On 01/14/2016 03:15 AM, Kyrill Tkachov wrote:
There are cases where calls to fold_rtx with a non-NULL insn don't
end up modifying src, ending up in src to not be always equal to
src_folded (I added an assert to that effect and saw it trigger). It
seems that fold_rtx is not *guaranteed* to modify
Hi Jeff,
On 13/01/16 19:28, Jeff Law wrote:
On 01/13/2016 04:33 AM, Kyrill Tkachov wrote:
I've been able to get it to do the right thing by changing the line
where it initially folds the source of the SET. That is line 4639 in
cse.c: /* Simplify and foldable subexpressions in SRC. Then get th
On 01/13/2016 04:33 AM, Kyrill Tkachov wrote:
I've been able to get it to do the right thing by changing the line
where it initially folds the source of the SET. That is line 4639 in
cse.c: /* Simplify and foldable subexpressions in SRC. Then get the
fully- simplified result, which may not nece
Hi all,
I'm looking at mitigating some of the codegen regressions on arm that come from
a proposed fix
to PR 65932 and I believe RTL CSE is getting in the way.
The problematic sequences involve sign-extending loads and sign-extending
multiplies.
From the gcc.target/arm/wmul-1.c case we have:
(
