Hi Mikael,
On 4/3/26 11:34, Mikael Morin wrote:
I agree, if the argument is a dummy procedure or dummy procedure pointer
we don't want the clobber, but if the argument is a local procedure
pointer, we do want the clobber, don't we?
in principle we do.
I think we should check that the var clobbered is a VAR_DECL, or maybe a
PARM_DECL or RESULT_DECL (or a conversion of such a DECL?).
If you don't have time to refine it further, please go ahead with your
patch as is. At least it prevents the ICE.
The cases excluding a clobber are already rather lengthy, so this calls
for outlining this part from "the beast" gfc_conv_procedure_call.
I've pushed the patch as-is as r16-8444-gae9b5fc970cd4c and will open
a PR on the missed optimization.
Thanks for the review!
Harald
Mikael
