Commit Graph

16 Commits

Author SHA1 Message Date
serge-sans-paille 3c8bf29f14
Reduce the number of attributes attached to each function
This takes advantage of the implicit default behavior to reduce the number of
attributes, which in turns reduces compilation time. I've observed -3% in
instruction count when compiling sqlite3 amalgamation with -O0

Differential Revision: https://reviews.llvm.org/D96400
2021-02-16 16:19:54 +01:00
Melanie Blower f4aaed3bf1 Reland D81869 "Modify FPFeatures to use delta not absolute settings"
This reverts commit defd43a5b3.
with correction to solve msan report

To solve https://bugs.llvm.org/show_bug.cgi?id=46166 where the
floating point settings in PCH files aren't compatible, rewrite
FPFeatures to use a delta in the settings rather than absolute settings.
With this patch, these floating point options can be benign.

Reviewers: rjmccall

Differential Revision: https://reviews.llvm.org/D81869
2020-06-27 01:34:57 -07:00
Melanie Blower defd43a5b3 Revert "Revert "Revert "Modify FPFeatures to use delta not absolute settings"""
This reverts commit 9518763d71.
Memory sanitizer fails in CGFPOptionsRAII::CGFPOptionsRAII dtor
2020-06-26 08:47:04 -07:00
Melanie Blower 9518763d71 Revert "Revert "Modify FPFeatures to use delta not absolute settings""
This reverts commit b55d723ed6.
Reapply Modify FPFeatures to use delta not absolute settings

To solve https://bugs.llvm.org/show_bug.cgi?id=46166 where the
floating point settings in PCH files aren't compatible, rewrite
FPFeatures to use a delta in the settings rather than absolute settings.
With this patch, these floating point options can be benign.

Reviewers: rjmccall

Differential Revision: https://reviews.llvm.org/D81869
2020-06-26 08:00:08 -07:00
Melanie Blower b55d723ed6 Revert "Modify FPFeatures to use delta not absolute settings"
This reverts commit 3a748cbf86.
I'm reverting this commit because I forgot to format the commit message
propertly. Sorry for the thrash.
2020-06-26 07:52:57 -07:00
Melanie Blower 3a748cbf86 Modify FPFeatures to use delta not absolute settings 2020-06-26 07:41:09 -07:00
John McCall 8a8d703be0 Fix how cc1 command line options are mapped into FP options.
Canonicalize on storing FP options in LangOptions instead of
redundantly in CodeGenOptions.  Incorporate -ffast-math directly
into the values of those LangOptions rather than considering it
separately when building FPOptions.  Build IR attributes from
those options rather than a mix of sources.

We should really simplify the driver/cc1 interaction here and have
the driver pass down options that cc1 directly honors.  That can
happen in a follow-up, though.

Patch by Michele Scandale!
https://reviews.llvm.org/D80315
2020-06-01 22:00:30 -04:00
Melanie Blower 827be690dc [clang] FastMathFlags.allowContract should be initialized only from FPFeatures.allowFPContractAcrossStatement
Summary: Fix bug introduced in D72841 adding support for pragma float_control

Reviewers: rjmccall, Anastasia

Differential Revision: https://reviews.llvm.org/D79903
2020-05-20 06:19:10 -07:00
Melanie Blower f5360d4bb3 Reapply "Add support for #pragma float_control" with buildbot fixes
Add support for #pragma float_control

Reviewers: rjmccall, erichkeane, sepavloff

Differential Revision: https://reviews.llvm.org/D72841

This reverts commit fce82c0ed3.
2020-05-04 05:51:25 -07:00
Melanie Blower fce82c0ed3 Revert "Reapply "Add support for #pragma float_control" with improvements to"
This reverts commit 69aacaf699.
2020-05-01 10:31:09 -07:00
Melanie Blower 69aacaf699 Reapply "Add support for #pragma float_control" with improvements to
test cases
Add support for #pragma float_control

Reviewers: rjmccall, erichkeane, sepavloff

Differential Revision: https://reviews.llvm.org/D72841

This reverts commit 85dc033cac, and makes
corrections to the test cases that failed on buildbots.
2020-05-01 10:03:30 -07:00
Melanie Blower 85dc033cac Revert "Add support for #pragma float_control"
This reverts commit 4f1e9a17e9.
due to fail on buildbot, sorry for the noise
2020-05-01 06:36:58 -07:00
Melanie Blower 4f1e9a17e9 Add support for #pragma float_control
Reviewers: rjmccall, erichkeane, sepavloff

Differential Revision: https://reviews.llvm.org/D72841
2020-05-01 06:14:24 -07:00
Egor Churaev 45c26ee0bf [OpenCL] Extended mapping of parcing CodeGen arguments
Summary: Enable cl_mad_enamle and cl_no_signed_zeros options when user turns on cl_unsafe_math_optimizations or cl_fast_relaxed_math options.

Reviewers: Anastasia, cfe-commits

Reviewed By: Anastasia

Subscribers: bader, yaxunl

Differential Revision: https://reviews.llvm.org/D31324

llvm-svn: 298838
2017-03-27 10:38:01 +00:00
Sanjay Patel a932bb93b0 Remove the cl-no-signed-zeros cc1 option
Use the driver flag -fno-signed-zeros instead.

This was recommended but not implemented in D6873:
http://reviews.llvm.org/D6873

which was checked in at r226915:
http://reviews.llvm.org/rL226915

llvm-svn: 234093
2015-04-04 14:54:24 +00:00
Pekka Jaaskelainen 3701450b06 OpenCL C: Add support for a set of floating point
arithmetic relaxation flags:

-cl-no-signed-zeros
-cl-unsafe-math-optimizations
-cl-finite-math-only
-cl-fast-relaxed-math

Propagate the info to FP instruction flags as well
as function attributes where they are available.

llvm-svn: 223928
2014-12-10 16:41:14 +00:00