Here are 2 small profile/count fixes I found while trying to fix pr90178.c (for -m32). These don't fix the testcase which seems like need a RA change but they will definitely help other places were there was mismatch happening.
Andrew Pinski (2): cfg: Fix count when creating new forwarder block ldist: Fix probability on loop exit gcc/tree-cfg.cc | 2 +- gcc/tree-loop-distribution.cc | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) -- 2.43.0
