This is an archive of the discontinued LLVM Phabricator instance.

[inline asm] dot operator while using imm generates wrong ir + asm
ClosedPublic

Authored by zizhar on May 10 2017, 7:13 AM.

Diff Detail

Repository
rL LLVM

Event Timeline

zizhar created this revision.May 10 2017, 7:13 AM
zizhar edited the summary of this revision. (Show Details)May 10 2017, 7:15 AM
ahatanak accepted this revision.May 10 2017, 3:26 PM

This looks correct.

This revision is now accepted and ready to land.May 10 2017, 3:26 PM
rnk accepted this revision.May 12 2017, 3:45 PM

Looks good.

I would recommend using the github monorepo to send and submit patches that cross clang and LLVM: http://llvm.org/docs/GettingStarted.html#for-developers-to-work-with-a-git-monorepo It should make it easier to review and submit these kinds of changes.

In the glorious future, if you are serious about improving the ms/Intel inline asm parser, we should come up with a test frontend so that we can move most of these tests back into LLVM. Until that day, I recommend the monorepo. :)

This revision was automatically updated to reflect the committed changes.