On Mon, Mar 7, 2016 at 1:03 PM, Jakub Jelinek <ja...@redhat.com> wrote: > On Mon, Mar 07, 2016 at 12:57:20PM +0100, Richard Biener wrote: >> > Honza, can you please repost the patch. Richard said on IRC that he may >> > reconsider his rejection after all. >> > >> > I've tested the patch on my Gentoo test machine and it fixes segfaults >> > in LLVM, QT, Chromium, Kdevelop. >> > >> > If the patch gets accepted, changes.html and porting_to.html need to be >> > updated to reflect the new flag. >> >> Note that I'll only re-consider if we want to make this flag enabled by >> default. > > By enabled by default you mean what exactly? That by default this will not > be treated as non-NULL anymore?
Yes. > I'm against that, while there are large > codebases that are coded in "C++", there are plenty of properly written > packages and by deferring this optimization we'd just delay the fixing of > firefox etc. True. Then I'm against a new flag. >> (no strong opinion about that - but for example the recent -flifetime-dse >> strengthening fallout is similar) > > -flifetime-dse change is different, we default to -flifetime-dse=2, but have > a way for broken packages to workaround their bugs, so that situation is the > same as with -fno-delete-null-pointer-checks. Well, if anybody would have asked I'd have said we don't want that fine-grained control either there - people could have just used -fno-lifetime-dse. Unless we want to change default behavior back to what GCC 5 did, but see above. Richard. > Jakub
