8bfc4a2ab4
This patch defines __libc_tbegin, __libc_tend, __libc_tabort and __libc_tx_nesting_depth in htm.h which replaces the direct usage of equivalent gcc builtins. We have to use an own inline assembly instead of __builtin_tbegin, as tbegin has to filter program interruptions which can't be done with the builtin. Before this change, e.g. a segmentation fault within a transaction, leads to a coredump where the instruction pointer points behind the tbegin instruction instead of real failing one. Now the transaction aborts and the code should be reexecuted by the fallback path without transactions. The segmentation fault will produce a coredump with the real failing instruction. The fpc is not saved before starting the transaction. If e.g. the rounging mode is changed and the transaction is aborting afterwards, the builtin will not restore the fpc. This is now done with the __libc_tbegin macro. Now the call saved fprs have to be saved / restored in the __libc_tbegin macro. Using the gcc builtin had forced the saving / restoring of fprs at begin / end of e.g. __lll_lock_elision function. The new macro saves these fprs before tbegin instruction and only restores them on a transaction abort. Restoring is not needed on a successfully started transaction. The used inline assembly does not clobber the fprs / vrs! Clobbering the latter ones would force the compiler to save / restore the call saved fprs as those overlap with the vrs, but they only need to be restored if the transaction fails. Thus the user of the tbegin macros has to compile the file / function with -msoft-float. It prevents gcc from using fprs / vrs. ChangeLog: * sysdeps/unix/sysv/linux/s390/Makefile (elision-CFLAGS): Add -msoft-float. * sysdeps/unix/sysv/linux/s390/htm.h: New File. * sysdeps/unix/sysv/linux/s390/elision-lock.c: Use __libc_t* transaction macros instead of __builtin_t*. * sysdeps/unix/sysv/linux/s390/elision-trylock.c: Likewise. * sysdeps/unix/sysv/linux/s390/elision-unlock.c: Likewise.
95 lines
3.3 KiB
C
95 lines
3.3 KiB
C
/* Elided pthread mutex trylock.
|
|
Copyright (C) 2014-2016 Free Software Foundation, Inc.
|
|
This file is part of the GNU C Library.
|
|
|
|
The GNU C Library is free software; you can redistribute it and/or
|
|
modify it under the terms of the GNU Lesser General Public
|
|
License as published by the Free Software Foundation; either
|
|
version 2.1 of the License, or (at your option) any later version.
|
|
|
|
The GNU C Library is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
Lesser General Public License for more details.
|
|
|
|
You should have received a copy of the GNU Lesser General Public
|
|
License along with the GNU C Library; if not, see
|
|
<http://www.gnu.org/licenses/>. */
|
|
|
|
#include <pthread.h>
|
|
#include <pthreadP.h>
|
|
#include <lowlevellock.h>
|
|
#include <htm.h>
|
|
#include <elision-conf.h>
|
|
|
|
#define aconf __elision_aconf
|
|
|
|
/* Try to elide a futex trylock. FUTEX is the futex variable. ADAPT_COUNT is
|
|
the adaptation counter in the mutex. */
|
|
|
|
int
|
|
__lll_trylock_elision (int *futex, short *adapt_count)
|
|
{
|
|
/* Implement POSIX semantics by forbiding nesting elided trylocks.
|
|
Sorry. After the abort the code is re-executed
|
|
non transactional and if the lock was already locked
|
|
return an error. */
|
|
if (__libc_tx_nesting_depth () > 0)
|
|
{
|
|
/* Note that this abort may terminate an outermost transaction that
|
|
was created outside glibc.
|
|
This persistently aborts the current transactions to force
|
|
them to use the default lock instead of retrying transactions
|
|
until their try_tbegin is zero.
|
|
*/
|
|
__libc_tabort (_HTM_FIRST_USER_ABORT_CODE | 1);
|
|
}
|
|
|
|
/* Only try a transaction if it's worth it. See __lll_lock_elision for
|
|
why we need atomic accesses. Relaxed MO is sufficient because this is
|
|
just a hint. */
|
|
if (atomic_load_relaxed (adapt_count) <= 0)
|
|
{
|
|
int status;
|
|
|
|
if (__builtin_expect
|
|
((status = __libc_tbegin ((void *) 0)) == _HTM_TBEGIN_STARTED, 1))
|
|
{
|
|
if (*futex == 0)
|
|
return 0;
|
|
/* Lock was busy. Fall back to normal locking. */
|
|
/* Since we are in a non-nested transaction there is no need to abort,
|
|
which is expensive. */
|
|
__libc_tend ();
|
|
/* Note: Changing the adapt_count here might abort a transaction on a
|
|
different cpu, but that could happen anyway when the futex is
|
|
acquired, so there's no need to check the nesting depth here.
|
|
See above for why relaxed MO is sufficient. */
|
|
if (aconf.skip_lock_busy > 0)
|
|
atomic_store_relaxed (adapt_count, aconf.skip_lock_busy);
|
|
}
|
|
else
|
|
{
|
|
if (status != _HTM_TBEGIN_TRANSIENT)
|
|
{
|
|
/* A persistent abort (cc 1 or 3) indicates that a retry is
|
|
probably futile. Use the normal locking now and for the
|
|
next couple of calls.
|
|
Be careful to avoid writing to the lock. */
|
|
if (aconf.skip_trylock_internal_abort > 0)
|
|
*adapt_count = aconf.skip_trylock_internal_abort;
|
|
}
|
|
}
|
|
/* Could do some retries here. */
|
|
}
|
|
else
|
|
{
|
|
/* Lost updates are possible, but harmless. Due to races this might lead
|
|
to *adapt_count becoming less than zero. */
|
|
atomic_store_relaxed (adapt_count,
|
|
atomic_load_relaxed (adapt_count) - 1);
|
|
}
|
|
|
|
return lll_trylock (*futex);
|
|
}
|