* [PATCH] fs/proc/array.c: slightly improve render_sigset_t
@ 2016-09-20 22:28 Rasmus Villemoes
2016-09-21 0:16 ` Kees Cook
2016-11-09 21:40 ` Andrei Vagin
0 siblings, 2 replies; 6+ messages in thread
From: Rasmus Villemoes @ 2016-09-20 22:28 UTC (permalink / raw)
To: Andrew Morton, Kees Cook; +Cc: linux-kernel, Rasmus Villemoes
format_decode and vsnprintf occasionally show up in perf top, so I
went looking for places that might not need the full printf
power. With the help of kprobes, I gathered some statistics on which
format strings we mostly pass to vsnprintf. On a trivial desktop
workload, I hit "%x" 25% of the time, so something apparently reads
/proc/pid/status (which does 5*16 printf("%x") calls) a lot.
With this patch, reading /proc/pid/status is 30% faster according to
this microbenchmark:
char buf[4096];
int i, fd;
for (i = 0; i < 10000; ++i) {
fd = open("/proc/self/status", O_RDONLY);
read(fd, buf, sizeof(buf));
close(fd);
}
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
---
fs/proc/array.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 88c7de12197b..7f73b689a15c 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -251,7 +251,7 @@ void render_sigset_t(struct seq_file *m, const char *header,
if (sigismember(set, i+2)) x |= 2;
if (sigismember(set, i+3)) x |= 4;
if (sigismember(set, i+4)) x |= 8;
- seq_printf(m, "%x", x);
+ seq_putc(m, hex_asc[x]);
} while (i >= 4);
seq_putc(m, '\n');
--
2.1.4
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH] fs/proc/array.c: slightly improve render_sigset_t
2016-09-20 22:28 [PATCH] fs/proc/array.c: slightly improve render_sigset_t Rasmus Villemoes
@ 2016-09-21 0:16 ` Kees Cook
2016-09-21 20:51 ` Rasmus Villemoes
2016-10-05 22:48 ` Rasmus Villemoes
2016-11-09 21:40 ` Andrei Vagin
1 sibling, 2 replies; 6+ messages in thread
From: Kees Cook @ 2016-09-21 0:16 UTC (permalink / raw)
To: Rasmus Villemoes; +Cc: Andrew Morton, LKML
On Tue, Sep 20, 2016 at 3:28 PM, Rasmus Villemoes
<linux@rasmusvillemoes.dk> wrote:
> format_decode and vsnprintf occasionally show up in perf top, so I
> went looking for places that might not need the full printf
> power. With the help of kprobes, I gathered some statistics on which
> format strings we mostly pass to vsnprintf. On a trivial desktop
> workload, I hit "%x" 25% of the time, so something apparently reads
> /proc/pid/status (which does 5*16 printf("%x") calls) a lot.
>
> With this patch, reading /proc/pid/status is 30% faster according to
> this microbenchmark:
>
> char buf[4096];
> int i, fd;
> for (i = 0; i < 10000; ++i) {
> fd = open("/proc/self/status", O_RDONLY);
> read(fd, buf, sizeof(buf));
> close(fd);
> }
>
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Heheh. Nice. :)
Acked-by: Kees Cook <keescook@chromium.org>
Out of curiosity, what other stuff ended up near the top? I'd be
curious to see your kprobes too.
-Kees
> ---
> fs/proc/array.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/proc/array.c b/fs/proc/array.c
> index 88c7de12197b..7f73b689a15c 100644
> --- a/fs/proc/array.c
> +++ b/fs/proc/array.c
> @@ -251,7 +251,7 @@ void render_sigset_t(struct seq_file *m, const char *header,
> if (sigismember(set, i+2)) x |= 2;
> if (sigismember(set, i+3)) x |= 4;
> if (sigismember(set, i+4)) x |= 8;
> - seq_printf(m, "%x", x);
> + seq_putc(m, hex_asc[x]);
> } while (i >= 4);
>
> seq_putc(m, '\n');
> --
> 2.1.4
>
--
Kees Cook
Nexus Security
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH] fs/proc/array.c: slightly improve render_sigset_t
2016-09-21 0:16 ` Kees Cook
@ 2016-09-21 20:51 ` Rasmus Villemoes
2016-10-05 22:48 ` Rasmus Villemoes
1 sibling, 0 replies; 6+ messages in thread
From: Rasmus Villemoes @ 2016-09-21 20:51 UTC (permalink / raw)
To: Kees Cook; +Cc: Andrew Morton, LKML
On Wed, Sep 21 2016, Kees Cook <keescook@chromium.org> wrote:
> On Tue, Sep 20, 2016 at 3:28 PM, Rasmus Villemoes
> <linux@rasmusvillemoes.dk> wrote:
>> format_decode and vsnprintf occasionally show up in perf top, so I
>> went looking for places that might not need the full printf
>> power. With the help of kprobes, I gathered some statistics on which
>> format strings we mostly pass to vsnprintf. On a trivial desktop
>> workload, I hit "%x" 25% of the time, so something apparently reads
>> /proc/pid/status (which does 5*16 printf("%x") calls) a lot.
>>
>> With this patch, reading /proc/pid/status is 30% faster according to
>> this microbenchmark:
>>
>> char buf[4096];
>> int i, fd;
>> for (i = 0; i < 10000; ++i) {
>> fd = open("/proc/self/status", O_RDONLY);
>> read(fd, buf, sizeof(buf));
>> close(fd);
>> }
>>
>> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
>
> Heheh. Nice. :)
>
> Acked-by: Kees Cook <keescook@chromium.org>
>
> Out of curiosity, what other stuff ended up near the top? I'd be
> curious to see your kprobes too.
I don't have the results from the other machine handy, but it very much
depends on what one is doing. I ran a 'find $HOME ...', and "%.2x" ended
up accounting for 99%. Turns out ecryptfs does a lot of bin-to-hex
conversions, but very inefficiently (calling sprintf("%.2x") in a
loop...). Patch sent.
Other than that, the top consists of stuff like
"%c%c " (/proc/pid/smaps, VmFlags:)
" %s" (/proc/cpuinfo, the flags: line)
"%*s" (maybe from seq_pad, but not sure who the user of that is)
and a lot of other short strings which are rather hard to trace to their
source, but presumably most are related to some /proc file.
The kprobe is just
echo 'p:vsnprintf vsnprintf fmt=+0(%dx):string' > /sys/kernel/debug/tracing/kprobe_events
This doesn't escape the strings, so embedded newlines mess up the trace
buffer slightly. I used this little piece of line noise to extract the
format strings.
use File::Slurp;
my $txt = read_file($ARGV[0]);
while ($txt =~ m/fmt="(.*?)"\n/mg) {
$_ = $1;
s/\\/\\\\/g; s/\n/\\n/g;
s/\t/\\t/g; s/"/\\"/g;
print "\"$_\"\n";
}
Rasmus
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH] fs/proc/array.c: slightly improve render_sigset_t
2016-09-21 0:16 ` Kees Cook
2016-09-21 20:51 ` Rasmus Villemoes
@ 2016-10-05 22:48 ` Rasmus Villemoes
2016-10-06 22:35 ` Andrew Morton
1 sibling, 1 reply; 6+ messages in thread
From: Rasmus Villemoes @ 2016-10-05 22:48 UTC (permalink / raw)
To: Kees Cook; +Cc: Andrew Morton, LKML
On Wed, Sep 21 2016, Kees Cook <keescook@chromium.org> wrote:
> On Tue, Sep 20, 2016 at 3:28 PM, Rasmus Villemoes
> <linux@rasmusvillemoes.dk> wrote:
>
> Heheh. Nice. :)
>
> Acked-by: Kees Cook <keescook@chromium.org>
>
Andrew, ping?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] fs/proc/array.c: slightly improve render_sigset_t
2016-10-05 22:48 ` Rasmus Villemoes
@ 2016-10-06 22:35 ` Andrew Morton
0 siblings, 0 replies; 6+ messages in thread
From: Andrew Morton @ 2016-10-06 22:35 UTC (permalink / raw)
To: Rasmus Villemoes; +Cc: Kees Cook, LKML
On Thu, 06 Oct 2016 00:48:59 +0200 Rasmus Villemoes <linux@rasmusvillemoes.dk> wrote:
> On Wed, Sep 21 2016, Kees Cook <keescook@chromium.org> wrote:
>
> > On Tue, Sep 20, 2016 at 3:28 PM, Rasmus Villemoes
> > <linux@rasmusvillemoes.dk> wrote:
> >
> > Heheh. Nice. :)
> >
> > Acked-by: Kees Cook <keescook@chromium.org>
> >
>
> Andrew, ping?
Yup, I have this queued for processing after -rc1.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] fs/proc/array.c: slightly improve render_sigset_t
2016-09-20 22:28 [PATCH] fs/proc/array.c: slightly improve render_sigset_t Rasmus Villemoes
2016-09-21 0:16 ` Kees Cook
@ 2016-11-09 21:40 ` Andrei Vagin
1 sibling, 0 replies; 6+ messages in thread
From: Andrei Vagin @ 2016-11-09 21:40 UTC (permalink / raw)
To: Rasmus Villemoes; +Cc: Andrew Morton, Kees Cook, LKML
On Tue, Sep 20, 2016 at 3:28 PM, Rasmus Villemoes
<linux@rasmusvillemoes.dk> wrote:
> format_decode and vsnprintf occasionally show up in perf top, so I
> went looking for places that might not need the full printf
> power. With the help of kprobes, I gathered some statistics on which
> format strings we mostly pass to vsnprintf. On a trivial desktop
> workload, I hit "%x" 25% of the time, so something apparently reads
> /proc/pid/status (which does 5*16 printf("%x") calls) a lot.
>
> With this patch, reading /proc/pid/status is 30% faster according to
> this microbenchmark:
>
> char buf[4096];
> int i, fd;
> for (i = 0; i < 10000; ++i) {
> fd = open("/proc/self/status", O_RDONLY);
> read(fd, buf, sizeof(buf));
> close(fd);
> }
>
Acked-by: Andrei Vagin <avagin@openvz.org>
Andrew, could you replace my patch
"proc-optimize-render_sigset_t.patch" to this one.
Thanks,
Andrei
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
> ---
> fs/proc/array.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/proc/array.c b/fs/proc/array.c
> index 88c7de12197b..7f73b689a15c 100644
> --- a/fs/proc/array.c
> +++ b/fs/proc/array.c
> @@ -251,7 +251,7 @@ void render_sigset_t(struct seq_file *m, const char *header,
> if (sigismember(set, i+2)) x |= 2;
> if (sigismember(set, i+3)) x |= 4;
> if (sigismember(set, i+4)) x |= 8;
> - seq_printf(m, "%x", x);
> + seq_putc(m, hex_asc[x]);
> } while (i >= 4);
>
> seq_putc(m, '\n');
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2016-11-09 21:40 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-09-20 22:28 [PATCH] fs/proc/array.c: slightly improve render_sigset_t Rasmus Villemoes
2016-09-21 0:16 ` Kees Cook
2016-09-21 20:51 ` Rasmus Villemoes
2016-10-05 22:48 ` Rasmus Villemoes
2016-10-06 22:35 ` Andrew Morton
2016-11-09 21:40 ` Andrei Vagin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox