r366212 - [OpenCL] Fixing sampler initialisations for C++ mode.

2019-07-16 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Tue Jul 16 07:57:32 2019 New Revision: 366212 URL: http://llvm.org/viewvc/llvm-project?rev=366212&view=rev Log: [OpenCL] Fixing sampler initialisations for C++ mode. Allow conversions between integer and sampler type. Differential Revision: https://reviews.llvm.org/D647

r289650 - Fixing cast condition for removing casts from builtin FPClassification.

2016-12-14 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Wed Dec 14 07:18:48 2016 New Revision: 289650 URL: http://llvm.org/viewvc/llvm-project?rev=289650&view=rev Log: Fixing cast condition for removing casts from builtin FPClassification. The function SemaBuiltinFPClassification removed superfluous float to double casts, th

r289552 - Fixing build failure by adding triple option to new test condition.

2016-12-13 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Tue Dec 13 11:04:33 2016 New Revision: 289552 URL: http://llvm.org/viewvc/llvm-project?rev=289552&view=rev Log: Fixing build failure by adding triple option to new test condition. Adding -triple option to ensure target supports double for fpmath test. Modified: cfe

RE: r286815 - Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64.

2016-12-13 Thread Neil Hickey via cfe-commits
exercise > >> double-rounding, especially near the overflow boundary? > >> > >> e.g. What is the expected value of x if the target does not support fp64?: > >> > >>float x = 340282356779733661637539395458142568447.0; > >> > >> –

r289544 - Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64.

2016-12-13 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Tue Dec 13 10:22:50 2016 New Revision: 289544 URL: http://llvm.org/viewvc/llvm-project?rev=289544&view=rev Log: Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64. This change makes sure single-precision floati

[PATCH] D24235: [OpenCL] Improve floating point literal handling

2016-11-15 Thread Neil Hickey via cfe-commits
neil.hickey retitled this revision from "[OpenCL] Improve double literal handling" to "[OpenCL] Improve floating point literal handling". neil.hickey updated the summary for this revision. neil.hickey updated this revision to Diff 77975. neil.hickey added a comment. Fixing indentation and improvi

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-11-15 Thread Neil Hickey via cfe-commits
neil.hickey removed rL LLVM as the repository for this revision. neil.hickey updated this revision to Diff 77961. neil.hickey added a comment. Fixes to tests and removal of incorrect check to stop float to float casts if types match. This was still needed as it was an lvalue to rvalue cast. Added

RE: r286815 - Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64.

2016-11-15 Thread Neil Hickey via cfe-commits
Can you add some non-trivial test cases that exercise double-rounding, > especially near the overflow boundary? > > e.g. What is the expected value of x if the target does not support fp64?: > > float x = 340282356779733661637539395458142568447.0; > > – Steve >

r286815 - Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64.

2016-11-14 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Mon Nov 14 05:15:51 2016 New Revision: 286815 URL: http://llvm.org/viewvc/llvm-project?rev=286815&view=rev Log: Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64. This change makes sure single-precision floati

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-11-14 Thread Neil Hickey via cfe-commits
This revision was automatically updated to reflect the committed changes. Closed by commit rL286815: Improve handling of floating point literals in OpenCL to only use double… (authored by neil.hickey). Changed prior to commit: https://reviews.llvm.org/D24235?vs=77600&id=77783#toc Repository:

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-11-11 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 77600. neil.hickey added a comment. Improving now confusing comment. https://reviews.llvm.org/D24235 Files: lib/Sema/SemaExpr.cpp lib/Sema/SemaType.cpp test/CodeGenOpenCL/fpmath.cl test/SemaOpenCL/extensions.cl Index: test/SemaOpenCL/extensions

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-11-09 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 77321. neil.hickey added a comment. Moving location of no-diagnostics statement to beginning of file https://reviews.llvm.org/D24235 Files: lib/Sema/SemaExpr.cpp lib/Sema/SemaType.cpp test/CodeGenOpenCL/fpmath.cl test/SemaOpenCL/extensions.cl I

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-11-01 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 76587. neil.hickey added a comment. Sorry for the delay. It looks like the code to handle extensions was changed since Neil Henning added the check against opencl 1.2. Perhaps the best approach is just to remove the check. https://reviews.llvm.org/D242

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-20 Thread Neil Hickey via cfe-commits
neil.hickey added inline comments. Comment at: lib/Sema/SemaExpr.cpp:3431 @@ -3410,1 +3430,3 @@ +.getSupportedOpenCLOpts() +.cl_khr_fp64) || getOpenCLOptions().cl_khr_fp64)) { yaxunl wrote: > This

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-19 Thread Neil Hickey via cfe-commits
neil.hickey added a reviewer: tstellarAMD. neil.hickey updated this revision to Diff 71837. neil.hickey added a comment. Herald added subscribers: yaxunl, wdng. There was a bug whereby an implicitcast was being applied from float to float, this caused issues later on in builin processing which ca

r281899 - Reverting r281714 due to causing an assert when calling builtins that expect a double, from CL

2016-09-19 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Mon Sep 19 06:42:14 2016 New Revision: 281899 URL: http://llvm.org/viewvc/llvm-project?rev=281899&view=rev Log: Reverting r281714 due to causing an assert when calling builtins that expect a double, from CL Modified: cfe/trunk/lib/Sema/SemaExpr.cpp cfe/trunk/lib

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-16 Thread Neil Hickey via cfe-commits
neil.hickey added a comment. committed @ 281714 https://reviews.llvm.org/D24235 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-16 Thread Neil Hickey via cfe-commits
neil.hickey closed this revision. neil.hickey added a comment. Commit merged to trunk https://reviews.llvm.org/D24235 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r281714 - Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64

2016-09-16 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Fri Sep 16 05:15:06 2016 New Revision: 281714 URL: http://llvm.org/viewvc/llvm-project?rev=281714&view=rev Log: Improve handling of floating point literals in OpenCL to only use double precision if the target supports fp64 https://reviews.llvm.org/D24235 Modified:

r281711 - Testing commit rights. Removing trailing white space from test file.

2016-09-16 Thread Neil Hickey via cfe-commits
Author: neil.hickey Date: Fri Sep 16 04:38:11 2016 New Revision: 281711 URL: http://llvm.org/viewvc/llvm-project?rev=281711&view=rev Log: Testing commit rights. Removing trailing white space from test file. Modified: cfe/trunk/test/SemaCXX/attr-noreturn.cpp Modified: cfe/trunk/test/SemaCXX/

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-09 Thread Neil Hickey via cfe-commits
neil.hickey added inline comments. Comment at: lib/Sema/SemaExpr.cpp:837 @@ +836,3 @@ + .getSupportedOpenCLOpts() + .cl_khr_fp64) || + getOpenCLOptions().cl_khr_fp64)) { Anastasia wrote: > Could we merge this and two lines abov

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-09 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 70799. neil.hickey added a comment. Modifying formating to match LLVM style. Adding new test to confirm that the vararg for printf is cast up to a double is allowable. https://reviews.llvm.org/D24235 Files: lib/Sema/SemaExpr.cpp lib/Sema/SemaType.c

Re: [PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-07 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 70527. neil.hickey added a comment. Added a CodeGen test as well as fixed an issue with vararg type promotion https://reviews.llvm.org/D24235 Files: lib/Sema/SemaExpr.cpp lib/Sema/SemaType.cpp test/CodeGenOpenCL/fpmath.cl test/SemaOpenCL/extensi

[PATCH] D24235: [OpenCL] Improve double literal handling

2016-09-05 Thread Neil Hickey via cfe-commits
neil.hickey created this revision. neil.hickey added a reviewer: Anastasia. neil.hickey added a subscriber: cfe-commits. Improve handling of floating point literals to only use double precision if it is supported. https://reviews.llvm.org/D24235 Files: lib/Sema/SemaExpr.cpp lib/Sema/SemaTyp

[PATCH] D15691: [OpenCL] Improving OpenCL function pointer error checking to catch lone function designator

2015-12-21 Thread Neil Hickey via cfe-commits
neil.hickey created this revision. neil.hickey added a reviewer: cfe-commits. An undecorated function designator implies taking the address of a function, which is illegal in OpenCL. Implementing a check for this earlier to allow the error to be reported even in the presence of other more obviou

Re: [PATCH] D13349: [OpenCL] Casting boolean to an integer vector in OpenCL should set all bits if boolean is true

2015-10-02 Thread Neil Hickey via cfe-commits
neil.hickey updated this revision to Diff 36348. neil.hickey added a comment. Fixing formatting problems. http://reviews.llvm.org/D13349 Files: lib/CodeGen/CGExprScalar.cpp test/CodeGenOpenCL/bool_cast.cl Index: test/CodeGenOpenCL/bool_cast.cl ==

[PATCH] D13349: Casting boolean to an integer vector in OpenCL should set all bits if boolean is true

2015-10-01 Thread Neil Hickey via cfe-commits
neil.hickey created this revision. neil.hickey added a reviewer: cfe-commits. Changing behaviour of casting a true boolean to an integer vector for OpenCL. The spec (6.2.2) states that if the boolean is true, every bit in the result vector should be set. This change will treat the i1 value as si

[PATCH] D13280: Adding reserved operator logical xor for OpenCL

2015-09-30 Thread Neil Hickey via cfe-commits
neil.hickey created this revision. neil.hickey added a reviewer: cfe-commits. This patch adds the reserved operator ^^ when compiling for OpenCL, which results in a more meaningful error message. http://reviews.llvm.org/D13280 Files: include/clang/Basic/DiagnosticParseKinds.td include/clang