https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125800

--- Comment #5 from Zhongyao Chen <chenzhongyao.hit at gmail dot com> ---
Created attachment 64824
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=64824&action=edit
patch for test

Hi Josef,

I suspect the SubMtx function in CalculiX is the issue.

It seems to generate very bad vectorized code, maybe because the upthread
swappable condition in my commit is too loose.

I attached a small patch only for test. Could you help run a performance test
and see if it fixes this issue?

Reply via email to