On 18/05/2017 20:19, Naveen N. Rao wrote:
> Paolo Bonzini wrote:
>> The ARM and x86 architectures already use libdw, and it is useful to
>> have as much common code for the unwinder as possible. Porting PPC
>> to libdw only needs an architecture-specific hook to move the register
>> state from perf to libdw.
>
> Thanks. Ravi has had a similar patch locally, but from what I
> understand, there are issues with libdw based unwinding on powerpc64. I
> gave this a quick test and I don't see the user-space call trace being
> unwinded properly with libdw.
I don't see that problem:
- 99,98% 0,00% dd libc-2.17.so [.] generic_start_main.isra.0
generic_start_main.isra.0
- main
- 99,97% iread
- 97,82% sys_read
- 96,97% extract_entropy_user
89,44% powernv_get_random_long
4,63% sha_transform
2,07% extract_buf
1,15% _raw_spin_lock_irqsave
0,51% extract_buf
(This is "perf record dd if=/dev/urandom of=/dev/null bs=512").
Can you copy the contents of tools/perf/.config-detected here?
>> diff --git a/tools/perf/arch/powerpc/util/unwind-libdw.c
>> b/tools/perf/arch/powerpc/util/unwind-libdw.c
>> new file mode 100644
>> index 0000000..c61e1d8
>> --- /dev/null
>> +++ b/tools/perf/arch/powerpc/util/unwind-libdw.c
>> @@ -0,0 +1,73 @@
>> +#include <elfutils/libdwfl.h>
>> +#include "../../util/util.h"
>> +#include "../../util/unwind-libdw.h"
>> +#include "../../util/perf_regs.h"
>
> As an aside, we also need to include "event.h", without which there are
> a few compile errors.
Oops, I was compiling this on 4.11. Sorry.
Paolo
