The nptl already expects a Linux syscall internally. Also
__is_internal_signal is used and the DEBUGGING_P check is removed.
Checked on x86_64-linux-gnu.
Some Linux filesystems might not fully support 64 bit timestamps [1],
which make some Linux specific tests to fail when they check for the
functionality.
This patch adds a new libsupport function, support_path_support_time64,
that returns whether the target file supports or not 64 bit timestamps.
The support is checked by issuing a utimensat and verifying both the
last access and last modification time against a statx call.
The tests that might fail are also adjusted to check the file support
as well:
$ dd if=/dev/zero of=loopbackfile.img bs=100M count=1
1+0 records in
1+0 records out
104857600 bytes (105 MB, 100 MiB) copied, 0,0589568 s, 1,8 GB/s
$ sudo losetup -fP loopbackfile.img
$ mkfs.xfs loopbackfile.img
meta-data=loopbackfile.img isize=512 agcount=4, agsize=6400 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1
data = bsize=4096 blocks=25600, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=1368, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
$ mkdir loopfs
$ sudo mount -o loop /dev/loop0 loopfs/
$ sudo chown -R azanella:azanella loopfs
$ TMPDIR=loopfs/ ./testrun.sh misc/tst-utimes
error: ../sysdeps/unix/sysv/linux/tst-utimes.c:55: File loopfs//utimesfECsK1 does not support 64-bit timestamps
[1] https://bugzilla.redhat.com/show_bug.cgi?id=1795576
Now that fstat is implemented on top fstatat we need to handle negative
inputs. The implementation now rejects AT_FDCWD, which would otherwise
be accepted by the kernel.
Checked on x86_64-linux-gnu and on i686-linux-gnu.
This patch updates the kernel version in the test tst-mman-consts.py
to 5.11. (There are no new MAP_* constants covered by this test in
5.11 that need any other header changes.)
Tested with build-many-glibcs.py.
Now that compat_symbol_reference works in non-internal tests.
Also do not build and run the test at all on architectures which
do not have the pre-2.28 symbol version of fcntl.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
compat_symbol_reference works in non-internal tests now. Also
avoid building the test for unsupported configurations at all.
I verified by building with build-many-glibcs.py that GLIBC_2.1.3
works as the predecessor of GLIBC_2.2. (Symbol versions in
the early days are complex.)
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This code privides test to check if time on target machine is properly
adjusted.
The time is altered only when cross-test-ssh.sh is executed with
--allow-time-setting flag.
As the delta added to CLOCK_REALTIME is only 1 sec the original time is
not restored and further tests are executed with this bias.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
There were following problems discovered for tst-timerfd test:
1. Do not set the struct itimerspec's it_interval tv_sec to 2 seconds.
After this change the timerfd will trigger only once (the it_value is
only set in this case).
2. The 'val1' variable (including the call to timerfd_gettime) is not
needed anymore, as it is just enough to read the struct itimerspec
after sleep. As a consequence the 'val2' has been renamed to 'val'.
3. After calling timerfd_gettime, the value of struct itimerspec time,
when timer is running, is the remaining time. In the case of this test
it would be less than 1 second.
As a result the TEST_COMPARE macro logic had to be adjusted.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This patch provides test for utimes. It uses wrapper to read
access and modification times to compare them with ones written by
utimes.
Moreover, access and modification times beyond the Y2038 threshold
date (i.e. 32 bit time_t overflow) are also checked.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This patch provides test for utime. It uses wrapper to read access
and modification times to compare them with ones written by utime.
Moreover, access and modification times beyond the Y2038 threshold
date (i.e. 32 bit time_t overflow) are also checked.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This patch provides test for futimens. It uses wrapper, which
reads access and modification time to compare them with ones
written by futimens.
Moreover, access and modification times beyond the Y2038 threshold
date (i.e. 32 bit time_t overflow) are also checked.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This will be used to consolidate the libgcc_s access for backtrace
and pthread_cancel.
Unlike the existing backtrace implementations, it provides some
hardening based on pointer mangling.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
It turns out the startup code in csu/elf-init.c has a perfect pair of
ROP gadgets (see Marco-Gisbert and Ripoll-Ripoll, "return-to-csu: A
New Method to Bypass 64-bit Linux ASLR"). These functions are not
needed in dynamically-linked binaries because DT_INIT/DT_INIT_ARRAY
are already processed by the dynamic linker. However, the dynamic
linker skipped the main program for some reason. For maximum
backwards compatibility, this is not changed, and instead, the main
map is consulted from __libc_start_main if the init function argument
is a NULL pointer.
For statically linked binaries, the old approach based on linker
symbols is still used because there is nothing else available.
A new symbol version __libc_start_main@@GLIBC_2.34 is introduced because
new binaries running on an old libc would not run their ELF
constructors, leading to difficult-to-debug issues.
The elision interfaces are closely aligned between the targets that
implement them, so declare them in the generic <lowlevellock.h>
file.
Empty .c stubs are provided, so that fewer makefile updates
under sysdeps are needed. Also simplify initialization via
__libc_early_init.
The symbols __lll_clocklock_elision, __lll_lock_elision,
__lll_trylock_elision, __lll_unlock_elision, __pthread_force_elision
move into libc. For the time being, non-hidden references are used
from libpthread to access them, but once that part of libpthread
is moved into libc, hidden symbols will be used again. (Hidden
references seem desirable to reduce the likelihood of transactions
aborts.)
This was likely a mistake in the original aarch64 port copied over
from arm: on aarch64 tpidr_el0 register is always available.
The __read_tp symbol is visible with static linking, but it's not
part of the public ABI so it should be safe to remove.
Linux 5.11 has one new syscall, epoll_pwait2. Update
syscall-names.list and regenerate the arch-syscall.h headers with
build-many-glibcs.py update-syscalls.
Tested with build-many-glibcs.py.
If the linux asm/ptrace.h is included before sys/ptrace.h that
breaks the newly added declarations there, so undef the names
that may be defined as macros in the linux header.
The check is moved to LFS fstatat implementation (since it is the
code that actually implements the syscall).
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Remove the internal_statvfs64.c and open code the implementation
on internal_statvfs.c. The alpha is now unrequired, the generic
implementation also handles it.
Also, remove unused includes on internal_statvfs.c, and remove
unused arguments on __internal_statvfs{64}.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
There is no need to handle ENOSYS on fstatfs64 call, required only
for alpha (where is already fallbacks to fstatfs).
Checked on x86_64-linux-gnu.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
There is no need to handle ENOSYS on fstatfs64 call, required only
for alpha (where is already fallbacks to fstatfs). The wordsize
internal_statvfs64.c is removed, since how the LFS support is
provided by fstatvfs64.c (used on 64-bit architectures as well).
Checked on x86_64-linux-gnu.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
The __NR_statfs64 syscall is supported on all architectures but
aarch64, mips64, riscv64, and x86_64. And newer ABIs also uses
the new statfs64 interface (where the struct size is used as
second argument).
So the default implementation now uses:
1. __NR_statfs64 for non-LFS call and handle overflow directly
There is no need to handle __NR_statfs since all architectures
that only support are LFS only.
2. __NR_statfs if defined or __NR_statfs64 otherwise for LFS
call.
Alpha is the only outlier, since it is a 64-bit architecture which
provides non-LFS interface and only provides __NR_statfs64 on
newer kernels (v5.1+).
Checked on x86_64-linux-gnu.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
The __NR_fstatfs64 syscall is supported on all architectures but
aarch64, mips64, riscv64, and x86_64. And newer ABIs also uses
the new fstatfs64 interface (where the struct size is used as
first argument).
So the default implementation now uses:
1. __NR_fstatfs64 for non-LFS call and handle overflow directly
There is no need to handle __NR_fstatfs since all architectures
that only support are LFS only.
2. __NR_fstatfs if defined or __NR_fstatfs64 otherwise for LFS
call.
Alpha is the only outlier, it is a 64-bit architecture which
provides non-LFS interface and only provides __NR_fstatfs64 on
newer kernels (5.1+).
Checked on x86_64-linux-gnu.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Currently glibc has three different struct statfs{64} definitions:
1. Non-LFS support where non-LFS and LFS struct have different
size: alpha, arm, hppa, i686, m68k, microblaze, mips (all abis),
powerpc32, s390, sh4, and sparc.
2. Non-LFS support where non-LFS and LFS struct have the same
size: csky and nios2.
3. Only LFS support (where both struct have the same size): arc,
ia64, powerpc64 (including LE), riscv (both 32 and 64 bits),
s390x, sparc64, and x86 (including x32).
The STATFS_IS_STATFS64/__STATFS_MATCHES_STATFS64 does not tell apart
between 1. and 2. since for both the only difference is the struct
size (for 2. both non-LFS and LFS uses the same syscall, where for
1. the old non-LFS is used for [f]statfs).
This patch move the generic statfs.h for both csky and nios2, and
make the default definitions for newer ABIs to assume that only
LFS will be support (so there is no need to keep no-LFS and LFS
struct statfs with the same size, it will be implicit).
This patch does not change the code generation.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
The XSTAT_IS_XSTAT64 and STAT_IS_KERNEL_STAT flags are now set to 1 and
STATFS_IS_STATFS64 is set to __STATFS_MATCHES_STATFS64. This makes the
default ABI for newer ports to provide only LFS calls.
A copy of non-LFS support is provided to 32-bit ABIS with non-LFS
support (arm, csky, i386, m68k, nios2, s390, and sh). Is also allows
to remove the 64-bit ports, which already uses the default values.
This patch does not change the code generation.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
aarch64, arc, ia64, mips64, powerpc64, riscv32, riscv64, s390x, sparc64,
and x86_64 defines STATFS_IS_STATFS64 to 0, but all of them alias
statfs to statfs64 and the struct statfs has the same and layout of
struct statfs64.
The correct definition will be used on the [f]statfs[64] consolidation.
This patch does not change code generation since the symbols are
implemented using the auto-generation syscall for all the aforementioned
ABIs.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
The TUNABLE_SET interface took a primitive C type argument, which
resulted in inconsistent type conversions internally due to incorrect
dereferencing of types, especialy on 32-bit architectures. This
change simplifies the TUNABLE setting logic along with the interfaces.
Now all numeric tunable values are stored as signed numbers in
tunable_num_t, which is intmax_t. All calls to set tunables cast the
input value to its primitive type and then to tunable_num_t for
storage. This relies on gcc-specific (although I suspect other
compilers woul also do the same) unsigned to signed integer conversion
semantics, i.e. the bit pattern is conserved. The reverse conversion
is guaranteed by the standard.
Linux 5.10 adds PTRACE_PEEKMTETAGS and PTRACE_POKEMTETAGS for AArch64.
Adding those shows up that glibc is also missing PTRACE_SYSEMU and
PTRACE_SYSEMU_SINGLESTEP, for AArch64 (where they were added to Linux
in 5.3) and for PowerPC (where they were added in Linux 4.20); it
already has those two defines for x86. Add all those defines to
glibc's headers.
Tested with build-many-glibcs.py for aarch64-linux-gnu and
powerpc-linux-gnu.
Previously, glibc would pick an arbitrary tmpfs file system from
/proc/mounts if /dev/shm was not available. This could lead to
an unsuitable file system being picked for the backing storage for
shm_open, sem_open, and related functions.
This patch introduces a new function, __shm_get_name, which builds
the file name under the appropriate (now hard-coded) directory. It is
called from the various shm_* and sem_* function. Unlike the
SHM_GET_NAME macro it replaces, the callers handle the return values
and errno updates. shm-directory.c is moved directly into the posix
subdirectory because it can be implemented directly using POSIX
functionality. It resides in libc because it is needed by both
librt and nptl/htl.
In the sem_open implementation, tmpfname is initialized directly
from a string constant. This happens to remove one alloca call.
Checked on x86_64-linux-gnu.
This change adds new test to assess ppoll()'s timeout related
functionality (the struct pollfd does not provide valid fd to wait
for - just wait for timeout).
To be more specific - two use cases are checked:
- if ppoll() times out immediately when passed struct timespec has zero
values of tv_nsec and tv_sec.
- if ppoll() times out after timeout specified in passed argument
This change adds new test to assess functionality of timerfd_*
functions.
It creates new timer (operates on its file descriptor) and checks
if time before and after sleep is between expected values.
The shmmax expected value is tricky to check because kernel clamps it
to INT_MAX in two cases:
1. Compat symbols with IPC_64, i.e, 32-bit binaries running on 64-bit
kernels.
2. Default symbol without IPC_64 (defined as IPC_OLD within Linux) and
glibc always use IPC_64 for 32-bit ABIs (to support 64-bit time_t).
It means that 32-bit binaries running on 32-bit kernels will not see
shmmax being clamped.
And finding out whether the compat symbol is used would require checking
the underlying kernel against the current ABI. The shmall and shmmni
already provided enough coverage.
Checked on x86_64-linux-gnu and i686-linux-gnu. It should fix the
tst-sysvshm-linux failures on 32-bit kernels.
Add _SC_MINSIGSTKSZ for the minimum signal stack size derived from
AT_MINSIGSTKSZ, which is the minimum number of bytes of free stack
space required in order to gurantee successful, non-nested handling
of a single signal whose handler is an empty function, and _SC_SIGSTKSZ
which is the suggested minimum number of bytes of stack space required
for a signal stack.
If AT_MINSIGSTKSZ isn't available, sysconf (_SC_MINSIGSTKSZ) returns
MINSIGSTKSZ. On Linux/x86 with XSAVE, the signal frame used by kernel
is composed of the following areas and laid out as:
------------------------------
| alignment padding |
------------------------------
| xsave buffer |
------------------------------
| fsave header (32-bit only) |
------------------------------
| siginfo + ucontext |
------------------------------
Compute AT_MINSIGSTKSZ value as size of xsave buffer + size of fsave
header (32-bit only) + size of siginfo and ucontext + alignment padding.
If _SC_SIGSTKSZ_SOURCE or _GNU_SOURCE are defined, MINSIGSTKSZ and SIGSTKSZ
are redefined as
/* Default stack size for a signal handler: sysconf (SC_SIGSTKSZ). */
# undef SIGSTKSZ
# define SIGSTKSZ sysconf (_SC_SIGSTKSZ)
/* Minimum stack size for a signal handler: SIGSTKSZ. */
# undef MINSIGSTKSZ
# define MINSIGSTKSZ SIGSTKSZ
Compilation will fail if the source assumes constant MINSIGSTKSZ or
SIGSTKSZ.
The reason for not simply increasing the kernel's MINSIGSTKSZ #define
(apart from the fact that it is rarely used, due to glibc's shadowing
definitions) was that userspace binaries will have baked in the old
value of the constant and may be making assumptions about it.
For example, the type (char [MINSIGSTKSZ]) changes if this #define
changes. This could be a problem if an newly built library tries to
memcpy() or dump such an object defined by and old binary.
Bounds-checking and the stack sizes passed to things like sigaltstack()
and makecontext() could similarly go wrong.
brk used by statup before TCB is properly set, so we can't use
IA64_USE_NEW_STUB.
This patch fixes a regression introduced by 720480934a.
Checked on ia64-linux-gnu.
GCC mainline shows the following error:
../sysdeps/unix/sysv/linux/mips/mips64/getdents64.c: In function '__getdents64':
../sysdeps/unix/sysv/linux/mips/mips64/getdents64.c:121:7: error: 'memcpy' forming offset [4, 7] is out of the bounds [0, 4] [-Werror=array-bounds]
121 | memcpy (((char *) dp + offsetof (struct dirent64, d_ino)),
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
122 | KDP_MEMBER (kdp, d_ino), sizeof ((struct dirent64){0}.d_ino));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../sysdeps/unix/sysv/linux/mips/mips64/getdents64.c:123:7: error: 'memcpy' forming offset [4, 7] is out of the bounds [0, 4] [-Werror=array-bounds]
123 | memcpy (((char *) dp + offsetof (struct dirent64, d_off)),
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
124 | KDP_MEMBER (kdp, d_off), sizeof ((struct dirent64){0}.d_off));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The issue is due both d_ino and d_off fields for mips64-n32
kernel_dirent are 32-bits, while this is using memcpy to copy 64 bits
from it into the glibc dirent64.
The fix is to use a temporary buffer to read the correct type
from kernel_dirent.
Checked with a build-many-glibcs.py for mips64el-linux-gnu and I
also checked the tst-getdents64 on mips64el 4.1.4 kernel with
and without fallback enabled (by manually setting the
getdents64_supported).
