The previous code used to evaluate the preprocessor token is_lock_free to
a variable before starting a transaction. This behavior can cause an
error if another thread got the lock (without using a transaction)
between the evaluation of the token and the beginning of the transaction.
This bug can be triggered with the following order of events:
1. The lock accessed by is_lock_free is free.
2. Thread T1 evaluates is_lock_free and stores into register R1 that the
lock is free.
3. Thread T2 acquires the same lock used in is_lock_free.
4. T1 begins the transaction, creating a memory barrier where is_lock_free
is false, but R1 is true.
5. T1 reads R1 and doesn't abort the transaction.
6. T1 calls ELIDE_UNLOCK, which reads false from is_lock_free and decides
to unlock a lock acquired by T2, leading to undefined behavior.
This patch delays the evaluation of is_lock_free to inside a transaction
by moving this part of the code to the macro ELIDE_LOCK.
[BZ #18743]
* sysdeps/powerpc/nptl/elide.h (__elide_lock): Move most of this
code to...
(ELIDE_LOCK): ...here.
(__get_new_count): New function with part of the code from
__elide_lock that updates the value of adapt_count after a
transaction abort.
(__elided_trylock): Moved this code to...
(ELIDE_TRYLOCK): ...here.
sysdeps/nptl/configure.ac has code to give errors if certain tests in
the top-level configure failed. However, all those failure conditions
also produce errors in the top-level configure, so the errors in the
NPTL configure are completely redundant; this patch removes them.
(As suggested in
<https://sourceware.org/ml/libc-alpha/2015-10/msg00510.html>, I think
the top-level tests in question can be completely removed as
unnecessary given the version tests. But even without that there is
clearly no point in duplicating code that gives an error if the test
fails.)
Tested for x86_64 (testsuite, and that installed shared libraries are
unchanged by the patch).
* sysdeps/nptl/configure.ac: Do not give errors based on the
results of top-level configure tests.
* sysdeps/nptl/configure: Regenerated.
With TLE enabled, the adapt count variable update incurs
an 8% overhead before entering the critical section of an
elided mutex.
Instead, if it is done right after leaving the critical
section, this serialization can be avoided.
This alters the existing behavior of __lll_trylock_elision
as it will only decrement the adapt_count if it successfully
acquires the lock.
* sysdeps/unix/sysv/linux/powerpc/elision-lock.c
(__lll_lock_elision): Remove adapt_count decrement...
* sysdeps/unix/sysv/linux/powerpc/elision-trylock.c
(__lll_trylock_elision): Likewise.
* sysdeps/unix/sysv/linux/powerpc/elision-unlock.c
(__lll_unlock_elision): ... to here. And utilize
new adapt_count parameter.
* sysdeps/unix/sysv/linux/powerpc/lowlevellock.h
(__lll_unlock_elision): Update to include adapt_count
parameter.
(lll_unlock_elision): Pass pointer to adapt_count
variable.
Adding this parameter will give architectures more freedom in
how they choose to update this variable. This change has no
effect on architectures which choose not to use it.
* nptl/pthread_mutex_unlock.c(lll_unlock_elision):
Add elision adapt_count parameter to list of arguments.
* sysdeps/unix/sysv/linux/powerpc/lowlevellock.h
(lll_unlock_elision): Update with new parameter list
* sysdeps/unix/sysv/linux/s390/lowlevellock.h
(lll_unlock_elision): Likewise
* sysdeps/unix/sysv/linux/x86_64/lowlevellock.h
(lll_unlock_elision): Likewise
Since _dl_catch_error is only used internally in ld.so, it should be
declared in sysdeps/generic/ldsodefs.h, not include/dlfcn.h and it can
be made hidden.
[BZ #19122]
* include/dlfcn.h (_dl_catch_error): Moved to ...
* sysdeps/generic/ldsodefs.h (_dl_catch_error): Add
attribute_hidden.
Since internal _itoa functions are only used internally in ld.so and
libc.so, they can be made hidden.
[BZ #19122]
* sysdeps/generic/_itoa.h (_itoa): Add attribute_hidden.
(_itoa_word): Likewise.
Since _wordcopy_XXX functions are only used internally in ld.so and
libc.so, they can be made hidden.
[BZ #19122]
* sysdeps/generic/memcopy.h (_wordcopy_fwd_aligned): Add
attribute_hidden.
(_wordcopy_fwd_dest_aligned): Likewise.
(_wordcopy_bwd_aligned): Likewise.
(_wordcopy_bwd_dest_aligned): Likewise.
Since x86 _dl_unmap and _dl_make_tlsdesc_dynamic are only used
internally in ld.so, they can be made hidden.
[BZ #19122]
* sysdeps/i386/dl-lookupcfg.h (_dl_unmap): Add attribute_hidden.
* sysdeps/i386/dl-tlsdesc.h (_dl_make_tlsdesc_dynamic):
Likewise.
* sysdeps/x86_64/dl-tlsdesc.h (_dl_make_tlsdesc_dynamic):
Likewise.
* sysdeps/x86_64/dl-lookupcfg.h (_dl_unmap): Likewise.
There is a configure test for assembler support for -mtune=i686. This
option was added in binutils 2.18 so the test is obsolete; this patch
removes it.
Tested for x86 (testsuite, and that installed shared libraries are
unchanged by the patch).
* sysdeps/i386/configure.ac (libc_cv_as_i686): Remove configure
test.
* sysdeps/i386/configure: Regenerated.
* sysdeps/i386/i686/Makefile [$(config-asflags-i686) = yes]: Make
code unconditional.
Only i386 implements epoll_pwait in assembly code withot cancellation
support. All other architectures implement epoll_pwait in epoll_pwait.c
with
int epoll_pwait (int epfd, struct epoll_event *events,
int maxevents, int timeout,
const sigset_t *set)
{
return SYSCALL_CANCEL (epoll_pwait, epfd, events, maxevents,
timeout, set, _NSIG / 8);
}
Although there is no test for epoll_pwait in glibc, since SYSCALL_CANCEL
works on i386 and epoll_pwait.c works for other architectures, it is
safe to assume that epoll_pwait.c with SYSCALL_CANCEL also works on
i386.
[BZ #19137]
* sysdeps/unix/sysv/linux/i386/Makefile (CFLAGS-epoll_pwait.c):
Add -fomit-frame-pointer.
* sysdeps/unix/sysv/linux/i386/epoll_pwait.S: Remove file.
Honoring the LD_POINTER_GUARD environment variable in AT_SECURE mode
has security implications. This commit enables pointer guard
unconditionally, and the environment variable is now ignored.
[BZ #18928]
* sysdeps/generic/ldsodefs.h (struct rtld_global_ro): Remove
_dl_pointer_guard member.
* elf/rtld.c (_rtld_global_ro): Remove _dl_pointer_guard
initializer.
(security_init): Always set up pointer guard.
(process_envvars): Do not process LD_POINTER_GUARD.
The powerpc32 implementation of lround and lroundf can produce
spurious exceptions from adding 0.5 then converting to integer. This
includes "inexact" from the conversion to integer (not allowed for
integer arguments to these functions), and, for larger integer
arguments, "inexact", and "overflow" when rounding upward, from the
addition. In addition, "inexact" is not allowed together with
"invalid" and so inexact addition must be avoided when the integer
will be out of range of 32-bit long, whether or not the argument is an
integer.
This patch fixes these problems. As in the powerpc64 llround
implementation, a check is added for too-large arguments; in the
powerpc64 case that means arguments at least 2^52 in magnitude (so
that 0.5 cannot be added exactly), while in this case it means
arguments for which the result would overflow "long". In those cases
a suitable overflowing value is used for the integer conversion
without adding 0.5, while for smaller arguments it's tested whether
the argument is an integer (by adding and subtracting 2^52 to the
absolute value and comparing with the original absolute value) to
avoid adding 0.5 to integers and generating spurious "inexact".
This code is not used when the power5+ sysdeps directories are used,
as there's a separate power5+ version of these functions..
Tested for powerpc. This gets test-float (for a default powerpc32
hard-float build without any --with-cpu) back to the point where it
should pass once powerpc ulps are regenerated; test-double still needs
another problem with exceptions fixed to get back to that point (and I
haven't looked lately at what default powerpc64 results are like).
[BZ #19134]
* sysdeps/powerpc/powerpc32/fpu/s_lround.S (.LC1): New object.
(.LC2): Likewise.
(.LC3): Likewise.
(__lround): Do not add 0.5 to integer or out-of-range arguments.
_dl_tlsdesc_resolve_hold calls into a C function that clobbers r0,
but it assumes the original argument is still in r0 after the call.
This can cause crash in case of concurrent TLS access when TLSDESC
is in use (-mtls-dialect=gnu2).
Run into this while fixing BZ 18572.
Both r0 and r1 are saved/restored so the stack remains 8 byte aligned.
[BZ #19129]
* sysdeps/arm/dl-tlsdesc.S (_dl_tlsdesc_resolve_hold): Save and restore
r0 and r1.
This patch adds an internal entry for __sched_getaffinity_new so that
__sched_getaffinity_old calls __sched_getaffinity_new without going
through PLT.
[BZ #18822]
* sysdeps/unix/sysv/linux/sched_getaffinity.c
(__sched_getaffinity_new): Add libc_hidden_proto and
libc_hidden_def.
Linker in binutils 2.26 and newer generate GOT references instead
PLT references when -z now is passed to linker. We need to extend
scripts/localplt.awk to allow PLT or GOT references.
[BZ #19007]
* scripts/localplt.awk: Also allow GOT references.
* sysdeps/unix/sysv/linux/i386/localplt.data: Mark
_Unwind_Find_FDE, calloc, memalign, realloc and __libc_memalign
with "+ REL R_386_GLOB_DAT".
* sysdeps/x86_64/localplt.data: Mark calloc, memalign, realloc
and __libc_memalign with "+ RELA R_X86_64_GLOB_DAT".
This patch uses INTERNAL_SYSCALL and INLINE_SYSCALL_ERROR_RETURN_VALUE
to avoid reading and writing errno directly so that we don't need to
call __x86.get_pc_thunk.reg to load PC into reg in case there is an
error.
* sysdeps/unix/sysv/linux/i386/brk.c (__brk): Use
INLINE_SYSCALL_ERROR_RETURN_VALUE.
* sysdeps/unix/sysv/linux/i386/fxstatat.c (__fxstatat):
Likewise.
* sysdeps/unix/sysv/linux/i386/setegid.c (setegid): Likewise.
* sysdeps/unix/sysv/linux/i386/seteuid.c (seteuid): Likewise.
* sysdeps/unix/sysv/linux/i386/fxstat.c (__fxstat): Use
INTERNAL_SYSCALLINTERNAL_SYSCALL and
INLINE_SYSCALL_ERROR_RETURN_VALUE.
* sysdeps/unix/sysv/linux/i386/lockf64.c (lockf64): Likewise.
* sysdeps/unix/sysv/linux/i386/lxstat.c (__lxstat): Likewise.
* sysdeps/unix/sysv/linux/i386/sigaction.c (__libc_sigaction):
Likewise.
* sysdeps/unix/sysv/linux/i386/xstat.c (__xstat): Likewise.
The powerpc32 implementations of llroundf and llround produce spurious
and missing exceptions (some arising from such exceptions from
conversions to long long, some present even when fctidz is used).
This patch fixes those problems in a similar way to the llrint /
llrintf fixes. The spurious exceptions in the fctidz case for large
arguments arise from a converted value that saturated as LLONG_MAX
being converted back to float or double (the conversion back being
inexact, but "inexact" must not be raised together with "invalid"),
and from the subtraction x - xrf also being inexact for sufficiently
large arguments (whether the saturation was to LLONG_MAX or
LLONG_MIN); those are fixed by returning early if the argument is
large enough that no rounding is needed.
This code is not used for --with-cpu=power4 builds (I suspect the code
used in that case may also produce spurious "inexact" exceptions, but
that's something to investigate later).
Tested for powerpc.
[BZ #19125]
* sysdeps/powerpc/powerpc32/fpu/s_llround.c: Include <limits.h>,
<math_private.h> and <stdint.h>.
(__llround): Avoid conversions to and from long long int, and
subtractions, where those might raise spurious exceptions.
* sysdeps/powerpc/powerpc32/fpu/s_llroundf.c: Include
<math_private.h> and <stdint.h>.
(__llroundf): Avoid conversions to and from long long int, and
subtractions, where those might raise spurious exceptions.
When x86-64 assmebler doesn't support AVX512, we should make
_dl_runtime_resolve_avx512/_dl_runtime_profile_avx512 as aliases of
_dl_runtime_resolve_avx/_dl_runtime_profile_avx. Tested on x86-64
using GCC 5.2 with binutils 20151008 and GCC 4.8 with binutils 20130219.
There are no differences in ld.so with binutils 20151008. There are no
unexpected failures with binutils 20130219 and 20151008.
[BZ #19124]
* sysdeps/x86_64/dl-trampoline.S [!HAVE_AVX512_ASM_SUPPORT]
(_dl_runtime_resolve_avx512): Make it a hidden alias of
_dl_runtime_resolve_avx.
(_dl_runtime_profile_avx512): Make it a hidden alias of
_dl_runtime_profile_avx.
sysdeps/powerpc/fpu/ has versions of llround and llroundf that are
actually used only for powerpc32 because
sysdeps/powerpc/powerpc64/fpu/ has its own versions of those
functions. This patch moves them into sysdeps/powerpc/powerpc32/fpu
to reflect where they are actually used (in preparation for fixing
other problems with those functions).
Tested for powerpc that installed stripped shared libraries are
unchanged by this patch.
* sysdeps/powerpc/fpu/s_llround.c: Move to ....
* sysdeps/powerpc/powerpc32/fpu/s_llround.c: ...here.
* sysdeps/powerpc/fpu/s_llroundf.c: Move to ....
* sysdeps/powerpc/powerpc32/fpu/s_llroundf.c: ...here.
The versions of llrint and llrintf for older powerpc32 processors
convert the results of __rint / __rintf to long long int, resulting in
spurious exceptions from such casts in certain cases. This patch
makes glibc work around the problems with the libgcc conversions when
the compiler used to build glibc doesn't use the fctidz instruction
for them.
Tested for powerpc.
[BZ #16422]
* sysdeps/powerpc/powerpc32/fpu/configure.ac (libc_cv_ppc_fctidz):
New configure test.
* sysdeps/powerpc/powerpc32/fpu/configure: Regenerated.
* config.h.in [_LIBC] (HAVE_PPC_FCTIDZ): New macro.
* sysdeps/powerpc/powerpc32/fpu/s_llrint.c: Include <limits.h>,
<math_private.h> and <stdint.h>.
(__llrint): Avoid conversions to long long int where those might
raise spurious exceptions.
* sysdeps/powerpc/powerpc32/fpu/s_llrintf.c: Include
<math_private.h> and <stdint.h>.
(__llrintf): Avoid conversions to long long int where those might
raise spurious exceptions.
Similar to the recent fix for MIPS, ARM is also missing correct
exceptions on overflow from llrint and llround functions because casts
from floating-point types to long long do not result in correct
exceptions on overflow. This patch enables the fix for this for ARM.
Tested for ARM.
[BZ #15470]
* sysdeps/arm/fix-fp-int-convert-overflow.h: New file.
For 32-bit MIPS and some other systems, various of the lrint, llrint,
lround, llround functions can be missing exceptions on overflow
because casts do not (in current GCC) result in the proper
exceptions. In the MIPS case there are two problems here: MIPS I code
generation uses an assembler macro that doesn't raise exceptions,
while the libgcc conversions of floating-point values to long long
also do not raise "invalid" on all overflow cases (and can raise
spurious "inexact").
This patch adds support in the generic code (only the functions for
which this problem has actually been seen) for forcing the "invalid"
exception in the problem cases, and enables that support for the
affected MIPS cases.
Tested for MIPS; also tested for x86_64 and x86 that installed
stripped shared libraries are unchanged by this patch.
[BZ #16399]
* sysdeps/generic/fix-fp-int-convert-overflow.h: New file.
* sysdeps/ieee754/dbl-64/s_llrint.c: Include <fenv.h>, <limits.h>
and <fix-fp-int-convert-overflow.h>.
(__llrint) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/dbl-64/s_llround.c: Include <fenv.h>, <limits.h>
and <fix-fp-int-convert-overflow.h>.
(__llround) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/dbl-64/s_lrint.c: Include
<fix-fp-int-convert-overflow.h>.
(__lrint) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/dbl-64/s_lround.c: Include
<fix-fp-int-convert-overflow.h>.
(__lround) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/flt-32/s_llrintf.c: Include <fenv.h>, <limits.h>
and <fix-fp-int-convert-overflow.h>.
(__llrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/flt-32/s_llroundf.c: Include <fenv.h>,
<limits.h> and <fix-fp-int-convert-overflow.h>.
(__llroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/flt-32/s_lrintf.c: Include <fenv.h>, <limits.h>
and <fix-fp-int-convert-overflow.h>.
(__lrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/ieee754/flt-32/s_lroundf.c: Include <fenv.h>, <limits.h>
and <fix-fp-int-convert-overflow.h>.
(__lroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
FIX_DBL_LLONG_CONVERT_OVERFLOW.
* sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h: New file.
The x86_64 versions of lrint/lrintf/ lrintl are aliases for the long
long versions which isn't correct for x32, where exceptions must respect
overflow for 32-bit long. Separate versions of the long functions for
x32 that convert to 32-bit long and raise the right exceptions for that
conversion, while keeping the aliases in the non-x32 case.
Tested on x86_64 and x32. There are no code changes in libm.so on
x86_64.
* sysdeps/x86_64/fpu/s_llrint.S (__lrint): Add alias only if
__ILP32__ isn't defined.
(lrint): Likewise.
* sysdeps/x86_64/fpu/s_llrintf.S (__lrintf): Likewise.
(lrintf): Likewise.
* sysdeps/x86_64/fpu/s_llrintl.S (__lrintl): Likewise.
(lrintl): Likewise.
* sysdeps/x86_64/x32/fpu/s_lrint.S: New file.
* sysdeps/x86_64/x32/fpu/s_lrintf.S: Likewise.
* sysdeps/x86_64/x32/fpu/s_lrintl.S: Likewise.
This patch sets lseek/llseek for 64-bit, MIPS n32, and x86_32 as non-
cancelable. This make it consistant with 32-bit platform.
Tested on i686, x86_64, and x32.
* sysdeps/unix/sysv/linux/mips/mips64/syscalls.list (lseek): Set as
non-cancelable.
* sysdeps/unix/sysv/linux/wordsize-64/syscalls.list (llseek): Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/lseek.S (__libc_lseek64):
Likewise.
GCC added support for -mno-vzeroupper in version 4.6. Thus the
configure tests for this support are obsolete, and this patch removes
them.
Tested for x86_64 and x86 (testsuite, and that installed stripped
shared libraries are unchanged by this patch).
* sysdeps/i386/configure.ac (libc_cv_cc_novzeroupper): Remove
configure test.
* sysdeps/i386/configure: Regenerated.
* sysdeps/x86_64/configure.ac (libc_cv_cc_novzeroupper): Remove
configure test.
* sysdeps/x86_64/configure: Regenerated.
* sysdeps/x86_64/Makefile [$(config-cflags-novzeroupper) = yes]:
Make code unconditional.
The dbl-64 implementation of lrint produces incorrect results for some
arguments with 64-bit long because a 32-bit (unsigned) low part of the
mantissa is shifted left, losing high bits in the process. This patch
fixes this by casting to long int before shifting, as in lround (as
this case only applies for 64-bit long, there are no issues with
sign-extension).
Tested for mips64 (n64).
[BZ #19095]
* sysdeps/ieee754/dbl-64/s_lrint.c (__lrint): Cast low part of
mantissa to long int before shifting left.
The dbl-64, ldbl-96 and ldbl-128 implementations of lrint and llrint
fail to produce "invalid" exceptions in cases where the rounded result
overflows the target type, but truncating the floating-point argument
to the next integer towards zero does not overflow it (so in
particular casts do not produce such exceptions). (This issue cannot
arise for float, or for double with 64-bit target type, or for ldbl-96
with 64-bit target type and negative arguments, because of
insufficient precision in the floating-point type for arguments with
the relevant property to exist. It also obviously cannot arise in
FE_TOWARDZERO mode.)
This patch fixes these problems by inserting checks for the special
cases that can occur in each implementation, and explicitly raising
FE_INVALID (and avoiding the cast if it might raise spurious
FE_INEXACT, while raising FE_INEXACT explicitly in the cases where it
is needed; unlike lround and llround, FE_INEXACT is required, not
optional, for these functions for a within-range inexact result).
The fixes are conditional on FE_INVALID or FE_INEXACT being defined.
If any future architecture supports one but not both of those
exceptions, the code will fail to compile and need fixing to handle
that case (this seemed better than conditioning on both macros being
defined, resulting in code that would compile but quietly miss
exceptions on such a system).
Tested for x86_64, x86 and mips64. Tested the ldbl-96 changes (only
relevant for ia64, it appears) on x86_64 by removing the x86_64
versions of lrintl / llrintl.
[BZ #19094]
* sysdeps/ieee754/dbl-64/s_lrint.c: Include <fenv.h> and
<limits.h>.
(__lrint) [FE_INVALID || FE_INEXACT]: Force FE_INVALID exception
when result overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-128/s_llrintl.c: Include <fenv.h> and
<limits.h>.
(__llrintl) [FE_INVALID || FE_INEXACT]: Force FE_INVALID exception
when result overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-128/s_lrintl.c: Include <fenv.h> and
<limits.h>.
(__lrintl) [FE_INVALID || FE_INEXACT]: Force FE_INVALID exception
when result overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-96/s_llrintl.c: Include <fenv.h> and
<limits.h>.
(__llrintl) [FE_INVALID || FE_INEXACT]: Force FE_INVALID exception
when result overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-96/s_lrintl.c: Include <fenv.h> and
<limits.h>.
(__lrintl) [FE_INVALID || FE_INEXACT]: Force FE_INVALID exception
when result overflows but exception would not result from cast.
* math/libm-test.inc (lrint_test_data): Add more tests.
(llrint_test_data): Likewise.
The dbl-64, ldbl-96 and ldbl-128 implementations of lround and llround
fail to produce "invalid" exceptions in cases where the rounded result
overflows the target type, but truncating the floating-point argument
to the next integer towards zero does not overflow it (so in
particular casts do not produce such exceptions). (This issue cannot
arise for float, or for double with 64-bit target type, or for ldbl-96
with 64-bit target type and negative arguments, because of
insufficient precision in the floating-point type for arguments with
the relevant property to exist.)
This patch fixes these problems by inserting checks for the special
cases that can occur in each implementation, and explicitly raising
FE_INVALID (and avoiding the cast if it might raise spurious
FE_INEXACT).
Tested for x86_64, x86 and mips64.
[BZ #19088]
* sysdeps/ieee754/dbl-64/s_lround.c: Include <fenv.h> and
<limits.h>.
(__lround) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* sysdeps/ieee754/dbl-64/wordsize-64/s_lround.c: Include <fenv.h>
and <limits.h>.
(__lround) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-128/s_llroundl.c: Include <fenv.h> and
<limits.h>.
(__llroundl) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-128/s_lroundl.c: Include <fenv.h> and
<limits.h>.
(__lroundl) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-96/s_llroundl.c: Include <fenv.h> and
<limits.h>.
(__llroundl) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* sysdeps/ieee754/ldbl-96/s_lroundl.c: Include <fenv.h> and
<limits.h>.
(__lroundl) [FE_INVALID]: Force FE_INVALID exception when result
overflows but exception would not result from cast.
* math/libm-test.inc (lround_test_data): Add more tests.
(llround_test_data): Likewise.
The ldbl-128 implementations of lrintl and lroundl miss "invalid"
exceptions on systems with 32-bit long for arguments that overflow
long but have exponent below 48. This patch fixes this by rearranging
the sequence of tests in the code so the exponent < 48 case is only
used for exponents that don't overflow long.
Tested for mips64 (n32 and n64).
[BZ #19085]
* sysdeps/ieee754/ldbl-128/s_lrintl.c (__lrintl): Move test for
exponent below 48 inside case for non-overflowing exponent.
* sysdeps/ieee754/ldbl-128/s_lroundl.c (__lroundl): Likewise.
This patch enables use of sysdeps/ieee754/dbl-64/wordsize-64 for
MIPS64 (both n64 and n32), removing a #error in one case now that case
has been tested and found to work.
Tested for mips64 (n64 and n32).
* sysdeps/mips/mips64/Implies: Use ieee754/dbl-64/wordsize-64.
* sysdeps/ieee754/dbl-64/wordsize-64/s_issignaling.c
(__issignaling) [HIGH_ORDER_BIT_IS_SET_FOR_SNAN]: Remove #error.
The implementation of lround in dbl-64/wordsize-64 as an alias or
wrapper for llround is always incorrect when long is not 64-bit,
because it misses required exceptions in overflow cases, as shown by
my recently added tests. This patch removes that alias / wrapper in
the non-LP64 case, together with the REGISTER_CAST_INT32_TO_INT64
macro, restoring the previous version of lround for dbl-64/wordsize-64
(newly conditioned on !_LP64).
Tested for x86_64, and for mips64 with use of dbl-64/wordsize-64
enabled.
[BZ #19079]
* sysdeps/ieee754/dbl-64/wordsize-64/s_lround.c: Restore previous
file, conditioned on [!_LP64].
* sysdeps/ieee754/dbl-64/wordsize-64/s_llround.c
[!_LP64] (__lround): Do not define as function or alias.
[!_LP64] (lround): Likewise.
[!_LP64] (__lroundl): Likewise.
[!_LP64] (lroundl): Likewise.
* sysdeps/tile/sysdep.h (REGISTER_CAST_INT32_TO_INT64): Remove
macro.
* sysdeps/x86_64/x32/sysdep.h (REGISTER_CAST_INT32_TO_INT64):
Likewise.
GCC added support for -msse4 in version 4.3. Thus the configure tests
for it are obsolete, and this patch removes them.
Tested for x86_64 and x86 (testsuite, and that installed stripped
shared libraries are unchanged by this patch).
* sysdeps/i386/configure.ac (libc_cv_cc_sse4): Remove configure
test.
* sysdeps/i386/configure: Regenerated.
* sysdeps/i386/i686/multiarch/Makefile
[$(config-cflags-sse4) = yes]: Make code unconditional.
* sysdeps/i386/i686/multiarch/strcspn.S [HAVE_SSE4_SUPPORT]:
Likewise.
* sysdeps/i386/i686/multiarch/strspn.S [HAVE_SSE4_SUPPORT]:
Likewise.
* sysdeps/x86_64/configure.ac (libc_cv_cc_sse4): Remove configure
test.
* sysdeps/x86_64/configure: Regenerated.
* sysdeps/x86_64/multiarch/Makefile [$(config-cflags-sse4) = yes]:
Make code unconditional.
* sysdeps/x86_64/multiarch/strcspn.S [HAVE_SSE4_SUPPORT]:
Likewise.
* sysdeps/x86_64/multiarch/strspn.S [HAVE_SSE4_SUPPORT]: Likewise.
* config.h.in (HAVE_SSE4_SUPPORT): Remove #undef.
The ldbl-128ibm expl wrapper checks the argument to determine when to
call __kernel_standard_l, thereby overriding overflowing results from
__ieee754_expl that could otherwise (given appropriately patched
libgcc) be correct for the rounding mode. This patch changes it to
check the result of __ieee754_expl instead, as other versions of this
wrapper do.
Tested for powerpc.
[BZ #19078]
* sysdeps/ieee754/ldbl-128ibm/w_expl.c (o_thres): Remove variable.
(u_thres): Likewise.
(__expl): Determine whether to call __kernel_standard_l based on
value of result, not argument.
The ldbl-128ibm implementation of logl produces a zero with the wrong
sign for logl (1) in FE_DOWNWARD mode. This patch makes it explicitly
return 0.0L in that case, as in e.g. the ldbl-128 implementation.
Tested for powerpc.
[BZ #19077]
* sysdeps/ieee754/ldbl-128ibm/e_logl.c (__ieee754_logl): Return
0.0L for argument 1.0L.
The ldbl-128ibm implementation of log1pl produces an infinity with the
wrong sign for log1pl (-1) in FE_DOWNWARD mode. This patch fixes this
by changing a division (-1.0L / (x - x)) (incorrect in FE_DOWNWARD
mode) to (-1.0L / 0.0L) (correct in all rounding modes).
Tested for powerpc.
[BZ #19076]
* sysdeps/ieee754/ldbl-128ibm/s_log1pl.c (__log1pl): Divide by
constant 0.0L when computing infinite result.
The ldbl-96 version of lroundl is incorrect for systems with 64-bit
long when the argument's absolute value is just below a power of 2,
2^32 or more, and rounds up to the next integer; in such cases, it
returns 0. The problem is incrementing the high part of the mantissa
loses the high bit of the value (which is not an issue for any other
floating-point format, and is handled specially in lround when the bit
corresponding to 0.5 was in the high part rather than the low part).
This patch fixes this in a similar way to that used in llroundl:
storing the high part in an unsigned long variable before incrementing
it, so problems cannot occur in the case when this code is reachable.
I improved test coverage for both lround and llround by making them
use the same test inputs (appropriately conditioned on the size of
long in the lround case) - complete with the same comments, to make
comparison as easy as possible. (This test coverage improvement was
how I found the lroundl bug.)
Tested for x86_64 and x86.
[BZ #19071]
* sysdeps/ieee754/ldbl-96/s_lroundl.c (__lroundl): Use unsigned
long int variable to store possibly incremented high part of
mantissa.
* math/libm-test.inc (lround_test_data): Add tests used for
llround. Use [LONG_MAX > 0x7fffffff] consistently as condition
for tests requiring 64-bit long. Do not condition tests on
[TEST_FLOAT] unnecessarily.
(llround_test_data): Add tests used for lround. Add another
expectation for the "inexact" exception. Do not condition tests
on [TEST_FLOAT] unnecessarily.
On powerpc32 hard-float, older processors (ones where fcfid is not
available for 32-bit code), GCC generates conversions from integers to
floating point that wrongly convert integer 0 to -0 instead of +0 in
FE_DOWNWARD mode. This in turn results in logb and a few other
functions wrongly returning -0 when they should return +0.
This patch works around this issue in glibc as I proposed in
<https://sourceware.org/ml/libc-alpha/2015-09/msg00728.html>, so that
the affected functions can be correct and the affected tests pass in
the absence of a GCC fix for this longstanding issue (GCC bug 67771 -
if fixed, of course we can put in GCC version conditionals, and
eventually phase out the workarounds). A new macro
FIX_INT_FP_CONVERT_ZERO is added in a new sysdeps header
fix-int-fp-convert-zero.h, and the powerpc32/fpu version of that
header defines the macro based on the results of a configure test for
whether such conversions use the fcfid instruction.
Tested for x86_64 (that installed stripped shared libraries are
unchanged by the patch) and powerpc (that HAVE_PPC_FCFID comes out to
0 as expected and that the relevant tests are fixed). Also tested a
build with GCC configured for -mcpu=power4 and verified that
HAVE_PPC_FCFID comes out to 1 in that case.
There are still some other issues to fix to get test-float and
test-double passing cleanly for older powerpc32 processors (apart from
the need for an ulps regeneration for powerpc). (test-ldouble will be
harder to get passing cleanly, but with a combination of selected
fixes to ldbl-128ibm code that don't involve significant performance
issues, allowing spurious underflow and inexact exceptions for that
format, and lots of XFAILing for the default case of unpatched libgcc,
it should be doable.)
[BZ #887]
[BZ #19049]
[BZ #19050]
* sysdeps/generic/fix-int-fp-convert-zero.h: New file.
* sysdeps/ieee754/dbl-64/e_log10.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log10): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/e_log2.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log2): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/s_erf.c: Include
<fix-int-fp-convert-zero.h>.
(__erfc): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/s_logb.c: Include
<fix-int-fp-convert-zero.h>.
(__logb): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/e_log10f.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log10f): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/e_log2f.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log2f): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/s_erff.c: Include
<fix-int-fp-convert-zero.h>.
(__erfcf): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/s_logbf.c: Include
<fix-int-fp-convert-zero.h>.
(__logbf): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/ldbl-128ibm/s_erfl.c: Include
<fix-int-fp-convert-zero.h>.
(__erfcl): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/ldbl-128ibm/s_logbl.c: Include
<fix-int-fp-convert-zero.h>.
(__logbl): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/powerpc/powerpc32/fpu/configure.ac: New file.
* sysdeps/powerpc/powerpc32/fpu/configure: New generated file.
* sysdeps/powerpc/powerpc32/fpu/fix-int-fp-convert-zero.h: New
file.
* config.h.in [_LIBC] (HAVE_PPC_FCFID): New macro.
ISO C requires overflowing results from nexttoward to be the
appropriate infinity independent of the rounding mode, but some
implementations use a rounding-mode-dependent result (this is the same
issue as was fixed for nextafter in bug 16677). This patch fixes the
problem by making the nexttoward implementations discard the result
from the floating-point computation that forced an overflow exception
and then return the infinity previously computed with integer
arithmetic.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #19059]
* math/s_nexttowardf.c (__nexttowardf): Do not return value from
overflowing computation.
* sysdeps/i386/fpu/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/i386/fpu/s_nexttowardf.c (__nexttowardf): Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-opt/s_nexttowardfd.c (__nldbl_nexttowardf):
Likewise.
* math/libm-test.inc (nexttoward_test_data): Add more tests.
The file sysdeps/powerpc/sysdeps.h defines aliases for register operands,
which add the letter 'r' as a prefix to a register name. E.g.: register 20
can be written as 'r20', instead of '20'. On the one hand, this increases
readability, as it makes it easier for readers to know whether the operand is a
register or an immediate. On the other hand, this permits that immediate
operands be written as if they were registers, and vice-versa, thus reducing
the readability of the code.
This commit removes some of these unintentional misuses.
This commit also increases readability of the code by adding the prefix 'cr' to
some uses of the control register.
Both changes have no effect on the final code. Checked with objdump.
* sysdeps/powerpc/powerpc64/power8/strncpy.S: Remove or add register
prefix from operands.
The ldbl-128 / ldbl-128ibm implementation of lgamma has problems with
its handling of large arguments. It has an overflow threshold that is
correct only for ldbl-128, despite being used for both types - with
diagnostic control macros as a temporary measure to disable warnings
about that constant overflowing for ldbl-128ibm - and it has a
calculation that's roughly x * log(x) - x, resulting in overflows for
arguments that are roughly at most a factor 1/log(threshold) below the
overflow threshold.
This patch fixes both issues, using an overflow threshold appropriate
for the type in question and adding another case for large arguments
that avoids the possible intermediate overflow.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16347]
[BZ #19046]
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c: Do not include
<libc-internal.h>.
(MAXLGM): Do not use diagnostic control macros.
[LDBL_MANT_DIG == 106] (MAXLGM): Change value to overflow
threshold for ldbl-128ibm.
(__ieee754_lgammal_r): For large arguments, multiply by log - 1
instead of multiplying by log then subtracting.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
The ldbl-128ibm implementation of exp10l uses a version of log(10)
split into high and low parts - but the low part is negative, so
causing spurious overflows from __ieee754_expl (exp_high) in cases
close to the overflow threshold (I added relevant tests close to the
overflow threshold to the testsuite earlier today). The same issue
applies close to the underflow threshold as well (except that spurious
underflows in IBM long double arithmetic are harder to fix than the
other deficiencies, so we might end up permitting those for IBM long
double in the libm testsuite, as permitted by ISO C).
This patch fixes it to use a low part rounded downward to 48 bits
instead. (The choice of 48 instead of 53 bits is to make it more
obviously safe even when the low part of the argument is negative.)
Tested for powerpc. (Note that because of libgcc bugs with
multiplication very close to LDBL_MAX, libgcc also needs patching for
all the problem cases to be fixed, but this patch is still safe and
correct in the absence of such libgcc fixes.)
[BZ #16620]
* sysdeps/ieee754/ldbl-128ibm/e_exp10l.c (log10_high): Use value
of log (10) rounded downward to 48 bits.
(log10_low): Use corresponding low part of log (10).
The i386 versions of acoshf and acosh raise a spurious "invalid"
exception for an argument that is a quiet NaN with the sign bit set.
The integer arithmetic to detect arguments < 1 also detects -NaN, and
then the computation 0 / 0 in that case raises the exception. This
patch fixes this by using (x - x) / (x - x) as the computation in that
case instead, which will always raise the exception for non-NaN
arguments reaching that code, but not for quiet NaN arguments.
Tested for x86_64 and x86.
[BZ #19032]
* sysdeps/i386/fpu/e_acosh.S (__ieee754_acosh): For arguments < 1,
compute result as (x - x) / (x - x) not as 0 / 0.
* sysdeps/i386/fpu/e_acoshf.S (__ieee754_acoshf): Likewise.
* math/libm-test.inc (acosh_test_data): Add another test of acosh.
This patch improves test coverage of the real libm functions [a-e]*,
ensuring that special cases and ranges of input values of potential
significance (such as close to overflow and underflow thresholds) are
more systematically covered.
This is a followup to
<https://sourceware.org/ml/libc-alpha/2013-12/msg00757.html> which
covered [a-c]* (however, I found more weaknesses in the coverage of
those functions when preparing this patch, hence the additional tests
being added for them here).
Addition of a test for acosh (-qNaN) is temporarily deferred, to be
included as part of a fix for bug 19032 which was discovered in the
course of adding these tests (and which illustrates the use of testing
-qNaN as well as +qNaN as input even to functions for which the sign
of a NaN isn't meant to be significant).
Tested for x86_64 and x86.
* math/auto-libm-test-in: Add more tests of acos, acosh, asin,
atan, atan2, atanh, cbrt, cos, cosh, erf, erfc, exp, exp10, exp2
and expm1.
* math/auto-libm-test-out: Regenerated.
* math/libm-test.inc (acos_test_data): Add more tests.
(asin_test_data): Likewise.
(asinh_test_data): Likewise.
(atan_test_data): Likewise.
(atanh_test_data): Likewise.
(atan2_test_data): Likewise.
(cbrt_test_data): Likewise.
(ceil_test_data): Likewise.
(copysign_test_data): Likewise.
(cos_test_data): Likewise.
(cosh_test_data): Likewise.
(erf_test_data): Likewise.
(erfc_test_data): Likewise.
(exp_test_data): Likewise.
(exp10_test_data): Likewise.
(exp2_test_data): Likewise.
(expm1_test_data): Likewise.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
For arguments with X^2 + Y^2 close to 1, clog and clog10 avoid large
errors from log(hypot) by computing X^2 + Y^2 - 1 in a way that avoids
cancellation error and then using log1p.
However, the thresholds for using that approach still result in log
being used on argument as large as sqrt(13/16) > 0.9, leading to
significant errors, in some cases above the 9ulp maximum allowed in
glibc libm. This patch arranges for the approach using log1p to be
used in any cases where |X|, |Y| < 1 and X^2 + Y^2 >= 0.5 (with the
existing allowance for cases where one of X and Y is very small),
adjusting the __x2y2m1 functions to work with the wider range of
inputs. This way, log only gets used on arguments below sqrt(1/2) (or
substantially above 1), where the error involved is much less.
Tested for x86_64, x86, mips64 and powerpc. For the ulps regeneration
I removed the existing clog and clog10 ulps before regenerating to
allow any reduced ulps to appear. Tests added include those found by
random test generation to produce large ulps either before or after
the patch, and some found by trying inputs close to the (0.75, 0.5)
threshold where the potential errors from using log are largest.
[BZ #19016]
* sysdeps/generic/math_private.h (__x2y2m1f): Update comment to
allow more cases with X^2 + Y^2 >= 0.5.
* sysdeps/ieee754/dbl-64/x2y2m1.c (__x2y2m1): Likewise. Add -1 as
normal element in sum instead of special-casing based on values of
arguments.
* sysdeps/ieee754/dbl-64/x2y2m1f.c (__x2y2m1f): Update comment.
* sysdeps/ieee754/ldbl-128/x2y2m1l.c (__x2y2m1l): Likewise. Add
-1 as normal element in sum instead of special-casing based on
values of arguments.
* sysdeps/ieee754/ldbl-128ibm/x2y2m1l.c (__x2y2m1l): Likewise.
* sysdeps/ieee754/ldbl-96/x2y2m1.c [FLT_EVAL_METHOD != 0]
(__x2y2m1): Update comment.
* sysdeps/ieee754/ldbl-96/x2y2m1l.c (__x2y2m1l): Likewise. Add -1
as normal element in sum instead of special-casing based on values
of arguments.
* math/s_clog.c (__clog): Handle more cases using log1p without
hypot.
* math/s_clog10.c (__clog10): Likewise.
* math/s_clog10f.c (__clog10f): Likewise.
* math/s_clog10l.c (__clog10l): Likewise.
* math/s_clogf.c (__clogf): Likewise.
* math/s_clogl.c (__clogl): Likewise.
* math/auto-libm-test-in: Add more tests of clog and clog10.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
The flt-32 version of powf can be inaccurate because of bugs in the
extra-precision calculation of (x-1)/(x+1) or (x-1.5)/(x+1.5) as part
of calculating log(x) with extra precision: a constant used (as part
of adding 1 or 1.5 through integer arithmetic) is incorrect, and then
the code fails to mask a computed high part before using it in
arithmetic that relies on s_h*t_h being exactly representable. This
patch fixes these bugs.
Tested for x86_64 and x86. x86_64 ulps for powf removed and
regenerated to reflect reduced ulps from the increased accuracy for
existing tests.
[BZ #18956]
* sysdeps/ieee754/flt-32/e_powf.c (__ieee754_powf): Add 0x00400000
not 0x0040000 for high bit of mantissa. Mask with 0xfffff000 when
extracting high part.
* math/auto-libm-test-in: Add another test of pow.
* math/auto-libm-test-out: Regenerated.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
Similar to various other bugs in this area, pow functions can fail to
raise the underflow exception when the result is tiny and inexact but
one or more low bits of the intermediate result that is scaled down
(or, in the i386 case, converted from a wider evaluation format) are
zero. This patch forces the exception in a similar way to previous
fixes, thereby concluding the fixes for known bugs with missing
underflow exceptions currently filed in Bugzilla.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18825]
* sysdeps/i386/fpu/i386-math-asm.h (FLT_NARROW_EVAL_UFLOW_NONNAN):
New macro.
(DBL_NARROW_EVAL_UFLOW_NONNAN): Likewise.
(LDBL_CHECK_FORCE_UFLOW_NONNAN): Likewise.
* sysdeps/i386/fpu/e_pow.S: Use DEFINE_DBL_MIN.
(__ieee754_pow): Use DBL_NARROW_EVAL_UFLOW_NONNAN instead of
DBL_NARROW_EVAL, reloading the PIC register as needed.
* sysdeps/i386/fpu/e_powf.S: Use DEFINE_FLT_MIN.
(__ieee754_powf): Use FLT_NARROW_EVAL_UFLOW_NONNAN instead of
FLT_NARROW_EVAL. Use separate return path for case when first
argument is NaN.
* sysdeps/i386/fpu/e_powl.S: Include <i386-math-asm.h>. Use
DEFINE_LDBL_MIN.
(__ieee754_powl): Use LDBL_CHECK_FORCE_UFLOW_NONNAN, reloading the
PIC register.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Use
math_check_force_underflow_nonneg.
* sysdeps/ieee754/flt-32/e_powf.c (__ieee754_powf): Force
underflow for subnormal result.
* sysdeps/ieee754/ldbl-128/e_powl.c (__ieee754_powl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_powl.c (__ieee754_powl): Use
math_check_force_underflow_nonneg.
* sysdeps/x86/fpu/powl_helper.c (__powl_helper): Use
math_check_force_underflow.
* sysdeps/x86_64/fpu/x86_64-math-asm.h
(LDBL_CHECK_FORCE_UFLOW_NONNAN): New macro.
* sysdeps/x86_64/fpu/e_powl.S: Include <x86_64-math-asm.h>. Use
DEFINE_LDBL_MIN.
(__ieee754_powl): Use LDBL_CHECK_FORCE_UFLOW_NONNAN.
* math/auto-libm-test-in: Add more tests of pow.
* math/auto-libm-test-out: Regenerated.
Systems without floating-point exceptions and rounding modes should
use the soft-fp versions of fmaf and fma, not the sysdeps/ieee754
versions that rely on setting rounding to zero and testing for the
"inexact" exception; this has been noted on
<https://sourceware.org/glibc/wiki/PortStatus> for some time. This
patch makes no-FPU ColdFire use the soft-fp files; sfp-machine.h is
made to include the nios2 version of sfp-machine.h which seems
sufficiently generic for 32-bit systems.
[BZ #13304]
* sysdeps/m68k/coldfire/nofpu/s_fma.c: New file.
* sysdeps/m68k/coldfire/nofpu/s_fmaf.c: Likewise.
* sysdeps/m68k/coldfire/nofpu/sfp-machine.h: Likewise.
Systems without floating-point exceptions and rounding modes should
use the soft-fp versions of fmaf and fma, not the sysdeps/ieee754
versions that rely on setting rounding to zero and testing for the
"inexact" exception; this has been noted on
<https://sourceware.org/glibc/wiki/PortStatus> for some time. This
patch makes MicroBlaze use the soft-fp files; sfp-machine.h is made to
include the nios2 version of sfp-machine.h which seems sufficiently
generic for 32-bit systems.
[BZ #13304]
* sysdeps/microblaze/s_fma.c: New file.
* sysdeps/microblaze/s_fmaf.c: Likewise.
* sysdeps/microblaze/sfp-machine.h: Likewise.
Similar to various other bugs in this area, hypot functions can fail
to raise the underflow exception when the result is tiny and inexact
but one or more low bits of the intermediate result that is scaled
down (or, in the i386 case, converted from a wider evaluation format)
are zero. This patch forces the exception in a similar way to
previous fixes.
Note that this issue cannot arise for implementations of hypotf using
double (or wider) for intermediate evaluation (if hypotf should
underflow, that means the double square root is being computed of some
number of the form N*2^-298, for 0 < N < 2^46, which is exactly
represented as a double, and whatever the rounding mode such a square
root cannot have a mantissa with all zeroes after the initial 23
bits). Thus no changes are made to hypotf implementations in this
patch, only to hypot and hypotl.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18803]
* sysdeps/i386/fpu/e_hypot.S: Use DEFINE_DBL_MIN.
(MO): New macro.
(__ieee754_hypot) [PIC]: Load PIC register.
(__ieee754_hypot): Use DBL_NARROW_EVAL_UFLOW_NONNEG instead of
DBL_NARROW_EVAL.
* sysdeps/ieee754/dbl-64/e_hypot.c (__ieee754_hypot): Use
math_check_force_underflow_nonneg in case where result might be
tiny.
* sysdeps/ieee754/ldbl-128/e_hypotl.c (__ieee754_hypotl):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_hypotl.c (__ieee754_hypotl):
Likewise.
* sysdeps/ieee754/ldbl-96/e_hypotl.c (__ieee754_hypotl): Likewise.
* sysdeps/powerpc/fpu/e_hypot.c (__ieee754_hypot): Likewise.
* math/auto-libm-test-in: Add more tests of hypot.
* math/auto-libm-test-out: Regenerated.
This patch refactors code in sysdeps/x86_64/fpu that forces underflow
exceptions and closely follows corresponding i386 code to use common
macros in x86_64-math-asm.h for that purpose. This is mainly about
keeping the code similar to the i386 code as far as possible, since
each macro apart from DEFINE_LDBL_MIN ends up used only once. It
would be possible to do a further refactoring to share these macros
between i386 and x86_64 (with i386 using the fcomip / fucomip versions
when building for i686 and above), but I have no immediate plans to do
so.
Tested for x86_64.
* sysdeps/x86_64/fpu/x86_64-math-asm.h: New file.
* sysdeps/x86_64/fpu/e_exp2l.S: Include <x86_64-math-asm.h>.
(ldbl_min): Replace with use of DEFINE_LDBL_MIN.
(__ieee754_exp2l): Use LDBL_CHECK_FORCE_UFLOW_NONNEG_NAN.
* sysdeps/x86_64/fpu/e_expl.S: Include <x86_64-math-asm.h>.
[!USE_AS_EXPM1L] (cmin): Replace with use of DEFINE_LDBL_MIN.
(IEEE754_EXPL): Use LDBL_CHECK_FORCE_UFLOW_NONNEG.
sysdeps/i386/fpu/e_atanh.S, unlike all other functions in that
directory, loads the PIC register with its own code using
_GLOBAL_OFFSET_TABLE_, rather than with the LOAD_PIC_REG macro. I see
no good reason for the difference; this patch makes it use the common
macro.
Tested for x86.
* sysdeps/i386/fpu/e_atanh.S (__ieee754_atanh) [PIC]: Use
LOAD_PIC_REG.
This patch refactors code in sysdeps/i386/fpu that forces underflow
exceptions to use common macros for that purpose as far as possible.
(Although some of the macros end up used in only one place, I think
it's cleanest to define all these macros together so that all the code
forcing underflow uses such macros. Some more uses of such macros
will also be introduced when fixing remaining bugs about missing
underflow exceptions, and it would be possible to do further
refactoring of the macros in i386-math-asm.h to share more code by
using other macros internally. Places that test for underflow by
examining the representation of the argument with integer operations,
rather that using floating-point comparisons on the argument or
result, are unchanged by this patch.)
Most of this code uses a macro MO to abstract away the differences
between PIC and non-PIC memory references to constants. log1p
functions, however, hardcoded PIC conditionals for this. Because the
common macros rely on the use of MO, I changed the log1p functions to
use the normal style here, and, for consistency, also made that change
to log1pl which is otherwise unaffected by this patch.
Tested for x86.
* sysdeps/i386/fpu/i386-math-asm.h (DEFINE_LDBL_MIN): New macro.
(FLT_CHECK_FORCE_UFLOW): Likewise.
(DBL_CHECK_FORCE_UFLOW): Likewise.
(FLT_CHECK_FORCE_UFLOW_NARROW): Likewise.
(DBL_CHECK_FORCE_UFLOW_NARROW): Likewise.
(LDBL_CHECK_FORCE_UFLOW_NONNEG_NAN): Likewise.
(FLT_CHECK_FORCE_UFLOW_NONNAN): Likewise.
(DBL_CHECK_FORCE_UFLOW_NONNAN): Likewise.
(FLT_CHECK_FORCE_UFLOW_NONNEG): Likewise.
(DBL_CHECK_FORCE_UFLOW_NONNEG): Likewise.
(LDBL_CHECK_FORCE_UFLOW_NONNEG): Likewise.
* sysdeps/i386/fpu/e_asin.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_asin): Use DBL_CHECK_FORCE_UFLOW.
* sysdeps/i386/fpu/e_asinf.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_asinf): Use FLT_CHECK_FORCE_UFLOW.
* sysdeps/i386/fpu/e_atan2.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_atan2): Use DBL_CHECK_FORCE_UFLOW_NARROW.
* sysdeps/i386/fpu/e_atan2f.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_atan2f): Use FLT_CHECK_FORCE_UFLOW_NARROW.
* sysdeps/i386/fpu/e_atanh.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_atanh): Use DBL_CHECK_FORCE_UFLOW_NONNEG.
* sysdeps/i386/fpu/e_atanhf.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_atanhf): Use FLT_CHECK_FORCE_UFLOW_NONNEG.
* sysdeps/i386/fpu/e_exp2l.S: Include <i386-math-asm.h>.
(ldbl_min): Replace with use of DEFINE_LDBL_MIN.
(__ieee754_exp2l): Use LDBL_CHECK_FORCE_UFLOW_NONNEG_NAN.
* sysdeps/i386/fpu/e_expl.S: Include <i386-math-asm.h>.
[!USE_AS_EXPM1L] (cmin): Replace with use of DEFINE_LDBL_MIN.
(IEEE754_EXPL): Use LDBL_CHECK_FORCE_UFLOW_NONNEG.
* sysdeps/i386/fpu/s_atan.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__atan): Use DBL_CHECK_FORCE_UFLOW.
* sysdeps/i386/fpu/s_atanf.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__atanf): Use FLT_CHECK_FORCE_UFLOW.
* sysdeps/i386/fpu/s_expm1.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__expm1): Use DBL_CHECK_FORCE_UFLOW. Move underflow check after
main computation.
* sysdeps/i386/fpu/s_expm1f.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__expm1f): Use FLT_CHECK_FORCE_UFLOW. Move underflow check after
main computation.
* sysdeps/i386/fpu/s_log1p.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(MO): New macro.
(__log1p): Use MO. Use DBL_CHECK_FORCE_UFLOW_NONNAN.
* sysdeps/i386/fpu/s_log1pf.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(MO): New macro.
(__log1pf): Use MO. Use FLT_CHECK_FORCE_UFLOW_NONNAN.
* sysdeps/i386/fpu/s_log1pl.S (MO): New macro.
(__log1pl): Use MO.
The x86_64 fma4 version of pow fails to disable contraction of
operations other than those explicitly intended to use fma
instructions, so resulting in large ulps errors on processors with
fma4 instructions, as in bug 18104 (165ulp for the test added for that
bug; error originally reported by "blaaa" on #glibc). This patch adds
$(config-cflags-nofma) for e_pow-fma4.c, corresponding to the use for
e_pow.c in sysdeps/ieee754/dbl-64/Makefile.
Tested for x86_64 on a processor with fma4.
[BZ #19003]
* sysdeps/x86_64/fpu/multiarch/Makefile (CFLAGS-e_pow-fma4.c): Add
$(config-cflags-nofma).
sysdeps/ieee754/flt-32/e_exp2f.c declares two variable as "static
const volatile float". Maybe this use of "volatile" was originally
intended to inhibit optimization of underflowing / overflowing
operations such as TWOM100 * TWOM100; in any case, it's not currently
needed, as given -frounding-math constant folding of such expressions
is properly disabled when it would be unsafe. This patch removes the
unnecessary use of "volatile".
Tested for x86_64.
* sysdeps/ieee754/flt-32/e_exp2f.c (TWOM100): Remove volatile.
(TWO127): Likewise.
Where glibc code needs to avoid excess range and precision in
floating-point arithmetic, code variously uses either asms or volatile
to force the results of that arithmetic to memory; mostly this is
conditional on FLT_EVAL_METHOD, but in the case of lrint / llrint
functions some use of volatile is unconditional (and is present
unnecessarily in versions for long double). This patch make such code
use the recently-added math_narrow_eval macro consistently, removing
the unnecessary uses of volatile in long double lrint / llrint
implementations completely.
Tested for x86_64, x86, mips64 and powerpc.
* math/s_nexttowardf.c (__nexttowardf): Use math_narrow_eval.
* stdlib/strtod_l.c: Include <math_private.h>.
(overflow_value): Use math_narrow_eval.
(underflow_value): Likewise.
* sysdeps/i386/fpu/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/i386/fpu/s_nexttowardf.c (__nexttowardf): Likewise.
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Likewise.
(__ieee754_gamma_r): Likewise.
* sysdeps/ieee754/dbl-64/gamma_productf.c (__gamma_productf):
Likewise.
* sysdeps/ieee754/dbl-64/k_rem_pio2.c (__kernel_rem_pio2):
Likewise.
* sysdeps/ieee754/dbl-64/lgamma_neg.c (__lgamma_neg): Likewise.
* sysdeps/ieee754/dbl-64/s_erf.c (__erfc): Likewise.
* sysdeps/ieee754/dbl-64/s_llrint.c (__llrint): Likewise.
* sysdeps/ieee754/dbl-64/s_lrint.c (__lrint): Likewise.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Likewise.
(__ieee754_gammaf_r): Likewise.
* sysdeps/ieee754/flt-32/k_rem_pio2f.c (__kernel_rem_pio2f):
Likewise.
* sysdeps/ieee754/flt-32/lgamma_negf.c (__lgamma_negf): Likewise.
* sysdeps/ieee754/flt-32/s_erff.c (__erfcf): Likewise.
* sysdeps/ieee754/flt-32/s_llrintf.c (__llrintf): Likewise.
* sysdeps/ieee754/flt-32/s_lrintf.c (__lrintf): Likewise.
* sysdeps/ieee754/ldbl-128/s_llrintl.c (__llrintl): Do not use
volatile.
* sysdeps/ieee754/ldbl-128/s_lrintl.c (__lrintl): Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttoward.c (__nexttoward): Use
math_narrow_eval.
* sysdeps/ieee754/ldbl-128ibm/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-96/gamma_product.c (__gamma_product):
Likewise.
* sysdeps/ieee754/ldbl-96/s_llrintl.c (__llrintl): Do not use
volatile.
* sysdeps/ieee754/ldbl-96/s_lrintl.c (__lrintl): Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttoward.c (__nexttoward): Use
math_narrow_eval.
* sysdeps/ieee754/ldbl-96/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-opt/s_nexttowardfd.c (__nldbl_nexttowardf):
Likewise.
i386 exp, hypot and pow functions can return overflowing and
underflowing values with excess range and precision; ; Wilco
Dijkstra's patches to make isfinite etc. expand inline cause this
pre-existing issue to result in test failures.
This patch fixes those functions to avoid excess range and precision
in their return values. Appropriate macros are added for the repeated
code sequences; in future I'll add more such macros and refactor
existing code forcing underflow (with or without also eliminating
excess range and precision from the return value) to use such macros.
Tested for x86. If, after this patch, you still see x86 libm test
failures with excess range or precision, please file bugs in Bugzilla.
[BZ #18980]
* sysdeps/i386/fpu/i386-math-asm.h (DEFINE_FLT_MIN): New macro.
(DEFINE_DBL_MIN): Likewise.
(FLT_NARROW_EVAL_UFLOW_NONNEG_NAN): Likewise.
(DBL_NARROW_EVAL_UFLOW_NONNEG_NAN): Likewise.
(FLT_NARROW_EVAL_UFLOW_NONNEG): Likewise.
(DBL_NARROW_EVAL_UFLOW_NONNEG): Likewise.
* sysdeps/i386/fpu/e_exp.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_exp): Use DBL_NARROW_EVAL_UFLOW_NONNEG_NAN.
(__exp_finite): Use DBL_NARROW_EVAL_UFLOW_NONNEG.
* sysdeps/i386/fpu/e_exp10.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_exp10): Use DBL_NARROW_EVAL_UFLOW_NONNEG_NAN.
* sysdeps/i386/fpu/e_exp10f.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_exp10f): Use FLT_NARROW_EVAL_UFLOW_NONNEG_NAN.
* sysdeps/i386/fpu/e_exp2.S: Include <i386-math-asm.h>.
(dbl_min): Replace with use of DEFINE_DBL_MIN.
(__ieee754_exp2): Use DBL_NARROW_EVAL_UFLOW_NONNEG_NAN.
* sysdeps/i386/fpu/e_exp2f.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_exp2f): Use FLT_NARROW_EVAL_UFLOW_NONNEG_NAN.
* sysdeps/i386/fpu/e_expf.S: Include <i386-math-asm.h>.
(flt_min): Replace with use of DEFINE_FLT_MIN.
(__ieee754_expf): Use FLT_NARROW_EVAL_UFLOW_NONNEG_NAN.
(__expf_finite): Use FLT_NARROW_EVAL_UFLOW_NONNEG.
* sysdeps/i386/fpu/e_hypot.S: Include <i386-math-asm.h>.
(__ieee754_hypot): Use DBL_NARROW_EVAL.
* sysdeps/i386/fpu/e_hypotf.S: Include <i386-math-asm.h>.
(__ieee754_hypotf): Use FLT_NARROW_EVAL.
* sysdeps/i386/fpu/e_pow.S: Include <i386-math-asm.h>.
(__ieee754_pow): Use DBL_NARROW_EVAL.
* sysdeps/i386/fpu/e_powf.S: Include <i386-math-asm.h>.
(__ieee754_powf): Use FLT_NARROW_EVAL.
* sysdeps/i386/i686/fpu/multiarch/e_expf-sse2.S
(__ieee754_expf_sse2): Convert double-precision result to single
precision.
* sysdeps/i386/fpu/libm-test-ulps: Update.
i386 scalb / scalbn / scalbln (and thus ldexp) functions for float and
double can return results with excess range (and consequently excess
precision for subnormal results). As the results of these functions
are fully determined by reference to IEEE 754 operations, this is
unambiguously a bug, apart from the testsuite failures it causes.
This patch makes those functions store their results on the stack and
load them back to eliminate the excess range. Double rounding is not
a problem, as the only cases where it could occur are when the result
overflows or underflows for extended precision, and then the
double-rounded results are the same as the single-rounded results.
The new macros will be used for more functions, more such macros
added, and existing code refactored to use such macros, in subsequent
patches.
Tested for x86. Committed.
[BZ #18981]
* sysdeps/i386/fpu/i386-math-asm.h: New file.
* sysdeps/i386/fpu/e_scalb.S: Include <i386-math-asm.h>.
(__ieee754_scalb): Use DBL_NARROW_EVAL.
* sysdeps/i386/fpu/e_scalbf.S: Include <i386-math-asm.h>.
(__ieee754_scalbf): Use FLT_NARROW_EVAL.
* sysdeps/i386/fpu/s_scalbn.S: Include <i386-math-asm.h>.
(__scalbn): Use DBL_NARROW_EVAL.
* sysdeps/i386/fpu/s_scalbnf.S: Include <i386-math-asm.h>.
(__scalbnf): Use FLT_NARROW_EVAL.
Various i386 libm functions return values with excess range and
precision; Wilco Dijkstra's patches to make isfinite etc. expand
inline cause this pre-existing issue to result in test failures (when
e.g. a result that overflows float but not long double gets counted as
overflowing for some purposes but not others).
This patch addresses those cases arising from functions defined in C,
adding a math_narrow_eval macro that forces values to memory to
eliminate excess precision if FLT_EVAL_METHOD indicates this is
needed, and is a no-op otherwise. I'll convert existing uses of
volatile and asm for this purpose to use the new macro later, once
i386 has clean test results again (which requires fixes for .S files
as well).
Tested for x86_64 and x86. Committed.
[BZ #18980]
* sysdeps/generic/math_private.h: Include <float.h>.
(math_narrow_eval): New macro.
[FLT_EVAL_METHOD != 0] (excess_precision): Likewise.
* sysdeps/ieee754/dbl-64/e_cosh.c (__ieee754_cosh): Use
math_narrow_eval on overflowing return value.
* sysdeps/ieee754/dbl-64/e_lgamma_r.c (__ieee754_lgamma_r):
Likewise.
* sysdeps/ieee754/dbl-64/e_sinh.c (__ieee754_sinh): Likewise.
* sysdeps/ieee754/flt-32/e_coshf.c (__ieee754_coshf): Likewise.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (__ieee754_lgammaf_r):
Likewise.
* sysdeps/ieee754/flt-32/e_sinhf.c (__ieee754_sinhf): Likewise.
The logic in setjmp/__longjmp incorrectly uses "PIC" to figure out
whether the code is going into a shared library when it should be
using "SHARED". If you build glibc with a gcc version that has PIE
enabled by default, then the code will try to use symbols that are
only in the shared library.
URL: https://bugs.gentoo.org/336914
Since we require a new enough kernel all the time, the __ASSUME_FDATASYNC
define has been hardcoded to 1. That means we can delete the alpha file
for fdatasync now and rely on the syscalls list like other ports.
Bug 15384 notes that in __finite, two different constants are used
that could be the same constant (the result only depends on the
exponent of the floating-point representation), and that using the
same constant is better for architectures where constants need loading
from a constant pool. This patch implements that change.
Tested for x86_64, mips64 and powerpc.
[BZ #15384]
* sysdeps/ieee754/dbl-64/s_finite.c (FINITE): Use same constant as
bit-mask as in subtraction.
* sysdeps/ieee754/dbl-64/wordsize-64/s_finite.c (__finite):
Likewise.
* sysdeps/ieee754/flt-32/s_finitef.c (FINITEF): Likewise.
* sysdeps/ieee754/ldbl-128/s_finitel.c (__finitel): Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_finitel.c (__finitel): Likewise.
Similar to various other bugs in this area, tgamma functions can fail
to raise the underflow exception when the result is tiny and inexact
but one or more low bits of the intermediate result that is scaled
down are zero. This patch forces the exception in a similar way to
previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18951]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (__ieee754_gamma_r): Force
underflow exception for small results.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (__ieee754_gammaf_r):
Likewise.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* math/auto-libm-test-in: Add more tests of tgamma.
* math/auto-libm-test-out: Regenerated.
As noted in bug 6803, scalbn fails to set errno on overflow and
underflow. This patch fixes this by making scalbn an alias of ldexp,
which has exactly the same semantics (for floating-point types with
radix 2) and already has wrappers that deal with setting errno,
instead of an alias of the internal __scalbn (which ldexp calls).
Notes:
* Where compat symbols were defined for scalbn functions, I didn't
change what they point to (to keep the patch minimal), so such
compat symbols continue to go directly to the non-errno-setting
functions.
* Mike, I didn't do anything with the IA64 versions of these
functions, where I think both the ldexp and scalbn functions already
deal with setting errno. As a cleanup (not needed to fix this bug)
however you might want to make those functions into aliases for
IA64; there is no need for them to be separate function
implementations at all.
* This concludes the fix for bug 6803 since the scalb and scalbln
cases of that bug were fixed some time ago.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #6803]
* math/s_ldexp.c (scalbn): Define as weak alias of __ldexp.
[NO_LONG_DOUBLE] (scalbnl): Define as weak alias of __ldexp.
* math/s_ldexpf.c (scalbnf): Define as weak alias of __ldexpf.
* math/s_ldexpl.c (scalbnl): Define as weak alias of __ldexpl.
* sysdeps/i386/fpu/s_scalbn.S (scalbn): Remove alias.
* sysdeps/i386/fpu/s_scalbnf.S (scalbnf): Likewise.
* sysdeps/i386/fpu/s_scalbnl.S (scalbnl): Likewise.
* sysdeps/ieee754/dbl-64/s_scalbn.c (scalbn): Likewise.
[NO_LONG_DOUBLE] (scalbnl): Likewise.
* sysdeps/ieee754/dbl-64/wordsize-64/s_scalbn.c (scalbn):
Likewise.
[NO_LONG_DOUBLE] (scalbnl): Likewise.
* sysdeps/ieee754/flt-32/s_scalbnf.c (scalbnf): Likewise.
* sysdeps/ieee754/ldbl-128/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_scalbnl.c (scalbnl): Remove
long_double_symbol calls.
* sysdeps/ieee754/ldbl-64-128/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/ieee754/ldbl-opt/s_ldexpl.c (__ldexpl_2): Define as
strong alias of __ldexpl.
(scalbnl): Define using long_double_symbol.
* sysdeps/m68k/m680x0/fpu/s_scalbn.c (__CONCATX(scalbn,suffix)):
Remove alias.
* sysdeps/sparc/sparc64/soft-fp/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/x86_64/fpu/s_scalbnl.S (scalbnl): Likewise.
* math/libm-test.inc (scalbn_test_data): Add errno expectations.
(scalbln_test_data): Add more errno expectations.
This way we do not
need to call the kernel just to get the port. Furthermore, we no
longer increase the reference count on every invocation of
`mach_host_self'.
* mach/mach/mach_traps.h (__mach_host_self, mach_host_self):
Protect declarations against the macro expansion.
* mach/mach_init.c (__mach_host_self_): New variable.
(mach_init): Initialize `__mach_host_self_'.
* mach/mach_init.h (__mach_host_self_): New declaration.
(__mach_host_self, mach_host_self): New macros.
* sysdeps/mach/hurd/dl-sysdep.c (_dl_sysdep_start_cleanup):
Release reference.
The ldbl-128 and ldbl-128ibm expm1l implementations have code to
handle +Inf and finite arguments above an overflow threshold. Since
they now use __expl for large positive arguments to fix other
problems, this code is unreachable; this patch removes it.
Tested for mips64 and powerpc.
[BZ #16415]
* sysdeps/ieee754/ldbl-128/s_expm1l.c (maxlog): Remove variable.
(__expm1l): Remove code to handle positive infinity and overflow.
* sysdeps/ieee754/ldbl-128ibm/s_expm1l.c (maxlog): Remove
variable.
(__expm1l): Remove code to handle positive infinity and overflow.
The ldbl-128ibm implementation of nearbyintl wrongly uses signaling
comparisons such as "if (fabs (u.d[0].d) < TWO52)" on arguments that
might be NaNs, when "invalid" exceptions should not be raised. (For
hard float, this issue may be hidden by
<https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58684>, powerpc GCC
wrongly only using unordered comparison instructions.) This patch
fixes this by just returning the argument if it is not finite (because
of the arbitrary value of the low part of a NaN in IBM long double,
there are quite a lot of comparisons that could end up involving a NaN
when the argument to nearbyintl is a NaN, so excluding NaN arguments
at the start is the simplest and safest fix).
Tested for powerpc-nofpu, where it removes failures for spurious
"invalid" exceptions from nearbyintl.
[BZ #18857]
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c (__nearbyintl): Just
return non-finite argument without doing ordered comparisons on
it.
Bug 15918 points out that the handling of infinities in hypotf can be
simplified: it's enough to return the absolute value of the infinite
argument without first comparing it to the other argument and possibly
returning that other argument's absolute value. This patch makes that
cleanup (which should not change how hypotf behaves on any input).
Tested for x86_64.
[BZ #15918]
* sysdeps/ieee754/flt-32/e_hypotf.c (__ieee754_hypotf): Simplify
handling of cases where one argument is an infinity.
On i386, the double version of exp10 can miss underflow exceptions if
the result is in the subnormal range for double but the last 11 bits
of the 64-bit extended-precision mantissa happen to be zero. This
patch forces the exception in a similar way to previous fixes.
As with the exp2 and exp fixes, the exp10f changes may in fact not be
needed to ensure underflow exceptions, but are included for
consistency and to fix the exp10 part of bug 18875 by ensuring that
excess range and precision is removed from underflowing return values.
Tested for x86_64 and x86.
[BZ #18875]
[BZ #18966]
* sysdeps/i386/fpu/e_exp10.S (dbl_min): New object.
(MO): New macro.
(__ieee754_exp10): For small results, force underflow exception
and remove excess range and precision from return value.
* sysdeps/i386/fpu/e_exp10f.S (flt_min): New object.
(MO): New macro.
(__ieee754_exp10f): For small results, force underflow exception
and remove excess range and precision from return value.
* math/auto-libm-test-in: Add more tests of exp10.
* math/auto-libm-test-out: Regenerated.
On i386, the double version of exp can miss underflow exceptions if
the result is in the subnormal range for double but the last 11 bits
of the 64-bit extended-precision mantissa happen to be zero. This
patch forces the exception in a similar way to previous fixes.
As with the exp2 fixes, the expf changes may in fact not be needed to
ensure underflow exceptions, but are included for consistency and to
fix the exp part of bug 18875 by ensuring that excess range and
precision is removed from underflowing return values.
Tested for x86_64 and x86.
[BZ #18875]
[BZ #18961]
* sysdeps/i386/fpu/e_exp.S (dbl_min): New object.
(MO): New macro.
(__ieee754_exp): For small results, force underflow exception and
remove excess range and precision from return value.
(__exp_finite): Likewise.
* sysdeps/i386/fpu/e_expf.S (flt_min): New object.
(MO): New macro.
(__ieee754_expf): For small results, force underflow exception and
remove excess range and precision from return value.
(__expf_finite): Likewise.
* math/auto-libm-test-in: Add more tests of exp.
* math/auto-libm-test-out: Regenerated.
Various exp2 implementations in glibc can miss underflow exceptions
when the scaling down part of the calculation is exact (or, in the x86
case, when the conversion from extended precision to the target
precision is exact). This patch forces the exception in a similar way
to previous fixes.
The x86 exp2f changes may in fact not be needed for this purpose -
it's likely to be the case that no argument of type float has an exp2
result so close to an exact subnormal float value that it equals that
value when rounded to 64 bits (even taking account of variation
between different x86 implementations). However, they are included
for consistency with the changes to exp2 and so as to fix the exp2f
part of bug 18875 by ensuring that excess range and precision is
removed from underflowing return values.
Tested for x86_64, x86 and mips64.
[BZ #16521]
[BZ #18875]
* math/e_exp2l.c (__ieee754_exp2l): Force underflow exception for
small results.
* sysdeps/i386/fpu/e_exp2.S (dbl_min): New object.
(MO): New macro.
(__ieee754_exp2): For small results, force underflow exception and
remove excess range and precision from return value.
* sysdeps/i386/fpu/e_exp2f.S (flt_min): New object.
(MO): New macro.
(__ieee754_exp2f): For small results, force underflow exception
and remove excess range and precision from return value.
* sysdeps/i386/fpu/e_exp2l.S (ldbl_min): New object.
(MO): New macro.
(__ieee754_exp2l): Force underflow exception for small results.
* sysdeps/ieee754/dbl-64/e_exp2.c (__ieee754_exp2): Likewise.
* sysdeps/ieee754/flt-32/e_exp2f.c (__ieee754_exp2f): Likewise.
* sysdeps/x86_64/fpu/e_exp2l.S (ldbl_min): New object.
(MO): New macro.
(__ieee754_exp2l): Force underflow exception for small results.
* math/auto-libm-test-in: Add more tests or exp2.
* math/auto-libm-test-out: Regenerated.
Profiling git's test suite, Linus noted [1] that a disproportionately
large amount of time was spent reading /proc/meminfo. This is done by
the glibc functions get_phys_pages and get_avphys_pages, but they only
need the MemTotal and MemFree fields, respectively. That same
information can be obtained with a single syscall, sysinfo, instead of
six: open, fstat, mmap, read, close, munmap. While sysinfo also
provides more than necessary, it does a lot less work than what the
kernel needs to do to provide the entire /proc/meminfo. Both strace -T
and in-app microbenchmarks shows that the sysinfo() approach is
roughly an order of magnitude faster.
sysinfo() is much older than what glibc currently requires, so I don't
think there's any reason to keep the old parsing code. Moreover, this
makes get_[av]phys_pages work even in the absence of /proc.
Linus noted that something as simple as 'bash -c "echo"' would trigger
the reading of /proc/meminfo, but gdb says that many more applications
than just bash are affected:
Starting program: /bin/bash "-c" "echo"
Breakpoint 1, __get_phys_pages () at ../sysdeps/unix/sysv/linux/getsysstats.c:283
283 ../sysdeps/unix/sysv/linux/getsysstats.c: No such file or directory.
(gdb) bt
So it seems that any application that uses qsort on a moderately sized
array will incur this cost (once), which is obviously proportionately
more expensive for lots of short-lived processes (such as the git test
suite).
[1] http://thread.gmane.org/gmane.linux.kernel/2019285
Signed-off-by: Rasmus Villemoes <rv@rasmusvillemoes.dk>
* sysdeps/unix/sysv/linux/getsysstats.c (__get_phys_pages):
Use sysinfo system call instead of parsing /proc/meminfo.
* sysdeps/unix/sysv/linux/getsysstats.c (__get_avphys_pages):
Likewise.
This patch adds more libm test inputs found through random test
generation to increase previously known ulps. This particular test
generation was run for mips64, so most of the increased ulps are for
ldbl-128 (float and double having been fairly well covered by such
testing for x86_64), but there's the odd ulps increase for other
formats.
Tested for x86_64, x86 and mips64.
* math/auto-libm-test-in: Add more tests of acos, acosh, asin,
asinh, atan, atan2, atanh, cabs, carg, cos, csqrt, erfc, exp,
exp10, exp2, log, log1p, log2, pow, sin, sincos, sinh, tan and
tanh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/mips/mips32/libm-test-ulps: Likewise.
* sysdeps/mips/mips64/libm-test-ulps: Likewise.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/atomic.h to atomic-machine.h to follow that
convention.
This is the only change in this series that needs to change the
filename rather than simply removing a directory level (because both
atomic.h and bits/atomic.h exist at present).
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
[BZ #14912]
* sysdeps/aarch64/bits/atomic.h: Move to ...
* sysdeps/aarch64/atomic-machine.h: ...here.
(_AARCH64_BITS_ATOMIC_H): Rename macro to
_AARCH64_ATOMIC_MACHINE_H.
* sysdeps/alpha/bits/atomic.h: Move to ...
* sysdeps/alpha/atomic-machine.h: ...here.
* sysdeps/arm/bits/atomic.h: Move to ...
* sysdeps/arm/atomic-machine.h: ...here. Update comments.
* bits/atomic.h: Move to ...
* sysdeps/generic/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/i386/bits/atomic.h: Move to ...
* sysdeps/i386/atomic-machine.h: ...here.
* sysdeps/ia64/bits/atomic.h: Move to ...
* sysdeps/ia64/atomic-machine.h: ...here.
* sysdeps/m68k/coldfire/bits/atomic.h: Move to ...
* sysdeps/m68k/coldfire/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/m68k/m680x0/m68020/bits/atomic.h: Move to ...
* sysdeps/m68k/m680x0/m68020/atomic-machine.h: ...here.
* sysdeps/microblaze/bits/atomic.h: Move to ...
* sysdeps/microblaze/atomic-machine.h: ...here.
* sysdeps/mips/bits/atomic.h: Move to ...
* sysdeps/mips/atomic-machine.h: ...here.
(_MIPS_BITS_ATOMIC_H): Rename macro to _MIPS_ATOMIC_MACHINE_H.
* sysdeps/powerpc/bits/atomic.h: Move to ...
* sysdeps/powerpc/atomic-machine.h: ...here. Update comments.
* sysdeps/powerpc/powerpc32/bits/atomic.h: Move to ...
* sysdeps/powerpc/powerpc32/atomic-machine.h: ...here. Update
comments. Include <atomic-machine.h> instead of <bits/atomic.h>.
* sysdeps/powerpc/powerpc64/bits/atomic.h: Move to ...
* sysdeps/powerpc/powerpc64/atomic-machine.h: ...here. Include
<atomic-machine.h> instead of <bits/atomic.h>.
* sysdeps/s390/bits/atomic.h: Move to ...
* sysdeps/s390/atomic-machine.h: ...here.
* sysdeps/sparc/sparc32/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc32/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/sparc/sparc32/sparcv9/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc32/sparcv9/atomic-machine.h: ...here.
* sysdeps/sparc/sparc64/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc64/atomic-machine.h: ...here.
* sysdeps/tile/bits/atomic.h: Move to ...
* sysdeps/tile/atomic-machine.h: ...here.
* sysdeps/tile/tilegx/bits/atomic.h: Move to ...
* sysdeps/tile/tilegx/atomic-machine.h: ...here. Include
<sysdeps/tile/atomic-machine.h> instead of
<sysdeps/tile/bits/atomic.h>.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/tile/tilepro/bits/atomic.h: Move to ...
* sysdeps/tile/tilepro/atomic-machine.h: ...here. Include
<sysdeps/tile/atomic-machine.h> instead of
<sysdeps/tile/bits/atomic.h>.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/arm/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/arm/atomic-machine.h: ...here. Include
<sysdeps/arm/atomic-machine.h> instead of
<sysdeps/arm/bits/atomic.h>.
* sysdeps/unix/sysv/linux/hppa/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/hppa/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/m68k/coldfire/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/m68k/coldfire/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/nios2/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/nios2/atomic-machine.h: ...here.
(_NIOS2_BITS_ATOMIC_H): Rename macro to _NIOS2_ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/sh/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/sh/atomic-machine.h: ...here.
* sysdeps/x86_64/bits/atomic.h: Move to ...
* sysdeps/x86_64/atomic-machine.h: ...here.
* include/atomic.h: Include <atomic-machine.h> instead of
<bits/atomic.h>.
The ldbl-128 / ldbl-128ibm implementation of lgammal converts (the
floor of minus) non-integer negative arguments to int to determine the
value of signgam. When those values are outside the range of int,
this produces spurious "invalid" exceptions and incorrect values of
signgam. This patch fixes this by instead determining signgam through
comparing half the integer in question to floor of half the integer.
Tested for mips64, x86_64 and x86.
[BZ #18952]
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c (__ieee754_lgammal_r): Do
not convert non-integer negative arguments to int to determine the
value of signgam.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
This patch adds more libm test inputs found through random test
generation to increase observed ulps on x86_64.
Tested for x86_64 and x86.
* math/auto-libm-test-in: Add more tests of acosh, atanh, cbrt,
cosh, csqrt, erfc, expm1 and lgamma.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
The existing implementations of lgamma functions (except for the ia64
versions) use the reflection formula for negative arguments. This
suffers large inaccuracy from cancellation near zeros of lgamma (near
where the gamma function is +/- 1).
This patch fixes this inaccuracy. For arguments above -2, there are
no zeros and no large cancellation, while for sufficiently large
negative arguments the zeros are so close to integers that even for
integers +/- 1ulp the log(gamma(1-x)) term dominates and cancellation
is not significant. Thus, it is only necessary to take special care
about cancellation for arguments around a limited number of zeros.
Accordingly, this patch uses precomputed tables of relevant zeros,
expressed as the sum of two floating-point values. The log of the
ratio of two sines can be computed accurately using log1p in cases
where log would lose accuracy. The log of the ratio of two gamma(1-x)
values can be computed using Stirling's approximation (the difference
between two values of that approximation to lgamma being computable
without computing the two values and then subtracting), with
appropriate adjustments (which don't reduce accuracy too much) in
cases where 1-x is too small to use Stirling's approximation directly.
In the interval from -3 to -2, using the ratios of sines and of
gamma(1-x) can still produce too much cancellation between those two
parts of the computation (and that interval is also the worst interval
for computing the ratio between gamma(1-x) values, which computation
becomes more accurate, while being less critical for the final result,
for larger 1-x). Because this can result in errors slightly above
those accepted in glibc, this interval is instead dealt with by
polynomial approximations. Separate polynomial approximations to
(|gamma(x)|-1)(x-n)/(x-x0) are used for each interval of length 1/8
from -3 to -2, where n (-3 or -2) is the nearest integer to the
1/8-interval and x0 is the zero of lgamma in the relevant half-integer
interval (-3 to -2.5 or -2.5 to -2).
Together, the two approaches are intended to give sufficient accuracy
for all negative arguments in the problem range. Outside that range,
the previous implementation continues to be used.
Tested for x86_64, x86, mips64 and powerpc. The mips64 and powerpc
testing shows up pre-existing problems for ldbl-128 and ldbl-128ibm
with large negative arguments giving spurious "invalid" exceptions
(exposed by newly added tests for cases this patch doesn't affect the
logic for); I'll address those problems separately.
[BZ #2542]
[BZ #2543]
[BZ #2558]
* sysdeps/ieee754/dbl-64/e_lgamma_r.c (__ieee754_lgamma_r): Call
__lgamma_neg for arguments from -28.0 to -2.0.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (__ieee754_lgammaf_r): Call
__lgamma_negf for arguments from -15.0 to -2.0.
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c (__ieee754_lgammal_r):
Call __lgamma_negl for arguments from -48.0 or -50.0 to -2.0.
* sysdeps/ieee754/ldbl-96/e_lgammal_r.c (__ieee754_lgammal_r):
Call __lgamma_negl for arguments from -33.0 to -2.0.
* sysdeps/ieee754/dbl-64/lgamma_neg.c: New file.
* sysdeps/ieee754/dbl-64/lgamma_product.c: Likewise.
* sysdeps/ieee754/flt-32/lgamma_negf.c: Likewise.
* sysdeps/ieee754/flt-32/lgamma_productf.c: Likewise.
* sysdeps/ieee754/ldbl-128/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-128/lgamma_productl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/lgamma_productl.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_product.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_productl.c: Likewise.
* sysdeps/generic/math_private.h (__lgamma_negf): New prototype.
(__lgamma_neg): Likewise.
(__lgamma_negl): Likewise.
(__lgamma_product): Likewise.
(__lgamma_productl): Likewise.
* math/Makefile (libm-calls): Add lgamma_neg and lgamma_product.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
There are a few .set mips* assembler directives used in the MIPS specific
sysdep code that force an instruction to be assembled for a specific ISA.
The reason for these is to allow an instruction to be encoded when it might
not be supported in the current ISA (when the code is run the Linux kernel
will trap and emulate any unsupported instructions). Unfortunately forcing
a specific ISA means that when assembling for a newer ISA, where the
instruction has a different encoding, the wrong encoding will be used.
* sysdeps/mips/bits/atomic.h [_MIPS_SIM == _ABIO32] (MIPS_PUSH_MIPS2):
Only use .set mips2 if the current ISA is below mips2.
* sysdeps/mips/sys/tas.h [_MIPS_SIM == _ABIO32] (_test_and_set):
Likewise.
* sysdeps/mips/nptl/tls.h (READ_THREAD_POINTER): Only use .set
mips32r2 if the current ISA is below mips32r2.
* sysdeps/mips/tls-macros.h (TLS_RDHWR): New define.
(TLS_IE): Updated to use the TLD_RDHWR macro.
(TLS_LE): Likewise.
* sysdeps/unix/mips/sysdep.h (__mips_isa_rev): Moved out of #ifdef
__ASSEMBLER__ condition.
when initial make call has subdir= explicitly set.
* sysdeps/mach/Makefile ($(patsubst
mach%,m\%h%,$(mach-before-compile))): Force subdir to mach when
calling $(MAKE).
* sysdeps/mach/hurd/Makefile ($(patsubst %,$(hurd-objpfx)hurd/%.%,auth
io fs process)): Force subdir to hurd when calling $(MAKE).
($(common-objpfx)hurd/../mach/RPC_task_get_sampled_pcs.c): Force
subdir to mach when calling $(MAKE).
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/linkmap.h to plain linkmap.h to follow that
convention.
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
[BZ #14912]
* bits/linkmap.h: Move to ...
* sysdeps/generic/linkmap.h: ...here.
* sysdeps/aarch64/bits/linkmap.h: Move to ...
* sysdeps/aarch64/linkmap.h: ...here.
* sysdeps/arm/bits/linkmap.h: Move to ...
* sysdeps/arm/linkmap.h: ...here.
* sysdeps/hppa/bits/linkmap.h: Move to ...
* sysdeps/hppa/linkmap.h: ...here.
* sysdeps/ia64/bits/linkmap.h: Move to ...
* sysdeps/ia64/linkmap.h: ...here.
* sysdeps/mips/bits/linkmap.h: Move to ...
* sysdeps/mips/linkmap.h: ...here.
* sysdeps/s390/bits/linkmap.h: Move to ...
* sysdeps/s390/linkmap.h: ...here.
* sysdeps/sh/bits/linkmap.h: Move to ...
* sysdeps/sh/linkmap.h: ...here.
* sysdeps/x86/bits/linkmap.h: Move to ...
* sysdeps/x86/linkmap.h: ...here.
* include/link.h: Include <linkmap.h> instead of <bits/linkmap.h>.
Commit f4491417cc introduced some warnings
when building GLIBC with GCC 5.x. similar to those fixed by commit
dd6e8af6ba. This patch fixes those warnings.
* sysdeps/unix/sysv/linux/socketpair.c: Use the address of the
first member of struct sv in syscall macro.
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/stdio-lock.h to plain stdio-lock.h to follow
that convention.
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
[BZ #14912]
* bits/stdio-lock.h: Move to ...
* sysdeps/generic/stdio-lock.h: ...here.
(_BITS_STDIO_LOCK_H): Rename macro to _STDIO_LOCK_H.
* sysdeps/nptl/bits/stdio-lock.h: Move to ...
* sysdeps/nptl/stdio-lock.h: ...here.
(_BITS_STDIO_LOCK_H): Rename macro to _STDIO_LOCK_H.
* include/libio.h: Include <stdio-lock.h> instead of
<bits/stdio-lock.h>.
* sysdeps/nptl/fork.c: Likewise.
* sysdeps/pthread/flockfile.c: Likewise.
* sysdeps/pthread/ftrylockfile.c: Likewise.
* sysdeps/pthread/funlockfile.c: Likewise.
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/m68k-vdso.h to plain m68k-vdso.h to follow
that convention.
[BZ #14912]
* sysdeps/unix/sysv/linux/m68k/bits/m68k-vdso.h: Move to ...
* sysdeps/unix/sysv/linux/m68k/m68k-vdso.h: ...here.
* sysdeps/unix/sysv/linux/m68k/coldfire/bits/atomic.h: Include
<m68k-vdso.h> instead of <bits/m68k-vdso.h>.
* sysdeps/unix/sysv/linux/m68k/init-first.c: Likewise.
* sysdeps/unix/sysv/linux/m68k/m68k-helpers.S: Likewise.
* sysdeps/unix/sysv/linux/m68k/m68k-vdso.c: Likewise.
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/libc-tsd.h to plain libc-tsd.h to follow that
convention.
Tested for x86_64 (testing, and that installed stripped shared
libraries are unchanged by the patch).
[BZ #14912]
* bits/libc-tsd.h: Move to ...
* sysdeps/generic/libc-tsd.h: ...here.
(_GENERIC_BITS_LIBC_TSD_H): Rename macro to _GENERIC_LIBC_TSD_H.
* sysdeps/mach/hurd/bits/libc-tsd.h: Move to ...
* sysdeps/mach/hurd/libc-tsd.h: ...here.
(_BITS_LIBC_TSD_H): Rename macro to _LIBC_TSD_H.
* include/ctype.h: Include <libc-tsd.h> instead of
<bits/libc-tsd.h>.
* include/rpc/rpc.h: Likewise.
* locale/localeinfo.h: Likewise.
* sunrpc/rpc_thread.c: Likewise.
* sysdeps/mach/hurd/malloc-machine.h: Likewise.
* sysdeps/nptl/malloc-machine.h: Likewise.
This patch adds new constants from Linux 4.2 to netinet/in.h:
IPPROTO_MPLS and IP_BIND_ADDRESS_NO_PORT (both in
include/uapi/linux/in.h in Linux; one directly in netinet/in.h, one in
bits/in.h in glibc).
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
* inet/netinet/in.h (IPPROTO_MPLS): New enum value and macro.
* sysdeps/unix/sysv/linux/bits/in.h (IP_BIND_ADDRESS_NO_PORT): New
macro.
This patch adds move TCP_* values to sysdeps/gnu/netinet/tcp.h to
bring it up to date with Linux 4.2. TCP_SAVE_SYN and TCP_SAVED_SYN
are new in 4.2, TCP_NOTSENT_LOWAT and TCP_CC_INFO are older (Szabolcs
previously noted in
<https://sourceware.org/ml/libc-alpha/2015-06/msg00938.html> that this
header was out of date in glibc).
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
* sysdeps/gnu/netinet/tcp.h (TCP_NOTSENT_LOWAT): New macro.
(TCP_CC_INFO): Likewise.
(TCP_SAVE_SYN): Likewise.
(TCP_SAVED_SYN): Likewise.
This patch fixes the default wordsize-32 mmap implementation offset
calculation for negative values. Current code uses signed shift
operation to calculate the multiple size to use with syscall and
it is implementation defined. Change it to use a division base
on mmap page size (default being as before, 4096).
Tested on armv7hf.
[BZ #18877]
* posix/Makefile (tests): Add tst-mmap-offset.
* posix/tst-mmap.c: New file.
* sysdeps/unix/sysv/linux/generic/wordsize-32/mmap.c (__mmap): Fix
offset calculation for negative values.
We detect i586 and i686 features at run-time by checking CX8 and CMOV
CPUID features bits. We can use these information to select the best
implementation in ix86 multiarch. HAS_I586/HAS_I686 is true if i586/i686
instructions are available on the processor.
Due to the reordering and the other nifty extensions in i686, it is not
really good to use heavily i586 optimized code on an i686. It's better
to use i486 code if it isn't an i586. USE_I586/USE_I686 is true if
i586/i686 implementation should be used for the processor. USE_I586
is true only if i686 instructions aren't available. If i686 instructions
are available, we always choose i686 or i486 implementation, in that order,
and we never choose i586 implementation for i686-class processors.
* sysdeps/i386/init-arch.h: New file.
* sysdeps/i386/i586/init-arch.h: Likewise.
* sysdeps/i386/i686/init-arch.h: Likewise.
* sysdeps/x86/cpu-features.c (init_cpu_features): Set bit_I586
bit if CX8 is available. Set bit_I686 bit if CMOV is available.
* sysdeps/x86/cpu-features.h (bit_I586): New.
(bit_I686): Likewise.
(bit_CX8): Likewise.
(bit_CMOV): Likewise.
(index_CX8): Likewise.
(index_CMOV): Likewise.
(index_I586): Likewise.
(index_I686): Likewise.
(reg_CX8): Likewise.
(reg_CMOV): Likewise.
(HAS_I586): Defined as HAS_ARCH_FEATURE (I586) if i586 isn't
available at compile-time.
(HAS_I686): Defined as HAS_ARCH_FEATURE (I686) if i686 isn't
available at compile-time.
* sysdeps/x86/init-arch.h (USE_I586): New macro.
(USE_I686): Likewise.
Since glibc doesn't support i386 any more, we can remove i486 subdirectory.
* sysdeps/i386/i586/Implies: Removed.
* sysdeps/i386/i686/Implies: Likewise.
Since glibc doesn't support i386 any more, we can move i486/strlen.S
to strlen.S.
* sysdeps/i386/i486/strlen.S: Moved to ...
* sysdeps/i386/strlen.S: Here.
Since glibc doesn't support i386 any more, we can move i486/strcat.S
to strcat.S.
* sysdeps/i386/i486/strcat.S: Moved to ...
* sysdeps/i386/strcat.S: Here.
* sysdeps/i386/i686/multiarch/strcat.S: Updated.
Since glibc doesn't support i386 any more, we can move
i486/pthread_spin_trylock.S to pthread_spin_trylock.S
* sysdeps/i386/i486/pthread_spin_trylock.S: Moved to ...
* sysdeps/i386/pthread_spin_trylock.S: Here.
* sysdeps/i386/i586/pthread_spin_trylock.S: Removed.
* sysdeps/i386/i686/pthread_spin_trylock.S: Updated.
Since glibc doesn't support i386 any more, we can move
i486/string-inlines.c to string-inlines.c.
* sysdeps/i386/i486/string-inlines.c: Moved to ...
* sysdeps/i386/string-inlines.c: Here.
Since glibc doesn't support i386 any more, we can move i486/bits/atomic.h
to bits/atomic.h.
* sysdeps/i386/i486/bits/atomic.h: Moved to ...
* sysdeps/i386/bits/atomic.h: Here.
As with other spots in the code, GCC 4.8 unnecessarily complains about
an uninitialized variable in tanl calcs, so this patch disables. With
it, the library and sees the usual set of test passes.
* sysdeps/ieee754/ldbl-96/k_tanl.c: Include <libc-internal.h>.
(__kernel_tanl): Ignore uninitialized warnings around use of SIGN.
Linux commit b4b56f9ecab40f3b4ef53e130c9f6663be491894 introduced
a new HWCAP2 bit to indicate that the kernel now aborts a memory
transaction when a syscall is made. This patch adds that bit to
sysdeps/powerpc/bits/hwcap.h.
2015-08-26 Carlos Eduardo Seo <cseo@linux.vnet.ibm.com>
* sysdeps/powerpc/bits/hwcap.h: Add PPC_FEATURE2_HTM_NOSC.
* sysdeps/powerpc/dl-procinfo.c:
(_dl_powerpc_cap_flags): Added descriptor for this hwcap
feature so it shows when LD_SHOW_AUXV=1.
Power ISA 2.07B section B.5.5 relaxed the barrier requirement around a
TLE enabled lock. It is now identical to a traditional lock.
2015-08-26 Paul E. Murphy <murphyp@linux.vnet.ibm.com>
* sysdeps/unix/sysv/linux/powerpc/elision-lock.c
(__arch_compare_and_exchange_val_32_acq): Remove and use common
definition. ISA 2.07B no longer requires full sync.
Replace BZERO_P with USE_AS_BZERO in i586/i686 memset.S to support i386
multi-arch memset. Also we should check SHARED not PIC for libc.so
since libc.a may be compiled with PIC.
* sysdeps/i386/i586/bzero.S (USE_AS_BZERO): New.
* sysdeps/i386/i686/bzero.S (USE_AS_BZERO): Likewise.
* sysdeps/i386/i586/memset.S (BZERO_P): Removed.
Check USE_AS_BZERO/SHARED instead of BZERO_P/PIC.
(__memset_zero_constant_len_parameter): New.
* sysdeps/i386/i686/memset.S (BZERO_P): Removed.
Check USE_AS_BZERO/SHARED instead of BZERO_P/PIC.
(__memset_zero_constant_len_parameter): Don't define if
__memset_chk or USE_AS_BZERO are defined.
Replace MEMPCPY_P with USE_AS_MEMPCPY in i586 memcpy.S to support i386
multi-arch memcpy. Also we should check SHARED not PIC for libc.so
since libc.a may be compiled with PIC.
* sysdeps/i386/i586/memcpy.S (MEMPCPY_P): Removed.
Check USE_AS_MEMPCPY/SHARED instead of MEMPCPY_P/PIC.
* sysdeps/i386/i586/mempcpy.S (USE_AS_MEMPCPY): New.
Since x86-64 ld.so preserves vector registers now, we can use SSE in
x86-64 ld.so. We should run tst-ld-sse-use.sh only on i386.
* sysdeps/x86/Makefile [$(subdir) == elf] (CFLAGS-.os,
tests-special, $(objpfx)tst-ld-sse-use.out): Moved to ...
* sysdeps/i386/Makefile [$(subdir) == elf] (CFLAGS-.os,
tests-special, $(objpfx)tst-ld-sse-use.out): Here. Update
comments.
* sysdeps/x86_64/Makefile [$(subdir) == elf] (CFLAGS-.os): Add
-mno-mmx for $(all-rtld-routines).
* sysdeps/x86/tst-ld-sse-use.sh: Moved to ...
* sysdeps/i386/tst-ld-sse-use.sh: Here. Replace x86-64 with
i386.
Building glibc on s390-32 with gcc option -mzarch produces the error due to
sysdeps/s390/jmpbuf-unwind.h:37:10: (void *) (_Unwind_GetCFA (_context):
cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
Building on s390-32 in esa-mode or s390-64 is fine.
_Unwind_GetCFA returns an _Unwind_Word which is an unsigned
with a size of 4 bytes on s390-32 (esa-mode) and 8 bytes on s390-64.
On s390-32 (zarch-mode), _Unwind_Word has a size of 8 bytes, too.
_Unwind_Word is defined in sysdeps/generic/unwind.h as
typedef unsigned _Unwind_Word __attribute__((__mode__(__word__)));
In libgcc unwind header (<gcc-src>/libgcc/unwind-generic.h) this typedef has
changed to "typedef unsigned _Unwind_Word __attribute__((__mode__(__unwind_word__)));"
in June 2008.
With this mode, _Unwind_Word has a size of 4 bytes on s390-32 (zarch-mode).
The same change applies to _Unwind_Sword.
Thus this patch updates the unwind header according to these changes.
Afterwards, the int-to-pointer-cast-warning is gone away on s390-32 (zarch-mode)
and the testsuite runs with the same test-failures as s390-32 (esa-mode)
plus FAIL: c++-types-check. Here register_t is expected to has a size of 4 bytes,
but it has a size of 8 bytes due to:
posix/sys/types.h:205:typedef int register_t __attribute__ ((__mode__ (__word__)));
The libgcc-patch for gcc 4.4 can be found here:
"[PATCH, spu, unwind] Remove attribute ((mode (word))) from unwind.h"
https://gcc.gnu.org/ml/gcc-patches/2008-06/msg00969.html
ChangeLog:
* sysdeps/generic/unwind.h
(_Unwind_Word): Use __mode__(__unwind_word__)
instead of __mode__(__word__).
(_Unwind_Sword): Likewise.
This patch fixes the build error with gcc6:
array subscript is above array bounds [-Werror=array-bounds]
While including loop.c to construct the SINGLE(LOOPFCT) method
for converting from UTF-16 to UTF-8, the bytebuf array with length
MAX_NEEDED_INPUT is used as inptr. MAX_NEEDED_INPUT defaults to
MIN_NEEDED_INPUT if not defined before including loop.c.
Thus bytebuf has a length of 2.
This patch defines MAX_NEEDED_INPUT to MAX_NEEDED_TO, which is 4.
ChangeLog:
* sysdeps/s390/s390-64/utf8-utf16-z9.c
(MAX_NEEDED_INPUT): New define.
(MAX_NEEDED_OUTPUT): New define.
This patch provides optimized version of memrchr with the z13 vector
instructions.
ChangeLog:
* sysdeps/s390/multiarch/memrchr-c.c: New File.
* sysdeps/s390/multiarch/memrchr-vx.S: Likewise.
* sysdeps/s390/multiarch/memrchr.c: Likewise.
* sysdeps/s390/multiarch/Makefile
(sysdep_routines): Add memrchr functions.
* sysdeps/s390/multiarch/ifunc-impl-list-common.c
(__libc_ifunc_impl_list_common): Add ifunc test for memrchr.
This patch provides optimized versions of memccpy with the z13 vector
instructions.
ChangeLog:
* sysdeps/s390/multiarch/memccpy-c.c: New File.
* sysdeps/s390/multiarch/memccpy-vx.S: Likewise.
* sysdeps/s390/multiarch/memccpy.c: Likewise.
* sysdeps/s390/multiarch/Makefile
(sysdep_routines): Add memccpy functions.
* sysdeps/s390/multiarch/ifunc-impl-list-common.c
(__libc_ifunc_impl_list_common): Add ifunc test for memccpy.
* string/memccpy.c: Use MEMCCPY if defined.
This patch provides optimized versions of strcmp and wcscmp with the z13
vector instructions.
The architecture specific string.h had a typo, which leads to ommiting the
inline version in this file if __USE_STRING_INLINES is defined.
Tested this inline version by tweaking test-strcmp.c.
ChangeLog:
* sysdeps/s390/multiarch/strcmp-vx.S: New File.
* sysdeps/s390/multiarch/strcmp.c: Likewise.
* sysdeps/s390/multiarch/wcscmp-c.c: Likewise.
* sysdeps/s390/multiarch/wcscmp-vx.S: Likewise.
* sysdeps/s390/multiarch/wcscmp.c: Likewise.
* sysdeps/s390/s390-32/multiarch/strcmp.c: Likewise.
* sysdeps/s390/s390-64/multiarch/strcmp.c: Likewise.
* sysdeps/s390/multiarch/Makefile (sysdep_routines): Add strcmp and
wcscmp functions.
* sysdeps/s390/multiarch/ifunc-impl-list.c
(__libc_ifunc_impl_list): Add ifunc test for strcmp, wcscmp.
* string/strcmp.c (STRCMP): Define and use macro.
* benchtests/bench-wcscmp.c: New File.
* benchtests/Makefile (wcsmbs-bench): Add wcscmp.
* sysdeps/s390/bits/string.h: Fix typo: _HAVE_STRING_ARCH_strcmp
instead of _HAVE_STRING_ARCH_memchr.
This patch provides optimized versions of strlen and wcslen with the z13 vector
instructions.
The helper macro IFUNC_VX_IMPL is introduced and is used to register all
__<func>_c() and __<func>_vx() functions within __libc_ifunc_impl_list()
to the ifunc test framework.
ChangeLog:
* sysdeps/s390/multiarch/Makefile: New File.
* sysdeps/s390/multiarch/strlen-c.c: Likewise.
* sysdeps/s390/multiarch/strlen-vx.S: Likewise.
* sysdeps/s390/multiarch/strlen.c: Likewise.
* sysdeps/s390/multiarch/wcslen-c.c: Likewise.
* sysdeps/s390/multiarch/wcslen-vx.S: Likewise.
* sysdeps/s390/multiarch/wcslen.c: Likewise.
* string/strlen.c (STRLEN): Define and use macro.
* sysdeps/s390/multiarch/ifunc-impl-list.c
(IFUNC_VX_IMPL): New macro function.
(__libc_ifunc_impl_list): Add ifunc test for strlen, wcslen.
* benchtests/Makefile (wcsmbs-bench): New variable.
(string-bench-all): Added wcsmbs-bench.
* benchtests/bench-wcslen.c: New File.
This patch introduces a s390 specific ifunc resolver macro for 32/64bit,
which chooses <func>_vx with vector instructions if HWCAP_S390_VX flag
in hwcaps is set or <func>_c if not.
ChangeLog:
* sysdeps/s390/multiarch/ifunc-resolve.h (s390_vx_libc_ifunc,
s390_vx_libc_ifunc2): New macro function.
The S390 specific test checks if the assembler has support for the new z13
vector instructions by compiling a vector instruction. The .machine and
.machinemode directives are needed to compile the vector instruction without
-march=z13 option on 31/64 bit.
On success the macro HAVE_S390_VX_ASM_SUPPORT is defined. This macro is used
to determine if the optimized functions can be build without compile errors.
If the used assembler lacks vector support, then a warning is dumped while
configuring and only the common code functions are build.
The z13 instruction support was introduced in
"[Committed] S/390: Add support for IBM z13."
(https://sourceware.org/ml/binutils/2015-01/msg00197.html)
ChangeLog:
* config.h.in (HAVE_S390_VX_ASM_SUPPORT): New macro undefine.
* sysdeps/s390/configure.ac: Add test for S390 vector instruction
assembler support.
* sysdeps/s390/configure: Regenerated.
The new IBM z13 is added to platform string array.
The macro _DL_PLATFORMS_COUNT is incremented to 8,
because it was not incremented by commit
"S/390: Sync AUXV capabilities and archs with kernel".
ChangeLog:
* sysdeps/s390/dl-procinfo.c (_dl_s390_cap_flags): Add z13.
* sysdeps/s390/dl-procinfo.h (_DL_PLATFORMS_COUNT): Increased.
The HWCAP_S390_VX flag in hwcap field of auxiliary vector indicates
if the vector facility is available and the kernel is aware of it.
This can be tested with LD_SHOW_AUXV=1 <prog>.
Currently it does not show te, because it was not incremented
by commit "S/390: Add hwcap value for transactional execution.".
Thus _DL_HWCAP_COUNT is incremented by two.
ChangeLog:
* sysdeps/s390/dl-procinfo.c (_dl_s390_platforms): Add vector flag.
* sysdeps/s390/dl-procinfo.h: Add vector capability.
* sysdeps/unix/sysv/linux/s390/bits/hwcap.h (HWCAP_S390_VX): Define.
On s390 all ifunc resolvers were implemented in multiarch/ifunc-resolve.c.
The resulting single object files has undefined references to all ifunc-functions.
This patch introduces one multiarch/<func>.c file for each of memcpy, memcmp
and memset with the function specific ifunc resolver. The different function
implementations are now implemented in multiarch/<func>-s390x.S
(moved from multiarch/<func>.S).
The new multiarch/ifunc-resolve.h file contains the ifunc-resolver macro
and other helper-macros. They are merged and are now used in common for
32/64bit. Therefore the __<func>_g5/__<func>_z900 functions were renamed to
__<func>_default.
This patch also enables testing the ifunc implementations by implementing
the function __libc_ifunc_impl_list. It uses the helper-macros of ifunc-resolve.h.
ChangeLog:
* sysdeps/s390/s390-32/multiarch/Makefile (sysdep_routines):
Remove ifunc-resolve, add memset-s390, memcpy-s390, memcmp-s390.
* sysdeps/s390/s390-32/multiarch/ifunc-resolve.c: Delete File.
* sysdeps/s390/s390-32/multiarch/memcmp.S: Move to ...
* sysdeps/s390/s390-32/multiarch/memcmp-s390.S: ... here.
(memcmp, bcmp): Use __memcmp_default as alias source.
* sysdeps/s390/s390-32/multiarch/memcmp.c: New File.
* sysdeps/s390/s390-32/memcmp.S (__memcmp_g5):
Rename to __memcmp_default.
* sysdeps/s390/s390-32/multiarch/memcpy.S: Move to ...
* sysdeps/s390/s390-32/multiarch/memcpy-s390.S: ... here.
(memcpy): Use __memcpy_default as alias source.
* sysdeps/s390/s390-32/multiarch/memcpy.c: New File.
* sysdeps/s390/s390-32/memcpy.S (__memcpy_g5):
Rename to __memcpy_default.
* sysdeps/s390/s390-32/multiarch/memset.S: Move to ...
* sysdeps/s390/s390-32/multiarch/memset-s390.S: ... here.
(memset): Use __memset_default as alias source.
* sysdeps/s390/s390-32/multiarch/memset.c: New File.
* sysdeps/s390/s390-32/memset.S (__memset_g5):
Rename to __memset_default.
* sysdeps/s390/s390-64/multiarch/Makefile (sysdep_routines):
Remove ifunc-resolve, add memset-s390x, memcpy-s390x, memcmp-s390x.
* sysdeps/s390/s390-64/multiarch/ifunc-resolve.c: Delete File.
* sysdeps/s390/s390-64/multiarch/memcmp.S: Move to ...
* sysdeps/s390/s390-64/multiarch/memcmp-s390x.S: ... here.
(memcmp, bcmp): Use __memcmp_default as alias source.
* sysdeps/s390/s390-64/multiarch/memcmp.c: New File.
* sysdeps/s390/s390-64/memcmp.S (__memcmp_z900):
Rename to __memcmp_default.
* sysdeps/s390/s390-64/multiarch/memcpy.S: Move to ...
* sysdeps/s390/s390-64/multiarch/memcpy-s390x.S: ... here.
(memcpy): Use __memcpy_default as alias source.
* sysdeps/s390/s390-64/multiarch/memcpy.c: New File.
* sysdeps/s390/s390-64/memcpy.S (__memcpy_z900):
Rename to __memcpy_default.
* sysdeps/s390/s390-64/multiarch/memset.S: Move to ...
* sysdeps/s390/s390-64/multiarch/memset-s390x.S: ... here.
(memset): Use __memset_default as alias source.
* sysdeps/s390/s390-64/multiarch/memset.c: New File.
* sysdeps/s390/s390-64/memset.S (__memset_z900):
Rename to __memset_default.
* sysdeps/s390/multiarch/ifunc-resolve.h: New File.
* sysdeps/s390/multiarch/ifunc-impl-list.c: New File.
On s390, the DXC(data-exception-code)-byte in FPC(floating-point-control)-
register contains a code of the last occured exception.
If bits 6 and 7 of DXC-byte are zero, the bits 0-5 correspond to the
ieee-exception flag bits.
The current implementation always uses these bits as ieee-exception flag bits.
fetestexcept() reports any exception after the first usage of a
vector-instruction in a process, because it raises an "vector instruction
exception" with DXC-code 0xFE.
This patch fixes the handling of the DXC-byte. The DXC-Byte is only handled
if bits 6 and 7 are zero.
The #define _FPU_RESERVED is extended by the DXC-Byte.
Otherwise the tests math/test-fpucw-static and math/test-fpucw-ieee-static
fails, because DXC-Byte contains the vector instruction exception when reaching
main(). This exception was triggered by strrchr() call in __init_misc().
__init_misc() is called after __setfpucw () in __libc_init_first().
The field __ieee_instruction_pointer in struct fenv_t is renamed to __unused
because it is a relict from commit "Remove PTRACE_PEEKUSER"
(87b9b50f0d) and isn´t used anymore.
ChangeLog:
[BZ #18610]
* sysdeps/s390/fpu/bits/fenv.h (fenv_t): Rename
__ieee_instruction_pointer to __unused.
* sysdeps/s390/fpu/fesetenv.c (__fesetenv): Remove usage of
__ieee_instruction_pointer.
* sysdeps/s390/fpu/fclrexcpt.c (feclearexcept): Fix dxc-field handling.
* sysdeps/s390/fpu/fgetexcptflg.c (fegetexceptflag): Likewise.
* sysdeps/s390/fpu/fsetexcptflg.c (fesetexceptflag): Likewise.
* sysdeps/s390/fpu/ftestexcept.c (fetestexcept): Likewise.
* sysdeps/s390/fpu/fpu_control.h (_FPU_RESERVED):
Mark dxc-field as reserved.
Since ld.so preserves vector registers now, we can use the same SSE2
optimized strcmp in x86-64 libc and ld.so.
* sysdeps/x86_64/strcmp.S: Remove "#if !IS_IN (libc)".
Since _dl_x86_64_save_sse and _dl_x86_64_restore_sse are removed now,
we don't need to run tst-getpid2 with LD_BIND_NOW=1.
[BZ #11214]
* sysdeps/unix/sysv/linux/Makefile (tst-getpid2-ENV): Removed.
Explicit system calls for the socket operations were added in Linux kernel
in commit 86250b9d12ca for powerpc. This patch make use of those instead of
calling socketcall to save number of cycles on networking syscalls.
2015-08-25 Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
* sysdeps/unix/sysv/linux/powerpc/kernel-features.h: Define new macros.
* sysdeps/unix/sysv/linux/accept.c: Call direct system call.
* sysdeps/unix/sysv/linux/bind.c: Call direct system call.
* sysdeps/unix/sysv/linux/connect.c: Call direct system call.
* sysdeps/unix/sysv/linux/getpeername.c: Call direct system call.
* sysdeps/unix/sysv/linux/getsockname.c: Call direct system call.
* sysdeps/unix/sysv/linux/getsockopt.c: Call direct system call.
* sysdeps/unix/sysv/linux/listen.c: Call direct system call.
* sysdeps/unix/sysv/linux/recv.c: Call direct system call.
* sysdeps/unix/sysv/linux/recvfrom.c: Call direct system call.
* sysdeps/unix/sysv/linux/recvmsg.c: Call direct system call.
* sysdeps/unix/sysv/linux/send.c: Call direct system call.
* sysdeps/unix/sysv/linux/sendmsg.c: Call direct system call.
* sysdeps/unix/sysv/linux/sendto.c: Call direct system call.
* sysdeps/unix/sysv/linux/setsockopt.c: Call direct system call.
* sysdeps/unix/sysv/linux/shutdown.c: Call direct system call.
* sysdeps/unix/sysv/linux/socket.c: Call direct system call.
* sysdeps/unix/sysv/linux/socketpair.c: Call direct system call.
Fix usage of tabort in generated syscalls. r0 has special meaning
when used with this instruction, thus it will not generate
persistent errors, nor return an error code. This mitigates poor
CPU usage when performing elided critical sections.
Additionally, transactions should be aborted when entering a user
invoked syscall. Otherwise the results of the transaction may be
undefined.
2015-08-25 Paul E. Murphy <murphyp@linux.vnet.ibm.com>
* sysdeps/powerpc/powerpc32/sysdep.h (ABORT_TRANSACTION): Use
register other than r0 for tabort, it has special meaning.
* sysdeps/powerpc/powerpc64/sysdep.h (ABORT_TRANSACTION): Likewise
* sysdeps/unix.sysv/linux/powerpc/syscall.S (syscall): Abort
transaction before starting syscall.
Instead of checking needle length, constant 'n' number of comparisons
is checked to fall back to default implementation. This patch is tested
on powerpc64 and powerpc64le.
2015-08-25 Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
* sysdeps/powerpc/powerpc64/power7/strstr.S: Handle worst case.
Since ld.so preserves vector registers now, we can use %xmm[0-4] to
avoid the REX prefix.
* sysdeps/x86_64/strlen.S: Replace %xmm[8-12] with %xmm[0-4].
Hi, as I wrote in previous patches a performance of checked strcpy and
stpcpy is terrible as these don't use sse2 and are around four times
slower that strcpy and stpcpy now.
As this bug shows that these functions are not performance sensitive I
decided just to improve generic implementation instead for easier
maintainance.
* debug/strcpy_chk.c: Improve performance.
* debug/stpcpy_chk.c: Likewise.
* sysdeps/x86_64/strcpy_chk.S: Remove.
* sysdeps/x86_64/stpcpy_chk.S: Remove.
This patch adds SSE, AVX and AVX512 versions of _dl_runtime_resolve
and _dl_runtime_profile, which save and restore the first 8 vector
registers used for parameter passing. elf_machine_runtime_setup
selects the proper _dl_runtime_resolve or _dl_runtime_profile based
on _dl_x86_cpu_features. It avoids race condition caused by
FOREIGN_CALL macros, which are only used for x86-64.
Performance impact of saving and restoring 8 vector registers are
negligible on Nehalem, Sandy Bridge, Ivy Bridge and Haswell when
ld.so is optimized with SSE2.
[BZ #15128]
* sysdeps/x86_64/Makefile [$(subdir) == elf] (tests): Add
ifuncmain8.
(modules-names): Add ifuncmod8.
($(objpfx)ifuncmain8): New rule.
* sysdeps/x86_64/dl-machine.h: Include <dl-procinfo.h> and
<cpuid.h>.
(elf_machine_runtime_setup): Use _dl_runtime_resolve_sse,
_dl_runtime_resolve_avx, or _dl_runtime_resolve_avx512,
_dl_runtime_profile_sse, _dl_runtime_profile_avx, or
_dl_runtime_profile_avx512, based on HAS_ARCH_FEATURE.
* sysdeps/x86_64/dl-trampoline.S: Rewrite.
* sysdeps/x86_64/dl-trampoline.h: Likewise.
* sysdeps/x86_64/ifuncmain8.c: New file.
* sysdeps/x86_64/ifuncmod8.c: Likewise.
* sysdeps/x86_64/nptl/tcb-offsets.sym (RTLD_SAVESPACE_SSE):
Removed.
* sysdeps/x86_64/nptl/tls.h (__128bits): Removed.
(tcbhead_t): Change rtld_must_xmm_save to __glibc_unused1.
Change rtld_savespace_sse to __glibc_unused2.
(RTLD_CHECK_FOREIGN_CALL): Removed.
(RTLD_ENABLE_FOREIGN_CALL): Likewise.
(RTLD_PREPARE_FOREIGN_CALL): Likewise.
(RTLD_FINALIZE_FOREIGN_CALL): Likewise.
PowerPC has always used __IPC_64 like most other architectures, which
means that __ASSUME_IPC64 can be always true. Also, all other
architecture implementations that use the ipc syscall are effectively
identical to the generic version and can be removed.
In powerpc64, memchr was always pointing to the internal __GI_memchr
implementation. This patch fixes that and makes it use the
optimized POWER7 version when adequate.
* sysdeps/powerpc/powerpc64/multiarch/memchr-ppc64.c: Make
memchr not point to the internal __GI_memchr implementation.
sysdeps/i386/i686/multiarch/strcasestr-c.c became unused after
commit 1818483b15
Author: Andreas Schwab <schwab@suse.de>
Date: Wed Dec 18 11:53:27 2013 +1000
Remove use of SSE4.2 functions for strstr on i686
which contains
-sysdep_routines += strcspn-c strpbrk-c strspn-c strstr-c strcasestr-c
+sysdep_routines += strcspn-c strpbrk-c strspn-c
sysdeps/x86_64/multiarch/strcasestr.c became useless after
t 584b18eb4d
Author: Ondřej Bílka <neleai@seznam.cz>
Date: Sat Dec 14 19:33:56 2013 +0100
Add strstr with unaligned loads. Fixes bug 12100.
which changes sysdeps/x86_64/multiarch/strcasestr.c to
libc_ifunc (__strcasestr, __strcasestr_sse2);
This patch removes these file.
* i386/i686/multiarch/strcasestr-c.c: Removed.
* x86_64/multiarch/strcasestr.c: Likewise.
* x86_64/multiarch/ifunc-impl-list.c (__libc_ifunc_impl_list):
Remove strcasestr.
Removing the use of -Wno-uninitialized for math/ shows errors for
ldbl-128ibm:
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c: In function '__nearbyintl':
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c:119:34: error: 'low' may be used uninitialized in this function [-Werror=maybe-uninitialized]
u.d[1].d = high - u.d[0].d + low;
^
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c:119:23: error: 'high' may be used uninitialized in this function [-Werror=maybe-uninitialized]
u.d[1].d = high - u.d[0].d + low;
^
These errors are correct: if the high part of the argument is a NaN,
and the low part is nonzero but has absolute value less than 2^52,
those variables can be used uninitialized. This patch rearranges the
code so that the variables are always initialized with the natural
values, and then possibly modified later, to avoid this uninitialized
use. (Note that there are still other issues with this code and NaNs
that are not fixed by this patch.) No bug filed in Bugzilla or
testcase added for the uninitialized use since it wasn't user-visible
with the compiler I tried (that is, I still got a NaN result).
Tested for powerpc.
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c: Always initialize
variables for high and low parts before possibly modifying them.
Move sysdeps/x86_64/multiarch/init-arch.h to sysdeps/x86/init-arch.h
which can be used for both i386 and x86_64.
* sysdeps/i386/i686/multiarch/init-arch.h: Removed.
* sysdeps/unix/sysv/linux/x86/init-arch.h: Likewise.
* sysdeps/x86_64/cacheinfo.c: Include <init-arch.h> instead
of "multiarch/init-arch.h".
* sysdeps/x86_64/multiarch/init-arch.h: Renamed to ...
* sysdeps/x86/init-arch.h: This.
Both files include sysdeps/x86_64/multiarch/init-arch.c which has been
removed.
* sysdeps/i386/i686/multiarch/init-arch.c: Removed.
* sysdeps/unix/sysv/linux/x86/init-arch.c: Likewise.
The csqrt implementations in glibc can miss underflow exceptions when
the real or imaginary part of the result becomes tiny in the course of
scaling down (in particular, multiplication by 0.5) and that scaling
is exact although the relevant part of the mathematical result isn't.
This patch forces the exception in a similar way to previous fixes.
Tested for x86_64 and x86.
[BZ #18370]
* math/s_csqrt.c (__csqrt): Force underflow exception for results
whose real or imaginary part has small absolute value.
* math/s_csqrtf.c (__csqrtf): Likewise.
* math/s_csqrtl.c (__csqrtl): Likewise.
* math/auto-libm-test-in: Add more tests of csqrt.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
This patch adds extra inline functions to change the Program Priority
Register from ISA 2.07.
2015-08-19 Gabriel F. T. Gomes <gftg@linux.vnet.ibm.com>
* sysdeps/powerpc/sys/platform/ppc.h (__ppc_set_ppr_med_high,
__ppc_set_ppr_very_low): New functions.
* manual/platform.texi: Add documentation about
__ppc_set_ppr_med_high and __ppc_set_ppr_very_low.
Fix the bind-now case when DT_REL and DT_JMPREL sections are separate
and there is a gap between them.
[BZ #14341]
* elf/dynamic-link.h (elf_machine_lazy_rel): Properly handle the
case when there is a gap between DT_REL and DT_JMPREL sections.
* sysdeps/x86_64/Makefile (tests): Add tst-split-dynreloc.
(LDFLAGS-tst-split-dynreloc): New.
(tst-split-dynreloc-ENV): Likewise.
* sysdeps/x86_64/tst-split-dynreloc.c: New file.
* sysdeps/x86_64/tst-split-dynreloc.lds: Likewise.
__xstat_conv, __xstat64_conv and __xstat32_conv are internal to glibc.
They should be marked as hidden so that they can't be called without
PLT.
[BZ #18822]
* sysdeps/unix/sysv/linux/xstatconv.h (__xstat_conv): Add
attribute_hidden.
(__xstat64_conv): Likewise.
(__xstat32_conv): Likewise.
Since _dl_x86_cpu_features is always available, we can use x86-64
cacheinfo.c and sysconf.c for both i386 and x86-64.
* sysdeps/i386/i686/Makefile
[$(subdir) == string] (sysdep_routines): Moved to ...
* sysdeps/i386/Makefile: Here.
* sysdeps/i386/i686/cacheinfo.c: Moved to ...
* sysdeps/i386/cacheinfo.c: Here.
* sysdeps/unix/sysv/linux/i386/sysconf.c: Removed.
* sysdeps/unix/sysv/linux/i386/i686/sysconf.c: Likewise.
* sysdeps/unix/sysv/linux/x86_64/sysconf.c: Moved to ...
* sysdeps/unix/sysv/linux/x86/sysconf.c: Here.
This patch fixes -Wundef warnings relating to __mips_isa_rev being
undefined.
Tested for mips64 (all three ABIs) that there is a clean build and
testsuite run with -Wno-error=undef removed (and my other -Wundef
patches applied).
* sysdeps/mips/dl-machine.h [__mips_isa_rev < 6]: Change
conditionals to [!defined __mips_isa_rev || __mips_isa_rev < 6].
* sysdeps/mips/machine-gmon.h [__mips_isa_rev < 6]: Likewise.
Some features in hwcap.h do not have matching string descriptors
to be displayed when LD_SHOW_AUXV=1. This patch fixes the problem.
2015-08-13 Carlos Eduardo Seo <cseo@linux.vnet.ibm.com>
* sysdeps/powerpc/dl-procinfo.c:
(_dl_powerpc_cap_flags): Added missing strings for some
hwcap features.
* sysdeps/powerpc/dl-procinfo.h: Updated hwcap bit count.
cpuid, i586 and i686 instructions are available if the processor
specified by -march= supports them. We can use this information
to determine whether those instructions can be used safely.
* sysdeps/x86/cpu-features.c (init_cpu_features): Check
whether cpuid is available only if HAS_CPUID is 0.
* sysdeps/x86/cpu-features.h (HAS_CPUID): New.
(HAS_I586): Likewise.
(HAS_I686): Likewise.
This brings hppa inline with all the other arches and main code where we
require TLS support everywhere. That means dropping the defines USE_TLS
and USE___THREAD, and dropping the binutils check (since we already have
a version requirement that is new enough).
Various fma implementations have logic that, when computing fma (x, y,
z) where z is large (so care needs taking to avoid internal overflow)
but x * y is small, scale x * y up instead of down to avoid internal
underflows resulting from scaling down. (In these cases, x * y is
small enough that only its sign actually matters rather than the exact
value.)
The threshold for scaling up instead of down was correct for "if the
unscaled values were multiplied, the low part of the multiplication
could underflow", and the scaling was sufficient to ensure that the
low part of the multiplication did not underflow (given that cases of
very small x * y - less than half the least subnormal - were
previously dealt with). However, the choice in the functions wasn't
between scaling up or no scaling, but between scaling up and scaling
down (scaling down actually being needed when x * y isn't so small
compared to z and so the exact value does matter). Thus a larger
threshold is needed to ensure that scaling down doesn't produce values
the multiplication of whose low parts underflows. This patch
increases the thresholds accordingly.
Tested for x86_64, x86 and mips64 (with the MIPS version of s_fmal.c
removed so that the ldbl-128 version gets tested instead of the
soft-fp one).
[BZ #18824]
* sysdeps/ieee754/dbl-64/s_fma.c (__fma): Increase threshold for
scaling x * y up instead of down.
* sysdeps/ieee754/ldbl-128/s_fmal.c (__fmal): Likewise.
* sysdeps/ieee754/ldbl-96/s_fmal.c (__fmal): Likewise.
* math/auto-libm-test-in: Add more tests of fma.
* math/auto-libm-test-out: Regenerated.
The change in 0b5395f052 replaced calls
to __get_cpu_features@plt followed by a mov from rax to rdx, with a
single macro LOAD_RTLD_GLOBAL_RO_RDX. It is pretty clear that there
was a typo in s_floorf and __nearbyint due to which the (now incorrect)
mov was not removed. This patch removes that mov.
* sysdeps/x86_64/fpu/multiarch/s_floorf.S (__floorf): Remove
unnecessary movq.
* sysdeps/x86_64/fpu/multiarch/s_nearbyint.S (__nearbyint):
Likewise.
This patch adds more test inputs to various libm functions found
through random generation to have larger ulps errors than previously
listed in libm-test-ulp, on at least one of x86_64 and x86.
Tested for x86_64 and x86.
* math/auto-libm-test-in: Add more tests of acos, acosh, asin,
asinh, atan, atan2, atanh, cabs, cbrt, cosh, csqrt, erf, erfc,
exp, exp2, lgamma, log, log1p, log2, pow, sin, sincos, tan, tanh
and tgamma.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
Similar to various other bugs in this area, some tanh implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact. This patch forces the exception in a
similar way to previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16520]
* sysdeps/ieee754/dbl-64/s_tanh.c: Include <float.h>.
(__tanh): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/flt-32/s_tanhf.c: Include <float.h>.
(__tanhf): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-128/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-128ibm/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-96/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* math/auto-libm-test-in: Add more tests of tanh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
Since not all i486 processors support cpuid, we call __get_cpuid_max to
check if cpuid is available before using it if not compiling for i586,
i686 nor x86-64.
* sysdeps/x86/cpu-features.c (init_cpu_features): Call
__get_cpuid_max if not compiling for i586, i686 nor x86-64.
This patch updates x86 elision-conf.c to use the newly defined
HAS_CPU_FEATURE from <cpu-features.h>.
* sysdeps/unix/sysv/linux/x86/elision-conf.c (elision_init):
Replace HAS_RTM with HAS_CPU_FEATURE (RTM).
When building with --disable-multi-arch the memmove and strstr POWER7
optimization create and uses symbols that conflict with expect conform
tests.
* sysdeps/powerpc/powerpc64/power7/memmove.S (bcopy): Changing to
__bcopy and add a weak_alias to bcopy.
* sysdeps/powerpc/powerpc64/power7/strstr.S (strstr): Use __strnlen
for static build.
This patches uses the default strcpy/stpcpy implementation for
POWER7/PPC64. This is faster in mostly inputs for benchtests
and for multiarch the implementation uses the POWER7 strlen and
memcpy.
* string/stpcpy.c (__stpcpy): Use STPCPY to redefine symbol name and
cleanup macro usage.
* string/strcpy.c (strcpt): Use STRCPY to redefine symbol name.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy-power7.S: Remove file.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy-ppc64.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/strcpy-power7.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/strcpy-ppc64.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/stpcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/strcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/strcpy.c: Likewise.
* sysdeps/powerpc/powerpc64/stpcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/strcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
[SHARED && IS_IN (libc)]: Include <string/strcpy.c>.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
[SHARED && IS_IN (libc)]: Include <string/stpcpy.c>.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy-power7.c: New file.
* sysdeps/powerpc/powerpc64/multiarch/stpcpy-ppc64.c: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/strcpy-power7.c: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/strcpy-ppc64.c: Likewise.
* sysdeps/powerpc/powerpc64/power7/strcpy.c: Likewise.
This patch fixes the strstr build with --disable-multi-arch option.
The optimization calls the __strstr_ppc symbol, which always build
for multiarch config but not if it is disable. This patch fixes it
by adding the default C implementation object with the expected
symbol name.
* sysdeps/powerpc/powerpc64/power7/Makefile [$(subdir) = string]
(sysdep_routines): Add strstr-ppc64.
* sysdeps/powerpc/powerpc64/power7/strstr-ppc64.c: New file.
This patch adds more tests of various libm functions found through
random test generation to give increased ulps on 32-bit x86.
Tested for x86_64 and x86.
* math/auto-libm-test-in: Add more tests of acosh, asin, asinh,
atanh, cabs, carg, cbrt, cosh, csqrt, erf, erfc, exp, exp10,
expm1, hypot, log, log10, log1p, log2, pow, sinh, tan and tgamma.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
ldbl-128ibm tanhl uses a too-small threshold to decide when to return
+/-1, resulting in large errors. This patch changes it to a more
appropriate threshold (the requirement is for 2*exp(-2|x|) to be small
in terms of ulps of 1).
Tested for x86_64, x86 and powerpc.
[BZ #18790]
* sysdeps/ieee754/ldbl-128ibm/s_tanhl.c (__tanhl): Increase
threshold for returning +/- 1.
* math/auto-libm-test-in: Add more tests of tanh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
ldbl-128ibm sinhl uses a too-big threshold to decide when to return
the argument, resulting in large errors. This patch fixes it to use a
more appropriate threshold.
Tested for x86_64, x86 and powerpc.
[BZ #18789]
* sysdeps/ieee754/ldbl-128ibm/e_sinhl.c (__ieee754_sinhl): Use
smaller threshold for returning the argument.
* math/auto-libm-test-in: Add more tests of sinh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
No other arch exports these defines, and having them in the default
namespace causes conformance header tests to fail. Put them behind
the __USE_MISC define as that is what other arches seem to use.
The attached change fixes the miscompilation of sched_setaffinity() on
hppa. This is an old problem that was fixed on other architectures using
a similar approach to the attached change. See:
https://sourceware.org/ml/libc-hacker/2004-04/msg00016.html
Build tested on trunk. Patch has been applied to debian glibc for some time.
As noted in the bug, the asm operands need to be copied to register
variables to avoid operand reloads in the principal asm of the macro.
See the arm implementation for reference. Otherwise we get:
../sysdeps/unix/sysv/linux/hppa/bits/atomic.h:68:6: error:
can't find a register in class 'R1_REGS' while reloading 'asm'
Build tested on trunk with gcc-4.8. Similar patch has been tested
with 2.19 on Debian hppa-unknown-linux-gnu.
The semi-recent SYSCALL_CANCEL inclusion broke microblaze due to the
sysdep.h header not including the unix/sysdep.h header. Include it
here like all other ports.
