Fix concurrent changes on nscd aware files (BZ #23178)
As indicated by BZ#23178, concurrent access on some files read by nscd
may result non expected data send through service requisition. This is
due 'sendfile' Linux implementation where for sockets with zero-copy
support, callers must ensure the transferred portions of the the file
reffered by input file descriptor remain unmodified until the reader
on the other end of socket has consumed the transferred data.
I could not find any explicit documentation stating this behaviour on
Linux kernel documentation. However man-pages sendfile entry [1] states
in NOTES the aforementioned remark. It was initially pushed on man-pages
with an explicit testcase [2] that shows changing the file used in
'sendfile' call prior the socket input data consumption results in
previous data being lost.
From commit message it stated on tested Linux version (3.15) only TCP
socket showed this issues, however on recent kernels (4.4) I noticed the
same behaviour for local sockets as well.
Since sendfile on HURD is a read/write operation and the underlying
issue on Linux, the straightforward fix is just remove sendfile use
altogether. I am really skeptical it is hitting some hotstop (there
are indication over internet that sendfile is helpfull only for large
files, more than 10kb) here to justify that extra code complexity or
to pursuit other possible fix (through memory or file locks for
instance, which I am not sure it is doable).
Joseph Myers [Wed, 16 May 2018 00:05:28 +0000 (00:05 +0000)]
Add narrowing multiply functions.
This patch adds the narrowing multiply functions from TS 18661-1 to
glibc's libm: fmul, fmull, dmull, f32mulf64, f32mulf32x, f32xmulf64
for all configurations; f32mulf64x, f32mulf128, f64mulf64x,
f64mulf128, f32xmulf64x, f32xmulf128, f64xmulf128 for configurations
with _Float64x and _Float128; __nldbl_dmull for ldbl-opt.
The changes are mostly essentially the same as for the narrowing add
functions, so the description of those generally applies to this patch
as well. f32xmulf64 for i386 cannot use precision control as used for
add and subtract, because that would result in double rounding for
subnormal results, so that uses round-to-odd with long double
intermediate result instead. The soft-fp support involves adding a
new FP_TRUNC_COOKED since soft-fp multiplication uses cooked inputs
and outputs.
Tested for x86_64, x86, mips64 (all three ABIs, both hard and soft
float) and powerpc, and with build-many-glibcs.py.
powerpc: Fix the compiler type used with C++ when -mabi=ieeelongdouble
When compiling C++ code with -mabi=ieeelongdouble, KCtype is
unavailable and the long double type should be used instead.
This is also providing macro __HAVE_FLOAT128_UNLIKE_LDBL in order to
identify the kind of long double type is being used in the current
compilation unit.
Notice that bits/floatn.h cannot benefit from the new macro due to order
of header inclusion.
* bits/floatn-common.h: Define __HAVE_FLOAT128_UNLIKE_LDBL.
* math/math.h: Restrict the prototype definition for the functions
issignaling(_Float128) and iszero(_Float128); and template
__iseqsig_type<_Float128>, from __HAVE_DISTINCT_FLOAT128 to
__HAVE_FLOAT128_UNLIKE_LDBL.
* sysdeps/powerpc/bits/floatn.h [__HAVE_FLOAT128
&& (!__GNUC_PREREQ (7, 0) || defined __cplusplus)
&& __LDBL_MANT_DIG__ == 113]: Use long double suffix for
__f128() constants; define the type _Float128 as long double;
and reuse long double in __CFLOAT128.
Signed-off-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Joseph Myers [Fri, 11 May 2018 15:11:38 +0000 (15:11 +0000)]
Do not include math-barriers.h in math_private.h.
This patch continues the math_private.h cleanup by stopping
math_private.h from including math-barriers.h and making the users of
the barrier macros include the latter header directly. No attempt is
made to remove any math_private.h includes that are now unused, except
in strtod_l.c where that is done to avoid line number changes in
assertions, so that installed stripped shared libraries can be
compared before and after the patch. (I think the floating-point
environment support in math_private.h should also move out - some
architectures already have fenv_private.h as an architecture-internal
header included from their math_private.h - and after moving that out
might be a better time to identify unused math_private.h includes.)
Tested for x86_64 and x86, and tested with build-many-glibcs.py that
installed stripped shared libraries are unchanged by the patch.
aarch64,falkor: Ignore prefetcher tagging for smaller copies
For smaller and medium sized copies, the effect of hardware
prefetching are not as dominant as instruction level parallelism.
Hence it makes more sense to load data into multiple registers than to
try and route them to the same prefetch unit. This is also the case
for the loop exit where we are unable to latch on to the same prefetch
unit anyway so it makes more sense to have data loaded in parallel.
The performance results are a bit mixed with memcpy-random, with
numbers jumping between -1% and +3%, i.e. the numbers don't seem
repeatable. memcpy-walk sees a 70% improvement (i.e. > 2x) for 128
bytes and that improvement reduces down as the impact of the tail copy
decreases in comparison to the loop.
* sysdeps/aarch64/multiarch/memcpy_falkor.S (__memcpy_falkor):
Use multiple registers to copy data in loop tail.
aarch64,falkor: Ignore prefetcher hints for memmove tail
The tail of the copy loops are unable to train the falkor hardware
prefetcher because they load from a different base compared to the hot
loop. In this case avoid serializing the instructions by loading them
into different registers. Also peel the last iteration of the loop
into the tail (and have them use different registers) since it gives
better performance for medium sizes.
This results in performance improvements of between 3% and 20% over
the current falkor implementation for sizes between 128 bytes and 1K
on the memmove-walk benchmark, thus mostly covering the regressions
seen against the generic memmove.
* sysdeps/aarch64/multiarch/memmove_falkor.S
(__memmove_falkor): Use multiple registers to move data in
loop tail.
Joseph Myers [Thu, 10 May 2018 00:53:04 +0000 (00:53 +0000)]
Move math_check_force_underflow macros to separate math-underflow.h.
This patch continues cleaning up math_private.h by moving the
math_check_force_underflow set of macros to a separate header
math-underflow.h.
This header is included by the files that need it rather than from
math_private.h. Moving these macros to a separate file removes the
math_private.h uses of macros from float.h, so the inclusion of
float.h in math_private.h is also removed; files that were depending
on that inclusion are fixed to include float.h directly. The
inclusion of math-barriers.h from math_private.h will be removed in a
separate patch.
Tested for x86_64 and x86. Also tested with build-many-glibcs.py that
installed stripped shared libraries are unchanged by this patch.
Joseph Myers [Wed, 9 May 2018 19:45:47 +0000 (19:45 +0000)]
Move math_opt_barrier, math_force_eval to separate math-barriers.h.
This patch continues cleaning up math_private.h by moving the
math_opt_barrier and math_force_eval macros to a separate header
math-barriers.h.
At present, those macros are inside a "#ifndef math_opt_barrier" in
math_private.h to allow architectures to override them and then use
a separate math-barriers.h header, no such #ifndef or #include_next is
needed; architectures just have their own alternative version of
math-barriers.h when providing their own optimized versions that avoid
going through memory unnecessarily. The generic math-barriers.h has a
comment added to document these two macros.
In this patch, math_private.h is made to #include <math-barriers.h>,
so files using these macros do not need updating yet. That is because
of uses of math_force_eval in math_check_force_underflow and
math_check_force_underflow_nonneg, which are still defined in
math_private.h. Once those are moved out to a separate header, that
separate header can be made to include <math-barriers.h>, as can the
other files directly using these barrier macros, and then the include
of <math-barriers.h> from math_private.h can be removed.
Tested for x86_64 and x86. Also tested with build-many-glibcs.py that
installed stripped shared libraries are unchanged by this patch.
* sysdeps/generic/math-barriers.h: New file.
* sysdeps/generic/math_private.h [!math_opt_barrier]
(math_opt_barrier): Move to math-barriers.h.
[!math_opt_barrier] (math_force_eval): Likewise.
* sysdeps/aarch64/fpu/math-barriers.h: New file.
* sysdeps/aarch64/fpu/math_private.h (math_opt_barrier): Move to
math-barriers.h.
(math_force_eval): Likewise.
* sysdeps/alpha/fpu/math-barriers.h: New file.
* sysdeps/alpha/fpu/math_private.h (math_opt_barrier): Move to
math-barriers.h.
(math_force_eval): Likewise.
* sysdeps/x86/fpu/math-barriers.h: New file.
* sysdeps/i386/fpu/fenv_private.h (math_opt_barrier): Move to
math-barriers.h.
(math_force_eval): Likewise.
* sysdeps/m68k/m680x0/fpu/math_private.h: Move to....
* sysdeps/m68k/m680x0/fpu/math-barriers.h: ... here. Adjust
multiple-include guard for rename.
* sysdeps/powerpc/fpu/math-barriers.h: New file.
* sysdeps/powerpc/fpu/math_private.h (math_opt_barrier): Move to
math-barriers.h.
(math_force_eval): Likewise.
Joseph Myers [Wed, 9 May 2018 00:15:10 +0000 (00:15 +0000)]
Move math_narrow_eval to separate math-narrow-eval.h.
This patch continues cleaning up the math_private.h header, which
contains lots of different definitions many of which are only needed
by a limited subset of files using that header (and some of which are
overridden by architectures that only want to override selected parts
of the header), by moving the math_narrow_eval macro out to a separate
math-narrow-eval.h header, only included by those files that need it.
That header is placed in include/ (since it's used in stdlib/, not
just files built in math/, but no sysdeps variants are needed at
present).
Tested for x86_64, and with build-many-glibcs.py. (Installed stripped
shared libraries change because of line numbers in assertions in
strtod_l.c.)
benchtests: Move iterator declaration into loop header
This is a minor style change to move the definition of I to its usage
scope instead of at the top of the function. This is consistent with
glibc style guidelines and more importantly it was getting in the way
of my testing.
* benchtests/bench-memcpy-walk.c (do_test): Move declaration
of I into loop header.
* benchtests/bench-memmove-walk.c (do_test): Likewise.
Alexandre Oliva [Mon, 7 May 2018 04:37:37 +0000 (01:37 -0300)]
Revert:
2018-04-30 Raymond Nicholson <rain1@airmail.cc>
* manual/startup.texi (Aborting a Program): Remove inappropriate joke.
This complies with the decision of the project leader and primary and
ultimate maintainer, who partially delegated maintainership to myself
and others under certain constraints.
This is also in line with the community-agreed procedures.
It is obvious that we didn't have consensus on a decision to install
that patch, since both sides are still arguing over it.
As for the decision to reverse the deletion, if we even need one to
counter a move that did not have consensus, although nobody else offered
to install the reversal and restore the status prior to the fait
accompli, and some explicitly refused to do so themselves, nobody
objected when I offered to do so. Therefore, by the same reasoning that
led to the mistaken installation of the patch, and after a much longer
wait for objections, I understand there is consensus on my reverting it.
Paul Pluzhnikov [Sun, 6 May 2018 01:08:27 +0000 (18:08 -0700)]
Fix BZ 20419. A PT_NOTE in a binary could be arbitratily large, so using
alloca for it may cause stack overflow. If the note is larger than
__MAX_ALLOCA_CUTOFF, use dynamically allocated memory to read it in.
2018-05-05 Paul Pluzhnikov <ppluzhnikov@google.com>
Joseph Myers [Fri, 4 May 2018 15:46:32 +0000 (15:46 +0000)]
Ignore absolute symbols in ABI tests.
A recent binutils patch
<https://sourceware.org/ml/binutils/2018-04/msg00336.html> stops the
MIPS linker including the _gp_disp absolute symbol in dynamic symbol
tables.
With older binutils, it is included for MIPS o32 (despite the use of
symbol versioning), and this means that all the ABI test baselines for
MIPS o32 include "_gp_disp _gp_disp A". This symbol is not
meaningfully part of the ABI for shared libraries, since it always has
to be resolved at static link time to the local definition for the
linked object.
All the other absolute symbols in ABI test baselines are the names of
symbol versions. I don't think the mere existence of a symbol version
with a given name - as opposed to the contents of that version if
nonempty - is part of the ABI either. Thus, this patch allows the ABI
tests to pass both before and after the binutils change by changing
abilist.awk not to include absolute symbols in its output, and
changing the baselines accordingly.
Stefan Liebler [Fri, 4 May 2018 08:00:59 +0000 (10:00 +0200)]
Fix blocking pthread_join. [BZ #23137]
On s390 (31bit) if glibc is build with -Os, pthread_join sometimes
blocks indefinitely. This is e.g. observable with
testcase intl/tst-gettext6.
pthread_join is calling lll_wait_tid(tid), which performs the futex-wait
syscall in a loop as long as tid != 0 (thread is alive).
On s390 (and build with -Os), tid is loaded from memory before
comparing against zero and then the tid is loaded a second time
in order to pass it to the futex-wait-syscall.
If the thread exits in between, then the futex-wait-syscall is
called with the value zero and it waits until a futex-wake occurs.
As the thread is already exited, there won't be a futex-wake.
In lll_wait_tid, the tid is stored to the local variable __tid,
which is then used as argument for the futex-wait-syscall.
But unfortunately the compiler is allowed to reload the value
from memory.
With this patch, the tid is loaded with atomic_load_acquire.
Then the compiler is not allowed to reload the value for __tid from memory.
ChangeLog:
[BZ #23137]
* sysdeps/nptl/lowlevellock.h (lll_wait_tid):
Use atomic_load_acquire to load __tid.
H.J. Lu [Thu, 26 Apr 2018 12:01:20 +0000 (05:01 -0700)]
x86-64/swapcontext: Restore the pointer into %rdx after syscall
To prepare for shadow stack support, restore the pointer into %rdx after
syscall and use %rdx, instead of %rsi, to restore context. There is no
functional change.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/x86_64/swapcontext.S (__swapcontext):
Restore the pointer into %rdx, after syscall and use %rdx,
instead of %rsi, to restore context.
H.J. Lu [Wed, 2 May 2018 13:21:09 +0000 (06:21 -0700)]
x86-64/setcontext: Pop the pointer into %rdx after syscall
To prepare for shadow stack support, pop the pointer into %rdx after
syscall and use %rdx, instead of %rsi, to restore context. There is
no functional change.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* sysdeps/unix/sysv/linux/x86_64/setcontext.S (__setcontext):
Pop the pointer into %rdx after syscall and use %rdx, instead
of %rsi, to restore context.
H.J. Lu [Wed, 2 May 2018 13:17:20 +0000 (06:17 -0700)]
x86: Use pad in pthread_unwind_buf to preserve shadow stack register
The pad array in struct pthread_unwind_buf is used by setjmp to save
shadow stack register. We assert that size of struct pthread_unwind_buf
is no less than offset of shadow stack pointer + shadow stack pointer
size.
Since functions, like LIBC_START_MAIN, START_THREAD_DEFN as well as
these with thread cancellation, call setjmp, but never return after
__libc_unwind_longjmp, __libc_unwind_longjmp, which is defined as
__libc_longjmp on x86, doesn't need to restore shadow stack register.
__libc_longjmp, which is a private interface for thread cancellation
implementation in libpthread, is changed to call __longjmp_cancel,
instead of __longjmp. __longjmp_cancel is a new internal function
in libc, which is similar to __longjmp, but doesn't restore shadow
stack register.
The compatibility longjmp and siglongjmp in libpthread.so are changed
to call __libc_siglongjmp, instead of __libc_longjmp, so that they will
restore shadow stack register.
Tested with build-many-glibcs.py.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* nptl/pthread_create.c (START_THREAD_DEFN): Clear previous
handlers after setjmp.
* setjmp/longjmp.c (__libc_longjmp): Don't define alias if
defined.
* sysdeps/unix/sysv/linux/x86/setjmpP.h: Include
<libc-pointer-arith.h>.
(_JUMP_BUF_SIGSET_BITS_PER_WORD): New.
(_JUMP_BUF_SIGSET_NSIG): Changed to 96.
(_JUMP_BUF_SIGSET_NWORDS): Changed to use ALIGN_UP and
_JUMP_BUF_SIGSET_BITS_PER_WORD.
* sysdeps/x86/Makefile (sysdep_routines): Add __longjmp_cancel.
* sysdeps/x86/__longjmp_cancel.S: New file.
* sysdeps/x86/longjmp.c: Likewise.
* sysdeps/x86/nptl/pt-longjmp.c: Likewise.
As for sysctl, ustat has been deprecated in favor of {f}statfs. Also
some newer ports which uses generic interface builds a stub version that
returns ENOSYS.
This patch deprecates ustat interface by removing ustat.h related headers,
adding a compatibility symbol, and avoiding new ports to build and provide
the symbol.
Checked on x86_64-linux-gnu and i686-linux-gnu. Also checked with a
check-abi on all affected ABIs.
* NEWS: Add ustat.h deprecation entry.
* bits/ustat.h: Remove file.
* misc/sys/ustat.h: Likewise.
* misc/ustat.h: Likewise.
* sysdeps/unix/sysv/linux/generic/ustat.c: Likewise.
* misc/Makefile (headers): Remove ustat.h and sys/ustat.h.
* misc/ustat.c (__ustat): Rename to __old_ustat and export only in
compatibility mode.
* sysdeps/unix/sysv/linux/ustat.c (__ustat): Likewise.
* sysdeps/unix/sysv/linux/mips/ustat.c: Define DEV_TO_KDEV and use
generic Linux implementation.
This patch consolidate Linux readahead implementation on generic
sysdeps/unix/sysv/linux/readahead.c one. The changes are:
- Assume __NR_readahead existence with current minimum kernel of 3.2
for all architectures.
- Use INLINE_SYSCALL_CALL, __ALIGNMENT_ARG, and SYSCALL_LL64 to pass
the 64 bit offset. This allows architectures with different abis
to use the same implementation.
math-type-macros.h provides the macro M_FABS in order to reference
the correct fabs function for a specific type.
In most of the cases, M_FABS is identical to M_SUF (fabs), but that
may change in the future.
powerpc64le: Create divergent sysdep directory for powerpc64le.
allowed float128 to be enabled for powerpc64le (little-endian) and not
for powerpc64 (big-endian). Since the only intended difference between
them was the presence or absence of the float128 interface, the sysdeps
directory for powerpc64le explicitly reused the files from powerpc64
(through the use of Implies files).
Although this works, it also means that files under the powerpc64
directory might be preferred over files under powerpc64le. For
instance, on a build for powerpc64le with target set to power9, a file
from powerpc64/power5 might get built, even though a file with the same
name exists in powerpc64le/power8. That happens because the processor
hierarchy was only defined in the sysdeps directory for powerpc64 (and
borrowed by powerpc64le).
This patch fixes this behavior, by creating new subdirectories under
powerpc64 (i.e.: powerpc64/be and powerpc64/le) and creating new Implies
files to provide the hierarchy of processors for powerpc64 and
powerpc64le separately. These changes have no effect on installed,
stripped binaries (which remain unchanged).
Tested that installed stripped binaries are unchanged and that there are
no regressions on powerpc64 and powerpc64le.
Joseph Myers [Fri, 27 Apr 2018 19:11:24 +0000 (19:11 +0000)]
Remove tilegx port.
Since tile support has been removed from the Linux kernel for 4.17,
this patch removes the (unmaintained) port to tilegx from glibc (the
tilepro support having been previously removed). This reflects the
general principle that a glibc port needs upstream support for the
architecture in all the components it build-depends on (so binutils,
GCC and the Linux kernel, for the normal case of a port supporting the
Linux kernel but no other OS), in order to be maintainable.
Apart from removal of sysdeps/tile and sysdeps/unix/sysv/linux/tile,
there are updates to various comments referencing tile for which
removal of those references seemed appropriate. The configuration is
removed from README and from build-many-glibcs.py. contrib.texi keeps
mention of removed contributions, but I updated Chris Metcalf's entry
to reflect that he also contributed the non-removed support for the
generic Linux kernel syscall interface.
__ASSUME_FADVISE64_64_NO_ALIGN support is removed, as it was only used
by tile.
* sysdeps/tile: Remove.
* sysdeps/unix/sysv/linux/tile: Likewise.
* README (tilegx-*-linux-gnu): Remove from list of supported
configurations.
* manual/contrib.texi (Contributors): Mention Chris Metcalf's
contribution of support for generic Linux kernel syscall
interface.
* scripts/build-many-glibcs.py (Context.add_all_configs): Remove
tilegx configurations.
(Config.install_linux_headers): Do not handle tile.
* sysdeps/unix/sysv/linux/aarch64/ldsodefs.h: Do not mention Tile
in comment.
* sysdeps/unix/sysv/linux/nios2/Makefile: Likewise.
* sysdeps/unix/sysv/linux/posix_fadvise.c: Likewise.
[__ASSUME_FADVISE64_64_NO_ALIGN] (__ALIGNMENT_ARG): Remove
conditional undefine and redefine.
* sysdeps/unix/sysv/linux/posix_fadvise64.c: Do not mention Tile
in comment.
[__ASSUME_FADVISE64_64_NO_ALIGN] (__ALIGNMENT_ARG): Remove
conditional undefine and redefine.
This simple test uses sigaction to define a signal handler. It then
uses sigaction again to fetch the information about the same signal
handler, and check that they are consistent. This is enough to detect
mismatches between struct kernel_sigaction and the kernel version of
struct sigaction, like in BZ #23069.
Changelog:
* signal/tst-sigaction.c: New file to test BZ #23069.
* signal/Makefile (tests): Fix indentation. Add tst-sigaction.
This patch consolidates Linux getdirentries{64} implementation on just
the default sysdeps/unix/sysv/linux/getdirentries{64} ones. The default
implementation handles the Linux requirements:
* getdirentries is only built for _DIRENT_MATCHES_DIRENT64 being 0.
* getdirentries64 is always built and aliased to getdents for ABIs
that define _DIRENT_MATCHES_DIRENT64 to 1.
Checked on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
sparcv9-linux-gnu, sparc64-linux-gnu, powerpc-linux-gnu, and
powerpc64le-linux-gnu.
* sysdeps/unix/sysv/linux/getdirentries.c (getdirentries): Build iff
_DIRENT_MATCHES_DIRENT64 is not defined.
* sysdeps/unix/sysv/linux/getdirentries64.c (getdirentries64): Open
implementation and alias to getdirentries if _DIRENT_MATCHES_DIRENT64
is defined.
* sysdeps/unix/sysv/linux/wordsize-64/getdirentries.c: Remove file.
* sysdeps/unix/sysv/linux/wordsize-64/getdirentries64.c: Remove file.
Joseph Myers [Wed, 25 Apr 2018 15:05:01 +0000 (15:05 +0000)]
Use GCC 8 in build-many-glibcs.py by default.
Now that GCC 8 has branched, this patch makes build-many-glibcs.py
default to using GCC 8 branch instead of GCC 7. The effect should be
that all builds complete cleanly and the compilation parts of the
glibc testsuite complete cleanly except for on i686-gnu (with GCC 7
there were testsuite failures for some other configurations as well).
I've replaced my bot building using GCC 6 branch with one using GCC 8
branch. (Of course glibc should continue building with GCC 6 - and
for that matter GCC 5 and 4.9, which are no longer maintained, are
supported versions as well - but building with GCC 6 will no longer be
included in my bot testing.)
* scripts/build-many-glibcs.py (Context.checkout): Default GCC
version to GCC 8 branch.
Joseph Myers [Tue, 24 Apr 2018 16:33:47 +0000 (16:33 +0000)]
Fix Hurd glibc build with GCC 8.
The build of glibc for Hurd has been failing with GCC mainline because
of the checks that aliases have the same type as the symbol aliased;
the Hurd dl-sysdep.c has a macro that defines aliases without using
the proper type. When GCC 8 branches (soon), I intend to make it the
default version in build-many-glibcs.py, so these failures would mean
the default build-many-glibcs.py build fails for Hurd again.
This patch fixes the Hurd build with GCC 8 by changing the macro that
defines the problem aliases to use the correct type for them. An
include of <not-errno.h> is needed to avoid this use of typeof
resulting in an error for __access_noerrno not being declared.
Tested compilation for i686-gnu with build-many-glibcs.py.
* sysdeps/mach/hurd/dl-sysdep.c: Include <not-errno.h>.
(check_no_hidden): Use type of original function when declaring
alias.
Joseph Myers [Tue, 24 Apr 2018 12:11:35 +0000 (12:11 +0000)]
Add PTRACE_SECCOMP_GET_METADATA from Linux 4.16 to sys/ptrace.h.
This patch adds the PTRACE_SECCOMP_GET_METADATA constant from Linux
4.16 to all relevant sys/ptrace.h files. A type struct
__ptrace_seccomp_metadata, analogous to other such types, is also
added.
Consolidate alphasort{64} and versionsort{64} implementation
This patch consolidates both alphasort{64} and versionsort{64}
implementation on just the default dirent/alphasort{64}c and
dirent/versionsort{64} respectively. It changes the logic
to follow the conventions used on other code consolidation:
* the non-LFS variant is only built for _DIRENT_MATCHES_DIRENT64 being 0.
* the LFS variant is always built and aliased to getdents for ABIs
that define _DIRENT_MATCHES_DIRENT64 to 1.
Also on Linux the compat symbol for old non-LFS dirent64 definition
requires a platform-specific scandir64.c. For powerpc32 and sparcv9
it requires to add specific arch-implementation to override the
generic Linux one because neither ABI exports an compat symbol for
non-LFS alphasort64 and versionsort64 variant. It is most likely a
bug and it is also not one that can be fixed (in that there would be
existing binaries expecting both meanings of that symbol at its single
existing version, with binaries expecting the new meaning probably much
more common than those expecting the original meaning of that symbol at
that version).
Checked on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
sparcv9-linux-gnu, sparc64-linux-gnu, powerpc-linux-gnu, and
powerpc64le-linux-gnu.
Make the indentation of the "Deprecated and removed features" section
for 2.28 consistent with the indentation of the "Major new features"
section above. Also, consistently refer to "stdio functions" instead
of "stdio.h functions".
Joseph Myers [Fri, 20 Apr 2018 19:52:02 +0000 (19:52 +0000)]
Define XTABS to TAB3 on alpha to match Linux 4.16.
This patch makes the alpha bits/termios.h define XTABS to TAB3, so
matching a change made in Linux 4.16 as well as matching other
architectures where the values are already equal.
Tested with build-many-glibcs.py for alpha-linux-gnu.
* sysdeps/unix/sysv/linux/alpha/bits/termios.h [__USE_MISC]
(XTABS): Define to TAB3.
This patch consolidates scandir{at}{64} implementation on just
the default dirent/scandir{at}{64}{_r}.c ones. It changes the logic
to follow the conventions used on other code consolidation:
* scandir{at} is only built for _DIRENT_MATCHES_DIRENT64 being 0.
* scandir{at}{64} is always built and aliased to getdents for ABIs
that define _DIRENT_MATCHES_DIRENT64 to 1.
Also on Linux the compat symbol for old non-LFS dirent64 definition
requires a platform-specific scandir64.c.
Checked on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
sparcv9-linux-gnu, sparc64-linux-gnu, powerpc-linux-gnu, and
powerpc64le-linux-gnu.
* dirent/scandir-tail-common.c: New file.
* dirent/scandir-tail.c: Use scandir-tail-common.c.
(__scandir_tail): Build iff _DIRENT_MATCHES_DIRENT64 is not defined.
* dirent/scandir.c: Use scandir-tail-common.c.
* dirent/scandirat.c: Likewise.
* dirent/scandir64-tail.c: Use scandir-tail-common.c.
* dirent/scandir64.c (scandir64): Always build and alias to scandir
if _DIRENT_MATCHES_DIRENT64 is defined.
* dirent/scandirat64.c (scandirat64): Likewise.
* include/dirent.h (__scandir_tail): Only define iff
_DIRENT_MATCHES_DIRENT64 is not defined.
(__scandir64_tail): Define regardless.
(__scandirat, scandirat64): Remove libc_hidden_proto.
* sysdeps/unix/sysv/linux/arm/scandir64.c: Remove file.
* sysdeps/unix/sysv/linux/m68k/scandir64.c: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/scandir64.c: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/scandir64.c: Likewise.
* sysdeps/unix/sysv/linux/i386/scandir64.c: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/scandir64.c: Likewise.
* sysdeps/unix/sysv/linux/scandir64.c: New file.
Samuel Thibault [Thu, 19 Apr 2018 18:55:12 +0000 (20:55 +0200)]
hurd: fix sigevent's sigev_notify_attributes field type
* sysdeps/pthread/bits/types/sigevent_t.h: New file, based on the
generic version but include <bits/pthreadtypes.h> to make struct
sigevent's sigev_notify_attributes field a pthread_attr_t*.
Samuel Thibault [Thu, 19 Apr 2018 18:21:33 +0000 (20:21 +0200)]
hurd: Avoid exposing all <sched.h> symbols from sys/types.h
* bits/sched.h: Include <bits/types/struct_sched_param.h> and move struct
sched_param definition to it.
* sysdeps/unix/sysv/linux/bits/sched.h: Likewise.
* bits/types/struct_sched_param.h: New file.
* sysdeps/htl/bits/types/struct___pthread_attr.h: Include
<bits/types/struct_sched_param.h> instead of <sched.h>.
* posix/Makefile (headers): Add bits/types/struct_sched_param.h.
Samuel Thibault [Thu, 19 Apr 2018 18:14:45 +0000 (20:14 +0200)]
hurd: Fix symbols exposition
* bits/in.h [!__USE_MISC]: Do not define struct ip_opts.
* conform/data/netinet/in.h-data: Allow sin_ and sin6_ prefix.
* sysdeps/gnu/bits/msq.h (struct msqid_ds): Use __wait_queue struct
instead of wait_queue.
* sysdeps/gnu/bits/shm.h (struct shmid_ds): Use __vm_area_struct
instead of vm_area_struct.
This patch consolidates Linux getdents{64} implementation on just
the default sysdeps/unix/sysv/linux/getdents{64}{_r}.c ones.
Although this symbol is used only internally, the non-LFS version
still need to be build due the non-LFS getdirentries which requires
its semantic.
The non-LFS default implementation now uses the wordsize-32 as base
which uses getdents64 syscall plus adjustment for overflow (it allows
to use the same code for architectures that does not support non-LFS
getdents syscall). It has two main differences to wordsize-32 one:
- DIRENT_SET_DP_INO is added to handle alpha requirement to zero
the padding.
- alloca is removed by allocating a bounded temporary buffer (it
increases stack usage by roughly 276 bytes).
The default implementation handle the Linux requirements:
* getdents is only built for _DIRENT_MATCHES_DIRENT64 being 0.
* getdents64 is always built and aliased to getdents for ABIs
that define _DIRENT_MATCHES_DIRENT64 to 1.
* A compat symbol is added for getdents64 for ABI that used to
export the old non-LFS version.
Checked on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
sparcv9-linux-gnu, sparc64-linux-gnu, powerpc-linux-gnu, and
powerpc64le-linux-gnu.
* sysdeps/unix/sysv/linux/alpha/getdents.c: Add comments with alpha
requirements.
(_DIRENT_MATCHES_DIRENT64): Undef
* sysdeps/unix/sysv/linux/alpha/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/arm/getdents64.c: Remove file.
* sysdeps/unix/sysv/linux/generic/getdents.c: Likewise.
* sysdeps/unix/sysv/linux/generic/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/generic/wordsize-32/getdents.c: Likewise.
* sysdeps/unix/sysv/linux/getdents.c: Simplify implementation by
use getdents64 syscalls as base.
* sysdeps/unix/sysv/linux/getdents64.c: Likewise and add compatibility
symbol if required.
* sysdeps/unix/sysv/linux/hppa/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/i386/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/m68k/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/powerpc/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/wordsize-64/getdents.c: Likewise.
* sysdeps/unix/sysv/linux/wordsize-64/getdents64.c: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/get_clockfreq.c
(__get_clockfreq_via_proc_openprom): Use __getdents64.
* sysdeps/unix/sysv/linux/mips/mips64/getdents64.c: New file.
Stefan Liebler [Thu, 19 Apr 2018 10:38:37 +0000 (12:38 +0200)]
Disable lock elision for mutex pretty printer tests.
If e.g. the testcase nptl/test-mutex-printers is run
with enabled lock-elision, it fails on s390x with:
Error: Response does not match the expected pattern.
Command: print *mutex
Expected pattern: pthread_mutex_t
Response: No symbol "mutex" in current context.
(gdb)
See https://www.sourceware.org/ml/libc-alpha/2018-03/msg00583.html
for more details.
In fact the mutex pretty printer tests rely on looking at the
internal details of the lock, thus we disable it by setting up
the GLIB_TUNABLES environment variable inside gdb.
Stefan Liebler [Thu, 19 Apr 2018 10:32:50 +0000 (12:32 +0200)]
Use volatile global counters in test-tgmath.c.
If build with -Os on s390x, the test-tgmath fails with:
float functions not called often enough (-10000)
Within compile_testf(), the counter (count_float) is saved
before the complex functions are called.
Afterwards the saved counter differs to the current-counter.
But the tests with the complex functions do not increment count_float!
Instead count_float is saved to a register before calling totalorder
and totalordermag which both increment count_float.
The compiler is allowed to do that as totalorderf and totalordermagf
is declared with __attribute__ ((__const__)) in math/bits/mathcalls.h.
Thus this patch adjusts the global counters to be volatile.
Then count_float is saved after totalordermag.
Samuel Thibault [Thu, 19 Apr 2018 00:25:03 +0000 (02:25 +0200)]
hurd: Update struct statfs according to struct statvfs
* sysdeps/mach/hurd/bits/statfs.h (struct statfs): Make f_bsize,
f_namemax, f_frsize, and f_flag fields unsigned long int instead of
unsigned int.
(struct statfs64): Likewise.
Samuel Thibault [Thu, 19 Apr 2018 00:05:54 +0000 (02:05 +0200)]
hurd: Fix standard compliance of some statvfs fields
Standards require that the f_bsize, f_frsize, f_flag and f_namemax fields be
unsigned long. They used to be only unsigned on hurd, which happens to be
compatible with unsigned long on the only existing, 32bit, port. We can
thus merely fix the type.
* sysdeps/mach/hurd/bits/statvfs.h (struct statvfs): Make f_bsize,
f_namemax, f_frsize, and f_flag fields unsigned long int instead of
unsigned int.
(struct statvfs64): Likewise.
Joseph Myers [Wed, 18 Apr 2018 16:07:19 +0000 (16:07 +0000)]
Make build-many-glibcs.py build GCC for powerpcspe with --enable-obsolete.
The powerpcspe GCC port has been obsoleted in GCC 8 for not having had
the removal of code for non-SPE processors completed. This patch
accordingly arranges for build-many-glibcs.py to configure GCC with
--enable-obsolete for affected configurations. This is temporary;
either the port gets cleaned up and unobsoleted in GCC and the
configure option can be removed, or the port gets removed in GCC and
we should remove the corresponding glibc support.
Tested with build-many-glibcs.py for the affected configurations.
* scripts/build-many-glibcs.py (Context.add_all_configs): Use
--enable-obsolete for powerpc-linux-gnuspe.
Samuel Thibault [Wed, 18 Apr 2018 00:07:17 +0000 (02:07 +0200)]
hurd: Fix spurious installation of headers defining hidden prototypes
* sysdeps/mach/include/lock-intern.h: Move to include/.
* sysdeps/mach/include/mach.h: Move to include/.
* sysdeps/mach/include/mach/mig_support.h: Move to include/mach/.
* sysdeps/mach/include/mach_error.h: Move to include/.
arm: Remove ununsed ARM code in optimized implementation
This patch removes the ununsed ARM code path for armv6t2 memchr and
strlen and armv7 memch and strcmp. In all implementation, the ARM
code is not used in any possible build (unless glibc is explicit
build with the non-documented NO_THUMB compiler flag) and for armv7
the resulting code either produces wrong results (memchr) and throw
build error (strcmp).
Checked on arm-linux-gnueabihf built targeting both armv6 and
armv7.
Improves __ieee754_exp(x) performance by 18-37% when |x| < 1.0397
Adds a fast path to e_exp.c when |x| < 1.03972053527832.
When values are tested in isolation, reduction in execution
time is: aarch 30%, sparc 18%, x86 37%.
When comparing benchtests/bench.out which includes values
outside that range, the gains are:
aarch 8%, sparc 5%, x86 9%.
make check is clean (no increase in ulp for any math test).
Testing 20M values for each rounding mode in that range shows
approximately one in 200 values is off by 1 ulp. No value tested
for exp(x) changed by 2 or more ulp.
No observed change in performance or accuracy for x outside
fast path range.
These changes will be active for all platforms that don't provide
their own exp() routines. They will also be active for ieee754
versions of ccos, ccosh, cosh, csin, csinh, sinh, exp10, gamma, and
erf.
Stefan Liebler [Thu, 12 Apr 2018 07:37:26 +0000 (09:37 +0200)]
S390: Fix struct sigaction for 31bit in kernel_sigaction.h.
The recent commit b4a5d26d8835d972995f0a0a2f805a8845bafa0b
"linux: Consolidate sigaction implementation" changed the definition
of struct sigaction for s390 (31bit). Unfortunately the order of the
fields were wrong.
This leads to blocking testcases e.g. nptl/tst-sem11.
A thread which blocks due to sem_wait() is cancelled via pthread_cancel()
and the signal-handler sigcancel_handler (see <glibc-src>/nptl/nptl-init.c
is called.
But it just returns as the siginfo_t argument is not setup by the kernel.
Then the main-thread is blocking due to pthread_join().
The flag SA_SIGINFO is set in sa_flags in struct sigaction and
is copied to the "kernel_sigaction.h" struct by the sigaction() call,
but due to the wrong ordering of the struct fields,
the kernel does not recognize it.
This patch consolidates Linux readdir{64}{_r} implementation on just
the default sysdeps/unix/sysv/linux/readdir{64}{_r}.c ones. The
default implementation handle the Linux requirements:
* readdir{_r} is only built for _DIRENT_MATCHES_DIRENT64 being 0.
* readdir64{_r} is always built and aliased to readdir{_r} for
ABI that define _DIRENT_MATCHES_DIRENT64.
* A compat symbol is added for readdir64{_r} for ABI that used to
export the old non-LFS version.
Checked on aarch64-linux-gnu, x86_64-linux-gnu, i686-linux-gnu,
sparcv9-linux-gnu, sparc64-linux-gnu, powerpc-linux-gnu, and
powerpc64le-linux-gnu.
This patch consolidates all Linux sigaction implementations on the default
sysdeps/unix/sysv/linux/sigaction.c. The idea is remove redundant code
and simplify new ports addition by following the current generic
Linux User API (UAPI).
The UAPI for new ports defines a generic extensible sigaction struct as:
Where SA_RESTORER is just placed for compatibility reasons (news ports
should not add it). A similar definition is used on generic
kernel_sigaction.h.
The user exported sigaction definition is not changed, so for most
architectures it requires an adjustment to kernel expected one for the
syscall.
The main changes are:
- All architectures now define and use a kernel_sigaction struct meant
for the syscall, even for the architectures where the user sigaction
has the same layout of the kernel expected one (s390-64 and ia64).
Although it requires more work for these architectures, it simplifies
the generic implementation. Also, sigaction is hardly a hotspot where
micro optimization would play an important role.
- The generic kernel_sigaction definition is now aligned with expected
UAPI one for newer ports, where SA_RESTORER and sa_restorer are not
expected to be defined. This means adding kernel_sigaction for
current architectures that does define it (m68k, nios2, powerpc, s390,
sh, sparc, and tile) and which rely on previous generic definition.
- Remove old MIPS usage of sa_restorer. This was removed since 2.6.27
(2957c9e61ee9c - "[MIPS] IRIX: Goodbye and thanks for all the fish").
- The remaining arch-specific sigaction.c are to handle ABI idiosyncrasies
(like SPARC kernel ABI for rt_sigaction that requires an additional
stub argument).
So for new ports the generic implementation should work if its uses
Linux UAPI. If SA_RESTORER is still required (due some architecture
limitation), it should define its own kernel_sigaction.h, define it and
include generic header (assuming it still uses the default generic kernel
layout).
Checked on x86_64-linux-gnu, i686-linux-gnu, arm-linux-gnueabihf,
aarch64-linux-gnu, sparc64-linux-gnu, sparcv9-linux-gnu, powerpc-linux-gnu,
powerpc64-linux-gnu, ia64-linux-gnu and alpha-linux-gnu. I also checked the
build on all remaining affected ABIs.
* sysdeps/unix/sysv/linux/aarch64/sigaction.c: Use default Linux version
as base implementation.
* sysdeps/unix/sysv/linux/arm/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/i386/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/x86_64/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/alpha/kernel_sigaction.h: Add include guards,
remove unrequired definitions and update comments.
* sysdeps/unix/sysv/linux/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/mips/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/ia64/kernel_sigaction.h: New file.
* sysdeps/unix/sysv/linux/m68k/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/nios2/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/powerpc/kernel_sigaction: Likewise.
* sysdeps/unix/sysv/linux/s390/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/sh/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/sparc/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/tile/kernel_sigaction.h: Likewise.
* sysdeps/unix/sysv/linux/ia64/sigaction.c: Remove file.
* sysdeps/unix/sysv/linux/mips/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/sigaction.c: Likewise.
* sysdeps/unix/sysv/linux/sigaction.c: Add STUB, SET_SA_RESTORER,
and RESET_SA_RESTORER hooks.
manual: Various fixes to the mbstouwcs example, and mbrtowc update
The example did not work because the null byte was not converted, and
mbrtowc was called with a zero-length input string. This results in a
(size_t) -2 return value, so the function always returns NULL.
The size computation for the heap allocation of the result was
incorrect because it did not deal with integer overflow.
Error checking was missing, and the allocated memory was not freed on
error paths. All error returns now set errno. (Note that there is an
assumption that free does not clobber errno.)
The slightly unportable comparision against (size_t) -2 to catch both
(size_t) -1 and (size_t) -2 return values is gone as well.
A null wide character needs to be stored in the result explicitly, to
terminate it.
The description in the manual is updated to deal with these finer
points. The (size_t) -2 behavior (consuming the input bytes) matches
what is specified in ISO C11.
git://sourceware.org / glibc.git / log