perf record: Fix branch_stack type in perf_record_opts
authorStephane Eranian <eranian@google.com>
Fri, 25 May 2012 21:13:44 +0000 (23:13 +0200)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Fri, 25 May 2012 21:32:44 +0000 (18:32 -0300)
The attr.branch_sample_type field is defined as u64 by the API.  As
such, we need to ensure the variable holding the value of the branch
stack filters is also u64 otherwise we may lose bits in the future.

Note also that the bogus definition of the field in perf_record_opts
caused problems on big-endian PPC systems.  Thanks to Anshuman Khandual
for tracking the problem on PPC.

Reported-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120525211344.GA7729@quad
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/perf.h

index 14f1034f14f93efbef0b7e5a7ea3f642674fc856..f960ccb2edc6f38f8a7b7a3f0f1b740cbd351c2b 100644 (file)
@@ -227,7 +227,7 @@ struct perf_record_opts {
        unsigned int freq;
        unsigned int mmap_pages;
        unsigned int user_freq;
-       int          branch_stack;
+       u64          branch_stack;
        u64          default_interval;
        u64          user_interval;
 };