]> git.karo-electronics.de Git - karo-tx-linux.git/commit
perf powerpc: Fix kprobe and kretprobe handling with kallsyms on ppc64le
authorNaveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Tue, 12 Apr 2016 09:10:49 +0000 (14:40 +0530)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Fri, 6 May 2016 00:04:02 +0000 (21:04 -0300)
commit239aeba764092b29dd7cab177cd47f472390622e
treecc917bab88b9b11e60d98557831ea922d0103d2a
parent7cecb7fe8388d5c39708d7f07e642ed31bb9d4fe
perf powerpc: Fix kprobe and kretprobe handling with kallsyms on ppc64le

So far, we used to treat probe point offsets as being offset from the
LEP. However, userspace applications (objdump/readelf) always show
disassembly and offsets from the function GEP. This is confusing to the
user as we will end up probing at an address different from what the
user expects when looking at the function disassembly with
readelf/objdump. Fix this by changing how we modify probe address with
perf.

If only the function name is provided, we assume the user needs the LEP.
Otherwise, if an offset is specified, we assume that the user knows the
exact address to probe based on function disassembly, and so we just
place the probe from the GEP offset.

Finally, kretprobe was also broken with kallsyms as we were trying to
specify an offset. This patch also fixes that issue.

Reported-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Acked-by: Balbir Singh <bsingharora@gmail.com>
Cc: Mark Wielaard <mjw@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/75df860aad8216bf4b9bcd10c6351ecc0e3dee54.1460451721.git.naveen.n.rao@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/arch/powerpc/util/sym-handling.c