Add -m[no-]x87 and -m[no-]80387 options to control FeatureX87.
-m[no-]80387 options is added for compatibility with GCC.
Details
Details
- Reviewers
bruno echristo rsmith - Commits
- rGb7a29675fcd1: [X86] Add -m[no-]x87 and -m[no-]80387 options to control FeatureX87.
rGf98266e2112d: Add missing -mno-cx16 driver option.
rGd90884944b67: Add a test for driver options from m_x86_Features_Group.
rC268488: Add missing -mno-cx16 driver option.
rC268489: [X86] Add -m[no-]x87 and -m[no-]80387 options to control FeatureX87.
rC268487: Add a test for driver options from m_x86_Features_Group.
rL268489: [X86] Add -m[no-]x87 and -m[no-]80387 options to control FeatureX87.
rL268488: Add missing -mno-cx16 driver option.
rL268487: Add a test for driver options from m_x86_Features_Group.
Diff Detail
Diff Detail
- Repository
- rL LLVM
Event Timeline
Comment Actions
Hi,
Thanks for the review.
All m_x86_Features_Group options are handled in function handleTargetFeaturesGroup from lib/Driver/Tools.cpp, so there is no additional code needed.
There was no test for this, so I added a new one trying to cover all options from m_x86_Features_Group.
include/clang/Driver/Options.td | ||
---|---|---|
1406 ↗ | (On Diff #55579) | The test showed me that this flag is missing so I added it. |
Comment Actions
Nice. If you wouldn't mind splitting out into 3 patches (all ok):
a) The initial testcase,
b) the mno-cx16 patch and associated test
c) The x87 part and associated test.
You'll also want to put in that there's definitely a bit of "last one wins" on the testcases and so that the test case isn't exhaustive.
Thanks!
-eric
Comment Actions
Hi,
Thanks for the review.
Committed:
- Add a test for driver options from m_x86_Features_Group (http://reviews.llvm.org/rL268487)
- Add missing -mno-cx16 driver option (http://reviews.llvm.org/rL268488)
- Add -m[no-]x87 and -m[no-]80387 options to control FeatureX87 (http://reviews.llvm.org/rL268489)