According to glibc memset microbenchmark test results, for LSX and LASX
versions, A few cases with length less than 8 experience performace
degradation, overall, the LASX version could reduce the runtime about
15% - 75%, LSX version could reduce the runtime about 15%-50%.
The unaligned version uses unaligned memmory access to set data which
length is less than 64 and make address aligned with 8. For this part,
the performace is better than aligned version. Comparing with the generic
version, the performance is close when the length is larger than 128. When
the length is 8-128, the unaligned version could reduce the runtime about
30%-70%, the aligned version could reduce the runtime about 20%-50%.
According to glibc memrchr microbenchmark, this implementation could reduce
the runtime as following:
Name Percent of rutime reduced
memrchr-lasx 20%-83%
memrchr-lsx 20%-64%
According to glibc memchr microbenchmark, this implementation could reduce
the runtime as following:
Name Percent of runtime reduced
memchr-lasx 37%-83%
memchr-lsx 30%-66%
memchr-aligned 0%-15%
According to glibc rawmemchr microbenchmark, A few cases tested with
char '\0' experience performance degradation due to the lasx and lsx
versions don't handle the '\0' separately. Overall, rawmemchr-lasx
implementation could reduce the runtime about 40%-80%, rawmemchr-lsx
implementation could reduce the runtime about 40%-66%, rawmemchr-aligned
implementation could reduce the runtime about 20%-40%.
Based on the glibc microbenchmark, only a few short inputs with this
strncmp-aligned and strncmp-lsx implementation experience performance
degradation, overall, strncmp-aligned could reduce the runtime 0%-10%
for aligned comparision, 10%-25% for unaligend comparision, strncmp-lsx
could reduce the runtime about 0%-60%.
Based on the glibc microbenchmark, strcmp-aligned implementation could
reduce the runtime 0%-10% for aligned comparison, 10%-20% for unaligned
comparison, strcmp-lsx implemenation could reduce the runtime 0%-50%.
Based on the glibc microbenchmark, strnlen-aligned implementation could
reduce the runtime more than 10%, strnlen-lsx implementation could reduce
the runtime about 50%-78%, strnlen-lasx implementation could reduce the
runtime about 50%-88%.
These implementations improve the time to copy data in the glibc
microbenchmark as below:
memcpy-lasx reduces the runtime about 8%-76%
memcpy-lsx reduces the runtime about 8%-72%
memcpy-unaligned reduces the runtime of unaligned data copying up to 40%
memcpy-aligned reduece the runtime of unaligned data copying up to 25%
memmove-lasx reduces the runtime about 20%-73%
memmove-lsx reduces the runtime about 50%
memmove-unaligned reduces the runtime of unaligned data moving up to 40%
memmove-aligned reduces the runtime of unaligned data moving up to 25%
These implementations improve the time to run strchr{nul}
microbenchmark in glibc as below:
strchr-lasx reduces the runtime about 50%-83%
strchr-lsx reduces the runtime about 30%-67%
strchr-aligned reduces the runtime about 10%-20%
strchrnul-lasx reduces the runtime about 50%-83%
strchrnul-lsx reduces the runtime about 36%-65%
strchrnul-aligned reduces the runtime about 6%-10%
strlen-lasx is implemeted by LASX simd instructions(256bit)
strlen-lsx is implemeted by LSX simd instructions(128bit)
strlen-align is implemented by LA basic instructions and never use unaligned memory acess
LoongArch glibc can add some LASX/LSX vector instructions codes,
change the required minimum binutils version to 2.41 which could
support vector instructions. HAVE_LOONGARCH_VEC_ASM is removed
accordingly.
The following usage of macro LEAF/ENTRY are all feasible:
1. LEAF(fcn) -- the align value of fcn is .align 3(default value)
2. LEAF(fcn, 6) -- the align value of fcn is .align 6
Bump autoconf requirement to 2.71 to allow regenerating configure on
more recent distributions. autoconf 2.71 has been in Fedora since F36
and is the current version in Debian stable (bookworm). It appears to
be current in Gentoo as well.
All sysdeps configure and preconfigure scripts have also been
regenerated; all changes are trivial transformations that do not affect
functionality.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
This patch checks if assembler supports vector instructions to
generate LASX/LSX code or not, and then define HAVE_LOONGARCH_VEC_ASM macro
We have added support for vector instructions in binutils-2.41
See:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=75b2f521b101d974354f6ce9ed7c054d8b2e3b7a
commit 75b2f521b101d974354f6ce9ed7c054d8b2e3b7a
Author: mengqinggang <mengqinggang@loongson.cn>
Date: Thu Jun 22 10:35:28 2023 +0800
LoongArch: gas: Add lsx and lasx instructions support
gas/ChangeLog:
* config/tc-loongarch.c (md_parse_option): Add lsx and lasx option.
(loongarch_after_parse_args): Add lsx and lasx option.
opcodes/ChangeLog:
* loongarch-opc.c (struct loongarch_ase): Add lsx and lasx
instructions.
Before GCC r13-2728, it would produce a normal dynamic-linked executable
with -static-pie. I mistakely believed it would produce a static-linked
executable, so failed to detect the breakage. Then with Binutils 2.40
and (vanilla) GCC 12, libc_cv_static_pie_on_loongarch is mistakenly
enabled and cause a building failure with "undefined reference to
_DYNAMIC".
Fix the issue by disabling static PIE if -static-pie creates something
with a INTERP header.
This patch implements the LoongArch specific math barriers in order to omit
the store and load from stack if possible.
Signed-off-by: Xi Ruoyao <xry111@xry111.site>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Add inline assembler for the scalbn functions. Passes GLIBC regression.
GCC 13, LoongArch support ___builtin_scalbn{,f} with -fno-math-errno,
but only "libm" can use -fno-math-errno in GLIBC, and scalbn is in libc
instead of libm because __printf_fp calls it.
GCC 13 compiles these built-ins instead of generic
implementation for function logb.
Link: https://gcc.gnu.org/r13-3922
Co-Authored-By: Xi Ruoyao <xry111@xry111.site>
GCC 13 compiles these built-ins instead of generic
implementation for function llrint.
Link: https://gcc.gnu.org/r13-3920
Co-Authored-By: Xi Ruoyao <xry111@xry111.site>
GCC 13 compiles these built-ins instead of generic
implementation for function lrint.
Link: https://gcc.gnu.org/r13-3920
Co-Authored-By: Xi Ruoyao <xry111@xry111.site>
Use __builtin_{fma, fmaf} to implement function {fma, fmaf} instead of
the generic implementation.
* sysdeps/loongarch/fpu/math-use-builtins-fma.h: New file.
This makes it more likely that the compiler can compute the strlen
argument in _startup_fatal at compile time, which is required to
avoid a dependency on strlen this early during process startup.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
In the future, this will result in a compilation failure if the
macros are unexpectedly undefined (due to header inclusion ordering
or header inclusion missing altogether).
Assembler sources are more difficult to convert. In many cases,
they are hand-optimized for the mangling and no-mangling variants,
which is why they are not converted.
sysdeps/s390/s390-32/__longjmp.c and sysdeps/s390/s390-64/__longjmp.c
are special: These are C sources, but most of the implementation is
in assembler, so the PTR_DEMANGLE macro has to be undefined in some
cases, to match the assembler style.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This allows us to define a generic no-op version of PTR_MANGLE and
PTR_DEMANGLE. In the future, we can use PTR_MANGLE and PTR_DEMANGLE
unconditionally in C sources, avoiding an unintended loss of hardening
due to missing include files or unlucky header inclusion ordering.
In i386 and x86_64, we can avoid a <tls.h> dependency in the C
code by using the computed constant from <tcb-offsets.h>. <sysdep.h>
no longer includes these definitions, so there is no cyclic dependency
anymore when computing the <tcb-offsets.h> constants.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
A start.o compiled from start.S with -DPIC and no -DSHARED is used by
both crt1.o and rcrt1.o. So the LoongArch static PIE patch
unintentionally introduced PC-relative addressing for main and
__libc_start_main into crt1.o.
While the latest Binutils (trunk, which will be released as 2.40)
supports the PC-relative relocs against an external function by creating
a PLT entry, the 2.39 release branch doesn't (and won't) support this.
An error is raised:
"PLT stub does not represent and symbol not defined."
So, we need the following changes:
1. Check if ld supports the PC-relative relocs against an external
function. If it's not supported, we deem static PIE unsupported.
2. Change start.S. If static PIE is supported, use PC-relative
addressing for main and __libc_start_main and rely on the linker to
create PLT entries. Otherwise, restore the old behavior (using GOT
to address these functions).
An alternative would be adding a new "static-pie-start.S", and some
custom logic into Makefile to build rcrt1.o with it. And, restore
start.S to the state before static PIE change so crt1.o won't contain
PC-relative relocs against external symbols. But I can't see any
benefit of this alternative, so I'd just keep it simple.
Tested by building glibc with the following configurations:
1. Binutils trunk + GCC trunk. Static PIE enabled. All tests
passed.
2. Binutils 2.39 branch + GCC trunk. Static PIE disabled. Tests
related to ifunc failed (it's a known issue). All other tests
passed.
3. Binutils 2.39 branch + GCC 12 branch, cross compilation with
build-many-glibcs.py from x86_64-linux-gnu. Static PIE disabled.
Build succeeded.
If the compiler is new enough, enable static PIE support. In the static
PIE version of _start (in rcrt1.o), use la.pcrel instead of la.got
because in a static PIE we cannot use GOT entries until the dynamic
relocations for GOT are resolved.
Rename atomic_exchange_rel/acq to use atomic_exchange_release/acquire
since these map to the standard C11 atomic builtins.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
GCC 13 compiles these built-ins to {fmax,fmin}.{s/d} instruction, use
them instead of the generic implementation.
Link: https://gcc.gnu.org/r13-2085
Signed-off-by: Xi Ruoyao <xry111@xry111.site>
The start code can get linked into dynamic linked executables where
LGPL would require shipping the source or linkable binaries when the
executable is distributed.
On some targets the license exception was missing in start.S (which
is compiled into crt1.o and Scrt1.o which may end up linked into PDE
and PIE binaries).
I did not review what other code may end up in executables, just
fixed the start.S license inconsistency across targets.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
