On March 27, 2020 4:32:58 PM GMT+01:00, Jakub Jelinek <ja...@redhat.com> wrote: >On Fri, Mar 27, 2020 at 10:57:05AM +0100, Richard Biener wrote: >> > That would just mean it ICEs at -g0 too, this PR isn't a >-fcompare-debug >> > issue but a bug where we try to gimplify the same STATEMENT_LIST >multiple >> > times and second time ICE on it as it has been voidified and >emptied. >> >> But only because we made STATEMENT_LISTs not having side-effects >which >> they all did before all the changes. We could revert that... > >So perhaps we could try to do that only for the C/C++ statement >expressions >which seems to be the most common reason of all these -fcompare-debug >-gstatement-frontiers related PRs, while it wouldn't consume too much >memory >in the common case? >I.e. when parsing a statement expression in the C/C++ FEs, ensure the >result >is wrapped in a STATEMENT_LIST with TREE_SIDE_EFFECTS set, no matter if >it >contains just a single stmt or more inside of it?
Sounds worth a try. Richard. > Jakub
