Carlos O'Donell [Fri, 28 Jan 2022 20:14:29 +0000 (15:14 -0500)]
malloc: Fix -Wuse-after-free warning in tst-mallocalign1 [BZ #26779]
The test leaks bits from the freed pointer via the return value
in ret, and the compiler correctly identifies this issue.
We switch the test to use TEST_VERIFY and terminate the test
if any of the pointers return an unexpected alignment.
This fixes another -Wuse-after-free error when compiling glibc
with gcc 12.
Tested on x86_64 and i686 without regression.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Carlos O'Donell [Sun, 30 Jan 2022 14:46:55 +0000 (09:46 -0500)]
Update libc.pot for 2.35 release.
H.J. Lu [Sat, 29 Jan 2022 13:22:31 +0000 (05:22 -0800)]
tst-socket-timestamp-compat.c: Check __TIMESIZE [BZ #28837]
time_t size is defined by __TIMESIZE, not __WORDSIZE. Check __TIMESIZE,
instead of __WORDSIZE, for time_t size. This fixes BZ #28837.
Adhemerval Zanella [Mon, 24 Jan 2022 18:57:55 +0000 (15:57 -0300)]
Add prelink removal plan on NEWS
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Adhemerval Zanella [Thu, 27 Jan 2022 19:45:18 +0000 (16:45 -0300)]
Linux: Only generate 64 bit timestamps for 64 bit time_t recvmsg/recvmmsg
The timestamps created by __convert_scm_timestamps only make sense for
64 bit time_t programs, 32 bit time_t programs will ignore 64 bit time_t
timestamps since SO_TIMESTAMP will be defined to old values (either by
glibc or kernel headers).
Worse, if the buffer is not suffice MSG_CTRUNC is set to indicate it
(which breaks some programs [1]).
This patch makes only 64 bit time_t recvmsg and recvmmsg to call
__convert_scm_timestamps. Also, the assumption to called it is changed
from __ASSUME_TIME64_SYSCALLS to __TIMESIZE != 64 since the setsockopt
might be called by libraries built without __TIME_BITS=64. The
MSG_CTRUNC is only set for the 64 bit symbols, it should happen only
if 64 bit time_t programs run older kernels.
Checked on x86_64-linux-gnu and i686-linux-gnu.
[1] https://github.com/systemd/systemd/pull/20567
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Adhemerval Zanella [Mon, 24 Jan 2022 11:55:53 +0000 (08:55 -0300)]
linux: Fix ancillary 64-bit time timestamp conversion (BZ #28349, BZ#28350)
The __convert_scm_timestamps only updates the control message last
pointer for SOL_SOCKET type, so if the message control buffer contains
multiple ancillary message types the converted timestamp one might
overwrite a valid message.
The test checks if the extra ancillary space is correctly handled
by recvmsg/recvmmsg, where if there is no extra space for the 64-bit
time_t converted message the control buffer should be marked with
MSG_TRUNC. It also check if recvmsg/recvmmsg handle correctly multiple
ancillary data.
Checked on x86_64-linux and on i686-linux-gnu on both 5.11 and
4.15 kernel.
Co-authored-by: Fabian Vogt <fvogt@suse.de>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Adhemerval Zanella [Mon, 13 Dec 2021 14:36:42 +0000 (11:36 -0300)]
support: Add support_socket_so_timestamp_time64
Check if the socket support 64-bit network packages timestamps
(SO_TIMESTAMP and SO_TIMESTAMPNS). This will be used on recvmsg
and recvmmsg tests to check if the timestamp should be generated.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Szabolcs Nagy [Fri, 28 Jan 2022 10:51:49 +0000 (10:51 +0000)]
Fix elf/loadfail test build dependencies
There was no direct or indirect make dependency on testobj3.so so the
test could fail with
/B/elf/loadfail: failed to load shared object: testobj3.so: cannot open
shared object file: No such file or directory
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Florian Weimer [Thu, 27 Jan 2022 15:03:58 +0000 (16:03 +0100)]
Fix glibc 2.34 ABI omission (missing GLIBC_2.34 in dynamic loader)
The glibc 2.34 release really should have added a GLIBC_2.34
symbol to the dynamic loader. With it, we could move functions such
as dlopen or pthread_key_create that work on process-global state
into the dynamic loader (once we have fixed a longstanding issue
with static linking). Without the GLIBC_2.34 symbol, yet another
new symbol version would be needed because old glibc will fail to
load binaries due to the missing symbol version in ld.so that newly
linked programs will require.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
H.J. Lu [Tue, 25 Jan 2022 03:33:43 +0000 (19:33 -0800)]
x86: Use CHECK_FEATURE_PRESENT to check HLE [BZ #27398]
HLE is disabled on blacklisted CPUs. Use CHECK_FEATURE_PRESENT, instead
of CHECK_FEATURE_ACTIVE, to check HLE.
Mark Wielaard [Wed, 26 Jan 2022 17:57:29 +0000 (18:57 +0100)]
Guard tst-valgrind-smoke.out with run-built-tests
Prevent tst-valgrind-smoke from running when run-built-tests is not yes.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Wed, 26 Jan 2022 19:21:04 +0000 (16:21 -0300)]
hurd: Add posix_spawnattr_tc{get,set}pgrp_np on libc.abilist
Commit
342cc934a3bf74ac missed the update-abi for the ABI.
Martin Sebor [Tue, 25 Jan 2022 22:39:38 +0000 (15:39 -0700)]
Avoid -Wuse-after-free in tests [BZ #26779].
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Adhemerval Zanella [Tue, 25 Jan 2022 16:33:13 +0000 (13:33 -0300)]
elf: Replace tst-p_alignmod1-editX with a python script
This avoid the cross-compiling breakage when the test should not run
($(run-built-tests) equal to no).
Checked on x86_64-linux-gnu and i686-linux-gnu as well with a cross
compile to aarch64-linux-gnu and powerpc64-linux-gnu.
Martin Sebor [Wed, 26 Jan 2022 00:39:36 +0000 (17:39 -0700)]
stdlib: Avoid -Wuse-after-free in __add_to_environ [BZ #26779]
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Martin Sebor [Wed, 26 Jan 2022 00:39:02 +0000 (17:39 -0700)]
io: Fix use-after-free in ftw [BZ #26779]
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Martin Sebor [Wed, 26 Jan 2022 00:38:31 +0000 (17:38 -0700)]
intl: Avoid -Wuse-after-free [BZ #26779]
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Martin Sebor [Wed, 26 Jan 2022 00:37:56 +0000 (17:37 -0700)]
elf: Fix use-after-free in ldconfig [BZ #26779]
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Adhemerval Zanella [Mon, 14 Jun 2021 17:41:31 +0000 (14:41 -0300)]
posix: Add terminal control setting support for posix_spawn
Currently there is no proper way to set the controlling terminal through
posix_spawn in race free manner [1]. This forces shell implementations
to keep using fork+exec when launching background process groups,
even when using posix_spawn yields better performance.
This patch adds a new GNU extension so the creating process can
configure the created process terminal group. This is done with a new
flag, POSIX_SPAWN_TCSETPGROUP, along with two new attribute functions:
posix_spawnattr_tcsetpgrp_np, and posix_spawnattr_tcgetpgrp_np.
The function sets a new attribute, spawn-tcgroupfd, that references to
the controlling terminal.
The controlling terminal is set after the spawn-pgroup attribute, and
uses the spawn-tcgroupfd along with current creating process group
(so it is composable with POSIX_SPAWN_SETPGROUP).
To create a process and set the controlling terminal, one can use the
following sequence:
posix_spawnattr_t attr;
posix_spawnattr_init (&attr);
posix_spawnattr_setflags (&attr, POSIX_SPAWN_TCSETPGROUP);
posix_spawnattr_tcsetpgrp_np (&attr, tcfd);
If the idea is also to create a new process groups:
posix_spawnattr_t attr;
posix_spawnattr_init (&attr);
posix_spawnattr_setflags (&attr, POSIX_SPAWN_TCSETPGROUP
| POSIX_SPAWN_SETPGROUP);
posix_spawnattr_tcsetpgrp_np (&attr, tcfd);
posix_spawnattr_setpgroup (&attr, 0);
The controlling terminal file descriptor is ignored if the new flag is
not set.
This interface is slight different than the one provided by QNX [2],
which only provides the POSIX_SPAWN_TCSETPGROUP flag. The QNX
documentation does not specify how the controlling terminal is obtained
nor how it iteracts with POSIX_SPAWN_SETPGROUP. Since a glibc
implementation is library based, it is more straightforward and avoid
requires additional file descriptor operations to request the caller
to setup the controlling terminal file descriptor (and it also allows
a bit less error handling by posix_spawn).
Checked on x86_64-linux-gnu and i686-linux-gnu.
[1] https://github.com/ksh93/ksh/issues/79
[2] https://www.qnx.com/developers/docs/7.0.0/index.html#com.qnx.doc.neutrino.lib_ref/topic/p/posix_spawn.html
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Florian Weimer [Mon, 24 Jan 2022 17:14:24 +0000 (18:14 +0100)]
Linux: Detect user namespace support in io/tst-getcwd-smallbuff
Otherwise the test fails with certain container runtimes.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Andreas Schwab [Tue, 18 Jan 2022 15:31:39 +0000 (16:31 +0100)]
Fix handling of unterminated bracket expressions in fnmatch (bug 28792)
When fnmatch processes a bracket expression, and eventually finds it to be
unterminated, it should rescan it, treating the starting bracket as a
normal character. That didn't happen when a matching character was found
while scanning the bracket expression.
Siddhesh Poyarekar [Mon, 24 Jan 2022 16:06:41 +0000 (21:36 +0530)]
realpath: Avoid overwriting preexisting error (CVE-2021-3998)
Set errno and failure for paths that are too long only if no other error
occurred earlier.
Related: BZ #28770
Reviewed-by: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
H.J. Lu [Tue, 21 Dec 2021 21:42:28 +0000 (13:42 -0800)]
elf: Add a test for PT_LOAD segments with invalid p_align [BZ #28688]
Build tst-p_alignmod3.so with 256 byte page size and verify that it is
rejected with a proper error message.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Tue, 21 Dec 2021 20:25:20 +0000 (12:25 -0800)]
elf: Add a test for PT_LOAD segments with p_align == 1 [BZ #28688]
Add tst-p_alignmod2-edit to edit the copy of tst-p_alignmod-base.so to
set p_align of the first PT_LOAD segment to 1 and verify that the shared
library can be loaded normally.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Tue, 21 Dec 2021 02:33:55 +0000 (18:33 -0800)]
elf: Add a test for PT_LOAD segments with mixed p_align [BZ #28676]
Add tst-p_alignmod1-edit to edit the copy of tst-p_alignmod-base.so to
reduce p_align of the first PT_LOAD segment by half and verify that the
shared library is mapped with the maximum p_align of all PT_LOAD segments.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Fri, 15 Oct 2021 13:33:05 +0000 (06:33 -0700)]
Add and use link-test-modules-rpath-link [BZ #28455]
DT_RUNPATH is only used to find the immediate dependencies of the
executable or shared object containing the DT_RUNPATH entry:
1. Define link-test-modules-rpath-link if $(build-hardcoded-path-in-tests)
is yes.
2. Use $(link-test-modules-rpath-link) in build-module-helper so that
test modules can dlopen modules with DT_RUNPATH.
3. Add a test to show why link-test-modules-rpath-link is needed.
This partially fixes BZ #28455.
Siddhesh Poyarekar [Mon, 24 Jan 2022 05:27:09 +0000 (10:57 +0530)]
tst-realpath-toolong: Fix hurd build
Define PATH_MAX to a constant if it isn't already defined, like in hurd.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Siddhesh Poyarekar [Fri, 21 Jan 2022 18:02:56 +0000 (23:32 +0530)]
getcwd: Set errno to ERANGE for size == 1 (CVE-2021-3999)
No valid path returned by getcwd would fit into 1 byte, so reject the
size early and return NULL with errno set to ERANGE. This change is
prompted by CVE-2021-3999, which describes a single byte buffer
underflow and overflow when all of the following conditions are met:
- The buffer size (i.e. the second argument of getcwd) is 1 byte
- The current working directory is too long
- '/' is also mounted on the current working directory
Sequence of events:
- In sysdeps/unix/sysv/linux/getcwd.c, the syscall returns ENAMETOOLONG
because the linux kernel checks for name length before it checks
buffer size
- The code falls back to the generic getcwd in sysdeps/posix
- In the generic func, the buf[0] is set to '\0' on line 250
- this while loop on line 262 is bypassed:
while (!(thisdev == rootdev && thisino == rootino))
since the rootfs (/) is bind mounted onto the directory and the flow
goes on to line 449, where it puts a '/' in the byte before the
buffer.
- Finally on line 458, it moves 2 bytes (the underflowed byte and the
'\0') to the buf[0] and buf[1], resulting in a 1 byte buffer overflow.
- buf is returned on line 469 and errno is not set.
This resolves BZ #28769.
Reviewed-by: Andreas Schwab <schwab@linux-m68k.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Signed-off-by: Qualys Security Advisory <qsa@qualys.com>
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Alexandra Hájková [Tue, 26 Oct 2021 09:42:44 +0000 (11:42 +0200)]
Add valgrind smoke test
Check if whether valgrind is available in the test environment.
If not, skip the test. Run smoke tests with valgrind to verify dynamic loader.
First, check if algrind works with the system ld.so in the test
environment. Then run the actual test inside the test environment,
using the just build ld.so and new libraries.
Co-authored-by: Mark Wielaard <mark@klomp.org>
Samuel Thibault [Sat, 22 Jan 2022 00:12:05 +0000 (00:12 +0000)]
htl: Fix cleaning the reply port
If any RPC fails, the reply port will already be deallocated.
__pthread_thread_terminate thus has to defer taking its name until the very last
__thread_terminate_release which doesn't reply a message. But then we
have to read from the pthread structure.
This introduces __pthread_dealloc_finish() which does the recording of
the thread termination, so the slot can be reused really only just before
the __thread_terminate_release call. Only the real thread can set it, so
let's decouple this from the pthread_state by just removing the
PTHREAD_TERMINATED state and add a terminated field.
H.J. Lu [Wed, 15 Dec 2021 21:58:33 +0000 (13:58 -0800)]
elf: Properly align all PT_LOAD segments [BZ #28676]
Linker may set p_align of a PT_LOAD segment larger than p_align of the
first PT_LOAD segment to satisfy a section alignment:
Elf file type is DYN (Shared object file)
Entry point 0x0
There are 10 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000000000 0x0000000000000000 0x0000000000000000
0x0000000000000834 0x0000000000000834 R E 0x1000
LOAD 0x0000000000000e00 0x0000000000001e00 0x0000000000001e00
0x0000000000000230 0x0000000000000230 RW 0x1000
LOAD 0x0000000000400000 0x0000000000400000 0x0000000000400000
0x0000000000000004 0x0000000000000008 RW 0x400000
...
Section to Segment mapping:
Segment Sections...
00 .note.gnu.property .note.gnu.build-id .gnu.hash .dynsym .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init .plt .plt.got .text .fini .rodata .eh_frame_hdr .eh_frame
01 .init_array .fini_array .data.rel.ro .dynamic .got .got.plt
02 .data .bss
We should align the first PT_LOAD segment to the maximum p_align of all
PT_LOAD segments, similar to the kernel commit:
commit
ce81bb256a224259ab686742a6284930cbe4f1fa
Author: Chris Kennelly <ckennelly@google.com>
Date: Thu Oct 15 20:12:32 2020 -0700
fs/binfmt_elf: use PT_LOAD p_align values for suitable start address
This fixes BZ #28676.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Siddhesh Poyarekar [Thu, 13 Jan 2022 05:58:36 +0000 (11:28 +0530)]
realpath: Set errno to ENAMETOOLONG for result larger than PATH_MAX [BZ #28770]
realpath returns an allocated string when the result exceeds PATH_MAX,
which is unexpected when its second argument is not NULL. This results
in the second argument (resolved) being uninitialized and also results
in a memory leak since the caller expects resolved to be the same as the
returned value.
Return NULL and set errno to ENAMETOOLONG if the result exceeds
PATH_MAX. This fixes [BZ #28770], which is CVE-2021-3998.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Siddhesh Poyarekar [Tue, 18 Jan 2022 07:59:36 +0000 (13:29 +0530)]
support: Add helpers to create paths longer than PATH_MAX
Add new helpers support_create_and_chdir_toolong_temp_directory and
support_chdir_toolong_temp_directory to create and descend into
directory trees longer than PATH_MAX.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Jangwoong Kim [Tue, 14 Dec 2021 12:30:51 +0000 (21:30 +0900)]
nptl: Effectively skip CAS in spinlock loop
The commit:
"Add LLL_MUTEX_READ_LOCK [BZ #28537]"
SHA1:
d672a98a1af106bd68deb15576710cd61363f7a6
introduced LLL_MUTEX_READ_LOCK, to skip CAS in spinlock loop
if atomic load fails. But, "continue" inside of do-while loop
does not skip the evaluation of escape expression, thus CAS
is not skipped.
Replace do-while with while and skip LLL_MUTEX_TRYLOCK if
LLL_MUTEX_READ_LOCK fails.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Wed, 19 Jan 2022 19:11:55 +0000 (20:11 +0100)]
mips: Move DT_MIPS into <ldsodefs.h>
ELF_MACHINE_XHASH_SETUP in that file needs it.
Fixes commit
c90363403b57b3b7919061851cb3e6d9c85e784a
("elf: Move _dl_setup_hash to its own file").
Sunil K Pandey [Tue, 18 Jan 2022 15:07:44 +0000 (07:07 -0800)]
x86_64: Document libmvec vector functions accuracy [BZ #28766]
Document maximum 4 ulps accuracy for x86_64 libmvec functions.
This fixes BZ #28766.
Reviewed-By: Paul Zimmermann <Paul.Zimmermann@inria.fr>
H.J. Lu [Fri, 14 Jan 2022 22:48:01 +0000 (14:48 -0800)]
x86: Black list more Intel CPUs for TSX [BZ #27398]
Disable TSX and enable RTM_ALWAYS_ABORT for Intel CPUs listed in:
https://www.intel.com/content/www/us/en/support/articles/
000059422/processors.html
This fixes BZ #27398.
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
Adhemerval Zanella [Tue, 18 Jan 2022 17:36:45 +0000 (14:36 -0300)]
elf: Fix tst-align3
The elf/tst-align3.c declares the function using a wrong prototype.
Checked on aarch64-linux-gnu.
Florian Weimer [Tue, 18 Jan 2022 12:53:11 +0000 (13:53 +0100)]
elf: Move _dl_setup_hash to its own file
And compile it with the early CFLAGS. _dl_setup_hash is called
very early for the ld.so link map, so it should be compiled
differently.
Reviewed-by: Stefan Liebler <stli@linux.ibm.com>
Tested-by: Stefan Liebler <stli@linux.ibm.com>
Samuel Thibault [Mon, 17 Jan 2022 23:18:20 +0000 (23:18 +0000)]
htl: Fix build error in annexc
We were getting
../scripts/evaluate-test.sh posix/annexc $? true false > /usr/src/glibc-upstream/build/posix/annexc.test-result
In file included from ../include/pthread.h:1,
from <stdin>:1:
../sysdeps/htl/include/pthread.h:7:62: error: missing binary operator before token "("
7 | # if defined __USE_EXTERN_INLINES && defined _LIBC && !IS_IN (libsupport)
| ^
Adhemerval Zanella [Mon, 17 Jan 2022 20:16:33 +0000 (17:16 -0300)]
elf: Reinstate tst-audit17
9926f6e2eeb3 ("elf: Skip tst-auditlogmod-* if the linker doesn't support
--depaudit [BZ #28 151]") dropped the test by mistake.
Aurelien Jarno [Mon, 17 Jan 2022 18:41:40 +0000 (19:41 +0100)]
x86: use default cache size if it cannot be determined [BZ #28784]
In some cases (e.g QEMU, non-Intel/AMD CPU) the cache information can
not be retrieved and the corresponding values are set to 0.
Commit
2d651eb9265d ("x86: Move x86 processor cache info to
cpu_features") changed the behaviour in such case by defining the
__x86_shared_cache_size and __x86_data_cache_size variables to 0 instead
of using the default values. This cause an issue with the i686 SSE2
optimized bzero/routine which assumes that the cache size is at least
128 bytes, and otherwise tries to zero/set the whole address space minus
128 bytes.
Fix that by restoring the original code to only update
__x86_shared_cache_size and __x86_data_cache_size variables if the
corresponding cache sizes are not zero.
Fixes bug 28784
Fixes commit
2d651eb9265d
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Samuel Thibault [Sat, 15 Jan 2022 14:52:41 +0000 (15:52 +0100)]
rt/tst-mqueue*: Return UNSUPPORTED when mq_open fails with ENOSYS
Rather than returning 0 or a failure.
Adhemerval Zanella [Wed, 12 Jan 2022 14:31:53 +0000 (11:31 -0300)]
Linux: Add epoll_pwait2 (BZ #27359)
It is similar to epoll_wait, with the difference the timeout has
nanosecond resoluting by using struct timespec instead of int.
Although Linux interface only provides 64 bit time_t support, old
32 bit interface is also provided (so keep in sync with current
practice and to no force opt-in on 64 bit time_t).
Checked on x86_64-linux-gnu and i686-linux-gnu.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
H.J. Lu [Sat, 15 Jan 2022 22:23:20 +0000 (14:23 -0800)]
Properly handle --disable-default-pie [BZ #28780]
When --disable-default-pie is used, all glibc programs and the testsuite
should be built as position dependent executables (non-PIE), regardless
if the build compiler supports PIE or static PIE.
When --disable-default-pie is used, don't build static PIE by default.
This fixes BZ #28780.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Adhemerval Zanella [Wed, 29 Dec 2021 13:20:46 +0000 (10:20 -0300)]
elf: Fix 64 time_t support for installed statically binaries
The usage of internal static symbol for statically linked binaries
does not work correctly for objects built with -D_TIME_BITS=64,
since the internal definition does not provide the expected aliases.
This patch makes it to use the default stat functions instead (which
uses the default 64 time_t alias and types).
Checked on i686-linux-gnu.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Adhemerval Zanella [Mon, 17 Jan 2022 13:56:58 +0000 (10:56 -0300)]
Revert "elf: Fix 64 time_t support for installed statically binaries"
This reverts commit
0b8e83eb1455f3c0332eeb1f96fbc262fbd054e0.
Florian Weimer [Mon, 17 Jan 2022 09:21:34 +0000 (10:21 +0100)]
CVE-2022-23218: Buffer overflow in sunrpc svcunix_create (bug 28768)
The sunrpc function svcunix_create suffers from a stack-based buffer
overflow with overlong pathname arguments.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Martin Sebor [Mon, 17 Jan 2022 09:21:34 +0000 (10:21 +0100)]
sunrpc: Test case for clnt_create "unix" buffer overflow (bug 22542)
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Florian Weimer [Mon, 17 Jan 2022 09:21:34 +0000 (10:21 +0100)]
CVE-2022-23219: Buffer overflow in sunrpc clnt_create for "unix" (bug 22542)
Processing an overlong pathname in the sunrpc clnt_create function
results in a stack-based buffer overflow.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Florian Weimer [Mon, 17 Jan 2022 09:21:34 +0000 (10:21 +0100)]
socket: Add the __sockaddr_un_set function
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Florian Weimer [Mon, 17 Jan 2022 08:57:19 +0000 (09:57 +0100)]
elf/tst-dl_find_object: Disable subtests for non-contiguous maps (bug 28732)
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Mon, 17 Jan 2022 08:57:19 +0000 (09:57 +0100)]
elf: Set l_contiguous to 1 for the main map in more cases
l_contiguous was not initialized at all for the main map and
always 0. This commit adds code to check if the LOAD segments
are adjacent to each other, and sets l_contiguous accordingly.
This helps _dl_find_object because it is more efficient if the
main mapping is contiguous.
Note that not all (PIE or non-PIE) binaries are contiguous in this
way because BFD ld creates executables with LOAD holes:
ELF LOAD segments creating holes in the process image on GNU/Linux
https://sourceware.org/pipermail/binutils/2022-January/119082.html
https://sourceware.org/bugzilla/show_bug.cgi?id=28743
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Mon, 17 Jan 2022 08:57:19 +0000 (09:57 +0100)]
elf: Introduce rtld_setup_main_map
This function collects most of the processing needed to initialize
the link map for the main executable.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Samuel Thibault [Sun, 16 Jan 2022 18:28:40 +0000 (18:28 +0000)]
hurd: Make RPC input array parameters const
This follows mig's
cf4bcc3f1435 ("Also add const qualifiers on server
side")
Samuel Thibault [Sat, 15 Jan 2022 23:02:16 +0000 (00:02 +0100)]
hurd: optimize exec cleanup
When ports are nul we do not need to request their deallocation. It is
also useless to look for them in portnames.
Samuel Thibault [Sat, 15 Jan 2022 22:41:14 +0000 (23:41 +0100)]
hurd: Add __rtld_execve
It trivially execs with the same dtable, portarray and intarray, and only
has to take care of deallocating / destroying ports (file, notably).
Samuel Thibault [Sat, 15 Jan 2022 20:58:39 +0000 (21:58 +0100)]
hurd: Fix exec() leak on proc_task2proc failure
env is allocated after args, so should be freed before it.
Samuel Thibault [Sat, 15 Jan 2022 16:35:51 +0000 (17:35 +0100)]
htl: Hide __pthread_attr's __schedparam type [BZ #23088]
The content of the structure is only used internally, so we can make
__pthread_attr_getschedparam and __pthread_attr_setschedparam convert
between the public sched_param type and an internal __sched_param.
This allows to avoid to spuriously expose the sched_param type.
This fixes BZ #23088.
Samuel Thibault [Sat, 15 Jan 2022 20:30:17 +0000 (21:30 +0100)]
htl: Clear kernel_thread field before releasing the thread structure
Otherwise this is a use-after-free.
Samuel Thibault [Sat, 15 Jan 2022 16:42:14 +0000 (17:42 +0100)]
hurd: drop SA_SIGINFO availability xfail
BZ #23089 was fixed by
d865ff74ba09 ("hurd: implement SA_SIGINFO signal
handlers")
Samuel Thibault [Sat, 15 Jan 2022 14:37:03 +0000 (15:37 +0100)]
hurd: Fix timer/clock_getres crash on NULL res parameter
POSIX allows res to be NULL.
Samuel Thibault [Sat, 15 Jan 2022 13:55:23 +0000 (14:55 +0100)]
hurd: Fix pthread_kill on exiting/ted thread
We have to drop the kernel_thread port from the thread structure, to
avoid pthread_kill's call to _hurd_thread_sigstate trying to reference
it and fail.
Samuel Thibault [Sat, 15 Jan 2022 13:23:37 +0000 (14:23 +0100)]
[hurd] Drop spurious #ifdef SHARED
The whole file is already #ifdef SHARED
Samuel Thibault [Sat, 15 Jan 2022 13:21:53 +0000 (14:21 +0100)]
[hurd] Call _dl_sort_maps_init in _dl_sysdep_start
This follows
15a0c5730d1d ("elf: Fix slow DSO sorting behavior in
dynamic loader (BZ #17645)").
Samuel Thibault [Sat, 15 Jan 2022 12:58:56 +0000 (13:58 +0100)]
elf tst-dl_find_object: Fix typo
mod1 was xdlclose()d a few lines above.
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
s390x: Use <gcc-macros.h> in early HWCAP check
This is required so that the checks still work if $(early-cflags)
selects a different ISA level.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
x86: Add x86-64-vN check to early startup
This ISA level covers the glibc build itself. <dl-hwcap-check.h>
cannot be used because this check (by design) happens before
DL_PLATFORM_INIT and the x86 CPU flags initialization.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
powerpc64le: Use <gcc-macros.h> in early HWCAP check
This is required so that the checks still work if $(early-cflags)
selects a different ISA level.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
Add --with-rtld-early-cflags configure option
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
elf: Split dl-printf.c from dl-misc.c
This allows to use different compiler flags for the diagnostics
code.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Fri, 14 Jan 2022 19:16:05 +0000 (20:16 +0100)]
elf/Makefile: Reflow and sort most variable assignments
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Fri, 14 Jan 2022 15:09:20 +0000 (16:09 +0100)]
Generate gcc-macros.h
The file can be used to check the effect of the default compiler
flags on code generation even in areas of the build that uses
non-default compiler flags.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Fri, 14 Jan 2022 15:09:20 +0000 (16:09 +0100)]
x86: HAVE_X86_LAHF_SAHF, HAVE_X86_MOVBE and -march=x86-64-vN (bug 28782)
HAVE_X86_LAHF_SAHF is implied by x86-64-v2, and HAVE_X86_MOVBE by
x86-64-v3.
The individual flag does not appear in -fverbose-asm flag output
even if the ISA level implies it.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Sunil K Pandey [Wed, 12 Jan 2022 20:26:35 +0000 (12:26 -0800)]
math: Add more inputs to atan2 accuracy tests [BZ #28765]
This patch adds following inputs:
0x1.bcab29da0e947p-54 0x1.bc41f4d2294b8p-54
0x1.a11891ec004d4p-348 0x1.814830510be26p-348
0x1.b836ed678be29p-588 0x1.b7be6f5a03a8cp-588
0x1.a83f842ef3f73p-633 0x1.a799d8a6677ep-633
to atan2 tests and updates x86_64 double atan2 ulps.
This fixes BZ #28765.
Reviewed-By: Paul Zimmermann <Paul.Zimmermann@inria.fr>
H.J. Lu [Sun, 9 Jan 2022 17:06:15 +0000 (09:06 -0800)]
Disable debuginfod in printer tests [BZ #28757]
With gdb-11.1-6.fc35.x86_64, I got
FAIL: nptl/test-cond-printers
FAIL: nptl/test-condattr-printers
FAIL: nptl/test-mutex-printers
FAIL: nptl/test-mutexattr-printers
FAIL: nptl/test-rwlock-printers
FAIL: nptl/test-rwlockattr-printers
$ cat nptl/test-condattr-printers.out
Error: Response does not match the expected pattern.
Command: start
Expected pattern: main
Response: Temporary breakpoint 1 at 0x11d5: file test-condattr-printers.c, line 43.
Starting program: /export/build/gnu/tools-build/glibc-cet-gitlab/build-x86_64-linux/nptl/test-condattr-printers
This GDB supports auto-downloading debuginfo from the following URLs:
https://debuginfod.fedoraproject.org/
Enable debuginfod for this session? (y or [n])
Disable debuginfod to avoid GDB messages. This fixes BZ #28757.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Joseph Myers [Thu, 13 Jan 2022 22:18:13 +0000 (22:18 +0000)]
Update syscall lists for Linux 5.16
Linux 5.16 has one new syscall, futex_waitv. Update
syscall-names.list and regenerate the arch-syscall.h headers with
build-many-glibcs.py update-syscalls.
Tested with build-many-glibcs.py.
Florian Weimer [Thu, 13 Jan 2022 13:59:29 +0000 (14:59 +0100)]
i386: Remove broken CAN_USE_REGISTER_ASM_EBP (bug 28771)
The configure check for CAN_USE_REGISTER_ASM_EBP tried to compile a
simple function that uses %ebp as an inline assembly operand. If
compilation failed, CAN_USE_REGISTER_ASM_EBP was set 0, which
eventually had these consequences:
(1) %ebx was avoided as an inline assembly operand, with an
assembler macro hack to avoid unnecessary register moves.
(2) %ebp was avoided as an inline assembly operand, using an
out-of-line syscall function for 6-argument system calls.
(1) is no longer needed for any GCC version that is supported for
building glibc. %ebx can be used directly as a register operand.
Therefore, this commit removes the %ebx avoidance completely. This
avoids the assembler macro hack, which turns out to be incompatible
with the current Systemtap probe macros (which switch to .altmacro
unconditionally).
(2) is still needed in many build configurations. The existing
configure check cannot really capture that because the simple function
succeeds to compile, while the full glibc build still fails.
Therefore, this commit removes the check, the CAN_USE_REGISTER_ASM_EBP
macro, and uses the out-of-line syscall function for 6-argument system
calls unconditionally.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Siddhesh Poyarekar [Thu, 13 Jan 2022 13:20:55 +0000 (18:50 +0530)]
stdlib: Fix formatting of tests list in Makefile
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Siddhesh Poyarekar [Thu, 13 Jan 2022 05:04:37 +0000 (10:34 +0530)]
stdlib: Sort tests in Makefile
Put one test per line and sort them.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Sunil K Pandey [Wed, 12 Jan 2022 19:02:19 +0000 (11:02 -0800)]
x86_64: Fix SSE4.2 libmvec atan2 function accuracy [BZ #28765]
This patch fixes SSE4.2 libmvec atan2 function accuracy for following
inputs to less than 4 ulps.
{0x1.bcab29da0e947p-54,0x1.bc41f4d2294b8p-54} 4.19888 ulps
{0x1.b836ed678be29p-588,0x1.b7be6f5a03a8cp-588} 4.09889 ulps
This fixes BZ #28765.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Siddhesh Poyarekar [Wed, 12 Jan 2022 18:04:48 +0000 (23:34 +0530)]
debug: Synchronize feature guards in fortified functions [BZ #28746]
Some functions (e.g. stpcpy, pread64, etc.) had moved to POSIX in the
main headers as they got incorporated into the standard, but their
fortified variants remained under __USE_GNU. As a result, these
functions did not get fortified when _GNU_SOURCE was not defined.
Add test wrappers that check all functions tested in tst-chk0 at all
levels with _GNU_SOURCE undefined and then use the failures to (1)
exclude checks for _GNU_SOURCE functions in these tests and (2) Fix
feature macro guards in the fortified function headers so that they're
the same as the ones in the main headers.
This fixes BZ #28746.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Siddhesh Poyarekar [Wed, 12 Jan 2022 18:04:23 +0000 (23:34 +0530)]
debug: Autogenerate _FORTIFY_SOURCE tests
Rename debug/tst-chk1.c to debug/tst-fortify.c and add make hackery to
autogenerate tests with different macros enabled to build and run the
same test with different configurations as well as different
fortification levels.
The change also ends up expanding the -lfs tests to include
_FORTIFY_SOURCE=3.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Wed, 12 Jan 2022 15:48:20 +0000 (12:48 -0300)]
Do not build libresolv module with 64 bit time_t flags
It is not used to build installed programs. It also fixes
resolv/tst-p_secstodate on big endian machines.
Checked on i686-linux-gnu and powerpc-linux-gnu-power4.
Adhemerval Zanella [Wed, 12 Jan 2022 13:35:06 +0000 (10:35 -0300)]
Revert "linux: Fix ancillary 64-bit time timestamp conversion (BZ #28349, BZ #28350)"
This reverts commit
21e0f45c7d73df6fe30c77ffcc9f81410e2ee369.
Adhemerval Zanella [Wed, 12 Jan 2022 13:35:02 +0000 (10:35 -0300)]
Revert "support: Add support_socket_so_timestamp_time64"
This reverts commit
a4cf12360fb7a8eae4fa4923763309b7f10797c3.
Adhemerval Zanella [Mon, 10 Jan 2022 13:17:18 +0000 (10:17 -0300)]
timezone: Fix tst-bz28707 Makefile rule
The $(testdata)/XT5 rule is ambiguous and it may not be correct
evaluated.
Adhemerval Zanella [Mon, 27 Sep 2021 19:07:08 +0000 (16:07 -0300)]
linux: Fix ancillary 64-bit time timestamp conversion (BZ #28349, BZ #28350)
The __convert_scm_timestamps() only updates the control message last
pointer for SOL_SOCKET type, so if the message control buffer contains
multiple ancillary message types the converted timestamp one might
overwrite a valid message.
The test check if the extra ancillary space is correctly handled
by recvmsg/recvmmsg, where if there is no extra space for the 64-bit
time_t converted message the control buffer should be marked with
MSG_TRUNC. It also check if recvmsg/recvmmsg handle correctly multiple
ancillary data.
Checked on x86_64-linux and on i686-linux-gnu on both 5.11 and
4.15 kernel.
Co-authored-by: Fabian Vogt <fvogt@suse.de>
Adhemerval Zanella [Mon, 19 Jul 2021 14:38:15 +0000 (11:38 -0300)]
support: Add support_socket_so_timestamp_time64
Check if the socket support 64-bit network packages timestamps
(SO_TIMESTAMP and SO_TIMESTAMPNS). This will be used on recvmsg
and recvmmsg tests to check if the timestamp should be generated.
Adhemerval Zanella [Wed, 29 Dec 2021 13:20:46 +0000 (10:20 -0300)]
elf: Fix 64 time_t support for installed statically binaries
The usage of internal static symbol for statically linked binaries
does not work correctly for objects built with -D_TIME_BITS=64,
since the internal definition does not provide the expected aliases.
This patch makes it to use the default stat functions instead (which
uses the default 64 time_t alias and types).
Checked on i686-linux-gnu.
Siddhesh Poyarekar [Wed, 12 Jan 2022 13:16:28 +0000 (18:46 +0530)]
Enable _FORTIFY_SOURCE=3 for gcc 12 and above
gcc 12 now has support for the __builtin_dynamic_object_size builtin.
Adapt the macro checks to enable _FORTIFY_SOURCE=3 on gcc 12 and above.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Siddhesh Poyarekar [Wed, 12 Jan 2022 08:58:44 +0000 (14:28 +0530)]
manual: Drop obsolete @refill
The @refill command has been obsolete for a while and now texinfo has
started warning about it.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Szabolcs Nagy [Tue, 14 Dec 2021 11:15:07 +0000 (11:15 +0000)]
aarch64: Add HWCAP2_ECV from Linux 5.16
Indicates the availability of enhanced counter virtualization extension
of armv8.6-a with self-synchronized virtual counter CNTVCTSS_EL0 usable
in userspace.
Joseph Myers [Tue, 11 Jan 2022 15:38:06 +0000 (15:38 +0000)]
Use Linux 5.16 in build-many-glibcs.py
This patch makes build-many-glibcs.py use Linux 5.16.
Tested with build-many-glibcs.py (host-libraries, compilers and glibcs
builds).
Noah Goldstein [Sun, 9 Jan 2022 22:02:28 +0000 (16:02 -0600)]
x86: Fix __wcsncmp_evex in strcmp-evex.S [BZ# 28755]
Fixes [BZ# 28755] for wcsncmp by redirecting length >= 2^56 to
__wcscmp_evex. For x86_64 this covers the entire address range so any
length larger could not possibly be used to bound `s1` or `s2`.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Noah Goldstein [Sun, 9 Jan 2022 22:02:21 +0000 (16:02 -0600)]
x86: Fix __wcsncmp_avx2 in strcmp-avx2.S [BZ# 28755]
Fixes [BZ# 28755] for wcsncmp by redirecting length >= 2^56 to
__wcscmp_avx2. For x86_64 this covers the entire address range so any
length larger could not possibly be used to bound `s1` or `s2`.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Szabolcs Nagy [Fri, 31 Dec 2021 09:50:50 +0000 (09:50 +0000)]
math: Fix float conversion regressions with gcc-12 [BZ #28713]
Converting double precision constants to float is now affected by the
runtime dynamic rounding mode instead of being evaluated at compile
time with default rounding mode (except static object initializers).
This can change the computed result and cause performance regression.
The known correctness issues (increased ulp errors) are already fixed,
this patch fixes remaining cases of unnecessary runtime conversions.
Add float M_* macros to math.h as new GNU extension API. To avoid
conversions the new M_* macros are used and instead of casting double
literals to float, use float literals (only required if the conversion
is inexact).
The patch was tested on aarch64 where the following symbols had new
spurious conversion instructions that got fixed:
__clog10f
__gammaf_r_finite@GLIBC_2.17
__j0f_finite@GLIBC_2.17
__j1f_finite@GLIBC_2.17
__jnf_finite@GLIBC_2.17
__kernel_casinhf
__lgamma_negf
__log1pf
__y0f_finite@GLIBC_2.17
__y1f_finite@GLIBC_2.17
cacosf
cacoshf
casinhf
catanf
catanhf
clogf
gammaf_positive
Fixes bug 28713.
Reviewed-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Florian Weimer [Mon, 10 Jan 2022 12:31:39 +0000 (13:31 +0100)]
elf: Simplify software TM implementation in _dl_find_object
With the current set of fences, the version update at the start
of the TM write operation is redundant, and the version update
at the end does not need to use an atomic read-modify-write
operation.
Also use relaxed MO stores during the dlclose update, and skip any
version changes there.
Suggested-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Florian Weimer [Fri, 7 Jan 2022 13:47:31 +0000 (14:47 +0100)]
Restore ENTRY_POINT definition on hppa, ia64 (bug 28749)
ENTRY_POINT is still needed for elf/rtld.c. Fixes commit
4fb4e7e821e3
("csu: Always use __executable_start in gmon-start.c").
Florian Weimer [Fri, 7 Jan 2022 12:21:57 +0000 (13:21 +0100)]
elf: Fix fences in _dl_find_object_update (bug 28745)
As explained in Hans Boehm, Can Seqlocks Get Along with Programming
Language Memory Models?, an acquire fence is needed in
_dlfo_read_success. The lack of a fence resulted in an observable
bug on powerpc64le compile-time load reordering.
The fence in _dlfo_mappings_begin_update has been reordered, turning
the fence/store sequence into a release MO store equivalent.
Relaxed MO loads are used on the reader side, and relaxed MO stores
on the writer side for the shared data, to avoid formal data races.
This is just to be conservative; it should not actually be necessary
given how the data is used.
This commit also fixes the test run time. The intent was to run it
for 3 seconds, but 0.3 seconds was enough to uncover the bug very
occasionally (while 3 seconds did not reliably show the bug on every
test run).
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>