glibc: sched_getcpu(): use rseq cpu_id TLS on Linux (v5)

Started by Mathieu Desnoyers on Aug 7, 2019

Subsystems (1)

Maintainers (1)

DORMANTno replies

open the first message on its own page

Message details and related content
When available, use the cpu_id field from __rseq_abi on Linux to
implement sched_getcpu(). Fall-back on the vgetcpu vDSO if unavailable.

Benchmarks:

x86-64: Intel E5-2630 v3@2.40GHz, 16-core, hyperthreading

glibc sched_getcpu():                     13.7 ns (baseline)
glibc sched_getcpu() using rseq:           2.5 ns (speedup:  5.5x)
inline load cpuid from __rseq_abi TLS:     0.8 ns (speedup: 17.1x)

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
CC: Carlos O'Donell <redacted>
CC: Florian Weimer <redacted>
CC: Joseph Myers <redacted>
CC: Szabolcs Nagy <redacted>
CC: Thomas Gleixner <redacted>
CC: Ben Maurer <redacted>
CC: Peter Zijlstra <peterz@infradead.org>
CC: "Paul E. McKenney" <redacted>
CC: Boqun Feng <redacted>
CC: Will Deacon <redacted>
CC: Dave Watson <redacted>
CC: Paul Turner <redacted>
CC: <redacted>
CC: <redacted>
CC: <redacted>
---
Changes since v1:
- rseq is only used if both __NR_rseq and RSEQ_SIG are defined.

Changes since v2:
- remove duplicated __rseq_abi extern declaration.

Changes since v3:
- update ChangeLog.

Changes since v4:
- Use atomic_load_relaxed to load the __rseq_abi.cpu_id field, a
  consequence of the fact that __rseq_abi is not volatile anymore.
- Include atomic.h which provides atomic_load_relaxed.
---
 ChangeLog                              |  5 +++++
 sysdeps/unix/sysv/linux/sched_getcpu.c | 25 +++++++++++++++++++++++--
 2 files changed, 28 insertions(+), 2 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 9a1930248f..128a7b4baa 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+2019-08-06  Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
+
+	* sysdeps/unix/sysv/linux/sched_getcpu.c: use rseq cpu_id TLS on
+	Linux.
+
 2019-08-06  Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
 
 	* NEWS: Add Restartable Sequences feature description.
diff --git a/sysdeps/unix/sysv/linux/sched_getcpu.c b/sysdeps/unix/sysv/linux/sched_getcpu.c
index fb0d317f83..01a818f5b0 100644
--- a/sysdeps/unix/sysv/linux/sched_getcpu.c
+++ b/sysdeps/unix/sysv/linux/sched_getcpu.c
@@ -18,14 +18,15 @@
 #include <errno.h>
 #include <sched.h>
 #include <sysdep.h>
+#include <atomic.h>
 
 #ifdef HAVE_GETCPU_VSYSCALL
 # define HAVE_VSYSCALL
 #endif
 #include <sysdep-vdso.h>
 
-int
-sched_getcpu (void)
+static int
+vsyscall_sched_getcpu (void)
 {
 #ifdef __NR_getcpu
   unsigned int cpu;
@@ -37,3 +38,23 @@ sched_getcpu (void)
   return -1;
 #endif
 }
+
+#ifdef __NR_rseq
+#include <sys/rseq.h>
+#endif
+
+#if defined __NR_rseq && defined RSEQ_SIG
+int
+sched_getcpu (void)
+{
+  int cpu_id = atomic_load_relaxed (&__rseq_abi.cpu_id);
+
+  return cpu_id >= 0 ? cpu_id : vsyscall_sched_getcpu ();
+}
+#else
+int
+sched_getcpu (void)
+{
+  return vsyscall_sched_getcpu ();
+}
+#endif
-- 
2.17.1
Keyboard shortcuts
hback out one level
jnext message in thread / next series-diff position
kprevious message in thread / previous series-diff position
ldrill in
Escclose help / fold thread tree
?toggle this help