public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] perf annotate: Numeric assembly labels cause incorrect annotation
@ 2012-02-07  3:55 Anton Blanchard
  2012-02-07 14:25 ` Eric B Munson
  2012-02-07 14:40 ` Arnaldo Carvalho de Melo
  0 siblings, 2 replies; 7+ messages in thread
From: Anton Blanchard @ 2012-02-07  3:55 UTC (permalink / raw)
  To: Peter Zijlstra, Paul Mackerras, Ingo Molnar,
	Arnaldo Carvalho de Melo, Frederic Weisbecker, emunson, imunsie,
	eranian
  Cc: linux-kernel


A perf annotate of a kernel function written in assembly shows
very strange percentages:

         :      _GLOBAL(__copy_tofrom_user_base)

...

         :              addi    r3,r3,1
   99.67 :      c00000000004d78c:       addi    r3,r3,1
         :
    0.00 :      1:      bf      cr7*4+2,2f
    0.07 :      c00000000004d790:       bne-    cr7,c00000000004d7a4
         :      err1;   lhz     r0,0(r4)
    0.00 :      c00000000004d794:       lhz     r0,0(r4)
         :              addi    r4,r4,2
    0.00 :      c00000000004d798:       addi    r4,r4,2
         :      err1;   sth     r0,0(r3)
    0.00 :      c00000000004d79c:       sth     r0,0(r3)
         :              addi    r3,r3,2
   99.60 :      c00000000004d7a0:       addi    r3,r3,2
         :
    0.00 :      2:      bf      cr7*4+1,3f
    0.12 :      c00000000004d7a4:       ble-    cr7,c00000000004d7b8
         :      err1;   lwz     r0,0(r4)
    0.00 :      c00000000004d7a8:       lwz     r0,0(r4)
         :              addi    r4,r4,4
    0.00 :      c00000000004d7ac:       addi    r4,r4,4
         :      err1;   stw     r0,0(r3)
    0.00 :      c00000000004d7b0:       stw     r0,0(r3)
         :              addi    r3,r3,4
   99.48 :      c00000000004d7b4:       addi    r3,r3,4

~300% in one function. Urgh.

This is caused by the way we parse objdump -S output, eg:

        addi    r3,r3,1
c00000000004d78c:       addi    r3,r3,1

1:      bf      cr7*4+2,2f
c00000000004d790:       bne-    cr7,c00000000004d7a4 

We assume the asm label (1:) is an address, compute a bogus offset
into the function and then screw up the matching of samples to lines.
I notice this also fails with c inline assembly in a similar
manner.

We already have a sanity check that the address is not beyond the end
of the function, so add a check against the start too.

Signed-off-by: Anton Blanchard <anton@samba.org>
---
Cc: <stable@kernel.org>

Index: linux-tip/tools/perf/util/annotate.c
===================================================================
--- linux-tip.orig/tools/perf/util/annotate.c	2012-01-09 17:45:09.056373433 +1100
+++ linux-tip/tools/perf/util/annotate.c	2012-02-07 13:53:01.610970209 +1100
@@ -244,7 +244,7 @@ static int symbol__parse_objdump_line(st
 		    end = map__rip_2objdump(map, sym->end);
 
 		offset = line_ip - start;
-		if (offset < 0 || (u64)line_ip > end)
+		if (offset < 0 || (u64)line_ip < start || (u64)line_ip > end)
 			offset = -1;
 	}
 

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2012-02-09 15:50 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-02-07  3:55 [PATCH] perf annotate: Numeric assembly labels cause incorrect annotation Anton Blanchard
2012-02-07 14:25 ` Eric B Munson
2012-02-07 14:40 ` Arnaldo Carvalho de Melo
2012-02-09 10:10   ` Mike Galbraith
2012-02-09 13:31     ` Ingo Molnar
2012-02-09 14:50       ` Mike Galbraith
2012-02-09 15:50       ` Mike Galbraith

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox