This patch optimizes the generic spinlock code.
The type pthread_spinlock_t is a typedef to volatile int on all archs.
Passing a volatile pointer to the atomic macros which are not mapped to the
C11 atomic builtins can lead to extra stores and loads to stack if such
a macro creates a temporary variable by using "__typeof (*(mem)) tmp;".
Thus, those macros which are used by spinlock code - atomic_exchange_acquire,
atomic_load_relaxed, atomic_compare_exchange_weak - have to be adjusted.
According to the comment from Szabolcs Nagy, the type of a cast expression is
unqualified (see http://www.open-std.org/jtc1/sc22/wg14/www/docs/dr_423.htm):
__typeof ((__typeof (*(mem)) *(mem)) tmp;
Thus from spinlock perspective the variable tmp is of type int instead of
type volatile int. This patch adjusts those macros in include/atomic.h.
With this construct GCC >= 5 omits the extra stores and loads.
The atomic macros are replaced by the C11 like atomic macros and thus
the code is aligned to it. The pthread_spin_unlock implementation is now
using release memory order instead of sequentially consistent memory order.
The issue with passed volatile int pointers applies to the C11 like atomic
macros as well as the ones used before.
I've added a glibc_likely hint to the first atomic exchange in
pthread_spin_lock in order to return immediately to the caller if the lock is
free. Without the hint, there is an additional jump if the lock is free.
I've added the atomic_spin_nop macro within the loop of plain reads.
The plain reads are also realized by C11 like atomic_load_relaxed macro.
The new define ATOMIC_EXCHANGE_USES_CAS determines if the first try to acquire
the spinlock in pthread_spin_lock or pthread_spin_trylock is an exchange
or a CAS. This is defined in atomic-machine.h for all architectures.
The define SPIN_LOCK_READS_BETWEEN_CMPXCHG is now removed.
There is no technical reason for throwing in a CAS every now and then,
and so far we have no evidence that it can improve performance.
If that would be the case, we have to adjust other spin-waiting loops
elsewhere, too! Using a CAS loop without plain reads is not a good idea
on many targets and wasn't used by one. Thus there is now no option to
do so.
Architectures are now using the generic spinlock automatically if they
do not provide an own implementation. Thus the pthread_spin_lock.c files
in sysdeps folder are deleted.
ChangeLog:
* NEWS: Mention new spinlock implementation.
* include/atomic.h:
(__atomic_val_bysize): Cast type to omit volatile qualifier.
(atomic_exchange_acq): Likewise.
(atomic_load_relaxed): Likewise.
(ATOMIC_EXCHANGE_USES_CAS): Check definition.
* nptl/pthread_spin_init.c (pthread_spin_init):
Use atomic_store_relaxed.
* nptl/pthread_spin_lock.c (pthread_spin_lock):
Use C11-like atomic macros.
* nptl/pthread_spin_trylock.c (pthread_spin_trylock):
Likewise.
* nptl/pthread_spin_unlock.c (pthread_spin_unlock):
Use atomic_store_release.
* sysdeps/aarch64/nptl/pthread_spin_lock.c: Delete File.
* sysdeps/arm/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/hppa/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/m68k/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/microblaze/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/mips/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/nios2/nptl/pthread_spin_lock.c: Likewise.
* sysdeps/aarch64/atomic-machine.h (ATOMIC_EXCHANGE_USES_CAS): Define.
* sysdeps/alpha/atomic-machine.h: Likewise.
* sysdeps/arm/atomic-machine.h: Likewise.
* sysdeps/i386/atomic-machine.h: Likewise.
* sysdeps/ia64/atomic-machine.h: Likewise.
* sysdeps/m68k/coldfire/atomic-machine.h: Likewise.
* sysdeps/m68k/m680x0/m68020/atomic-machine.h: Likewise.
* sysdeps/microblaze/atomic-machine.h: Likewise.
* sysdeps/mips/atomic-machine.h: Likewise.
* sysdeps/powerpc/powerpc32/atomic-machine.h: Likewise.
* sysdeps/powerpc/powerpc64/atomic-machine.h: Likewise.
* sysdeps/s390/atomic-machine.h: Likewise.
* sysdeps/sparc/sparc32/atomic-machine.h: Likewise.
* sysdeps/sparc/sparc32/sparcv9/atomic-machine.h: Likewise.
* sysdeps/sparc/sparc64/atomic-machine.h: Likewise.
* sysdeps/tile/tilegx/atomic-machine.h: Likewise.
* sysdeps/tile/tilepro/atomic-machine.h: Likewise.
* sysdeps/unix/sysv/linux/hppa/atomic-machine.h: Likewise.
* sysdeps/unix/sysv/linux/m68k/coldfire/atomic-machine.h: Likewise.
* sysdeps/unix/sysv/linux/nios2/atomic-machine.h: Likewise.
* sysdeps/unix/sysv/linux/sh/atomic-machine.h: Likewise.
* sysdeps/x86_64/atomic-machine.h: Likewise.
This uses atomic operations to access lock elision metadata that is accessed
concurrently (ie, adapt_count fields). The size of the data is less than a
word but accessed only with atomic loads and stores; therefore, we add
support for shorter-size atomic load and stores too.
* include/atomic.h (__atomic_check_size_ls): New.
(atomic_load_relaxed, atomic_load_acquire, atomic_store_relaxed,
atomic_store_release): Use it.
* sysdeps/x86/elide.h (ACCESS_ONCE): Remove.
(elision_adapt, ELIDE_LOCK): Use atomics.
* sysdeps/unix/sysv/linux/x86/elision-lock.c (__lll_lock_elision): Use
atomics and improve code comments.
* sysdeps/unix/sysv/linux/x86/elision-trylock.c
(__lll_trylock_elision): Likewise.
No code uses atomic_fetch_xor_release except for the upcoming
conditional variable rewrite. Therefore there is no user
visible bug here. The use of atomic_compare_and_exchange_bool_rel
is removed (since it doesn't exist anymore), and is replaced
by atomic_compare_exchange_weak_release.
We use weak_release because it provides better performance in
the loop (the weak semantic) and because the xor is release MO
(the release semantic). We don't reload expected in the loop
because atomic_compare_and_exchange_weak_release does this for
us as part of the CAS failure.
It is otherwise a fairly plain conversion that fixes building
the new condvar for 32-bit x86. Passes all regression tests
for x86.
atomic_compare_and_exchange_bool_rel and
catomic_compare_and_exchange_bool_rel are removed and replaced with the
new C11-like atomic_compare_exchange_weak_release. The concurrent code
in nscd/cache.c has not been reviewed yet, so this patch does not add
detailed comments.
* nscd/cache.c (cache_add): Use new C11-like atomic operation instead
of atomic_compare_and_exchange_bool_rel.
* nptl/pthread_mutex_unlock.c (__pthread_mutex_unlock_full): Likewise.
* include/atomic.h (atomic_compare_and_exchange_bool_rel,
catomic_compare_and_exchange_bool_rel): Remove.
* sysdeps/aarch64/atomic-machine.h
(atomic_compare_and_exchange_bool_rel): Likewise.
* sysdeps/alpha/atomic-machine.h
(atomic_compare_and_exchange_bool_rel): Likewise.
* sysdeps/arm/atomic-machine.h
(atomic_compare_and_exchange_bool_rel): Likewise.
* sysdeps/mips/atomic-machine.h
(atomic_compare_and_exchange_bool_rel): Likewise.
* sysdeps/tile/atomic-machine.h
(atomic_compare_and_exchange_bool_rel): Likewise.
It was noted in
<https://sourceware.org/ml/libc-alpha/2012-09/msg00305.html> that the
bits/*.h naming scheme should only be used for installed headers.
This patch renames bits/atomic.h to atomic-machine.h to follow that
convention.
This is the only change in this series that needs to change the
filename rather than simply removing a directory level (because both
atomic.h and bits/atomic.h exist at present).
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch).
[BZ #14912]
* sysdeps/aarch64/bits/atomic.h: Move to ...
* sysdeps/aarch64/atomic-machine.h: ...here.
(_AARCH64_BITS_ATOMIC_H): Rename macro to
_AARCH64_ATOMIC_MACHINE_H.
* sysdeps/alpha/bits/atomic.h: Move to ...
* sysdeps/alpha/atomic-machine.h: ...here.
* sysdeps/arm/bits/atomic.h: Move to ...
* sysdeps/arm/atomic-machine.h: ...here. Update comments.
* bits/atomic.h: Move to ...
* sysdeps/generic/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/i386/bits/atomic.h: Move to ...
* sysdeps/i386/atomic-machine.h: ...here.
* sysdeps/ia64/bits/atomic.h: Move to ...
* sysdeps/ia64/atomic-machine.h: ...here.
* sysdeps/m68k/coldfire/bits/atomic.h: Move to ...
* sysdeps/m68k/coldfire/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/m68k/m680x0/m68020/bits/atomic.h: Move to ...
* sysdeps/m68k/m680x0/m68020/atomic-machine.h: ...here.
* sysdeps/microblaze/bits/atomic.h: Move to ...
* sysdeps/microblaze/atomic-machine.h: ...here.
* sysdeps/mips/bits/atomic.h: Move to ...
* sysdeps/mips/atomic-machine.h: ...here.
(_MIPS_BITS_ATOMIC_H): Rename macro to _MIPS_ATOMIC_MACHINE_H.
* sysdeps/powerpc/bits/atomic.h: Move to ...
* sysdeps/powerpc/atomic-machine.h: ...here. Update comments.
* sysdeps/powerpc/powerpc32/bits/atomic.h: Move to ...
* sysdeps/powerpc/powerpc32/atomic-machine.h: ...here. Update
comments. Include <atomic-machine.h> instead of <bits/atomic.h>.
* sysdeps/powerpc/powerpc64/bits/atomic.h: Move to ...
* sysdeps/powerpc/powerpc64/atomic-machine.h: ...here. Include
<atomic-machine.h> instead of <bits/atomic.h>.
* sysdeps/s390/bits/atomic.h: Move to ...
* sysdeps/s390/atomic-machine.h: ...here.
* sysdeps/sparc/sparc32/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc32/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/sparc/sparc32/sparcv9/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc32/sparcv9/atomic-machine.h: ...here.
* sysdeps/sparc/sparc64/bits/atomic.h: Move to ...
* sysdeps/sparc/sparc64/atomic-machine.h: ...here.
* sysdeps/tile/bits/atomic.h: Move to ...
* sysdeps/tile/atomic-machine.h: ...here.
* sysdeps/tile/tilegx/bits/atomic.h: Move to ...
* sysdeps/tile/tilegx/atomic-machine.h: ...here. Include
<sysdeps/tile/atomic-machine.h> instead of
<sysdeps/tile/bits/atomic.h>.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/tile/tilepro/bits/atomic.h: Move to ...
* sysdeps/tile/tilepro/atomic-machine.h: ...here. Include
<sysdeps/tile/atomic-machine.h> instead of
<sysdeps/tile/bits/atomic.h>.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/arm/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/arm/atomic-machine.h: ...here. Include
<sysdeps/arm/atomic-machine.h> instead of
<sysdeps/arm/bits/atomic.h>.
* sysdeps/unix/sysv/linux/hppa/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/hppa/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/m68k/coldfire/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/m68k/coldfire/atomic-machine.h: ...here.
(_BITS_ATOMIC_H): Rename macro to _ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/nios2/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/nios2/atomic-machine.h: ...here.
(_NIOS2_BITS_ATOMIC_H): Rename macro to _NIOS2_ATOMIC_MACHINE_H.
* sysdeps/unix/sysv/linux/sh/bits/atomic.h: Move to ...
* sysdeps/unix/sysv/linux/sh/atomic-machine.h: ...here.
* sysdeps/x86_64/bits/atomic.h: Move to ...
* sysdeps/x86_64/atomic-machine.h: ...here.
* include/atomic.h: Include <atomic-machine.h> instead of
<bits/atomic.h>.
This patch combines BUSY_WAIT_NOP and atomic_delay into a new
atomic_spin_nop function and adjusts all clients. The new function is
put into atomic.h because what is best done in a spin loop is
architecture-specific, and atomics must be used for spinning. The
function name is meant to tell users that this has no effect on
synchronization semantics but is a performance aid for spinning.
The following patch fixes catomic_compare_and_exchange_*_rel definitions
(which were never used and weren't correct) and uses
catomic_compare_and_exchange_val_rel in _int_free. Comparing to the
pre-2009-07-02 --enable-experimental-malloc state the generated code should
be identical on all arches other than ppc/ppc64 and on ppc/ppc64 should use
lwsync instead of isync barrier.
2006-10-17 Jakub Jelinek <jakub@redhat.com>
* sunrpc/xdr_mem.c (xdrmem_setpos): Don't compare addresses
as signed longs, check for x_base + pos overflow.
* sunrpc/Makefile (tests): Add tst-xdrmem2.
* sunrpc/tst-xdrmem2.c: New test.
2006-10-18 Ulrich Drepper <drepper@redhat.com>
* elf/dl-lookup.c (_dl_lookup_symbol_x): Add warning to
_dl_lookup_symbol_x code.
2006-10-17 Jakub Jelinek <jakub@redhat.com>
* elf/dl-runtime.c: Include sysdep-cancel.h.
(_dl_fixup, _dl_profile_fixup): Use __rtld_mrlock_* and
scoperec->nusers only if !SINGLE_THREAD_P. Use atomic_*
instead of catomic_* macros.
* elf/dl-sym.c: Include sysdep-cancel.h.
(do_sym): Use __rtld_mrlock_* and scoperec->nusers only
if !SINGLE_THREAD_P. Use atomic_* instead of catomic_* macros.
* elf/dl-close.c: Include sysdep-cancel.h.
(_dl_close): Use __rtld_mrlock_* and scoperec->nusers only
if !SINGLE_THREAD_P. Use atomic_* instead of catomic_* macros.
* elf/dl-open.c: Include sysdep-cancel.h.
(dl_open_worker): Use __rtld_mrlock_* and scoperec->nusers only
if !SINGLE_THREAD_P. Use atomic_* instead of catomic_* macros.
2006-10-17 Jakub Jelinek <jakub@redhat.com>
[BZ #3313]
* malloc/malloc.c (malloc_consolidate): Set maxfb to address of last
fastbin rather than end of fastbin array.
2006-10-18 Ulrich Drepper <drepper@redhat.com>
* sysdeps/i386/i486/bits/atomic.h (catomic_decrement): Use correct
body macro.
* sysdeps/x86_64/bits/atomic.h
(__arch_c_compare_and_exchange_val_64_acq): Add missing casts.
(catomic_decrement): Use correct body macro.
2006-10-17 Jakub Jelinek <jakub@redhat.com>
* include/atomic.h: Add a unique prefix to all local variables
in macros.
* csu/tst-atomic.c (do_test): Test also catomic_* macros.
* include/link.h: Include <rtld-lowlevel.h>. Define struct
Implement reference counting of scope records.
* elf/dl-close.c (_dl_close): Remove all scopes from removed objects
from the list in objects which remain. Always allocate new scope
record.
* elf/dl-open.c (dl_open_worker): When growing array for scopes,
don't resize, allocate a new one.
* elf/dl-runtime.c: Update reference counters before using a scope
array.
* elf/dl-sym.c: Likewise.
* elf/dl-libc.c: Adjust for l_scope name change.
* elf/dl-load.c: Likewise.
* elf/dl-object.c: Likewise.
* elf/rtld.c: Likewise.
* include/link.h: Inlcude <rtld-lowlevel.h>. Define struct
r_scoperec. Replace r_scope with pointer to r_scoperec structure.
Add l_scoperec_lock.
* sysdeps/generic/ldsodefs.h: Include <rtld-lowlevel.h>.
* sysdeps/generic/rtld-lowlevel.h: New file.
* include/atomic.h: Rename atomic_and to atomic_and_val and
atomic_or to atomic_or_val. Define new macros atomic_and and
atomic_or which do not return values.
* sysdeps/x86_64/bits/atomic.h: Define atomic_and and atomic_or.
Various cleanups.
* sysdeps/i386/i486/bits/atomic.h: Likewise.
2005-08-17 Robert Love <rml@novell.com>
* sysdeps/unix/sysv/linux/sys/inotify.h: Add IN_MOVE_SELF event.
2005-08-19 Bruno Haible <bruno@clisp.org>
* intl/localealias.c (read_alias_file): In case of failure, close the
file descriptor and sort the array before returning.
2005-08-19 Ulrich Drepper <drepper@redhat.com>
* malloc/Makefile: Link libmemusage.so with ld.so.
* malloc/memusage.c: Use atomic operations for all counter and size
computations. Handle thread stacks.
* sysdeps/generic/memusage.h: Define memusage_cntr_t and
memusage_size_t.
* sysdeps/i386/i686/memusage.h: New file.
* include/atomic.h: Define atomic_max and atomic_min.
* debug/tst-chk1.c: Add tests for mbstowcs and wcstombs.
atomic_compare_and_exchange_val_acq, atomic_add_zero,
atomic_compare_and_exchange_bool_acq and atomic_add_negative tests.
* include/atomic.h (atomic_add_negative, atomic_add_zero):
Prefix local variable so that it doesn't clash with the one
in atomic_exchange_and_add.
* sysdeps/ia64/bits/atomic.h (atomic_exchange): Fix for long/void *
pointers.
(atomic_exchange_and_add): Implement using __sync_fetch_and_add_?i.
* sysdeps/powerpc/bits/atomic.h (atomic_exchange_and_add): Force
value into register.
* sysdeps/s390/bits/atomic.h (__arch_compare_and_exchange_val_64_acq):
Cast newval to long.
* sysdeps/x86_64/bits/atomic.h
(__arch_compare_and_exchange_val_64_acq): Cast newval and oldval to
long.
(atomic_exchange): Cast newvalue to long if sizeof == 8.
(atomic_exchange_and_add): Cast value to long if sizeof == 8.
(atomic_add, atomic_add_negative, atomic_add_zero): Likewise.
(atomic_bit_set): Shift 1L up in all cases to shut up warnings.
2003-03-25 Ulrich Drepper <drepper@redhat.com>
* csu/tst-atomic.c: Adjust tests to what atomic_add_negative and
atomic_add_zero were supposed to do.
* include/atomic.h: Adjust atomic_add_negative and atomic_add_zero
to x86 behavior.
* sysdeps/generic/bits/typesizes.h (__TIMER_T_TYPE): Define as void*.
This matches the new timer implementation.
* sysdeps/unix/sysv/linux/bits/siginfo.h (struct siginfo): Adjust
timer info for what the kernel provides these days.
(struct sigevent): Add _tid field.
Define SIGEV_THREAD_ID.
* Versions.def (librt): Add GLIBC_2.3.3.
* abilist/libpthread.abilist: Update for nptl.
(__atomic_val_bysize, __atomic_bool_bysize): New macros.
(atomic_compare_and_exchange_val_acq): Use it.
(atomic_compare_and_exchange_bool_acq): Likewise.
(atomic_increment_and_test): Invert sense of test.
(atomic_decrement_and_test): Likewise.
* csu/tst-atomic.c: Update those tests to match.
Add comment.
(atomic_decrement_and_test): Add comment.
2003-03-22 Jakub Jelinek <jakub@redhat.com>
* include/atomic.h (atomic_compare_and_exchange_val_acq): Add comment.
Don't define if __arch_compare_and_exchange_val_32_acq is not defined.
(atomic_compare_and_exchange_bool_acq): Add comment. Don't use
__oldval variable in the macro, since it might be macro argument.
(atomic_decrement_if_positive): Initialize __memp, remove setting
of non-existent variable.
(atomic_bit_test_set): Cast 1 to __typeof (*mem) before shifting.
* sysdeps/ia64/bits/atomic.h (atomic_exchange_and_add): Implement
using atomic_compare_and_exchange_val_acq.
(atomic_decrement_if_positive, atomic_bit_test_set): Define.
* sysdeps/s390/bits/atomic.h (__arch_compare_and_exchange_val_8_acq):
Renamed from...
(__arch_compare_and_exchange_bool_8_acq): ... this.
(__arch_compare_and_exchange_val_16_acq): Renamed from...
(__arch_compare_and_exchange_bool_16_acq): ... this.
(__arch_compare_and_exchange_val_32_acq): Return old value. Renamed
from...
(__arch_compare_and_exchange_bool_32_acq): ... this.
(__arch_compare_and_exchange_val_64_acq): Return old value. Renamed
from...
(__arch_compare_and_exchange_bool_64_acq): ... this.
(__arch_compare_and_exchange_val_32_acq): Use __typeof for local
variables types instead of assuming int.
Change prefix of local variables to __arch.
* sysdeps/generic/bits/atomic.h (arch_compare_and_exchange_acq):
Remove.
(atomic_compare_and_exchange_val_acq,
atomic_compare_and_exchange_bool_acq): Define.
* csu/tst-atomic.c: New test.
* csu/tst-atomic-long.c: New test.
* csu/Makefile (tests): Add tst-atomic and tst-atomic-long.
* malloc/memusagestat.c (main): Kill warning if uint64_t is ulong.
* sysdeps/s390/Versions: Add trailing newline.
* sysdeps/unix/sysv/linux/sysconf.c (__sysconf): Kill warning
if INTERNAL_SYSCALL_ERROR_P doesn't use its first argument.
variable. Remove superfluous memory clobber.
* include/atomic.h: Syntax braino fix.
* posix/tst-nice.c (do_test): Use %m formats instead of printing errno
in decimal. Don't bail if niced at start. Just check that nice call
bumps the total at all.
Joseph Myers
Stefan Liebler
Torvald Riegel
Carlos O'Donell
Ondřej Bílka
Allan McRae
Maxim Kuvyrkov
Paul Eggert
Ulrich Drepper
Jakub Jelinek
Ulrich Drepper
Roland McGrath
Andreas Schwab