Sync with compat and remove dup implementations of ldexp and modf, switching
mips and m68k to the compat copies.

branches: 1.34.2;
cache stuff needs _KERNTYPES

Properly support the MIPS TLS ABI.

Make the MIPS N32/N64 ABIs properly support 128-bit long doubles. With this
change, we should be fully conformant with the N32 and N64 ABIs.
Add {fpclassify,infinity,isnan,ininf,signbit}l_ieee754.c back to lib/libc/gen.
Note that infinityl_ieee754.c will work with either 64-bit, 80-bit, or
128-bit long doubles.

Pullup fixes from matt-nb5-mips64.
This makes setjmp/longjmp work again.

Merge from matt-nb5-mips64.

Rename ${SRCS.{alpha,arm,...}.gen} to ${LSRCS.{alpha,arm,...}.gen} because
they ended up in ${LSRCS}, not ${SRCS}. ${SRCS.*} namespace will be used
for more useful things. No functional changes intended.

branches: 1.28.34;
build frexp_ieee754 in the "compat" subtree
(There are assembler versions of ldexp and modf which need to be checked.
I'm in doubt they are metter than those in libm.)

Follow compat libc changes on i386:
> _setjmp and _longjmp are not compat functions. Noticed by Havard Eidnes

Switch mips to the new libc compat code layout.

Push back the descriptions of NaN formats, and descriptions of the
distinction between signalling NaNs and quiet NaNs back into the
machine-dependent headers; treat the implementation of __nanf in the
same spirit.

IEEE 754 leaves the distinction between signalling NaNs and quiet NANs
to the implementation, and unlike our headers used to suggest they're
not identical in the interpretation of the fraction's MSb; in due
course, make those of hppa, mips, sh3, and sh5 reflect reality.

branches: 1.24.6;
Actually compile __longjmp14.c.

* Turn isinf(3) and isnan(3) into C99-style macros.
* Make it possible for ports to override these (i.e., VAX).
* Remove isnanl(), which was internal to libc only.

Add C99 fpclassify(), isfinite(), isnormal(), and signbit() macros.

Add __infinityf and __infinityl, float respectively long double analogs
of __infinity.

Correct another typo.

Pasto in previous.

* Since there is no single portable IEEE 754 format for a long double,
keep a common implementation of isinfl() and isnanl() to be used by
platforms where `long double' == `double'; move others into
machine-dependent code.
* In due course, consider __VFP_FP__ on arm.

Rework how dependency generation is performed:

* DPSRCS contains extra dependencies, but is _NOT_ added to CLEANFILES.
This is a change of behaviour. If a Makefile wants the clean semantics
it must specifically append to CLEANFILES.
Resolves PR toolchain/5204.

* To recap: .d (depend) files are generated for all files in SRCS and DPSRCS
that have a suffix of: .c .m .s .S .C .cc .cpp .cxx

* If YHEADER is set, automatically add the .y->.h to DPSRCS & CLEANFILES

* Ensure that ${OBJS} ${POBJS} ${LOBJS} ${SOBJS} *.d depend upon ${DPSRCS}

* Deprecate the (short lived) DEPENDSRCS


Update the various Makefiles to these new semantics; generally either
adding to CLEANFILES (because DPSRCS doesn't do that anymore), or replacing
specific .o dependencies with DPSRCS entries.

Tested with "make -j 8 distribution" and "make distribution".

Build _isinfl() and _isnanl().

Rename ieee754_function.c to function_ieee754.c, following the convention
used in exec_format.c, loadfile_format.c, subsys_machdep.c etc.
Per discussion with Christos.

Merge the nathanw_sa branch.

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.12.6;
remove MIPS SOFTFLOAT.

C99: Define a NAN macro in <math.h> which evaulates to a constant expression of
a single-precision quiet NaN; only to be defined on platforms that do support
this value.

Hpcmips needs completely FPU instruction free user land.
XXX, this modification is kinda ad-hoc.

branches: 1.9.2;
Use C versions of isinf() and isnan(). The assembler code is basically
identical, but was endian-dependent.

* Fix typo in Makefile.inc -- byte_swap_2.4 -> byte_swap_4.S
* Change bytes_swap_4.S to have consistent NLEAF and END for swap routine.

Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

mips libc stubs for sig*14 signal API (128bit sigset_t),
plus backward-compatibility for the old 1.3 libc entrypoints.

Cacheflush support code for mips CPUs.
Used by GCC's lexical-closure trampoline code.

Add sigsetjmp.S for mips, and add it to the makefile.

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

don't forget inifinity.c

need these for new gen/Makefile

Switch mips to the new libc compat code layout.

Add missing <compat/include/setjmp.h>

Copy the sparc/sparc64 __longjmp14() way of dealing with signal masks:
* Restore the signal mask with sigprocmask() instead of _UC_SIGMASK,
* since libpthread may want to interpose on signal handling.

Fixes regress/lib/libpthread/siglongjmp1.

Catch up with changes to __fpregset_t.
In __longjmp14(), copy the FP CSR from the correct array and array slot.

Completes fix for PR port-mips/25942.

branches: 1.2.2;
Restore the PC as well; also use mcontext register defines for accessing
the mcontext.

Use setcontext() instead of sigreturn() to implement longjmp().

Switch mips to the new libc compat code layout.

Use setcontext() instead of sigreturn() to implement longjmp().

Hide the register number constants behind an _R_ prefix, and also
rename FPBASE to _FPBASE, so that we avoid polluting the user's
name space when e.g. <sys/ptrace.h> is included. Previously, the
PC symbol in mips/regnum.h would conflict with the declaration of
the external variable by the same name in termcap.h, as discovered
by the ``okheaders'' regression test.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.6.6;
remove MIPS SOFTFLOAT.

adjust #endif position (SOFTFLOAT)

Hpcmips needs completely FPU instruction free user land.
XXX, this modification is kinda ad-hoc.

branches: 1.3.4;
Regularize with removal of genpubassym.cf. Structure offsets
now obtained from signal.h

Userland support for 64-bit safe sigcontexts.

mips libc stubs for sig*14 signal API (128bit sigset_t),
plus backward-compatibility for the old 1.3 libc entrypoints.

Switch mips to the new libc compat code layout.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.2.12;
Userland support for 64-bit safe sigcontexts.

mips libc stubs for sig*14 signal API (128bit sigset_t),
plus backward-compatibility for the old 1.3 libc entrypoints.

mips _lwp_makecontext: Align sp.

PR kern/59327: user stack pointer is not aligned properly

lib/libc/arch/mips/gen/_lwp.c: Sort includes.

No functional change intended.

Preparation for:

PR kern/59327: user stack pointer is not aligned properly

Create a new header lwp_private.h to contain _lwp_getprivate_fast,
_lwp_gettcb_fast, _lwp_settcb and remove them from mcontext.h, so that:
1. we don't need special hacks to hide them
2. we can include <lwp.h> where needed to get the necessary prototypes
without redefining them locally.

branches: 1.8.56;
Use correct CALLFRAME_SIZ depending on ABI.

Properly support the MIPS TLS ABI.

Allow storing and receiving the LWP private pointer via ucontext_t
on all platforms except VAX and IA64. Add fast access via register for
AMD64, i386 and SH3 ports. Use this fast access in libpthread to replace
the stack based pthread_self(). Implement skeleton support for Alpha,
HPPA, PowerPC, SPARC and SPARC64, but leave it disabled.

Ports that support this feature provide __HAVE____LWP_GETPRIVATE_FAST in
machine/types.h and a corresponding __lwp_getprivate_fast in
machine/mcontext.h.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

branches: 1.5.8; 1.5.14; 1.5.20;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.20;
Add missing __RCSID()

Make sure we reference getcontext() by its internal name.

Merge the nathanw_sa branch.

branches: 1.1.2;
file _lwp.c was initially added on branch nathanw_sa.

mips __resumecontext: Align private stack frame size.

On mips64, UCONTEXT_SIZE is 664=0x298 (n32) or 744=0x2e8 (n64), but
sp is required to have 16-byte alignment in the ABI, so we need to
round the stack frame allocation up to 0x2a0 or 0x2f0.

No functional change intended on mips o32. Unlikely to have a
practical change on mips n32/n64, but possible if signal delivery
wins a race -- until we fix sp alignment in signal handler calls,
coming soon.

Based on a patch by rin@.

PR kern/59327: user stack pointer is not aligned properly

branches: 1.13.8;
Remove '_OFFSETOF' prefix for genassm(1) generate CPP identifers for
consistency with other arches.

NFCI and libc.so is the same before and after.

Trailing whitespace

Fix for [on]{32,64}

/usr/tests/lib/libc/sys/t_getcontext
/usr/tests/lib/libc/sys/t_swapcontext
/usr/tests/lib/libc/sys/t_ucontext
/usr/tests/lib/libpthread/t_swapcontext

All pass now.

Changes amount to

- saving GP from caller in context for n{32,64}
- performing (the equivalient of) PIC_PROLOGUE for swapcontext and
__resumecontext
- Call setcontext via the PLT in __resumecontext

Add a NOP_L

PIC_TAILCALL on n32/n64 would mess up GP, so just use SYSTRAP to call
setcontext.

Fixes tests/lib/libc/sys/t_getcontext.c:setcontext_link

Trailing whitespace

branches: 1.7.14;
setcontext() used to be incompatible with -lpthread since it affected
the TLS pointer, therefore wrecking the pthread environement.

Some ports had _UC_TLSBASE flag or equivalent (_UC_UNIQUE on alpha)
that controlled whether setcontext() would change the TLS pointer.
This change let libpthread override setcontext() with its own version
that unsets _UC_TLSBASE, enabling safe usage of setcontext() with
-lpthread.

We also have the following required changes here:
- rename alpha's _UC_UNIQUE into _UC_TLSBASE
- add _UC_TLSBASE definition in header file for all ports
(powerpc, sh3, sparc and sparc64 lack the implementation for now)
- introduce a libc stub that can be overriden for setcontext()
- modify MD libcs swapcontext() implementations so that they use the
setcontext() libc stub instead of doing a plain system call.

While we are there:
- document various MD _UC_* flags in header file
- add libc and libpthread tests for swapcontext() behavior
(hopefully helpful to spot MD problems introduced with this change)

Future work:
- Deciding whether kernel support or _UC_TLSBASE should be added for
powerpc, sh3, sparc and sparc64 is left to portmasters
sparc64

Approved by core@

branches: 1.6.6;
Merge from matt-nb5-mips64.

branches: 1.5.8; 1.5.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.32;
optimize this: we can use %ra for relocation directly
(saves 1 insn)

get the offset to gp right

Merge the nathanw_sa branch.

branches: 1.1.2;
file _resumecontext.S was initially added on branch nathanw_sa.

mips: Fix _longjmp(..., 0) so _setjmp returns 1 as intended.

_setjmp (like setjmp and sigsetjmp) is only permitted to return 0 the
first time around.

PR port-mips/59285: _longjmp(..., 0) makes setjmp return 0, not 1

branches: 1.26.8;
Remove '_OFFSETOF' prefix for genassm(1) generate CPP identifers for
consistency with other arches.

NFCI and libc.so is the same before and after.

Trailing whitespace

Slight optimization.

branches: 1.23.4;
Reorder insns to fix more load delay hazard botches on R3000.
Fixes floating point exceptions in _longjmp() during /etc/rc and
now NWS-3470D boots up to multi user with -current userland binaries.

Also fill BDslots in error path properly.

Pullup fixes from matt-nb5-mips64.
This makes setjmp/longjmp work again.

Merge from matt-nb5-mips64.

branches: 1.20.34;
Follow compat libc changes on i386:
> _setjmp and _longjmp are not compat functions. Noticed by Havard Eidnes

Switch mips to the new libc compat code layout.

Hide the register number constants behind an _R_ prefix, and also
rename FPBASE to _FPBASE, so that we avoid polluting the user's
name space when e.g. <sys/ptrace.h> is included. Previously, the
PC symbol in mips/regnum.h would conflict with the declaration of
the external variable by the same name in termcap.h, as discovered
by the ``okheaders'' regression test.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.15.6;
remove MIPS SOFTFLOAT.

remove extra lw (SOFTFLOAT)

adjust #endif position (SOFTFLOAT)

Hpcmips needs completely FPU instruction free user land.
XXX, this modification is kinda ad-hoc.

branches: 1.11.4;
_setjmp.S needs OFFSETOF_SC_{FP,}REGS, so include <machine/signal.h>
even tho _setjmp()/_longjmp() don't manipulate signals. (more thought?)

Userland support for 64-bit safe sigcontexts.

Remove single ' from comment to keep cross-assemblers happy; fixes PR 5451.

Back out previous revision, which is reported to break signal-handing
in the GNU awk in our tree.
Leave the change in but commented out, it may be a ld.so problem.

Add .set noreorder/.set reorder around .cpload

* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Change mips assembly code for {,_,sig}setjmp.S to get register names
from <machine/regnum.h>, not <machine/reg.h> (which also defines structs.)

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Use reach-over sources from common/lib/libc

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.3.2;
make these more amenable to being copied into libkern. (Gee, why wasn't
this done before? it's not like it was hard...)

pull up <machine/endian.h> and use BYTE_ORDER to decide endianness,
rather than MIPSEL and MIPSEB definitions.

Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

Use reach-over sources from common/lib/libc

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.4.2;
make these more amenable to being copied into libkern. (Gee, why wasn't
this done before? it's not like it was hard...)

pull up <machine/endian.h> and use BYTE_ORDER to decide endianness,
rather than MIPSEL and MIPSEB definitions.

* Fix typo in Makefile.inc -- byte_swap_2.4 -> byte_swap_4.S
* Change bytes_swap_4.S to have consistent NLEAF and END for swap routine.

Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

nbytes is now size_t

fix lint, should nbytes be changed to size_t to match with the struct passed?

branches: 1.4.4; 1.4.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.3.46;
s/vm_offset_t/vaddr_t/

change second parm of sysarch() from char * to void *

Cacheflush support code for mips CPUs.
Used by GCC's lexical-closure trampoline code.

On ports without __HAVE_LONG_DOUBLE, make fabsl alias fabs.

For some reason, fabs lives in libc, not in libm, and our tests now
detect when fabs or fabsl is missing from libm. For those ports that
sometimes have long double and sometimes don't, make it conditional.

Still missing: fabs _and_ fabsl on ia64. Need help from an itanium
wizard! Other portmasters: Please take a look and see if I missed
any ports that might have long double where this alias will not work.

branches: 1.8.44; 1.8.46;
Merge from matt-nb5-mips64.

branches: 1.7.46;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.5.16;
* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Call libc's fpgetround.

Explicitly set noat around instructions using $at.

branches: 1.7.10;
fix old style definitions; XXX: gcc should have picked them up but it did not.

branches: 1.6.6;
Pullup fixes from matt-nb5-mips64.
This makes setjmp/longjmp work again.

branches: 1.5.34;
__asm__ -> __asm

Add missing __RCSID()

Prototype __flt_rounds().

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

Explicitly set noat around instructions using $at.

Revert unneeded casts.

Cast to unsigned int before doing shift operations, to appease lint.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

Explicitly set noat around instructions using $at.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

Explicitly set noat around instructions using $at.

Revert unneeded casts.

Cast to unsigned int before doing shift operations, to appease lint.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

fix asm operands: "ctc1" uses the register as an input, not an output.

branches: 1.9.2; 1.9.4;
Explicitly set noat around instructions using $at.

Revert unneeded casts.

Cast to unsigned int before doing shift operations, to appease lint.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

fix asm operands: "ctc1" uses the register as an input, not an output.

branches: 1.7.2; 1.7.4;
Explicitly set noat around instructions using $at.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

fix asm operands: "ctc1" uses the register as an input, not an output.

branches: 1.9.2; 1.9.4;
Explicitly set noat around instructions using $at.

Revert unneeded casts.

Cast to unsigned int before doing shift operations, to appease lint.

Convert to C89 definitions.
Cleanup up type usage.

branches: 1.5.44;
__asm__ -> __asm

Add missing __RCSID()

Add internal names for fp{get,set}{mask,round,sticky}(), and
externally-visible weak aliases for the internal names.

branches: 1.2.16;
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Added __flt_rounds, fp{get,set}{mask,round,sticky}.

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.6.10;
Trivial KNF.

Make the C version of frexp() consitent across platforms, and handle the
Inf/NaN case.

Use ieee.h.

Fix RCSID's

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

libc support for mips. libkern hooks are disabled

remove unused files (MIPS SOFTFLOAT).

branches: 1.1.1;
Initial revision

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.5.12;
delint

const poisoning.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

*** empty log message ***

+Infinity for MIPS

Use C versions of isinf() and isnan(). The assembler code is basically
identical, but was endian-dependent.

Need internal names for isinf() and isnan().

* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.2.6;
Delint.

Use C versions of isinf() and isnan(). The assembler code is basically
identical, but was endian-dependent.

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.2.6;
Delint.

Use C versions of isinf() and isnan(). The assembler code is basically
identical, but was endian-dependent.

Use canonical three operand form of add.d.

Merge from matt-nb5-mips64.

branches: 1.8.46;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

t4 -> ta0

branches: 1.5.16;
* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

mips longjmp: Use _UC_SIGMASK to restore signal mask.

This way, restoring the signal mask and restoring the stack pointer
happen atomically with respect to signal handler calls, whereas using
sigprocmask would restore the signal mask _before_ the stack pointer,
breaking sigaltstack.

The motivation for using sigprocmask first and then setcontext later,
rather than _UC_SIGMASK in setcontext, was to get SA-based libpthead
sigprocmask interposition. But that's long gone and unlikely to come
back.

PR lib/57946: longjmp fails to restore stack first before restoring
signal mask on most architectures

branches: 1.6.28;
Use namespace protected types.

make constant explicitly unsigned

branches: 1.4.6;
Pullup fixes from matt-nb5-mips64.
This makes setjmp/longjmp work again.

fix -Wsign-compare issue

branches: 1.2.8; 1.2.10; 1.2.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.1.20;
Switch mips to the new libc compat code layout.

mips/makecontext.c: Use STACK_ALIGNBYTES for clarity.

No functional change intended.

Tidying to make stack alignment more obvious, prompted by:

PR kern/59327: user stack pointer is not aligned properly

Part of a patch by rin@.

mips/makecontext.c: Sort includes.

No functional change intended.

Preparation for:

PR kern/59327: user stack pointer is not aligned properly

branches: 1.8.26;
Fix comments

Explicitly add back __resumecontext wrapper

Mark _resumecontext as dead. ANSIfy all declarations for it.

Merge from matt-nb5-mips64.

branches: 1.4.8; 1.4.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.3.32;
Cast with uintptr_t instead of u_long.

Merge the nathanw_sa branch.

branches: 1.1.2;
file makecontext.c was initially added on branch nathanw_sa.

Merge from matt-nb5-mips64.

branches: 1.10.46;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.8.14;
Do a .cpload if compiling with PIC.
Examining the .so output from binutils 2.8 and 2.8.1 shows the
li.d $f2, 4503599627370496e0
is (incorrectly?) emitted as GP-relative loads when PIC is in effect.

* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Copy the input argument to $f0 so we handle negative inputs properly
in the tail.

Doesn't yet handle +/- inf as cleanly as other implementations.

The old mips modf didn't work (pr 1048). This replaces it with
a version that does work on little-endian MIPS CPUs. Not tested on
big-endian MIPS; the "li.d" may be byteorder-dependent.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Make NAN a compile time constant (with help from the compiler). Suggested
by krister. Fixes PR 40695.
Make references to the old construct (which we can't remove for binary
compatibility reasons) emit a linker warning.

branches: 1.3.2; 1.3.32; 1.3.36;
Push back the descriptions of NaN formats, and descriptions of the
distinction between signalling NaNs and quiet NaNs back into the
machine-dependent headers; treat the implementation of __nanf in the
same spirit.

IEEE 754 leaves the distinction between signalling NaNs and quiet NANs
to the implementation, and unlike our headers used to suggest they're
not identical in the interpretation of the fraction's MSb; in due
course, make those of hppa, mips, sh3, and sh5 reflect reality.

Clean up some rampant code duplication wrt ieee number handling:
- Add alignment-safe double and float unions.
- Use the above for the __infinity and __nan constants on all
architectures that use the standard ieee754 representation of
those constants.
- Add a single copy of various ieee754 math functions (frexp, isinf,
isnan, ldexp and modf) that had numerous duplicates among the
arch-specific directories.
- Use the above functions on all architectures where the generic C
versions where used. Architectures that had local assembly
routines are untouched (for those functions only).

branches: 1.1.6;
C99: Define a NAN macro in <math.h> which evaulates to a constant expression of
a single-precision quiet NaN; only to be defined on platforms that do support
this value.

Mark _resumecontext as dead. ANSIfy all declarations for it.

branches: 1.4.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.3.12;
Sync all resumecontext.c versions with the changes originally made
by Matthias Drochner to i386 and alpha:

-check for a NULL uc_link before calling setcontext(2); do a real exit(3)
instead with exit handlers and all that
-If setcontext(2) returns (which it shouldn't), do an _exit(-1) rather
than relying on a sensible return value to use as exit code. Makes
it less likely that an ucontext corruption goes unnoticed.

branches: 1.2.16;
Merge the nathanw_sa branch.

branches: 1.1.2;
file resumecontext.c was initially added on branch nathanw_sa.

Remove '_OFFSETOF' prefix for genassm(1) generate CPP identifers for
consistency with other arches.

NFCI and libc.so is the same before and after.

Pullup fixes from matt-nb5-mips64.
This makes setjmp/longjmp work again.

Merge from matt-nb5-mips64.

branches: 1.17.34;
Switch mips to the new libc compat code layout.

Hide the register number constants behind an _R_ prefix, and also
rename FPBASE to _FPBASE, so that we avoid polluting the user's
name space when e.g. <sys/ptrace.h> is included. Previously, the
PC symbol in mips/regnum.h would conflict with the declaration of
the external variable by the same name in termcap.h, as discovered
by the ``okheaders'' regression test.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.13.6;
remove MIPS SOFTFLOAT.

adjust #endif position (SOFTFLOAT)

Hpcmips needs completely FPU instruction free user land.
XXX, this modification is kinda ad-hoc.

branches: 1.10.4;
Regularize with removal of genpubassym.cf. Structure offsets
now obtained from signal.h

Userland support for 64-bit safe sigcontexts.

mips libc stubs for sig*14 signal API (128bit sigset_t),
plus backward-compatibility for the old 1.3 libc entrypoints.

Use __sigaltstack14(), not sigaltstack().

* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Change mips assembly code for {,_,sig}setjmp.S to get register names
from <machine/regnum.h>, not <machine/reg.h> (which also defines structs.)

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

Clean up deleted files.

More underscores...

Prepended underscores

libc support for mips. libkern hooks are disabled

remove unused files (MIPS SOFTFLOAT).

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Remove pointless #ifdef.

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Fix bogons.

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Fix bogons.

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Fix bogons.

Do the minimum work to make this compile with -Werror.

fix name space pollution by adding prefix '_mips_'.
closes PR 8770.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

Delint.

branches: 1.1.1;
Initial revision

remove unused files (MIPS SOFTFLOAT).

branches: 1.1.1;
Initial revision

merge christos-time_t

branches: 1.3.2; 1.3.16;
Fix doubled 'the's.

remove unused files (MIPS SOFTFLOAT).

branches: 1.1.1;
Initial revision

Merge from matt-nb5-mips64.

branches: 1.8.34;
Switch mips to the new libc compat code layout.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

ABICALLS -> __ABICALLS__

branches: 1.5.12;
Regularize with removal of genpubassym.cf. Structure offsets
now obtained from signal.h

* Add Mips ABI changes for ELf shared library support.
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.

Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
from "Id:" to "NetBSD:".

Change mips assembly code for {,_,sig}setjmp.S to get register names
from <machine/regnum.h>, not <machine/reg.h> (which also defines structs.)

Add sigsetjmp.S for mips, and add it to the makefile.

Remove '_OFFSETOF' prefix for genassm(1) generate CPP identifers for
consistency with other arches.

NFCI and libc.so is the same before and after.

Fix for [on]{32,64}

/usr/tests/lib/libc/sys/t_getcontext
/usr/tests/lib/libc/sys/t_swapcontext
/usr/tests/lib/libc/sys/t_ucontext
/usr/tests/lib/libpthread/t_swapcontext

All pass now.

Changes amount to

- saving GP from caller in context for n{32,64}
- performing (the equivalient of) PIC_PROLOGUE for swapcontext and
__resumecontext
- Call setcontext via the PLT in __resumecontext

Trailing whitespace.

branches: 1.5.14;
setcontext() used to be incompatible with -lpthread since it affected
the TLS pointer, therefore wrecking the pthread environement.

Some ports had _UC_TLSBASE flag or equivalent (_UC_UNIQUE on alpha)
that controlled whether setcontext() would change the TLS pointer.
This change let libpthread override setcontext() with its own version
that unsets _UC_TLSBASE, enabling safe usage of setcontext() with
-lpthread.

We also have the following required changes here:
- rename alpha's _UC_UNIQUE into _UC_TLSBASE
- add _UC_TLSBASE definition in header file for all ports
(powerpc, sh3, sparc and sparc64 lack the implementation for now)
- introduce a libc stub that can be overriden for setcontext()
- modify MD libcs swapcontext() implementations so that they use the
setcontext() libc stub instead of doing a plain system call.

While we are there:
- document various MD _UC_* flags in header file
- add libc and libpthread tests for swapcontext() behavior
(hopefully helpful to spot MD problems introduced with this change)

Future work:
- Deciding whether kernel support or _UC_TLSBASE should be added for
powerpc, sh3, sparc and sparc64 is left to portmasters
sparc64

Approved by core@

branches: 1.4.6;
Merge from matt-nb5-mips64.

branches: 1.3.8; 1.3.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.2.32;
Merge the nathanw_sa branch.

branches: 1.1.2;
file swapcontext.S was initially added on branch nathanw_sa.