rpms/glibc
5
0
Fork 2
Code Issues Pull Requests Releases Activity

Sync with upstream branch release/2.39/master (CVE-2025-0395)

Browse Source 
Upstream commit: 808a84a8b81468b517a4d721fdc62069cb8c211f

- Fix underallocation of abort_msg_s struct (CVE-2025-0395)
- x86/string: Fixup alignment of main loop in str{n}cmp-evex [BZ #32212]
- x86: Improve large memset perf with non-temporal stores [RHEL-29312]
- x86: Avoid integer truncation with large cache sizes (bug 32470)
- math: Exclude internal math symbols for tests [BZ #32414]
- malloc: add indirection for malloc(-like) functions in tests [BZ #32366]
- Pass -nostdlib -nostartfiles together with -r [BZ #31753]
- nptl: initialize cpu_id_start prior to rseq registration
- nptl: initialize rseq area prior to registration
This commit is contained in:
Florian Weimer 2025-01-23 09:08:39 +01:00
parent 071446cc90
commit 7e7241f620
10 changed files with 831 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
glibc-upstream-2.39-138.patch Normal file
View File

@ -0,0 +1,55 @@
commit 9a0e174a39a3a65f628c6a55e29fe35f6d67bf42
Author: Michael Jeanson <mjeanson@efficios.com>
Date: Thu Nov 7 22:23:49 2024 +0100
nptl: initialize rseq area prior to registration
Per the rseq syscall documentation, 3 fields are required to be
initialized by userspace prior to registration, they are 'cpu_id',
'rseq_cs' and 'flags'. Since we have no guarantee that 'struct pthread'
is cleared on all architectures, explicitly set those 3 fields prior to
registration.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
(cherry picked from commit 97f60abd25628425971f07e9b0e7f8eec0741235)
diff --git a/nptl/descr.h b/nptl/descr.h
index 4697f633e16c7359..a83df327e4bcba2e 100644
--- a/nptl/descr.h
+++ b/nptl/descr.h
@@ -417,6 +417,8 @@ struct pthread
{
uint32_t cpu_id_start;
uint32_t cpu_id;
+ uint64_t rseq_cs;
+ uint32_t flags;
};
char pad[32]; /* Original rseq area size. */
} rseq_area __attribute__ ((aligned (32)));
diff --git a/sysdeps/unix/sysv/linux/rseq-internal.h b/sysdeps/unix/sysv/linux/rseq-internal.h
index 7ea935b4adab8c20..37a8f630b6519ff0 100644
--- a/sysdeps/unix/sysv/linux/rseq-internal.h
+++ b/sysdeps/unix/sysv/linux/rseq-internal.h
@@ -51,11 +51,21 @@ rseq_register_current_thread (struct pthread *self, bool do_rseq)
/* The initial implementation used only 20 bytes out of 32,
but still expected size 32. */
size = RSEQ_AREA_SIZE_INITIAL;
+
+ /* Initialize the rseq fields that are read by the kernel on
+ registration, there is no guarantee that struct pthread is
+ cleared on all architectures. */
+ THREAD_SETMEM (self, rseq_area.cpu_id, RSEQ_CPU_ID_UNINITIALIZED);
+ THREAD_SETMEM (self, rseq_area.rseq_cs, 0);
+ THREAD_SETMEM (self, rseq_area.flags, 0);
+
int ret = INTERNAL_SYSCALL_CALL (rseq, &self->rseq_area,
size, 0, RSEQ_SIG);
if (!INTERNAL_SYSCALL_ERROR_P (ret))
return true;
}
+ /* When rseq is disabled by tunables or the registration fails, inform
+ userspace by setting 'cpu_id' to RSEQ_CPU_ID_REGISTRATION_FAILED. */
THREAD_SETMEM (self, rseq_area.cpu_id, RSEQ_CPU_ID_REGISTRATION_FAILED);
return false;
}

29
glibc-upstream-2.39-139.patch Normal file
View File

@ -0,0 +1,29 @@
commit 350db2839387659e1500a54d276e401c9c6b2dee
Author: Michael Jeanson <mjeanson@efficios.com>
Date: Wed Nov 20 14:15:42 2024 -0500
nptl: initialize cpu_id_start prior to rseq registration
When adding explicit initialization of rseq fields prior to
registration, I glossed over the fact that 'cpu_id_start' is also
documented as initialized by user-space.
While current kernels don't validate the content of this field on
registration, future ones could.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
(cherry picked from commit d9f40387d3305d97e30a8cf8724218c42a63680a)
diff --git a/sysdeps/unix/sysv/linux/rseq-internal.h b/sysdeps/unix/sysv/linux/rseq-internal.h
index 37a8f630b6519ff0..ef3eab1fefd4d90d 100644
--- a/sysdeps/unix/sysv/linux/rseq-internal.h
+++ b/sysdeps/unix/sysv/linux/rseq-internal.h
@@ -56,6 +56,7 @@ rseq_register_current_thread (struct pthread *self, bool do_rseq)
registration, there is no guarantee that struct pthread is
cleared on all architectures. */
THREAD_SETMEM (self, rseq_area.cpu_id, RSEQ_CPU_ID_UNINITIALIZED);
+ THREAD_SETMEM (self, rseq_area.cpu_id_start, 0);
THREAD_SETMEM (self, rseq_area.rseq_cs, 0);
THREAD_SETMEM (self, rseq_area.flags, 0);

28
glibc-upstream-2.39-140.patch Normal file
View File

@ -0,0 +1,28 @@
commit aa8768999e94fcee1695feb766c69dd8a93b706b
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Fri May 17 20:00:38 2024 -0700
Pass -nostdlib -nostartfiles together with -r [BZ #31753]
Since -r in GCC 6/7/8 doesn't imply -nostdlib -nostartfiles, update the
link-static-libc.out rule to also pass -nostdlib -nostartfiles. This
fixes BZ #31753.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
(cherry picked from commit 2be3352f0b1ebaa39596393fffe1062275186669)
diff --git a/Makefile b/Makefile
index 37bf70aa4ad4403f..ae9bc09327dd2d5b 100644
--- a/Makefile
+++ b/Makefile
@@ -581,7 +581,8 @@ $(objpfx)lint-makefiles.out: scripts/lint-makefiles.sh
# definitions of any symbols.
tests-special += $(objpfx)link-static-libc.out
$(objpfx)link-static-libc.out:
- $(LINK.o) $(whole-archive) -r $(objpfx)libc.a -o /dev/null > $@ 2>&1; \
+ $(LINK.o) $(whole-archive) -nostdlib -nostartfiles -r \
+ $(objpfx)libc.a -o /dev/null > $@ 2>&1; \
$(evaluate-test)
# Print test summary for tests in $1 .sum file;

172
glibc-upstream-2.39-141.patch Normal file
View File

@ -0,0 +1,172 @@
commit 51da74a97e0f024fd89b57304b3ab010a3cfaef1
Author: Sam James <sam@gentoo.org>
Date: Mon Dec 9 23:11:25 2024 +0000
malloc: add indirection for malloc(-like) functions in tests [BZ #32366]
GCC 15 introduces allocation dead code removal (DCE) for PR117370 in
r15-5255-g7828dc070510f8. This breaks various glibc tests which want
to assert various properties of the allocator without doing anything
obviously useful with the allocated memory.
Alexander Monakov rightly pointed out that we can and should do better
than passing -fno-malloc-dce to paper over the problem. Not least because
GCC 14 already does such DCE where there's no testing of malloc's return
value against NULL, and LLVM has such optimisations too.
Handle this by providing malloc (and friends) wrappers with a volatile
function pointer to obscure that we're calling malloc (et. al) from the
compiler.
Reviewed-by: Paul Eggert <eggert@cs.ucla.edu>
(cherry picked from commit a9944a52c967ce76a5894c30d0274b824df43c7a)
diff --git a/malloc/tst-aligned-alloc.c b/malloc/tst-aligned-alloc.c
index 91167d1392c0e626..b0f05a8fec78d5e8 100644
--- a/malloc/tst-aligned-alloc.c
+++ b/malloc/tst-aligned-alloc.c
@@ -25,6 +25,8 @@
#include <libc-diag.h>
#include <support/check.h>
+#include "tst-malloc-aux.h"
+
static int
do_test (void)
{
diff --git a/malloc/tst-compathooks-off.c b/malloc/tst-compathooks-off.c
index d0106f3fb74ff3b1..4cce6e5a8076f6b6 100644
--- a/malloc/tst-compathooks-off.c
+++ b/malloc/tst-compathooks-off.c
@@ -25,6 +25,8 @@
#include <support/check.h>
#include <support/support.h>
+#include "tst-malloc-aux.h"
+
extern void (*volatile __free_hook) (void *, const void *);
extern void *(*volatile __malloc_hook)(size_t, const void *);
extern void *(*volatile __realloc_hook)(void *, size_t, const void *);
diff --git a/malloc/tst-malloc-aux.h b/malloc/tst-malloc-aux.h
new file mode 100644
index 0000000000000000..54908b4a2464d510
--- /dev/null
+++ b/malloc/tst-malloc-aux.h
@@ -0,0 +1,41 @@
+/* Wrappers for malloc-like functions to allow testing the implementation
+ without optimization.
+ Copyright (C) 2024 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public License as
+ published by the Free Software Foundation; either version 2.1 of the
+ License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; see the file COPYING.LIB. If
+ not, see <https://www.gnu.org/licenses/>. */
+
+#ifndef TST_MALLOC_AUX_H
+#define TST_MALLOC_AUX_H
+
+#include <stddef.h>
+#include <stdlib.h>
+
+static void *(*volatile aligned_alloc_indirect)(size_t, size_t) = aligned_alloc;
+static void *(*volatile calloc_indirect)(size_t, size_t) = calloc;
+static void *(*volatile malloc_indirect)(size_t) = malloc;
+static void *(*volatile realloc_indirect)(void*, size_t) = realloc;
+
+#undef aligned_alloc
+#undef calloc
+#undef malloc
+#undef realloc
+
+#define aligned_alloc aligned_alloc_indirect
+#define calloc calloc_indirect
+#define malloc malloc_indirect
+#define realloc realloc_indirect
+
+#endif /* TST_MALLOC_AUX_H */
diff --git a/malloc/tst-malloc-check.c b/malloc/tst-malloc-check.c
index fde8863ad7561a71..cc88bff3b39a421c 100644
--- a/malloc/tst-malloc-check.c
+++ b/malloc/tst-malloc-check.c
@@ -20,6 +20,8 @@
#include <stdlib.h>
#include <libc-diag.h>
+#include "tst-malloc-aux.h"
+
static int errors = 0;
static void
diff --git a/malloc/tst-malloc-too-large.c b/malloc/tst-malloc-too-large.c
index 8e9e0d5fa2b4b907..2b91377e54cdc485 100644
--- a/malloc/tst-malloc-too-large.c
+++ b/malloc/tst-malloc-too-large.c
@@ -43,6 +43,7 @@
#include <unistd.h>
#include <sys/param.h>
+#include "tst-malloc-aux.h"
/* This function prepares for each 'too-large memory allocation' test by
performing a small successful malloc/free and resetting errno prior to
diff --git a/malloc/tst-malloc.c b/malloc/tst-malloc.c
index f7a6e4654c374d01..68af399022543111 100644
--- a/malloc/tst-malloc.c
+++ b/malloc/tst-malloc.c
@@ -22,6 +22,8 @@
#include <libc-diag.h>
#include <time.h>
+#include "tst-malloc-aux.h"
+
static int errors = 0;
static void
diff --git a/malloc/tst-realloc.c b/malloc/tst-realloc.c
index f50499ecb114d574..74a28fb45ed80bf5 100644
--- a/malloc/tst-realloc.c
+++ b/malloc/tst-realloc.c
@@ -23,6 +23,8 @@
#include <libc-diag.h>
#include <support/check.h>
+#include "tst-malloc-aux.h"
+
static int
do_test (void)
{
diff --git a/support/support.h b/support/support.h
index ba21ec9b5add7c02..1a77f7979330d60c 100644
--- a/support/support.h
+++ b/support/support.h
@@ -113,7 +113,7 @@ void *xposix_memalign (size_t alignment, size_t n)
__attribute_malloc__ __attribute_alloc_align__ ((1))
__attribute_alloc_size__ ((2)) __attr_dealloc_free __returns_nonnull;
char *xasprintf (const char *format, ...)
- __attribute__ ((format (printf, 1, 2), malloc)) __attr_dealloc_free
+ __attribute__ ((format (printf, 1, 2), __malloc__)) __attr_dealloc_free
__returns_nonnull;
char *xstrdup (const char *) __attr_dealloc_free __returns_nonnull;
char *xstrndup (const char *, size_t) __attr_dealloc_free __returns_nonnull;
diff --git a/test-skeleton.c b/test-skeleton.c
index ae185a4f2821de00..690f26e7cf229622 100644
--- a/test-skeleton.c
+++ b/test-skeleton.c
@@ -27,7 +27,6 @@
#include <errno.h>
#include <fcntl.h>
#include <getopt.h>
-#include <malloc.h>
#include <paths.h>
#include <search.h>
#include <signal.h>

44
glibc-upstream-2.39-142.patch Normal file
View File

@ -0,0 +1,44 @@
commit 2c882bf9c15d206aaf04766d1b8e3ae5b1002cc2
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Thu Dec 5 08:39:44 2024 +0800
math: Exclude internal math symbols for tests [BZ #32414]
Since internal tests don't have access to internal symbols in libm,
exclude them for internal tests. Also make tst-strtod5 and tst-strtod5i
depend on $(libm) to support older versions of GCC which can't inline
copysign family functions. This fixes BZ #32414.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
Reviewed-by: Sunil K Pandey <skpgkp2@gmail.com>
(cherry picked from commit 5df09b444835fca6e64b3d4b4a5beb19b3b2ba21)
diff --git a/include/math.h b/include/math.h
index fa11a710a6c152a4..035fd160ffb9e032 100644
--- a/include/math.h
+++ b/include/math.h
@@ -130,7 +130,10 @@ fabsf128 (_Float128 x)
}
# endif
-# if !(defined __FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
+
+/* NB: Internal tests don't have access to internal symbols. */
+# if !IS_IN (testsuite_internal) \
+ && !(defined __FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
# ifndef NO_MATH_REDIRECT
/* Declare some functions for use within GLIBC. Compilers typically
inline those functions as a single instruction. Use an asm to
diff --git a/stdlib/Makefile b/stdlib/Makefile
index 70d7291c6e3454a8..ff1418f5bb2ea5c9 100644
--- a/stdlib/Makefile
+++ b/stdlib/Makefile
@@ -607,6 +607,8 @@ $(objpfx)bug-strtod2: $(libm)
$(objpfx)tst-strtod-round: $(libm)
$(objpfx)tst-tininess: $(libm)
$(objpfx)tst-strtod-underflow: $(libm)
+$(objpfx)tst-strtod5: $(libm)
+$(objpfx)tst-strtod5i: $(libm)
$(objpfx)tst-strtod6: $(libm)
$(objpfx)tst-strtod-nan-locale: $(libm)
$(objpfx)tst-strtod-nan-sign: $(libm)

30
glibc-upstream-2.39-143.patch Normal file
View File

@ -0,0 +1,30 @@
commit 2c8a7f14fac3628b6a06cc76cdfda54a7ac20386
Author: Florian Weimer <fweimer@redhat.com>
Date: Tue Dec 17 18:12:03 2024 +0100
x86: Avoid integer truncation with large cache sizes (bug 32470)
Some hypervisors report 1 TiB L3 cache size. This results
in some variables incorrectly getting zeroed, causing crashes
in memcpy/memmove because invariants are violated.
(cherry picked from commit 61c3450db96dce96ad2b24b4f0b548e6a46d68e5)
diff --git a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
index 5a98f70364220da4..1f68968a9a457586 100644
--- a/sysdeps/x86/dl-cacheinfo.h
+++ b/sysdeps/x86/dl-cacheinfo.h
@@ -959,11 +959,11 @@ dl_init_cacheinfo (struct cpu_features *cpu_features)
non_temporal_threshold = maximum_non_temporal_threshold;
/* NB: The REP MOVSB threshold must be greater than VEC_SIZE * 8. */
- unsigned int minimum_rep_movsb_threshold;
+ unsigned long int minimum_rep_movsb_threshold;
/* NB: The default REP MOVSB threshold is 4096 * (VEC_SIZE / 16) for
VEC_SIZE == 64 or 32. For VEC_SIZE == 16, the default REP MOVSB
threshold is 2048 * (VEC_SIZE / 16). */
- unsigned int rep_movsb_threshold;
+ unsigned long int rep_movsb_threshold;
if (CPU_FEATURE_USABLE_P (cpu_features, AVX512F)
&& !CPU_FEATURE_PREFERRED_P (cpu_features, Prefer_No_AVX512))
{

250
glibc-upstream-2.39-144.patch Normal file
View File

@ -0,0 +1,250 @@
commit 61daaa76390e0ff73eade3a688d3626b7e7e0c20
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Fri May 24 12:38:50 2024 -0500
x86: Improve large memset perf with non-temporal stores [RHEL-29312]
Previously we use `rep stosb` for all medium/large memsets. This is
notably worse than non-temporal stores for large (above a
few MBs) memsets.
See:
https://docs.google.com/spreadsheets/d/1opzukzvum4n6-RUVHTGddV6RjAEil4P2uMjjQGLbLcU/edit?usp=sharing
For data using different stategies for large memset on ICX and SKX.
Using non-temporal stores can be up to 3x faster on ICX and 2x faster
on SKX. Historically, these numbers would not have been so good
because of the zero-over-zero writeback optimization that `rep stosb`
is able to do. But, the zero-over-zero writeback optimization has been
removed as a potential side-channel attack, so there is no longer any
good reason to only rely on `rep stosb` for large memsets. On the flip
size, non-temporal writes can avoid data in their RFO requests saving
memory bandwidth.
All of the other changes to the file are to re-organize the
code-blocks to maintain "good" alignment given the new code added in
the `L(stosb_local)` case.
The results from running the GLIBC memset benchmarks on TGL-client for
N=20 runs:
Geometric Mean across the suite New / Old EXEX256: 0.979
Geometric Mean across the suite New / Old EXEX512: 0.979
Geometric Mean across the suite New / Old AVX2 : 0.986
Geometric Mean across the suite New / Old SSE2 : 0.979
Most of the cases are essentially unchanged, this is mostly to show
that adding the non-temporal case didn't add any regressions to the
other cases.
The results on the memset-large benchmark suite on TGL-client for N=20
runs:
Geometric Mean across the suite New / Old EXEX256: 0.926
Geometric Mean across the suite New / Old EXEX512: 0.925
Geometric Mean across the suite New / Old AVX2 : 0.928
Geometric Mean across the suite New / Old SSE2 : 0.924
So roughly a 7.5% speedup. This is lower than what we see on servers
(likely because clients typically have faster single-core bandwidth so
saving bandwidth on RFOs is less impactful), but still advantageous.
Full test-suite passes on x86_64 w/ and w/o multiarch.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 5bf0ab80573d66e4ae5d94b094659094336da90f)
diff --git a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
index 97839a22483b0613..637caadb406b2544 100644
--- a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
+++ b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
@@ -21,10 +21,13 @@
2. If size is less than VEC, use integer register stores.
3. If size is from VEC_SIZE to 2 * VEC_SIZE, use 2 VEC stores.
4. If size is from 2 * VEC_SIZE to 4 * VEC_SIZE, use 4 VEC stores.
- 5. On machines ERMS feature, if size is greater or equal than
- __x86_rep_stosb_threshold then REP STOSB will be used.
- 6. If size is more to 4 * VEC_SIZE, align to 4 * VEC_SIZE with
- 4 VEC stores and store 4 * VEC at a time until done. */
+ 5. If size is more to 4 * VEC_SIZE, align to 1 * VEC_SIZE with
+ 4 VEC stores and store 4 * VEC at a time until done.
+ 6. On machines ERMS feature, if size is range
+ [__x86_rep_stosb_threshold, __x86_shared_non_temporal_threshold)
+ then REP STOSB will be used.
+ 7. If size >= __x86_shared_non_temporal_threshold, use a
+ non-temporal stores. */
#include <sysdep.h>
@@ -147,6 +150,41 @@ L(entry_from_wmemset):
VMOVU %VMM(0), -VEC_SIZE(%rdi,%rdx)
VMOVU %VMM(0), (%rdi)
VZEROUPPER_RETURN
+
+ /* If have AVX512 mask instructions put L(less_vec) close to
+ entry as it doesn't take much space and is likely a hot target. */
+#ifdef USE_LESS_VEC_MASK_STORE
+ /* Align to ensure the L(less_vec) logic all fits in 1x cache lines. */
+ .p2align 6,, 47
+ .p2align 4
+L(less_vec):
+L(less_vec_from_wmemset):
+ /* Less than 1 VEC. */
+# if VEC_SIZE != 16 && VEC_SIZE != 32 && VEC_SIZE != 64
+# error Unsupported VEC_SIZE!
+# endif
+ /* Clear high bits from edi. Only keeping bits relevant to page
+ cross check. Note that we are using rax which is set in
+ MEMSET_VDUP_TO_VEC0_AND_SET_RETURN as ptr from here on out. */
+ andl $(PAGE_SIZE - 1), %edi
+ /* Check if VEC_SIZE store cross page. Mask stores suffer
+ serious performance degradation when it has to fault suppress. */
+ cmpl $(PAGE_SIZE - VEC_SIZE), %edi
+ /* This is generally considered a cold target. */
+ ja L(cross_page)
+# if VEC_SIZE > 32
+ movq $-1, %rcx
+ bzhiq %rdx, %rcx, %rcx
+ kmovq %rcx, %k1
+# else
+ movl $-1, %ecx
+ bzhil %edx, %ecx, %ecx
+ kmovd %ecx, %k1
+# endif
+ vmovdqu8 %VMM(0), (%rax){%k1}
+ VZEROUPPER_RETURN
+#endif
+
#if defined USE_MULTIARCH && IS_IN (libc)
END (MEMSET_SYMBOL (__memset, unaligned))
@@ -185,54 +223,6 @@ L(last_2x_vec):
#endif
VZEROUPPER_RETURN
- /* If have AVX512 mask instructions put L(less_vec) close to
- entry as it doesn't take much space and is likely a hot target.
- */
-#ifdef USE_LESS_VEC_MASK_STORE
- .p2align 4,, 10
-L(less_vec):
-L(less_vec_from_wmemset):
- /* Less than 1 VEC. */
-# if VEC_SIZE != 16 && VEC_SIZE != 32 && VEC_SIZE != 64
-# error Unsupported VEC_SIZE!
-# endif
- /* Clear high bits from edi. Only keeping bits relevant to page
- cross check. Note that we are using rax which is set in
- MEMSET_VDUP_TO_VEC0_AND_SET_RETURN as ptr from here on out. */
- andl $(PAGE_SIZE - 1), %edi
- /* Check if VEC_SIZE store cross page. Mask stores suffer
- serious performance degradation when it has to fault suppress.
- */
- cmpl $(PAGE_SIZE - VEC_SIZE), %edi
- /* This is generally considered a cold target. */
- ja L(cross_page)
-# if VEC_SIZE > 32
- movq $-1, %rcx
- bzhiq %rdx, %rcx, %rcx
- kmovq %rcx, %k1
-# else
- movl $-1, %ecx
- bzhil %edx, %ecx, %ecx
- kmovd %ecx, %k1
-# endif
- vmovdqu8 %VMM(0), (%rax){%k1}
- VZEROUPPER_RETURN
-
-# if defined USE_MULTIARCH && IS_IN (libc)
- /* Include L(stosb_local) here if including L(less_vec) between
- L(stosb_more_2x_vec) and ENTRY. This is to cache align the
- L(stosb_more_2x_vec) target. */
- .p2align 4,, 10
-L(stosb_local):
- movzbl %sil, %eax
- mov %RDX_LP, %RCX_LP
- mov %RDI_LP, %RDX_LP
- rep stosb
- mov %RDX_LP, %RAX_LP
- VZEROUPPER_RETURN
-# endif
-#endif
-
#if defined USE_MULTIARCH && IS_IN (libc)
.p2align 4
L(stosb_more_2x_vec):
@@ -318,21 +308,33 @@ L(return_vzeroupper):
ret
#endif
- .p2align 4,, 10
-#ifndef USE_LESS_VEC_MASK_STORE
-# if defined USE_MULTIARCH && IS_IN (libc)
+#ifdef USE_WITH_AVX2
+ .p2align 4
+#else
+ .p2align 4,, 4
+#endif
+
+#if defined USE_MULTIARCH && IS_IN (libc)
/* If no USE_LESS_VEC_MASK put L(stosb_local) here. Will be in
range for 2-byte jump encoding. */
L(stosb_local):
+ cmp __x86_shared_non_temporal_threshold(%rip), %RDX_LP
+ jae L(nt_memset)
movzbl %sil, %eax
mov %RDX_LP, %RCX_LP
mov %RDI_LP, %RDX_LP
rep stosb
+# if (defined USE_WITH_SSE2) || (defined USE_WITH_AVX512)
+ /* Use xchg to save 1-byte (this helps align targets below). */
+ xchg %RDX_LP, %RAX_LP
+# else
mov %RDX_LP, %RAX_LP
- VZEROUPPER_RETURN
# endif
+ VZEROUPPER_RETURN
+#endif
+#ifndef USE_LESS_VEC_MASK_STORE
/* Define L(less_vec) only if not otherwise defined. */
- .p2align 4
+ .p2align 4,, 12
L(less_vec):
/* Broadcast esi to partial register (i.e VEC_SIZE == 32 broadcast to
xmm). This is only does anything for AVX2. */
@@ -423,4 +425,35 @@ L(between_2_3):
movb %SET_REG8, -1(%LESS_VEC_REG, %rdx)
#endif
ret
-END (MEMSET_SYMBOL (__memset, unaligned_erms))
+
+#if defined USE_MULTIARCH && IS_IN (libc)
+# ifdef USE_WITH_AVX512
+ /* Force align so the loop doesn't cross a cache-line. */
+ .p2align 4
+# endif
+ .p2align 4,, 7
+ /* Memset using non-temporal stores. */
+L(nt_memset):
+ VMOVU %VMM(0), (VEC_SIZE * 0)(%rdi)
+ leaq (VEC_SIZE * -4)(%rdi, %rdx), %rdx
+ /* Align DST. */
+ orq $(VEC_SIZE * 1 - 1), %rdi
+ incq %rdi
+ .p2align 4,, 7
+L(nt_loop):
+ VMOVNT %VMM(0), (VEC_SIZE * 0)(%rdi)
+ VMOVNT %VMM(0), (VEC_SIZE * 1)(%rdi)
+ VMOVNT %VMM(0), (VEC_SIZE * 2)(%rdi)
+ VMOVNT %VMM(0), (VEC_SIZE * 3)(%rdi)
+ subq $(VEC_SIZE * -4), %rdi
+ cmpq %rdx, %rdi
+ jb L(nt_loop)
+ sfence
+ VMOVU %VMM(0), (VEC_SIZE * 0)(%rdx)
+ VMOVU %VMM(0), (VEC_SIZE * 1)(%rdx)
+ VMOVU %VMM(0), (VEC_SIZE * 2)(%rdx)
+ VMOVU %VMM(0), (VEC_SIZE * 3)(%rdx)
+ VZEROUPPER_RETURN
+#endif
+
+END(MEMSET_SYMBOL(__memset, unaligned_erms))

143
glibc-upstream-2.39-145.patch Normal file
View File

@ -0,0 +1,143 @@
commit 994b129a35ca5218ecddd1add74aea68f1314560
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Fri Sep 27 15:50:10 2024 -0700
x86/string: Fixup alignment of main loop in str{n}cmp-evex [BZ #32212]
The loop should be aligned to 32-bytes so that it can ideally run out
the DSB. This is particularly important on Skylake-Server where
deficiencies in it's DSB implementation make it prone to not being
able to run loops out of the DSB.
For example running strcmp-evex on 200Mb string:
32-byte aligned loop:
- 43,399,578,766 idq.dsb_uops
not 32-byte aligned loop:
- 6,060,139,704 idq.dsb_uops
This results in a 25% performance degradation for the non-aligned
version.
The fix is to just ensure the code layout is such that the loop is
aligned. (Which was previously the case but was accidentally dropped
in 84e7c46df).
NB: The fix was actually 64-byte alignment. This is because 64-byte
alignment generally produces more stable performance than 32-byte
aligned code (cache line crosses can affect perf), so if we are going
past 16-byte alignmnent, might as well go to 64. 64-byte alignment
also matches most other functions we over-align, so it creates a
common point of optimization.
Times are reported as ratio of Time_With_Patch /
Time_Without_Patch. Lower is better.
The values being reported is the geometric mean of the ratio across
all tests in bench-strcmp and bench-strncmp.
Note this patch is only attempting to improve the Skylake-Server
strcmp for long strings. The rest of the numbers are only to test for
regressions.
Tigerlake Results Strings <= 512:
strcmp : 1.026
strncmp: 0.949
Tigerlake Results Strings > 512:
strcmp : 0.994
strncmp: 0.998
Skylake-Server Results Strings <= 512:
strcmp : 0.945
strncmp: 0.943
Skylake-Server Results Strings > 512:
strcmp : 0.778
strncmp: 1.000
The 2.6% regression on TGL-strcmp is due to slowdowns caused by
changes in alignment of code handling small sizes (most on the
page-cross logic). These should be safe to ignore because 1) We
previously only 16-byte aligned the function so this behavior is not
new and was essentially up to chance before this patch and 2) this
type of alignment related regression on small sizes really only comes
up in tight micro-benchmark loops and is unlikely to have any affect
on realworld performance.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 483443d3211532903d7e790211af5a1d55fdb1f3)
diff --git a/sysdeps/x86_64/multiarch/strcmp-evex.S b/sysdeps/x86_64/multiarch/strcmp-evex.S
index 06730ab2a18f72a0..cea034f394ab45e2 100644
--- a/sysdeps/x86_64/multiarch/strcmp-evex.S
+++ b/sysdeps/x86_64/multiarch/strcmp-evex.S
@@ -209,7 +209,9 @@
returned. */
.section SECTION(.text), "ax", @progbits
- .align 16
+ /* Align 64 bytes here. This is to get the L(loop) block ideally
+ aligned for the DSB. */
+ .align 64
.type STRCMP, @function
.globl STRCMP
# ifdef USE_AS_STRCASECMP_L
@@ -509,9 +511,7 @@ L(ret4):
ret
# endif
- /* 32 byte align here ensures the main loop is ideally aligned
- for DSB. */
- .p2align 5
+ .p2align 4,, 4
L(more_3x_vec):
/* Safe to compare 4x vectors. */
VMOVU (VEC_SIZE)(%rdi), %VMM(0)
@@ -1426,10 +1426,9 @@ L(less_32_till_page):
L(ret_zero_page_cross_slow_case0):
xorl %eax, %eax
ret
-# endif
-
-
+# else
.p2align 4,, 10
+# endif
L(less_16_till_page):
cmpl $((VEC_SIZE - 8) / SIZE_OF_CHAR), %eax
ja L(less_8_till_page)
@@ -1482,8 +1481,12 @@ L(less_16_till_page):
# endif
jmp L(prepare_loop_aligned)
-
-
+# ifndef USE_AS_STRNCMP
+ /* Fits in aligning bytes. */
+L(ret_zero_4_loop):
+ xorl %eax, %eax
+ ret
+# endif
.p2align 4,, 10
L(less_8_till_page):
@@ -1554,6 +1557,7 @@ L(ret_less_8_wcs):
# ifdef USE_AS_STRNCMP
.p2align 4,, 2
+L(ret_zero_4_loop):
L(ret_zero_page_cross_slow_case1):
xorl %eax, %eax
ret
@@ -1586,10 +1590,6 @@ L(less_4_loop):
subq $-(CHAR_PER_VEC * 4), %rdx
# endif
jmp L(prepare_loop_aligned)
-
-L(ret_zero_4_loop):
- xorl %eax, %eax
- ret
L(ret_less_4_loop):
xorl %r8d, %eax
subl %r8d, %eax

57
glibc-upstream-2.39-146.patch Normal file
View File

@ -0,0 +1,57 @@
commit 808a84a8b81468b517a4d721fdc62069cb8c211f
Author: Siddhesh Poyarekar <siddhesh@sourceware.org>
Date: Tue Jan 21 16:11:06 2025 -0500
Fix underallocation of abort_msg_s struct (CVE-2025-0395)
Include the space needed to store the length of the message itself, in
addition to the message string. This resolves BZ #32582.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed: Adhemerval Zanella <adhemerval.zanella@linaro.org>
(cherry picked from commit 68ee0f704cb81e9ad0a78c644a83e1e9cd2ee578)
diff --git a/assert/assert.c b/assert/assert.c
index c29629f5f68921a0..b6e37d694cf4b779 100644
--- a/assert/assert.c
+++ b/assert/assert.c
@@ -18,6 +18,7 @@
#include <assert.h>
#include <atomic.h>
#include <ldsodefs.h>
+#include <libc-pointer-arith.h>
#include <libintl.h>
#include <stdio.h>
#include <stdlib.h>
@@ -65,7 +66,8 @@ __assert_fail_base (const char *fmt, const char *assertion, const char *file,
(void) __fxprintf (NULL, "%s", str);
(void) fflush (stderr);
- total = (total + 1 + GLRO(dl_pagesize) - 1) & ~(GLRO(dl_pagesize) - 1);
+ total = ALIGN_UP (total + sizeof (struct abort_msg_s) + 1,
+ GLRO(dl_pagesize));
struct abort_msg_s *buf = __mmap (NULL, total, PROT_READ | PROT_WRITE,
MAP_ANON | MAP_PRIVATE, -1, 0);
if (__glibc_likely (buf != MAP_FAILED))
diff --git a/sysdeps/posix/libc_fatal.c b/sysdeps/posix/libc_fatal.c
index f9e3425e04496a26..089c47b04b8af049 100644
--- a/sysdeps/posix/libc_fatal.c
+++ b/sysdeps/posix/libc_fatal.c
@@ -20,6 +20,7 @@
#include <errno.h>
#include <fcntl.h>
#include <ldsodefs.h>
+#include <libc-pointer-arith.h>
#include <paths.h>
#include <stdarg.h>
#include <stdbool.h>
@@ -105,7 +106,8 @@ __libc_message_impl (const char *fmt, ...)
{
WRITEV_FOR_FATAL (fd, iov, iovcnt, total);
- total = (total + 1 + GLRO(dl_pagesize) - 1) & ~(GLRO(dl_pagesize) - 1);
+ total = ALIGN_UP (total + sizeof (struct abort_msg_s) + 1,
+ GLRO(dl_pagesize));
struct abort_msg_s *buf = __mmap (NULL, total,
PROT_READ | PROT_WRITE,
MAP_ANON | MAP_PRIVATE, -1, 0);

24
glibc.spec
View File

@ -145,7 +145,7 @@ Version: %{glibcversion}
# - It allows using the Release number without the %%dist tag in the dependency # - It allows using the Release number without the %%dist tag in the dependency
# generator to make the generated requires interchangeable between Rawhide # generator to make the generated requires interchangeable between Rawhide
# and ELN (.elnYY < .fcXX). # and ELN (.elnYY < .fcXX).
%global baserelease 33 %global baserelease 34
Release: %{baserelease}%{?dist} Release: %{baserelease}%{?dist}
# Licenses: # Licenses:
@ -495,6 +495,15 @@ Patch177: glibc-RHEL-71530-7.patch
Patch178: glibc-RHEL-71530-8.patch Patch178: glibc-RHEL-71530-8.patch
Patch179: glibc-RHEL-71530-9.patch Patch179: glibc-RHEL-71530-9.patch
Patch180: glibc-RHEL-71530-10.patch Patch180: glibc-RHEL-71530-10.patch
Patch181: glibc-upstream-2.39-138.patch
Patch182: glibc-upstream-2.39-139.patch
Patch183: glibc-upstream-2.39-140.patch
Patch184: glibc-upstream-2.39-141.patch
Patch185: glibc-upstream-2.39-142.patch
Patch186: glibc-upstream-2.39-143.patch
Patch187: glibc-upstream-2.39-144.patch
Patch188: glibc-upstream-2.39-145.patch
Patch189: glibc-upstream-2.39-146.patch
############################################################################## ##############################################################################
# Continued list of core "glibc" package information: # Continued list of core "glibc" package information:
@ -2490,6 +2499,19 @@ update_gconv_modules_cache ()
%endif %endif
%changelog %changelog
* Thu Jan 23 2025 Florian Weimer <fweimer@redhat.com> - 2.39-34
- Sync with upstream branch release/2.39/master,
commit 808a84a8b81468b517a4d721fdc62069cb8c211f:
- Fix underallocation of abort_msg_s struct (CVE-2025-0395)
- x86/string: Fixup alignment of main loop in str{n}cmp-evex [BZ #32212]
- x86: Improve large memset perf with non-temporal stores [RHEL-29312]
- x86: Avoid integer truncation with large cache sizes (bug 32470)
- math: Exclude internal math symbols for tests [BZ #32414]
- malloc: add indirection for malloc(-like) functions in tests [BZ #32366]
- Pass -nostdlib -nostartfiles together with -r [BZ #31753]
- nptl: initialize cpu_id_start prior to rseq registration
- nptl: initialize rseq area prior to registration
* Mon Dec 23 2024 Florian Weimer <fweimer@redhat.com> - 2.39-33 * Mon Dec 23 2024 Florian Weimer <fweimer@redhat.com> - 2.39-33
- Support in-place file conversion in the iconv tool (RHEL-71530) - Support in-place file conversion in the iconv tool (RHEL-71530)