This operation can be simplified to use simpler multiply-round-convert
sequence, which uses fewer instructions and constants.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Rearrange operations so MOV is not necessary in reduction or around
the special-case handler. Reduce memory access by using more indexed
MLAs in polynomial.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
log1pf is quite register-intensive - use fewer registers for the
polynomial, and make various changes to shorten dependency chains in
parent routines. There is now no spilling with GCC 14. Accuracy moves
around a little - comments adjusted accordingly but does not require
regen-ulps.
Use the helper in log1pf as well, instead of having separate
implementations. The more accurate polynomial means special-casing can
be simplified, and the shorter dependency chain avoids the usual dance
around v0, which is otherwise difficult.
There is a small duplication of vectors containing 1.0f (or 0x3f800000) -
GCC is not currently able to efficiently handle values which fit in FMOV
but not MOVI, and are reinterpreted to integer. There may be potential
for more optimisation if this is fixed.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Reduce MOVPRFXs by using unpredicated (non-destructive) instructions
where possible. Similar to the recent change to AdvSIMD F32 logs,
adjust special-case arguments and bounds to allow for more optimal
register usage. For all 3 routines one MOVPRFX remains in the
reduction, which cannot be avoided as immediate AND and ASR are both
destructive.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Reduce MOV and MOVPRFX by improving special-case handling. Use inline
helper to duplicate the entire computation between the special- and
non-special case branches, removing the contention for z0 between x
and the return value.
Also rearrange some MLAs and MLSs - by making the multiplicand the
destination we can avoid a MOVPRFX in several cases. Also change which
constants go in the vector used for lanewise ops - the last lane is no
longer wasted.
Spotted that shift was incorrect in exp2f and exp10f, w.r.t. to the
comment that explains it. Fixed - worst-case ULP for exp2f moves
around but it doesn't change significantly for either routine.
Worst-case error for coshf increases due to passing x to exp rather
than abs(x) - updated the comment, but does not require regen-ulps.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
It is not necessary to do the conversion at the getdents64
layer for readdir64_r. Doing it piecewise for readdir64
is slightly simpler and allows deleting __old_getdents64.
This fixes bug 32128 because readdir64_r handles the length
check correctly.
Reviewed-by: DJ Delorie <dj@redhat.com>
This enables vectorisation of C23 logp1, which is an alias for log1p.
There are no new tests or ulp entries because the new symbols are simply
aliases.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
A common use case of access () / faccessat () is checking for file
existence, not any specific access permissions. In that case, we can
avoid doing the file_check_access () RPC; whether the given path had
been successfully resolved to a file is all we need to know to answer.
This is prompted by GLib switching to use faccessat (F_OK) to implement
g_file_query_exists () for local files.
https://gitlab.gnome.org/GNOME/glib/-/merge_requests/4272
Signed-off-by: Sergey Bugaev <bugaevc@gmail.com>
Message-ID: <20240919101439.179663-1-bugaevc@gmail.com>
And struct sched_attr.
In sysdeps/unix/sysv/linux/bits/sched.h, the hack that defines
sched_param around the inclusion of <linux/sched/types.h> is quite
ugly, but the definition of struct sched_param has already been
dropped by the kernel, so there is nothing else we can do and maintain
compatibility of <sched.h> with a wide range of kernel header
versions. (An alternative would involve introducing a separate header
for this functionality, but this seems unnecessary.)
The existing sched_* functions that change scheduler parameters
are already incompatible with PTHREAD_PRIO_PROTECT mutexes, so
there is no harm in adding more functionality in this area.
The documentation mostly defers to the Linux manual pages.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Improve small memsets by avoiding branches and use overlapping stores.
Use DC ZVA for copies over 128 bytes. Remove unnecessary code for ZVA sizes
other than 64 and 128. Performance of random memset benchmark improves by 24%
on Neoverse N1.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Since the last operation is destructive, the first argument to the FMA
also has to be the first argument to the special-case in order to
avoid unnecessary MOVs. Reorder arguments and adjust special-case
bounds to facilitate this.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Upon error, return the errno value set by the __getdents call
in __readdir_unlocked. Previously, kernel-reported errors
were ignored.
Reviewed-by: DJ Delorie <dj@redhat.com>
In __syscall_cancel_arch, there's a tail call to __syscall_do_cancel.
On P10, since the caller uses the TOC and the callee is using
PC-relative addressing, there's only a branch instruction with no NOPs
to restore the TOC, which causes the build error. The fix involves adding
the NOTOC directive to the branch instruction, informing the linker
not to generate a TOC stub, thus resolving the issue.
This patch uses 'Avoid_Non_Temporal_Memset' flag to access
the non-temporal memset implementation for hygon processors.
Test Results:
hygon1 arch
x86_memset_non_temporal_threshold = 8MB
size new performance time / old performance time
1MB 0.994
4MB 0.996
8MB 0.670
16MB 0.343
32MB 0.355
hygon2 arch
x86_memset_non_temporal_threshold = 8MB
size new performance time / old performance time
1MB 1
4MB 1
8MB 1.312
16MB 0.822
32MB 0.830
hygon3 arch
x86_memset_non_temporal_threshold = 8MB
size new performance time / old performance time
1MB 1
4MB 0.990
8MB 0.737
16MB 0.390
32MB 0.401
For hygon arch with this patch, non-temporal stores can improve
performance by 20% - 65%.
Signed-off-by: Feifei Wang <wangfeifei@hygon.cn>
Reviewed-by: Jing Li <lijing@hygon.cn>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Add hygon branch in dl_init_cacheinfo function to initialize
cache size variables for hygon processors. In the meanwhile,
add handle_hygon() function to get cache information.
Signed-off-by: Feifei Wang <wangfeifei@hygon.cn>
Reviewed-by: Jing Li <lijing@hygon.cn>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Add a new architecture type arch_kind_hygon to spilt Hygon branch
from AMD. This is to facilitate the Hygon processors to make settings
that are suitable for its own characteristics.
Signed-off-by: Feifei Wang <wangfeifei@hygon.cn>
Reviewed-by: Jing Li <lijing@hygon.cn>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
One can be very unlucky to call time_now first just before a second switch,
and mach_msg sleep just a bit more enough for the second time_now call to
count one second too many (or even more if scheduling is really unlucky).
So we have to protect against returning a bogus negative value in such case.
This patch modifies the current Power9 implementation of strcpy and
stpcpy to optimize it for Power9 and Power10.
No new Power10 instructions are used, so the original Power9 strcpy
is modified instead of creating a new implementation for Power10.
The changes also affect stpcpy, which uses the same implementation
with some additional code before returning.
Improvements compared to the old Power9 version:
Use simple comparisons for the first ~512 bytes:
The main loop is good for long strings, but comparing 16B each time is
better for shorter strings. After aligning the address to 16 bytes, we
unroll the loop four times, checking 128 bytes each time. There may be
some overlap with the main loop for unaligned strings, but it is better
for shorter strings.
Loop with 64 bytes for longer bytes:
Use 4 consecutive lxv/stxv instructions.
Showed an average improvement of 13%.
Reviewed-by: Paul E. Murphy <murphyp@linux.ibm.com>
Reviewed-by: Peter Bergner <bergner@linux.ibm.com>
The current racy approach is to enable asynchronous cancellation
before making the syscall and restore the previous cancellation
type once the syscall returns, and check if cancellation has happen
during the cancellation entrypoint.
As described in BZ#12683, this approach shows 2 problems:
1. Cancellation can act after the syscall has returned from the
kernel, but before userspace saves the return value. It might
result in a resource leak if the syscall allocated a resource or a
side effect (partial read/write), and there is no way to program
handle it with cancellation handlers.
2. If a signal is handled while the thread is blocked at a cancellable
syscall, the entire signal handler runs with asynchronous
cancellation enabled. This can lead to issues if the signal
handler call functions which are async-signal-safe but not
async-cancel-safe.
For the cancellation to work correctly, there are 5 points at which the
cancellation signal could arrive:
[ ... )[ ... )[ syscall ]( ...
1 2 3 4 5
1. Before initial testcancel, e.g. [*... testcancel)
2. Between testcancel and syscall start, e.g. [testcancel...syscall start)
3. While syscall is blocked and no side effects have yet taken
place, e.g. [ syscall ]
4. Same as 3 but with side-effects having occurred (e.g. a partial
read or write).
5. After syscall end e.g. (syscall end...*]
And libc wants to act on cancellation in cases 1, 2, and 3 but not
in cases 4 or 5. For the 4 and 5 cases, the cancellation will eventually
happen in the next cancellable entrypoint without any further external
event.
The proposed solution for each case is:
1. Do a conditional branch based on whether the thread has received
a cancellation request;
2. It can be caught by the signal handler determining that the saved
program counter (from the ucontext_t) is in some address range
beginning just before the "testcancel" and ending with the
syscall instruction.
3. SIGCANCEL can be caught by the signal handler and determine that
the saved program counter (from the ucontext_t) is in the address
range beginning just before "testcancel" and ending with the first
uninterruptable (via a signal) syscall instruction that enters the
kernel.
4. In this case, except for certain syscalls that ALWAYS fail with
EINTR even for non-interrupting signals, the kernel will reset
the program counter to point at the syscall instruction during
signal handling, so that the syscall is restarted when the signal
handler returns. So, from the signal handler's standpoint, this
looks the same as case 2, and thus it's taken care of.
5. For syscalls with side-effects, the kernel cannot restart the
syscall; when it's interrupted by a signal, the kernel must cause
the syscall to return with whatever partial result is obtained
(e.g. partial read or write).
6. The saved program counter points just after the syscall
instruction, so the signal handler won't act on cancellation.
This is similar to 4. since the program counter is past the syscall
instruction.
So The proposed fixes are:
1. Remove the enable_asynccancel/disable_asynccancel function usage in
cancellable syscall definition and instead make them call a common
symbol that will check if cancellation is enabled (__syscall_cancel
at nptl/cancellation.c), call the arch-specific cancellable
entry-point (__syscall_cancel_arch), and cancel the thread when
required.
2. Provide an arch-specific generic system call wrapper function
that contains global markers. These markers will be used in
SIGCANCEL signal handler to check if the interruption has been
called in a valid syscall and if the syscalls has side-effects.
A reference implementation sysdeps/unix/sysv/linux/syscall_cancel.c
is provided. However, the markers may not be set on correct
expected places depending on how INTERNAL_SYSCALL_NCS is
implemented by the architecture. It is expected that all
architectures add an arch-specific implementation.
3. Rewrite SIGCANCEL asynchronous handler to check for both canceling
type and if current IP from signal handler falls between the global
markers and act accordingly.
4. Adjust libc code to replace LIBC_CANCEL_ASYNC/LIBC_CANCEL_RESET to
use the appropriate cancelable syscalls.
5. Adjust 'lowlevellock-futex.h' arch-specific implementations to
provide cancelable futex calls.
Some architectures require specific support on syscall handling:
* On i386 the syscall cancel bridge needs to use the old int80
instruction because the optimized vDSO symbol the resulting PC value
for an interrupted syscall points to an address outside the expected
markers in __syscall_cancel_arch. It has been discussed in LKML [1]
on how kernel could help userland to accomplish it, but afaik
discussion has stalled.
Also, sysenter should not be used directly by libc since its calling
convention is set by the kernel depending of the underlying x86 chip
(check kernel commit 30bfa7b3488bfb1bb75c9f50a5fcac1832970c60).
* mips o32 is the only kABI that requires 7 argument syscall, and to
avoid add a requirement on all architectures to support it, mips
support is added with extra internal defines.
Checked on aarch64-linux-gnu, arm-linux-gnueabihf, powerpc-linux-gnu,
powerpc64-linux-gnu, powerpc64le-linux-gnu, i686-linux-gnu, and
x86_64-linux-gnu.
[1] https://lkml.org/lkml/2016/3/8/1105
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
With ia64 removal, the function descriptor supports is only used
by HPPA and new architectures do not seem leaning towards this
design.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
This commit uses a common implementation 'strnlen-evex-base.S' for both
'strnlen-evex' and 'strnlen-evex512'
This patch serves both to reduce the number of implementations, and it also does some small optimizations that benefit strnlen-evex and strnlen-evex512.
All tests pass on x86.
Benchmarks were taken on SKX.
https://www.intel.com/content/www/us/en/products/sku/123613/intel-core-i97900x-xseries-processor-13-75m-cache-up-to-4-30-ghz/specifications.html
Geometric mean for strnlen-evex over all benchmarks (N=10) was (new/old) 0.881
Geometric mean for strnlen-evex512 over all benchmarks (N=10) was (new/old) 0.953
Code Size Changes:
strnlen-evex : +31 bytes
strnlen-evex512 : +156 bytes
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
Fix an issue with commit 8f4632deb3 ("Linux: rseq registration tests")
and prevent testing from being run in the process of the test driver
itself rather than just the test child where one has been forked. The
problem here is the unguarded use of a destructor to call a part of the
testing. The destructor function, 'do_rseq_destructor_test' is called
implicitly at program completion, however because it is associated with
the executable itself rather than an individual process, it is called
both in the test child *and* in the test driver itself.
Prevent this from happening by providing a guard variable that only
enables test invocation from 'do_rseq_destructor_test' in the process
that has first run 'do_test'. Consequently extra testing is invoked
from 'do_rseq_destructor_test' only once and in the correct process,
regardless of the use or the lack of of the '--direct' option. Where
called in the controlling test driver process that has neved called
'do_test' the destructor function silently returns right away without
taking any further actions, letting the test driver fail gracefully
where applicable.
This arrangement prevents 'tst-rseq-nptl' from ever causing testing to
hang forever and never complete, such as currently happening with the
'mips-linux-gnu' (o32 ABI) target.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Previously if the setaffinity wrapper failed the rest of the subtest
would not execute and the current subtest would be reported as passing.
Now if the setaffinity wrapper fails the subtest is correctly reported
as faling. Tested manually by changing the conditions of the affinity
call including setting size to zero, or checking the wrong condition.
No regressions on x86_64.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
The goal of this flag is to allow targets which don't prefer/have ERMS
to still access the non-temporal memset implementation.
There are 4 cases for tuning memset:
1) `Avoid_STOSB && Avoid_Non_Temporal_Memset`
- Memset with temporal stores
2) `Avoid_STOSB && !Avoid_Non_Temporal_Memset`
- Memset with temporal/non-temporal stores. Non-temporal path
goes through `rep stosb` path. We accomplish this by setting
`x86_rep_stosb_threshold` to
`x86_memset_non_temporal_threshold`.
3) `!Avoid_STOSB && Avoid_Non_Temporal_Memset`
- Memset with temporal stores/`rep stosb`
3) `!Avoid_STOSB && !Avoid_Non_Temporal_Memset`
- Memset with temporal stores/`rep stosb`/non-temporal stores.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
This is just a refactor and there should be no behavioral change from
this commit.
The goal is to make `Avoid_Non_Temporal_Memset` a more universal knob
for controlling whether we use non-temporal memset rather than having
extra logic based on vendor.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Issue was we were expecting not matches with CHAR before the start of
the string in the page cross case.
The check code in the page cross case:
```
and $0xffffffffffffffc0,%rax
vmovdqa64 (%rax),%zmm17
vpcmpneqb %zmm17,%zmm16,%k1
vptestmb %zmm17,%zmm17,%k0{%k1}
kmovq %k0,%rax
inc %rax
shr %cl,%rax
je L(continue)
```
expects that all characters that neither match null nor CHAR will be
1s in `rax` prior to the `inc`. Then the `inc` will overflow all of
the 1s where no relevant match was found.
This is incorrect in the page-cross case, as the
`vmovdqa64 (%rax),%zmm17` loads from before the start of the input
string.
If there are matches with CHAR before the start of the string, `rax`
won't properly overflow.
The fix is quite simple. Just replace:
```
inc %rax
shr %cl,%rax
```
With:
```
sar %cl,%rax
inc %rax
```
The arithmetic shift will clear any matches prior to the start of the
string while maintaining the signbit so the 1s can properly overflow
to zero in the case of no matches.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
In _dl_tlsdesc_dynamic, there are three 'addi.d sp, sp, -size'
instructions to allocate stack size for Float/LSX/LASX registers.
Every 'addi.d sp, sp, -size' needs a cfi_adjust_cfa_offset because
of sp is used to compute CFA. But only one 'addi.d sp, sp, -size'
will be run according to HWCAP value. And all cfi_adjust_cfa_offset
will be executed in stack unwinding, it result in incorrect CFA.
Change _dl_tlsdesc_dynamic to _dl_tlsdesc_dynamic,
_dl_tlsdesc_dynamic_lsx and _dl_tlsdesc_dynamic_lasx.
Conflicting cfi instructions can be distributed to the three functions.
And cfi instructions can correspond to stack down instructions.
Improve performance by handling another 16 bytes before entering the loop.
Use ADDHN in the loop to avoid SHRN+FMOV when it terminates. Change final
size computation to avoid increasing latency. On Neoverse V1 performance
of the random strlen benchmark improves by 4.6%.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
