Amrita H S [Fri, 15 Dec 2023 16:48:17 +0000 (11:48 -0500)]
powerpc: Fix performance issues of strcmp power10
Current implementation of strcmp for power10 has
performance regression for multiple small sizes
and alignment combination.
Most of these performance issues are fixed by this
patch. The compare loop is unrolled and page crosses
of unrolled loop is handled.
Thanks to Paul E. Murphy for helping in fixing the
performance issues.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com> Co-Authored-By: Paul E. Murphy <murphyp@linux.ibm.com> Reviewed-by: Rajalakshmi Srinivasaraghavan <rajis@linux.ibm.com>
MAHESH BODAPATI [Tue, 12 Dec 2023 14:52:45 +0000 (08:52 -0600)]
powerpc : Add optimized memchr for POWER10
Optimized memchr for POWER10 based on existing rawmemchr and strlen.
Reordering instructions and loop unrolling helped in getting better performance. Reviewed-by: Rajalakshmi Srinivasaraghavan <rajis@linux.ibm.com>
Bruno Haible [Tue, 12 Dec 2023 08:45:16 +0000 (09:45 +0100)]
intl: Treat C.UTF-8 locale like C locale, part 2 (BZ# 16621)
The previous commit was incomplete: gettext() still returns a translation
if the file /usr/share/locale/C/LC_MESSAGES/<domain>.mo exists. This patch
prohibits the translation also in this case.
* gettext-runtime/intl/dcigettext.c (DCIGETTEXT): Treat C.<encoding> locale
like the C locale.
Ludwig Rydberg [Mon, 11 Dec 2023 12:50:41 +0000 (13:50 +0100)]
resolv: Fix a few unaligned accesses to fields in HEADER
After refactoring the alloca usage in 40c0add7d4 ("resolve: Remove
__res_context_query alloca usage") a few unaligned accesses to HEADER
fields surfaced. These unaligned accesses led to problems when running
the resolv test suite on sparc32-linux (leon) as many tests failed due to
SIGBUS crashes.
The issue(s) occured during T_QUERY_A_AND_AAAA queries as the second query
now can start on an unaligned address (previously it was explicitly aligned).
With this patch the unaligned accesses are now fixed by using the
UHEADER instead to ensure the fields are accessed with byte
loads/stores.
The patch has been verfied by running the resolv test suite on sparc32
and x86_64.
Signed-off-by: Ludwig Rydberg <ludwig.rydberg@gaisler.com> Signed-off-by: Andreas Larsson <andreas@gaisler.com> Reviewed-by: Florian Weimer <fweimer@redhat.com>
H.J. Lu [Thu, 7 Dec 2023 21:05:39 +0000 (13:05 -0800)]
x86: Check PT_GNU_PROPERTY early
The PT_GNU_PROPERTY segment is scanned before PT_NOTE. For binaries
with the PT_GNU_PROPERTY segment, we can check it to avoid scan of
the PT_NOTE segment. Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
Florian Weimer [Fri, 8 Dec 2023 08:51:34 +0000 (09:51 +0100)]
elf: Initialize GLRO(dl_lazy) before relocating libc in dynamic startup
GLRO(dl_lazy) is used to set the parameters for the early
_dl_relocate_object call, so the consider_profiling setting has to
be applied before the call.
One of the requirements to becoming a CVE Numbering Authority (CNA) is
to publish advisories. Do this by maintaining a file for each CVE fixed
in the advisories directory in the source tree. Links to the advisories
can then be shared as:
The file format at the moment is rudimentary and derives from the git
commit format, i.e. a subject line and a potentially multi-paragraph
description and then tags to describe some meta information. This is a
loose format at the moment and could change as we evolve this.
Also add a script process-fixed-cves.sh that processes these advisories
and generates a list to add to NEWS at release time.
Amrita H S [Wed, 6 Dec 2023 16:43:11 +0000 (11:43 -0500)]
powerpc: Optimized strcmp for power10
This patch is based on __strcmp_power9 and __strlen_power10.
Improvements from __strcmp_power9:
1. Uses new POWER10 instructions
- This code uses lxvp to decrease contention on load
by loading 32 bytes per instruction.
2. Performance implication
- This version has around 30% better performance on average.
- Performance regression is seen for a specific combination
of sizes and alignments. Some of them is observed without
changes also, while rest may be induced by the patch.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com> Reviewed-by: Paul E. Murphy <murphyp@linux.ibm.com>
It splits between process_envvars_secure and process_envvars_default,
with the former used to process arguments for __libc_enable_secure.
It does not have any semantic change, just simplify the code so there
is no need to handle __libc_enable_secure on each len switch.
Checked on x86_64-linux-gnu and aarch64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Loader already ignores LD_DEBUG, LD_DEBUG_OUTPUT, and
LD_TRACE_LOADED_OBJECTS. Both LD_WARN and LD_VERBOSE are similar to
LD_DEBUG, in the sense they enable additional checks and debug
information, so it makes sense to disable them.
Also add both LD_VERBOSE and LD_WARN on filtered environment variables
for setuid binaries.
Checked on x86_64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Call the document a "Security Policy" to disambiguate it from the
security *process* documented in the security page. Also, point to the
security page for bug reporting and CVE assignment.
Andreas Schwab [Thu, 23 Nov 2023 17:23:46 +0000 (18:23 +0100)]
aarch64: correct CFI in rawmemchr (bug 31113)
The .cfi_return_column directive changes the return column for the whole
FDE range. But the actual intent is to tell the unwinder that the value
in x30 (lr) now resides in x15 after the move, and that is expressed by
the .cfi_register directive.
Joe Ramsay [Fri, 1 Dec 2023 09:49:45 +0000 (09:49 +0000)]
math: Add new exp10 implementation
New implementation is based on the existing exp/exp2, with different
reduction constants and polynomial. Worst-case error in round-to-
nearest is 0.513 ULP.
The exp/exp2 shared table is reused for exp10 - .rodata size of
e_exp_data increases by 64 bytes.
As for exp/exp2, targets with single-instruction rounding/conversion
intrinsics can use them by toggling TOINT_INTRINSICS=1 and adding the
necessary code to their math_private.h.
Improvements on Neoverse V1 compared to current GLIBC master:
exp10 thruput: 3.3x in [-0x1.439b746e36b52p+8 0x1.34413509f79ffp+8]
exp10 latency: 1.8x in [-0x1.439b746e36b52p+8 0x1.34413509f79ffp+8]
Tested on:
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction)
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Flavio Cruz [Fri, 24 Nov 2023 21:30:38 +0000 (16:30 -0500)]
Update code to handle the new ABI for sending inlined port rights.
For i686, this change is no op but for x86_64 it forces all inlined port
rights to be 8 bytes long.
Message-ID: <20231124213041.952886-2-flaviocruz@gmail.com>
It is not strictly required by the POSIX, since O_PATH is a Linux
extension, but it is QoI to fail early instead of at readdir. Also
the check is free, since fdopendir already checks if the file
descriptor is opened for read.
Stefan Liebler [Tue, 21 Nov 2023 12:33:03 +0000 (13:33 +0100)]
Avoid padding in _init and _fini. [BZ #31042]
The linker just concatenates the .init and .fini sections which
results in the complete _init and _fini functions. If needed the
linker adds padding bytes due to an alignment. GNU ld is adding
NOPs, which is fine. But e.g. mold is adding traps which results
in broken _init and _fini functions.
Thus this patch removes the alignment in .init and .fini sections
in crtn.S files.
We keep the 4 byte function alignment in crti.S files. As the
assembler now also outputs the start of _init and _fini functions
as multiples of 4 byte, it perhaps has to fill it. Although GNU as
is using NOPs here, to be sure, we just keep the alignment with
0x07 (=NOPs) at the end of crti.S.
In order to avoid an obvious NOP slide in _fini, this patch also
uses an lg instead of lgr instruction. Then the emitted instructions
needs a multiple of 4 bytes.
malloc: Improve MAP_HUGETLB with glibc.malloc.hugetlb=2
Even for explicit large page support, allocation might use mmap without
the hugepage bit set if the requested size is smaller than
mmap_threshold. For this case where mmap is issued, MAP_HUGETLB is set
iff the allocation size is larger than the used large page.
To force such allocations to use large pages, also tune the mmap_threhold
(if it is not explicit set by a tunable). This forces allocation to
follow the sbrk path, which will fall back to mmap (which will try large
pages before galling back to default mmap).
Checked on x86_64-linux-gnu. Reviewed-by: DJ Delorie <dj@redhat.com> Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
elf: Add a way to check if tunable is set (BZ 27069)
The patch adds two new macros, TUNABLE_GET_DEFAULT and TUNABLE_IS_INITIALIZED,
here the former get the default value with a signature similar to
TUNABLE_GET, while the later returns whether the tunable was set by
the environment variable.
Checked on x86_64-linux-gnu. Reviewed-by: DJ Delorie <dj@redhat.com> Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Noah Goldstein [Wed, 1 Nov 2023 20:30:26 +0000 (15:30 -0500)]
x86: Only align destination to 1x VEC_SIZE in memset 4x loop
Current code aligns to 2x VEC_SIZE. Aligning to 2x has no affect on
performance other than potentially resulting in an additional
iteration of the loop.
1x maintains aligned stores (the only reason to align in this case)
and doesn't incur any unnecessary loop iterations. Reviewed-by: Sunil K Pandey <skpgkp2@gmail.com>
_dl_assign_tls_modid() assigns a slotinfo entry for a new module, but
does *not* do anything to the generation counter. The first time this
happens, the generation is zero and map_generation() returns the current
generation to be used during relocation processing. However, if
a slotinfo entry is later reused, it will already have a generation
assigned. If this generation has fallen behind the current global max
generation, then this causes an obsolete generation to be assigned
during relocation processing, as map_generation() returns this
generation if nonzero. _dl_add_to_slotinfo() eventually resets the
generation, but by then it is too late. This causes DTV updates to be
skipped, leading to NULL or broken TLS slot pointers and segfaults.
Fix this by resetting the generation to zero in _dl_assign_tls_modid(),
so it behaves the same as the first time a slot is assigned.
_dl_add_to_slotinfo() will still assign the correct static generation
later during module load, but relocation processing will no longer use
an obsolete generation.
Note that slotinfo entry (aka modid) reuse typically happens after a
dlclose and only TLS access via dynamic tlsdesc is affected. Because
tlsdesc is optimized to use the optional part of static TLS, dynamic
tlsdesc can be avoided by increasing the glibc.rtld.optional_static_tls
tunable to a large enough value, or by LD_PRELOAD-ing the affected
modules.
Fixes bug 29039.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Tobias Klauser [Tue, 28 Nov 2023 11:06:43 +0000 (12:06 +0100)]
Add TCP_MD5SIG_FLAG_IFINDEX from Linux 5.6 to netinet/tcp.h.
This patch adds the TCP_MD5SIG_FLAG_IFINDEX constant from Linux 5.6 to
sysdeps/gnu/netinet/tcp.h and updates struct tcp_md5sig accordingly to
contain the device index.
Joseph Myers [Thu, 23 Nov 2023 19:01:32 +0000 (19:01 +0000)]
Remove __access_noerrno
A recent commit, apparently commit 6c6fce572fb8f583f14d898e54fd7d25ae91cf56 "elf: Remove /etc/suid-debug
support", resulted in localplt failures for i686-gnu and x86_64-gnu:
After that commit, __access_noerrno is actually no longer used at all.
So rather than just removing the localplt expectation for that symbol
for Hurd, completely remove all definitions of and references to that
symbol.
Tested for x86_64, and with build-many-glibcs.py for i686-gnu and
x86_64-gnu.
This restore the 2.33 semantic for arena_get2. It was changed by 11a02b035b46 to avoid arena_get2 call malloc (back when __get_nproc
was refactored to use an scratch_buffer - 903bc7dcc2acafc). The
__get_nproc was refactored over then and now it also avoid to call
malloc.
The 11a02b035b46 did not take in consideration any performance
implication, which should have been discussed properly. The
__get_nprocs_sched is still used as a fallback mechanism if procfs
and sysfs is not acessible.
Checked on x86_64-linux-gnu. Reviewed-by: DJ Delorie <dj@redhat.com>
Joe Ramsay [Tue, 21 Nov 2023 14:39:39 +0000 (14:39 +0000)]
aarch64: Fix libmvec benchmarks
These were broken by the new atan2 functions, as they were only
set up for univariate functions. Arity is now detected from the
input file - this revealed a mistake that the double-precision
inputs were being used for both single- and double-precision
routines, which is now remedied.
Florian Weimer [Wed, 22 Nov 2023 07:38:33 +0000 (08:38 +0100)]
posix: Revert the removal of the crypt prototype from <unistd.h>
Many applications still rely on this prototype. Rebuilds without
this prototype result in an implicit function declaration, which can
introduce security vulnerabilities due to 32-bit pointer truncation.
The _dl_non_dynamic_init does not parse LD_PROFILE, which does not
enable profile for dlopen objects. Since dlopen is deprecated for
static objects, it is better to remove the support.
It also allows to trim down libc.a of profile support.
Checked on x86_64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Loader does not ignore LD_PROFILE in secure-execution mode (different
than man-page states [1]), rather it uses a different path
(/var/profile) and ignore LD_PROFILE_OUTPUT.
Allowing secure-execution profiling is already a non good security
boundary, since it enables different code paths and extra OS access by
the process. But by ignoring LD_PROFILE_OUTPUT, the resulting profile
file might also be acceded in a racy manner since the file name does not
use any process-specific information (such as pid, timing, etc.).
Another side-effect is it forces lazy binding even on libraries that
might be with DF_BIND_NOW.
Using the memcmp symbol directly allows the compile to inline the
memcmp calls (especially because _dl_tunable_set_hwcaps uses constants
values), generating better code.
Checked with tst-tunables on s390x-linux-gnu (qemu system). Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
The dl-symbol-redir-ifunc.h redirects compiler-generated libcalls to
arch-specific memory implementations to avoid ifunc calls where it is not
yet possible. The memcmp-isa-default-impl.h aims to fix the same issue
by calling the specific memset implementation directly.
Using the memcmp symbol directly allows the compiler to inline the memset
calls (especially because _dl_tunable_set_hwcaps uses constants values),
generating better code.
Instead of ignoring ill-formatted tunable strings, first, check all the
tunable definitions are correct and then set each tunable value. It
means that partially invalid strings, like "key1=value1:key2=key2=value'
or 'key1=value':key2=value2=value2' do not enable 'key1=value1'. It
avoids possible user-defined errors in tunable definitions.
Checked on x86_64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Tunable definitions with more than one '=' on are parsed and enabled,
and any subsequent '=' are ignored. It means that tunables in the form
'tunable=tunable=value' or 'tunable=value=value' are handled as
'tunable=value'. These inputs are likely user input errors, which
should not be accepted.
Checked on x86_64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Some environment variables allow alteration of allocator behavior
across setuid boundaries, where a setuid program may ignore the
tunable, but its non-setuid child can read it and adjust the memory
allocator behavior accordingly.
Most library behavior tunings is limited to the current process and does
not bleed in scope; so it is unclear how pratical this misfeature is.
If behavior change across privilege boundaries is desirable, it would be
better done with a wrapper program around the non-setuid child that sets
these envvars, instead of using the setuid process as the messenger.
The patch as fixes tst-env-setuid, where it fail if any unsecvars is
set. It also adds a dynamic test, although it requires
--enable-hardcoded-path-in-tests so kernel correctly sets the setuid
bit (using the loader command directly would require to set the
setuid bit on the loader itself, which is not a usual deployment).
Co-authored-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Checked on x86_64-linux-gnu. Reviewed-by: DJ Delorie <dj@redhat.com>
elf: Ignore GLIBC_TUNABLES for setuid/setgid binaries
The tunable privilege levels were a retrofit to try and keep the malloc
tunable environment variables' behavior unchanged across security
boundaries. However, CVE-2023-4911 shows how tricky can be
tunable parsing in a security-sensitive environment.
Not only parsing, but the malloc tunable essentially changes some
semantics on setuid/setgid processes. Although it is not a direct
security issue, allowing users to change setuid/setgid semantics is not
a good security practice, and requires extra code and analysis to check
if each tunable is safe to use on all security boundaries.
It also means that security opt-in features, like aarch64 MTE, would
need to be explicit enabled by an administrator with a wrapper script
or with a possible future system-wide tunable setting.
Co-authored-by: Siddhesh Poyarekar <siddhesh@sourceware.org> Reviewed-by: DJ Delorie <dj@redhat.com>
setuid/setgid process now ignores any glibc tunables, and filters out
all environment variables that might changes its behavior. This patch
also adds GLIBC_TUNABLES, so any spawned process by setuid/setgid
processes should set tunable explicitly.
Since malloc debug support moved to a different library
(libc_malloc_debug.so), the glibc.malloc.check requires preloading the
debug library to enable it. It means that suid-debug support has not
been working since 2.34.
To restore its support, it would require to add additional information
and parsing to where to find libc_malloc_debug.so.
It is one thing less that might change AT_SECURE binaries' behavior
due to environment configurations.
Checked on x86_64-linux-gnu. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Florian Weimer [Tue, 21 Nov 2023 15:45:35 +0000 (16:45 +0100)]
stdlib: The qsort implementation needs to use heapsort in more cases
The existing logic avoided internal stack overflow. To avoid
a denial-of-service condition with adversarial input, it is necessary
to fall over to heapsort if tail-recursing deeply, too, which does
not result in a deep stack of pending partitions.
The new test stdlib/tst-qsort5 is based on Douglas McIlroy's paper
on this subject.
Florian Weimer [Tue, 21 Nov 2023 15:45:35 +0000 (16:45 +0100)]
stdlib: Handle various corner cases in the fallback heapsort for qsort
The previous implementation did not consistently apply the rule that
the child nodes of node K are at 2 * K + 1 and 2 * K + 2, or
that the parent node is at (K - 1) / 2.
Add an internal test that targets the heapsort implementation
directly.
Florian Weimer [Tue, 21 Nov 2023 15:45:35 +0000 (16:45 +0100)]
stdlib: Avoid another self-comparison in qsort
In the insertion phase, we could run off the start of the array if the
comparison function never runs zero. In that case, it never finds the
initial element that terminates the iteration.
linux: Use fchmodat2 on fchmod for flags different than 0 (BZ 26401)
Linux 6.6 (09da082b07bbae1c) added support for fchmodat2, which has
similar semantics as fchmodat with an extra flag argument. This
allows fchmodat to implement AT_SYMLINK_NOFOLLOW and AT_EMPTY_PATH
without the need for procfs.
The syscall is registered on all architectures (with value of 452
except on alpha which is 562, commit 78252deb023cf087).
The tst-lchmod.c requires a small fix where fchmodat checks two
contradictory assertions ('(st.st_mode & 0777) == 2' and
'(st.st_mode & 0777) == 3').
Jan Palus [Mon, 30 Oct 2023 18:57:01 +0000 (19:57 +0100)]
resolv: free only initialized items from gai pool
pool_max_size denotes total allocated rows in pool but possibly not yet
initialized. it's pool_size that represents number of actually occupied
rows hence use it when freeing pool to avoid freeing random addresses.
Signed-off-by: Jan Palus <jpalus@fastmail.com> Reviewed-by: Florian Weimer <fweimer@redhat.com>
Flavio Cruz [Sun, 19 Nov 2023 05:08:47 +0000 (00:08 -0500)]
Remove untyped mach RPC code.
Existing MiG does not support untyped messages and the Hurd will
continue to use typed messages for the foreseeable future.
Message-ID: <ZVmYX6j4pYNUfqn4@jupiter.tail36e24.ts.net>
The `ty` pointer is only set at the end of the loop so that
`msgtl_header.msgt_inline` and `msgtl_header.msgt_deallocate` remain
valid. Also, when deallocating memory, we use the length from the
message directly rather than hard coding mach_port_t since we want to
deallocate any kind of OOL data.
Message-ID: <ZVlGVD6eEN-dXsOr@jupiter.tail36e24.ts.net>
Florian Weimer [Thu, 16 Nov 2023 18:55:35 +0000 (19:55 +0100)]
elf: Fix force_first handling in dlclose (bug 30981)
The force_first parameter was ineffective because the dlclose'd
object was not necessarily the first in the maps array. Also
enable force_first handling unconditionally, regardless of namespace.
The initial object in a namespace should be destructed first, too.
The _dl_sort_maps_dfs function had early returns for relocation
dependency processing which broke force_first handling, too, and
this is fixed in this change as well.
elf: Handle non-directory name in search path (BZ 31035)
The open_path stops if a relative path in search path contains a
component that is a non directory (for instance, if the component
is an existing file).
For instance:
$ cat > lib.c <<EOF
> void foo (void) {}
> EOF
$ gcc -shared -fPIC -o lib.so lib.c
$ cat > main.c <<EOF
extern void foo ();
int main () { foo (); return 0; }
EOF
$ gcc -o main main.c lib.so
$ LD_LIBRARY_PATH=. ./main
$ LD_LIBRARY_PATH=non-existing/path:. ./main
$ LD_LIBRARY_PATH=$(pwd)/main:. ./main
$ LD_LIBRARY_PATH=./main:. ./main
./main: error while loading shared libraries: lib.so: cannot open shared object file: No such file or directory
The invalid './main' should be ignored as a non-existent one,
instead as a valid but non accessible file.
Absolute paths do not trigger this issue because their status are
initialized as 'unknown' and open_path check if this is a directory.
Checked on x86_64-linux-gnu.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Noah Goldstein [Wed, 1 Nov 2023 21:42:07 +0000 (16:42 -0500)]
x86: Fix unchecked AVX512-VBMI2 usage in strrchr-evex-base.S
strrchr-evex-base used `vpcompress{b|d}` in the page cross logic but
was missing the CPU_FEATURE checks for VBMI2 in the
ifunc/ifunc-impl-list.
The fix is either to add those checks or change the logic to not use
`vpcompress{b|d}`. Choosing the latter here so that the strrchr-evex
implementation is usable on SKX.
New implementation is a bit slower, but this is in a cold path so its
probably okay.
Samuel Thibault [Tue, 14 Nov 2023 01:03:35 +0000 (02:03 +0100)]
hurd: Make _hurd_intr_rpc_mach_msg avoid returning MACH_SEND_INTERRUPTED
When the given options do not include MACH_SEND_INTERRUPT,
_hurd_intr_rpc_mach_msg (aka mach_msg) is not supposed to return
MACH_SEND_INTERRUPTED. In such a case we thus have to retry sending the
message.
This was observed to fix various occurrences of spurious
"(ipc/send) interrupted" errors when running haskell programs.
Wilco Dijkstra [Thu, 26 Oct 2023 16:30:36 +0000 (17:30 +0100)]
AArch64: Remove Falkor memcpy
The latest implementations of memcpy are actually faster than the Falkor
implementations [1], so remove the falkor/phecda ifuncs for memcpy and
the now unused IS_FALKOR/IS_PHECDA defines.
Wilco Dijkstra [Thu, 26 Oct 2023 16:07:21 +0000 (17:07 +0100)]
AArch64: Add memset_zva64
Add a specialized memset for the common ZVA size of 64 to avoid the
overhead of reading the ZVA size. Since the code is identical to
__memset_falkor, remove the latter.
Florian Weimer [Wed, 8 Nov 2023 14:18:02 +0000 (15:18 +0100)]
stdlib: Avoid element self-comparisons in qsort
This improves compatibility with applications which assume that qsort
does not invoke the comparison function with equal pointer arguments.
The newly introduced branches should be predictable, as leading to a
call to the comparison function. If the prediction fails, we avoid
calling the function.
The PR_SET_VMA_ANON_NAME support is only enabled through a configurable
kernel switch, mainly because assigning a name to a
anonymous virtual memory area might prevent that area from being
merged with adjacent virtual memory areas.
The kernel will potentially merge both mappings resulting in only one
segment of size 0x800000. If the segment is names with
PR_SET_VMA_ANON_NAME with different names, it results in two mappings.
Although this will unlikely be an issue for pthread stacks and malloc
arenas (since for pthread stacks the guard page will result in
a PROT_NONE segment, similar to the alignment requirement for the arena
block), it still might prevent the mmap memory allocated for detail
malloc.
There is also another potential scalability issue, where the prctl
requires
to take the mmap global lock which is still not fully fixed in Linux
[1] (for pthread stacks and arenas, it is mitigated by the stack
cached and the arena reuse).
So this patch disables anonymous mapping annotations as default and
add a new tunable, glibc.mem.decorate_maps, can be used to enable
it.
[1] https://lwn.net/Articles/906852/
Checked on x86_64-linux-gnu and aarch64-linux-gnu. Reviewed-by: DJ Delorie <dj@redhat.com>