[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

ro added a comment. That claim of 16-byte alignment on Solaris is half-wrong: it's definitely wrong on Solaris, but seems to be true on Illumos. However, there's currently no way to distinguish the two other than checking `uname -v` at runtime in a native compiler. The configure triples

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

ro added a comment. In D87615#2271646 , @efriedma wrote: > > X86Subtarget::initSubtargetFeatures claims that that stack is 16-byte aligned > on Solaris; if that's wrong, we should fix it, not force stack realignment. I missed that (my usual issue of

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

joerg requested changes to this revision. joerg added a comment. This revision now requires changes to proceed. I don't think this is the right place for this at all. Look at `X86Subtarget::initSubtargetFeatures` please. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

efriedma added a comment. In D87615#2271297 , @ro wrote: > In D87615#2271268 , @RKSimon wrote: > >> Would it be possible to add some tests? > > Probably not reliably: the tests that usually fail bye the hundreds

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

ro added a comment. In D87615#2271268 , @RKSimon wrote: > Would it be possible to add some tests? Probably not reliably: the tests that usually fail bye the hundreds happen to `PASS` once in a while when the stack pointer is 16-byte aligned by accident.

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

RKSimon added a comment. Would it be possible to add some tests? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D87615/new/ https://reviews.llvm.org/D87615 ___ cfe-commits mailing list

[PATCH] D87615: [clang][Driver] Force stack realignment on 32-bit Solaris/x86

ro created this revision. ro added reviewers: MaskRay, rahmanl, efriedma, RKSimon. ro added a project: clang. Herald added a subscriber: fedor.sergeev. ro requested review of this revision. On Solaris/x86, several hundred 32-bit tests `FAIL`, all in the same way: env