Cross Reference: /src/sys/arch/arm/arm32/pmap.c

History log of /src/sys/arch/arm/arm32/pmap.c
Revision	Date	Author	Comments
1.443	13-Apr-2024	skrll	port-arm/58135: reproducible pmap KASSERT failure for armv7 with NFS root Don't unconditionally set XN in pmap_clearbit - only set it if a mapping exists VM_PROT_EXEC is being cleared. I've simplified the #ifdefs in the patch from the PR.
1.442	13-Apr-2024	skrll	Restore a space I accidentally removed from a copyright with $NetBSD: pmap.c,v 1.396 2020/03/13 16:14:18 skrll Exp $
1.441	13-Dec-2023	rin	arm: pmap: Fix clang build without DIAGNOSTIC
1.440	12-Oct-2023	skrll	Fix non-DIAGNOSTIC builds
1.439	20-Apr-2023	skrll	Provide a shared pmap_devmap implementation and convert all pmap_devmap arrays to use DEVMAP_ENTRY{,_END}
1.438	18-Dec-2022	skrll	Appease KDASSERT / LOCKDEBUG. Tested by mlelstv.
1.437	03-May-2022	skrll	branches: 1.437.4; Catch up with aarch64 TTBR0 handling in pmap_{,de}activate_efirt and kpreempt_{en,dis}able.
1.436	09-Apr-2022	riastradh	sys: Use membar_release/acquire around reference drop. This just goes through my recent reference count membar audit and changes membar_exit to membar_release and membar_enter to membar_acquire -- this should make everything cheaper on most CPUs without hurting correctness, because membar_acquire is generally cheaper than membar_enter.
1.435	02-Apr-2022	skrll	Update to support EFI runtime outside the kernel virtual address space by creating an EFI RT pmap that can be activated / deactivated when required. Adds support for EFI RT to ARM_MMU_EXTENDED (ASID) 32-bit Arm machines. On Arm64 the usage of pmapboot_enter is reduced and the mappings are created much later in the boot process -- now in cpu_startup_hook. Backward compatiblity for KVA mapped RT from old bootaa64.efi is maintained. Adding support to other platforms should be easier as a result.
1.434	19-Mar-2022	skrll	Alight code re-organisation so it better matches the VPRINTF headings it is under. NFCI.
1.433	12-Mar-2022	riastradh	sys: Membar audit around reference count releases. If two threads are using an object that is freed when the reference count goes to zero, we need to ensure that all memory operations related to the object happen before freeing the object. Using an atomic_dec_uint_nv(&refcnt) == 0 ensures that only one thread takes responsibility for freeing, but it's not enough to ensure that the other thread's memory operations happen before the freeing. Consider: Thread A Thread B obj->foo = 42; obj->baz = 73; mumble(&obj->bar); grumble(&obj->quux); /* membar_exit(); / / membar_exit(); / atomic_dec -- not last atomic_dec -- last / membar_enter(); / KASSERT(invariant(obj->foo, obj->bar)); free_stuff(obj); The memory barriers ensure that obj->foo = 42; mumble(&obj->bar); in thread A happens before KASSERT(invariant(obj->foo, obj->bar)); free_stuff(obj); in thread B. Without them, this ordering is not guaranteed. So in general it is necessary to do membar_exit(); if (atomic_dec_uint_nv(&obj->refcnt) != 0) return; membar_enter(); to release a reference, for the `last one out hit the lights' style of reference counting. (This is in contrast to the style where one thread blocks new references and then waits under a lock for existing ones to drain with a condvar -- no membar needed thanks to mutex(9).) I searched for atomic_dec to find all these. Obviously we ought to have a better abstraction for this because there's so much copypasta. This is a stop-gap measure to fix actual bugs until we have that. It would be nice if an abstraction could gracefully handle the different styles of reference counting in use -- some years ago I drafted an API for this, but making it cover everything got a little out of hand (particularly with struct vnode::v_usecount) and I ended up setting it aside to work on psref/localcount instead for better scalability. I got bored of adding #ifdef __HAVE_ATOMIC_AS_MEMBAR everywhere, so I only put it on things that look performance-critical on 5sec review. We should really adopt membar_enter_preatomic/membar_exit_postatomic or something (except they are applicable only to atomic r/m/w, not to atomic_load/store_, making the naming annoying) and get rid of all the ifdefs.
1.432	02-Jan-2022	riastradh	arm: Remove #ifdef DIAGNOSTIC now wrong after KASSERT change. Objects in question aren't volatile here so access is flushable.
1.431	01-Jan-2022	christos	KASSERT now always uses the expression, so don't protect with DIAGNOSTIC
1.430	26-Aug-2021	skrll	Improve a comment
1.429	24-Aug-2021	skrll	Remove '\n' from UVMHIST format
1.428	23-Mar-2021	skrll	Another missing kpreempt_enable
1.427	23-Mar-2021	skrll	Re-enable kpreemption in an error path. Spotted by nat@
1.426	14-Mar-2021	skrll	branches: 1.426.2; Sprinkle kpreempt_{dis,en}able ready for when preemption gets turned on.
1.425	01-Feb-2021	skrll	"as appropos" -> "as appropriate" in comments
1.424	29-Jan-2021	skrll	More debug
1.423	24-Jan-2021	skrll	More debug
1.422	30-Oct-2020	skrll	branches: 1.422.2; Retire arm_[di]sb in favour of the isb() and dsb(sy) macro invocations.
1.421	12-Aug-2020	skrll	Whack-a-mole
1.420	11-Aug-2020	skrll	s/pmaphist/maphist/
1.419	10-Aug-2020	skrll	More UVMHIST_LOG and a new KASSERT
1.418	10-Aug-2020	skrll	In pmag_page_remove initilise pvp after taking the page lock
1.417	10-Jul-2020	skrll	Add support for KASAN on ARMv[67] Thanks to maxv for many pointers and reviews.
1.416	03-Jul-2020	skrll	1 page is enough for memhook now since the merge of the rmind-uvmplock branch and the removal of sys/arch/arm/arm32/mem.c The last users of memhook don't care about number of CPUs or page colours.
1.415	21-Jun-2020	skrll	Use howmany(). NFCI.
1.414	27-May-2020	skrll	KNF
1.413	27-May-2020	skrll	Remove duplicate #include
1.412	21-Apr-2020	skrll	There is no fast spoon^Wfast path in pmap_clearbit. PR port-arm/55186: tests crash arm pmap Tested by martin@
1.411	19-Apr-2020	ad	PR port-arm/55186: tests crash arm pmap pmap_clearbit(): take execbits into account in the fastpath.
1.410	19-Apr-2020	kre	Give UVMHIST_LOG() the 6 args it requires. Unbreak builds.
1.409	19-Apr-2020	skrll	Fix typo in UVMHIST_LOG
1.408	18-Apr-2020	skrll	Remove PMAP_DEBUG by converting to UVMHIST
1.407	17-Apr-2020	skrll	Fix build after PV locking change
1.406	17-Apr-2020	skrll	Use UVMHIST_CALLARGS
1.405	16-Apr-2020	ad	With the right timing, V->P operations could change stuff behind the back of callers working in the opposite direction - fix it. Tested by skrll@.
1.404	14-Apr-2020	skrll	Fix a comment. From ad@
1.403	13-Apr-2020	chs	slightly change and fix the semantics of pool_setwat(), pool_sethardlimit() and pool_prime() (and their pool_cache_ counterparts): - the pool_set*wat() APIs are supposed to specify thresholds for the count of free items in the pool before pool pages are automatically allocated or freed during pool_get() / pool_put(), whereas pool_sethardlimit() and pool_prime() are supposed to specify minimum and maximum numbers of total items in the pool (both free and allocated). these were somewhat conflated in the existing code, so separate them as they were intended. - change pool_prime() to take an absolute number of items to preallocate rather than an increment over whatever was done before, and wait for any memory allocations to succeed. since pool_prime() can no longer fail after this, change its return value to void and adjust all callers. - pool_setlowat() is documented as not immediately attempting to allocate any memory, but it was changed some time ago to immediately try to allocate up to the lowat level, so just fix the manpage to describe the current behaviour. - add a pool_cache_prime() to complete the API set.
1.402	29-Mar-2020	skrll	branches: 1.402.2; PTE_SYNC before TLB flush and no need to PTE_SYNC after an unmap.
1.401	29-Mar-2020	skrll	Support PMAP_FAULTINFO on arm
1.400	23-Mar-2020	skrll	Reduce #ifdefs
1.399	14-Mar-2020	ad	pmap_remove_all(): Return a boolean value to indicate the behaviour. If true, all mappings have been removed, the pmap is totally cleared out, and UVM can then avoid doing the work to call pmap_remove() for each map entry. If false, either nothing has been done, or some helpful arch-specific voodoo has taken place.
1.398	13-Mar-2020	skrll	Enhance the DIAGNOSTICs around pmap_grow_map
1.397	13-Mar-2020	skrll	Re-indent a function call. NFCI.
1.396	13-Mar-2020	skrll	Oops... remove a stray <space>
1.395	13-Mar-2020	skrll	Fixup some comments
1.394	24-Feb-2020	ad	Adjust for UVM locking changes
1.393	23-Feb-2020	skrll	type in comment
1.392	12-Feb-2020	skrll	Convert the DEBUG code in pmap_grow_map to DIAGNOSTIC
1.391	12-Feb-2020	skrll	Add a CTASSERT to pmap_grow_l2_bucket that PAGE_SIZE is a multiple of L2_TABLE_SIZE_REAL.
1.390	12-Feb-2020	skrll	Add a KASERT that we're not overwriting anything in pmap_growkernel
1.389	12-Feb-2020	skrll	Minor changes to make pmap_grow_l2_bucket look more like pmap_alloc_l2_bucket. NFCI.
1.388	05-Feb-2020	skrll	Fix the armv[67] memory attributes for uncached memory. Previously it was mapped as strongly-ordered which meant that unaligned accesses would fault. armv7_generic_bs_map now maps pages with PMAP_DEV which is treated as SO bus_dma continues to use PMAP_NOCACHE as appropriate, but this now get mapped to the correct memory attribute bits for armv[67] DEVMAP_ENTRY usees a new flag PTE_DEV. The workaround for the unaligned access faults is now removed. XXX Other armv[67] boards bus_space implementations should be checked. XXX There is scope to reduce the difference to aarch64
1.387	02-Feb-2020	skrll	More KNF
1.386	02-Feb-2020	skrll	Always call pmap_grow_map with a page aligned new VA. KASSERT that this happenes.
1.385	02-Feb-2020	skrll	Always pmap_kenter_pa the page in pmap_grow_map regardless of how we got it.
1.384	02-Feb-2020	skrll	KNF
1.383	02-Feb-2020	skrll	Fix comment
1.382	25-Jan-2020	skrll	A fix and an optimisation to pmap_l1tt_free - in the !__HAVE_MM_MD_DIRECT_MAPPED_PHYS case pass UVM_KMF_WIRED so that the mappings are removed and the KVA is released. Fixes the KASSERT seen in the automated test runs. - in the __HAVE_MM_MD_DIRECT_MAPPED_PHYS case we can work out pa much easier than caling pmap_extract.
1.381	19-Jan-2020	skrll	Fix non-ARM_MMU_EXTENDED buildx
1.380	18-Jan-2020	skrll	Use 4K pages on ARM_MMU_EXTENDED platforms (all armv[67] except RPI) by creating a new pool l1ttpl for the userland L1 translation table which needs to be 8KB and 8KB aligned. Limit the pool to maxproc and add hooks to allow the sysctl changing of maxproc to adjust the pool. This comes at a 5% performance penalty for build.sh -j8 kernel on a Tegra TK1.
1.379	18-Jan-2020	skrll	Trailing whitespace
1.378	17-Jan-2020	skrll	Fix a bug introduced in 1.271 where pmap_grow_map would no longer map the allocated page for the uvm.page_init_done == false case when PMAP_STEAL_MEMORY is not defined.
1.377	17-Jan-2020	skrll	Typo in comment
1.376	17-Jan-2020	skrll	Update PMAP_STEAL_MEMORY code to uvm_hotplug
1.375	31-Dec-2019	skrll	branches: 1.375.2; Improve a comment
1.374	25-Sep-2019	skrll	Convert a __CTASSERT into a KASSERT as L1_S_CACHE_MASK may not be a compile time constant if ARM_NMMUS > 1
1.373	23-Apr-2019	bouyer	branches: 1.373.2; Fix a deadlock between the pool and pmap codes: - cpu0 grabs the kernel lock (e.g. from a non-MPSAFE interrupt) and calls pool_get(). - cpu1 does a pool_get() on the same pool from MPSAFE code, which needs a pool_page_alloc(), which ends up in pmap_extract_coherency(). So cpu0 holds the kernel_lock and wants the pool lock. cpu1 holds the pool lock and wants the kernel_lock in pmap_extract_coherency(). The pmap code should not rely on kernel_lock. Intead make the pmap_kernel()->pm_obj_lock a IPL_VM lock and use it as pmap lock (thus dropping the pmap test pmap_{acquire,release}_pmap_lock()). This needs to be a IPL_VM because unlike user pmaps, this can be locked from interrupt context. Add a IPL_NONE lock for pmap_growkernel(). We can't use pmap_kernel()->pm_obj_lock here because pmap_grow_map() may sleep. Make pmap_lock (which may be locked with pm_obj_lock held) a IPL_VM lock in all case. reorder a few things to not call pool_get()/pool_put() (which may sleep) with pm_obj_lock held. Patch initially posted to port-arm@ on April 19, improved patch (per suggestions from Nick Hudson and Jason Thorpe) on April 21.
1.372	23-Apr-2019	bouyer	Don't try to aquire/release the pmap lock when in ddb. Avoids a deadlock when entering ddb, or on "mach cpu n" ddb command (the pmap lock may already be held by another CPU, which is halted when entering ddb). Posted to port-arm@ on April 19.
1.371	28-Oct-2018	skrll	Fix the comment near pmap_bootstrap itself as well
1.370	18-Oct-2018	skrll	Provide generic start code that assumes the MMU is off and caches are disabled as per the linux booting protocol for ARMv6 and ARMv7 boards. u-boot image type should be changed to 'linux' for correct behaviour. The new start code builds a minimal "bootstrap" L1PT with cached access disabled and uses the same table for all processors. AP startup is performed in less steps and more code is written in C. The bootstrap tables and stack are placed into an (orphaned) section "_init_memory" which is given to uvm when it is no longer used. Various kernels have been converted to use this code and tested. Some boards were provided by TNF. Thanks! The GENERIC kernel now boots on boards using the TEGRA, SUNXI and EXYNOS kernels. The GENERIC kernel will also work on RPI2 using u-boot. Thanks to martin@ and aymeric@ for testing on parallella and nanosoc respectively
1.369	14-Oct-2018	skrll	Use __nothing
1.368	03-Sep-2018	riastradh	Rename min/max -> uimin/uimax for better honesty. These functions are defined on unsigned int. The generic name min/max should not silently truncate to 32 bits on 64-bit systems. This is purely a name change -- no functional change intended. HOWEVER! Some subsystems have #define min(a, b) ((a) < (b) ? (a) : (b)) #define max(a, b) ((a) > (b) ? (a) : (b)) even though our standard name for that is MIN/MAX. Although these may invite multiple evaluation bugs, these do _not_ cause integer truncation. To avoid `fixing' these cases, I first changed the name in libkern, and then compile-tested every file where min/max occurred in order to confirm that it failed -- and thus confirm that nothing shadowed min/max -- before changing it. I have left a handful of bootloaders that are too annoying to compile-test, and some dead code: cobalt ews4800mips hp300 hppa ia64 luna68k vax acorn32/if_ie.c (not included in any kernels) macppc/if_gm.c (superseded by gem(4)) It should be easy to fix the fallout once identified -- this way of doing things fails safe, and the goal here, after all, is to _avoid_ silent integer truncations, not introduce them. Maybe one day we can reintroduce min/max as type-generic things that never silently truncate. But we should avoid doing that for a while, so that existing code has a chance to be detected by the compiler for conversion to uimin/uimax without changing the semantics until we can properly audit it all. (Who knows, maybe in some cases integer truncation is actually intended!)
1.367	28-Aug-2018	skrll	Fix arguments to align and align_offset for pmap_l2ptp_cache, i.e.align is now L2_TABLE_SIZE_REAL and align_offset is now 0. (There were reversed previously)
1.366	31-Jul-2018	skrll	Define and use VPRINTF
1.365	01-Apr-2018	ryo	branches: 1.365.2; Add initial support for ARMv8 (AARCH64) (by nisimura@ and ryo@) - sys/arch/evbarm64 is gone and integrated into sys/arch/evbarm. (by skrll@) - add support fdt. evbarm/conf/GENERIC64 fdt (bcm2837,sunxi,tegra) based generic 64bit kernel config. (by skrll@, jmcneill@)
1.364	22-Jan-2018	skrll	branches: 1.364.2; Add PMAP_WRITE_COMBINE: to the list of flags supported by pmap_kenter_pa
1.363	22-Jan-2018	skrll	Trailing whitespace
1.362	17-Jan-2018	skrll	G/C ARM32_NEW_VM_LAYOUT
1.361	01-Nov-2017	skrll	Unwrap two lines. NFC.
1.360	01-Nov-2017	skrll	Fix a comment
1.359	28-Oct-2017	pgoyette	Update the kernhist(9) kernel history code to address issues identified in PR kern/52639, as well as some general cleaning-up... (As proposed on tech-kern@ with additional changes and enhancements.) Details of changes: * All history arguments are now stored as uintmax_t values[1], both in the kernel and in the structures used for exporting the history data to userland via sysctl(9). This avoids problems on some architectures where passing a 64-bit (or larger) value to printf(3) can cause it to process the value as multiple arguments. (This can be particularly problematic when printf()'s format string is not a literal, since in that case the compiler cannot know how large each argument should be.) * Update the data structures used for exporting kernel history data to include a version number as well as the length of history arguments. * All [2] existing users of kernhist(9) have had their format strings updated. Each format specifier now includes an explicit length modifier 'j' to refer to numeric values of the size of uintmax_t. * All [2] existing users of kernhist(9) have had their format strings updated to replace uses of "%p" with "%#jx", and the pointer arguments are now cast to (uintptr_t) before being subsequently cast to (uintmax_t). This is needed to avoid compiler warnings about casting "pointer to integer of a different size." * All [2] existing users of kernhist(9) have had instances of "%s" or "%c" format strings replaced with numeric formats; several instances of mis-match between format string and argument list have been fixed. * vmstat(1) has been modified to handle the new size of arguments in the history data as exported by sysctl(9). * vmstat(1) now provides a warning message if the history requested with the -u option does not exist (previously, this condition was silently ignored, with only a single blank line being printed). * vmstat(1) now checks the version and argument length included in the data exported via sysctl(9) and exits if they do not match the values with which vmstat was built. * The kernhist(9) man-page has been updated to note the additional requirements imposed on the format strings, along with several other minor changes and enhancements. [1] It would have been possible to use an explicit length (for example, uint64_t) for the history arguments. But that would require another "rototill" of all the users in the future when we add support for an architecture that supports a larger size. Also, the printf(3) format specifiers for explicitly-sized values, such as "%"PRIu64, are much more verbose (and less aesthetically appealing, IMHO) than simply using "%ju". [2] I've tried very hard to find "all [the] existing users of kernhist(9)" but it is possible that I've missed some of them. I would be glad to update any stragglers that anyone identifies.
1.358	08-Oct-2017	flxd	Revert attempt at tracking unmanaged mappings for VIVT as it was incomplete and buggy. PR port-shark/52102 From skrll@. Tested by martin@ and me.
1.357	06-Sep-2017	skrll	Remove useless KASSERT - we can't assert !mutex_owned()
1.356	02-Sep-2017	skrll	Perform tracking of unmanaged mappings for VIVT and call vac_me_harder as appropriate. PR/52102 shark: ffs_newvnode panic when unpacking sets installing -current Thanks to Felix Deichmann for bisecting the problem and testing the fix.
1.355	02-Sep-2017	skrll	KNF
1.354	02-Sep-2017	skrll	Remove an #ifdef PMAP_CACHE_VIVT that's inside #ifdef PMAP_CACHE_VIVT already
1.353	24-Aug-2017	jmcneill	Do runtime detection of MP extensions to allow using a MULTIPROCESSOR kernel on CPUs without the MP extensions feature (like Cortex-A8).
1.352	27-Jul-2017	skrll	Macro confusion - fortunately this has no functional change.
1.351	27-Jul-2017	skrll	Remove redundant test - the compiler realised anyway as we get same binary output
1.350	10-Jul-2017	skrll	No need for '\n' in UVMHIST_LOG
1.349	24-May-2017	skrll	branches: 1.349.2; Check the "Havard TLB" maintenance operations if the "Unified TLB" maintenance opeations don't include invalidate by ASID. Some CPUs, e.g. Cortex-A8, have Havard TLBs and report ASID operations this way.
1.348	24-May-2017	skrll	Move closer to the common pmap by using the same pmap_remove_all optimisation where TLB flushes are avoided by clever ASID assignment. pmap_remove_all_complete can now be removed.
1.347	22-May-2017	skrll	Indent comment to match code it's describing
1.346	21-May-2017	skrll	KNF
1.345	17-Apr-2017	skrll	Perform icache syncs for ARM_MMU_EXTENDED as well. This helps the PT_STEP code in pr/52119 and probably other things.
1.344	25-Feb-2017	christos	fix unused.
1.343	23-Feb-2017	skrll	Fixup the compile time decisions around PMAP_{INCLUDE,NEEDS}_PTE_SYNC and fix the options for xscale boards which require the code in pmap_l2ptp_ctor marked as #ifndef PMAP_INCLUDE_PTE_SYNC. Fix the typo (pte -> opte) in this code block and consistently use opte elsewhere. PR/51990: Regression data_abort_handler: data_aborts fsr=0x406 far=0xbfffeff5 on copyout in init
1.342	23-Dec-2016	cherry	branches: 1.342.2; "Make NetBSD great again!" Introduce uvm_hotplug(9) to the kernel. Many thanks, in no particular order to: TNF, for funding the project. Chuck Silvers - for multiple API reviews and feedback. Nick Hudson - for testing on multiple architectures and bugfix patches. Everyone who helped with boot testing. KeK (http://www.kek.org.in) for hosting the primary developers.
1.341	17-Dec-2016	flxd	Fix typo "one the" and architecture where appropriate.
1.340	04-Aug-2016	skrll	provide and use 'ci' in pmap_remove_all_complete.
1.339	03-Aug-2016	skrll	Don't touch pm_onproc at all in pmap_{update,destroy} and adjust KASSERT to suit. Update to cover the PMAP_TLB_MAX > 1 case as well while I'm here.
1.338	01-Aug-2016	skrll	pm_remove_all needs handling in pmap_destroy as well as pmap_update
1.337	29-Jul-2016	skrll	more debug
1.336	29-Jul-2016	skrll	Simplify ARM_MMU_EXTENDED #ifdefs No functional change
1.335	14-Jul-2016	skrll	branches: 1.335.2; Adapt this to the recent sys/uvm/pmap/pmap_tlb.c changes. MP kernels now don't trigger KASSERTs
1.334	14-Jul-2016	skrll	Spell PMAP_TLB_NEED_SHOOTDOWN correctly
1.333	12-Jul-2016	skrll	Fix harmless (?) typo
1.332	14-Dec-2015	skrll	Delete pmap_pmaps and its only user pmap_dump_all. The list wasn't being updated in an MP-safe manner.
1.331	25-Nov-2015	skrll	In the ARM_MMU_EXTENDED case make sure the kernel mappings are marked with XN (eXecute Never) appropriately. The XN logic is inconsistent and could do with changing to XN always set and cleared for VM_PROT_EXECUTE (or the inverse) everywhere.
1.330	13-Nov-2015	skrll	Tweak the KASSERTs in pmap_update in the ARM_MMU_EXTENDED case - Lazy updating is not done on the kernel pmap. PR port-arm/50420: curcpu()->ci_pmap_cur != pm \|\| pm->pm_pai[0].pai_asid == curcpu()->ci_pmap_asid_cur" failed
1.329	13-Nov-2015	skrll	Wrap long lines.
1.328	11-Nov-2015	skrll	Support pmap_pv_track and friends
1.327	06-Nov-2015	skrll	Trailing whitespace
1.326	26-Jul-2015	matt	KASSERT->KASSERTMSG
1.325	09-Jun-2015	skrll	Use armreg_auxctl_{read,write} instead of inline asm. No functional change.
1.324	01-Jun-2015	matt	pmap_directbase is a failed experiment, nuke it.
1.323	30-May-2015	matt	Support directmapped systems with >1GB that start memory at 0x80000000.
1.322	13-May-2015	skrll	Use PDE_SYNC when syncing pdeps
1.321	11-May-2015	matt	Make sure nptes is a multiple of PAGE_SIZE / L2_S_SIZE.
1.320	13-Apr-2015	matt	Add pmap locking to pmap_kenter_pa/kremove
1.319	11-Apr-2015	skrll	#include "opt_arm_debug.h" for VERBOSE_INIT_ARM
1.318	07-Apr-2015	matt	Fix two bugs. pmap_is_cached fix for MULTIPROCESSOR (not just ASID on local cpu -> any valid ASID on any cpu). pmap_deactivate: update curcpu()->ci_pmap_cur_asid to KERNEL_PID too.
1.317	25-Feb-2015	joerg	Improve inline asm around dsb/dmb/isb: - always use volatile and mark them as memory barrier - use the common version from locore.h in all places not included from userland
1.316	10-Nov-2014	skrll	branches: 1.316.2; Update PTE_SYNC_CURRENT to add a dsb for armv7 - part of the break-before-make fix.
1.315	10-Nov-2014	skrll	Remove an XXXNH comment.
1.314	10-Nov-2014	skrll	Remove an unnecessary flush that sneaked in as part of break-before-make change.
1.313	10-Nov-2014	skrll	In pmap_fault_fixup re-instate the TLB flush for the shared L1 case that occurs for non-ARM_MMU_EXTENDED kernels. This fixes rump/rumpkern/t_sp:stress_killer on rpi which is currently non-ARM_MMU_EXTENDED
1.312	08-Nov-2014	matt	When allocing a l1page, if a page isn't available, use uvm_wait to wait for one to become available. Should fix PR/49364.
1.311	07-Nov-2014	skrll	The fixup debug message can report false positives on MULTIPROCESSOR, so disable them there.
1.310	07-Nov-2014	skrll	In pmap_enter only flush the TLB if really necessary
1.309	05-Nov-2014	skrll	Fix typo in comment
1.308	04-Nov-2014	matt	pmap_tlb_flushD is for !ARM_MMU_EXTENDED only now
1.307	29-Oct-2014	skrll	Apply a bunch of break-before-make, i.e. set PTEs to zero, flush the TLB (across all CPUs) and set new PTE value. Mostly from matt@ with some updates from me. Flush the branch predictor in pmap_update.
1.306	29-Oct-2014	skrll	In pmap_deactivate don't set TTBR with the ASID of the deactivated LWP - use KERNEL_PID instead. This is probably a no-op as TTBCR_S_PD0 should (still) be set at the time of the call to cpu_setttb.
1.305	25-Oct-2014	skrll	Remove katelib.h and references to it. {Read,Write}{Word,Byte} macros are provided in the files that still use them. Someone(tm) should convert them to bus_space(9)
1.304	20-Oct-2014	skrll	Reduce code a little. Reviewed by gimpy.
1.303	14-Oct-2014	skrll	Typo in comment
1.302	23-Sep-2014	matt	Use right conditional for checking nG.
1.301	23-Sep-2014	nonaka	nG bit exists only in ARMv6 and above.
1.300	21-Sep-2014	christos	remove dead code
1.299	05-Sep-2014	matt	Don't nest structure definitions.
1.298	30-Aug-2014	kiyohara	It expects return value 'va != NULL' from uvm_km_alloc().
1.297	13-Aug-2014	matt	Fix for PR/49061 only kassert in pmap_kenter_pa if PMAP_CACHE_PIVT && !ARM_MMU_EXTENDED
1.296	13-Aug-2014	matt	Fix for PR/49107. Make sure pmap_copy_page_xscale clears the ptes afters its done with them.
1.295	25-Jul-2014	matt	branches: 1.295.2; PTE_SYNC_RANGE a newly allocated L1 page for ARM_MMU_EXTENDED
1.294	15-Jun-2014	ozaki-r	Add missing semicolon
1.293	05-Jun-2014	matt	Fix occupancy bug.
1.292	30-Apr-2014	joerg	pmap_is_current might be unused, tag it as such.
1.291	22-Apr-2014	skrll	Remove stray comment.
1.290	22-Apr-2014	skrll	Call pmap_free_l2_bucket for the kernel pmap so that l2b_occupancy is updated. pmap_free_l2_bucket already deals with the kernel pmap properly. ok matt@ Should fix the panic: kernel debugging assertion "mappings <= l2b->l2b_occupancy" failed: file "/usr/src/sys/arch/arm/arm32/pmap.c", line 3838 problem reported on port-arm by Joachim Thiemann
1.289	22-Apr-2014	skrll	Fix a format string to actually print the prot
1.288	20-Apr-2014	matt	Reduce resident_count by number of pages, not number of ptes.
1.287	12-Apr-2014	matt	KDASSERT -> KDASSERTMSG
1.286	12-Apr-2014	skrll	Trailing whitespace
1.285	12-Apr-2014	skrll	Fix typo and unbreak the build for various arm kernel builds.
1.284	11-Apr-2014	matt	Add a kernel for the CUBIETRUCK (CUBIEBOARD3). Allow direct mapping of all memory (but for now allow the memory mapped above KERNEL_BASE to used for poolpages).
1.283	10-Apr-2014	matt	Fix pasto.
1.282	10-Apr-2014	matt	Fix pmap_extract to deal with non-4KB pages.
1.281	05-Apr-2014	skrll	Add a missing pmap_release_page_lock
1.280	02-Apr-2014	matt	branches: 1.280.2; Mark omd as diagused for the ARM_MMU_EXTENDED case.
1.279	02-Apr-2014	skrll	Fix non-DIAGNOSTIC build
1.278	02-Apr-2014	matt	Init the page_lock to IPL_VM iff VIPT && arm_cache_prefer_mask != 0 otherwise use IPL_NONE. Don't bother with page_lock for KMPAGEs.
1.277	02-Apr-2014	matt	braces are your friends. use them.
1.276	02-Apr-2014	matt	Really fix locking this time.
1.275	02-Apr-2014	matt	Because vector_page might be 0, we have to process the entry L1pt. Make sure to map the new l2 area after we've grown the kernel. If msgbufaddr is set, don't alloc space for msgbuf.
1.274	01-Apr-2014	matt	For ARM_MMU_EXTENDED, if the page being mapped asked for EXEC but is not EXEC, sync it.
1.273	31-Mar-2014	matt	tag a variable with __diagused
1.272	30-Mar-2014	matt	Fix debug code in fault_fixup
1.271	30-Mar-2014	matt	Make the ARM pmap use ASIDs, split TTBRs, and MP safe. This only happens for ARMv6 or later CPUs. This means that on context switch that the TLBs and caches no longer to cleaned/flushed. Also, eXecute Never (XN) protection has been added so non-exec pages can not be run. Change the page size for ARMv6+ to be 8KB while allows a L1PT to be a normal page. This means that the L1PT is not special. Use the XN support to only sync pages that are executed from.
1.270	27-Feb-2014	joerg	Member will never be null, so remove check.
1.269	26-Feb-2014	matt	Only track kenter_pa'ed pages if arm_cache_prefer_mask is non-zero.
1.268	26-Feb-2014	matt	Don't do ref/mod emulation for KENTRY pages. l2pte_minidata -> l2pte_minidata_p
1.267	26-Feb-2014	matt	Move pmap_recent_user to ci->ci_pmap_lastuser and pmap_previous_active_lwp to ci->ci_lastlwp. Fix some comments.
1.266	26-Feb-2014	matt	l2pte_valid -> l2pte_valid_p
1.265	26-Feb-2014	matt	Add support for PMAP_PTE to pmap_kenter_pa
1.264	12-Sep-2013	kiyohara	Fix VIVT cache operation. Tested on Kirkwood machines.
1.263	18-Aug-2013	matt	Include <arm/locore.h>
1.262	03-Jul-2013	matt	Add l2pte_set and l2pte_reset inlines to set/reset a pte. These will be used to support > 4KB pages sizes. Don't use >> L1_S_SHIFT, use L1_IDX() instead.
1.261	03-Jul-2013	matt	restore deleted conditional
1.260	03-Jul-2013	matt	Fix c&p error
1.259	03-Jul-2013	matt	Collapse multiple inlines and move repeated tests into them. No functional change.
1.258	03-Jul-2013	matt	Add pmap_domain and pmap_l1_kva inlines.
1.257	12-Jun-2013	matt	branches: 1.257.2; Add a ARM_HAS_VBAR option which forces the use of the VBAR register. This allows much code to deal with vector_page mappings to be eliminated. On a BEAGLEBONE kernel, this saves 8KB of text and instructions that never have to be executed. (The PJ4B has VBAR but doesn't implement the security extensions it is part of so a method was needed to allow it use VBAR with relying on the default test for the security extensions.)
1.256	12-Jun-2013	matt	If the vector_page is not ARM_VECTORS_{LOW,HIGH}, assume it's in kernel text and don't do anything special to map it.
1.255	11-May-2013	skrll	Fix !DDB build.
1.254	29-Mar-2013	matt	Fix pmap_flush_page to also flush the secondary cache, if there is one. This solves a bus_dma problem with DMA from uncached pages.
1.253	13-Feb-2013	matt	Some armv7 fixes for speculative tlb loads.
1.252	04-Feb-2013	macallan	support ARM32_MMAP_WRITECOMBINE for managed pages as well
1.251	01-Feb-2013	matt	cleanup PVF_WRITE & pvh_attrs interaction.
1.250	31-Jan-2013	skrll	Another typo in a comment.
1.249	31-Jan-2013	skrll	Typo in comment.
1.248	19-Jan-2013	matt	Switch from only UVM_PGA_STRAT_ONLY to UVM_PGA_STRAT_FALLBACK in arm_pmap_alloc_poolpage.
1.247	11-Jan-2013	matt	Fix a bug in pmap_modify_pv where we didn't set PVF_WRITE on a page after changing its mapping to writeable. Add more KASSERTS Don't go into DDB by default in pmap_fixup.
1.246	11-Dec-2012	matt	Fix C&P bug.
1.245	11-Dec-2012	matt	Fix c&p error.
1.244	11-Dec-2012	matt	Optimize pmap_{copy,zerp}_page_generic to understand that when all of memory is always mapped that you don't have create mappings dynamically.
1.243	10-Dec-2012	matt	Change a KASSERT to a KASSERTMSG
1.242	12-Nov-2012	skrll	C99 types
1.241	17-Oct-2012	matt	Add a PMAP_NEEDS_ALLOC_POOLPAGE / PMAP_ALLOC_POOLPAGE hook so systems can allocate pool pages from a specific VM freelist.
1.240	17-Oct-2012	matt	Need to set pmap_needs_pte_sync before calling PTE_SYNC
1.239	17-Oct-2012	matt	When setting pmap_needs_pte_sync to 1 be sure to sync pte that caused the issue.
1.238	26-Sep-2012	matt	If we get a fault we shouldn't have, set pmap_needs_pte_sync and retry.
1.237	22-Sep-2012	matt	Don't use an asm in pmap_activate to update the TTBR, use cpu_setttb instead but add a second argument to it to indicate whether the TLB/caches need to be flushed. Default cortex to pmap_needs_fixup = 1. But check the MMFR3 field to see if the fixed can be skipped. Use a cf_flag bit 0 to indicate whether the A9 L2 cache should disable (bit 0 = 1) or enabeld (bit = 0). With these changes, the A9 MMU can use traverse caches to do MMU tablewalks Also, make sure all memory has the shareable bit for the A9.
1.236	02-Sep-2012	matt	branches: 1.236.2; Supersections are on armv6 too.
1.235	30-Aug-2012	matt	Teach pmap_extract about supersections.
1.234	29-Aug-2012	matt	Make all cortex and arm11 cpus uses writeback cached memories for pagetables
1.233	29-Aug-2012	matt	Support PMAP_NOCACHE in pmap_kenter_pa
1.232	29-Aug-2012	matt	Use the correct prot mask in vector_page_setprot
1.231	27-Aug-2012	matt	Use the correct (L1_S_PROT, not L2_S_PROT) macro for setting the protection of the vector page. This fixes a nasty little bug that shows up on armv7 systems when the vector page physical address changes (bit 12 of the address is cleared) and then any exception causes the system to hang.
1.230	20-Aug-2012	matt	Add support for mapping SuperSection on armv6 and armv7. These always a domain of 0 so move the kernel from domain 15 to domain 0.
1.229	13-Jul-2012	skrll	Fix a KASSERT. From/OK'ed by matt@
1.228	29-Jan-2012	he	branches: 1.228.2; Only declare helper variable which is only used in KASSERT() if DIAGNOSTIC is defined, to avoid "variable defined but never used" warning if DIAGNOSTIC isn't defined, and KASSERT() expands to nothing.
1.227	28-Jan-2012	matt	Since we don't do MULTIPROCESSOR, and hence preemption, locking the pvlists doesn't really matter.
1.226	28-Jan-2012	matt	Don't use simple locks.
1.225	27-Jan-2012	para	extending vmem(9) to be able to allocated resources for it's own needs. simplifying uvm_map handling (no special kernel entries anymore no relocking) make malloc(9) a thin wrapper around kmem(9) (with private interface for interrupt safety reasons) releng@ acknowledged
1.224	01-Jul-2011	dyoung	branches: 1.224.2; 1.224.6; #include <sys/bus.h> instead of <machine/bus.h>.
1.223	30-Jun-2011	wiz	dependant -> dependent
1.222	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.
1.221	10-Mar-2011	bsh	branches: 1.221.2; Preliminary ARM11 MPCore support. I have confirmed this commit doesn't affect existing evbarm kernels by comparing binaries.
1.220	28-Feb-2011	macallan	implement arm32_pmap_flags() to allow mappings with write buffering enabled, mostly for video memory Tested on shark
1.219	12-Nov-2010	uebayasi	branches: 1.219.2; 1.219.4; Put VM_PAGE_TO_MD() definition in one place. No functional changes.
1.218	10-Nov-2010	uebayasi	Use more VM_PHYSMEM_*() accessors. No functional changes.
1.217	03-Nov-2010	uebayasi	Fix build of IXM1200 too. Pointed out by cegger, thanks.
1.216	02-Nov-2010	uebayasi	Fix build of XScale.
1.215	30-Oct-2010	uebayasi	Pass struct vm_page_md * where possible. This causes 1% code increase, mainly because additional argument (paddr_t) affects register usage. This will be fixed when per-page data structure (struct vm_page) is redone, and physical address can be retrieved from struct vm_page_md *. Tested on (uncommitted) i.MX35 (ARM1136).
1.214	16-Jun-2010	jmcneill	PR port-arm/43299: Support added for igepv2/cortexa8/omap3530 Apply patch from PR, with build fixes. ok skrll, matt
1.213	14-May-2010	cegger	Move PMAP_KMPAGE to be used in pmap_kenter_pa flags argument. 'Looks good to me' gimpy@
1.212	15-Feb-2010	skrll	branches: 1.212.2; Typo in comment.
1.211	02-Jan-2010	he	branches: 1.211.2; Remove a shadowed and unused local declaration so that this builds again.
1.210	01-Jan-2010	uebayasi	Sprinkle assertions after calling pmap_get_l2_bucket().
1.209	31-Dec-2009	uebayasi	Use pmap_is_current() where appropriate. No functional changes.
1.208	31-Dec-2009	uebayasi	pmap_page_remove(): remove an unused local variable; no functional changes.
1.207	31-Dec-2009	uebayasi	Correct assertions for PMAP_CACHE_VIPT code; if page coloring is disabled (arm_cache_prefer_mask == 0), pvh_attrs has never PVF_COLORED, so don't assert it. No functional changes.
1.206	28-Dec-2009	uebayasi	Indent.
1.205	28-Dec-2009	uebayasi	Always name struct pv_entry * as pv, not pve. No binary change.
1.204	27-Dec-2009	uebayasi	Add write-through cache work-around for ARM11 as well as ARM9/ARM10. Analyzed & tested on i.MX35 with help from Tsubai Masanari.
1.203	28-Nov-2009	scw	Apply some band-aid to pmap_activate() for PR kern/41058: There's a corner case here which can leave turds in the cache as reported in kern/41058. They're probably left over during tear-down and switching away from an exiting process. Until the root cause is identified and fixed, zap the cache when switching pmaps. This will result in a few unnecessary cache flushes, but that's better than silently corrupting data. Also remove an extraneous return statement in pmap_page_protect() which crept in during the matt-armv6 merge.
1.202	21-Nov-2009	rmind	Use lwp_getpcb() on ARM (and acorn26/32), clean from struct user usage.
1.201	07-Nov-2009	cegger	Add a flags argument to pmap_kenter_pa(9). Patch showed on tech-kern@ http://mail-index.netbsd.org/tech-kern/2009/11/04/msg006434.html No objections.
1.200	22-Oct-2009	rmind	Simplify pmap_remove() a little by avoiding pmap_do_remove() layer, since possibility to skip wired mappings is not needed anymore. Apart from that, no functional differences are intended.
1.199	21-Oct-2009	rmind	Remove uarea swap-out functionality: - Addresses the issue described in PR/38828. - Some simplification in threading and sleepq subsystems. - Eliminates pmap_collect() and, as a side note, allows pmap optimisations. - Eliminates XS_CTL_DATA_ONSTACK in scsipi code. - Avoids few scans on LWP list and thus potentially long holds of proc_lock. - Cuts ~1.5k lines of code. Reduces amd64 kernel size by ~4k. - Removes __SWAP_BROKEN cases. Tested on x86, mips, acorn32 (thanks <mpumford>) and partly tested on acorn26 (thanks to <bjh21>). Discussed on <tech-kern>, reviewed by <ad>.
1.198	21-Apr-2009	cegger	change pmap flags argument from int to u_int. discussed with christos@ on source-changes-d@
1.197	15-Mar-2009	cegger	ansify function definitions
1.196	09-Mar-2009	nonaka	avail_start and avail_end is paddr_t.
1.195	08-Jan-2009	matt	branches: 1.195.2; When allocating a KMPAGE and it has multiple colors, make sure to flush them. Add some more KASSERTs for multiple colors (or lack thereof).
1.194	30-Dec-2008	matt	Reclaim PVF_KNC in VIPT to be PVF_MULTCLR (page has multiple colors). Track when a page is mapping in multiple colors and deal with the ramifications. When a page's MOD attribute is cleared, clean it from the cache. Fix a logic inversion. With these changes, the TI SDP2420 H4 board can successfully natively build a TISDP2420 kernel.
1.193	10-Dec-2008	pooka	Make kernel_pmap_ptr a const. Requested by steve_martin.
1.192	09-Dec-2008	pooka	Make pmap_kernel() a MI macro for struct pmap *kernel_pmap_ptr, which is now the "API" provided by the pmap module. pmap_kernel() remains as the syntactic sugar. Bonus cosmetics round: move all the pmap_t pointer typedefs into uvm_pmap.h. Thanks to Greg Oster for providing cpu muscle for doing test builds.
1.191	19-Nov-2008	matt	Allocate /dev/mem's page in pmap_init. Mark /dev/mem as MPSAFE. Ansify.
1.190	12-Nov-2008	ad	Remove LKMs and switch to the module framework, pass 1. Proposed on tech-kern@.
1.189	04-Nov-2008	matt	Use a mutex to control access DEV_MEM.
1.188	04-Nov-2008	matt	Protect some code with if (pg) { .. }
1.187	28-Sep-2008	skrll	branches: 1.187.2; 1.187.4; 1.187.8; Typo in comment.
1.186	14-Aug-2008	matt	Now that we track pages used by any of the kernel memory allocators, keep a count of them and export it as a sysctl node machdep.kmpages
1.185	13-Aug-2008	matt	Fix a few more corner cases. Always KMPAGE or pages with unmanaged writeable kernel mappings as modified. Only ever set DIRTY bit is DMOD is true and NC is false. Don't modify unmanaged mappings in pmap_clearbit.
1.184	08-Aug-2008	dogcow	fix "warning: 'npv' may be used uninitialized in this function"
1.183	06-Aug-2008	matt	Change pv_entries to use SLIST. For VIPT caches, keep track of when pages are dirty so that their content can be flushed back to main memory. This is done when the page is read-only mapped by more than 1 color. Pages become when either their modified bit gets set or an unmanaged writeable page is mapped. When a page in unmapped or changed to read-only, run pmap_vac_me_harder in case the page can be mapped read-only. Thanks are given to Imre Deak for giving me the idea to assert for PVF_DIRTY.
1.182	16-Jul-2008	matt	Revamp bookkeeping for pages entered by pmap_kenter_pa. Keep track of them on pvlists so that the cacheability can be properly tracked.
1.181	09-Jul-2008	scw	When dealing with 'cleanlist_idx == PMAP_REMOVE_CLEAN_LIST_SIZE' in pmap_do_remove(), make sure to use PTE_SYNC() for each rolled-back PTE on the list. Fixes potential MMU inconsistencies on some ARM platforms where page-tables are mapped write-back.
1.180	03-Jul-2008	matt	branches: 1.180.2; Use the same login in pmap_kenter_pa when removing an unmamanged mapping as in pmap_kremove (otherwise kro_mappings will become incorrect).
1.179	03-Jul-2008	matt	For armv6(VIPT), change the rules for mapping via kenter_pa. Allow readonly pages to be mapped by different cache color indexes.
1.178	24-Jun-2008	scw	In pmap_deactivate(), if the process is exiting make sure the next call to pmap_activate() performs a full MMU context-switch and cache flush, which might otherwise be skipped. Fixes ARM_LOW_VECTORS problem reported in PR port-arm/38950.
1.177	17-Jun-2008	chris	Fix two KASSERT(value \| (C1\|C2)) to KASSERT(value & (C1\|C2)) so that it tests for something, rather than always being true. Pointed out by Andy Shevchenko in: http://mail-index.netbsd.org/port-arm/2008/06/17/msg000255.html
1.176	04-Jun-2008	ad	branches: 1.176.2; vm_page: put TAILQ_ENTRY into a union with LIST_ENTRY, so we can use both.
1.175	28-Apr-2008	martin	branches: 1.175.2; Remove clause 3 and 4 from TNF licenses
1.174	27-Apr-2008	matt	Merge kernel changes in matt-armv6 to HEAD.
1.173	20-Apr-2008	scw	branches: 1.173.2; There's really no need to switch VM contexts within cpu_switchto() as MI code always calls pmap_deactivate/pmap_activate on context switch. Instead, just record the last active lwp (or NULL if it exited) and defer switching VM context to pmap_activate(). This saves an additional function call overhead in cpu_switchto(). While here, g/c unused cpuswitch.S local .Lblock_userspace_access.
1.172	29-Mar-2008	chris	branches: 1.172.2; Fix LOCKDEBUG build on arm by: * converting simple_{un}lock to mutex_enter/exit * Using UVM_OBJ_INIT & DESTROY for the uvm_object in the pmap structure
1.171	06-Jan-2008	matt	branches: 1.171.6; current_intr_depth is dead. Make sure we don't use it anymore.
1.170	01-Jan-2008	chris	Add support for kcore headers to arm32 kernel core dumps. The kcore code is based on i386's kcore header handling. Having an asm stub for dumpsys, to dump the registers onto the stack, and then call the C code to do the memory dump is based on amd64's core dump code. This allows a successful core dump on cats. Part of fixing PR cats/18026.
1.169	08-Nov-2007	matt	branches: 1.169.6; Make this compile again.
1.168	07-Nov-2007	ad	Merge from vmlocking: - pool_cache changes. - Debugger/procfs locking fixes. - Other minor changes.
1.167	17-Oct-2007	garbled	branches: 1.167.2; 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.
1.166	10-Oct-2007	ad	branches: 1.166.2; Comment out references to spinlockmgr().
1.165	15-Sep-2007	scw	branches: 1.165.2; ARM cpu_switchto() has been partially broken since yamt-idlelwp was merged as its cache/tlb management smarts relied too heavily on pre-merge context- switch behaviour. See PR kern/36548 for one manifestation of the breakage. To address this: - Ditch the shadow pmap variables in the PCB (pagedir, l1vec, dacr, cstate) as it was too easy for them to get out of sync with the pmap. - Re-write (and fix) the convoluted cpuswitch.S cache/tlb ASM code in C. It's only slightly less efficient, but is much more readable/maintainable. - Document cpufuncs.cf_context_switch() as being C-callable. - pmap_activate() becomes a no-op if the lwp's vmspace is already active. (Good performance win, since pmap_activate() is now invoked on every context-switch, even though ARM's cpu_switchto() already does all the grunt work) XXX: Some CPU-specific armXX_context_switch() implementations (arm67, arm7tdmi, arm8) always flush the I+D caches. This should not be necessary. Someone with access to hardware (acorn32?) needs to deal with this.
1.164	17-May-2007	yamt	branches: 1.164.6; 1.164.8; 1.164.10; 1.164.12; 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.
1.163	09-Apr-2007	chris	branches: 1.163.4; In pmap_activate restore interrupts to their previous state rather than always enabling them. It's not clear if this actually caused any problems, but it seems safer to restore to the previous state in case pmap_activate is ever called with interrupts disabled.
1.162	12-Mar-2007	ad	branches: 1.162.2; Pass an ipl argument to pool_init/POOL_INIT to be used when initializing the pool's lock.
1.161	04-Mar-2007	christos	branches: 1.161.2; Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.
1.160	22-Feb-2007	thorpej	TRUE -> true, FALSE -> false
1.159	21-Feb-2007	thorpej	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.
1.158	06-Jan-2007	christos	branches: 1.158.2; From Scott Allen in http://mail-index.netbsd.org/port-arm/2006/07/26/0000.html I ran into a problem when I tried to set up a mapping that started at virtual address 0xFFF00000 and was 0x00100000 long. In other words, the mapping should have gone to the end of the 32 bit address space. The mapping was made with no problem, but pmap_devmap_find_va() wouldn't find an address within the mapping. For example, if I told it to find a mapping for 0x1000 bytes at 0xFFF01000, it would try to make sure that 0xFFF01000 was greater than 0xFFF00000 and that (0xFFF01000+0x1000) was less than (0xFFF00000+0x00100000). However, that last expression (0xFFF00000+0x00100000) wrapped around to be simply 0x00000000 so it wasn't found. This patch fixes this problem in pmap_devmap_find_va() and pmap_devmap_find_pa() by subtracting one off of the sizes to be compared, so in my example, (0xFFF01000+0x1000-1) will be less than (0xFFF00000+0x00100000-1).
1.157	24-Dec-2005	perry	branches: 1.157.20; 1.157.24; Remove leading __ from __(const\|inline\|signed\|volatile) -- it is obsolete.
1.156	10-Dec-2005	scw	Implement pmap_collect() for arm32.
1.155	08-Dec-2005	yamt	use VM_PAGE_TO_PHYS macro.
1.154	04-Jul-2005	bsh	branches: 1.154.2; The first step to support Intel PXA270. kernel config option CPU_XSCALE_PXA2X0 is now obsoleted by CPU_XSCALE_PXA250 and CPU_XSCALE_PXA270. If both of them are defined, CPU is determined run-time.
1.153	24-Jun-2005	scw	In pmap_devmap_find_pa(), use 64-bit arithmetic to handle the case where 'pa + size' == 0x0. As in, if we're passed details of a region right at the top of physical address space. Otherwise we'll likely hit a false-positive due to 32-bit wrap-around.
1.152	26-Apr-2005	scw	Since pmap_page_remove() is called from pmap_page_protect(), don't modify the current pmap's pm_cstate if we have to flush the TLB, as callers of pmap_page_protect() are not required to invoke pmap_update() afterwards. Otherwise we can end up with a pm_cstate which is inconsistent with reality in the TLB, which can lead to future TLB flushes being erroneously skipped.
1.151	01-Apr-2005	yamt	merge yamt-km branch. - don't use managed mappings/backing objects for wired memory allocations. save some resources like pv_entry. also fix (most of) PR/27030. - simplify kernel memory management API. - simplify pmap bootstrap of some ports. - some related cleanups.
1.150	14-Jan-2005	joff	branches: 1.150.2; 1.150.4; Dont try freeing bootstrap pages with pool_page_free. Fixes kern/28869
1.149	02-Jan-2005	chris	Remove direct references to TAILQ internal structures. No functional change, just tidying up code.
1.148	03-Apr-2004	bsh	pmap_pte_init_arm9() is necessary only when defined(CPU_ARM9) and defined(ARM9_CACHE_WRITE_THROUGH)
1.147	18-Jan-2004	scw	Fix ARM_VECTORS_LOW fallout caused by the recent reaper removal. Just before removing the vector page mapping, switch to the kernel pmap's L1/vector page mapping so as not to pull the rug out from under ourselves. To prevent the stale L1/vector page mapping from being restored by cpu_switch, replace the relevant fields of the dying process' pcb with those of lwp0's pcb.
1.146	01-Nov-2003	jdolecek	avoid stong words in comments
1.145	29-Oct-2003	mycroft	Whitespace.
1.144	29-Oct-2003	mycroft	The previous patch was wrong -- mcr does not output anything. Instead give a junk input (it's not used when the last argument is 0).
1.143	28-Oct-2003	scw	Fix an uninitialised variable warning, reported by Shoichi Miyake in port-arm/23293.
1.142	26-Oct-2003	chris	Fix up some unitialised variables.
1.141	13-Oct-2003	scw	On Xscale, define PMAP_UAREA() and use it to tweak uarea mappings so they use the mini D$. This results in a small performance boost on xscale platforms, since flushing the main cache on a context switch won't affect the kernel stack/pcb.
1.140	05-Oct-2003	matt	Add SA_SIGINFO support for ARM (from Chris Gilbert).
1.139	21-Sep-2003	matt	Change some type-punning detected by gcc 3.3.1 to (void *).
1.138	06-Sep-2003	rearnsha	Support for initializing ARM10 processors in write-through mode.
1.137	23-Jun-2003	martin	branches: 1.137.2; Make sure to include opt_foo.h if a defflag option FOO is used.
1.136	15-Jun-2003	thorpej	Add another devmap routine that allows bootstrap code to register a devmap reflecting mappings that are created by really early bootstrap code before pmap_devmap_bootstrap() is called.
1.135	15-Jun-2003	thorpej	Replace the ad-hoc "section mapping table" for static device mappings with a more generic "devmap" structure that can also handle mappings made with large and small pages. Add new pmap routines to enter these mappings during bootstrap (and "remember" the devmap), and routines to look up the static mappings once the kernel is running.
1.134	21-May-2003	thorpej	Move the new pmap from arm32/pmap_new.c to arm32/pmap.c, fully replacing the old.
1.133	10-May-2003	thorpej	Back out the following chagne: http://mail-index.netbsd.org/source-changes/2003/05/08/0068.html There were some side-effects that I didn't anticipate, and fixing them is proving to be more difficult than I thought, do just eject for now. Maybe one day we can look at this again. Fixes PR kern/21517.
1.132	08-May-2003	thorpej	Simplify the way the bounds of the managed kernel virtual address space is advertised to UVM by making virtual_avail and virtual_end first-class exported variables by UVM. Machine-dependent code is responsible for initializing them before main() is called. Anything that steals KVA must adjust these variables accordingly. This reduces the number of instances of this info from 3 to 1, and simplifies the pmap(9) interface by removing the pmap_virtual_space() function call, and removing two arguments from pmap_steal_memory(). This also eliminates some kludges such as having to burn kernel_map entries on space used by the kernel and stolen KVA. This also eliminates use of VM_{MIN,MAX}_KERNEL_ADDRESS from MI code, this giving MD code greater flexibility over the bounds of the managed kernel virtual address space if a given port's specific platforms can vary in this regard (this is especially true of the evb* ports).
1.131	22-Apr-2003	thorpej	Some ARM32_PMAP_NEW-related cleanup: * Define a new "MMU type", ARM_MMU_SA1. While the SA-1's MMU is basically compatible with the generic, the SA-1 cache does not have a write-through mode, and it is useful to know have an indication of this. * Add a new PMAP_NEEDS_PTE_SYNC indicator, and try to evaluate it at compile time. We evaluate it like so: - If SA-1-style MMU is the only type configured -> 1 - If SA-1-style MMU is not configured -> 0 - Otherwise, defer to a run-time variable. If PMAP_NEEDS_PTE_SYNC might evaluate to true (SA-1 only or run-time check), then we also define PMAP_INCLUDE_PTE_SYNC so that e.g. assembly code can include the necessary run-time support. PMAP_INCLUDE_PTE_SYNC largely replaces the ARM32_PMAP_NEEDS_PTE_SYNC manual setting Steve included with the original new pmap. * In the new pmap, make pmap_pte_init_generic() check to see if the CPU has a write-back cache. If so, init the PT cache mode to C=1,B=0 to get write-through mode. Otherwise, init the PT cache mode to C=1,B=1. * Add a new pmap_pte_init_arm8(). Old pmap, same as generic. New pmap, sets page table cacheability to 0 (ARM8 has a write-back cache, but flushing it is quite expensive). * In the new pmap, make pmap_pte_init_arm9() reset the PT cache mode to C=1,B=0, since the write-back check in generic gets it wrong for ARM9, since we use write-through mode all the time on ARM9 right now. (What this really tells me is that the test for write-through cache is less than perfect, but we can fix that later.) * Add a new pmap_pte_init_sa1(). Old pmap, same as generic. New pmap, does generic initialization, then resets page table cache mode to C=1,B=1, since C=1,B=0 does not produce write-through on the SA-1.
1.130	01-Apr-2003	thorpej	Use PAGE_SIZE rather than NBPG.
1.129	29-Mar-2003	bsh	for Intel PXA2[15][05] processors, select write-back/write-through cache based on CPU id. write-through on PXA2[15]0 B2 stepping and earlier. write-back on C0 and C1 stepping (a.k.a PXA2[15]5 A0) options XSCALE_CACHE_WRITE_{THROUGH,BACK} can override it. for other XScale CPUs than PXA2xx, XSCALE_CACHE_WRITE_THROUGH works same as before.
1.128	27-Mar-2003	mycroft	Remove references to variables that aren't used here.
1.127	23-Mar-2003	chris	Garbage collect pmap_map, the last (and only?) use has been removed.
1.126	23-Feb-2003	thorpej	Change pcb32_pagedir to a paddr_t (after all, it's used as a paddr_t everywhere in the code).
1.125	21-Feb-2003	chris	Convert a few types into things that are more accurate, mostly: int's to unsigned int/u_int where they shouldn't go negative. int's to boolean_t's where they're being used as bools. No real functional change (in the produced asm a few condition codes changed)
1.124	17-Jan-2003	thorpej	Merge the nathanw_sa branch.
1.123	24-Nov-2002	chris	Add a debug assert that wired pages provide protection flags in the flags argument as well. Also update a couple of debug messages to NPDEBUG.
1.122	12-Nov-2002	chris	Tweak a few minor things: when looking to reenable caching, only do so if all the pages aren't already cached. Convert some ints to unsigned int. (scarily this actually shows the biggest decrease in timing for my benchmark, I guess the compiler can optimise better)
1.121	11-Nov-2002	chris	gratuitous whitespace and de-__P'ing. No functional change.
1.120	11-Nov-2002	chris	A few minor tweaks. Use pmap_free_pvs in pmap_remove, should save on the overhead of freeing each pv on it's own. Correctly set ptp when calling pmap_enter_pv, this adds more overhead, but the effect is minimal. Timings show that it increases gmake's make configure step from 2:07.90 to 2:08.90. I've more optimisations planned that should negate this increase.
1.119	11-Nov-2002	chris	Remove unused pa variable (it's assigned but not used any more)
1.118	27-Sep-2002	provos	remove trailing \n in panic(). approved perry.
1.117	22-Sep-2002	chs	rename the existing pmap_remove_all() here to pmap_page_remove() (ala the x86 pmap) to avoid conflicting with the new pmap interface function of the same name.
1.116	05-Sep-2002	jdolecek	whitespace fix past __KERNEL_RCSID()
1.115	24-Aug-2002	thorpej	In pmap_map_in_l1() and pmap_unmap_in_l1(), make sure that the VA that is passed in is already aligned to a 4M super-section.
1.114	24-Aug-2002	thorpej	When we allocate a PTP, make sure the offset we specify is for the 4M super-section that the PTP will map, not some random 1M chunk of it. This gives the PTP hint code a much better chance to working properly, and allows us to tidy up the code that flushes a PTP from the cache in pmap_destroy().
1.113	24-Aug-2002	thorpej	Enable caching on kernel and user page tables. This saves having to do uncached memory access during VM operations (which can be quite expensive on some CPUs). We currently write-back PTEs as soon as they're modified; there is some room for optimization (to write them back in larger chunks). For PTEs in the APTE space (i.e. PTEs for pmaps that describe another process's address space), PTEs must also be evicted from the cache complete (PTEs in PTE space will be evicted durint a context switch).
1.112	22-Aug-2002	thorpej	* Add PTE_SYNC() and PTE_SYNC_RANGE() macros. These don't actually do anything yet. * Use PTE_SYNC() and PTE_SYNC_RANGE() in some obvious places, i.e. where vtopte() is used.
1.111	21-Aug-2002	thorpej	Use a pool cache for PT-PTs.
1.110	21-Aug-2002	thorpej	Do cached memory access to L1 tables, making sure to write-back the cache after any L1 table modifications.
1.109	13-Aug-2002	thorpej	Add the brutal hack that allows us to limp along using the read/write cache line allocation policy on XScale CPUs: in pmap_enter(), if the pmap is the kernel pmap, clear the X-bit in the PTE, thus disabling read/write-allocate for managed kernel mappings. Yes, this is ugly. But it makes userland code run with r/w-allocate, which is a huge improvement on systems with low core memory performance.
1.108	10-Aug-2002	thorpej	Tidy up pmap_clean_page() a little, and reenable some code that was disabled previously: Skip cleaning mappings which are read-only, because the pmap (now) does clean pages on a r/w -> r/o transition.
1.107	10-Aug-2002	thorpej	Clean up some warts in pmap_protect().
1.106	09-Aug-2002	thorpej	Add an XSCALE_CACHE_READ_WRITE_ALLOCATE option for people who want to play fast-and-loose.
1.105	09-Aug-2002	thorpej	Add some code, conditional on PMAP_ALIAS_DEBUG, that can be used to hunt for virtual aliases between managed (pmap_enter) and non-managed (pmap_kenter_pa) mappings.
1.104	06-Aug-2002	thorpej	- pmap_remove(): unmap the PTEs after we have finished with the page tables. - pmap_enter(): if making a mapping for the same PA rw->ro, write-back the cache before doing so. - pmap_clearbit(): if revoking REF on a page, make sure to wbinv the cache if the page has write permission, else inv the cache if the page's PTE is valid (XXX we actually wbinv in this case, as well, due to lack of idcache_inv_range()). Only flush the TLB if the PTE changed.
1.103	31-Jul-2002	thorpej	Overhaul how DMA ranges work in the ARM bus_dma implementation. A new "arm32_dma_range" structure now describes a DMA window, with a system address base, bus address base, and length. In addition to providing info about which memory regions are legal for DMA, the new structure provides address translation support, as well. As before, if a tag does not list any ranges, then all addresses are considered valid, and no DMA address translation is performed. This allows us to remove a large chunk of code which was duplicated and tweaked slightly (to do the address translation) from the stock ARM bus_dma in the XScale IOP and ARM Integrator ports. Test compiled on all ARM platforms, test booted on Intel IQ80321 and Shark.
1.102	31-Jul-2002	thorpej	Move the calls to uvm_page_physload() out of pmap_bootstrap() and into platform-specific initialization code, giving platform-specific code control over which free list a given chunk of memory gets put onto. Changes are essentially mechanical. Test compiled for all ARM platforms, test booted on Intel IQ80321 and Shark. Discussed some time ago on port-arm.
1.101	30-Jul-2002	thorpej	Move the uvm_setpagesize() call to platform-dependent code in preparation for other changes to pmap_bootstrap().
1.100	30-Jul-2002	thorpej	Don't use pmap_kenter_pa() in pmap_map(); doing so causes an assertion failure in pmap_kenter_pa().
1.99	02-Jun-2002	drochner	move initialization of the "struct pglist" returned by uvm_pglistalloc() from the calling code into uvm_pglistalloc() itself for consistency and easier error handling
1.98	01-Jun-2002	lukem	SIMPLEQ rototill: - implement SIMPLEQ_REMOVE(head, elm, type, field). whilst it's O(n), this mirrors the functionality of SLIST_REMOVE() (the other singly-linked list type) and FreeBSD's STAILQ_REMOVE() - remove the unnecessary elm arg from SIMPLEQ_REMOVE_HEAD(). this mirrors the functionality of SLIST_REMOVE_HEAD() (the other singly-linked list type) and FreeBSD's STAILQ_REMOVE_HEAD() - remove notes about SIMPLEQ not supporting arbitrary element removal - use SIMPLEQ_FOREACH() instead of home-grown for loops - use SIMPLEQ_EMPTY() appropriately - use SIMPLEQ_*() instead of accessing sqh_first,sqh_last,sqe_next directly - reorder manual page; be consistent about how the types are listed - other minor cleanups
1.97	14-May-2002	chris	branches: 1.97.2; 1.97.4; Implement scheduler lock protocol, this fixes PR arm/10863. Also add correct locking when freeing pages in pmap_destroy (fix from potr) This now means that arm32 kernels can be built with LOCKDEBUG enabled. (only tested on cats though)
1.96	24-Apr-2002	thorpej	* pmap_clean_page(): Clean up a comment. * pmap_protect(): write back the range when doing a r/w -> r/o transition. (Still leave the block concerned with this in pmap_clean_page() disabled, for now.) * pmap_pte_init_xscale(): Disable read/write-allocate for now, until we figure out why sometimes cache lines of NULs get deposited into file data. Also, make sure ECC protection of page table access is disabled for now. * xscale_setup_minidata(): Make sure the mini-data cache is configured write-back with read/write-allocate.
1.95	12-Apr-2002	thorpej	Default all XScale core processors to the read/write-allocate write-back cache mode. Add a new XSCALE_CACHE_WRITE_THROUGH option for people who are paranoid about the cache-related errata (you do have to line up the planets correctly to trip them, but having the option is useful).
1.94	10-Apr-2002	thorpej	On XScale processors where we use write-back caching, use are read/write-allocate line allocation policy. On the i80321, this improves nearly every lmbench benchmark, dramatically so the ones that are sensitive to memory bandwidth (100-300% improvement for these).
1.93	10-Apr-2002	thorpej	Add a new function, pmap_alloc_ptpt(), that allocates the PTPT and maps it the way we want, rather than using uvm_km_zalloc() and playing the "revoke cacheability" song-and-dance.
1.92	10-Apr-2002	thorpej	pmap_alloc_l1pt(): Just enter the mappings for the L1 table by hand, rather than calling pmap_kenter_pa() and then revoking cacheability in the PTE.
1.91	10-Apr-2002	thorpej	Use L2_S_CACHE_MASK in places where we revoke cacheability.
1.90	10-Apr-2002	thorpej	pmap_kenter_pa(): Obey the "prot" argument, rather than simply making all mappings r/w (!!).
1.89	10-Apr-2002	thorpej	In pmap_copy_page_xscale(), put the source page in the mini-data cache, as well. The mini-data cache is 2-way, so src and dst won't clobber each other, and the smallness of the cache doesn't matter, since we access each page once sequentially. While we still have to do the initial clean of the source page, this saves another 4K of main D$ pollution, and also means we don't have to do 2 cache passes after the copy is complete (i.e. we can skip the invalidation of the source page in the main cache, since it's no longer there).
1.88	10-Apr-2002	thorpej	Add separate pmap_{zero,copy}_page() functions for generic ARM vs. XScale. Use the mini-data cache for the destination on XScale, thus saving tossing out 4K of possible-useful data from the main data cache each time. This significantly improves every test in lmbench.
1.87	09-Apr-2002	thorpej	* Move the code that cleans the XScale mini-data cache into its own function. * Add a new function which sets up the mini-data cache clean area properly.
1.86	09-Apr-2002	thorpej	* Split pte_cache_mode into pte_l1_s_cache_mode, pte_l2_l_cache_mode, and pte_l2_s_cache_mode. The cache-meaningful bits are different for these descriptor types on some processor models. * Add pte_*_cache_mask, corresponding to each above, which has a mask of the cache-meangful bits, and define those for generic and XScale MMU classes. Note, the L2_S_CACHE_MASK_xscale definition requires use of the Extended Small Page L2 descriptor (the "X" bit overlaps with AP bits otherwise).
1.85	09-Apr-2002	thorpej	Define 2 classes of ARM MMUs: 1. Generic (compatible with ARM6) 1. XScale (can be used as generic, but also has certainly nifty extensions). Define abstract PTE bit defintions for each MMU class. If only one MMU class is configured into the kernel (based on CPU_* options), then we get the constants for that MMU class. Otherwise we indirect through varaibles set up via set_cpufuncs(). XXX The XScale bits are currently the same as the generic bits. Baby steps.
1.84	09-Apr-2002	thorpej	L2_TYPE_S -> L2_S_PROTO
1.83	09-Apr-2002	thorpej	Use abstract names for the protection and PTE type bits in L1 and L2 descriptors. This will allow us to support different PTE layouts that enable the use of extensions on different processor models.
1.82	05-Apr-2002	thorpej	Back-out rev 1.75 (pmap_extract() rewrite), and fix the (minor) bug that revision intended to fix properly.
1.81	05-Apr-2002	thorpej	* Rewrite the 32-bit ARM pte.h based on the ARM architecture manual. Significant cleanup, here, including better PTE bit names. * Add XScale PTE extensions (ECC enable, write-allocate cache mode). * Mechanical changes everywhere else to update for new pte.h. While doing this, two bugs (as a result of typos) were fixed in arm/arm32/bus_dma.c evbarm/integrator/int_bus_dma.c
1.80	04-Apr-2002	thorpej	Eliminate a mask against PD_MASK.
1.79	04-Apr-2002	thorpej	There is no need to mask VAs and PAs w/ PG_FRAME to clear the lower bits; UVM provides us page-aligned addresses for everything. For the paranoid, we'll leave KDASSERT()'s in that check for this if the kernel is built with DEBUG. Low-hanging fruit that shaves some cycles.
1.78	04-Apr-2002	thorpej	Rename flags that are really part of the pv_entry/mdpage into pmap.h and give them more descriptive names and better comments: * PT_M -> PVF_MOD (page is modified) * PT_H -> PVF_REF (page is referenced) * PT_W -> PVF_WIRED (mapping is wired) * PT_Wr -> PVF_WRITE (mapping is writable) * PT_NC -> PVF_NC (mapping is non-cacheable; multiple mappings)
1.77	04-Apr-2002	thorpej	Catch a couple more vector page mapping manipulations.
1.76	03-Apr-2002	thorpej	Clean up handling of the vector page on 32-bit ARM systems: * Don't refer to VA 0, instead refer to a new variable: vector_page * Delete the old zero_page_() functions, replacing them with a new one: vector_page_setprot(). When manipulating vector page mappings in user pmaps, only do so if the vector page is below KERNEL_BASE (if it's above KERNEL_BASE, the vector page is mapped by the kernel pmap). * Add a new function, arm32_vector_init(), which takes the virtual address of the vector page (which MUST be valid when the function is called) and a bitmask of vectors the kernel is going to take over, and performs all vector page initialization, including setting the V bit in the CPU Control register ("relocate vectors to high address"), if necessary.
1.75	03-Apr-2002	reinoud	Rototil and fix the pmap_extract function. It wouldn't even return data when the part being quiried was mapped with a section (!) giving weird results and had become a mess of goto's. Complete rewrite and cleaned up the `goto'-jungle entirely ... ripped all goto's. The resulting code is much better to read and might even have a small performance gain.
1.74	25-Mar-2002	thorpej	Fix reporting of the kernel virtual address space range to UVM.
1.73	25-Mar-2002	thorpej	* Some cleanup. * Delete the call to pmap_copy() in pmap.h
1.72	25-Mar-2002	thorpej	Clean up pmap_map_ptes() and pmap_unmap_ptes() a little, and add a debug assertion that curproc is never NULL if mapping a non-current pmap.
1.71	25-Mar-2002	thorpej	The target page of pmap_zero_page(), pmap_pageidlezero(), and pmap_copy_page() will never have any mappings. Therefore, it is unnecessary to do a cache clean for that page. Add assertions in #ifdef DEBUG that assert this invariant. This shaves some cycles off the frequently-called pmap_zero_page() and pmap_copy_page() (no need to look up the dst page's vm_page structure, and one less function call to clean the page).
1.70	25-Mar-2002	thorpej	* Fix use of pmap_curmaxkvaddr. * Use the PTP hint in the pmap.
1.69	25-Mar-2002	thorpej	Move some private pmap data structures into pmap.c
1.68	24-Mar-2002	thorpej	Garbage-collect pmap_pte() (and good riddance!)
1.67	24-Mar-2002	chris	remove pointless pg = NULL in else part of if (pg != NULL)
1.66	24-Mar-2002	thorpej	pmap_enter(): Use pmap_map_ptes() correctly.
1.65	24-Mar-2002	chris	Update pmap_copy_page to only map in the src readonly and only invalidate it after the copy, no need for it to flush the wb.
1.64	24-Mar-2002	thorpej	pmap_allocpagedir(): Don't use pmap_pte(), and simplify a little.
1.63	24-Mar-2002	thorpej	pmap_handled_emulation(): Fix locking protocol botch. XXX Should we traverse the PV list and enable all PTEs?
1.62	24-Mar-2002	thorpej	pmap_handled_emulation(): Use pmap_map_ptes() correctly.
1.61	24-Mar-2002	thorpej	pmap_modified_emulation(): Use pmap_map_ptes() correctly.
1.60	24-Mar-2002	thorpej	pmap_unwire(): Use pmap_map_ptes() correctly.
1.59	24-Mar-2002	thorpej	pmap_clearbit(): Use pmap_map_ptes() correctly.
1.58	24-Mar-2002	thorpej	Use pmap_is_curpmap() consistently.
1.57	24-Mar-2002	thorpej	Clean up the PTP allocation functions a bit.
1.56	24-Mar-2002	thorpej	* arm_byte_to_page() -> arm_btop() * arm_page_to_byte() -> arm_ptob()
1.55	24-Mar-2002	thorpej	Remove some redundant tests in pmap_enter().
1.54	23-Mar-2002	thorpej	Garbage-collect the "pagehook" stuff.
1.53	23-Mar-2002	thorpej	* Rename PROCESS_PAGE_TBLS_BASE -> PTE_BASE * Rename ALT_PAGE_TBLS_BASE -> APTE_BASE * Garbage-collect PAGE_TABLE_SPACE_START
1.52	08-Mar-2002	thorpej	Pool deals fairly well with physical memory shortage, but it doesn't deal with shortages of the VM maps where the backing pages are mapped (usually kmem_map). Try to deal with this: * Group all information about the backend allocator for a pool in a separate structure. The pool references this structure, rather than the individual fields. * Change the pool_init() API accordingly, and adjust all callers. * Link all pools using the same backend allocator on a list. * The backend allocator is responsible for waiting for physical memory to become available, but will still fail if it cannot callocate KVA space for the pages. If this happens, carefully drain all pools using the same backend allocator, so that some KVA space can be freed. * Change pool_reclaim() to indicate if it actually succeeded in freeing some pages, and use that information to make draining easier and more efficient. * Get rid of PR_URGENT. There was only one use of it, and it could be dealt with by the caller. From art@openbsd.org.
1.51	06-Mar-2002	chris	Mostly style changes to stop us directly referencing tqh_first, and use TAILQ_FIRST instead. Based on rev 1.130 of the i386 pmap.c.
1.50	05-Mar-2002	thorpej	* Make pmap_is_{modified,referenced}() macros in pmap.h that just test the attributes in the vm_page_md directly. * Clean up pmap_clear_{modified,referenced}(). * Delete now-unused pmap_testbit().
1.49	05-Mar-2002	thorpej	Switch back to using vm_page_md (thanks chuq for finding the bug in the code that made it unstable before!)
1.48	03-Mar-2002	chris	Implement pmap_growkernel for arm32 based ports. Note that this has been compiled on some systems, cats, IQ80310, IPAQ, netwinder and shark (note that shark's build is currently broken due to other reasons), but only actually run on cats. Shark doesn't make use of the functionality as I believe there has to be a correlation between OFW and the kernel tables so that calls into OFW work.
1.47	22-Feb-2002	thorpej	Change pmap_map_entry() to work like pmap_map_chunk(): take a pointer to the L1 table and a virtual address, and no pointer to the L2 table. The L2 table will be looked up by pmap_map_entry(), which will panic if the there is no L2 table for the requested VA. NOTE: IT IS EXTREMELY IMPORTANT THAT THE CORRECT VIRTUAL ADDRESS BE PROVIDED TO pmap_map_entry()! Notably, the code that mapped the kernel L2 tables into the kernel PT mapping L2 table were not passing actual virtual addresses, but rather offsets into the range mapped by the L2 table. I have fixed up all of these call sites, and tested the resulting kernel on both an IQ80310 and a Shark. Other portmasters should examine their pmap_map_entry() calls if their new kernels fail.
1.46	21-Feb-2002	thorpej	Keep track of which kernel PTs are available during bootstrap, and let pmap_map_chunk() lookup the correct one to use for the current VA. Eliminate the "l2table" argument to pmap_map_chunk(). Add a second L2 table for mapping kernel text/data/bss on the IQ80310 (fixes booting kernels with ramdisks).
1.45	21-Feb-2002	thorpej	In pmap_map_chunk(), if we can't use a section mapping, then make sure that the L1 slot for the current VA points to an L2 table, and panic if it doesn't.
1.44	21-Feb-2002	thorpej	Always pass the L1 table to pmap_map_chunk(). This allows pmap_map_chunk() to perform some error checking.
1.43	21-Feb-2002	thorpej	map_chunk() -> pmap_map_chunk(), and move it to pmap.c
1.42	20-Feb-2002	thorpej	map_pagetable() -> pmap_link_l2pt(), and move it to pmap.c
1.41	20-Feb-2002	thorpej	Collapse map_entry{,ro,nc}() into a single pmap_map_entry() that takes a prot and a "cacheable" indicator.
1.40	20-Feb-2002	thorpej	Rename map_section() to pmap_map_section(), move it to pmap.c, and give it an extra argument (prot - specifies protection of the mapping).
1.39	06-Feb-2002	thorpej	Back out all the vm_page_md changes. They are causing some mysterious problems (a similar change to the i386 pmap causes mysterious problems there, as well), and the issue needs to be investigated more.
1.38	06-Feb-2002	thorpej	Efficiency tweaks, some made possible by vm_page_md.
1.37	05-Feb-2002	thorpej	Use vm_page_md rather than pmap_physseg. Saves lots of cycles in common operations.
1.36	25-Jan-2002	thorpej	Overhaul of the ARM cache code. This is mostly a simplification pass. Rather than providing a whole slew of cache operations that aren't ever used, distill them down to some useful primitives: icache_sync_all Synchronize I-cache icache_sync_range Synchronize I-cache range dcache_wbinv_all Write-back and Invalidate D-cache dcache_wbinv_range Write-back and Invalidate D-cache range dcache_inv_range Invalidate D-cache range dcache_wb_range Write-back D-cache range idcache_wbinv_all Write-back and Invalidate D-cache, Invalidate I-cache idcache_wbinv_range Write-back and Invalidate D-cache, Invalidate I-cache range Note: This does not yet include an overhaul of the actual asm files that implement the primitives. Instead, we've provided a safe default for each CPU type, and the individual CPU types can now be optimized one at a time.
1.35	20-Jan-2002	thorpej	Some prototype cleanup.
1.34	17-Jan-2002	thorpej	Teach pmap_extract() about section mappings.
1.33	05-Jan-2002	chris	Make some of the arm32 files build with LOOSE_PROTOTYPES not set in the makefile. Turned up a few mismatched functions. Note that this isn't all of the arm32 files. Aim will be to get arm32 kernels built with LOOSE_PROTOTYPES not set.
1.32	22-Nov-2001	thorpej	Add cpu_cpwait() calls after TLB updates that are not expected to be followed up by a pmap_update().
1.31	19-Nov-2001	thorpej	Implement pmap_update(). Currently it just calls cpu_cpwait(), which ensures that TLB/cache operations have completed.
1.30	03-Nov-2001	rearnsha	branches: 1.30.2; Replace most uses of pmap_pde_p with pmap_pde_page, since that is what we need later in the code. This fixes a fatal kernel fault in pmap_modified_emulation if a user application tries to access a kernel address that is section-mapped. Add a diagnostic that detects attempts to call pmap_kenter_pa with a va that is section-mapped.
1.29	01-Nov-2001	rearnsha	When clearing the modified bit for modified emulation, don't turn caching on for a page just because we are clearing the writable bit in the PTE: this is incompatible with the way pmap_vac_me_harder works, and the code in the modified emulation handler doesn't know about recalculating the cachable attributes (nor should it, IMO). Also, if we are invalidating a page, flush its TLB entry; for some reason we were only doing this when clearing the Write or modified bits. These patches together seem to solve the random seg-faults that were still occuring occasionally under heavy paging.
1.28	18-Oct-2001	rearnsha	branches: 1.28.2; Add a comment describing the logic implemented by pmap_vac_me_harder.
1.27	18-Oct-2001	rearnsha	On processors that support both write-through and write-back cacheing (eg ARM920), the mode in which the processor operates is governed by the use of both the PT_C and PT_B bits: PT_C=1,PT_B=1 -> Write-back PT_C=1,PT_B=0 -> Write-through To support this define pte_cache_mode (initialized to PT_C\|PT_B) and use that when enabling cacheing for a page.
1.26	18-Oct-2001	rearnsha	With a diagnostic kernel, printing out a message each time we fail to allocate a L1 pt is often enough to bring the system to its knees: so make the messages PDEBUG(0,...). However, even with this step having more than a small number of processes searching for a L1 pt can still be enough to bring the system down, since they all run at high priority and sleep for very little time, thus blocking out user code from completing. So implement an exponential backoff when waiting for a page table, so that we don't hog the CPU when memory is scarce. Tested by running a make of the C compiler with "gnumake -j30" (and plenty of swap space).
1.25	18-Oct-2001	rearnsha	Fix pmap_vac_me_harder to take into account pages that are mapped into both kernel and user space. Fixes port-arm32/13998.
1.24	29-Sep-2001	chris	Add a couple of simplelocks to make sure we call pagealloc with the uvm_object locked. Found while running a LOCKDEBUG kernel on cats.
1.23	15-Sep-2001	chs	a whole bunch of changes to improve performance and robustness under load: - remove special treatment of pager_map mappings in pmaps. this is required now, since I've removed the globals that expose the address range. pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's no longer any need to special-case it. - eliminate struct uvm_vnode by moving its fields into struct vnode. - rewrite the pageout path. the pager is now responsible for handling the high-level requests instead of only getting control after a bunch of work has already been done on its behalf. this will allow us to UBCify LFS, which needs tighter control over its pages than other filesystems do. writing a page to disk no longer requires making it read-only, which allows us to write wired pages without causing all kinds of havoc. - use a new PG_PAGEOUT flag to indicate that a page should be freed on behalf of the pagedaemon when it's unlocked. this flag is very similar to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the pageout fails due to eg. an indirect-block buffer being locked. this allows us to remove the "version" field from struct vm_page, and together with shrinking "loan_count" from 32 bits to 16, struct vm_page is now 4 bytes smaller. - no longer use PG_RELEASED for swap-backed pages. if the page is busy because it's being paged out, we can't release the swap slot to be reallocated until that write is complete, but unlike with vnodes we don't keep a count of in-progress writes so there's no good way to know when the write is done. instead, when we need to free a busy swap-backed page, just sleep until we can get it busy ourselves. - implement a fast-path for extending writes which allows us to avoid zeroing new pages. this substantially reduces cpu usage. - encapsulate the data used by the genfs code in a struct genfs_node, which must be the first element of the filesystem-specific vnode data for filesystems which use genfs_{get,put}pages(). - eliminate many of the UVM pagerops, since they aren't needed anymore now that the pager "put" operation is a higher-level operation. - enhance the genfs code to allow NFS to use the genfs_{get,put}pages instead of a modified copy. - clean up struct vnode by removing all the fields that used to be used by the vfs_cluster.c code (which we don't use anymore with UBC). - remove kmem_object and mb_object since they were useless. instead of allocating pages to these objects, we now just allocate pages with no object. such pages are mapped in the kernel until they are freed, so we can use the mapping to find the page to free it. this allows us to remove splvm() protection in several places. The sum of all these changes improves write throughput on my decstation 5000/200 to within 1% of the rate of NetBSD 1.5 and reduces the elapsed time for "make release" of a NetBSD 1.5 source tree on my 128MB pc to 10% less than a 1.5 kernel took.
1.22	13-Sep-2001	chris	Sprinkle some static and inline into a couple of functions. Remove dead entries from pmap.h.
1.21	13-Sep-2001	chris	Update pmap_clearbit to flush the cache if the area is being made readonly. Also only do this and the tlb flush if the pmap is the current pmap.
1.20	10-Sep-2001	chris	Update the pmap following some comments from Chuck Silvers: Remove some overzealous locking of HEAD_TO_MAP Remove a potential deadlock in pmap_copy_page Change alloc and free l1pt to use kenter/kremove. Update pmap_map to use kenter (only actually used by dumpsys, so no matching kremove)
1.19	10-Sep-2001	chris	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.
1.18	11-Aug-2001	chris	branches: 1.18.2; Fix compile without DIAGNOSTICs enabled
1.17	11-Aug-2001	chris	Checking a whole host of pmap changes: Improved locking (not that we actually use it on a uniprocessor, but one day :) Removed unneeded splvm's tweaked pmap_clean_page code to only flush the cache if the page is mapped in the current pmap (based on diff from richard E) Adopted pv entry allocation mechanism from i386. Laid framework for returning ptp's when we've finished with them rather than holding onto them till the process exits. ptp's are now allocated with a uvm object for the pmap, means that we can walk a list to free them off in pmap_release, until they get freed off by pmap_remove. Also implemented a page zeroing function when the processor is idling. Note that hpcarm may wish to disable this. I believe this code to be stable, if anyone has any problems please shout up.
1.16	29-Jul-2001	chris	Rework the pmap_release code to not have to walk the ptpt, it now uses a uvm_object to track the allocated vm_pages, this means it can free off the entries in the uvm_object. Testing shows that it's about 5% faster on the make configure step for gmake.
1.15	28-Jul-2001	chris	A couple of tidy ups to pmap: pmap_t -> struct pmap * in pmap.c and pmap.h kernel_pmap -> pmap_kernel() everywhere. Compiled and booted on riscpc and cats.
1.14	08-Jul-2001	chs	branches: 1.14.2; clean up pmap_k{enter_pa,remove}(): a page is allowed to be mapped normally when k-mapped. use UVM_PGA_ZERO to get zeroed pages rather than zeroing here.
1.13	06-Jul-2001	chris	Implement proper versions of kenter_pa and kremove, I've based them on versions provided by Richard E.
1.12	25-Jun-2001	chris	Improve the vac_me_harder function, it is now slightly faster, however pmap_enter_pv is also now slightly slower, so they appear to balance out. Note that I've some ideas in the works on how to improve the pv handling, so the slow down is short term only. Also added non-advertising licence and copyright to myself and richard.
1.11	24-Jun-2001	chris	Add a pmap_map_ptes function to map another process page table entries into the current pmap (similair to pmap_pte) but this allows multiple use of it, rather than repeated calls of pmap_pte for pte, map_ptes returns a pointer to the va where the entries were mapped so it can be reused ptes can be found by indexing into it. Update some of the functions that use pmap_pte to pmap_map_ptes. Note that there's a dummy macro for pmap_unmap_ptes, this is because at some point locking will be needed, so we need to be able to unlock them. Performance gain seems to be minimal, however long term it should help improve things. This is similair to the i386 pmap_map_ptes, however it's based on a version from Richard Earnshaw.
1.10	22-Jun-2001	chris	Use a pool to store pmap structs.
1.9	26-May-2001	chs	replace vm_page_t with struct vm_page *.
1.8	25-Apr-2001	thorpej	Garbage-collect pmap_page_index().
1.7	24-Apr-2001	thorpej	Sprinkle pmap_update() calls after calls to: - pmap_enter() - pmap_remove() - pmap_protect() - pmap_kenter_pa() - pmap_kremove() as described in pmap(9). These calls are relatively conservative. It may be possible to optimize these a little more.
1.6	22-Apr-2001	thorpej	Remove pmap_kenter_pgs(). It was never really adopted by anything, and the interface itself wasn't as flexible as callers would have probably liked.
1.5	20-Apr-2001	toshii	In pmap_allocpagedir(), check if uvm_km_zalloc of ptpt is successful, and handle alloc failure case.
1.4	15-Mar-2001	chs	eliminate the KERN_* error codes in favor of the traditional E* codes. the mapping is: KERN_SUCCESS 0 KERN_INVALID_ADDRESS EFAULT KERN_PROTECTION_FAILURE EACCES KERN_NO_SPACE ENOMEM KERN_INVALID_ARGUMENT EINVAL KERN_FAILURE various, mostly turn into KASSERTs KERN_RESOURCE_SHORTAGE ENOMEM KERN_NOT_RECEIVER <unused> KERN_NO_ACCESS <unused> KERN_PAGES_LOCKED <unused>
1.3	04-Mar-2001	matt	branches: 1.3.2; 1.3.4; more vm_offset_t/vm_size_t -> {p,v}{addr,size}_t changes move pmap_* declarations to pmap.h. fix conflicts this exposed (e.g different definitions for pmap_bootstrap).
1.2	04-Mar-2001	matt	Convert some vm_size_t to vsize_t/psize_t. Change vaddr_t to paddr_t where appropriate.
1.1	04-Mar-2001	matt	Move from arm32/arm32. s/vm_offset_t/vaddr_t/g
1.3.4.4	23-Apr-2001	bouyer	Sync with HEAD.
1.3.4.3	27-Mar-2001	bouyer	Sync with HEAD.
1.3.4.2	12-Mar-2001	bouyer	Sync with HEAD.
1.3.4.1	04-Mar-2001	bouyer	file pmap.c was added on branch thorpej_scsipi on 2001-03-12 13:27:20 +0000
1.3.2.2	21-Jun-2001	nathanw	Catch up to -current.
1.3.2.1	09-Apr-2001	nathanw	Catch up with -current.
1.14.2.9	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
1.14.2.8	06-Sep-2002	jdolecek	sync kqueue branch with HEAD
1.14.2.7	23-Jun-2002	jdolecek	catch up with -current on kqueue branch
1.14.2.6	16-Mar-2002	jdolecek	Catch up with -current.
1.14.2.5	11-Feb-2002	jdolecek	Sync w/ -current.
1.14.2.4	10-Jan-2002	thorpej	Sync kqueue branch with -current.
1.14.2.3	13-Sep-2001	thorpej	Update the kqueue branch to HEAD.
1.14.2.2	25-Aug-2001	thorpej	Merge Aug 24 -current into the kqueue branch.
1.14.2.1	03-Aug-2001	lukem	update to -current
1.18.2.1	01-Oct-2001	fvdl	Catch up with -current.
1.28.2.1	12-Nov-2001	thorpej	Sync the thorpej-mips-cache branch with -current.
1.30.2.19	11-Dec-2002	thorpej	Sync with HEAD.
1.30.2.18	11-Nov-2002	nathanw	Catch up to -current
1.30.2.17	18-Oct-2002	nathanw	Catch up to -current.
1.30.2.16	17-Sep-2002	nathanw	Catch up to -current.
1.30.2.15	28-Aug-2002	wrstuden	There is no 'p', there is only 'l'.
1.30.2.14	27-Aug-2002	thorpej	Sync with -current.
1.30.2.13	13-Aug-2002	thorpej	From trunk: Add the brutal hack that allows us to limp along using the read/write cache line allocation policy on XScale CPUs: in pmap_enter(), if the pmap is the kernel pmap, clear the X-bit in the PTE, thus disabling read/write-allocate for managed kernel mappings. Yes, this is ugly. But it makes userland code run with r/w-allocate, which is a huge improvement on systems with low core memory performance.
1.30.2.12	13-Aug-2002	nathanw	Catch up to -current.
1.30.2.11	01-Aug-2002	nathanw	Catch up to -current.
1.30.2.10	02-Jul-2002	nathanw	Several curlwp references here can revert to curproc.
1.30.2.9	24-Jun-2002	nathanw	Curproc->curlwp renaming. Change uses of "curproc->l_proc" back to "curproc", which is more like the original use. Bare uses of "curproc" are now "curlwp". "curproc" is now #defined in proc.h as ((curlwp) ? (curlwp)->l_proc) : NULL) so that it is always safe to reference curproc (dereferencing curproc is another story, but that's always been true).
1.30.2.8	20-Jun-2002	nathanw	Catch up to -current.
1.30.2.7	17-Apr-2002	nathanw	Catch up to -current.
1.30.2.6	01-Apr-2002	nathanw	Catch up to -current. (CVS: It's not just a program. It's an adventure!)
1.30.2.5	28-Feb-2002	nathanw	Catch up to -current.
1.30.2.4	11-Jan-2002	nathanw	More catchup.
1.30.2.3	08-Jan-2002	nathanw	Catch up to -current.
1.30.2.2	15-Nov-2001	thorpej	Machine-dependent kernel mods for scheduler activations on 32-bit ARM processors. Kernel boots multi-user on an XScale, but upcalls not yet tested.
1.30.2.1	03-Nov-2001	thorpej	file pmap.c was added on branch nathanw_sa on 2001-11-15 06:39:22 +0000
1.97.4.6	14-Feb-2003	he	Revert pull-up of revision 1.110 (requested by rearnsha in ticket #1170): Revert cached memory access to L1 tables, this causes instability on the release branch.
1.97.4.5	07-Dec-2002	he	Pull up revision 1.111 (requested by thorpej in ticket #714): Use a pool cache for PT-PTs.
1.97.4.4	07-Dec-2002	he	Pull up revision 1.105 (requested by thorpej in ticket #714): Add code, conditional on PMAP_ALIAS_DEBUG, which can be used to hunt for virtual aliases between managed (pmap_enter) and unmanaged (pmap_kenter_pa) mappings.
1.97.4.3	21-Nov-2002	he	Pull up revision 1.110 (requested by thorpej in ticket #712): Do cached memory access to L1 tables, making sure to write-back the cache after any L1 table modifications.
1.97.4.2	16-Nov-2002	he	Pull up revision 1.107 (requested by thorpej in ticket #662): Clean up some warts in pmap_protect().
1.97.4.1	31-Jul-2002	lukem	Pull up revision 1.100 (requested by thorpej in ticket #587): Don't use pmap_kenter_pa() in pmap_map(); doing so causes an assertion failure in pmap_kenter_pa().
1.97.2.2	30-Aug-2002	gehenna	catch up with -current.
1.97.2.1	14-Jul-2002	gehenna	catch up with -current.
1.137.2.7	11-Dec-2005	christos	Sync with head.
1.137.2.6	10-Nov-2005	skrll	Sync with HEAD. Here we go again...
1.137.2.5	01-Apr-2005	skrll	Sync with HEAD.
1.137.2.4	17-Jan-2005	skrll	Sync with HEAD.
1.137.2.3	21-Sep-2004	skrll	Fix the sync with head I botched.
1.137.2.2	18-Sep-2004	skrll	Sync with HEAD.
1.137.2.1	03-Aug-2004	skrll	Sync with HEAD
1.150.4.1	28-Jan-2005	yamt	convert arch/arm to new apis.
1.150.2.1	29-Apr-2005	kent	sync with -current
1.154.2.6	21-Jan-2008	yamt	sync with head
1.154.2.5	15-Nov-2007	yamt	sync with head.
1.154.2.4	27-Oct-2007	yamt	sync with head.
1.154.2.3	03-Sep-2007	yamt	sync with head.
1.154.2.2	26-Feb-2007	yamt	sync with head.
1.154.2.1	21-Jun-2006	yamt	sync with head.
1.157.24.1	21-Feb-2007	snj	branches: 1.157.24.1.4; Pull up following revision(s) (requested by matt in ticket #457): sys/arch/arm/arm32/pmap.c: revision 1.158 From Scott Allan in http://mail-index.netbsd.org/port-arm/2006/07/26/0000.html I ran into a problem when I tried to set up a mapping that started at virtual address 0xFFF00000 and was 0x00100000 long. In other words, the mapping should have gone to the end of the 32 bit address space. The mapping was made with no problem, but pmap_devmap_find_va() wouldn't find an address within the mapping. For example, if I told it to find a mapping for 0x1000 bytes at 0xFFF01000, it would try to make sure that 0xFFF01000 was greater than 0xFFF00000 and that (0xFFF01000+0x1000) was less than (0xFFF00000 +0x00100000). However, that last expression (0xFFF00000+0x00100000) wrapped around to be simply 0x00000000 so it wasn't found. This patch fixes this problem in pmap_devmap_find_va() and pmap_devmap_find_pa() by subtracting one off of the
1.157.24.1.4.3	10-Nov-2007	matt	Refix thinko. (deal with non-page-aligned starts/lengths)
1.157.24.1.4.2	10-Nov-2007	matt	Fix thinko.
1.157.24.1.4.1	10-Nov-2007	matt	Add AT91 support from Sami Kantoluoto Add TI OMAP2430 support from Marty Fouts @ Danger Inc
1.157.20.1	12-Jan-2007	ad	Sync with head.
1.158.2.5	15-Apr-2007	yamt	sync with head.
1.158.2.4	29-Mar-2007	skrll	Adapt arm32. Thanks to scw for helping out. Tested on my cats (SA1) XXX hydra should die. i've made some changes, but no guarantees.
1.158.2.3	24-Mar-2007	yamt	sync with head.
1.158.2.2	12-Mar-2007	rmind	Sync with HEAD.
1.158.2.1	27-Feb-2007	yamt	- sync with head. - move sched_changepri back to kern_synch.c as it doesn't know PPQ anymore.
1.161.2.6	03-Dec-2007	ad	Sync with HEAD.
1.161.2.5	19-Oct-2007	ad	Sync with head.
1.161.2.4	09-Oct-2007	ad	Sync with head.
1.161.2.3	27-May-2007	ad	Sync with head.
1.161.2.2	10-Apr-2007	ad	Sync with head.
1.161.2.1	13-Mar-2007	ad	Sync with head.
1.162.2.1	11-Jul-2007	mjf	Sync with head.
1.163.4.3	16-Oct-2007	garbled	Sync with HEAD
1.163.4.2	03-Oct-2007	garbled	Sync with HEAD
1.163.4.1	22-May-2007	matt	Update to HEAD.
1.164.12.10	04-Mar-2008	matt	Fix a bug in pmap_flush_page which could nuke a PTE being used by pmap_zero_page or pmap_copy_page.
1.164.12.9	09-Jan-2008	matt	sync with HEAD
1.164.12.8	10-Nov-2007	matt	Fix thinko. (need to deal non-page-aligned size/lengths)
1.164.12.7	09-Nov-2007	matt	Add pmap_icache_sync_range and change arm32_sync_icache to use it. This will only invalidate va that have valid PTEs. This avoids cleaning unneeded cache lines.
1.164.12.6	09-Nov-2007	matt	Make all the evbarm kernels build again. Fix lossage from rebase.
1.164.12.5	08-Nov-2007	matt	sync with -HEAD
1.164.12.4	07-Nov-2007	matt	Fix botched color assertion
1.164.12.3	06-Nov-2007	matt	sync with HEAD
1.164.12.2	12-Oct-2007	matt	Import TI OMAP 2430 and ARM11/ARMv6 support. Now on ARMv6, the cache is no longer purged on context switches.
1.164.12.1	29-Aug-2007	matt	Use ci_intr_depth
1.164.10.2	20-Jan-2008	chris	Sync to HEAD.
1.164.10.1	01-Jan-2008	chris	Sync with HEAD.
1.164.8.3	11-Nov-2007	joerg	Sync with HEAD.
1.164.8.2	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.
1.164.8.1	02-Oct-2007	joerg	Sync with HEAD.
1.164.6.3	28-Feb-2008	rjs	Sync with HEAD.
1.164.6.2	26-Dec-2007	rjs	Sync with HEAD.
1.164.6.1	01-Nov-2007	rjs	Sync with HEAD.
1.165.2.1	14-Oct-2007	yamt	sync with head.
1.166.2.1	13-Nov-2007	bouyer	Sync with HEAD
1.167.2.2	18-Feb-2008	mjf	Sync with HEAD.
1.167.2.1	19-Nov-2007	mjf	Sync with HEAD.
1.169.6.2	08-Jan-2008	bouyer	Sync with HEAD
1.169.6.1	02-Jan-2008	bouyer	Sync with HEAD
1.171.6.7	17-Jan-2009	mjf	Sync with HEAD.
1.171.6.6	05-Oct-2008	mjf	Sync with HEAD.
1.171.6.5	28-Sep-2008	mjf	Sync with HEAD.
1.171.6.4	29-Jun-2008	mjf	Sync with HEAD.
1.171.6.3	05-Jun-2008	mjf	Sync with HEAD. Also fix build.
1.171.6.2	02-Jun-2008	mjf	Sync with HEAD.
1.171.6.1	03-Apr-2008	mjf	Sync with HEAD.
1.172.2.2	17-Jun-2008	yamt	sync with head.
1.172.2.1	18-May-2008	yamt	sync with head.
1.173.2.4	11-Aug-2010	yamt	sync with head.
1.173.2.3	11-Mar-2010	yamt	sync with head
1.173.2.2	04-May-2009	yamt	sync with head.
1.173.2.1	16-May-2008	yamt	sync with head.
1.175.2.3	10-Oct-2008	skrll	Sync with HEAD.
1.175.2.2	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.175.2.1	23-Jun-2008	wrstuden	Sync w/ -current. 34 merge conflicts to follow.
1.176.2.4	18-Jul-2008	simonb	Sync with head.
1.176.2.3	03-Jul-2008	simonb	Sync with head.
1.176.2.2	27-Jun-2008	simonb	Sync with head.
1.176.2.1	18-Jun-2008	simonb	Sync with head.
1.180.2.2	13-Dec-2008	haad	Update haad-dm branch to haad-dm-base2.
1.180.2.1	19-Oct-2008	haad	Sync with HEAD.
1.187.8.2	15-Feb-2014	matt	Merge armv7 support from HEAD, specifically support for the BCM5301X and BCM56340 evbarm kernels.
1.187.8.1	21-Apr-2010	matt	sync to netbsd-5
1.187.4.1	03-Dec-2009	sborrill	Pull up the following revisions(s) (requested by scw in ticket #1168): sys/arch/arm/arm32/pmap.c: revision 1.203 Work-around a possible process exit corner case which can leave stale data in the cache after a context-switch. Addresses kern/41058.
1.187.2.2	28-Apr-2009	skrll	Sync with HEAD.
1.187.2.1	19-Jan-2009	skrll	Sync with HEAD.
1.195.2.1	13-May-2009	jym	Sync with HEAD. Commit is split, to avoid a "too many arguments" protocol error.
1.211.2.23	15-Nov-2010	uebayasi	Revert xmd(4).
1.211.2.22	10-Nov-2010	uebayasi	Fix thinko; make vm_physseg ptr swap really work.
1.211.2.21	10-Nov-2010	uebayasi	opt_xip.h isn't needed any more here.
1.211.2.20	10-Nov-2010	uebayasi	Always use VM_PHYSMEM_PTR().
1.211.2.19	06-Nov-2010	uebayasi	Sync with HEAD.
1.211.2.18	31-Oct-2010	uebayasi	We already have a flag PMAP_NOCACHE. s/PMAP_UNMANAGED/PMAN_NOCACHE/. Pointed out by Chuck Silvers, thanks.
1.211.2.17	31-Oct-2010	uebayasi	More struct vm_page * -> struct vm_page_md * adjustments.
1.211.2.16	30-Oct-2010	uebayasi	Implement pmap_physload_device(9) to replace xmd(4) MD backend. Implement pmap_mmap(9) and use it from mem(4) and xmd(4).
1.211.2.15	17-Aug-2010	uebayasi	Sync with HEAD.
1.211.2.14	07-Jul-2010	uebayasi	Clean up; merge options DIRECT_PAGE into options XIP.
1.211.2.13	31-May-2010	uebayasi	Re-define the definition of "device page"; device pages are pages of device memory. Pages which don't have vm_page (== can't be used for generic use), but whose PV are tracked, are called "direct pages" from now.
1.211.2.12	30-Apr-2010	uebayasi	Sync with HEAD.
1.211.2.11	28-Apr-2010	uebayasi	Always use struct vm_physseg *vm_physmem_ptrs[] in MD code.
1.211.2.10	27-Apr-2010	uebayasi	Support PMAP_UNMANAGED in some pmaps. (Others should be converted eventually, but no problem while managed device page is not used.)
1.211.2.9	25-Feb-2010	uebayasi	A few more VM_PAGE_TO_MD().
1.211.2.8	20-Feb-2010	uebayasi	Fix \!DIAGNOSTIC build.
1.211.2.7	10-Feb-2010	uebayasi	Adjust previous.
1.211.2.6	10-Feb-2010	uebayasi	Replace all remaining pg->mdpage references with VM_PAGE_TO_MD(). Now struct vm_page * is fully opaque.
1.211.2.5	10-Feb-2010	uebayasi	Fix previous again & use VM_PAGE_TO_MD() where appropriate.
1.211.2.4	10-Feb-2010	uebayasi	Convert pmap_enter() and pmap_vac_me_harder().
1.211.2.3	10-Feb-2010	uebayasi	Convert pmap_enter_pv().
1.211.2.2	10-Feb-2010	uebayasi	Convert pmap_remove_pv() / pmap_modify_pv() to take struct vm_page_md *.
1.211.2.1	10-Feb-2010	uebayasi	Start changing this to be ready for device page (XIP). The basic rule is device pages don't have struct vm_page * objects. Instead per-page data (mainly PV mappings) is rooted from the global struct vm_page_md array. Convert 2 functions to take struct vm_page_md * instead of struct vm_page *.
1.212.2.6	19-May-2011	rmind	Implement sharing of vnode_t::v_interlock amongst vnodes: - Lock is shared amongst UVM objects using uvm_obj_setlock() or getnewvnode(). - Adjust vnode cache to handle unsharing, add VI_LOCKSHARE flag for that. - Use sharing in tmpfs and layerfs for underlying object. - Simplify locking in ubc_fault(). - Sprinkle some asserts. Discussed with ad@.
1.212.2.5	21-Apr-2011	rmind	sync with head
1.212.2.4	05-Mar-2011	rmind	sync with head
1.212.2.3	03-Jul-2010	rmind	sync with head
1.212.2.2	30-May-2010	rmind	sync with head
1.212.2.1	16-Mar-2010	rmind	Change struct uvm_object::vmobjlock to be dynamically allocated with mutex_obj_alloc(). It allows us to share the locks among UVM objects.
1.219.4.1	05-Mar-2011	bouyer	Sync with HEAD
1.219.2.1	06-Jun-2011	jruoho	Sync with HEAD.
1.221.2.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.224.6.1	18-Feb-2012	mrg	merge to -current.
1.224.2.5	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")
1.224.2.4	23-Jan-2013	yamt	sync with head
1.224.2.3	16-Jan-2013	yamt	sync with (a bit old) head
1.224.2.2	30-Oct-2012	yamt	sync with head
1.224.2.1	17-Apr-2012	yamt	sync with head
1.228.2.2	09-Feb-2013	riz	Pull up following revision(s) (requested by msaitoh in ticket #803):
1.251	01-Feb-2013	matt	cleanup PVF_WRITE & pvh_attrs interaction.
1.250	31-Jan-2013	skrll	Another typo in a comment.
1.249	31-Jan-2013	skrll	Typo in comment.
1.247	11-Jan-2013	matt	Fix a bug in pmap_modify_pv where we didnt set PVF_WRITE on a page after changing its mapping to writeable. Add more KASSERTS Dont go into DDB by default in pmap_fixup.
1.243	10-Dec-2012	matt	Change a KASSERT to a KASSERTMSG
1.233	29-Aug-2012	matt	Support PMAP_NOCACHE in pmap_kenter_pa
1.232	29-Aug-2012	matt	Use the correct prot mask in vector_page_setprot
1.228.2.1	09-Aug-2012	jdc	branches: 1.228.2.1.2; Pull up revisions: src/sys/arch/evbarm/dev/plcomreg.h revisions 1.2,1.3,1.4 src/sys/arch/evbarm/conf/INTEGRATOR revision 1.65 src/sys/arch/evbarm/dev/plcom.c revisions 1.34,1.35,1.36,1.37,1.38,1.39,1.40 src/sys/arch/evbarm/ifpga/plcom_ifpga.c revisions 1.12,1.13,1.14 src/sys/arch/evbarm/dev/plcomvar.h revisions 1.9,1.10,1.11 src/sys/arch/evbarm/ifpga/plcom_ifpgavar.h revision 1.2 src/sys/arch/arm/arm/cpufunc.c revisions 1.105,1.108 src/sys/arch/arm/arm32/cpu.c revision 1.79 src/sys/arch/arm/include/armreg.h revisions 1.49,1.54 src/sys/arch/arm/arm32/pmap.c revision 1.229 src/sys/arch/arm/arm32/arm32_machdep.c revision 1.77 src/sys/arch/arm/include/cpu.h revision 1.64 src/sys/arch/arm/arm/cpufunc_asm_arm1136.S revision 1.3 src/sys/arch/arm/arm/cpufunc_asm_arm11x6.S revision 1.1 src/sys/arch/arm/conf/files.arm revision 1.106 src/sys/arch/arm/include/cpufunc.h revision 1.57 src/sys/dev/sdmmc/sdhc.c revisions 1.14,1.24 src/sys/dev/sdmmc/sdhcvar.h revisions 1.7,1.8 src/sys/arch/evbarm/ifpga/ifpgareg.h revision 1.4 src/sys/arch/evbarm/integrator/integrator_machdep.c revision 1.69 src/sys/arch/arm/broadcom/bcm2835_dma.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_emmc.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_intr.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_intr.h revision 1.1 src/sys/arch/arm/broadcom/bcm2835_obio.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_plcom.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_pm.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_pmvar.h revision 1.1 src/sys/arch/arm/broadcom/bcm2835_space.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835_tmr.c revision 1.1 src/sys/arch/arm/broadcom/bcm2835reg.h revision 1.1 src/sys/arch/arm/broadcom/bcm2835var.h revision 1.1 src/sys/arch/arm/broadcom/bcm_amba.h revision 1.1 src/sys/arch/arm/broadcom/files.bcm2835 revision 1.1 src/sys/arch/evbarm/Makefile revision 1.9 src/sys/arch/evbarm/conf/RPI revision 1.1 src/sys/arch/evbarm/conf/files.rpi revision 1.1 src/sys/arch/evbarm/conf/mk.rpi revision 1.1 src/sys/arch/evbarm/conf/std.rpi revision 1.1 src/sys/arch/evbarm/rpi/genassym.cf revision 1.1 src/sys/arch/evbarm/rpi/rpi.h revision 1.1 src/sys/arch/evbarm/rpi/rpi_machdep.c revision 1.1 src/sys/arch/evbarm/rpi/rpi_start.S revision 1.1,1.2 src/etc/etc.evbarm/Makefile.inc revision 1.28 (requested by skrll in ticket #454). don't mix #define<TAB> and #define<SPACE> in a file. avoid warning with options PLCOM_DEBUG for INTEGRATOR. Rename register values. No functional change - same code before and after. Existing names are prefixed with PL01X_ where they're common between the PL010 and the PL011. The PL010_/PL011_ prefixes are added where they're found only on the respective chips. Replace the simple_lock with a kmutex_t. Update the locking to match com(4) in the few places it didn't already. DOH. Replace a line that got accidently deleted in the last commit. device_t/softc split struct device * -> device_t struct cfdata * -> cfdata_t Add the 'Z' to the 1176 cpu product name. ok matt@ Fix locking botch introduced in 1.36. Fix a KASSERT. From/OK'ed by matt@ Fix racy softint dispatch that lead to KASSERT(si->si_active) in softint_execute Discussed with matt@. "Looks good to me" Add the documented ARM11[37]6 Auxiliary control register defines. Add support for the ARM1176JZS Add a flag for the lack of LED_ON in HOST_CTL (ti omap3 doesn't do that). Provide a method for attachments to specify capabilites. Add support for the PL011 to plcom. Pull across a bunch of fixes from com(4) while I'm here and do some other tidyup. Tested on a RaspberryPi. PL010 not tested. Initial commit of support for the RaspberryPI (www.raspberrypi.org) This is enough for serial console via the gpio header pins and to get to multiuser. A huge thank you to Matt Thomas for all his help. Add RPI to KERNEL_SETS Remove #if 0 code.
1.228.2.1.2.3	13-Feb-2013	matt	Sync with HEAD
1.228.2.1.2.2	07-Feb-2013	matt	pullup pmap changes from HEAD
1.228.2.1.2.1	28-Nov-2012	matt	Merge improved arm support (especially Cortex) from HEAD including OMAP and BCM53xx support.
1.236.2.5	03-Dec-2017	jdolecek	update from HEAD
1.236.2.4	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.236.2.3	23-Jun-2013	tls	resync from head
1.236.2.2	25-Feb-2013	tls	resync with head
1.236.2.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.257.2.2	18-May-2014	rmind	sync with head
1.257.2.1	28-Aug-2013	rmind	sync with head
1.280.2.1	10-Aug-2014	tls	Rebase.
1.295.2.11	03-Jun-2017	snj	Pull up following revision(s) (requested by skrll in ticket #1424): sys/arch/arm/arm32/pmap.c: revision 1.345 Perform icache syncs for ARM_MMU_EXTENDED as well. This helps the PT_STEP code in pr/52119 and probably other things.
1.295.2.10	11-Mar-2017	snj	fix fallout from ticket #1366: __nothing is nothing on netbsd-7
1.295.2.9	11-Mar-2017	snj	Pull up following revision(s) (requested by skrll in ticket #1366): sys/arch/arm/include/arm32/pmap.h: 1.145 sys/arch/arm/arm32/pmap.c: 1.343, 1.344 sys/arch/evbarm/gumstix/gumstix_machdep.c: 1.58 via patch Fixup the compile time decisions around PMAP_{INCLUDE,NEEDS}_PTE_SYNC and fix the options for xscale boards which require the code in pmap_l2ptp_ctor marked as #ifndef PMAP_INCLUDE_PTE_SYNC. Fix the typo (pte -> opte) in this code block and consistently use opte elsewhere. PR/51990: Regression data_abort_handler: data_aborts fsr=0x406 far=0xbfffeff5 on copyout in init -- fix unused. -- Set xscale_cache_clean_addr appropriately and re-arrange default KVA layout to allow direct map for all boards. OVERO/DUOVERO/PEPPER aren't tested. PR/52010: Regression: Gumstix Verdex is hanging in enabling cache + KASSERT ram_size
1.295.2.8	26-Feb-2016	snj	branches: 1.295.2.8.2; Pull up following revision(s) (requested by skrll in ticket #1106): sys/arch/arm/arm32/pmap.c: revision 1.332 Delete pmap_pmaps and its only user pmap_dump_all. The list wasn't being updated in an MP-safe manner.
1.295.2.7	27-May-2015	msaitoh	Pull up following revision(s) (requested by skrll in ticket #805): sys/arch/arm/include/arm32/pmap.h: revision 1.138 sys/arch/arm/arm/cpufunc.c: revision 1.151 sys/arch/arm/arm32/bus_dma.c: revision 1.90 sys/arch/arm/broadcom/bcm53xx_pax.c: revision 1.14 sys/arch/arm/arm32/bus_dma.c: revision 1.91 sys/arch/arm/samsung/exynos_space.c: revision 1.2 sys/arch/arm/arm32/db_machdep.c: revision 1.23 sys/arch/arm/allwinner/awin_space.c: revision 1.4 sys/arch/arm/include/rwlock.h: revision 1.9 sys/arch/arm/amlogic/amlogic_space.c: revision 1.2 sys/arch/arm/zynq/zynq_space.c: revision 1.2 sys/arch/arm/broadcom/bcm2835_space.c: revision 1.7 sys/arch/arm/arm32/pmap.c: revision 1.317 sys/arch/arm/include/locore.h: revision 1.19 sys/arch/arm/include/mutex.h: revision 1.20 sys/arch/arm/include/lock.h: revision 1.31 sys/arch/arm/include/lock.h: revision 1.32 sys/arch/arm/broadcom/bcmgen_space.c: revision 1.5 - Kill redundant semicolons. - Indentation. - Improve inline asm around dsb/dmb/isb: - always use volatile and mark them as memory barrier - use the common version from locore.h in all places not included from userland
1.295.2.6	26-May-2015	msaitoh	Pull up following revision(s) (requested by skrll in ticket #800): sys/arch/arm/arm32/pmap.c: revision 1.320 sys/arch/arm/arm32/pmap.c: revision 1.321 sys/arch/arm/arm32/pmap.c: revision 1.322 sys/arch/arm/arm32/pmap.c: revision 1.319 - include "opt_arm_debug.h" for VERBOSE_INIT_ARM - Add pmap locking to pmap_kenter_pa/kremove - Make sure nptes is a multiple of PAGE_SIZE / L2_S_SIZE. - Use PDE_SYNC when syncing pdeps
1.295.2.5	19-May-2015	snj	Pull up following revision(s) (requested by joerg in ticket #777): sys/arch/arm/arm32/pmap.c: revision 1.308 pmap_tlb_flushD is for !ARM_MMU_EXTENDED only now
1.295.2.4	10-Apr-2015	snj	Pull up following revision(s) (requested by skrll in ticket #669): sys/arch/arm/arm32/pmap.c: revision 1.318 Fix two bugs. pmap_is_cached fix for MULTIPROCESSOR (not just ASID on local cpu -> any valid ASID on any cpu). pmap_deactivate: update curcpu()->ci_pmap_cur_asid to KERNEL_PID too.
1.295.2.3	10-Nov-2014	martin	Pull up following revision(s) (requested by skrll in ticket #209): sys/arch/arm/pic/pic.c: revision 1.25 sys/arch/arm/arm/cpufunc_asm_armv7.S: revision 1.21 sys/arch/arm/arm32/pmap.c: revision 1.312 sys/arch/arm/arm32/bus_dma.c: revision 1.89 sys/arch/arm/arm32/pmap.c: revision 1.313 sys/arch/arm/arm32/pmap.c: revision 1.314 sys/arch/arm/arm32/pmap.c: revision 1.315 sys/arch/arm/arm32/pmap.c: revision 1.316 Include opt_multiprocessor.h When allocing a l1page, if a page isn't available, use uvm_wait to wait for one to become available. Should fix PR/49364. Post a dmb before invalidating the cache in the post-{read,write} operations to ensure that any/all cachelines brought in via speculation are really flushed. Ensure all memory operations are complete by before wfi. For example, the cpu could have just been in uvm_pageidlezero. In pmap_fault_fixup re-instate the TLB flush for the shared L1 case that occurs for non-ARM_MMU_EXTENDED kernels. This fixes rump/rumpkern/t_sp:stress_killer on rpi which is currently non-ARM_MMU_EXTENDED Remove an unnecessary flush that sneaked in as part of break-before-make change. Remove an XXXNH comment. Update PTE_SYNC_CURRENT to add a dsb for armv7 - part of the break-before-make fix.
1.295.2.2	09-Nov-2014	martin	Pull up following revision(s) (requested by skrll in ticket #188): sys/arch/arm/include/arm32/pmap.h: revision 1.136 sys/arch/arm/include/armreg.h: revision 1.100 sys/arch/arm/cortex/gic.c: revision 1.11 sys/arch/arm/arm32/db_interface.c: revision 1.54 sys/arch/arm/include/armreg.h: revision 1.101 sys/arch/arm/cortex/gic.c: revision 1.12 sys/arch/arm/arm32/arm32_machdep.c: revision 1.107 sys/arch/arm/arm/cpufunc_asm_armv7.S: revision 1.19 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.20 sys/arch/evbarm/conf/BPI: revision 1.5 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.21 sys/arch/arm/arm32/pmap.c: revision 1.306 sys/arch/arm/arm32/db_machdep.c: revision 1.22 sys/arch/arm/arm32/arm32_tlb.c: revision 1.3 sys/arch/arm/arm/undefined.c: revision 1.55 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.22 sys/arch/arm/arm32/pmap.c: revision 1.307 sys/arch/arm/arm32/arm32_tlb.c: revision 1.4 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.23 sys/arch/arm/arm32/arm32_tlb.c: revision 1.5 sys/arch/evbarm/conf/BPI: revision 1.8 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.24 sys/arch/arm/arm32/arm32_tlb.c: revision 1.6 sys/arch/arm/arm32/arm32_tlb.c: revision 1.7 sys/arch/evbarm/conf/CUBIETRUCK: revision 1.5 sys/arch/arm/pic/pic.c: revision 1.23 sys/arch/arm/pic/pic.c: revision 1.24 sys/arch/arm/pic/picvar.h: revision 1.11 sys/arch/arm/arm/cpufunc_asm_armv7.S: revision 1.20 sys/arch/arm/mainbus/cpu_mainbus.c: revision 1.16 sys/arch/arm/arm32/pmap.c: revision 1.298 sys/arch/arm/arm/cpufunc_asm_arm11.S: revision 1.17 sys/arch/arm/arm/cpufunc_asm_pj4b.S: revision 1.5 sys/arch/arm/arm32/pmap.c: revision 1.310 sys/arch/arm/arm32/pmap.c: revision 1.311 sys/arch/arm/arm32/arm32_kvminit.c: revision 1.32 sys/arch/arm/cortex/a9_mpsubr.S: revision 1.19 sys/arch/arm/arm32/arm32_boot.c: revision 1.10 sys/arch/arm/arm/ast.c: revision 1.25 sys/arch/arm/include/armreg.h: revision 1.98 sys/uvm/pmap/pmap_tlb.c: revision 1.10 sys/arch/arm/arm32/arm32_boot.c: revision 1.8 sys/arch/arm/arm32/arm32_boot.c: revision 1.9 sys/arch/arm/arm/arm_machdep.c: revision 1.43 Various ARM MP fixes.
1.295.2.1	30-Oct-2014	martin	Pull up following revision(s) (requested by maxv in ticket #165): sys/arch/newsmips/stand/boot/netif_news.c: revision 1.9 sys/arch/mvme68k/stand/installboot/installboot.c: revision 1.19 sys/arch/arm/arm32/pmap.c: revision 1.300 sys/arch/amiga/dev/siop2.c: revision 1.43 sys/arch/amiga/amiga/disksubr.c: revision 1.62 sys/arch/news68k/news68k/bus_space.c: revision 1.13 sys/arch/amiga/dev/siop.c: revision 1.69 sys/arch/x86/x86/x86_autoconf.c: revision 1.72 Remove dead code in various places under arch/.
1.295.2.8.2.1	13-Mar-2017	skrll	Sync with netbsd-7-1-RELEASE
1.316.2.7	28-Aug-2017	skrll	Sync with HEAD
1.316.2.6	05-Feb-2017	skrll	Sync with HEAD
1.316.2.5	05-Oct-2016	skrll	Sync with HEAD
1.316.2.4	27-Dec-2015	skrll	Sync with HEAD (as of 26th Dec)
1.316.2.3	22-Sep-2015	skrll	Sync with HEAD
1.316.2.2	06-Jun-2015	skrll	Sync with HEAD
1.316.2.1	06-Apr-2015	skrll	Sync with HEAD
1.335.2.4	26-Apr-2017	pgoyette	Sync with HEAD
1.335.2.3	20-Mar-2017	pgoyette	Sync with HEAD
1.335.2.2	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.335.2.1	06-Aug-2016	pgoyette	Sync with HEAD
1.342.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.349.2.3	23-Apr-2019	martin	Pull up following revision(s) (requested by bouyer in ticket #1246): sys/arch/arm/arm32/pmap.c: revision 1.373 (via patch) Fix a deadlock between the pool and pmap codes: - cpu0 grabs the kernel lock (e.g. from a non-MPSAFE interrupt) and calls pool_get(). - cpu1 does a pool_get() on the same pool from MPSAFE code, which needs a pool_page_alloc(), which ends up in pmap_extract_coherency(). So cpu0 holds the kernel_lock and wants the pool lock. cpu1 holds the pool lock and wants the kernel_lock in pmap_extract_coherency(). The pmap code should not rely on kernel_lock. Intead make the pmap_kernel()->pm_obj_lock a IPL_VM lock and use it as pmap lock (thus dropping the pmap test pmap_{acquire,release}_pmap_lock()). This needs to be a IPL_VM because unlike user pmaps, this can be locked from interrupt context. Add a IPL_NONE lock for pmap_growkernel(). We can't use pmap_kernel()->pm_obj_lock here because pmap_grow_map() may sleep. Make pmap_lock (which may be locked with pm_obj_lock held) a IPL_VM lock in all case. reorder a few things to not call pool_get()/pool_put() (which may sleep) with pm_obj_lock held. Patch initially posted to port-arm@ on April 19, improved patch (per suggestions from Nick Hudson and Jason Thorpe) on April 21.
1.349.2.2	23-Apr-2019	martin	Pull up following revision(s) (requested by bouyer in ticket #1245): sys/arch/arm/arm32/pmap.c: revision 1.372 Don't try to aquire/release the pmap lock when in ddb. Avoids a deadlock when entering ddb, or on "mach cpu n" ddb command (the pmap lock may already be held by another CPU, which is halted when entering ddb). Posted to port-arm@ on April 19.
1.349.2.1	02-Nov-2017	snj	Pull up following revision(s) (requested by pgoyette in ticket #335): share/man/man9/kernhist.9: 1.5-1.8 sys/arch/acorn26/acorn26/pmap.c: 1.39 sys/arch/arm/arm32/fault.c: 1.105 via patch sys/arch/arm/arm32/pmap.c: 1.350, 1.359 sys/arch/arm/broadcom/bcm2835_bsc.c: 1.7 sys/arch/arm/omap/if_cpsw.c: 1.20 sys/arch/arm/omap/tiotg.c: 1.7 sys/arch/evbarm/conf/RPI2_INSTALL: 1.3 sys/dev/ic/sl811hs.c: 1.98 sys/dev/usb/ehci.c: 1.256 sys/dev/usb/if_axe.c: 1.83 sys/dev/usb/motg.c: 1.18 sys/dev/usb/ohci.c: 1.274 sys/dev/usb/ucom.c: 1.119 sys/dev/usb/uhci.c: 1.277 sys/dev/usb/uhub.c: 1.137 sys/dev/usb/umass.c: 1.160-1.162 sys/dev/usb/umass_quirks.c: 1.100 sys/dev/usb/umass_scsipi.c: 1.55 sys/dev/usb/usb.c: 1.168 sys/dev/usb/usb_mem.c: 1.70 sys/dev/usb/usb_subr.c: 1.221 sys/dev/usb/usbdi.c: 1.175 sys/dev/usb/usbdi_util.c: 1.67-1.70 sys/dev/usb/usbroothub.c: 1.3 sys/dev/usb/xhci.c: 1.75 sys/external/bsd/drm2/dist/drm/i915/i915_gem.c: 1.34 sys/kern/kern_history.c: 1.15 sys/kern/kern_xxx.c: 1.74 sys/kern/vfs_bio.c: 1.275-1.276 sys/miscfs/genfs/genfs_io.c: 1.71 sys/sys/kernhist.h: 1.21 sys/ufs/ffs/ffs_balloc.c: 1.63 sys/ufs/lfs/lfs_vfsops.c: 1.361 sys/ufs/lfs/ulfs_inode.c: 1.21 sys/ufs/lfs/ulfs_vnops.c: 1.52 sys/ufs/ufs/ufs_inode.c: 1.102 sys/ufs/ufs/ufs_vnops.c: 1.239 sys/uvm/pmap/pmap.c: 1.37-1.39 sys/uvm/pmap/pmap_tlb.c: 1.22 sys/uvm/uvm_amap.c: 1.108 sys/uvm/uvm_anon.c: 1.64 sys/uvm/uvm_aobj.c: 1.126 sys/uvm/uvm_bio.c: 1.91 sys/uvm/uvm_device.c: 1.66 sys/uvm/uvm_fault.c: 1.201 sys/uvm/uvm_km.c: 1.144 sys/uvm/uvm_loan.c: 1.85 sys/uvm/uvm_map.c: 1.353 sys/uvm/uvm_page.c: 1.194 sys/uvm/uvm_pager.c: 1.111 sys/uvm/uvm_pdaemon.c: 1.109 sys/uvm/uvm_swap.c: 1.175 sys/uvm/uvm_vnode.c: 1.103 usr.bin/vmstat/vmstat.c: 1.219 Reorder to test for null before null deref in debug code -- Reorder to test for null before null deref in debug code -- KNF -- No need for '\n' in UVMHIST_LOG -- normalise a BIOHIST log message -- Update the kernhist(9) kernel history code to address issues identified in PR kern/52639, as well as some general cleaning-up... (As proposed on tech-kern@ with additional changes and enhancements.) Details of changes: * All history arguments are now stored as uintmax_t values[1], both in the kernel and in the structures used for exporting the history data to userland via sysctl(9). This avoids problems on some architectures where passing a 64-bit (or larger) value to printf(3) can cause it to process the value as multiple arguments. (This can be particularly problematic when printf()'s format string is not a literal, since in that case the compiler cannot know how large each argument should be.) * Update the data structures used for exporting kernel history data to include a version number as well as the length of history arguments. * All [2] existing users of kernhist(9) have had their format strings updated. Each format specifier now includes an explicit length modifier 'j' to refer to numeric values of the size of uintmax_t. * All [2] existing users of kernhist(9) have had their format strings updated to replace uses of "%p" with "%#jx", and the pointer arguments are now cast to (uintptr_t) before being subsequently cast to (uintmax_t). This is needed to avoid compiler warnings about casting "pointer to integer of a different size." * All [2] existing users of kernhist(9) have had instances of "%s" or "%c" format strings replaced with numeric formats; several instances of mis-match between format string and argument list have been fixed. * vmstat(1) has been modified to handle the new size of arguments in the history data as exported by sysctl(9). * vmstat(1) now provides a warning message if the history requested with the -u option does not exist (previously, this condition was silently ignored, with only a single blank line being printed). * vmstat(1) now checks the version and argument length included in the data exported via sysctl(9) and exits if they do not match the values with which vmstat was built. * The kernhist(9) man-page has been updated to note the additional requirements imposed on the format strings, along with several other minor changes and enhancements. [1] It would have been possible to use an explicit length (for example, uint64_t) for the history arguments. But that would require another "rototill" of all the users in the future when we add support for an architecture that supports a larger size. Also, the printf(3) format specifiers for explicitly-sized values, such as "%"PRIu64, are much more verbose (and less aesthetically appealing, IMHO) than simply using "%ju". [2] I've tried very hard to find "all [the] existing users of kernhist(9)" but it is possible that I've missed some of them. I would be glad to update any stragglers that anyone identifies. -- For some reason this single kernel seems to have outgrown its declared size as a result of the kernhist(9) changes. Bump the size. XXX The amount of increase may be excessive - anyone with more detailed XXX knowledge please feel free to further adjust the value appropriately. -- Misssed one cast of pointer --> uintptr_t in previous kernhist(9) commit -- And yet another one. :( -- Use correct mark-up for NetBSD version. -- More improvements in grammar and readability. -- Remove a stray '"' (obvious typo) and add a couple of casts that are probably needed. -- And replace an instance of "%p" conversion with "%#jx" -- Whitespace fix. Give Bl tag table a width. Fix Xr.
1.364.2.4	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.364.2.3	20-Oct-2018	pgoyette	Sync with head
1.364.2.2	06-Sep-2018	pgoyette	Sync with HEAD Resolve a couple of conflicts (result of the uimin/uimax changes)
1.364.2.1	07-Apr-2018	pgoyette	Sync with HEAD. 77 conflicts resolved - all of them $NetBSD$
1.365.2.4	21-Apr-2020	martin	Sync with HEAD
1.365.2.3	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.365.2.2	08-Apr-2020	martin	Merge changes from current as of 20200406
1.365.2.1	10-Jun-2019	christos	Sync with HEAD
1.373.2.2	27-Feb-2020	martin	Pull up following revision(s) (requested by skrll in ticket #742): sys/arch/arm/arm32/pmap.c: revision 1.388 sys/arch/arm/arm32/armv7_generic_space.c: revision 1.11 sys/arch/arm/arm/cpufunc.c: revision 1.176 sys/arch/arm/conf/Makefile.arm: revision 1.54 (via patch) sys/arch/arm/include/arm32/pmap.h: revision 1.161 Fix the armv[67] memory attributes for uncached memory. Previously it was mapped as strongly-ordered which meant that unaligned accesses would fault. armv7_generic_bs_map now maps pages with PMAP_DEV which is treated as SO bus_dma continues to use PMAP_NOCACHE as appropriate, but this now get mapped to the correct memory attribute bits for armv[67] DEVMAP_ENTRY usees a new flag PTE_DEV. The workaround for the unaligned access faults is now removed. XXX Other armv[67] boards bus_space implementations should be checked. XXX There is scope to reduce the difference to aarch64
1.373.2.1	10-Feb-2020	martin	Pull up following revision(s) (requested by skrll in ticket #691): sys/arch/arm/arm32/pmap.c: revision 1.383 sys/arch/arm/arm32/pmap.c: revision 1.385 sys/arch/arm/arm32/pmap.c: revision 1.386 sys/arch/arm/arm32/pmap.c: revision 1.387 sys/arch/arm/arm32/pmap.c: revision 1.374 sys/arch/arm/arm32/pmap.c: revision 1.375 sys/arch/arm/arm32/pmap.c: revision 1.376 sys/arch/arm/arm32/pmap.c: revision 1.377 sys/arch/arm/arm32/pmap.c: revision 1.378 sys/arch/arm/arm32/pmap.c: revision 1.379 Convert a __CTASSERT into a KASSERT as L1_S_CACHE_MASK may not be a compile time constant if ARM_NMMUS > 1 Improve a comment Update PMAP_STEAL_MEMORY code to uvm_hotplug Typo in comment Fix a bug introduced in 1.271 where pmap_grow_map would no longer map the allocated page for the uvm.page_init_done == false case when PMAP_STEAL_MEMORY is not defined. Trailing whitespace Fix comment Always pmap_kenter_pa the page in pmap_grow_map regardless of how we got it. Always call pmap_grow_map with a page aligned new VA. KASSERT that this happenes. More KNF
1.375.2.4	29-Feb-2020	ad	Sync with head.
1.375.2.3	29-Feb-2020	ad	Sync with head.
1.375.2.2	25-Jan-2020	ad	Sync with head.
1.375.2.1	17-Jan-2020	ad	Sync with head.
1.402.2.2	25-Apr-2020	bouyer	Sync with bouyer-xenpvh-base2 (HEAD)
1.402.2.1	20-Apr-2020	bouyer	Sync with HEAD
1.422.2.1	03-Apr-2021	thorpej	Sync with HEAD.
1.426.2.1	03-Apr-2021	thorpej	Sync with HEAD.
1.437.4.4	18-Apr-2024	martin	Pull up following revision(s) (requested by skrll in ticket #666): sys/arch/arm/arm32/pmap.c: revision 1.443 port-arm/58135: reproducible pmap KASSERT failure for armv7 with NFS root Don't unconditionally set XN in pmap_clearbit - only set it if a mapping exists VM_PROT_EXEC is being cleared. I've simplified the #ifdefs in the patch from the PR.
1.437.4.3	14-Dec-2023	martin	Pull up following revision(s) (requested by rin in ticket #495): sys/arch/arm/arm/smccc.c: revision 1.4 sys/arch/arm/arm32/pmap.c: revision 1.441 smccc: Adjust SMCCC_ARCH_ATTRIBUTE for clang/arm Conditionally use (1) __attribute__ ((target("arch=armv7ve"))) (2) __attribute__ ((target("armv7ve"))) for gcc and clang, respectively. While gcc does not accept (2), clang accepts (1) but silently ignores it :( arm: pmap: Fix clang build without DIAGNOSTIC
1.437.4.2	14-Oct-2023	martin	Pull up following revision(s) (requested by skrll in ticket #411): sys/arch/evbarm/nslu2/nslu2_machdep.c: revision 1.41 sys/arch/evbarm/gumstix/gumstix_machdep.c: revision 1.75 sys/arch/evbarm/iq80321/iq80321_machdep.c: revision 1.66 sys/arch/iyonix/iyonix/iyonix_machdep.c: revision 1.34 sys/arch/zaurus/zaurus/machdep.c: revision 1.52 sys/arch/evbarm/g42xxeb/g42xxeb_machdep.c: revision 1.41 sys/arch/hpcarm/hpcarm/pxa2x0_hpc_machdep.c: revision 1.33 sys/arch/evbarm/iq80310/iq80310_machdep.c: revision 1.96 sys/arch/evbarm/adi_brh/brh_machdep.c: revision 1.53 sys/arch/arm/include/arm32/pmap.h: revision 1.177 sys/arch/evbarm/viper/viper_machdep.c: revision 1.34 sys/arch/evbarm/iyonix/iyonix_machdep.c: revision 1.5 sys/arch/evbarm/npwr_fc/npwr_fc_machdep.c: revision 1.30 sys/arch/evbarm/hdl_g/hdlg_machdep.c: revision 1.35 sys/arch/arm/arm32/pmap.c: revision 1.440 sys/arch/evbarm/lubbock/lubbock_machdep.c: revision 1.45 sys/arch/evbarm/ixdp425/ixdp425_machdep.c: revision 1.47 Fix non-DIAGNOSTIC builds
1.437.4.1	19-Dec-2022	martin	Pull up following revision(s) (requested by skrll in ticket #4): sys/arch/arm/arm32/pmap.c: revision 1.438 sys/arch/arm/arm/efi_machdep.c: revision 1.3 Appease KDASSERT / LOCKDEBUG. Tested by mlelstv.

OpenGrok