[2/5] glibc: sched_getcpu(): use rseq cpu_id TLS on Linux (v2)
Commit Message
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 <carlos@redhat.com>
CC: Florian Weimer <fweimer@redhat.com>
CC: Joseph Myers <joseph@codesourcery.com>
CC: Szabolcs Nagy <szabolcs.nagy@arm.com>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Ben Maurer <bmaurer@fb.com>
CC: Peter Zijlstra <peterz@infradead.org>
CC: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
CC: Boqun Feng <boqun.feng@gmail.com>
CC: Will Deacon <will.deacon@arm.com>
CC: Dave Watson <davejwatson@fb.com>
CC: Paul Turner <pjt@google.com>
CC: libc-alpha@sourceware.org
CC: linux-kernel@vger.kernel.org
CC: linux-api@vger.kernel.org
---
Changes since v1:
- rseq is only used if both __NR_rseq and RSEQ_SIG are defined.
---
sysdeps/unix/sysv/linux/sched_getcpu.c | 27 ++++++++++++++++++++++++--
1 file changed, 25 insertions(+), 2 deletions(-)
Comments
On 16/04/2019 18:32, Mathieu Desnoyers wrote:
> --- a/sysdeps/unix/sysv/linux/sched_getcpu.c
> +++ b/sysdeps/unix/sysv/linux/sched_getcpu.c
> @@ -37,3 +37,26 @@ sched_getcpu (void)
> return -1;
> #endif
> }
> +
> +#ifdef __NR_rseq
> +#include <sys/rseq.h>
> +#endif
> +
> +#if defined __NR_rseq && defined RSEQ_SIG
> +extern __attribute__ ((tls_model ("initial-exec")))
> +__thread volatile struct rseq __rseq_abi;
i'd expect sys/rseq.h to provide this declaration.
> +
> +int
> +sched_getcpu (void)
> +{
> + int cpu_id = __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
>
----- On Apr 18, 2019, at 11:33 AM, Szabolcs Nagy Szabolcs.Nagy@arm.com wrote:
> On 16/04/2019 18:32, Mathieu Desnoyers wrote:
>> --- a/sysdeps/unix/sysv/linux/sched_getcpu.c
>> +++ b/sysdeps/unix/sysv/linux/sched_getcpu.c
>> @@ -37,3 +37,26 @@ sched_getcpu (void)
>> return -1;
>> #endif
>> }
>> +
>> +#ifdef __NR_rseq
>> +#include <sys/rseq.h>
>> +#endif
>> +
>> +#if defined __NR_rseq && defined RSEQ_SIG
>> +extern __attribute__ ((tls_model ("initial-exec")))
>> +__thread volatile struct rseq __rseq_abi;
>
> i'd expect sys/rseq.h to provide this declaration.
And it actually does! Will remove this duplicate.
Thanks,
Mathieu
>
>> +
>> +int
>> +sched_getcpu (void)
>> +{
>> + int cpu_id = __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
@@ -24,8 +24,8 @@
#endif
#include <sysdep-vdso.h>
-int
-sched_getcpu (void)
+static int
+vsyscall_sched_getcpu (void)
{
#ifdef __NR_getcpu
unsigned int cpu;
@@ -37,3 +37,26 @@ sched_getcpu (void)
return -1;
#endif
}
+
+#ifdef __NR_rseq
+#include <sys/rseq.h>
+#endif
+
+#if defined __NR_rseq && defined RSEQ_SIG
+extern __attribute__ ((tls_model ("initial-exec")))
+__thread volatile struct rseq __rseq_abi;
+
+int
+sched_getcpu (void)
+{
+ int cpu_id = __rseq_abi.cpu_id;
+
+ return cpu_id >= 0 ? cpu_id : vsyscall_sched_getcpu ();
+}
+#else
+int
+sched_getcpu (void)
+{
+ return vsyscall_sched_getcpu ();
+}
+#endif