On Wed, 22 Nov 2017, Jakub Jelinek wrote:
On Wed, Nov 22, 2017 at 06:34:08PM +0100, Marc Glisse wrote:
Hello,
I hadn't implemented this simplification because I think it is invalid code,
but apparently people do write it, so we might as well handle it sensibly.
This also happens to work around PR 83104 (already fixed).
bootstrap+regtest on powerpc64le-unknown-linux-gnu.
I guess we'll need to disable it for -fsanitize=pointer-subtraction
once/if it makes it in, because it should be still diagnosed as invalid.
I am not sure you'll need to disable it. From what I've seen, the patch
inserts a call to BUILT_IN_ASAN_POINTER_SUBTRACT before the
POINTER_DIFF_EXPR is created, and I don't see why simplifying
POINTER_DIFF_EXPR would affect that call. But then again, I don't know how
the sanitizers work, so I can easily misinterpret things.
--
Marc Glisse
