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.7	24-Dec-2005	perry	branches: 1.7.2; 1.7.4; 1.7.6; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.6	31-Oct-2003	cl	branches: 1.6.16; Reduce code duplication by adding mi_userret() in sys/userret.h containing signal posting, kernel-exit handling and sa_upcall processing. XXX the pc532, sparc, sparc64 and vax ports should have their XXX userret() code rearranged to use this.
1.5	27-Oct-2003	junyoung	Nuke __P().
1.4	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.3	17-Jan-2003	thorpej	branches: 1.3.2; Merge the nathanw_sa branch.
1.2	10-Dec-2000	mycroft	branches: 1.2.2; 1.2.4; 1.2.6; Introduce PROC_PC(), which is used to get a process's user PC. If this is defined, call addupc_intr() directly from statclock() in the system time case, using the same P_OWEUPC path if the copyin/copyout fails. Use this in i386 to remove profiling code from the normal userret() path.
1.1	09-Dec-2000	jdolecek	put userret() into header file machine/userret.h and pull it where needed
1.2.6.6	26-Sep-2002	nathanw	Change "if (l->l_flag & L_SA_UPCALL)" to "while (l->l_flag & L_SA_UPCALL)" in userret() functions or equivalent, to permit delivery of multiple upcalls in a single kernel entry. XXX It's getting crowded in here. Collapsing posting signals, upcalls, and XXX kernel-exit handling into one mechanism would be nice.
1.2.6.5	17-Dec-2001	nathanw	Call sa_upcall_userret(), not cpu_upcall().
1.2.6.4	17-Nov-2001	nathanw	Check and obey user-return exit hook (replaces test for P_WEXIT).
1.2.6.3	19-Jul-2001	nathanw	Check for P_WEXIT after taking signals, in case we were stopped in CURSIG().
1.2.6.2	09-Jul-2001	nathanw	Move upcall processing to the point of userret() rather than invocation time.
1.2.6.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.2.4.2	10-Dec-2000	mycroft	Introduce PROC_PC(), which is used to get a process's user PC. If this is defined, call addupc_intr() directly from statclock() in the system time case, using the same P_OWEUPC path if the copyin/copyout fails. Use this in i386 to remove profiling code from the normal userret() path.
1.2.4.1	10-Dec-2000	mycroft	file userret.h was added on branch sommerfeld_i386mp_1 on 2000-12-10 19:29:32 +0000
1.2.2.2	13-Dec-2000	bouyer	Sync with HEAD (for UBC fixes).
1.2.2.1	10-Dec-2000	bouyer	file userret.h was added on branch thorpej_scsipi on 2000-12-13 15:49:31 +0000
1.3.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.3.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.3.2.1	03-Aug-2004	skrll	Sync with HEAD
1.6.16.2	26-Feb-2007	yamt	sync with head.
1.6.16.1	21-Jun-2006	yamt	sync with head.
1.7.6.1	22-Apr-2006	simonb	Sync with head.
1.7.4.1	09-Sep-2006	rpaulo	sync with head
1.7.2.1	18-Feb-2006	yamt	sync with head.
1.8.6.1	24-May-2006	tron	Merge 2006-05-24 NetBSD-current into the "peter-altq" branch.
1.8.4.1	19-Apr-2006	elad	sync with head - hopefully this will work
1.8.2.1	24-May-2006	yamt	sync with head.
1.9.8.1	20-Oct-2006	ad	Do the priority adjustment in mi_userret().
1.10.48.1	16-May-2008	yamt	sync with head.
1.10.46.1	18-May-2008	yamt	sync with head.
1.10.44.1	02-Jun-2008	mjf	Sync with HEAD.
1.11.68.2	20-Mar-2017	pgoyette	Sync with HEAD
1.11.68.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.11.64.2	28-Aug-2017	skrll	Sync with HEAD
1.11.64.1	05-Feb-2017	skrll	Sync with HEAD
1.11.44.1	03-Dec-2017	jdolecek	update from HEAD
1.12.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.14.14.1	10-Jun-2019	christos	Sync with HEAD
1.14.12.1	28-Jul-2018	pgoyette	Sync with HEAD
1.19	17-Jul-2011	joerg	Retire varargs.h support. Move machine/stdarg.h logic into MI sys/stdarg.h and expect compiler to provide proper builtins, defaulting to the GCC interface. lint still has a special fallback. Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and derive va_list as required by standards.
1.18	07-Aug-2003	agc	branches: 1.18.120; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.17	12-May-2000	thorpej	branches: 1.17.6; 1.17.30; Only use __builtin_va_alist_t if using GCC 2.96 or later. Fixes pkg/10080, from Matthias Scheler.
1.16	05-May-2000	thorpej	Use __GNUC_PREREQ__().
1.15	03-May-2000	thorpej	Make varargs and stdarg work with GCC 2.96 and later.
1.14	26-Dec-1995	mycroft	branches: 1.14.30; Make the type of __builtin_va_list a long.
1.13	26-Dec-1995	mycroft	Use __builtin_va_alist.
1.12	25-Dec-1995	mycroft	Stylistic changes.
1.11	25-Dec-1995	mycroft	Update for GCC 2.7, and fix bugs.
1.10	28-Mar-1995	jtc	KERNEL -> _KERNEL
1.9	28-Jan-1995	jtc	ANSI says that <stdarg.h>'s va_end macro must expand to a void expression. For consistancy, I'm changing <varargs.h> too.
1.8	28-Dec-1994	cgd	do the right thing with strange-sized varargs objects.
1.7	27-Oct-1994	cgd	new RCS ID format.
1.6	15-Oct-1994	cgd	make <stdarg.h> a symlink, and clean up ports' stdarg.h and varargs.h files.
1.5	09-Oct-1994	mycroft	Make multiple-inclusion protection consistent.
1.4	29-Jun-1994	deraadt	branches: 1.4.2; _MACHINE_VARGS_H_
1.3	21-May-1994	cgd	update from lite
1.2	16-May-1994	cgd	USL copyright foo
1.1	27-Jan-1994	pk	branches: 1.1.2; move varargs stuff to arch/<machine>/include.
1.1.2.2	27-Jan-1994	pk	move varargs stuff to arch/<machine>/include.
1.1.2.1	27-Jan-1994	pk	file varargs.h was added on branch magnum on 1994-01-27 00:18:24 +0000
1.4.2.1	11-Oct-1994	mycroft	Update from trunk.
1.14.30.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.17.30.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.17.30.2	18-Sep-2004	skrll	Sync with HEAD.
1.17.30.1	03-Aug-2004	skrll	Sync with HEAD
1.17.6.2	12-May-2000	thorpej	Only use __builtin_va_alist_t if using GCC 2.96 or later. Fixes pkg/10080, from Matthias Scheler.
1.17.6.1	12-May-2000	thorpej	file varargs.h was added on branch sommerfeld_i386mp_1 on 2000-05-12 23:49:13 +0000
1.18.120.1	27-Aug-2011	jym	Add/remove files, like in HEAD.
1.20	12-Aug-2017	maxv	Remove vm86. Pass 4.
1.19	21-Nov-2009	rmind	branches: 1.19.22; 1.19.40; Use lwp_getpcb() on x86 MD code, clean from struct user usage.
1.18	16-Mar-2009	cegger	ansify function definitions
1.17	14-Nov-2008	ad	branches: 1.17.4; Remove some compat ifdefs that aren't really needed.
1.16	28-Apr-2008	martin	branches: 1.16.6; 1.16.8; Remove clause 3 and 4 from TNF licenses
1.15	16-Apr-2007	ad	branches: 1.15.32; 1.15.34; 1.15.36; Share the sysarch stuff between the x86 ports. PR kern/36046.
1.14	16-Feb-2006	perry	branches: 1.14.20; 1.14.24; 1.14.26; Change "inline" back to "__inline" in .h files -- C99 is still too new, and some apps compile things in C89 mode. C89 keywords stay. As per core@.
1.13	24-Dec-2005	perry	branches: 1.13.2; 1.13.4; 1.13.6; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.12	27-Oct-2003	junyoung	branches: 1.12.16; Nuke __P().
1.11	11-Sep-2003	christos	make vm86 use __gregset_t
1.10	17-Jan-2003	thorpej	branches: 1.10.2; Merge the nathanw_sa branch.
1.9	09-Oct-1997	jtc	branches: 1.9.30; Fix tipo inherited from old version of TNF copyright template.
1.8	03-May-1996	christos	branches: 1.8.12; Add missing prototypes.
1.7	30-Apr-1996	mycroft	Remove BIOSSEG. Don't mess with the flags when entering a software interrupt. Use IF, not VIF. NOTE: These changes break dosemu, but are required for proper emulation.
1.6	18-Apr-1996	mycroft	Make some of the flag handling clearer. No functional change.
1.5	12-Apr-1996	mycroft	Use the flag mask in a somewhat more sane fashion.
1.4	11-Apr-1996	mycroft	Actually keep track of the naughty PSL bits. Use VIF, rather than IF, by default, to be compatible with dosemu. XXXX
1.3	11-Apr-1996	mycroft	Export [gs]et_vflags(), and always use them when saving/restoring registers in VM86 mode. Allow changing of all PSL bits in VM86 mode; some applications don't work right otherwise, and all the nasty bits are virtualized anyway. Make sure PSL_VIF, PSL_VIP, and PSL_VM don't show up in the user's view of the PSL.
1.2	27-Feb-1996	jtc	Copyright assigned to the NetBSD Foundation
1.1	08-Jan-1996	mycroft	VM86 support, by John Kohl, touched up a bit by me.
1.8.12.1	14-Oct-1997	thorpej	Update marc-pcmcia branch from trunk.
1.9.30.1	05-Mar-2001	nathanw	Initial commit of scheduler activations and lightweight process support.
1.10.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.10.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.10.2.1	03-Aug-2004	skrll	Sync with HEAD
1.12.16.1	03-Sep-2007	yamt	sync with head.
1.13.6.1	22-Apr-2006	simonb	Sync with head.
1.13.4.1	09-Sep-2006	rpaulo	sync with head
1.13.2.1	18-Feb-2006	yamt	sync with head.
1.14.26.1	11-Jul-2007	mjf	Sync with head.
1.14.24.1	27-May-2007	ad	Sync with head.
1.14.20.1	07-May-2007	yamt	sync with head.
1.15.36.3	11-Mar-2010	yamt	sync with head
1.15.36.2	04-May-2009	yamt	sync with head.
1.15.36.1	16-May-2008	yamt	sync with head.
1.15.34.1	18-May-2008	yamt	sync with head.
1.15.32.2	17-Jan-2009	mjf	Sync with HEAD.
1.15.32.1	02-Jun-2008	mjf	Sync with HEAD.
1.16.8.2	28-Apr-2009	skrll	Sync with HEAD.
1.16.8.1	19-Jan-2009	skrll	Sync with HEAD.
1.16.6.1	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.17.4.3	28-Mar-2011	jym	Cure sync hiccups. Code with compile errors is not really useful, heh.
1.17.4.2	01-Nov-2009	jym	Sync with HEAD.
1.17.4.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.19.40.1	28-Aug-2017	skrll	Sync with HEAD
1.19.22.1	03-Dec-2017	jdolecek	update from HEAD
1.88	21-Aug-2022	riastradh	i386/vmparam.h: Need some constants from i386/pte.h. These are used by VM_MIN/MAX_KERNEL_ADDRESS. pte.h is small and likely stable enough that it's not worthwhile to migrate them to vmparam.h instead.
1.87	20-Aug-2022	riastradh	x86: Split most of pmap.h into pmap_private.h or vmparam.h. This way pmap.h only contains the MD definition of the MI pmap(9) API, which loads of things in the kernel rely on, so changing x86 pmap internals no longer requires recompiling the entire kernel every time. Callers needing these internals must now use machine/pmap_private.h. Note: This is not x86/pmap_private.h because it contains three parts: 1. CPU-specific (different for i386/amd64) definitions used by... 2. common definitions, including Xenisms like xpmap_ptetomach, further used by... 3. more CPU-specific inlines for pmap_pte_* operations So {amd64,i386}/pmap_private.h defines 1, includes x86/pmap_private.h for 2, and then defines 3. Maybe we should split that out into a new pmap_pte.h to reduce this trouble. No functional change intended, other than that some .c files must include machine/pmap_private.h when previously uvm/uvm_pmap.h polluted the namespace with pmap internals. Note: This migrates part of i386/pmap.h into i386/vmparam.h -- specifically the parts that are needed for several constants defined in vmparam.h: VM_MAXUSER_ADDRESS VM_MAX_ADDRESS VM_MAX_KERNEL_ADDRESS VM_MIN_KERNEL_ADDRESS Since i386 needs PDP_SIZE in vmparam.h, I added it there on amd64 too, just to keep things parallel.
1.86	11-Feb-2019	cherry	We reorganise definitions for XEN source support as follows: XEN - common sources required for baseline XEN support. XENPV - sources required for support of XEN in PV mode. XENPVHVM - sources required for support for XEN in HVM mode. XENPVH - sources required for support for XEN in PVH mode.
1.85	23-Jun-2017	joerg	branches: 1.85.6; Recommit exec_subr.c revision 1.79: Always include a 1MB guard area beyond the end of stack. While ASLR will normally create a guard area as well, this provides a deterministic area for all binaries. Mitigates the rest of CVE-2017-1000374 and CVE-2017-1000375 from Qualys. Additionally, change VM_DEFAULT_ADDRESS_TOPDOWN to include user_stack_guard_size in the size reservation.
1.84	11-Feb-2017	maxv	branches: 1.84.6; Remove VM_MAX_KERNEL_BUF (unused). Looks like several other ports could do the same.
1.83	19-Nov-2016	maxv	branches: 1.83.2; Put a one-page redzone between userland and the PTE space on amd64 and i386. The PTE space is a critical region that maps the page tree, and bugs have been found in both amd64 and i386 where the kernel would wrongly overflow userland data on this area. This kind of bug is terrible, since it allows userland to overwrite some entries of the page tree, which makes it easy to patch the kernel text and get ring0 privileges.
1.82	07-Aug-2016	dholland	Remove unused <sys/tree.h>.
1.81	24-Jul-2014	riastradh	branches: 1.81.4; 1.81.8; Add a FIRST1G page freelist to x86, for old graphics devices.
1.80	12-Jun-2014	riastradh	Tweak x86 page freelists and add x86_select_freelist. - Add 4G freelist to i386 -- there may be higher addresses if PAE. - Add 64G and 1T freelists to amd64. - Simplify freelist setup code and condense it into a table. - Add x86_select_freelist to get a freelist guaranteed to yield addresses no greater than a prescribed maximum address. x86_select_freelist takes a uint64_t, not a paddr_t or bus_addr_t, so that you can pass in, e.g., a 36-bit maximum address without needing to write conditionals for i386/PAE. No objections on port-x86: https://mail-index.netbsd.org/port-i386/2014/05/21/msg003277.html https://mail-index.netbsd.org/port-amd64/2014/05/21/msg002062.html
1.79	28-Jan-2014	christos	branches: 1.79.2; only limit bottom-up allocations, put back 3G limit on datasize.
1.78	28-Jan-2014	christos	Cut down MAXDSIZE from 3G to 2.5G otherwise bottomup allocation ends up supplying an out of bounds hint for sigcode (c001e000 > bf000000). Makes a.out binaries work again.
1.77	25-Jan-2014	christos	delete VM_DEFAULT_ADDRESS; some of those should be GC'ed because they match the default definition.
1.76	13-Nov-2012	chs	branches: 1.76.2; bump VM_PHYSSEG_MAX to 32, we've seen a system where 16 wasn't enough.
1.75	15-Aug-2012	sborrill	branches: 1.75.2; Bump VM_PHYSSEG_MAX to 16 from 10. Modern IBM hardware requires VM_PHYSSEG_MAX to be turned up to 11 to avoid an early panic.
1.74	07-May-2012	joerg	Raise per-image text size limit to 256MB. 64MB has seen already, so provide some margin of grows.
1.73	19-Feb-2012	cherry	Removing remaining references to the alternate PTE space. Modify documentation appropriately
1.72	14-Nov-2010	uebayasi	branches: 1.72.8; 1.72.12; 1.72.14; Move struct vm_page_md definition from vmparam.h to pmap.h, because it's used only by pmap. vmparam.h has definitions for wider audience. All GENERIC kernels build tested, except ia64. powerpc/include/booke/vmparam.h has one too, but it has no pmap.h, so it's left as is.
1.71	06-Nov-2010	uebayasi	Remove incomplete, never worked dynamic run-time memory registration (uvm_page_physload(9)). This functionality will be re-added later.
1.70	06-Mar-2009	joerg	branches: 1.70.2; 1.70.4; Remove SHMMAXPGS from all kernel configs. Dynamically compute the initial limit as 1/4 of the physical memory. Ensure the limit is at least 1024 pages, the old default on most platforms.
1.69	13-Dec-2008	pooka	branches: 1.69.2; _VMPARAM_H_ -> _$MACHINE_VMPARAM_H_
1.68	23-Jan-2008	bouyer	branches: 1.68.6; 1.68.10; 1.68.18; Merge the bouyer-xeni386 branch. This brings in PAE support to NetBSD xeni386 (domU only). PAE support is enabled by 'options PAE', see the new XEN3PAE_DOMU and INSTALL_XEN3PAE_DOMU kernel config files. See the comments in arch/i386/include/{pte.h,pmap.h} to see how it works. In short, we still handle it as a 2-level MMU, with the second level page directory being 4 pages in size. pmap switching is done by switching the L2 pages in the L3 entries, instead of loading %cr3. This is almost required by Xen, which handle the last L2 page (the one mapping 0xc0000000 - 0xffffffff) in a very special way. But this approach should also work for native PAE support if ever supported (in fact, the pmap should almost suport native PAE, what's missing is bootstrap code in locore.S).
1.67	20-Jan-2008	yamt	- rewrite P->V tracking. - use a hash rather than SPLAY trees. SPLAY tree is a wrong algorithm to use here. will be revisited if it slows down anything other than micro-benchmarks. - optimize the single mapping case (it's a common case) by embedding an entry into mdpage. - don't keep a pmap pointer as it can be obtained from ptp. (discussed on port-i386 some years ago.) ideally, a single paddr_t should be enough to describe a pte. but it needs some more thoughts as it can increase computational costs. - pmap_enter: simplify and fix races with pmap_sync_pv. - don't bother to lock pm_obj[i] where i > 0, unless DIAGNOSTIC. - kill mp_link to save space. - add many KASSERTs.
1.66	11-Jan-2008	bouyer	Merge the bouyer-xeni386 branch to head, at tag bouyer-xeni386-merge1 (the branch is still active and will see i386PAE support developement). Sumary of changes: - switch xeni386 to the x86/x86/pmap.c, and the xen/x86/x86_xpmap.c pmap bootstrap. - merge back most of xen/i386/ to i386/i386 - change the build to reduce diffs between i386 and amd64 in file locations - remove include files that were identical to the i386/amd64 counterparts, the build will find them via the xen-ma/machine link.
1.65	06-Jan-2008	ad	#include <sys/mutex.h>
1.64	18-Oct-2007	yamt	branches: 1.64.2; 1.64.8; merge yamt-x86pmap branch. - reduce differences between amd64 and i386. notably, share pmap.c between them. it makes several i386 pmap improvements available to amd64, including tlb shootdown reduction and bug fixes from Stephan Uphoff. - implement deferred pmap switching for amd64. - remove LARGEPAGES option. always use large pages if available. also, make it work on amd64.
1.63	17-Oct-2007	garbled	Merge the ppcoea-renovation branch to HEAD. This branch was a major cleanup and rototill of many of the various OEA cpu based PPC ports that focused on sharing as much code as possible between the various ports to eliminate near-identical copies of files in every tree. Additionally there is a new PIC system that unifies the interface to interrupt code for all different OEA ppc arches. The work for this branch was done by a variety of people, too long to list here. TODO: bebox still needs work to complete the transition to -renovation. ofppc still needs a bunch of work, which I will be looking at. ev64260 still needs to be renovated amigappc was not attempted. NOTES: pmppc was removed as an arch, and moved to a evbppc target.
1.62	29-Aug-2007	ad	branches: 1.62.2; 1.62.4; Merge most x86 changes from the vmlocking branch, except the threaded soft interrupt stuff. This is mostly comprised of changes to the pmap modules to work on multiprocessor systems without kernel_lock, and changes to speed up tlb shootdowns.
1.61	27-Sep-2006	cube	branches: 1.61.8; 1.61.16; 1.61.22; 1.61.26; 1.61.28; This is again that time of the millenium where we have to crank up a few static limits to meet modern bloat requirements. VM_PHYSSEG_MAX needs it to run on Intel's D946GZIS motherboard, as reported by rix on #NetBSD-code on freenode. This has a consequence on the initial number of possible extent allocations for iomem_ex, so increase that value too. While there, clarify the action to be taken when VM_PHYSSEG_MAX is maxed out. Do that on both amd64 and i386 because the causes, the effects and the code are mostly the same.
1.60	20-Sep-2005	christos	branches: 1.60.26; 1.60.28; - Bump default data size from 128M to 256M, allows the java interpreter to start up without unlimit. - Bump max data size from 2G to 3G. The actual space we are allowed to allocate is somewhere between 2G and 3G, so trying to allocate above that will fail. - Bump max stack size from 32M to 64M. Approved by fvdl
1.59	17-Jul-2005	christos	Increase the default datasize limit to 2GB.
1.58	10-Jan-2005	mycroft	branches: 1.58.10; Now that countless UVM bugs have been fixed, enable "topdown" memory allocation by default.
1.57	12-May-2004	yamt	add a comment on mp_attrs.
1.56	23-Oct-2003	chs	rename pv_next to pv_node. use SPLAY_RIGHT() instead of expanding it inline.
1.55	23-Oct-2003	provos	changed linked list in pmap_remove_pv to a splay tree; approved: fvdl@
1.54	24-Aug-2003	chs	add support for non-executable mappings (where the hardware allows this) and make the stack and heap non-executable by default. the changes fall into two basic catagories: - pmap and trap-handler changes. these are all MD: = alpha: we already track per-page execute permission with the (software) PG_EXEC bit, so just have the trap handler pay attention to it. = i386: use a new GDT segment for %cs for processes that have no executable mappings above a certain threshold (currently the bottom of the stack). track per-page execute permission with the last unused PTE bit. = powerpc/ibm4xx: just use the hardware exec bit. = powerpc/oea: we already track per-page exec bits, but the hardware only implements non-exec mappings at the segment level. so track the number of executable mappings in each segment and turn on the no-exec segment bit iff the count is 0. adjust the trap handler to deal. = sparc (sun4m): fix our use of the hardware protection bits. fix the trap handler to recognize text faults. = sparc64: split the existing unified TSB into data and instruction TSBs, and only load TTEs into the appropriate TSB(s) for the permissions. fix the trap handler to check for execute permission. = not yet implemented: amd64, hppa, sh5 - changes in all the emulations that put a signal trampoline on the stack. instead, we now put the trampoline into a uvm_aobj and map that into the process separately. originally from openbsd, adapted for netbsd by me.
1.53	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22364, verified by myself.
1.52	22-Jul-2003	yamt	move pv_head and page attributes into md part of vm_page structure instead of having separate arrays for them. (i.e. use __HAVE_VM_PAGE_MD instead of __HAVE_PMAP_PHYSSEG.)
1.51	02-Apr-2003	thorpej	branches: 1.51.2; Use PAGE_SIZE rather than NBPG.
1.50	16-Mar-2003	cjep	kernal -> kernel
1.49	23-Feb-2003	atatat	#include opt_uvm.h in machine/vmparam.h (for those ports offering the topdown option) so that including it directly before including uvm/uvm_param.h (or uvm/uvm_extern.h which includes uvm/uvm_param.h) and attempting to use topdown won't result in a compiler error. Problem noted in private email.
1.48	20-Feb-2003	atatat	Introduce "top down" memory management for mmap()ed allocations. This means that the dynamic linker gets mapped in at the top of available user virtual memory (typically just below the stack), shared libraries get mapped downwards from that point, and calls to mmap() that don't specify a preferred address will get mapped in below those. This means that the heap and the mmap()ed allocations will grow towards each other, allowing one or the other to grow larger than before. Previously, the heap was limited to MAXDSIZ by the placement of the dynamic linker (and the process's rlimits) and the space available to mmap was hobbled by this reservation. This is currently only enabled via an option for the i386 platform (though other platforms are expected to follow). Add "options USE_TOPDOWN_VM" to your kernel config file, rerun config, and rebuild your kernel to take advantage of this. Note that the pmap_prefer() interface has not yet been modified to play nicely with this, so those platforms require a bit more work (most notably the sparc) before they can use this new memory arrangement. This change also introduces a VM_DEFAULT_ADDRESS() macro that picks the appropriate default address based on the size of the allocation or the size of the process's text segment accordingly. Several drivers and the SYSV SHM address assignment were changed to use this instead of each one picking their own "default".
1.47	29-Jan-2003	drochner	give some VM which was reserved for the UAREA but not used anymore back to the user and adjust some comments (which were not accurate anyway since NOREDZONE) binary compatibility note: changing VM_MAXUSER_ADDRESS might influence some sanity check in kvm_proc, where arguments on the stack are dealt with, but it was a variable anyway and noone did care...
1.46	10-Dec-2002	thorpej	Use __LDPGSZ (which must be == USRTEXT) as the text address for a.out executables, and eliminate the USRTEXT constant, which was only used by the a.out exec code.
1.45	15-Nov-2001	soren	branches: 1.45.2; MAXSLP is defined to be a machine-independent scheduling parameter, so move it into sys/param.h.
1.44	20-Oct-2001	fvdl	Until early last year, the i386 port had 256M of kvm, and the max. amount of kvm used for buffers was set at 70%, some 188M. Then the total amount of kvm became 1G, and the amount for buffers thus became some 716M. This is really too much, and some device drivers want to map quite a bit of kvm these days. So, cap it at 384M, which gives each buffer a little over 8k (the default FFS blocksize) physical in an 1G physram configuration.
1.43	25-Jun-2001	thorpej	branches: 1.43.4; Compute the VM_* constants directly, rather than hard-coding them (with the formulas in comments).
1.42	01-May-2001	thorpej	Per discussion w/ chuck and chuck, restructure the md page stuff to use a structure called "vm_page_md", and use __HAVE_VM_PAGE_MD and __HAVE_PMAP_PHYSSEG.
1.41	29-Apr-2001	thorpej	Add a VM_MDPAGE_MEMBERS macro that defines pmap-specific data for each vm_page structure. Add a VM_MDPAGE_INIT() macro to init this data when pages are initialized by UVM. These macros are mandatory, but ports may #define them to nothing if they are not needed/used. This deprecates struct pmap_physseg. As a transitional measure, allow a port to #define PMAP_PHYSSEG so that it can continue to use it until its pmap is converted to use VM_MDPAGE_MEMBERS. Use all this stuff to eliminate a lot of extra work in the Alpha pmap module (it's smaller and faster now). Changes to other pmap modules will follow.
1.40	03-Jan-2001	fvdl	branches: 1.40.4; Bump shm max pages value to a more realistic one.
1.39	14-Nov-2000	thorpej	The i386 page size is fixed -- override PAGE_SIZE, et al, so that they are compile-time constants.
1.38	09-Nov-2000	kim	BIOS memory loading messages can be enabled with DEBUG_MEMLOAD. Increase VM_PHYSSEG_MAX from 3 to 5 to avoid a panic. Suggest increasing VM_PHYSSEG_MAX in the panic message.
1.37	18-Jun-2000	castor	branches: 1.37.2; Fix integer overflow in calculation of VM_MAX_KERNEL_BUF.
1.36	11-Feb-2000	thorpej	branches: 1.36.4; Update for the NKMEMPAGES changes.
1.35	11-Feb-2000	thorpej	Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1G of virtual address space, leaving userland with 3G, and update comments to match the new reality. We knew we were going to have to bite this bullet eventually, and there are a couple of outstanding PRs related to this issue (9389 and 9313). Complete solution to those PRs is going to involve some sort of run-time decision on how large kmem_map should be, as well as changing some data structure allocation strategies in UVM. However, this change will at least allow the PR submitter to simply throw resources at the problem.
1.34	26-Jan-2000	tsutsui	Remove obsoleted macros.
1.33	04-Dec-1999	ragge	CL* discarding.
1.32	17-Jun-1999	thorpej	branches: 1.32.2; 1.32.8; PMAP_NEW is no longer optional on the i386; the old pmap's page table allocation strategy no longer works at all. Move pmap.new.* to pmap.. To read the revision history of PMAP_NEW up until this merge, use cvs rlog of the old pmap.new. files.
1.31	26-Apr-1999	thorpej	Garbage-collect the VM_MBUF_SIZE constant. Instead, use the size (nmbclusters * mclbytes), so that the right amount of KVA space is allocated if those variables are patched.
1.30	16-Jan-1999	chuck	branches: 1.30.2; 1.30.4; remove non MNN code
1.29	13-Aug-1998	thorpej	vm_offset_t -> {vaddr_t,paddr_t}, vm_size_t -> vsize_t
1.28	08-Jul-1998	thorpej	Define 2 free lists: a default, and a "first16", which is where the first 16M of RAM goes (in an effort to protect ISA DMA-safe memory).
1.27	04-May-1998	thorpej	Whitespace.
1.26	10-Feb-1998	mrg	- add defopt's for UVM, UVMHIST and PMAP_NEW. - remove unnecessary UVMHIST_DECL's.
1.25	06-Feb-1998	mrg	add the i386 MD portions for UVM.
1.24	15-Jan-1998	thorpej	Just commit to using MACHINE_NEW_NONCONTIG.
1.23	13-Jan-1998	mrg	this is the i386 MD portion of the MACHINE_NEW_NONCONTIG code, as written by chuck cranor. thanks to mycroft for helping me find the one little line of code i accidentally deleted while merging it. this is not enabled by default. `options MACHINE_NEW_NONCONTIG' will use this code. eventually, this should go into <machine/vmparam.h> insteaed of MACHINE_NONCONTIG.
1.22	20-Oct-1997	fvdl	Fix typo in VM_MAX_ADDRESS.
1.21	20-Oct-1997	fvdl	Change various constants that depend on the kernel VM base address. Lower it to 0xf0000000 to give it more breathing space.
1.20	12-Jul-1997	perry	update comment from 1981 on memory and disk prices -- pr-2754 from Curt Sampson
1.19	12-Jun-1997	mrg	bring mrg-vm-swap2 onto mainilne.
1.18	05-Mar-1997	fvdl	branches: 1.18.2; Bring limits up to 'modern' standards.
1.17	09-Dec-1996	fvdl	branches: 1.17.4; 1.17.6; Include VM_MAX_KERNEL_BUF constant to put a cap on buffer cache KVM usage. Override via config file possible. XXX this is a stopgap measure.
1.16	15-Nov-1996	fvdl	Add MACHINE_NONCONTIG as a define.
1.15	27-Oct-1994	cgd	new RCS ID format.
1.14	16-Jan-1994	cgd	branches: 1.14.2; SHMMAXPGS now pegged at 1024, where it belongs
1.13	05-Jan-1994	mycroft	Move kernel to 0xf8000000.
1.12	04-Jan-1994	mycroft	Fix VM_{MIN,MAX}_KERNEL_ADDRESS.
1.11	03-Jan-1994	mycroft	Garbage collection.
1.10	22-Dec-1993	cgd	bump MAXSSIZ back down to 8M (from 256M, where charles put it). execve() currently allocated MAXSSIZ of address space (AND PAGE TABLES) for the stack. until that's changed, and the stack-handling code is updated to match, MAXSSIZ==256M will make execve() significantly slower than MAXSSIZ==8M.
1.9	20-Dec-1993	mycroft	Cleanup and garbage collection; nothing significant. From magnum branch.
1.8	15-Nov-1993	cgd	cut down SHMMAXPGS (temporary)
1.7	14-Nov-1993	cgd	Add the System V message queue and semaphore facilities. Implemented by Daniel Boulet <danny@BouletFermat.ab.ca>
1.6	04-Sep-1993	cgd	branches: 1.6.2; change maximum stack size to 8M.
1.5	04-Sep-1993	cgd	decouple MAXDSIZ and MAXSSIZ. MAXSSIZ now is 2M; shouldn't need more, and saves time in exec.
1.4	13-Jul-1993	cgd	up default data size limit to 16M, like on freefall, so everything will compile properly with gcc2.
1.3	18-Jun-1993	cgd	convert magic numbers to network byte order, and attendent changes
1.2	22-May-1993	cgd	add rcsids to everything and clean up headers
1.1	21-Mar-1993	cgd	branches: 1.1.1; Initial revision
1.1.1.1	21-Mar-1993	cgd	initial import of 386bsd-0.1 sources
1.6.2.4	26-Oct-1993	mycroft	Add a default SHMMAXPGS.
1.6.2.3	15-Oct-1993	mycroft	I386_CR3PAT could not justify its existence.
1.6.2.2	15-Oct-1993	mycroft	Major cleanup of include files and constants.
1.6.2.1	10-Oct-1993	mycroft	Remove old and non-working macros.
1.14.2.2	16-Jan-1994	cgd	SHMMAXPGS now pegged at 1024, where it belongs
1.14.2.1	16-Jan-1994	cgd	file vmparam.h was added on branch magnum on 1994-01-16 01:36:08 +0000
1.17.6.1	12-Mar-1997	is	Merge in Changes from The Trunk.
1.17.4.1	12-Feb-1997	mrg	initial work for dynamic swap additions.
1.18.2.1	04-May-1997	mrg	re-merge mrg-vm-swap into -current, and call it mrg-vm-swap2.
1.30.4.2	04-Jul-1999	chs	add UBC parameters.
1.30.4.1	21-Jun-1999	thorpej	Sync w/ -current.
1.30.2.1	19-Feb-2000	he	Apply patch (requested by thorpej): Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1GB of virtual address space, leaving userland with 3GB, and update comments to match the new reality.
1.32.8.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.32.2.3	05-Jan-2001	bouyer	Sync with HEAD
1.32.2.2	22-Nov-2000	bouyer	Sync with HEAD.
1.32.2.1	20-Nov-2000	bouyer	Update thorpej_scsipi to -current as of a month ago A i386 GENERIC kernel compiles without the siop, ahc and bha drivers (will be updated later). i386 IDE/ATAPI and ncr work, as well as sparc/esp_sbus. alpha should work as well (untested yet). siop, ahc and bha will be updated once I've updated the branch to current -current, as well as machine-dependant code.
1.36.4.1	22-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.37.2.1	30-Mar-2001	he	Pull up revision 1.40 (requested by fvdl): Add some required Linux emulation bits to support the Linux version of VMware.
1.40.4.5	11-Dec-2002	thorpej	Sync with HEAD.
1.40.4.4	08-Jan-2002	nathanw	Catch up to -current.
1.40.4.3	22-Oct-2001	nathanw	Catch up to -current.
1.40.4.2	24-Aug-2001	nathanw	Catch up with -current.
1.40.4.1	21-Jun-2001	nathanw	Catch up to -current.
1.43.4.1	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.45.2.2	15-Nov-2001	soren	MAXSLP is defined to be a machine-independent scheduling parameter, so move it into sys/param.h.
1.45.2.1	15-Nov-2001	soren	file vmparam.h was added on branch sommerfeld_i386mp_1 on 2001-11-15 18:06:15 +0000
1.51.2.5	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.51.2.4	17-Jan-2005	skrll	Sync with HEAD.
1.51.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.51.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.51.2.1	03-Aug-2004	skrll	Sync with HEAD
1.58.10.6	04-Feb-2008	yamt	sync with head.
1.58.10.5	21-Jan-2008	yamt	sync with head
1.58.10.4	27-Oct-2007	yamt	sync with head.
1.58.10.3	03-Sep-2007	yamt	sync with head.
1.58.10.2	30-Dec-2006	yamt	sync with head.
1.58.10.1	21-Jun-2006	yamt	sync with head.
1.60.28.1	22-Oct-2006	yamt	sync with head
1.60.26.1	18-Nov-2006	ad	Sync with head.
1.61.28.3	23-Mar-2008	matt	sync with HEAD
1.61.28.2	09-Jan-2008	matt	sync with HEAD
1.61.28.1	06-Nov-2007	matt	sync with HEAD
1.61.26.2	26-Oct-2007	joerg	Sync with HEAD. Follow the merge of pmap.c on i386 and amd64 and move pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup code to restore CR4 before jumping back into kernel space as the large page option might cover that.
1.61.26.1	03-Sep-2007	jmcneill	Sync with HEAD.
1.61.22.1	03-Sep-2007	skrll	Sync with HEAD.
1.61.16.1	03-Oct-2007	garbled	Sync with HEAD
1.61.8.3	23-Oct-2007	ad	Sync with head.
1.61.8.2	05-Apr-2007	ad	Fix the pmap locking.
1.61.8.1	13-Mar-2007	ad	Pull in the initial set of changes for the vmlocking branch.
1.62.4.1	25-Oct-2007	bouyer	Sync with HEAD.
1.62.2.1	23-Sep-2007	yamt	- sync with amd64. - pmap_alloc_level: remove fast call special case. - tweak some printf to work for both of LP64 and LP32. - remove machdep.nkpde sysctl. - remove automatic adjustment of nkpde for now. will revisit later.
1.64.8.5	20-Jan-2008	bouyer	Sync with HEAD
1.64.8.4	13-Jan-2008	bouyer	Reorganize so that there are less #ifdef XEN
1.64.8.3	10-Jan-2008	bouyer	Change xeni386 build to match xenamd64: - machine/ points to xen/include/i386 - remove includes that were identical to the i386/include/ ones; the build will find them though the xen-ma/machine link. - include xen files using xen/ not machine/
1.64.8.2	08-Jan-2008	bouyer	Sync with HEAD
1.64.8.1	11-Dec-2007	bouyer	Switch i386 to x86/x86/pmap.c
1.64.2.1	18-Feb-2008	mjf	Sync with HEAD.
1.68.18.2	28-Apr-2009	skrll	Sync with HEAD.
1.68.18.1	19-Jan-2009	skrll	Sync with HEAD.
1.68.10.1	04-May-2009	yamt	sync with head.
1.68.6.1	17-Jan-2009	mjf	Sync with HEAD.
1.69.2.3	10-Jan-2011	jym	Sync with HEAD
1.69.2.2	01-Nov-2009	jym	Sync with HEAD.
1.69.2.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.70.4.1	05-Mar-2011	rmind	sync with head
1.70.2.3	16-Nov-2010	uebayasi	Sync with HEAD.
1.70.2.2	26-Apr-2010	uebayasi	Remove the unfinished code to add a memory segment after uvm_page_init(). It doesn't even compile. (In the future, we should allocate struct vm_page [] on the added memory segment for NUMA's sake.)
1.70.2.1	23-Feb-2010	uebayasi	Convert all VM_MDPAGE_INIT()'s to take struct vm_page_md * and paddr_t.
1.72.14.1	15-Aug-2012	riz	Pull up following revision(s) (requested by sborrill in ticket #501): sys/arch/amd64/include/vmparam.h: revision 1.31 sys/arch/i386/include/vmparam.h: revision 1.75 Bump VM_PHYSSEG_MAX to 16 from 10. Modern IBM hardware requires VM_PHYSSEG_MAX to be turned up to 11 to avoid an early panic. Bump VM_PHYSSEG_MAX to 16 from 10. Modern IBM hardware requires VM_PHYSSEG_MAX to be turned up to 11 to avoid an early panic.
1.72.12.2	02-Jun-2012	mrg	sync to latest -current.
1.72.12.1	24-Feb-2012	mrg	sync to -current.
1.72.8.5	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.72.8.4	16-Jan-2013	yamt	sync with (a bit old) head
1.72.8.3	30-Oct-2012	yamt	sync with head
1.72.8.2	23-May-2012	yamt	sync with head.
1.72.8.1	17-Apr-2012	yamt	sync with head
1.75.2.3	03-Dec-2017	jdolecek	update from HEAD
1.75.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.75.2.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.76.2.1	18-May-2014	rmind	sync with head
1.79.2.1	10-Aug-2014	tls	Rebase.
1.81.8.2	20-Mar-2017	pgoyette	Sync with HEAD
1.81.8.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.81.4.3	28-Aug-2017	skrll	Sync with HEAD
1.81.4.2	05-Dec-2016	skrll	Sync with HEAD
1.81.4.1	05-Oct-2016	skrll	Sync with HEAD
1.83.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.84.6.1	31-Aug-2017	bouyer	Pull up following revision(s) (requested by joerg in ticket #234): sys/arch/amd64/include/vmparam.h: revision 1.43 sys/kern/exec_subr.c: revision 1.79 lib/libpthread/pthread_int.h: revision 1.94 sys/arch/mips/include/vmparam.h: revision 1.58 sys/arch/mips/include/vmparam.h: revision 1.59 lib/libpthread/TODO: revision 1.19 sys/arch/powerpc/include/vmparam.h: revision 1.20 sys/arch/riscv/include/vmparam.h: revision 1.2 sys/arch/riscv/include/vmparam.h: revision 1.3 sys/arch/i386/include/vmparam.h: revision 1.85 tests/lib/libpthread/t_join.c: revision 1.9 sys/uvm/uvm_meter.c: revision 1.66 sys/uvm/uvm_param.h: revision 1.36 sys/kern/exec_subr.c: revision 1.80 sys/uvm/uvm_param.h: revision 1.37 sys/kern/exec_subr.c: revision 1.81 sys/kern/exec_subr.c: revision 1.82 lib/libpthread/pthread_attr_getguardsize.3: revision 1.4 lib/libpthread/pthread.c: revision 1.148 lib/libpthread/pthread_attr.c: revision 1.17 sys/arch/amd64/include/vmparam.h: revision 1.42 Always include a 1MB guard area beyond the end of stack. While ASLR will normally create a guard area as well, this provides a deterministic area for all binaries. Mitigates the rest of CVE-2017-1000374 and CVE-2017-1000375 from Qualys. Revert for the moment, creates problems on i386. Recommit exec_subr.c revision 1.79: Always include a 1MB guard area beyond the end of stack. While ASLR will normally create a guard area as well, this provides a deterministic area for all binaries. Mitigates the rest of CVE-2017-1000374 and CVE-2017-1000375 from Qualys. Additionally, change VM_DEFAULT_ADDRESS_TOPDOWN to include user_stack_guard_size in the size reservation. Update VM_DEFAULT_ADDRESS32_TOPDOWN to include guard area. Export the guard size of the main thread via vm.guard_size. Add a complementary writable sysctl for the initial guard size of threads created via pthread_create. Let the existing attribut accessors do the right thing. Raise the default guard size for threads to 64KB.
1.85.6.1	10-Jun-2019	christos	Sync with HEAD
1.2	28-Apr-2008	martin	Remove clause 3 and 4 from TNF licenses
1.1	08-May-2004	kleink	branches: 1.1.2; 1.1.98; 1.1.100; 1.1.102; Factor out W{CHAR,INT}_{MAX,MIN} into their own header file.
1.1.102.1	16-May-2008	yamt	sync with head.
1.1.100.1	18-May-2008	yamt	sync with head.
1.1.98.1	02-Jun-2008	mjf	Sync with HEAD.
1.1.2.4	21-Sep-2004	skrll	Fix the sync with head I botched.
1.1.2.3	18-Sep-2004	skrll	Sync with HEAD.
1.1.2.2	03-Aug-2004	skrll	Sync with HEAD
1.1.2.1	08-May-2004	skrll	file wchar_limits.h was added on branch ktrace-lwp on 2004-08-03 10:36:05 +0000
1.4	18-Nov-2011	jmcneill	remove Xbox support
1.3	07-Jan-2007	jmcneill	branches: 1.3.2; 1.3.6; 1.3.86; Remove nForce ethernet hack from machdep and instead call xbox_startup from cpu_startup, which applies the patch.
1.2	06-Jan-2007	jmcneill	Use the pic16lc@iic driver for setting the system LED, rebooting, and powering off the machine instead of the assembly code in pic16l.S
1.1	04-Jan-2007	jmcneill	Microsoft Xbox support, from Andrew Gillham. Still requires some bug fixes in the nfe(4) driver; works with USB ethernet. Requires Xromwell / Cromwell 2.41dev or above to boot. Tested both CD boot and etherboot. xboxfb(4) currently does not work properly with component video cables.
1.3.86.1	17-Apr-2012	yamt	sync with head
1.3.6.2	26-Feb-2007	yamt	sync with head.
1.3.6.1	07-Jan-2007	yamt	file xbox.h was added on branch yamt-lazymbuf on 2007-02-26 09:07:03 +0000
1.3.2.2	12-Jan-2007	ad	Sync with head.
1.3.2.1	07-Jan-2007	ad	file xbox.h was added on branch newlock2 on 2007-01-12 01:00:50 +0000
1.2	25-Apr-2020	bouyer	Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM guests in GENERIC. Xen support can be disabled at runtime with boot -c disable hypervisor
1.1	16-Apr-2020	bouyer	branches: 1.1.2; file hypercalls.h was initially added on branch bouyer-xenpvh.
1.1.2.2	25-Apr-2020	bouyer	Include changes in sys/arch/xen/include/ between bouyer-xenpvh-base1 and bouyer-xenpvh-base2.
1.1.2.1	16-Apr-2020	bouyer	Reorganise sources to make it possible to include Xen PVHVM support in native kernels. Among others: - move xen/include/amd64/hypercall.h to amd64/include/xen and xen/include/i386/hypercall.h to i386/include/xen - exclude some native files from the build for xenpv - add xen to "machine" config statement for amd64 and i386 - split arch/xen/conf/files.xen to arch/xen/conf/files.xen (for pv drivers) and arch/xen/conf/files.xen.pv (for full pv support) - add GENERIC_XENHVM kernel config which includes GENERIC and add Xen PV drivers.

OpenGrok