https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85381
Tom de Vries <vries at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #43992|0 |1
is obsolete| |
--- Comment #9 from Tom de Vries <vries at gcc dot gnu.org> ---
Created attachment 43999
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43999&action=edit
Tentative patch inserting dummy instructions
(In reply to Tom de Vries from comment #7)
> We may wanna insert dummy ops inbetween (it would be nice if something less
> heavy than a membar.cta will work).
This patch implements this approach. I've reported the bug at nvidia and see if
they come up with a better or more restricted workaround.
