This patch updates struct signalfd_siginfo in sys/signalfd.h with new
members from Linux 4.18 (plus ssi_addr_lsb, added to the kernel in
2.6.37 without being added to sys/signalfd.h at that time). The
__pad2 member name follows the kernel and the existing __pad name.
Tested for x86_64.
* sysdeps/unix/sysv/linux/sys/signalfd.h (struct
signalfd_siginfo): Add ssi_addr_lsb, ssi_syscall, ssi_call_addr
and ssi_arch members.
New generic optimization of sinf and cosf introduced by commit
599cf39766 shows improvement
compared to powerpc specific assembly version. Hence removing
the powerpc assembly versions to make use of generic code.
This patch moves little endian specific POWER9 optimization files to
sysdeps/powerpc/powerpc64/le and creates POWER9 ifunc functions
only for little endian.
This variant of strlen uses vector loads and operations to reduce the
size of the code and also eliminate the non-ascii fallback. This
works very well for falkor because of its two vector units and
efficient vector ops. In the best case it reduces latency of cases in
bench-strlen by 48%, with gains throughout the benchmark.
strlen-walk also sees uniform gains in the 5%-15% range.
Overall the routine appears to work better than the stock one for falkor
regardless of the benchmark, length of string or cache state.
The same cannot be said of a53 and a72 though. a53 performance was
greatly reduced and for a72 it was a bit of a mixed bag, slightly on the
negative side but I reckon it might be fast in some situations.
* sysdeps/aarch64/strlen.S (__strlen): Rename to STRLEN.
[!STRLEN](STRLEN): Set to __strlen.
* sysdeps/aarch64/multiarch/strlen.c: New file.
* sysdeps/aarch64/multiarch/strlen_generic.S: Likewise.
* sysdeps/aarch64/multiarch/strlen_asimd.S: Likewise.
* sysdeps/aarch64/multiarch/ifunc-impl-list.c
(__libc_ifunc_impl_list): Add strlen.
* sysdeps/aarch64/multiarch/Makefile (sysdep_routines): Add
strlen_generic and strlen_asimd.
Reviewed-By: szabolcs.nagy@arm.com
CC: pinskia@gmail.com
The internal functions __kernel_sinf and __kernel_cosf are used only by
lgammaf_r. Removing the internal functions and using the generic sinf
and cosf is better overall. Benchmarking on Cortex-A72 shows the generic
sinf and cosf are 1.4x and 2.3x faster in the range |x| < PI/4, and 0.66x
and 1.1x for |x| < PI/2, so it should make lgammaf_r faster on average.
GLIBC regression tests pass on AArch64.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (sin_pif): Use __sinf/__cosf.
* sysdeps/ieee754/flt-32/k_cosf.c (__kernel_cosf): Remove all code.
* sysdeps/ieee754/flt-32/k_sinf.c (__kernel_sinf): Likewise.
Fix a few missing spaces, it's now identical to the regenerated version.
Passes GLIBC tests on x64.
* sysdeps/x86_64/fpu/libm-test-ulps: Regenerate to fix spaces.
The second patch improves performance of sinf and cosf using the same
algorithms and polynomials. The returned values are identical to sincosf
for the same input. ULP definitions for AArch64 and x64 are updated.
sinf/cosf througput gains on Cortex-A72:
* |x| < 0x1p-12 : 1.2x
* |x| < M_PI_4 : 1.8x
* |x| < 2 * M_PI: 1.7x
* |x| < 120.0 : 2.3x
* |x| < Inf : 3.0x
* NEWS: Mention sinf, cosf, sincosf.
* sysdeps/aarch64/libm-test-ulps: Update ULP for sinf, cosf, sincosf.
* sysdeps/x86_64/fpu/libm-test-ulps: Update ULP for sinf and cosf.
* sysdeps/x86_64/fpu/multiarch/s_sincosf-fma.c: Add definitions of
constants rather than including generic sincosf.h.
* sysdeps/x86_64/fpu/s_sincosf_data.c: Remove.
* sysdeps/ieee754/flt-32/s_cosf.c (cosf): Rewrite.
* sysdeps/ieee754/flt-32/s_sincosf.h (reduced_sin): Remove.
(reduced_cos): Remove.
(sinf_poly): New function.
* sysdeps/ieee754/flt-32/s_sinf.c (sinf): Rewrite.
This patch updates sysdeps/unix/sysv/linux/syscall-names.list for
Linux 4.18. The io_pgetevents and rseq syscalls are added to the
kernel on various architectures, so need to be mentioned in this file.
Tested with build-many-glibcs.py.
* sysdeps/unix/sysv/linux/syscall-names.list: Update kernel
version to 4.18.
(io_pgetevents): New syscall.
(rseq): Likewise.
Continuing moving macros out of math-tests.h to smaller headers
following typo-proof conventions instead of using #ifndef, this patch
moves the SNAN_TESTS_* macros for individual types out to their own
sysdeps header (while the type-generic SNAN_TESTS wrapper for those
macros remains in math-tests.h).
Tested for x86_64 and x86, and with build-many-glibcs.py.
* sysdeps/generic/math-tests-snan.h: New file.
* sysdeps/generic/math-tests.h: Include <math-tests-snan.h>.
(SNAN_TESTS_float): Do not define here.
(SNAN_TESTS_double): Likewise.
(SNAN_TESTS_long_double): Likewise.
(SNAN_TESTS_float128): Likewise.
* sysdeps/i386/fpu/math-tests-snan.h: New file.
* sysdeps/i386/fpu/math-tests.h: Remove file.
* sysdeps/ia64/math-tests-snan.h: New file.
* sysdeps/ia64/math-tests.h: Remove file.
* sysdeps/x86/math-tests.h: Likewise.
* sysdeps/x86_64/fpu/math-tests-snan.h: New file.
This patch is a complete rewrite of sincosf. The new version is
significantly faster, as well as simple and accurate.
The worst-case ULP is 0.5607, maximum relative error is 0.5303 * 2^-23 over
all 4 billion inputs. In non-nearest rounding modes the error is 1ULP.
The algorithm uses 3 main cases: small inputs which don't need argument
reduction, small inputs which need a simple range reduction and large inputs
requiring complex range reduction. The code uses approximate integer
comparisons to quickly decide between these cases.
The small range reducer uses a single reduction step to handle values up to
120.0. It is fastest on targets which support inlined round instructions.
The large range reducer uses integer arithmetic for simplicity. It does a
32x96 bit multiply to compute a 64-bit modulo result. This is more than
accurate enough to handle the worst-case cancellation for values close to
an integer multiple of PI/4. It could be further optimized, however it is
already much faster than necessary.
sincosf throughput gains on Cortex-A72:
* |x| < 0x1p-12 : 1.6x
* |x| < M_PI_4 : 1.7x
* |x| < 2 * M_PI: 1.5x
* |x| < 120.0 : 1.8x
* |x| < Inf : 2.3x
* math/Makefile: Add s_sincosf_data.c.
* sysdeps/ia64/fpu/s_sincosf_data.c: New file.
* sysdeps/ieee754/flt-32/s_sincosf.h (abstop12): Add new function.
(sincosf_poly): Likewise.
(reduce_small): Likewise.
(reduce_large): Likewise.
* sysdeps/ieee754/flt-32/s_sincosf.c (sincosf): Rewrite.
* sysdeps/ieee754/flt-32/s_sincosf_data.c: New file with sincosf data.
* sysdeps/m68k/m680x0/fpu/s_sincosf_data.c: New file.
* sysdeps/x86_64/fpu/s_sincosf_data.c: New file.
This patch currently only affects aarch64.
The roundtoint and converttoint internal functions are only called with small
values, so 32 bit result is enough for converttoint and it is a signed int
conversion so the return type is changed to int32_t.
The original idea was to help the compiler keeping the result in uint64_t,
then it's clear that no sign extension is needed and there is no accidental
undefined or implementation defined signed int arithmetics.
But it turns out gcc does a good job with inlining so changing the type has
no overhead and the semantics of the conversion is less surprising this way.
Since we want to allow the asuint64 (x + 0x1.8p52) style conversion, the top
bits were never usable and the existing code ensures that only the bottom
32 bits of the conversion result are used.
On aarch64 the neon intrinsics (which round ties to even) are changed to
round and lround (which round ties away from zero) this does not affect the
results in a significant way, but more portable (relies on round and lround
being inlined which works with -fno-math-errno).
The TOINT_SHIFT and TOINT_RINT macros were removed, only keep separate code
paths for TOINT_INTRINSICS and !TOINT_INTRINSICS.
* sysdeps/aarch64/fpu/math_private.h (roundtoint): Use round.
(converttoint): Use lround.
* sysdeps/ieee754/flt-32/math_config.h (roundtoint): Declare and
document the semantics when TOINT_INTRINSICS is set.
(converttoint): Likewise.
(TOINT_RINT): Remove.
(TOINT_SHIFT): Remove.
* sysdeps/ieee754/flt-32/e_expf.c (__expf): Remove the TOINT_RINT code
path.
Commit 298d0e3129 ("Consolidate Linux
getdents{64} implementation") broke the implementation because it does
not take into account struct offset differences.
The new implementation is close to the old one, before the
consolidation, but has been cleaned up slightly.
* Since __fentry__ is almost the same as _mcount, reuse the code by
#including it twice with different #defines around.
* Remove LA usages - they are needed in 31-bit mode to clear the top
bit, but in 64-bit they appear to do nothing.
* Add CFI rule for the nonstandard return register. This rule applies
to the current function (binutils generates a new CIE - see
gas/dw2gencfi.c:select_cie_for_fde()), so it is not necessary to put
__fentry__ into a new file.
* Fix CFI offset for %r14.
* Add CFI rule for %r0.
* Fix unwound value of %r15 being off by 244 bytes.
* Unwinding in __fentry__@plt does not work, no plan to fix it - it
would require asking linker to generate CFI for return address in
%r0. From functional perspective keeping it broken is fine, since
the callee did not have a chance to do anything yet. From
convenience perspective it would be possible to enhance GDB in the
future to treat __fentry__@plt in a special way.
* Fix whitespace.
* Fix offsets in comments, which were copied from 32-bit code.
* 32-bit version will not be implemented, since it's not compatible
with the corresponding PLT stubs: they assume %r12 points to GOT,
which is not the case for gcc-emitted __fentry__ stub, which runs
before the prolog.
This patch adds the runtime support in glibc for the -mfentry
gcc feature introduced in [1] and [2].
[1] https://gcc.gnu.org/ml/gcc-patches/2018-07/msg00784.html
[2] https://gcc.gnu.org/ml/gcc-patches/2018-07/msg00912.html
ChangeLog:
* sysdeps/s390/s390-64/Versions (__fentry__): Add.
* sysdeps/s390/s390-64/s390x-mcount.S: Move the common
code to s390x-mcount.h and #include it.
* sysdeps/s390/s390-64/s390x-mcount.h: New file.
* sysdeps/unix/sysv/linux/s390/s390-64/libc.abilist
(__fentry__): Add.
__fentry__ symbol is currently not defined for other architectures.
Attempts to introduce it cause abicheck to fail, because it will be
available since 2.29 earliest, and not 2.13, which is the case for
Intel. With the new code, abicheck passes for i686-linux-gnu,
x86_64-linux-gnu and x86_64-linux-gnu32 triples.
ChangeLog:
* stdlib/Versions: Remove __fentry__.
* sysdeps/i386/Versions: Add __fentry__.
* sysdeps/x86_64/Versions: Add __fentry__.
The following combinations need to be tested:
* 32- (g5, esa and zarch) and 64-bit
* linux32 glibc/configure CC='gcc -m31 -march=g5'
* linux32 glibc/configure CC='gcc -m31'
* linux32 glibc/configure CC='gcc -m31 -mzarch'
* With and without VX:
* glibc/configure libc_cv_asm_s390_vx=no
* With and without profiling (using LD_PROFILE)
* With and without pltexit (using LD_AUDIT)
ChangeLog:
* sysdeps/s390/Makefile: Register the new tests.
* sysdeps/s390/tst-dl-runtime-mod.S: New file.
* sysdeps/s390/tst-dl-runtime-profile-audit.c: New file.
* sysdeps/s390/tst-dl-runtime-profile-noaudit.c: New file.
* sysdeps/s390/tst-dl-runtime-resolve-audit.c: New file.
* sysdeps/s390/tst-dl-runtime-resolve-noaudit.c: New file.
* sysdeps/s390/tst-dl-runtime.c: New file.
Following the recent discussion of using Python instead of Perl and
Awk for glibc build / test, this patch replaces gen-libm-test.pl with
a new gen-libm-test.py script. This script should work with all
Python versions supported by glibc (tested by hand with Python 2.7,
tested in the build system with Python 3.5; configure prefers Python 3
if available).
This script is designed to give identical output to gen-libm-test.pl
for ease of verification of the change, except for generated comments
referring to .py instead of .pl. (That is, identical for actual
inputs passed to the script, not necessarily for all possible input;
for example, this version more precisely follows the C standard syntax
for floating-point constants when deciding when to add LIT macro
calls.) In one place a comment notes that the generation of
NON_FINITE flags is replicating a bug in the Perl script to assist in
such comparisons (with the expectation that this bug can then be
separately fixed in the Python script later).
Tested for x86_64, including comparison of generated files (and hand
testing of the case of generating a sorted libm-test-ulps file, which
isn't covered by normal "make check").
I'd expect to follow this up by extending the new script to produce
the ulps tables for the manual as well (replacing
manual/libm-err-tab.pl, so that then we just have one ulps file
parser) - at which point the manual build would depend on both Perl
and Python (eliminating the Perl dependency would require someone to
rewrite summary.pl in Python, and that would only eliminate the
*direct* Perl dependency; current makeinfo is written in Perl so there
would still be an indirect dependency).
I think install.texi is more or less equally out-of-date regarding
Perl and Python uses before and after this patch, so I don't think
this patch depends on my patch
<https://sourceware.org/ml/libc-alpha/2018-08/msg00133.html> to update
install.texi regarding such uses (pending review).
* math/gen-libm-test.py: New file.
* math/gen-libm-test.pl: Remove.
* math/Makefile [$(PERL) != no]: Change condition to [PYTHON].
($(objpfx)libm-test-ulps.h): Use gen-libm-test.py instead of
gen-libm-test.pl.
($(libm-test-c-noauto-obj)): Likewise.
($(libm-test-c-auto-obj)): Likewise.
($(libm-test-c-narrow-obj)): Likewise.
(regen-ulps): Likewise.
* math/README.libm-test: Update references to gen-libm-test.pl.
* math/libm-test-driver.c (struct test_fj_f_data): Update comment
referencing gen-libm-test.pl.
* math/libm-test-nexttoward.inc (nexttoward_test_data): Likewise.
* math/libm-test-support.c: Likewise.
* math/libm-test-support.h: Likewise.
* sysdeps/generic/libm-test-ulps: Likewise.
MIN_PAGE_SIZE is normally set to 4096 but for testing it can be set to
16 so that it exercises the page crossing code for every misaligned
access. The value was set to 15, which is obviously wrong, so fixed
as obvious and tested.
* sysdeps/aarch64/strlen.S [TEST_PAGE_CROSS](MIN_PAGE_SIZE):
Fix value.
Since RISC-V stores the thread pointer in a general register libthread_db
can just ask the debugger for the register contents instead of trying to
call ps_get_thread_area. This enables thread debugging in gdb.
* sysdeps/riscv/nptl/tls.h (DB_THREAD_SELF): Use REGISTER instead
of CONST_THREAD_AREA.
Move STATE_SAVE_OFFSET and STATE_SAVE_MASK to sysdep.h to make
sysdeps/x86/cpu-features.h a C header file.
* sysdeps/x86/cpu-features.h (STATE_SAVE_OFFSET): Removed.
(STATE_SAVE_MASK): Likewise.
Don't check __ASSEMBLER__ to include <cpu-features-offsets.h>.
* sysdeps/x86/sysdep.h (STATE_SAVE_OFFSET): New.
(STATE_SAVE_MASK): Likewise.
* sysdeps/x86_64/dl-trampoline.S: Include <cpu-features-offsets.h>
instead of <cpu-features.h>.
* sysdeps/riscv/rv64/rvd/libm-test-ulps: Update.
Note: I regen'd these from scratch, but I'm only committing the
increases, as I only tested on hardware. There were a few 2->1
decreases that I omitted, possibly "for now".
* sysdeps/riscv/rvf/math_private.h (libc_feholdexcept_setround_riscv):
Fix rounding save-restore bug.
Fixes about a hundred off-by-ULP failures in the math testsuite.
There is no need to include <init-arch.h> in assembly codes since all
x86 IFUNC selector functions are written in C. Tested on i686 and
x86-64. There is no code change in libc.so, ld.so and libmvec.so.
* sysdeps/i386/i686/multiarch/bzero-ia32.S: Don't include
<init-arch.h>.
* sysdeps/x86_64/fpu/multiarch/svml_d_sin8_core-avx2.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/svml_s_expf16_core-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S: Likewise.
The glibc.tune namespace is vaguely named since it is a 'tunable', so
give it a more specific name that describes what it refers to. Rename
the tunable namespace to 'cpu' to more accurately reflect what it
encompasses. Also rename glibc.tune.cpu to glibc.cpu.name since
glibc.cpu.cpu is weird.
* NEWS: Mention the change.
* elf/dl-tunables.list: Rename tune namespace to cpu.
* sysdeps/powerpc/dl-tunables.list: Likewise.
* sysdeps/x86/dl-tunables.list: Likewise.
* sysdeps/aarch64/dl-tunables.list: Rename tune.cpu to
cpu.name.
* elf/dl-hwcaps.c (_dl_important_hwcaps): Adjust.
* elf/dl-hwcaps.h (GET_HWCAP_MASK): Likewise.
* manual/README.tunables: Likewise.
* manual/tunables.texi: Likewise.
* sysdeps/powerpc/cpu-features.c: Likewise.
* sysdeps/unix/sysv/linux/aarch64/cpu-features.c
(init_cpu_features): Likewise.
* sysdeps/x86/cpu-features.c: Likewise.
* sysdeps/x86/cpu-features.h: Likewise.
* sysdeps/x86/cpu-tunables.c: Likewise.
* sysdeps/x86_64/Makefile: Likewise.
* sysdeps/x86/dl-cet.c: Likewise.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
MIPS soft-float glibc does not support floating-point exceptions and
rounding modes, and uses a different ABI from hard-float so a
soft-float compilation cannot use a glibc that does support
floating-point exceptions and rounding modes. Thus, bits/fenv.h
should not, when compiling for soft-float, define macros for the
unsupported features.
This patch changes it accordingly to define those macros only for
hard-float. None of the exception macros are defined for soft-float,
with FE_ALL_EXCEPT defined to 0 in that case, and only FE_TONEAREST is
defined of the rounding-mode macros, and FE_NOMASK_ENV is not defined;
this is consistent with how architectures lacking exception and
rounding mode support generally define things in this header. As well
as making the header more correct for this case, this also means the
generic math_private.h optimizations for this case automatically apply
(inlining libm-internal fenv.h function calls that are trivial when
exceptions and rounding modes are not supported).
The mips64 sfp-machine.h then needs similar changes to disable more of
the exception and rounding mode handling for soft-float. (The mips32
sfp-machine.h is already used only for soft-float, has no integration
with hardware exceptions or rounding modes and so needs no changes.)
Existing binaries might use the old FE_NOMASK_ENV value as an argument
to fesetenv / feupdateenv and expect an error for it (given that it
was defined in a header that also defined FE_ALL_EXCEPT to a nonzero
value). To preserve that error, wrappers for the fallback fesetenv
and feupdateenv are created in sysdeps/mips/nofpu/.
Tested for mips64 (hard-float and soft-float, all three ABIs).
[BZ #23479]
* sysdeps/mips/bits/fenv.h (FE_INEXACT): Define only if
[__mips_hard_float].
(FE_UNDERFLOW): Likewise.
(FE_OVERFLOW): Likewise.
(FE_DIVBYZERO): Likewise.
(FE_INVALID): Likewise.
(FE_ALL_EXCEPT): Define to 0 if [!__mips_hard_float].
(FE_TOWARDZERO): Define only if [__mips_hard_float].
(FE_UPWARD): Likewise.
(FE_DOWNWARD): Likewise.
(__FE_UNDEFINED): Define if [!__mips_hard_float]
(FE_NOMASK_ENV): Define only if [__mips_hard_float].
* sysdeps/mips/mips64/sfp-machine.h (_FP_DECL_EX): Define only if
[__mips_hard_float].
(FP_ROUNDMODE): Likewise.
(FP_RND_NEAREST): Likewise.
(FP_RND_ZERO): Likewise.
(FP_RND_PINF): Likewise.
(FP_RND_MINF): Likewise.
(FP_EX_INVALID): Likewise.
(FP_EX_OVERFLOW): Likewise.
(FP_EX_UNDERFLOW): Likewise.
(FP_EX_DIVZERO): Likewise.
(FP_EX_INEXACT): Likewise.
(FP_INIT_ROUNDMODE): Likewise.
* sysdeps/mips/nofpu/fesetenv.c: New file.
* sysdeps/mips/nofpu/feupdateenv.c: Likewise.
Continuing moving macros out of math-tests.h to smaller headers
following typo-proof conventions instead of using #ifndef, this patch
moves SNAN_TESTS_PRESERVE_PAYLOAD out to its own sysdeps header.
Tested with build-many-glibcs.py.
* sysdeps/generic/math-tests-snan-payload.h: New file.
* sysdeps/hppa/math-tests-snan-payload.h: Likewise.
* sysdeps/mips/math-tests-snan-payload.h: Likewise.
* sysdeps/riscv/math-tests-snan-payload.h: Likewise.
* sysdeps/generic/math-tests.h: Include
<math-tests-snan-payload.h>.
(SNAN_TESTS_PRESERVE_PAYLOAD): Do not define macro here.
* sysdeps/hppa/math-tests.h: Remove file.
* sysdeps/mips/math-tests.h [!__mips_nan2008]
(SNAN_TESTS_PRESERVE_PAYLOAD): Do not define macro here.
* sysdeps/riscv/math-tests.h (SNAN_TESTS_PRESERVE_PAYLOAD):
Likewise.
The math-tests.h header has many different macros and groups of
macros, defined using #ifndef in the generic version which is included
by architecture versions with #include_next after possibly defining
non-default versions of some of those macros.
This use of #ifndef is contrary to our normal typo-proof conventions
for macro definitions. This patch moves one of the macros,
SNAN_TESTS_TYPE_CAST, out to its own sysdeps header, to follow those
typo-proof conventions more closely.
Tested with build-many-glibcs.py.
2018-08-01 Joseph Myers <joseph@codesourcery.com>
* sysdeps/generic/math-tests-snan-cast.h: New file.
* sysdeps/powerpc/math-tests-snan-cast.h: Likewise.
* sysdeps/generic/math-tests.h: Include <math-tests-snan-cast.h>.
(SNAN_TESTS_TYPE_CAST): Do not define macro here.
* sysdeps/powerpc/math-tests.h (SNAN_TESTS_TYPE_CAST): Likewise.
Exec needs that mach_setup_thread does *not* set up TLS since it works on
another task, so we have to split this into mach_setup_tls.
* mach/mach.h (__mach_setup_tls, mach_setup_tls): Add prototypes.
* mach/setup-thread.c (__mach_setup_thread): Move TLS setup to...
(__mach_setup_tls): ... new function.
(mach_setup_tls): New alias.
* hurd/hurdsig.c (_hurdsig_init): Call __mach_setup_tls after
__mach_setup_thread.
* sysdeps/mach/hurd/profil.c (update_waiter): Likewise.
* sysdeps/mach/hurd/setitimer.c (setitimer_locked): Likewise.
* mach/Versions [libc] (mach_setup_tls): Add symbol.
* sysdeps/mach/hurd/i386/libc.abilist (mach_setup_tls): Likewise.
GNU_PROPERTY_X86_FEATURE_1_AND may not be the first property item. We
need to check each property item until we reach the end of the property
or find GNU_PROPERTY_X86_FEATURE_1_AND.
This patch adds 2 tests. The first test checks if IBT is enabled and
the second test reads the output from the first test to check if IBT
is is enabled. The second second test fails if IBT isn't enabled
properly.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
[BZ #23467]
* sysdeps/unix/sysv/linux/x86/Makefile (tests): Add
tst-cet-property-1 and tst-cet-property-2 if CET is enabled.
(CFLAGS-tst-cet-property-1.o): New.
(ASFLAGS-tst-cet-property-dep-2.o): Likewise.
($(objpfx)tst-cet-property-2): Likewise.
($(objpfx)tst-cet-property-2.out): Likewise.
* sysdeps/unix/sysv/linux/x86/tst-cet-property-1.c: New file.
* sysdeps/unix/sysv/linux/x86/tst-cet-property-2.c: Likewise.
* sysdeps/unix/sysv/linux/x86/tst-cet-property-dep-2.S: Likewise.
* sysdeps/x86/dl-prop.h (_dl_process_cet_property_note): Parse
each property item until GNU_PROPERTY_X86_FEATURE_1_AND is found.
All tests should be added to $(tests).
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
[BZ #23458]
* sysdeps/x86/Makefile (tests): Add tst-get-cpu-features-static.
ld.so symbols to be overriden by libc need to be extern to really get
overriden.
* sysdeps/mach/hurd/not-errno.h: New file.
* sysdeps/mach/hurd/i386/localplt.data: Update accordingly.
ld.so symbols to be overriden by libc need to be extern to really get
overriden.
* sysdeps/mach/hurd/dl-unistd.h (__access, __brk, __lseek, __read,
__sbrk): Do not set attribute_hidden.
* sysdeps/mach/hurd/i386/ld.abilist: Update accordingly.
* sysdeps/mach/hurd/i386/localplt.data: Update accordingly.
Simply check if "ptr < ptr_end" since "ptr" is always incremented by 8.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/x86/dl-prop.h (_dl_process_cet_property_note): Don't
parse beyond the note end.
This patch make the OFD tests return unsupported if kernel does not
support OFD locks (it was added on 3.15).
Checked on a ia64-linux-gnu with Linux 3.14.
* sysdeps/unix/sysv/linux/tst-ofdlocks.c: Return unsupported if
kernel does not support OFD locks.
* sysdeps/unix/sysv/linux/tst-ofdlocks-compat.c: Likewise.
ld.so symbols to be overriden by libc need to be extern to really get
overriden.
More fixes are needed to avoid the hidden attribute.
* sysdeps/mach/hurd/Versions (libc): Make __access and
__access_noerrno external so they can override the ld symbols.
(ld): Make __access, __read, __sbrk, __strtoul_internal, __write,
__writev, __open64, __access_noerrno extern so they can be overrided.
* sysdeps/mach/hurd/i386/libc.abilist: Update accordingly.
* sysdeps/mach/hurd/i386/ld.abilist: Update accordingly.
cpu-features.h has
#define bit_cpu_LZCNT (1 << 5)
#define index_cpu_LZCNT COMMON_CPUID_INDEX_1
#define reg_LZCNT
But the LZCNT feature bit is in COMMON_CPUID_INDEX_80000001:
Initial EAX Value: 80000001H
ECX Extended Processor Signature and Feature Bits:
Bit 05: LZCNT available
index_cpu_LZCNT should be COMMON_CPUID_INDEX_80000001, not
COMMON_CPUID_INDEX_1. The VMX feature bit is in COMMON_CPUID_INDEX_1:
Initial EAX Value: 01H
Feature Information Returned in the ECX Register:
5 VMX
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
[BZ # 23456]
* sysdeps/x86/cpu-features.h (index_cpu_LZCNT): Set to
COMMON_CPUID_INDEX_80000001.
Set installed NPTL header as the expected one (instead of an
internal one for glibc testsuite) and add a hurd specific
stdc-predef with __STDC_NO_THREADS__.
Checked on both i686-linux-gnu and i686-gnu that both threads.h
and stdc-predef.h are the expected ones.
* nptl/threads.h: Move to ...
* sysdeps/nptl/threads.h: ... here.
* sysdeps/hurd/stdc-predef.h: New file.
Verify that setcontext works with gaps above and below the newly
allocated shadow stack.
* sysdeps/unix/sysv/linux/x86/Makefile (tests): Add
tst-cet-setcontext-1 if CET is enabled.
(CFLAGS-tst-cet-setcontext-1.c): Add -mshstk.
* sysdeps/unix/sysv/linux/x86/tst-cet-setcontext-1.c: New file.
Remove conformace assumption of NPTL implementation for ISO C threads
and revert wrong libcrypt addition on linknamespace-libs-XPG4.
The i686-gnu target now shows two new conformance failures:
FAIL: conform/ISO11/threads.h/conform
FAIL: conform/ISO11/threads.h/linknamespace
It is expected due missing HTL ISO C threads support and both conformance
.out files indicates the reason ("#error "HTL does not implement ISO C
threads").
Checked on i686-linux-gnu and i686-gnu.
* include/threads.h: Move to ...
* sysdeps/nptl/threads.h: ... here.
* sysdeps/htl/threads.h: New file.
* conform/Makefile (linknamespace-libs-ISO11): Use
static-thread-library instead of linking libpthread.
(linknamespace-libs-XPG4): Revert wrong libcrypt.a addition.
This patch adds a field to ucontext_t to save shadow stack:
1. getcontext and swapcontext are updated to save the caller's shadow
stack pointer and return addresses.
2. setcontext and swapcontext are updated to restore shadow stack and
jump to new context directly.
3. makecontext is updated to allocate a new shadow stack and set the
caller's return address to __start_context.
Since makecontext allocates a new shadow stack when making a new
context and kernel allocates a new shadow stack for clone/fork/vfork
syscalls, we track the current shadow stack base. In setcontext and
swapcontext, if the target shadow stack base is the same as the current
shadow stack base, we unwind the shadow stack. Otherwise it is a stack
switch and we look for a restore token.
We enable shadow stack at run-time only if program and all used shared
objects, including dlopened ones, are shadow stack enabled, which means
that they must be compiled with GCC 8 or above and glibc 2.28 or above.
We need to save and restore shadow stack only if shadow stack is enabled.
When caller of getcontext, setcontext, swapcontext and makecontext is
compiled with smaller ucontext_t, shadow stack won't be enabled at
run-time. We check if shadow stack is enabled before accessing the
extended field in ucontext_t.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/x86/sys/ucontext.h (ucontext_t): Add
__ssp.
* sysdeps/unix/sysv/linux/x86_64/__start_context.S: Include
<asm/prctl.h> and "ucontext_i.h" when shadow stack is enabled.
(__push___start_context): New.
* sysdeps/unix/sysv/linux/x86_64/getcontext.S: Include
<asm/prctl.h>.
(__getcontext): Record the current shadow stack base. Save the
caller's shadow stack pointer and base.
* sysdeps/unix/sysv/linux/x86_64/makecontext.c: Include
<pthread.h>, <libc-pointer-arith.h> and <sys/prctl.h>.
(__push___start_context): New prototype.
(__makecontext): Call __push___start_context to allocate a new
shadow stack, push __start_context onto the new stack as well
as the new shadow stack.
* sysdeps/unix/sysv/linux/x86_64/setcontext.S: Include
<asm/prctl.h>.
(__setcontext): Restore the target shadow stack.
* sysdeps/unix/sysv/linux/x86_64/swapcontext.S: Include
<asm/prctl.h>.
(__swapcontext): Record the current shadow stack base. Save
the caller's shadow stack pointer and base. Restore the target
shadow stack.
* sysdeps/unix/sysv/linux/x86_64/sysdep.h
(STACK_SIZE_TO_SHADOW_STACK_SIZE_SHIFT): New.
* sysdeps/unix/sysv/linux/x86_64/ucontext_i.sym (oSSP): New.
This will be used to record the current shadow stack base for shadow
stack switching by getcontext, makecontext, setcontext and swapcontext.
If the target shadow stack base is the same as the current shadow stack
base, we unwind the shadow stack. Otherwise it is a stack switch and
we look for a restore token to restore the target shadow stack.
* sysdeps/i386/nptl/tcb-offsets.sym (SSP_BASE_OFFSET): New.
* sysdeps/i386/nptl/tls.h (tcbhead_t): Replace __glibc_reserved2
with ssp_base.
* sysdeps/x86_64/nptl/tcb-offsets.sym (SSP_BASE_OFFSET): New.
* sysdeps/x86_64/nptl/tls.h (tcbhead_t): Replace __glibc_reserved2
with ssp_base.
CET arch_prctl bits should be defined in <asm/prctl.h> from Linux kernel
header files. Add x86 <include/asm/prctl.h> for pre-CET kernel header
files.
Note: sysdeps/unix/sysv/linux/x86/include/asm/prctl.h should be removed
if <asm/prctl.h> from the required kernel header files contains CET
arch_prctl bits.
/* CET features:
IBT: GNU_PROPERTY_X86_FEATURE_1_IBT
SHSTK: GNU_PROPERTY_X86_FEATURE_1_SHSTK
*/
/* Return CET features in unsigned long long *addr:
features: addr[0].
shadow stack base address: addr[1].
shadow stack size: addr[2].
*/
# define ARCH_CET_STATUS 0x3001
/* Disable CET features in unsigned int features. */
# define ARCH_CET_DISABLE 0x3002
/* Lock all CET features. */
# define ARCH_CET_LOCK 0x3003
/* Allocate a new shadow stack with unsigned long long *addr:
IN: requested shadow stack size: *addr.
OUT: allocated shadow stack address: *addr.
*/
# define ARCH_CET_ALLOC_SHSTK 0x3004
/* Return legacy region bitmap info in unsigned long long *addr:
address: addr[0].
size: addr[1].
*/
# define ARCH_CET_LEGACY_BITMAP 0x3005
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/x86/include/asm/prctl.h: New file.
* sysdeps/unix/sysv/linux/x86/cpu-features.c: Include
<sys/prctl.h> and <asm/prctl.h>.
(get_cet_status): Call arch_prctl with ARCH_CET_STATUS.
* sysdeps/unix/sysv/linux/x86/dl-cet.h: Include <sys/prctl.h>
and <asm/prctl.h>.
(dl_cet_allocate_legacy_bitmap): Call arch_prctl with
ARCH_CET_LEGACY_BITMAP.
(dl_cet_disable_cet): Call arch_prctl with ARCH_CET_DISABLE.
(dl_cet_lock_cet): Call arch_prctl with ARCH_CET_LOCK.
* sysdeps/x86/libc-start.c: Include <startup.h>.
This patch adds the thrd_* definitions from C11 threads (ISO/IEC 9899:2011),
more specifically thrd_create, thrd_curent, rhd_detach, thrd_equal,
thrd_exit, thrd_join, thrd_sleep, thrd_yield, and required types.
Mostly of the definitions are composed based on POSIX conterparts, such as
thrd_t (using pthread_t). For thrd_* function internally direct
POSIX pthread call are used with the exceptions:
1. thrd_start uses pthread_create internal implementation, but changes
how to actually calls the start routine. This is due the difference
in signature between POSIX and C11, where former return a 'void *'
and latter 'int'.
To avoid calling convention issues due 'void *' to int cast, routines
from C11 threads are started slight different than default pthread one.
Explicit cast to expected return are used internally on pthread_create
and the result is stored back to void also with an explicit cast.
2. thrd_sleep uses nanosleep internal direct syscall to avoid clobbering
errno and to handle expected standard return codes. It is a
cancellation entrypoint to be consistent with both thrd_join and
cnd_{timed}wait.
3. thrd_yield also uses internal direct syscall to avoid errno clobbering.
Checked with a build for all major ABI (aarch64-linux-gnu, alpha-linux-gnu,
arm-linux-gnueabi, i386-linux-gnu, ia64-linux-gnu, m68k-linux-gnu,
microblaze-linux-gnu [1], mips{64}-linux-gnu, nios2-linux-gnu,
powerpc{64le}-linux-gnu, s390{x}-linux-gnu, sparc{64}-linux-gnu,
and x86_64-linux-gnu).
Also ran a full check on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
arm-linux-gnueabhf, and powerpc64le-linux-gnu.
[BZ #14092]
* conform/Makefile (conformtest-headers-ISO11): Add threads.h.
(linknamespace-libs-ISO11): Add libpthread.a.
* conform/data/threads.h-data: New file: add C11 thrd_* types and
functions.
* include/stdc-predef.h (__STDC_NO_THREADS__): Remove definition.
* nptl/Makefile (headers): Add threads.h.
(libpthread-routines): Add new C11 thread thrd_create, thrd_current,
thrd_detach, thrd_equal, thrd_exit, thrd_join, thrd_sleep, and
thrd_yield.
* nptl/Versions (libpthread) [GLIBC_2.28]): Add new C11 thread
thrd_create, thrd_current, thrd_detach, thrd_equal, thrd_exit,
thrd_join, thrd_sleep, and thrd_yield symbols.
* nptl/descr.h (struct pthread): Add c11 field.
* nptl/pthreadP.h (ATTR_C11_THREAD): New define.
* nptl/pthread_create.c (START_THREAD_DEFN): Call C11 thread start
routine with expected function prototype.
(__pthread_create_2_1): Add C11 threads check based on attribute
value.
* sysdeps/unix/sysdep.h (INTERNAL_SYSCALL_CANCEL): New macro.
* nptl/thrd_create.c: New file.
* nptl/thrd_current.c: Likewise.
* nptl/thrd_detach.c: Likewise.
* nptl/thrd_equal.c: Likewise.
* nptl/thrd_exit.c: Likewise.
* nptl/thrd_join.c: Likewise.
* nptl/thrd_priv.h: Likewise.
* nptl/thrd_sleep.c: Likewise.
* nptl/thrd_yield.c: Likewise.
* include/threads.h: Likewise.
Add <bits/indirect-return.h> and include it in <ucontext.h>.
__INDIRECT_RETURN defined in <bits/indirect-return.h> indicates if
swapcontext requires special compiler treatment. The default
__INDIRECT_RETURN is empty.
On x86, when shadow stack is enabled, __INDIRECT_RETURN is defined
with indirect_return attribute, which has been added to GCC 9, to
indicate that swapcontext returns via indirect branch. Otherwise
__INDIRECT_RETURN is defined with returns_twice attribute.
When shadow stack is enabled, remove always_inline attribute from
prepare_test_buffer in string/tst-xbzero-opt.c to avoid:
tst-xbzero-opt.c: In function ‘prepare_test_buffer’:
tst-xbzero-opt.c:105:1: error: function ‘prepare_test_buffer’ can never be inlined because it uses setjmp
prepare_test_buffer (unsigned char *buf)
when indirect_return attribute isn't available.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* bits/indirect-return.h: New file.
* misc/sys/cdefs.h (__glibc_has_attribute): New.
* sysdeps/x86/bits/indirect-return.h: Likewise.
* stdlib/Makefile (headers): Add bits/indirect-return.h.
* stdlib/ucontext.h: Include <bits/indirect-return.h>.
(swapcontext): Add __INDIRECT_RETURN.
* string/tst-xbzero-opt.c (ALWAYS_INLINE): New.
(prepare_test_buffer): Use it.
The shadow stack prevents us from pushing the saved return PC onto
the stack and returning normally. Instead we pop the shadow stack
and return directly. This is the safest way to return and ensures
any stack manipulations done by the vfork'd child doesn't cause the
parent to terminate when CET is enabled.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/i386/vfork.S (SYSCALL_ERROR_HANDLER):
Redefine if shadow stack is enabled.
(SYSCALL_ERROR_LABEL): Likewise.
(__vfork): Pop shadow stack and jump back to to caller directly
when shadow stack is in use.
* sysdeps/unix/sysv/linux/x86_64/vfork.S (SYSCALL_ERROR_HANDLER):
Redefine if shadow stack is enabled.
(SYSCALL_ERROR_LABEL): Likewise.
(__vfork): Pop shadow stack and jump back to to caller directly
when shadow stack is in use.
Add endbr64 to tst-quadmod1.S and tst-quadmod2.S so that func and foo
can be called indirectly.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/x86_64/tst-quadmod1.S (func): Add endbr64 if IBT is
enabled.
(foo): Likewise.
* sysdeps/x86_64/tst-quadmod2.S (func) : Likewise.
(foo): Likewise.
* scripts/check-execstack.awk: Consider `xfail' variable containing a
list
of libraries whose stack executability is expected.
* elf/Makefile ($(objpfx)check-execstack.out): Pass
$(check-execstack-xfail) to check-execstack.awk through `xfail'
variable.
* sysdeps/mach/hurd/i386/Makefile (check-execstack-xfail): Set to ld.so
libc.so libpthread.so.
* sysdeps/mach/hurd/pipe2.c: New file, copy from pipe.c. Evolve it to
implement __pipe2.
* sysdeps/mach/hurd/pipe.c (__pipe): Reimplement using __pipe2.
i386 add_n.S and sub_n.S use a trick to implment jump tables with LEA.
We can't use conditional branches nor normal jump tables since jump
table entries use EFLAGS set by jump table index. This patch adds
_CET_ENDBR to indirect jump targets and adjust destination for
_CET_ENDBR.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/i386/add_n.S: Include <sysdep.h>, instead of
"sysdep.h".
(__mpn_add_n): Save and restore %ebx if IBT is enabed. Add
_CET_ENDBR to indirect jump targets and adjust jump destination
for _CET_ENDBR.
* sysdeps/i386/i686/add_n.S: Include <sysdep.h>, instead of
"sysdep.h".
(__mpn_add_n): Save and restore %ebx if IBT is enabed. Add
_CET_ENDBR to indirect jump targets and adjust jump destination
for _CET_ENDBR.
* sysdeps/i386/sub_n.S: Include <sysdep.h>, instead of
"sysdep.h".
(__mpn_sub_n): Save and restore %ebx if IBT is enabed. Add
_CET_ENDBR to indirect jump targets and adjust jump destination
for _CET_ENDBR.
Add _CET_ENDBR to STRCMP_SSE42, which is called indirectly, to support
IBT.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/x86_64/multiarch/strcmp-sse42.S (STRCMP_SSE42): Add
_CET_ENDBR.
Add _CET_ENDBR to functions in crti.S, which are called indirectly, to
support IBT.
Tested on i686 and x86-64.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/i386/crti.S (_init): Add _CET_ENDBR.
(_fini): Likewise.
* sysdeps/x86_64/crti.S (_init): Likewise.
(_fini): Likewise.
Always include <dl-cet.h> and cet-tunables.h> when CET is enabled.
Otherwise, configure glibc with --enable-cet --disable-tunables will
fail to build.
* sysdeps/x86/cpu-features.c: Always include <dl-cet.h> and
cet-tunables.h> when CET is enabled.
Intel Control-flow Enforcement Technology (CET) instructions:
https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-en
forcement-technology-preview.pdf
includes Indirect Branch Tracking (IBT) and Shadow Stack (SHSTK).
GNU_PROPERTY_X86_FEATURE_1_IBT is added to GNU program property to
indicate that all executable sections are compatible with IBT when
ENDBR instruction starts each valid target where an indirect branch
instruction can land. Linker sets GNU_PROPERTY_X86_FEATURE_1_IBT on
output only if it is set on all relocatable inputs.
On an IBT capable processor, the following steps should be taken:
1. When loading an executable without an interpreter, enable IBT and
lock IBT if GNU_PROPERTY_X86_FEATURE_1_IBT is set on the executable.
2. When loading an executable with an interpreter, enable IBT if
GNU_PROPERTY_X86_FEATURE_1_IBT is set on the interpreter.
a. If GNU_PROPERTY_X86_FEATURE_1_IBT isn't set on the executable,
disable IBT.
b. Lock IBT.
3. If IBT is enabled, when loading a shared object without
GNU_PROPERTY_X86_FEATURE_1_IBT:
a. If legacy interwork is allowed, then mark all pages in executable
PT_LOAD segments in legacy code page bitmap. Failure of legacy code
page bitmap allocation causes an error.
b. If legacy interwork isn't allowed, it causes an error.
GNU_PROPERTY_X86_FEATURE_1_SHSTK is added to GNU program property to
indicate that all executable sections are compatible with SHSTK where
return address popped from shadow stack always matches return address
popped from normal stack. Linker sets GNU_PROPERTY_X86_FEATURE_1_SHSTK
on output only if it is set on all relocatable inputs.
On a SHSTK capable processor, the following steps should be taken:
1. When loading an executable without an interpreter, enable SHSTK if
GNU_PROPERTY_X86_FEATURE_1_SHSTK is set on the executable.
2. When loading an executable with an interpreter, enable SHSTK if
GNU_PROPERTY_X86_FEATURE_1_SHSTK is set on interpreter.
a. If GNU_PROPERTY_X86_FEATURE_1_SHSTK isn't set on the executable
or any shared objects loaded via the DT_NEEDED tag, disable SHSTK.
b. Otherwise lock SHSTK.
3. After SHSTK is enabled, it is an error to load a shared object
without GNU_PROPERTY_X86_FEATURE_1_SHSTK.
To enable CET support in glibc, --enable-cet is required to configure
glibc. When CET is enabled, both compiler and assembler must support
CET. Otherwise, it is a configure-time error.
To support CET run-time control,
1. _dl_x86_feature_1 is added to the writable ld.so namespace to indicate
if IBT or SHSTK are enabled at run-time. It should be initialized by
init_cpu_features.
2. For dynamic executables:
a. A l_cet field is added to struct link_map to indicate if IBT or
SHSTK is enabled in an ELF module. _dl_process_pt_note or
_rtld_process_pt_note is called to process PT_NOTE segment for
GNU program property and set l_cet.
b. _dl_open_check is added to check IBT and SHSTK compatibilty when
dlopening a shared object.
3. Replace i386 _dl_runtime_resolve and _dl_runtime_profile with
_dl_runtime_resolve_shstk and _dl_runtime_profile_shstk, respectively if
SHSTK is enabled.
CET run-time control can be changed via GLIBC_TUNABLES with
$ export GLIBC_TUNABLES=glibc.tune.x86_shstk=[permissive|on|off]
$ export GLIBC_TUNABLES=glibc.tune.x86_ibt=[permissive|on|off]
1. permissive: SHSTK is disabled when dlopening a legacy ELF module.
2. on: IBT or SHSTK are always enabled, regardless if there are IBT or
SHSTK bits in GNU program property.
3. off: IBT or SHSTK are always disabled, regardless if there are IBT or
SHSTK bits in GNU program property.
<cet.h> from CET-enabled GCC is automatically included by assembly codes
to add GNU_PROPERTY_X86_FEATURE_1_IBT and GNU_PROPERTY_X86_FEATURE_1_SHSTK
to GNU program property. _CET_ENDBR is added at the entrance of all
assembly functions whose address may be taken. _CET_NOTRACK is used to
insert NOTRACK prefix with indirect jump table to support IBT. It is
defined as notrack when _CET_NOTRACK is defined in <cet.h>.
[BZ #21598]
* configure.ac: Add --enable-cet.
* configure: Regenerated.
* elf/Makefille (all-built-dso): Add a comment.
* elf/dl-load.c (filebuf): Moved before "dynamic-link.h".
Include <dl-prop.h>.
(_dl_map_object_from_fd): Call _dl_process_pt_note on PT_NOTE
segment.
* elf/dl-open.c: Include <dl-prop.h>.
(dl_open_worker): Call _dl_open_check.
* elf/rtld.c: Include <dl-prop.h>.
(dl_main): Call _rtld_process_pt_note on PT_NOTE segment. Call
_rtld_main_check.
* sysdeps/generic/dl-prop.h: New file.
* sysdeps/i386/dl-cet.c: Likewise.
* sysdeps/unix/sysv/linux/x86/cpu-features.c: Likewise.
* sysdeps/unix/sysv/linux/x86/dl-cet.h: Likewise.
* sysdeps/x86/cet-tunables.h: Likewise.
* sysdeps/x86/check-cet.awk: Likewise.
* sysdeps/x86/configure: Likewise.
* sysdeps/x86/configure.ac: Likewise.
* sysdeps/x86/dl-cet.c: Likewise.
* sysdeps/x86/dl-procruntime.c: Likewise.
* sysdeps/x86/dl-prop.h: Likewise.
* sysdeps/x86/libc-start.h: Likewise.
* sysdeps/x86/link_map.h: Likewise.
* sysdeps/i386/dl-trampoline.S (_dl_runtime_resolve): Add
_CET_ENDBR.
(_dl_runtime_profile): Likewise.
(_dl_runtime_resolve_shstk): New.
(_dl_runtime_profile_shstk): Likewise.
* sysdeps/linux/x86/Makefile (sysdep-dl-routines): Add dl-cet
if CET is enabled.
(CFLAGS-.o): Add -fcf-protection if CET is enabled.
(CFLAGS-.os): Likewise.
(CFLAGS-.op): Likewise.
(CFLAGS-.oS): Likewise.
(asm-CPPFLAGS): Add -fcf-protection -include cet.h if CET
is enabled.
(tests-special): Add $(objpfx)check-cet.out.
(cet-built-dso): New.
(+$(cet-built-dso:=.note)): Likewise.
(common-generated): Add $(cet-built-dso:$(common-objpfx)%=%.note).
($(objpfx)check-cet.out): New.
(generated): Add check-cet.out.
* sysdeps/x86/cpu-features.c: Include <dl-cet.h> and
<cet-tunables.h>.
(TUNABLE_CALLBACK (set_x86_ibt)): New prototype.
(TUNABLE_CALLBACK (set_x86_shstk)): Likewise.
(init_cpu_features): Call get_cet_status to check CET status
and update dl_x86_feature_1 with CET status. Call
TUNABLE_CALLBACK (set_x86_ibt) and TUNABLE_CALLBACK
(set_x86_shstk). Disable and lock CET in libc.a.
* sysdeps/x86/cpu-tunables.c: Include <cet-tunables.h>.
(TUNABLE_CALLBACK (set_x86_ibt)): New function.
(TUNABLE_CALLBACK (set_x86_shstk)): Likewise.
* sysdeps/x86/sysdep.h (_CET_NOTRACK): New.
(_CET_ENDBR): Define if not defined.
(ENTRY): Add _CET_ENDBR.
* sysdeps/x86/dl-tunables.list (glibc.tune): Add x86_ibt and
x86_shstk.
* sysdeps/x86_64/dl-trampoline.h (_dl_runtime_resolve): Add
_CET_ENDBR.
(_dl_runtime_profile): Likewise.
This patch changes longjmp to always restore the TOC pointer (r2 register)
to the caller frame on powerpc64 and powerpc64le. This is related to bug
21895 that reports a situation where you have a static longjmp to a
shared object file.
[BZ #21895]
* sysdeps/powerpc/powerpc64/__longjmp-common.S: Remove condition code for
restoring r2 in longjmp.
* sysdeps/powerpc/powerpc64/Makefile: Added tst-setjmp-bug21895-static to
test list.
Added rules to build test tst-setjmp-bug21895-static.
Added module setjmp-bug21895 and rules to build a shared object from it.
* sysdeps/powerpc/powerpc64/setjmp-bug21895.c: New test file.
* sysdeps/powerpc/powerpc64/tst-setjmp-bug21895-static.c: New test file.
Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Since SHADOW_STACK_POINTER_OFFSET is defined in jmp_buf-ssp.h, we must
undef SHADOW_STACK_POINTER_OFFSET after including <jmp_buf-ssp.h>.
* sysdeps/unix/sysv/linux/x86_64/____longjmp_chk.S: Undef
SHADOW_STACK_POINTER_OFFSET after including <jmp_buf-ssp.h>.
Save and restore shadow stack pointer in setjmp and longjmp to support
shadow stack in Intel CET. Use feature_1 in tcbhead_t to check if
shadow stack is enabled before saving and restoring shadow stack pointer.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/i386/__longjmp.S: Include <jmp_buf-ssp.h>.
(__longjmp): Restore shadow stack pointer if shadow stack is
enabled, SHADOW_STACK_POINTER_OFFSET is defined and __longjmp
isn't defined for __longjmp_cancel.
* sysdeps/i386/bsd-_setjmp.S: Include <jmp_buf-ssp.h>.
(_setjmp): Save shadow stack pointer if shadow stack is enabled
and SHADOW_STACK_POINTER_OFFSET is defined.
* sysdeps/i386/bsd-setjmp.S: Include <jmp_buf-ssp.h>.
(setjmp): Save shadow stack pointer if shadow stack is enabled
and SHADOW_STACK_POINTER_OFFSET is defined.
* sysdeps/i386/setjmp.S: Include <jmp_buf-ssp.h>.
(__sigsetjmp): Save shadow stack pointer if shadow stack is
enabled and SHADOW_STACK_POINTER_OFFSET is defined.
* sysdeps/unix/sysv/linux/i386/____longjmp_chk.S: Include
<jmp_buf-ssp.h>.
(____longjmp_chk): Restore shadow stack pointer if shadow stack
is enabled and SHADOW_STACK_POINTER_OFFSET is defined.
* sysdeps/unix/sysv/linux/x86/Makefile (gen-as-const-headers):
Remove jmp_buf-ssp.sym.
* sysdeps/unix/sysv/linux/x86_64/____longjmp_chk.S: Include
<jmp_buf-ssp.h>.
(____longjmp_chk): Restore shadow stack pointer if shadow stack
is enabled and SHADOW_STACK_POINTER_OFFSET is defined.
* sysdeps/x86/Makefile (gen-as-const-headers): Add
jmp_buf-ssp.sym.
* sysdeps/x86/jmp_buf-ssp.sym: New dummy file.
* sysdeps/x86_64/__longjmp.S: Include <jmp_buf-ssp.h>.
(__longjmp): Restore shadow stack pointer if shadow stack is
enabled, SHADOW_STACK_POINTER_OFFSET is defined and __longjmp
isn't defined for __longjmp_cancel.
* sysdeps/x86_64/setjmp.S: Include <jmp_buf-ssp.h>.
(__sigsetjmp): Save shadow stack pointer if shadow stack is
enabled and SHADOW_STACK_POINTER_OFFSET is defined.
feature_1 has X86_FEATURE_1_IBT and X86_FEATURE_1_SHSTK bits for CET
run-time control.
CET_ENABLED, IBT_ENABLED and SHSTK_ENABLED are defined to 1 or 0 to
indicate that if CET, IBT and SHSTK are enabled.
<tls-setup.h> is added to set up thread-local data.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
[BZ #22563]
* nptl/pthread_create.c: Include <tls-setup.h>.
(__pthread_create_2_1): Call tls_setup_tcbhead.
* sysdeps/generic/tls-setup.h: New file.
* sysdeps/x86/nptl/tls-setup.h: Likewise.
* sysdeps/i386/nptl/tcb-offsets.sym (FEATURE_1_OFFSET): New.
* sysdeps/x86_64/nptl/tcb-offsets.sym (FEATURE_1_OFFSET):
Likewise.
* sysdeps/i386/nptl/tls.h (tcbhead_t): Rename __glibc_reserved1
to feature_1.
* sysdeps/x86_64/nptl/tls.h (tcbhead_t): Likewise.
* sysdeps/x86/sysdep.h (X86_FEATURE_1_IBT): New.
(X86_FEATURE_1_SHSTK): Likewise.
(CET_ENABLED): Likewise.
(IBT_ENABLED): Likewise.
(SHSTK_ENABLED): Likewise.
As pointed out in a libc-alpha thread [1], the misc/tst-ofdlocks-compat
may fail in some specific Linux releases. This patch adds a comment
along with a link to discussion in the test source code.
No changes are expected.
* sysdeps/unix/sysv/linux/tst-ofdlocks-compat.c: Add a comment about
a kernel issue which lead to test failure in some cases.
[1] https://sourceware.org/ml/libc-alpha/2018-07/msg00243.html
This enables searching shared libraries in atomics/ when the hardware
supports LSE atomics of armv8.1 so one can provide optimized variants
of libraries in a portable way.
LSE atomics does not affect library abi, the new instructions can
interoperate with old ones.
I considered the earlier comments on the patch
https://sourceware.org/ml/libc-alpha/2018-04/msg00400.htmlhttps://sourceware.org/ml/libc-alpha/2018-04/msg00625.html
It turns out that the way glibc dynamic linker decides on the search
path is not very flexible: it wants to use hwcap bits and associated
strings. So some targets reuse hwcap bits for glibc internal purposes
to affect the search logic. But hwcap is an interface with the kernel,
glibc should not allocate bits in it for its internal logic as that
limits future hwcap extensions and confusing to users who expect to see
hwcap bits in ifunc resolvers. Instead of rewriting the dynamic linker
path logic (which affects all targets) this patch just uses the existing
mechanism, however this means that the path name has to be the hwcap
name "atomics" and cannot be changed to something more meaningful to
users.
It is hard to tell how much performance benefit this can give, in
principle armv8.1 atomics can be better optimized in the hardware, so it
can make a difference for synchronization heavy code. On some systems
such multilib setup may be the only viable way to get optimized
libraries used.
* sysdeps/unix/sysv/linux/aarch64/dl-procinfo.h (HWCAP_IMPORTANT): Add
HWCAP_ATOMICS.
This partially reverts
commit f82e9672ad
Author: Siddhesh Poyarekar <siddhesh@sourceware.org>
aarch64: Allow overriding HWCAP_CPUID feature check using HWCAP_MASK
The idea was to make it possible to disable cpuid based ifunc resolution
in glibc by changing the hwcap mask which the user could already control.
However the hwcap mask has an orthogonal role: it specifies additional
library search paths for the dynamic linker. So "cpuid" got added to
the search paths when it was set in the default mask (HWCAP_IMPORTANT),
which is not useful behaviour, the hwcap masking should not be reused
in the cpu features code.
Meanwhile there is a tunable to set the cpu explicitly so it is possible
to disable the cpuid based dispatch without using a hwcap mask:
GLIBC_TUNABLES=glibc.tune.cpu=generic
* sysdeps/unix/sysv/linux/aarch64/cpu-features.c (init_cpu_features):
Use dl_hwcap without masking.
* sysdeps/unix/sysv/linux/aarch64/dl-procinfo.h (HWCAP_IMPORTANT):
Remove HWCAP_CPUID.
From Zen onwards this will be enabled. It was disabled for the
Excavator case and will remain disabled.
Reviewd-by: Carlos O'Donell <carlos@redhat.com>
Define a new ABSOLUTE ABI for static linker's use with EI_ABIVERSION
where correct absolute (SHN_ABS) symbol run-time load semantics is
required. This way it can be ensured at static link time that a program
or DSO will not suffer from previous semantics where absolute symbols
were relocated by the base address, or symbols whose `st_value' is zero
silently ignored leading to a confusing "undefined symbol" error message
at load time, and instead "ELF file ABI version invalid" is printed with
old dynamic loaders, making it clear that there is an ABI version
incompatibility.
[BZ #19818]
[BZ #23307]
* libc-abis (ABSOLUTE): New ABI.
* sysdeps/unix/sysv/linux/mips/libc-abis (ABSOLUTE): New ABI.
* NEWS: Mention the new ABI.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
The implementation falls back to renameat if renameat2 is not available
in the kernel (or in the kernel headers) and the flags argument is zero.
Without kernel support, a non-zero argument returns EINVAL, not ENOSYS.
This mirrors what the kernel does for invalid renameat2 flags.
Different than Linux, hurd does not need the OFD locks fix from
06ab719d30 (since OFD locks are current Linux specific). This in
turn allows hurd to not provide a fcntl compat symbol.
Checked on a i686-gnu with check-abi.
* sysdeps/mach/hurd/i386/libc.abilist [GLIBC_2.28] (fcntl): Remove
symbol.
Since the addition of the _Float128 API, strfromf128 and printf_size use
__printf_fp to print _Float128 values. This is achieved by setting the
'is_binary128' member of the 'printf_info' structure to one. Now that
the format of long double on powerpc64le is getting a third option, this
mechanism is reused for long double values that have binary128 format
(i.e.: when -mabi=ieeelongdouble).
This patch adds __printf_sizeieee128 as an exported symbol, but doesn't
provide redirections from printf_size, yet. All redirections will be
installed in a future commit, once all other functions that print or
read long double values with binary128 format are ready. In
__printf_fp, when 'is_binary128' is one, the floating-point argument is
treated as if it was of _Float128 type, regardless of the value of
'is_long_double', thus __printf_sizeieee128 sets 'is_binary128' to the
same value of 'is_long_double'. Otherwise, double values would not be
printed correctly.
Tested for powerpc64le.
Ideally sign should be bool, but sometimes (e.g. in powf) it's more
efficient to pass a non-zero value than 1 to indicate that the sign
should be set. The fixed size int is less ambigous than unsigned
long.
* sysdeps/ieee754/flt-32/e_powf.c (__powf): Use uint32_t.
(exp2f_inline): Likewise.
* sysdeps/ieee754/flt-32/math_config.h (__math_oflowf): Likewise.
(__math_uflowf): Likewise.
(__math_may_uflowf): Likewise.
(__math_divzerof): Likewise.
(__math_invalidf): Likewise.
* sysdeps/ieee754/flt-32/math_errf.c (xflowf): Likewise.
(__math_oflowf): Likewise.
(__math_uflowf): Likewise.
(__math_may_uflowf): Likewise.
(__math_divzerof): Likewise.
(__math_invalidf): Likewise.
The __libc_freeres framework does not extend to non-libc.so objects.
This causes problems in general for valgrind and mtrace detecting
unfreed objects in both libdl.so and libpthread.so. This change is
a pre-requisite to properly moving the malloc hooks out of malloc
since such a move now requires precise accounting of all allocated
data before destructors are run.
This commit adds a proper hook in libc.so.6 for both libdl.so and
for libpthread.so, this ensures that shm-directory.c which uses
freeit () to free memory is called properly. We also remove the
nptl_freeres hook and fall back to using weak-ref-and-check idiom
for a loaded libpthread.so, thus making this process similar for
all DSOs.
Lastly we follow best practice and use explicit free calls for
both libdl.so and libpthread.so instead of the generic hook process
which has undefined order.
Tested on x86_64 with no regressions.
Signed-off-by: DJ Delorie <dj@redhat.com>
Signed-off-by: Carlos O'Donell <carlos@redhat.com>
Vector registers perform better than scalar register pairs for copying
data so prefer them instead. This results in a time reduction of over
50% (i.e. 2x speed improvemnet) for some smaller sizes for memcpy-walk.
Larger sizes show improvements of around 1% to 2%. memcpy-random shows
a very small improvement, in the range of 1-2%.
* sysdeps/aarch64/multiarch/memcpy_falkor.S (__memcpy_falkor):
Use vector registers.
Vector registers perform much better for moves compared to pairs of
registers on falkor, so use them instead. This results in a time
reduction of up to 50% (i.e. 2x improvement) for a lot of the smaller
sizes, i.e. up to 1K in memmove-walk. Improvements for larger sizes are
smaller, at about 1%-2%.
* sysdeps/aarch64/multiarch/memmove_falkor.S
(__memcpy_falkor): Use vector registers.
A lookup operation in map_newlink could turn into an insert because of
holes in the interface part of the map. This leads to incorrectly set
the name of the interface to NULL when the interface is not present
for the address being processed (most likely because the interface was
added between the RTM_GETLINK and RTM_GETADDR calls to the kernel).
When such changes are detected by the kernel, it'll mark the dump as
"inconsistent" by setting NLM_F_DUMP_INTR flag on the next netlink
message.
This patch checks this condition and retries the whole operation.
Hopes are that next time the interface corresponding to the address
entry is present in the list and correct name is returned.
This patch adds __*ieee128 symbols for strfrom, strtold, strtold_l, wcstold
and wcstold_l functions. Redirection from *l to *ieee128 will be handled
in separate patch once we start building these new files.
2018-06-28 Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
* sysdeps/ieee754/ldbl-128ibm-compat/Versions: Add __strfromieee128,
__strtoieee128, __strtoieee128_l,__wcstoieee128 and __wcstoieee128_l.
* sysdeps/ieee754/ldbl-128ibm-compat/strfromf128.c: New file.
* sysdeps/ieee754/ldbl-128ibm-compat/strtof128.c: New file.
* sysdeps/ieee754/ldbl-128ibm-compat/strtof128_l.c: New file.
* sysdeps/ieee754/ldbl-128ibm-compat/wcstof128.c: New file.
* sysdeps/ieee754/ldbl-128ibm-compat/wcstof128_l.c: New file.
This patch fixes the OFD ("file private") locks for architectures that
support non-LFS flock definition (__USE_FILE_OFFSET64 not defined). The
issue in this case is both F_OFD_{GETLK,SETLK,SETLKW} and
F_{SET,GET}L{W}K64 expects a flock64 argument and when using old
F_OFD_* flags with a non LFS flock argument the kernel might interpret
the underlying data wrongly. Kernel idea originally was to avoid using
such flags in non-LFS syscall, but since GLIBC uses fcntl with LFS
semantic as default it is possible to provide the functionality and
avoid the bogus struct kernel passing by adjusting the struct manually
for the required flags.
The idea follows other LFS interfaces that provide two symbols:
1. A new LFS fcntl64 is added on default ABI with the usual macros to
select it for FILE_OFFSET_BITS=64.
2. The Linux non-LFS fcntl use a stack allocated struct flock64 for
F_OFD_{GETLK,SETLK,SETLKW} copy the results on the user provided
struct.
3. Keep a compat symbol with old broken semantic for architectures
that do not define __OFF_T_MATCHES_OFF64_T.
So for architectures which defines __USE_FILE_OFFSET64, fcntl64 will
aliased to fcntl and no adjustment would be required. So to actually
use F_OFD_* with LFS support the source must be built with LFS support
(_FILE_OFFSET_BITS=64).
Also F_OFD_SETLKW command is handled a cancellation point, as for
F_SETLKW{64}.
Checked on x86_64-linux-gnu and i686-linux-gnu.
[BZ #20251]
* NEWS: Mention fcntl64 addition.
* csu/check_fds.c: Replace __fcntl_nocancel by __fcntl64_nocancel.
* login/utmp_file.c: Likewise.
* sysdeps/posix/fdopendir.c: Likewise.
* sysdeps/posix/opendir.c: Likewise.
* sysdeps/unix/pt-fcntl.c: Likewise.
* include/fcntl.h (__libc_fcntl64, __fcntl64,
__fcntl64_nocancel_adjusted): New prototype.
(__fcntl_nocancel_adjusted): Remove prototype.
* io/Makefile (routines): Add fcntl64.
(CFLAGS-fcntl64.c): New rule.
* io/Versions [GLIBC_2.28] (fcntl64): New symbol.
[GLIBC_PRIVATE] (__libc_fcntl): Rename to __libc_fcntl64.
* io/fcntl.h (fcntl64): Add prototype and redirect if
__USE_FILE_OFFSET64 is defined.
* io/fcntl64.c: New file.
* manual/llio.text: Add a note for which commands fcntl acts a
cancellation point.
* nptl/Makefile (CFLAGS-fcntl64.c): New rule.
* sysdeps/mach/hurd/fcntl.c: Alias fcntl to fcntl64 symbols.
* sysdeps/mach/hurd/i386/libc.abilist [GLIBC_2.28] (fcntl, fcntl64):
New symbols.
* sysdeps/unix/sysv/linux/fcntl.c (__libc_fcntl): Fix F_GETLK64,
F_OFD_GETLK, F_SETLK64, F_SETLKW64, F_OFD_SETLK, and F_OFD_SETLKW for
non-LFS case.
* sysdeps/unix/sysv/linux/fcntl64.c: New file.
* sysdeps/unix/sysv/linux/fcntl_nocancel.c (__fcntl_nocancel): Rename
to __fcntl64_nocancel.
(__fcntl_nocancel_adjusted): Rename to __fcntl64_nocancel_adjusted.
* sysdeps/unix/sysv/linux/not-cancel.h (__fcntl_nocancel): Rename
to __fcntl64_nocancel.
* sysdeps/unix/sysv/linux/tst-ofdlocks.c: New file.
* sysdeps/unix/sysv/linux/tst-ofdlocks-compat.c: Likewise.
* sysdeps/unix/sysv/linux/Makefile (tests): Add tst-ofdlocks.
(tests-internal): Add tst-ofdlocks-compat.
* sysdeps/unix/sysv/linux/aarch64/libc.abilist [GLIBC_2.28]
(fcntl64): New symbol.
* sysdeps/unix/sysv/linux/alpha/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libc-le.abilist: Likewise.
* sysdeps/unix/sysv/linux/riscv/rv64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/arm/libc.abilist [GLIBC_2.28] (fcntl,
fcntl64): Likewise.
* sysdeps/unix/sysv/linux/hppa/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libc.abilis: Likewise.
* sysdeps/unix/sysv/linux/m68k/coldfire/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/m680x0/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/microblaze/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/fpu/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/nofpu/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/nios2/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sh/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libc.abilist: Likewise.
Commit 5e79e0292b broke m68k after
s_significand.c became available in the build directory. All m68k
implementations of log1p and significand were including s_significand.c
and stopped working after the inclusion of the the auto-generated file.
This patch reorganizes the implementation of log1p and significand for
m680x0 in order to avoid hitting this problem.
* sysdeps/m68k/m680x0/fpu/s_log1p.c: Set as the generic file for
all log1p and significand functions on m680x0.
* sysdeps/m68k/m680x0/fpu/s_log1pf.c: Include s_log1p.c instead
of s_significand.c..
* sysdeps/m68k/m680x0/fpu/s_log1pl.c: Likewise.
* sysdeps/m68k/m680x0/fpu/s_significandf.c: Likewise.
* sysdeps/m68k/m680x0/fpu/s_significandl.c: Likewise.
* sysdeps/m68k/m680x0/fpu/s_significand.c: Move all the code to
s_log1p.c and include it..
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Add a new libm-alias-float128.h in order to provide the __*ieee128
aliases for the existing *f128 that do not have a globally exported
symbol.
* sysdeps/ieee754/ldbl-128ibm-compat/Versions: New file.
* sysdeps/ieee754/ldbl-128ibm-compat/libm-alias-float128.h: New file.
Move declare_mgen_finite_alias, declare_mgen_finite_alias_s and
declare_mgen_finite_alias_x to a shared place in order to reuse them in
other files that also declare _finite aliases.
* math/e_exp2_template.c (declare_mgen_finite_alias,
declare_mgen_finite_alias_s, declare_mgen_finite_alias_x): Move to...
* sysdeps/generic/math-type-macros.h (declare_mgen_finite_alias,
declare_mgen_finite_alias_s, declare_mgen_finite_alias_x): ... here.
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
This patch updates the hppa definition of MAP_TYPE to reflect a
corresponding change in the Linux kernel in 4.17 (so the value now has
four bits set, as it does on other architectures, although they are
different from other architectures because of hppa differences in
other MAP_* bits).
Tested with build-many-glibcs.py for hppa.
* sysdeps/unix/sysv/linux/hppa/bits/mman.h [__USE_MISC]
(MAP_TYPE): Change value to 0x2b.
My recent nan-sign tests fail to build for powerpc64le with GCC 8
because of the special compile / link options needed there for any
test using _Float128. This patch arranges for these tests to be
handled on powerpc64le similarly to other such tests.
Tested with build-many-glibcs.py for powerpc64le.
[BZ #23303]
* sysdeps/powerpc/powerpc64/le/Makefile
(CFLAGS-tst-strtod-nan-sign.c): Add -mfloat128.
(CFLAGS-tst-wcstod-nan-sign.c): Likewise.
(gnulib-tests): Also add $(f128-loader-link) for
tst-strtod-nan-sign abd tst-wcstod-nan-sign.
* sysdeps/mach/include/mach-shortcuts-hidden.h: New file.
* mach/shortcut.awk: Make syscall stubs include
<mach-shortcuts-hidden.h> and add hidden definition.
* sysdeps/mach/include/mach.h: Include <mach-shortcuts-hidden.h>.
* sysdeps/mach/hurd/lseek.c: Include <errno.h>.
* sysdeps/mach/hurd/lseek.c (__libc_lseek): Check that the value returned
by __lseek64 can fit off_t, return EOVERFLOW otherwise.
126b3ec370 ("hurd: Avoid PLTs for __mach_thread_self and
__mach_reply_port") made mach traps hidden, but htl actually uses two of
them. Re-expose them for now. Exposing them properly will be more involved
since their definition is generated.
* sysdeps/mach/include/mach/mach_traps.h (__mach_thread_self,
__mach_task_self): Remove attribute_hidden.
This patch uses an ifunc to implement gettimeofday in the shared libc.
This is faster compared to the vsyscall mechanism that has to check a
global pointer, demangle it and call it indirectly when the VDSO is
present. Resolving the gettimeofday symbol directly to the VDSO code
is safe because there are no failures that the libc has to handle by
setting errno like in a generic vsyscall (the only failure when the
VDSO code falls back to a syscall is EFAULT, but passing an invalid
pointer is undefined behaviour so returning -EFAULT is fine).
If the kernel supports the VDSO interface we use it for extern calls,
otherwise the old vsyscall method is used which falls back to a syscall.
The static version of gettimeofday continues to use a syscall, libc.so
internal calls use the old vsyscall method.
* sysdeps/unix/sysv/linux/aarch64/gettimeofday.c: New file.
after 329ea513b4 ("Avoid cancellable I/O primitives in ld.so.")
* sysdeps/mach/hurd/localplt.data (ld.so): Add __open64, rename
__libc_read and __libc_write to __read and __write.
They need more work to implement, see bug 23286.
* sysdeps/mach/hurd/i386/Makefile (test-xfail-check-abi-libhurduser,
test-xfail-check-abi-libmachuser): Add.
Neither the <dlfcn.h> entry points, nor lazy symbol resolution, nor
initial shared library load-up, are cancellation points, so ld.so
should exclusively use I/O primitives that are not cancellable. We
currently achieve this by having the cancellation hooks compile as
no-ops when IS_IN(rtld); this patch changes to using exclusively
_nocancel primitives in the source code instead, which makes the
intent clearer and significantly reduces the amount of code compiled
under IS_IN(rtld) as well as IS_IN(libc) -- in particular,
elf/Makefile no longer thinks we require a copy of unwind.c in
rtld-libc.a. (The older mechanism is preserved as a backstop.)
The bulk of the change is splitting up the files that define the
_nocancel I/O functions, so they don't also define the variants that
*are* cancellation points; after which, the existing logic for picking
out the bits of libc that need to be recompiled as part of ld.so Just
Works. I did this for all of the _nocancel functions, not just the
ones used by ld.so, for consistency.
fcntl was a little tricky because it's only a cancellation point for
certain opcodes (F_SETLKW(64), which can block), and the existing
__fcntl_nocancel wasn't applying the FCNTL_ADJUST_CMD hook, which
strikes me as asking for trouble, especially as the only nontrivial
definition of FCNTL_ADJUST_CMD (for powerpc64) changes F_*LK* opcodes.
To fix this, fcntl_common moves to fcntl_nocancel.c along with
__fcntl_nocancel, and changes its name to the extern (but hidden)
symbol __fcntl_nocancel_adjusted, so that regular fcntl can continue
calling it. __fcntl_nocancel now applies FCNTL_ADJUST_CMD; so that
both both fcntl.c and fcntl_nocancel.c can see it, the only nontrivial
definition moves from sysdeps/u/s/l/powerpc/powerpc64/fcntl.c to
.../powerpc64/sysdep.h and becomes entirely a macro, instead of a macro
that calls an inline function.
The nptl version of libpthread also changes a little, because its
"compat-routines" formerly included files that defined all the
_nocancel functions it uses; instead of continuing to duplicate them,
I exported the relevant ones from libc.so as GLIBC_PRIVATE. Since the
Linux fcntl.c calls a function defined by fcntl_nocancel.c, it can no
longer be used from libpthread.so; instead, introduce a custom
forwarder, pt-fcntl.c, and export __libc_fcntl from libc.so as
GLIBC_PRIVATE. The nios2-linux ABI doesn't include a copy of vfork()
in libpthread, and it was handling that by manipulating
libpthread-routines in .../linux/nios2/Makefile; it is cleaner to do
what other such ports do, and have a pt-vfork.S that defines no symbols.
Right now, it appears that Hurd does not implement _nocancel I/O, so
sysdeps/generic/not-cancel.h will forward everything back to the
regular functions. This changed the names of some of the functions
that sysdeps/mach/hurd/dl-sysdep.c needs to interpose.
* elf/dl-load.c, elf/dl-misc.c, elf/dl-profile.c, elf/rtld.c
* sysdeps/unix/sysv/linux/dl-sysdep.c
Include not-cancel.h. Use __close_nocancel instead of __close,
__open64_nocancel instead of __open, __read_nocancel instead of
__libc_read, and __write_nocancel instead of __libc_write.
* csu/check_fds.c (check_one_fd)
* sysdeps/posix/fdopendir.c (__fdopendir)
* sysdeps/posix/opendir.c (__alloc_dir): Use __fcntl_nocancel
instead of __fcntl and/or __libc_fcntl.
* sysdeps/unix/sysv/linux/pthread_setname.c (pthread_setname_np)
* sysdeps/unix/sysv/linux/pthread_getname.c (pthread_getname_np)
* sysdeps/unix/sysv/linux/i386/smp.h (is_smp_system):
Use __open64_nocancel instead of __open_nocancel.
* sysdeps/unix/sysv/linux/not-cancel.h: Move all of the
hidden_proto declarations to the end and issue them if either
IS_IN(libc) or IS_IN(rtld).
* sysdeps/unix/sysv/linux/Makefile [subdir=io] (sysdep_routines):
Add close_nocancel, fcntl_nocancel, nanosleep_nocancel,
open_nocancel, open64_nocancel, openat_nocancel, pause_nocancel,
read_nocancel, waitpid_nocancel, write_nocancel.
* io/Versions [GLIBC_PRIVATE]: Add __libc_fcntl,
__fcntl_nocancel, __open64_nocancel, __write_nocancel.
* posix/Versions: Add __nanosleep_nocancel, __pause_nocancel.
* nptl/pt-fcntl.c: New file.
* nptl/Makefile (pthread-compat-wrappers): Remove fcntl.
(libpthread-routines): Add pt-fcntl.
* include/fcntl.h (__fcntl_nocancel_adjusted): New function.
(__libc_fcntl): Remove attribute_hidden.
* sysdeps/unix/sysv/linux/fcntl.c (__libc_fcntl): Call
__fcntl_nocancel_adjusted, not fcntl_common.
(__fcntl_nocancel): Move to new file fcntl_nocancel.c.
(fcntl_common): Rename to __fcntl_nocancel_adjusted; also move
to fcntl_nocancel.c.
* sysdeps/unix/sysv/linux/fcntl_nocancel.c: New file.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/fcntl.c: Remove file.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/sysdep.h:
Define FCNTL_ADJUST_CMD here, as a self-contained macro.
* sysdeps/unix/sysv/linux/close.c: Move __close_nocancel to...
* sysdeps/unix/sysv/linux/close_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/nanosleep.c: Move __nanosleep_nocancel to...
* sysdeps/unix/sysv/linux/nanosleep_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/open.c: Move __open_nocancel to...
* sysdeps/unix/sysv/linux/open_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/open64.c: Move __open64_nocancel to...
* sysdeps/unix/sysv/linux/open64_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/openat.c: Move __openat_nocancel to...
* sysdeps/unix/sysv/linux/openat_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/openat64.c: Move __openat64_nocancel to...
* sysdeps/unix/sysv/linux/openat64_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/pause.c: Move __pause_nocancel to...
* sysdeps/unix/sysv/linux/pause_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/read.c: Move __read_nocancel to...
* sysdeps/unix/sysv/linux/read_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/waitpid.c: Move __waitpid_nocancel to...
* sysdeps/unix/sysv/linux/waitpid_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/write.c: Move __write_nocancel to...
* sysdeps/unix/sysv/linux/write_nocancel.c: ...this new file.
* sysdeps/unix/sysv/linux/nios2/Makefile: Don't override
libpthread-routines.
* sysdeps/unix/sysv/linux/nios2/pt-vfork.S: New file which
defines nothing.
* sysdeps/mach/hurd/dl-sysdep.c: Define __read instead of
__libc_read, and __write instead of __libc_write. Define
__open64 in addition to __open.
sysdeps/i386/nptl/tls.h has
typedef struct
{
void *tcb; /* Pointer to the TCB. Not necessarily the
thread descriptor used by libpthread. */
dtv_t *dtv;
void *self; /* Pointer to the thread descriptor. */
int multiple_threads;
uintptr_t sysinfo;
uintptr_t stack_guard;
uintptr_t pointer_guard;
int gscope_flag;
int __glibc_reserved1;
/* Reservation of some values for the TM ABI. */
void *__private_tm[4];
/* GCC split stack support. */
void *__private_ss;
} tcbhead_t;
The offset of __private_ss is 0x34. But GCC defines
/* We steal the last transactional memory word. */
#define TARGET_THREAD_SPLIT_STACK_OFFSET 0x30
and libgcc/config/i386/morestack.S has
cmpl %gs:0x30,%eax # See if we have enough space.
movl %eax,%gs:0x30 # Save the new stack boundary.
movl %eax,%gs:0x30 # Save the new stack boundary.
movl %ecx,%gs:0x30 # Save new stack boundary.
movl %eax,%gs:0x30
movl %gs:0x30,%eax
movl %eax,%gs:0x30
Since update TARGET_THREAD_SPLIT_STACK_OFFSET changes split stack ABI,
this patch updates tcbhead_t to match GCC.
[BZ #23250]
[BZ #10686]
* sysdeps/i386/nptl/tls.h (tcbhead_t): Change __private_tm[4]
to _private_tm[3] and add __glibc_reserved2.
Add _Static_assert of offset of __private_ss == 0x30.
* sysdeps/x86_64/nptl/tls.h: Add _Static_assert of offset of
__private_ss == 0x40 for ILP32 and == 0x70 for LP64.
Due to the way the conditions were written, the rtld build of strncmp
ended up with no definition of the strncmp symbol at all: The
implementations were renamed for use within an IFUNC resolver, but the
IFUNC resolver itself was missing (because rtld does not use IFUNCs).
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
As reported in bug 23272, the ldbl-96 implementation of fma (fma for
double, in terms of ldbl-96 as the internal arithmetic type, as used
on 32-bit x86) is missing some of the special-case handling for
non-finite arguments, resulting in incorrect NaN results when the
first two arguments are infinities, the third is finite and so the
infinities go through the logic for finite arguments. This patch
fixes it by handling all cases of non-finite arguments up front, with
additional fma tests for the problem cases being added to the
testsuite.
Tested for x86_64 and x86.
[BZ #23272]
* sysdeps/ieee754/ldbl-96/s_fma.c (__fma): Start by handling all
cases of non-finite arguments.
* math/libm-test-fma.inc (fma_test_data): Add more tests.
syscall restarts and signal returns. Thus, we need to xfail the
check-execstack test.
[BZ #23174]
* sysdeps/unix/sysv/linux/hppa/Makefile: xfail check-execstack.
Current posix_spawnp implementation wrongly tries to execute invalid
binaries (for instance script without shebang) as a shell script in
non compat mode. It was a regression introduced by
9ff72da471 when __spawni started to use
__execvpe instead of __execve (glibc __execvpe try to execute ENOEXEC
as shell script regardless).
This patch fixes it by using an internal symbol (__execvpex) with the
faulty semantic (since compat mode is handled by spawni.c itself).
It was reported by Daniel Drake on libc-help [1].
Checked on x86_64-linux-gnu and i686-linux-gnu.
[BZ #23264]
* include/unistd.h (__execvpex): New prototype.
* posix/Makefile (tests): Add tst-spawn4.
(tests-internal): Add tst-spawn4-compat.
* posix/execvpe.c (__execvpe_common, __execvpex): New functions.
* posix/tst-spawn4-compat.c: New file.
* posix/tst-spawn4.c: Likewise.
* sysdeps/unix/sysv/linux/spawni.c (__spawni): Do not interpret invalid
binaries as shell scripts.
* sysdeps/posix/spawni.c (__spawni): Likewise.
[1] https://sourceware.org/ml/libc-help/2018-06/msg00012.html
_init and _fini are special functions provided by glibc for linker to
define DT_INIT and DT_FINI in executable and shared library. They
should never be put in dynamic symbol table. This patch marks them as
hidden to remove them from dynamic symbol table.
Tested with build-many-glibcs.py.
[BZ #23145]
* elf/Makefile (tests-special): Add $(objpfx)check-initfini.out.
($(all-built-dso:=.dynsym): New target.
(common-generated): Add $(all-built-dso:$(common-objpfx)%=%.dynsym).
($(objpfx)check-initfini.out): New target.
(generated): Add check-initfini.out.
* scripts/check-initfini.awk: New file.
* sysdeps/aarch64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/alpha/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/arm/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/hppa/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/i386/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/ia64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/m68k/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/microblaze/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/mips/mips32/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/mips/mips64/n32/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/mips/mips64/n64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/nios2/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/powerpc/powerpc32/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/powerpc/powerpc64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/s390/s390-32/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/s390/s390-64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/sh/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/sparc/crti.S (_init): Mark as hidden.
(_fini): Likewise.
* sysdeps/x86_64/crti.S (_init): Mark as hidden.
(_fini): Likewise.
When building with -mlong-double-128 or -mabi=ibmlongdouble, TFtype
represents the IBM 128-bit extended floating point type, while KFtype
represents the IEEE 128-bit floating point type.
The soft float implementation of e_sqrtf128 had to redefine TFtype and
TF in order to workaround this issue. However, this behavior changes
when -mabi=ieeelongdouble is used and the macros are not necessary.
* sysdeps/powerpc/powerpc64/le/fpu/e_sqrtf128.c
[__HAVE_FLOAT128_UNLIKE_LDBL] (TFtype, TF): Restrict TFtype
and TF redirection to KFtype and KF only when the default
long double type is not the IEEE 128-bit floating point type.
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Linux 4.17 adds four new AArch64 hwcap values. This patch adds them
to glibc's AArch64 bits/hwcap.h, with corresponding dl-procinfo.c
updates.
Tested with build-many-glibcs.py for aarch64.
* sysdeps/unix/sysv/linux/aarch64/bits/hwcap.h (HWCAP_DIT): New
macro.
(HWCAP_USCAT): Likewise.
(HWCAP_ILRCPC): Likewise.
(HWCAP_FLAGM): Likewise.
* sysdeps/unix/sysv/linux/aarch64/dl-procinfo.c (_DL_HWCAP_COUNT):
Increase to 28.
(_dl_aarch64_cap_flags): Add new flag names.
As far as I can tell, Linux 4.17 does not add any new syscalls; this
patch updates the version number in syscall-names.list to reflect that
it's still current for 4.17.
Tested for x86_64-linux-gnu with build-many-glibcs.py.
* sysdeps/unix/sysv/linux/syscall-names.list: Update kernel
version to 4.17.
Optimize x86-64 strcmp/wcscmp and strncmp/wcsncmp with AVX2. It uses vector
comparison as much as possible. Peak performance observed on a SkyLake
machine: 9x, 3x, 2.5x and 5.5x for strcmp, strncmp, wcscmp and wcsncmp,
respectively. The larger the comparison length, the more benefit using
avx2 functions, except on the strcmp, where peak is observed at length
== 32 bytes. Select AVX2 strcmp/wcscmp on AVX2 machines where vzeroupper
is preferred and AVX unaligned load is fast.
NB: It uses TZCNT instead of BSF since TZCNT produces the same result
as BSF for non-zero input. TZCNT is faster than BSF and is executed
as BSF if machine doesn't support TZCNT.
* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Add
strcmp-avx2, strncmp-avx2, wcscmp-avx2, wcscmp-sse2, wcsncmp-avx2 and
wcsncmp-sse2.
* sysdeps/x86_64/multiarch/ifunc-impl-list.c
(__libc_ifunc_impl_list): Add tests for __strcmp_avx2,
__strncmp_avx2, __wcscmp_avx2, __wcsncmp_avx2, __wcscmp_sse2
and __wcsncmp_sse2.
* sysdeps/x86_64/multiarch/strcmp.c (OPTIMIZE (avx2)):
(IFUNC_SELECTOR): Return OPTIMIZE (avx2) on AVX 2 machines if
AVX unaligned load is fast and vzeroupper is preferred.
* sysdeps/x86_64/multiarch/strncmp.c: Likewise.
* sysdeps/x86_64/multiarch/strcmp-avx2.S: New file.
* sysdeps/x86_64/multiarch/strncmp-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/wcscmp-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/wcscmp-sse2.S: Likewise.
* sysdeps/x86_64/multiarch/wcscmp.c: Likewise.
* sysdeps/x86_64/multiarch/wcsncmp-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/wcsncmp-sse2.c: Likewise.
* sysdeps/x86_64/multiarch/wcsncmp.c: Likewise.
* sysdeps/x86_64/wcscmp.S (__wcscmp): Add alias only if __wcscmp
is undefined.
The results are from configuring with --disable-multi-arch, building
with “-march=x86-64 -mtune=generic -mfpmath=sse” and running the
testsuite on a Haswell-era CPU.
powerpc-nofpu libc exports __sqrtsf2 and __sqrtdf2 symbols. The
export of these soft-fp symbols is a mistake; they aren't part of the
libgcc interface and GCC will never generate code that calls them.
This patch makes them into compat symbols (no code built for static
libc), moving their sources from the generic soft-fp sources to
sysdeps/powerpc/nofpu (the underlying soft-fp FP_SQRT functionality
remains of use to implement actual sqrt public interfaces, such as
sqrtl / sqrtf128 for which it is used on various platforms, but
__sqrt[sdt]f2 are not such interfaces).
Tested with build-many-glibcs.py for relevant platforms.
[BZ #18473]
* soft-fp/sqrttf2.c: Remove file.
* soft-fp/sqrtdf2.c: Move to ....
* sysdeps/powerpc/nofpu/sqrtdf2.c: ... here. Include
<shlib-compat.h>.
(__sqrtdf2): Make conditional on
[SHLIB_COMPAT (libc, GLIBC_2_3_2, GLIBC_2_28)]. Define as compat
symbol.
* soft-fp/sqrtsf2.c: Move to ....
* sysdeps/powerpc/nofpu/sqrtsf2.c: ... here. Include
<shlib-compat.h>.
(__sqrtsf2): Make conditional on
[SHLIB_COMPAT (libc, GLIBC_2_3_2, GLIBC_2_28)]. Define as compat
symbol.
* soft-fp/Makefile (gcc-single-routines): Remove sqrtsf2.
(gcc-double-routines): Remove sqrtdf2.
(gcc-quad-routines): Remove sqrttf2.
* sysdeps/nios2/Makefile [$(subdir) = soft-fp] (sysdep_routines):
Do not filter out sqrtsf2 and sqrtdf2.
* sysdeps/powerpc/nofpu/Makefile [$(subdir) = soft-fp]
(sysdep_routines): Add sqrtsf2 and sqrtdf2.
This patch creates ifunc for sqrtf128() to make use of new xssqrtqp
instruction for POWER9 when --enable-multi-arch and --with-cpu=power8
options are used on power9 system. This is achieved by explicitly
adding -mcpu=power9 flag for sqrtf128-power9.
As per <https://sourceware.org/ml/libc-alpha/2014-10/msg00369.html>,
there should not be separate sysdeps/<arch>/soft-fp directories when
those are used by all configurations that use sysdeps/<arch>, and,
more generally, should not be sysdeps/foo/Implies files pointing to a
subdirectory foo/bar. This patch eliminates the
sysdeps/sparc/sparc64/soft-fp directory accordingly, merging its
contents into sysdeps/sparc/sparc64. This completes removing the
unnecessary <arch>/soft-fp sysdeps directories.
sysdeps/sparc/sparc64/soft-fp/e_ilogbl.c is removed rather than moved.
It was not in fact used previously - the ldbl-128 version of
e_ilogbl.c was used instead - and moving it into sysdeps/sparc/sparc64
results in it being used, but causing a build failure because of
FP_DECL_EX declaring an unused variable (as I noted in
<https://sourceware.org/ml/libc-alpha/2013-10/msg00457.html> that file
doesn't appear to use FP_DECL_EX). Given that the file was previously
unused and so presumably not tested recently, removing it is the safe
way to avoid this patch changing what actually gets built into glibc
(if this file should turn out more efficient than the ldbl-128
e_ilogbl.c, it can always be added back in future with the build
failure fixed).
Tested with build-many-glibcs.py that installed stripped shared
libraries for sparc configurations are unchanged by this patch.
* sysdeps/sparc/sparc64/Implies: Remove sparc/sparc64/soft-fp.
* sysdeps/sparc/sparc64/Makefile [$(subdir) = soft-fp]
(sparc64-quad-routines): New variable. Moved from ....
[$(subdir) = soft-fp] (sysdep_routines): Add
$(sparc64-quad-routines). Moved from ....
[$(subdir) = math] (CPPFLAGS): Add -I../soft-fp/. Moved from ....
* sysdeps/sparc/sparc64/soft-fp/Makefile: ... here. Remove file.
* sysdeps/sparc/sparc64/Versions (libc): Add GLIBC_2.2 symbols
moved from ....
* sysdeps/sparc/sparc64/soft-fp/Versions: ... here. Remove file.
* sysdeps/sparc/sparc64/soft-fp/e_ilogbl.c: Remove file.
* sysdeps/sparc/sparc64/soft-fp/qp_add.c: Move to ....
* sysdeps/sparc/sparc64/qp_add.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_cmp.c: Move to ....
* sysdeps/sparc/sparc64/qp_cmp.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_cmpe.c: Move to ....
* sysdeps/sparc/sparc64/qp_cmpe.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_div.c: Move to ....
* sysdeps/sparc/sparc64/qp_div.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_dtoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_dtoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_feq.c: Move to ....
* sysdeps/sparc/sparc64/qp_feq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_fge.c: Move to ....
* sysdeps/sparc/sparc64/qp_fge.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_fgt.c: Move to ....
* sysdeps/sparc/sparc64/qp_fgt.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_fle.c: Move to ....
* sysdeps/sparc/sparc64/qp_fle.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_flt.c: Move to ....
* sysdeps/sparc/sparc64/qp_flt.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_fne.c: Move to ....
* sysdeps/sparc/sparc64/qp_fne.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_itoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_itoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_mul.c: Move to ....
* sysdeps/sparc/sparc64/qp_mul.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_neg.S: Move to ....
* sysdeps/sparc/sparc64/qp_neg.S: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtod.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtod.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtoi.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtoi.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtos.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtos.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtoui.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtoui.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtoux.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtoux.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_qtox.c: Move to ....
* sysdeps/sparc/sparc64/qp_qtox.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_sqrt.c: Move to ....
* sysdeps/sparc/sparc64/qp_sqrt.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_stoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_stoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_sub.c: Move to ....
* sysdeps/sparc/sparc64/qp_sub.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_uitoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_uitoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_util.c: Move to ....
* sysdeps/sparc/sparc64/qp_util.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_uxtoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_uxtoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/qp_xtoq.c: Move to ....
* sysdeps/sparc/sparc64/qp_xtoq.c: ... here.
* sysdeps/sparc/sparc64/soft-fp/sfp-machine.h: Move to ....
* sysdeps/sparc/sparc64/sfp-machine.h: ... here.
Currently, powerpc, powerpc64, and powerpc64le imply the same set of
subdirectories from sysdeps/ieee754: flt-32, dbl-64, ldbl-128ibm, and
ldbl-opt. In preparation for the transition of the long double format -
from IBM Extended Precision to IEEE 754 128-bits floating-point - on
powerpc64le, this patch splits the shared Implies file into three
separate files (one for each of the powerpc architectures), without
changing their contents. Future patches will modify powerpc64le.
* sysdeps/powerpc/Implies: Removed. Previous contents copied to...
* sysdeps/powerpc/powerpc32/Implies-after: ... here.
* sysdeps/powerpc/powerpc64/be/Implies-after: ... here.
* sysdeps/powerpc/powerpc64/le/Implies-before: ... and here.
As per <https://sourceware.org/ml/libc-alpha/2014-10/msg00369.html>,
there should not be separate sysdeps/<arch>/soft-fp directories when
those are used by all configurations that use sysdeps/<arch>, and,
more generally, should not be sysdeps/foo/Implies files pointing to a
subdirectory foo/bar.
sysdeps/powerpc/soft-fp isn't quite such a case, as the Implies files
pointing to it are
sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/Implies and
sysdeps/unix/sysv/linux/powerpc/powerpc32/e500/nofpu/Implies (and
indeed there is a different sfp-machine.h used for powerpc64le).
However, the same principle applies: there is no need for this
directory because sfp-machine.h, the only file in it, can most
naturally go in sysdeps/powerpc/nofpu, which is used by exactly the
same configurations (and there is a close dependence between the files
there and the sfp-machine.h implementation). This patch eliminates
the sysdeps/powerpc/soft-fp directory accordingly.
Tested with build-many-glibcs.py that installed stripped shared
libraries for powerpc configurations are unchanged by this patch.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/Implies: Remove
powerpc/soft-fp.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/e500/nofpu/Implies:
Likewise.
* sysdeps/powerpc/soft-fp/sfp-machine.h: Move to ....
* sysdeps/powerpc/nofpu/sfp-machine.h: ... here.
As per <https://sourceware.org/ml/libc-alpha/2014-10/msg00369.html>,
there should not be separate sysdeps/<arch>/soft-fp directories when
those are used by all configurations that use sysdeps/<arch>, and,
more generally, should not be sysdeps/foo/Implies files pointing to a
subdirectory foo/bar. This patch eliminates the sysdeps/sh/soft-fp
directory accordingly, merging its contents into sysdeps/sh.
Tested with build-many-glibcs.py that installed stripped shared
libraries for sh configurations are unchanged by this patch.
* sysdeps/sh/Implies: Remove sh/soft-fp.
* sysdeps/sh/soft-fp/sfp-machine.h: Move to ....
* sysdeps/sh/sfp-machine.h: ... here.
This patch skips zero length in __mempcpy_erms, __memmove_erms and
__memset_erms.
Tested on x86-64.
* sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
(__mempcpy_erms): Skip zero length.
(__memmove_erms): Likewise.
* sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
(__memset_erms): Likewise.
As per <https://sourceware.org/ml/libc-alpha/2014-10/msg00369.html>,
there should not be separate sysdeps/<arch>/soft-fp directories when
those are used by all configurations that use sysdeps/<arch>, and,
more generally, should not be sysdeps/foo/Implies files pointing to a
subdirectory foo/bar. This patch eliminates the
sysdeps/alpha/soft-fp directory accordingly, merging its contents
into sysdeps/alpha.
Tested with build-many-glibcs.py that installed stripped shared
libraries for alpha-linux-gnu are unchanged by this patch.
* sysdeps/alpha/Implies: Remove alpha/soft-fp.
* sysdeps/alpha/Makefile [$(subdir) = soft-fp] (sysdep_routines):
Add functions moved from ....
[$(subdir) = math] (CPPFLAGS): Add -I../soft-fp. Moved from ....
* sysdeps/alpha/soft-fp/Makefile: ... here. Remove file.
* sysdeps/alpha/Versions (libc): Add GLIBC_2.3.4 symbols moved
from ....
* sysdeps/alpha/soft-fp/Versions: ... here. Remove file.
* sysdeps/alpha/soft-fp/e_sqrtl.c: Move to ....
* sysdeps/alpha/e_sqrtl.c: ... here.
* sysdeps/alpha/soft-fp/local-soft-fp.h: Move to ....
* sysdeps/alpha/local-soft-fp.h: ... here.
* sysdeps/alpha/soft-fp/ots_add.c: Move to ....
* sysdeps/alpha/ots_add.c: ... here.
* sysdeps/alpha/soft-fp/ots_cmp.c: Move to ....
* sysdeps/alpha/ots_cmp.c: ... here.
* sysdeps/alpha/soft-fp/ots_cmpe.c: Move to ....
* sysdeps/alpha/ots_cmpe.c: ... here.
* sysdeps/alpha/soft-fp/ots_cvtqux.c: Move to ....
* sysdeps/alpha/ots_cvtqux.c: ... here.
* sysdeps/alpha/soft-fp/ots_cvtqx.c: Move to ....
* sysdeps/alpha/ots_cvtqx.c: ... here.
* sysdeps/alpha/soft-fp/ots_cvttx.c: Move to ....
* sysdeps/alpha/ots_cvttx.c: ... here.
* sysdeps/alpha/soft-fp/ots_cvtxq.c: Move to ....
* sysdeps/alpha/ots_cvtxq.c: ... here.
* sysdeps/alpha/soft-fp/ots_cvtxt.c: Move to ....
* sysdeps/alpha/ots_cvtxt.c: ... here.
* sysdeps/alpha/soft-fp/ots_div.c: Move to ....
* sysdeps/alpha/ots_div.c: ... here.
* sysdeps/alpha/soft-fp/ots_mul.c: Move to ....
* sysdeps/alpha/ots_mul.c: ... here.
* sysdeps/alpha/soft-fp/ots_nintxq.c: Move to ....
* sysdeps/alpha/ots_nintxq.c: ... here.
* sysdeps/alpha/soft-fp/ots_sub.c: Move to ....
* sysdeps/alpha/ots_sub.c: ... here.
* sysdeps/alpha/soft-fp/sfp-machine.h: Move to ....
* sysdeps/alpha/sfp-machine.h: ... here.
As per <https://sourceware.org/ml/libc-alpha/2014-10/msg00369.html>,
there should not be separate sysdeps/<arch>/soft-fp directories when
those are used by all configurations that use sysdeps/<arch>, and,
more generally, should not be sysdeps/foo/Implies files pointing to a
subdirectory foo/bar. This patch eliminates the
sysdeps/aarch64/soft-fp directory accordingly, merging its contents
into sysdeps/aarch64.
Tested with build-many-glibcs.py that installed stripped shared
libraries for aarch64 configurations are unchanged by this patch.
* sysdeps/aarch64/Implies: Remove aarch64/soft-fp.
* sysdeps/aarch64/Makefile [$(subdir) = math] (CPPFLAGS): Add
-I../soft-fp. Moved from ....
* sysdeps/aarch64/soft-fp/Makefile: ... here. Remove file.
* sysdeps/aarch64/soft-fp/e_sqrtl.c: Move to ....
* sysdeps/aarch64/e_sqrtl.c: ... here.
* sysdeps/aarch64/soft-fp/sfp-machine.h: Move to ....
* sysdeps/aarch64/sfp-machine.h: ... here.
Building with recent GCC mainline for i686-linux-gnu is failing with:
../sysdeps/ieee754/flt-32/k_rem_pio2f.c: In function '__kernel_rem_pio2f':
../sysdeps/ieee754/flt-32/k_rem_pio2f.c:186:28: error: 'fq[0]' may be used uninitialized in this function [-Werror=maybe-uninitialized]
fv = math_narrow_eval (fq[0]-fv);
^
and
../sysdeps/ieee754/dbl-64/k_rem_pio2.c: In function '__kernel_rem_pio2':
../sysdeps/ieee754/dbl-64/k_rem_pio2.c:333:32: error: 'fq[0]' may be used uninitialized in this function [-Werror=maybe-uninitialized]
fv = math_narrow_eval (fq[0] - fv);
^
These are similar to -Warray-bounds cases for which the DIAG_* macros
are already used in those files: the array element is in fact always
initialized, but the reasoning that it is depends on another array not
having been all zero at an earlier point, which depends on the
functions not being called with zero arguments. Thus, this patch uses
DIAG_* to disable -Wmaybe-uninitialized for this code.
(The warning may be i686-specific because of math_narrow_eval somehow
perturbing what the compiler does with this code enough to cause the
warning. I don't know why it doesn't appear for i686-gnu.)
Tested with build-many-glibcs.py that this fixes the i686 build in
this configuration.
* sysdeps/ieee754/dbl-64/k_rem_pio2.c (__kernel_rem_pio2): Ignore
-Wmaybe-uninitialized around access to fq[0].
* sysdeps/ieee754/flt-32/k_rem_pio2f.c (__kernel_rem_pio2f):
Likewise.
The llseek function name is an obsolete, Linux-specific, unprototyped
name for lseek64 with a link-time warning. This patch completes the
obsoletion of this function name by making it into a compat symbol,
not available for newly linked programs and not included in the ABI
for new ports.
When a compat symbol is defined in syscalls.list, the code for that
function is not built at all for static linking unless some non-compat
symbol for that function is also defined with an explicit symbol
version, so an explicit symbol version for lseek64 is added to the
MIPS n32 syscalls.list. The case in make-syscalls.sh that handles
such explicit non-compat symbol versions then needs to be changed to
use weak_alias instead of strong_alias when the syscall is built
outside of libc, to avoid linknamespace failures from a strong lseek64
symbol in static libpthread.
The x32 llseek.S was as far as I could tell already unused (nothing
builds an llseek.* source file, at least since the lseek / lseek64 /
llseek consolidation), so is removed in this patch as well.
Tested for x86_64 and x86, and with build-many-glibcs.py.
[BZ #18471]
* sysdeps/unix/make-syscalls.sh (emit_weak_aliases): Use weak
aliases for non-libc case of versioned symbols.
* sysdeps/unix/sysv/linux/lseek64.c: Include <shlib-compat.h>.
(llseek): Define as compat symbol if
[SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_28)], not as weak alias
with link warning.
* sysdeps/unix/sysv/linux/mips/mips64/n32/syscalls.list (llseek):
Make into a compat symbol, disabled for minimum symbol version
GLIBC_2.28 and later.
* sysdeps/unix/sysv/linux/x86_64/x32/llseek.S: Remove file.
Although the REP MOVSB implementations of memmove, memcpy and mempcpy
aren't used by the current processors, this patch adds Prefer_FSRM
check in ifunc-memmove.h so that they can be used in the future.
* sysdeps/x86/cpu-features.h (bit_arch_Prefer_FSRM): New.
(index_arch_Prefer_FSRM): Likewise.
* sysdeps/x86/cpu-tunables.c (TUNABLE_CALLBACK (set_hwcaps)):
Also check Prefer_FSRM.
* sysdeps/x86_64/multiarch/ifunc-memmove.h (IFUNC_SELECTOR):
Also return OPTIMIZE (erms) for Prefer_FSRM.
The newer Intel processors support Fast Short REP MOVSB which has a
feature bit in CPUID. This patch adds the Fast Short REP MOVSB (FSRM)
bit to x86 cpu-features.
* sysdeps/x86/cpu-features.h (bit_cpu_FSRM): New.
(index_cpu_FSRM): Likewise.
(reg_FSRM): Likewise.
The Linux nfsservctl syscall was removed in Linux 3.1. Since the
minimum kernel version for use with glibc is 3.2, the glibc wrapper
for this syscall can no longer usefully be called. This patch makes
it into a compat symbol, not provided at all for static linking or new
ports. (It was already the case that there was no header declaration
of this function.)
Tested for x86_64.
* sysdeps/unix/sysv/linux/syscalls.list (nfsservctl): Make into a
compat symbol, disabled for minimum symbol version GLIBC_2.28 and
later.
_Float128 is defined for certain compilers indirectly from
<libm-alias-double.h>, and <ieee754_float128.h> (included from
<math-nan-payload-float128.h>) needs this definition.
As indicated by BZ#23178, concurrent access on some files read by nscd
may result non expected data send through service requisition. This is
due 'sendfile' Linux implementation where for sockets with zero-copy
support, callers must ensure the transferred portions of the the file
reffered by input file descriptor remain unmodified until the reader
on the other end of socket has consumed the transferred data.
I could not find any explicit documentation stating this behaviour on
Linux kernel documentation. However man-pages sendfile entry [1] states
in NOTES the aforementioned remark. It was initially pushed on man-pages
with an explicit testcase [2] that shows changing the file used in
'sendfile' call prior the socket input data consumption results in
previous data being lost.
From commit message it stated on tested Linux version (3.15) only TCP
socket showed this issues, however on recent kernels (4.4) I noticed the
same behaviour for local sockets as well.
Since sendfile on HURD is a read/write operation and the underlying
issue on Linux, the straightforward fix is just remove sendfile use
altogether. I am really skeptical it is hitting some hotstop (there
are indication over internet that sendfile is helpfull only for large
files, more than 10kb) here to justify that extra code complexity or
to pursuit other possible fix (through memory or file locks for
instance, which I am not sure it is doable).
Checked on x86_64-linux-gnu.
[BZ #23178]
* nscd/nscd-client.h (sendfileall): Remove prototype.
* nscd/connections.c [HAVE_SENDFILE] (sendfileall): Remove function.
(handle_request): Use writeall instead of sendfileall.
* nscd/aicache.c (addhstaiX): Likewise.
* nscd/grpcache.c (cache_addgr): Likewise.
* nscd/hstcache.c (cache_addhst): Likewise.
* nscd/initgrcache.c (addinitgroupsX): Likewise.
* nscd/netgroupcache.c (addgetnetgrentX, addinnetgrX): Likewise.
* nscd/pwdcache.c (cache_addpw): Likewise.
* nscd/servicescache.c (cache_addserv): Likewise.
* sysdeps/unix/sysv/linux/Makefile [$(subdir) == nscd]
(sysdep-CFLAGS): Remove -DHAVE_SENDFILE.
* sysdeps/unix/sysv/linux/kernel-features.h (__ASSUME_SENDFILE):
Remove define.
[1] http://man7.org/linux/man-pages/man2/sendfile.2.html
[2] 7b6a329977 (diff-efd6af3a70f0f07c578e85b51e83b3c3)
Unlike i386, we can call hidden IFUNC functions inside libc.so since
x86-64 PLT is always PIC.
Tested on x86-64.
* sysdeps/x86_64/multiarch/strncat-c.c (STRNCAT_PRIMARY): Removed.
Include <string/strncat.c>.
* sysdeps/x86_64/multiarch/strncat.c (__strncat): New strong
alias.
(__GI___strncat): New hidden alias.
Since we have loaded address of PREINIT_FUNCTION into %eax, we can
avoid extra branch to PLT slot.
* sysdeps/i386/crti.S (_init): Replace PREINIT_FUNCTION@PLT
with *%eax in call.
Acked-by: Christian Brauner (Ubuntu) <christian@brauner.io>
Since the result of testl is never used, this patch removes it.
Tested on 64-bit AVX2 machine.
* sysdeps/x86_64/multiarch/strlen-avx2.S (STRLEN): Remove the
unnecessary testl.
When compiling C++ code with -mabi=ieeelongdouble, KCtype is
unavailable and the long double type should be used instead.
This is also providing macro __HAVE_FLOAT128_UNLIKE_LDBL in order to
identify the kind of long double type is being used in the current
compilation unit.
Notice that bits/floatn.h cannot benefit from the new macro due to order
of header inclusion.
* bits/floatn-common.h: Define __HAVE_FLOAT128_UNLIKE_LDBL.
* math/math.h: Restrict the prototype definition for the functions
issignaling(_Float128) and iszero(_Float128); and template
__iseqsig_type<_Float128>, from __HAVE_DISTINCT_FLOAT128 to
__HAVE_FLOAT128_UNLIKE_LDBL.
* sysdeps/powerpc/bits/floatn.h [__HAVE_FLOAT128
&& (!__GNUC_PREREQ (7, 0) || defined __cplusplus)
&& __LDBL_MANT_DIG__ == 113]: Use long double suffix for
__f128() constants; define the type _Float128 as long double;
and reuse long double in __CFLOAT128.
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
This patch continues the math_private.h cleanup by stopping
math_private.h from including math-barriers.h and making the users of
the barrier macros include the latter header directly. No attempt is
made to remove any math_private.h includes that are now unused, except
in strtod_l.c where that is done to avoid line number changes in
assertions, so that installed stripped shared libraries can be
compared before and after the patch. (I think the floating-point
environment support in math_private.h should also move out - some
architectures already have fenv_private.h as an architecture-internal
header included from their math_private.h - and after moving that out
might be a better time to identify unused math_private.h includes.)
Tested for x86_64 and x86, and tested with build-many-glibcs.py that
installed stripped shared libraries are unchanged by the patch.
* sysdeps/generic/math_private.h: Do not include
<math-barriers.h>.
* stdlib/strtod_l.c: Include <math-barriers.h> instead of
<math_private.h>.
* math/fromfp.h: Include <math-barriers.h>.
* math/math-narrow.h: Likewise.
* math/s_nextafter.c: Likewise.
* math/s_nexttowardf.c: Likewise.
* sysdeps/aarch64/fpu/s_llrint.c: Likewise.
* sysdeps/aarch64/fpu/s_llrintf.c: Likewise.
* sysdeps/aarch64/fpu/s_lrint.c: Likewise.
* sysdeps/aarch64/fpu/s_lrintf.c: Likewise.
* sysdeps/i386/fpu/s_nextafterl.c: Likewise.
* sysdeps/i386/fpu/s_nexttoward.c: Likewise.
* sysdeps/i386/fpu/s_nexttowardf.c: Likewise.
* sysdeps/ieee754/dbl-64/e_atan2.c: Likewise.
* sysdeps/ieee754/dbl-64/e_atanh.c: Likewise.
* sysdeps/ieee754/dbl-64/e_exp.c: Likewise.
* sysdeps/ieee754/dbl-64/e_exp2.c: Likewise.
* sysdeps/ieee754/dbl-64/e_j0.c: Likewise.
* sysdeps/ieee754/dbl-64/e_sqrt.c: Likewise.
* sysdeps/ieee754/dbl-64/s_expm1.c: Likewise.
* sysdeps/ieee754/dbl-64/s_fma.c: Likewise.
* sysdeps/ieee754/dbl-64/s_fmaf.c: Likewise.
* sysdeps/ieee754/dbl-64/s_log1p.c: Likewise.
* sysdeps/ieee754/dbl-64/s_nearbyint.c: Likewise.
* sysdeps/ieee754/dbl-64/wordsize-64/s_nearbyint.c: Likewise.
* sysdeps/ieee754/flt-32/e_atanhf.c: Likewise.
* sysdeps/ieee754/flt-32/e_j0f.c: Likewise.
* sysdeps/ieee754/flt-32/s_expm1f.c: Likewise.
* sysdeps/ieee754/flt-32/s_log1pf.c: Likewise.
* sysdeps/ieee754/flt-32/s_nearbyintf.c: Likewise.
* sysdeps/ieee754/flt-32/s_nextafterf.c: Likewise.
* sysdeps/ieee754/k_standardl.c: Likewise.
* sysdeps/ieee754/ldbl-128/e_asinl.c: Likewise.
* sysdeps/ieee754/ldbl-128/e_expl.c: Likewise.
* sysdeps/ieee754/ldbl-128/e_powl.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_fmal.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_nearbyintl.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_nextafterl.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttoward.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttowardf.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_asinl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_fmal.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nextafterl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttoward.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttowardf.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_rintl.c: Likewise.
* sysdeps/ieee754/ldbl-96/e_atanhl.c: Likewise.
* sysdeps/ieee754/ldbl-96/e_j0l.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_fma.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_fmal.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttoward.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttowardf.c: Likewise.
* sysdeps/ieee754/ldbl-opt/s_nexttowardfd.c: Likewise.
* sysdeps/m68k/m680x0/fpu/s_nextafterl.c: Likewise.
For smaller and medium sized copies, the effect of hardware
prefetching are not as dominant as instruction level parallelism.
Hence it makes more sense to load data into multiple registers than to
try and route them to the same prefetch unit. This is also the case
for the loop exit where we are unable to latch on to the same prefetch
unit anyway so it makes more sense to have data loaded in parallel.
The performance results are a bit mixed with memcpy-random, with
numbers jumping between -1% and +3%, i.e. the numbers don't seem
repeatable. memcpy-walk sees a 70% improvement (i.e. > 2x) for 128
bytes and that improvement reduces down as the impact of the tail copy
decreases in comparison to the loop.
* sysdeps/aarch64/multiarch/memcpy_falkor.S (__memcpy_falkor):
Use multiple registers to copy data in loop tail.
The tail of the copy loops are unable to train the falkor hardware
prefetcher because they load from a different base compared to the hot
loop. In this case avoid serializing the instructions by loading them
into different registers. Also peel the last iteration of the loop
into the tail (and have them use different registers) since it gives
better performance for medium sizes.
This results in performance improvements of between 3% and 20% over
the current falkor implementation for sizes between 128 bytes and 1K
on the memmove-walk benchmark, thus mostly covering the regressions
seen against the generic memmove.
* sysdeps/aarch64/multiarch/memmove_falkor.S
(__memmove_falkor): Use multiple registers to move data in
loop tail.
