From e5366c12d0453ab754ab5dc2d1e7b98966b0988c Mon Sep 17 00:00:00 2001 From: Adhemerval Zanella Date: Mon, 27 Aug 2018 09:42:50 -0300 Subject: powerpc: Only enable TLE with PPC_FEATURE2_HTM_NOSC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Linux from 3.9 through 4.2 does not abort HTM transaction on syscalls, instead it suspend and resume it when leaving the kernel. The side-effects of the syscall will always remain visible, even if the transaction is aborted. This is an issue when transaction is used along with futex syscall, on pthread_cond_wait for instance, where the futex call might succeed but the transaction is rolled back leading the pthread_cond object in an inconsistent state. Glibc used to prevent it by always aborting a transaction before issuing a syscall. Linux 4.2 also decided to abort active transaction in syscalls which makes the glibc workaround superfluous. Worse, glibc transaction abortion leads to a performance issue on recent kernels where the HTM state is saved/restore lazily (v4.9). By aborting a transaction on every syscalls, regardless whether a transaction has being initiated before, GLIBS makes the kernel always save/restore HTM state (it can not even lazily disable it after a certain number of syscall iterations). Because of this shortcoming, Transactional Lock Elision is just enabled when it has been explicitly set (either by tunables of by a configure switch) and if kernel aborts HTM transactions on syscalls (PPC_FEATURE2_HTM_NOSC). It is reported that using simple benchmark [1], the context-switch is about 5% faster by not issuing a tabort in every syscall in newer kernels. Checked on powerpc64le-linux-gnu with 4.4.0 kernel (Ubuntu 16.04). * NEWS: Add note about new TLE support on powerpc64le. * sysdeps/powerpc/nptl/tcb-offsets.sym (TM_CAPABLE): Remove. * sysdeps/powerpc/nptl/tls.h (tcbhead_t): Rename tm_capable to __ununsed1. (TLS_INIT_TP, TLS_DEFINE_INIT_TP): Remove tm_capable setup. (THREAD_GET_TM_CAPABLE, THREAD_SET_TM_CAPABLE): Remove macros. * sysdeps/powerpc/powerpc32/sysdep.h, sysdeps/powerpc/powerpc64/sysdep.h (ABORT_TRANSACTION_IMPL, ABORT_TRANSACTION): Remove macros. * sysdeps/powerpc/sysdep.h (ABORT_TRANSACTION): Likewise. * sysdeps/unix/sysv/linux/powerpc/elision-conf.c (elision_init): Set __pthread_force_elision iff PPC_FEATURE2_HTM_NOSC is set. * sysdeps/unix/sysv/linux/powerpc/powerpc32/sysdep.h, sysdeps/unix/sysv/linux/powerpc/powerpc64/sysdep.h sysdeps/unix/sysv/linux/powerpc/syscall.S (ABORT_TRANSACTION): Remove usage. * sysdeps/unix/sysv/linux/powerpc/not-errno.h: Remove file. Reported-by: Breno Leitão (cherry picked from commit f0458cf4f9ff3d870c43b624e6dccaaf657d5e83) --- sysdeps/unix/sysv/linux/powerpc/elision-conf.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'sysdeps/unix/sysv/linux/powerpc/elision-conf.c') diff --git a/sysdeps/unix/sysv/linux/powerpc/elision-conf.c b/sysdeps/unix/sysv/linux/powerpc/elision-conf.c index 906882a65e..fc82bd1ad8 100644 --- a/sysdeps/unix/sysv/linux/powerpc/elision-conf.c +++ b/sysdeps/unix/sysv/linux/powerpc/elision-conf.c @@ -127,6 +127,26 @@ elision_init (int argc __attribute__ ((unused)), TUNABLE_CALLBACK (set_elision_skip_trylock_internal_abort)); #endif + /* Linux from 3.9 through 4.2 do not abort HTM transaction on syscalls, + instead it suspends the transaction and resumes it when returning to + usercode. The side-effects of the syscall will always remain visible, + even if the transaction is aborted. This is an issue when a transaction + is used along with futex syscall, on pthread_cond_wait for instance, + where futex might succeed but the transaction is rolled back leading + the condition variable object in an inconsistent state. + + Glibc used to prevent it by always aborting a transaction before issuing + a syscall. Linux 4.2 also decided to abort active transaction in + syscalls which makes the glibc workaround superflours. Worse, glibc + transaction abortions leads to a performance issues on recent kernels. + + So Lock Elision is just enabled when it has been explict set (either + by tunables of by a configure switch) and if kernel aborts HTM + transactions on syscalls (PPC_FEATURE2_HTM_NOSC) */ + + __pthread_force_elision = (__pthread_force_elision + && GLRO (dl_hwcap2) & PPC_FEATURE2_HTM_NOSC); + if (!__pthread_force_elision) __elision_aconf.try_tbegin = 0; /* Disable elision on rwlocks. */ } -- cgit v1.2.3-70-g09d2