On 04/28/2016 09:57 AM, Segher Boessenkool wrote:
On various PowerPC configurations, the top frame is often mentioned
twice in the backtrace, making many asan tests fail. I see no particular
reason the asan tests want to check the frame number, so this patch
makes it check for " #. " instead of " #1 ", etc., in all of the
c-c++-common/asan tests.
Tested on powerpc64-linux, also -m32; is this okay for trunk?
Segher
2016-04-28 Segher Boessenkool <seg...@kernel.crashing.org>
gcc/testsuite/
* c-c++-common/asan/global-overflow-1.c: Don't check frame numbers.
* c-c++-common/asan/heap-overflow-1.c: Ditto.
* c-c++-common/asan/memcmp-1.c: Ditto.
* c-c++-common/asan/misalign-1.c: Ditto.
* c-c++-common/asan/misalign-2.c: Ditto.
* c-c++-common/asan/null-deref-1.c: Ditto.
* c-c++-common/asan/pr64820.c: Ditto.
* c-c++-common/asan/sanity-check-pure-c-1.c: Ditto.
* c-c++-common/asan/stack-overflow-1.c: Ditto.
* c-c++-common/asan/strip-path-prefix-1.c: Ditto.
* c-c++-common/asan/strlen-overflow-1.c: Ditto.
* c-c++-common/asan/strncpy-overflow-1.c: Ditto.
* c-c++-common/asan/use-after-free-1.c: Ditto.
* c-c++-common/asan/use-after-return-1.c: Ditto.
One could argue that testing the frame numbers tests is a QofI test for
our backtrace generation and is thus valid.
Has anyone looked into fixing the unwinder so that we're not getting
duplicate frames?
I'd rather not dumb down the test since it is showing a real issue.
jeff
