report the per-irq statistics on all arches
Ravikiran G Thirumalai [Wed, 17 Oct 2007 06:26:28 +0000 (23:26 -0700)]
Commit 4004c69ad68dd03733179277280ea2946990ba36 avoids too many remote cpu
references while reporting per-irq stats.  Since we will not have the same
performance penalty of bringing in remote cpu cachelines while reporting
per-irq stats anymore, we can now afford to be consistent and report this
statistic on all arches, all configs.

akpm: affects ia64, alpha and ppc64, mainly.

Kiran earlier said:

Read to /proc/stat takes:
Plain:  2.622832
With speedup patch: 0.013194
With the per-irq stats commented out: 0.008124

So the performance problems which originally caused those architectures to
disable this statistic should now be fixed up.

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

fs/proc/proc_misc.c

index 5de7f87..d6dc72c 100644 (file)
@@ -526,11 +526,8 @@ static int show_stat(struct seq_file *p, void *v)
        }
        seq_printf(p, "intr %llu", (unsigned long long)sum);
 
-#ifndef CONFIG_SMP
-       /* Touches too many cache lines on SMP setups */
        for (i = 0; i < NR_IRQS; i++)
                seq_printf(p, " %u", per_irq_sum[i]);
-#endif
 
        seq_printf(p,
                "\nctxt %llu\n"