Tidy up how __sigaction14_sigtramp.c is added to SRCS.

remove various HAVE_GCC=45 fragments.

do not need lint overrides anymore

We do not care about portability when instantiating softfloats for our
concrete environment - add appropriate lintflags.

Fix so that float128_to_uint64_round_to_zero is defined for sparc64.
(isn't tha a namespace violation?)

apply some -Wno-error/-Wno-enum-compare/-Wno-array-bounds.

this should be looked at more closely.

Remove the sparc-v7 compatibility wrapper functions (.div, .mul, ... .urem),
the compiler never generates (or did generate in the past) 64bit code to call
them, and they are still present in our 32bit compat library.

While we use hardware (for float and double), but cooperate with softfloat
(to do long double), we need to not only handle the hardware exception
mask and cummulated flags, but also update/query the softfloat variables.
Since this may go away sometime (to make us psABI compliant), ifdef it
properly.

Set ${SRCS.softfloat}.

branches: 1.12.4;
define and use LIBCDIR in terms of .CURDIR. fixes compat/lib/libc build.

G/c KMSRCS and KMINCLUDES now that we have common/lib/libc

Bring sparc64 to the new compat layout.

Don't use 'm4' here. Use $TOOL_M4 instead.

Add __sigtramp2

remove setjmp/longjmp from libkern, they're not used.

Enable the libc-provided signal trampoline. The old __sigaction14
entry point has been replaced by the new __sigaction14() which is a
wrapper around the __sigaction_sigtramp system call (which is itself
not directly exposed outside of libc).

No shared library version change is necessary; no new symbols
are exposed outside of libc.

Adapt for recent lib/libc/softfloat changes.

branches: 1.4.2;
Move names for softfloat lib that _Qp functions call into _softfloat_
namespace

Enable the _Qp* routines to be built always and allow full softfloat to be
pulled in if MKSOFTFLOAT gets enabled.

Remove -Wa,-Av9a hack from <sys.mk> and relegate it to the places that need
it (kernel and libc).

The current version of the gas assembler in the tree (2.11.2) already
defaults to generating object files for "-Av9 -64", supporting V9
instructions in ELF64 object format. "-Av9a" is only needed for specific
parts of the NetBSD base sources, and not for all third-party code.

branches: 1.1.12;
Add sparc64-specific libc.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.14.4; 1.14.10;
ABI police: avoid cluttering global registers %g2 and %g7. Not quite there
yet for %g2, but the target is not very well defined. %g7 should be clean
now.

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

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

Revert previous: .S files are currently compiled with -traditional-cpp,
so __STDC__ is _not_ always defined.

__STDC__ is always defined on NetBSD.

Go back from using %g2 (an application register) to %g7 (a system register)
for syscall returns.

Add support for large -fPIC relocations.

Add a .register directive to appease newer gas.

* Add WSYSCALL() to SYS.h, which creates a system call with an
internal name and a weak alias for the name.
* Add a WEAKASM syscall list, and process it.
* Make sysarch() have an internal name (_sysarch()).

From rev. 1.11 of the 32-bit version:

Add a macro "CALL()" to call a function correctly in the pic and non pic
case. Write ERROR() in terms of call.

adjust the PIC case for ERROR() to handle __cerror being at a larger offset
than will fit in the immediate field of the LDX instruction.

branches: 1.4.2;
Fix PIC bugs.

Add a _SYSCALL(entry,syscall) macro like the one found on other
architectures.

branches: 1.2.10;
* On architectures using ELF, and conditionally on those in transition, rename
_ASM_LABEL(cerror) and _ASM_LABEL(curbrk) to _C_LABEL(__cerror) and
_C_LABEL(__curbrk) (or their respective architecture-specific equivalents) to
avoid possible name clashes with identifiers used in user applications.
* Do the same for minbrk on all architectures to avoid a GCC-specific (and
on ELF architectures effectively useless) symbol reference renaming in MI code.

Add sparc64-specific libc.

Add strtof(3) and strtold(3); welcome libc 12.137.

Add machine-dependent definitions for gdtoa.

Add machine-dependent definitions for gdtoa.

Unify signbitl implementations.
(Unclear why they were different, it was only the comments that
differed)

Add a comment describing why the generic version works for both 80-bit
and 128-bit double.

branches: 1.19.46;
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.

build frexp_ieee754 and ldexp_ieee754 in the "compat" subtree
(There is an assembler version of modf which needs to be checked.)

rename the files in gen like we did for i386.

Bring sparc64 to the new compat 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: 1.14.6;
* 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.

Avoid using the depreceated __sigreturn14 for longjmp by adapting
Christian Limpachs arch/m68k implementation using setcontext().

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

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

* 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: 1.4.6;
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.

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.

Fix libc signal breakage (I hope).

Add sparc64-specific libc.

rename the files in gen like we did for i386.

Add missing <compat/..> include

Simplify slightly and fix a comment.

Forgot to mention in last commit message: remove some accidently commited
debug code.

Unfortunately we can not restore the signal mask with setcontext() - this
breaks libpthread, which tries to intercept the weak sigprocmask symbol.
So use an explicit sigprocmask call instead.

Avoid using the depreceated __sigreturn14 for longjmp by adapting
Christian Limpachs arch/m68k implementation using setcontext().

rename the files in gen like we did for i386.

Call PIC_PROLOGUE only once.

Store the correct boolean value in sc_onstack.

Too sad - inlining syscall stubs on sparc64 is so much easier than doing
PIC calls to the libc stubs, but unfortunately we need to do the latter-
otherwise we break weak symbol semantics and make libpthread fail.
Thanks to Christian Limpach for the analyzis.

Avoid using the depreceated __sigreturn14 for longjmp by adapting
Christian Limpachs arch/m68k implementation using setcontext().

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

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

Fix layout of fields in jmp_buf. [N.B. we could make JMP_BUFLEN 9 if we want.]

branches: 1.2.10;
More fixes to signal handling code...

Fix libc signal breakage (I hope).

rename the files in gen like we did for i386.

Store the "signal mask saved" flag at the end of the jmp_buf - we now use
more of it than before (and I bet the old place just is a non translated
32bit sparc value).

Fix two inverted comparisions and a few comments.
This makes sparc64 pass the sigsetjmp regression tests.

Use large (-fPIC) relocations.

branches: 1.4.2;
Fix layout of fields in jmp_buf. [N.B. we could make JMP_BUFLEN 9 if we want.]

Update some more assembly code for PIC issues.

Fix PIC bugs.

branches: 1.1.10;
Fix libc signal breakage (I hope).

Avoid lint warnings

branches: 1.6.4;
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; 1.5.20;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.18;
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.

Save/restore some global registers (sigsetjmp compatible). Does not realy
make a difference currently, but is a hard to spot blackhole if something
ever starts using the application registers.

Replace the stack unwinding by a much simpler solution taken from
../../arch/sparc/gen/_setjmp.S rev. 1.6. This ended up to be ~identical
to the OpenBSD change triggered by the same sparc change.
Thanks to Stoned Elipot for pointing out the problem.

branches: 1.5.2; 1.5.4;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Fix libc signal breakage (I hope).

Libc fixes, mostly to work around the new signal stuff.

Bring in definition for stack BIAS.

Add sparc64-specific libc.

Remove the sparc-v7 compatibility wrapper functions (.div, .mul, ... .urem),
the compiler never generates (or did generate in the past) 64bit code to call
them, and they are still present in our 32bit compat library.

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

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

move includes to the top so that this builds in libc context too.

use %g5 instead of %g7 (since we want to use %g7 for the cpu_info pointer
in the kernel). resync libc and libkern versions of this file.

branches: 1.1.12;
Add sparc64-specific libc.

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

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

Floating point argument should be in %f0 (sparcv9 ABI says).

Check how the compiler expects the return value of this function
by testing SOFTFLOATSPARC64_FOR_GCC.

Should fix PR port-sparc64/20140.

branches: 1.2.4;
Fix stack BIAS issues.

branches: 1.1.10;
Add sparc64-specific libc.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.4.56; 1.4.62;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Make __dtoul() actually DTRT. The calling convention was totally wrong, and
the `big' value should be 2^63, not 2^31.
Also, nuke __fixunsdfsi(), since this is never output by the sparc64 compiler.

branches: 1.2.8;
Add _dtoul() which gcc generates a lot. Maybe it should go in another file.

Add sparc64-specific libc.

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

delint

branches: 1.4.44;
__asm__ -> __asm

Add missing __RCSID()

Corrent map order to match sparc/ieeefp.h.

Add sparc64-specific libc.

PR 48384: reorganize struct ieee_ext for sparc64 to only use a high/low
part for the mantissa.
Adjust arch specific code accordingly.

branches: 1.2.4; 1.2.8; 1.2.28;
Remove clause 3 and 4 from TNF licenses

branches: 1.1.30;
C99 7.12.3.1: add IEEE library portions of fpclassify. Don't build these
yet as the VAX implementation is still subject to discussion.

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

delint

branches: 1.4.44;
__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: 1.1.12;
Add sparc64-specific libc.

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

delint

branches: 1.4.44;
__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: 1.1.12;
Add sparc64-specific libc.

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

delint

branches: 1.5.4;
While we use hardware (for float and double), but cooperate with softfloat
(to do long double), we need to not only handle the hardware exception
mask and cummulated flags, but also update/query the softfloat variables.
Since this may go away sometime (to make us psABI compliant), ifdef it
properly.

__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: 1.1.12;
Add sparc64-specific libc.

convert to ANSI C definitions.

branches: 1.6.2;
delint

branches: 1.5.4;
While we use hardware (for float and double), but cooperate with softfloat
(to do long double), we need to not only handle the hardware exception
mask and cummulated flags, but also update/query the softfloat variables.
Since this may go away sometime (to make us psABI compliant), ifdef it
properly.

__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: 1.1.12;
Add sparc64-specific libc.

sparc64/fpsetround: avoid shifting into the sign bit

Lint had warned about the constant expression '0x03 << 30' but not about
the structurally equal nonconstant expression '(rnd_dir & 0x03) << 30'.

No binary change.

fix lint

convert to ANSI C definitions.

While we are using softfloat for 128 bit long double values, make fpsetround
update the softfloat rounding mode as well.

branches: 1.5.2;
delint

branches: 1.4.44; 1.4.46;
__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: 1.1.12;
Add sparc64-specific libc.

convert to ANSI C definitions.

branches: 1.6.2;
delint

branches: 1.5.4;
While we use hardware (for float and double), but cooperate with softfloat
(to do long double), we need to not only handle the hardware exception
mask and cummulated flags, but also update/query the softfloat variables.
Since this may go away sometime (to make us psABI compliant), ifdef it
properly.

__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: 1.1.12;
Add sparc64-specific libc.

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: 1.4.10;
Trivial KNF.

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

Just some minor cleanup.

Add sparc64-specific libc.

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: 1.3.2;
check LIBC_SCCS

delint

Add sparc64-specific libc.

Add missing __RCSID()

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

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

branches: 1.1.30;
C99 7.12.3.2: add library portions of isfinite, but don't build these just
yet.

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: 1.5.6;
Delint.

Separate isnan() into a separate module, and make isinf()/isnan() use ieee.h.

delint

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

Add sparc64-specific libc.

PR 48384: reorganize struct ieee_ext for sparc64 to only use a high/low
part for the mantissa.
Adjust arch specific code accordingly.

branches: 1.4.56; 1.4.62;
* 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.

Err, rename some members added in previous to make them reflect their
semantics better.

Retire local union { floating-type; struct ieee_prec; } in favor of those
available from <machine/ieee.h>

* 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.

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: 1.4.6;
Delint.

Separate isnan() into a separate module, and make isinf()/isnan() use ieee.h.

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

Add sparc64-specific libc.

PR 48384: reorganize struct ieee_ext for sparc64 to only use a high/low
part for the mantissa.
Adjust arch specific code accordingly.

branches: 1.4.56; 1.4.62;
* 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.

Err, rename some members added in previous to make them reflect their
semantics better.

Retire local union { floating-type; struct ieee_prec; } in favor of those
available from <machine/ieee.h>

* 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.

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).

fix for -Wshadow

branches: 1.8.10;
Another microoptimization. Take advantage of the addition already having set
the flags for the <=0 comparison.

Reuse the trick of loading 0.0 (rather than 1.0) and then modifying the
exponent. It's a bit faster on some machines.

Remove more unused mulexp variables.

ldexp(denormal, exp>1023) would generate the wrong result in all non-overflow
cases. Totally rework this code to fix this bug *and* make it faster.

Don't ever return -0.0.

Minor style change.

Just some minor cleanup.

Add sparc64-specific libc.

sparc, sparc64 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: 1.7.44;
delint

branches: 1.6.4;
Fix previous: the way we calculate, we need to test the size of sigjmp_buf

Add another compile time assert: catch overflows of jmp_buf

Rearange jmp_buf content again: storing the sigsetjmp() save_mask midway
inside the environment is not a good idea, we may overwrite it when
fetching other details of the sigjmp environment.
Instead move it to the very end again.
Do not bother to save volatile (caller saved) global registers.
Fixes PR port-sparc64/44918.

Fix offsets used in the assembly code to save global registers into a
jmp_buf, add a compile time assert to catch this kind of divergence
in the C code.
Move the fixed "%npc = %pc + 4" computation to longjmp, this frees
the sigcontext sc_npc value. Use this space to save the savemask
for siglongjmp (we have run out of space behind the globals).
Fixes the tests/lib/libc/setjmp/t_threadjmp tests, and also hopefully
fixes PR port-sparc64/44902.

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

branches: 1.1.18;
rename the files in gen like we did for i386.

delint

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

branches: 1.3.18;
in makecontext(), sp is unsigned * so don't assign signed *'s to it.

Merge the nathanw_sa branch.

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

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.4.56; 1.4.62;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Fix modf() (I hope. Can't be any worse than the old version.)

Fix stack BIAS issues.

branches: 1.1.10;
Add sparc64-specific libc.

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.

Add sparc64-specific libc.

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;
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: 1.3.10;
Pull over rev. 1.5 of the i386 version (../../i386/resumecontext.S),
quoting Matthias Drochner:

> -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.

This also fixes regress/lib/libc/context for sparc64.

branches: 1.2.16;
Merge the nathanw_sa branch.

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

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.

Update some more assembly code for PIC issues.

branches: 1.1.10;
Add sparc64-specific libc.

Consistently pass 0 as first and ignored argument to sigprocmask in the
setjmp implementations.

NFCI.

Trailing whitespace

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.9.4; 1.9.10;
Rearange jmp_buf content again: storing the sigsetjmp() save_mask midway
inside the environment is not a good idea, we may overwrite it when
fetching other details of the sigjmp environment.
Instead move it to the very end again.
Do not bother to save volatile (caller saved) global registers.
Fixes PR port-sparc64/44918.

Fix offsets used in the assembly code to save global registers into a
jmp_buf, add a compile time assert to catch this kind of divergence
in the C code.
Move the fixed "%npc = %pc + 4" computation to longjmp, this frees
the sigcontext sc_npc value. Use this space to save the savemask
for siglongjmp (we have run out of space behind the globals).
Fixes the tests/lib/libc/setjmp/t_threadjmp tests, and also hopefully
fixes PR port-sparc64/44902.

rename the files in gen like we did for i386.

Bring sparc64 to the new compat layout.

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

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

Fix layout of fields in jmp_buf. [N.B. we could make JMP_BUFLEN 9 if we want.]

branches: 1.3.10;
Fix libc signal breakage (I hope).

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

Unify signbitl implementations.
(Unclear why they were different, it was only the comments that
differed)

Add a comment describing why the generic version works for both 80-bit
and 128-bit double.

branches: 1.2.8; 1.2.64;
Remove clause 3 and 4 from TNF licenses

branches: 1.1.30;
Add C99 fpclassify(), isfinite(), isnormal(), and signbit() macros.

Redo previous to fix build break.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.10.4; 1.10.10;
Rearange jmp_buf content again: storing the sigsetjmp() save_mask midway
inside the environment is not a good idea, we may overwrite it when
fetching other details of the sigjmp environment.
Instead move it to the very end again.
Do not bother to save volatile (caller saved) global registers.
Fixes PR port-sparc64/44918.

Fix offsets used in the assembly code to save global registers into a
jmp_buf, add a compile time assert to catch this kind of divergence
in the C code.
Move the fixed "%npc = %pc + 4" computation to longjmp, this frees
the sigcontext sc_npc value. Use this space to save the savemask
for siglongjmp (we have run out of space behind the globals).
Fixes the tests/lib/libc/setjmp/t_threadjmp tests, and also hopefully
fixes PR port-sparc64/44902.

rename the files in gen like we did for i386.

Bring sparc64 to the new compat layout.

Use large (-fPIC) relocations.

branches: 1.5.2;
Fix layout of fields in jmp_buf. [N.B. we could make JMP_BUFLEN 9 if we want.]

Update some more assembly code for PIC issues.

Fix PIC bugs.

branches: 1.2.10;
Fix libc signal breakage (I hope).

Add sparc64-specific libc.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

Fix register usage

branches: 1.6.2;
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: 1.5.4;
ABI police: avoid cluttering global registers %g2 and %g7. Not quite there
yet for %g2, but the target is not very well defined. %g7 should be clean
now.

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

branches: 1.3.30;
Declare global register.

Merge the nathanw_sa branch.

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

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.

Add sparc64-specific libc.

Add sparc64-specific libc.

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.

Add sparc64-specific libc.

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.

Add sparc64-specific libc.

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.

Add sparc64-specific libc.

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.

Add sparc64-specific libc.

branches: 1.1.2;
Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

Fix unsigned 64 bit int to long double conversion for numbers that would not
fit into a signed 64 bit int. Found by latest t_floatunditf test case
failuer, hint from Matt Thomas.

Make all conversions from 128bit (long double) floats to integral types
use rounding towards zero (casts to int/long are not affected by current
rounding mode).
Fixes PR port-sparc64/47535.

branches: 1.9.2;
Avoid lint warnings

Rename __zero to __sf128_zero to avoid clashes with the strange new
inline function in cdefs.h

branches: 1.7.4; 1.7.8; 1.7.24;
Remove clause 3 and 4 from TNF licenses

branches: 1.6.6;
When converting long double values to integer types, explicitly use the
"round to zero" variants of the softfloat conversion functions.
Add a variant to convert long double to unsigned long - the "to long"
variant checked for overflows that do not apply to unsigned results.
This fixes the regress/lib/libc/convfp tests for sparc64.

branches: 1.5.22;
Make _Qp_qtoui return the right value for inputs that do not fit into
an int value, but are valid unsigned ints. This should fix PR 21414.
The same problem exists for _Qp_qtoux, but there is no similar trivial
solution to that (I could think of).
Pointed out by David Laight.

Correct unsigned to float conversions, add license, adjust style.

branches: 1.3.2;
Fix typo in previous commit

Correct mistaken signature for _Qp_dtoq. It gets a double passed in, not
a pointer to double which causes obvious errors on evaluation

branches: 1.1.2;
Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

When converting long double values to integer types, explicitly use the
"round to zero" variants of the softfloat conversion functions.
Add a variant to convert long double to unsigned long - the "to long"
variant checked for overflows that do not apply to unsigned results.
This fixes the regress/lib/libc/convfp tests for sparc64.

branches: 1.1.2; 1.1.26;
Move names for softfloat lib that _Qp functions call into _softfloat_
namespace

All the versions of softfloat-specialize were practically identical, so replace
them with a single one.

branches: 1.2.2;
Move names for softfloat lib that _Qp functions call into _softfloat_
namespace

Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

branches: 1.1.2;
Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

Don't define a new enum. Just #define round_* to their FP_Rx instead.

Add __floatunsidf __floatunsisf __floatunsitf routines.
XXX i think they are correct but not sure.

When converting long double values to integer types, explicitly use the
"round to zero" variants of the softfloat conversion functions.
Add a variant to convert long double to unsigned long - the "to long"
variant checked for overflows that do not apply to unsigned results.
This fixes the regress/lib/libc/convfp tests for sparc64.

branches: 1.5.10;
if SOFTFLOAT_FOR_GCC is not defined, don't declare float_detect_tininess.

Add SIGFPE support to SoftFloat, so fpsetmask/fpgetmask now work. A new
variable, float_exception_mask (#defined to _softfloat_float_exception_mask)
contains the current exception mask.
Also make the argument to float_raise into an fp_except.
Also synchronise file header comments between files.

Consistently declare the types of float_rounding_mode and float_exception_flags
as fp_rnd and fp_except respectively.

branches: 1.2.2;
Move names for softfloat lib that _Qp functions call into _softfloat_
namespace

Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

branches: 1.1.2;
Add softfloat and quad softfloat setup for sparc64 (thanks to Andrey Petrov
for most of this. I just integrated the build).

This adds proper functions for the _Qp* parts of the sparc64 ABI (which
handle quad softfloat). The routines work but need additional testing.
However if the compiler calls these currently bad things (core dumps)
will happen as gcc 2.95.3 generates bad calls for these.

Until that problem is fixed in the toolchain -msoft-quad-float cannot be
enabled as the default option for all builds (but the routines need to be
in libc as libgcc contains some references to these and anything including
it via --whole-archive will need the symbols to at least resolve).

Add the 'abs', 'div' and 'erand48' sources to stdlib/Makefile.inc.
Change the arch files only include the .S files.
This adds imaxdiv() to some archs where it was missing.

branches: 1.4.6;
Add C99 functions imaxabs and imaxdiv.

branches: 1.3.20;
build erand48_ieee754.c on everything but vax

Add ISO C99 long long integer general utility interfaces; partially addresses
PR standards/9482.

Add sparc64-specific libc.

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

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

Add sparc64-specific libc.

Change NO_SRCS lines to specify .S in order to exclude the .c file from
SRCS but add it to LSRCS (sort of makes sense!).

Stop bcopy.c and bzero.c being compiled

If SRCS contains any .S files then remove the corresponding .c file from
SRCS and add to LSRCS (for lint).
Change the 'string' Makefiles so that the arch/*/string/Makefile.inc
need only specify the .S files for that architecture and not the .c
files for all the files they don't override.

Remove references to index.c and rindex.c (strchr.c and strrchr.c
now provide the definitions).

Replace bzero.S and bcopy.S with memset.S and memcpy.S to make the makefiles
happy.

Add an optimized asssembly bcopy/memcpy.

Add an assembly version of memset(3).

branches: 1.1.12;
Add sparc64-specific libc.

Replace bzero.S and bcopy.S with memset.S and memcpy.S to make the makefiles
happy.

Fix more boundary issues.

Fix some boundary condition issues.

Re-enable block insns for cahe issues.

Fix return value and fetches beyond the source range that could cause
segmentation faults in some cases. And disable the VIS version which
seems to be slower.

Oops. Fix typo.

I seem to have stumbled on an even faster bcopy implementation....

Add an optimized asssembly bcopy/memcpy.

Replace bzero.S and bcopy.S with memset.S and memcpy.S to make the makefiles
happy.

Re-enable block insns since they don't allocate in the caches.

Disable slower VIS version.

branches: 1.8.2;
Don't flush the zero-ed data from the cache.

Offset the %i2 stack slot from %sp, not %fp. Otherwise we may modify the
caller's registers!

There's no good reason not to assume a 64-bit stack here.

Don't bother saving %f16-%f31, since we don't use them.

Just use a store/load pair to get the pattern into the FP reg. This eliminates
a bunch of cruft and avoids using a v9a instruction.
In addition, eliminate 8 of the fmovda's, which we are not using the result of
anyway.
Net result is that this should be faster in all cases.

Block align pointer *AFTER* fixing up possible stack bias.

branches: 1.2.4;
New bzero() using block store insns.

Add sparc64-specific libc.

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.

Removed unused PIC setup code.

branches: 1.4.2;
Fix PIC bugs.

branches: 1.3.4;
Sample code to implement ffs() using the popc insn (if it's ever implemented).

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

now live in common/lib/libc/arch/sparc64/string

branches: 1.5.2; 1.5.8;
rename sparc64 BLOCK_SIZE and BLOCK_ALIGN to have SPARC64_ prefixes.
for the assembler files, define the old names to the new names
since using the new names cause ugliness due to longer identifer
names, and reduces churn.

fixes build issues in dtv and vaguely makes <machine/psl.h> slightly
less name-space invasive.

use _ALIGN consistently.

make this compile with DEBUG (_ALIGN was not defined).

branches: 1.2.2;
Use unsigned compares for addresses.

Replace bzero.S and bcopy.S with memset.S and memcpy.S to make the makefiles
happy.

now live in common/lib/libc/arch/sparc64/string

branches: 1.4.2; 1.4.60; 1.4.66;
Improved version of bzero.

Fix some corner cases.

Replace bzero.S and bcopy.S with memset.S and memcpy.S to make the makefiles
happy.

Add an assembly version of memset(3).

Use reach-over sources from common/lib/libc

The mask for 8-byte alignment is 0x7 not 0xf.

Remove test harnass bits.

Better, faster strlen().

branches: 1.1.12;
Add sparc64-specific libc.

Bring sparc64 to the new compat layout.

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

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

Use cpp comment style.

Warn about references to the compatibility vfork() (i.e. the vfork() with
the 4.4BSD non-shared-address-space semantics), and direct the user to
include <unistd.h> to generate the correct reference.

This warning isn't about an ABI compatibility issue, but the new vfork() is
considerably faster.

Add sparc64-specific libc.

branches: 1.7.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.6.10;
After discussing this further with uwe, rearange stack usage slightly and
replace a comment with words suggested by him.

When setting up the __clone()'s userland stack, make sure to allocate
an initial frame. Fixes PR 33075.

branches: 1.4.16;
Restore stack frames on error exit. This makes us pass dynamic and static
linked versions of the clonetest regression test.
While here, some cosmetics: branch prediction, avoid depreceated instructions.

Use CC64FSZ instead of magic constant for stack frame size.
Frob the clone's stack pointer to be BIASed - it is 64 bit code too.
Use "carry set" as test for a failed syscall.

Use large (-fPIC) relocations.

branches: 1.1.2;
sparc64 version, cloned (har har!) from the sparc version and tweaked.

Bring sparc64 to the new compat layout.

Add __msgctl13(), __semctl13(), and __shmctl13() (symbols are __RENAME()'d
in their respective header files). Add PSEUDO stubs for the old versions
of these system calls.

Note that __semctl13() doesn't require the extra stub that the old
semctl()/__semctl() did, as the SUSv2 version takes variable arguments,
and thus does not require the extra indirection.

Merge all MD __sigaction14_sigtramp.c copies into one:
- sparc and sparc64 were not using version 0 sigcontext when there were
no arguments in the signal version. This was probably a bug.
- vax is using +1 the version numbers of the other archs.
- Only hppa was defining __LIBC12_SOURCE__ so it was getting a working
sigcontext before. all the other ports that supported sigcontext had
the compat code disabled.
[pointed out by thorpej, thanks!]
If we want to remove sigcontext support from userland at least now there
is less work to do so.

merge christos-time_t

Remove __LIBC12_SOURCE__

branches: 1.7.6; 1.7.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.6.4; 1.6.18;
make these consistent.

Add missing __RCSID()

save errno while trying to install a !SA_SIGINFO trampoline -- this
is expected to fail with !COMPAT_16 kernels, and we shouldn't
overwrite errno in this case

Signal trampoline stuff for siginfo.
XXX: I don't understand why I am off by 48 bytes.

Merge the nathanw_sa branch.

branches: 1.1.2;
New __sigaction14() system call stub which registers the signal
trampoline, and the signal trampolines themselves.

NOTE: These are not yet enabled; they will be enabled after further
testing on more architectures.

Bring sparc64 to the new compat layout.

fix references to the syscall formerly known as SYS___sigreturn14

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

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

Fix libc signal breakage (I hope).

Bring sparc64 to the new compat layout.

fix references to the syscall formerly known as SYS___sigreturn14

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

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

Compile with an ISO C preprocessor.

branches: 1.2.2;
Also need <machine/fsr.h> and <machine/psl.h>.

New __sigaction14() system call stub which registers the signal
trampoline, and the signal trampolines themselves.

NOTE: These are not yet enabled; they will be enabled after further
testing on more architectures.

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

branches: 1.2.30;
Explain the stack layout correctly now that that I understand it :-)

Signal trampoline stuff for siginfo.
XXX: I don't understand why I am off by 48 bytes.

Add simple __syscall.S to make libc compile on sparc64 again.

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

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

Use cpp comment style.

Add sparc64-specific libc.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.12.56; 1.12.62;
Use an unsigned comparison against minbrk.
From OpenBSD, changed by Miod Vallat.

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

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

Use large (-fPIC) relocations.

branches: 1.9.2;
Store the whole 64-bit pointer...

Use cpp comment style.

Update some more assembly code for PIC issues.

Fix PIC bugs.

Thinko in previous: we do always use an internal name.

Add internal names for brk() and sbrk().

branches: 1.3.10;
Change references to `end' to `_end' as not to invade (or being invaded by)
the user's name space.

* On architectures using ELF, and conditionally on those in transition, rename
_ASM_LABEL(cerror) and _ASM_LABEL(curbrk) to _C_LABEL(__cerror) and
_C_LABEL(__curbrk) (or their respective architecture-specific equivalents) to
avoid possible name clashes with identifiers used in user applications.
* Do the same for minbrk on all architectures to avoid a GCC-specific (and
on ELF architectures effectively useless) symbol reference renaming in MI code.

Add sparc64-specific libc.

Redo previous to fix build break.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.6.56; 1.6.62;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Use large (-fPIC) relocations.

branches: 1.4.2;
Need to reserve a full 64-bit stackframe, not a 32-bit one.

Fix PIC bugs.

branches: 1.2.10;
* On architectures using ELF, and conditionally on those in transition, rename
_ASM_LABEL(cerror) and _ASM_LABEL(curbrk) to _C_LABEL(__cerror) and
_C_LABEL(__curbrk) (or their respective architecture-specific equivalents) to
avoid possible name clashes with identifiers used in user applications.
* Do the same for minbrk on all architectures to avoid a GCC-specific (and
on ELF architectures effectively useless) symbol reference renaming in MI code.

Add sparc64-specific libc.

Mark exect(3) obsolete and bind it to plain execve(2) on all platforms

The original exect(2) from BSD4.2 was enabling bit for tracing
(single-step mode) and calling execve(2). The purpose of it was to generate
a signal for a tracer once the application will change its image to a new
program.

This approach no longer works as:
- exect(2) traces (single-steps) libc and it requires hundreds or
thousands steps before entering a new image
- it's vax and x86 specific code
- this functionality has been moved to the kernel - once a process is
traced it will generate SIGTRAP with si_code TRAP_EXEC and route it to
its debugger
- the side effects and unportability make this interface unusable
- there are no known users of this interface
- it apparently never worked better since day0 of NetBSD ("day0 bug")

Users are requested to move to other execve(2) variants. Calling current
execve(2) as it is the most similar behavior to this one from BSD4.2.

Discussed several times on mailing lists and in PR/51700.

Add warning to exect(3) telling about marking this function obsolete.

This function is prepared to be removed in next libc major bump.

Sponsored by <The NetBSD Foundation>

branches: 1.2.80; 1.2.84;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Add sparc64-specific libc.

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

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

Implement pthread_atfork() (in libc, because the required threadlib
stub behavior is exactly the same as the usual behavior).

Use cpp comment style.

Need an internal name for fork().

branches: 1.1.10;
Add sparc64-specific libc.

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

branches: 1.2.30;
Merge the nathanw_sa branch.

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

Bring sparc64 to the new compat layout.

Add __msgctl13(), __semctl13(), and __shmctl13() (symbols are __RENAME()'d
in their respective header files). Add PSEUDO stubs for the old versions
of these system calls.

Note that __semctl13() doesn't require the extra stub that the old
semctl()/__semctl() did, as the SUSv2 version takes variable arguments,
and thus does not require the extra indirection.

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

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

Need an internal name for pipe(2).

Use cpp comment style.

branches: 1.1.10;
Add sparc64-specific libc.

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

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

_errno -> __errno

Update some more assembly code for PIC issues.

Fix PIC bugs.

branches: 1.1.10;
Add sparc64-specific libc.

Redo previous to fix build break.

Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.9.56; 1.9.62;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Use large (-fPIC) relocations.

branches: 1.7.2;
Use cpp comment style.

Fix PIC bugs.

Thinko in previous: we do always use an internal name.

Add internal names for brk() and sbrk().

branches: 1.3.10;
Change references to `end' to `_end' as not to invade (or being invaded by)
the user's name space.

* On architectures using ELF, and conditionally on those in transition, rename
_ASM_LABEL(cerror) and _ASM_LABEL(curbrk) to _C_LABEL(__cerror) and
_C_LABEL(__curbrk) (or their respective architecture-specific equivalents) to
avoid possible name clashes with identifiers used in user applications.
* Do the same for minbrk on all architectures to avoid a GCC-specific (and
on ELF architectures effectively useless) symbol reference renaming in MI code.

Add sparc64-specific libc.

Merge the nathanw_sa branch.

Use large (-fPIC) relocations.

branches: 1.3.2;
Update some more assembly code for PIC issues.

Fix PIC bugs.

branches: 1.1.10;
Add sparc64-specific libc.

Add special treatment of shmat(2).
On m68k ELF calling standard (__SVR4_ABI__), if a function
returns a pointer, the return value should go to a0 as well as d0.

Bring sparc64 to the new compat layout.

Add __msgctl13(), __semctl13(), and __shmctl13() (symbols are __RENAME()'d
in their respective header files). Add PSEUDO stubs for the old versions
of these system calls.

Note that __semctl13() doesn't require the extra stub that the old
semctl()/__semctl() did, as the SUSv2 version takes variable arguments,
and thus does not require the extra indirection.

Bring sparc64 to the new compat layout.

Warn about references to the compatibility() sigaction(), sigpending(),
sigprocmask(), sigreturn(), and sigsuspend(), and direct the user to
include <signal.h> to generate the correct reference.

Fix libc signal breakage (I hope).

Bring sparc64 to the new compat layout.

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

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

Use cpp comment style.

Warn about references to the compatibility() sigaction(), sigpending(),
sigprocmask(), sigreturn(), and sigsuspend(), and direct the user to
include <signal.h> to generate the correct reference.

Fix libc signal breakage (I hope).

Back out previous change; it was incorrect.

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

Bring sparc64 to the new compat layout.

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

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

Use cpp comment style.

Update some more assembly code for PIC issues.

branches: 1.5.10;
Warn about references to the compatibility() sigaction(), sigpending(),
sigprocmask(), sigreturn(), and sigsuspend(), and direct the user to
include <signal.h> to generate the correct reference.

Fix libc signal breakage (I hope).

Back out previous change; it was incorrect.

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

Bring sparc64 to the new compat layout.

the WARN_REFERENCES text "include <signal.h> for correct reference" is pointless
as <signal.h> won't help you, so remove it

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

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

Warn about references to the compatibility() sigaction(), sigpending(),
sigprocmask(), sigreturn(), and sigsuspend(), and direct the user to
include <signal.h> to generate the correct reference.

Fix libc signal breakage (I hope).

Back out previous change; it was incorrect.

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

Bring sparc64 to the new compat layout.

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

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

Use cpp comment style.

Warn about references to the compatibility() sigaction(), sigpending(),
sigprocmask(), sigreturn(), and sigsuspend(), and direct the user to
include <signal.h> to generate the correct reference.

Fix libc signal breakage (I hope).

Back out previous change; it was incorrect.

Libc fixes, mostly to work around the new signal stuff.

Add sparc64-specific libc.

Rename "syscall" to "_syscall" and provide "syscall" as a weak alias.

branches: 1.2.98;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

Add sparc64-specific libc.