Author: Krzysztof Parzyszek
Date: 2025-05-28T08:13:31-05:00
New Revision: 11e804fcabce11a9fdfb4033263ef1e502cc8a72
URL:
https://github.com/llvm/llvm-project/commit/11e804fcabce11a9fdfb4033263ef1e502cc8a72
DIFF:
https://github.com/llvm/llvm-project/commit/11e804fcabce11a9fdfb4033263ef1e502cc8a72
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/139961
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/139961
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/139793
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/139793
The "workshare" construct is only present in Fortran. The common OpenMP code
does treat it as any other directive, but in clang we need to reject it, and do
so gracefully before it encounters an internal asser
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/139131
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/139131
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/139115
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/139115
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/138179
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
kparzysz wrote:
@alexey-bataev Are you ok with the change in SemaOpenMP?
https://github.com/llvm/llvm-project/pull/138179
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/138179
>From d991e966f12229fc4a9cc1114e3fa4651bc9d59d Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Thu, 1 May 2025 11:42:12 -0500
Subject: [PATCH 1/2] [OpenMP] Remove "alternativeName" from Clause and
Dir
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/138179
The "alternativeName" was introduced ~5 years ago in D82405, and at the moment
it has only one use, the one that the original change introduced.
OpenMP 6.0 spec has introduced different spellings of some direc
@@ -2759,6 +2759,19 @@ StmtResult
Parser::ParseOpenMPDeclarativeOrExecutableDirective(
OpenMPClauseKind CKind = Tok.isAnnotation()
? OMPC_unknown
: getOpenMPClauseKind(PP.getSpelling(Tok));
+ // C
@@ -236,6 +236,8 @@ enum class OpenMPOffloadMappingFlags : uint64_t {
// dynamic.
// This is an OpenMP extension for the sake of OpenACC support.
OMP_MAP_OMPX_HOLD = 0x2000,
+ /// Self directs mapping without creating a separate device copy.
+ OMP_MAP_SELF = 0x4000,
---
@@ -1657,6 +1657,10 @@ def err_omp_expected_colon : Error<"missing ':' in %0">;
def err_omp_missing_comma : Error< "missing ',' after %0">;
def err_omp_expected_context_selector
: Error<"expected valid context selector in %0">;
+def err_omp_unknown_clause
+: Error<"unkn
kparzysz wrote:
> In general, we aim to only depend on libClangDriver and move it to LLVM so
> that Flang need not depend on Clang or its libraries. If this patch increases
> the dependency on Clang it might not be acceptable.
Agreed. 👍
@JDPailleux: It may a good idea to extract the diagnost
https://github.com/kparzysz approved this pull request.
Thanks!
https://github.com/llvm/llvm-project/pull/101701
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz approved this pull request.
LGTM, but please address Alexey's comments before merging.
https://github.com/llvm/llvm-project/pull/119891
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/m
kparzysz wrote:
> All that said, there are two cases to consider wrt. the standard:
>
> 1. The initial device is the CPU and the code compiled here is just part of a
> GPU library, or
> 2. the initial device is the GPU and the code compiled here is just part of
> the "host code".
>
> For 1),
@@ -343,6 +343,124 @@ implementation.
| task | nowait clause on taskwait
| :part:`partial` | parsing/sema done: D131830, D141531
|
+--+---
kparzysz wrote:
To me this looks like compilation for a host, except the GPU is the host. The
only functions that could be called from such a CU would be the top-level ones,
not any of the auto-generated one.
Additionally, the host wouldn't support offload, so we'd need to do something
about
@@ -2158,6 +2158,10 @@ enum CXCursorKind {
*/
CXCursor_OMPAssumeDirective = 309,
+ /** OpenMP assume directive.
+ */
+ CXCursor_OMPStripeDirective = 310,
kparzysz wrote:
Please add this to `clang/bindings/python/clang/cindex.py` as well.
https://git
kparzysz wrote:
Is NONE special enough to have a dedicated option, or could we have
`-fopenmp-default=` for any dsa?
https://github.com/llvm/llvm-project/pull/120287
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bi
kparzysz wrote:
Here's another thing---could there be tools that try to parse the messages
(e.g. something that runs clang and presents the messages to the user in some
form)? Having a policy such as "single quotes only come in pairs" could make
it easier. I don't know if that's something we
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/116623
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
kparzysz wrote:
I (visually) looked at all occurrences of `can\>` in the td files, and I didn't
see any occurring at the end of a line, or in a place where a `not` could be
pasted into the message following the `can`. It was a "best effort"
inspection, but there weren't too many of these, so
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/116623
See
https://discourse.llvm.org/t/cant-cannot-can-not-in-diagnostic-messages/83171
>From 489d5a0c1f9cb53e87cbbf30e8ddec83820e0326 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 18 Nov 2024 07:51
kparzysz wrote:
I was able to see the internal bug report. This shows up when some code on
Windows runs clang (via a library call), and it ends up crashing due to stack
overflow. The crash happens in this function, in "__chkstk" to be exact.
https://github.com/llvm/llvm-project/pull/114285
_
kparzysz wrote:
> Don't see how or why this would introduce stackusage. std::string temporaries
> should be avoided whenever possible
The code that's included afterwards contains a lot of comparisons with string
literals:
```
if (Name == "aarch64_sve_pcs" && getSyntax() == AttributeCommonInfo:
kparzysz wrote:
Thanks! I was getting these warning too.
https://github.com/llvm/llvm-project/pull/112942
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz approved this pull request.
https://github.com/llvm/llvm-project/pull/112942
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
kparzysz wrote:
> Right, use `git clang-format`, that will take care of formatting the file
> over time without creating churn.
I think the number of changes is relatively small for a file this size. The
churn here is comparable to that created by a "normal" patch...
https://github.com/llvm/
kparzysz wrote:
> Yeah I was actually wondering if I should go for something like this:
>
> ```c
> #ifdef __NVPTX__
> uint32_t nvptx_get_thread_id_x() { return __nvvm_ptx_read_sreg_tid_x(); }
> #define IMPL nvptx
> #endif
> uint32_t gpu_get_thread_id_x() { return ##IMPL##_get_thread_id_x(); }
>
kparzysz wrote:
Right now it's about 400 check-clang failures for all changes together. Adding
AST nodes for directives in templates causes 12 failures (mostly in slightly
differing diagnostics), adding capturing regions for all constituents will have
its own share, but the majority comes fro
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/109003
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2861,113 +2861,120 @@ void SemaOpenMP::EndOpenMPDSABlock(Stmt
*CurDirective) {
// clause requires an accessible, unambiguous default constructor for the
// class type, unless the list item is also specified in a firstprivate
// clause.
- if (const auto *D = dyn_c
@@ -2861,113 +2861,120 @@ void SemaOpenMP::EndOpenMPDSABlock(Stmt
*CurDirective) {
// clause requires an accessible, unambiguous default constructor for the
// class type, unless the list item is also specified in a firstprivate
// clause.
- if (const auto *D = dyn_c
@@ -2861,113 +2861,120 @@ void SemaOpenMP::EndOpenMPDSABlock(Stmt
*CurDirective) {
// clause requires an accessible, unambiguous default constructor for the
// class type, unless the list item is also specified in a firstprivate
// clause.
- if (const auto *D = dyn_c
@@ -2861,113 +2861,120 @@ void SemaOpenMP::EndOpenMPDSABlock(Stmt
*CurDirective) {
// clause requires an accessible, unambiguous default constructor for the
// class type, unless the list item is also specified in a firstprivate
// clause.
- if (const auto *D = dyn_c
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/109003
>From d9bb31da5c897aad0dbc55781df2341db75aad6d Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 12 Aug 2024 08:46:12 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Slightly refactor EndOpenMPDSABlock f
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/109003
Change the loop
```
if (isOpenMPExecutableDirective)
for (Clause)
if (Clause is kind1)
multi
line
do
something1;
else if (Clause is kind2)
...
...
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/106786
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -6060,69 +6057,55 @@ StmtResult SemaOpenMP::ActOnOpenMPExecutableDirective(
return StmtError();
// Generate list of implicitly defined firstprivate variables.
VarsWithInheritedDSA = DSAChecker.getVarsWithInheritedDSA();
+VariableImplicitInfo ImpInfo = DSACh
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/106786
>From 92e82b249836b4f17d1e13a927ca3e8300f671ee Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 30 Aug 2024 13:57:13 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Simplify handling of implicit DSA/map
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/106787
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/106786
>From 92e82b249836b4f17d1e13a927ca3e8300f671ee Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 30 Aug 2024 13:57:13 -0500
Subject: [PATCH] [clang][OpenMP] Simplify handling of implicit DSA/mapping
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/106787
>From 053374ee88af47515a54e5c5ba3b680e252862bf Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Tue, 30 Jul 2024 15:02:38 -0500
Subject: [PATCH] [clang][OpenMP] `masked` and `master` are not capturing
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/106786
Create `VariableImplicitInfo` to hold the data. Most of it is used all at once,
so aggregating it seems like a good idea.
>From 92e82b249836b4f17d1e13a927ca3e8300f671ee Mon Sep 17 00:00:00 2001
From: Krzysztof
Author: Krzysztof Parzyszek
Date: 2024-08-20T10:07:19-05:00
New Revision: 1c3955f9dff9b653be88d5dea5bc3cd007904e90
URL:
https://github.com/llvm/llvm-project/commit/1c3955f9dff9b653be88d5dea5bc3cd007904e90
DIFF:
https://github.com/llvm/llvm-project/commit/1c3955f9dff9b653be88d5dea5bc3cd007904e90
https://github.com/kparzysz approved this pull request.
LGTM. We will complain when someone tries to capture a structure binding
anyway.
https://github.com/llvm/llvm-project/pull/104822
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://
kparzysz wrote:
> I think you're right for this particular test case, but I don't think that
> solution works in general. Consider something like:
>
> ```
> int arr1[N][N], arr2[N];
>
> #pragma omp parallel for collapse(2)
> {
> for (int i = 0; i < N; i++) {
> arr2[i] = i;
> for (int
Author: Krzysztof Parzyszek
Date: 2024-08-18T10:48:49-05:00
New Revision: de5ea2d122c31e1551654ff506c33df299f351b8
URL:
https://github.com/llvm/llvm-project/commit/de5ea2d122c31e1551654ff506c33df299f351b8
DIFF:
https://github.com/llvm/llvm-project/commit/de5ea2d122c31e1551654ff506c33df299f351b8
Author: Krzysztof Parzyszek
Date: 2024-08-18T10:25:52-05:00
New Revision: e05307f6633ca405834a4fd24d858ffb676c9170
URL:
https://github.com/llvm/llvm-project/commit/e05307f6633ca405834a4fd24d858ffb676c9170
DIFF:
https://github.com/llvm/llvm-project/commit/e05307f6633ca405834a4fd24d858ffb676c9170
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/101058
>From e6286e542ede1dfe5b63ce3b8d86a78fc8c8c895 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 29 Jul 2024 13:29:09 -0500
Subject: [PATCH 1/3] [clang][OpenMP] Rename `varlists` to `varlist`, NFC
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/101058
>From e6286e542ede1dfe5b63ce3b8d86a78fc8c8c895 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 29 Jul 2024 13:29:09 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Rename `varlists` to `varlist`, NFC
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/101058
It returns a range of variables (via Expr*), not a range of lists.
>From e6286e542ede1dfe5b63ce3b8d86a78fc8c8c895 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 29 Jul 2024 13:29:09 -0500
Subjec
kparzysz wrote:
Linux failure went away, but a Windows issue was introduced upstream in the
meantime. None of that has to do with this PR.
https://github.com/llvm/llvm-project/pull/100162
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https:
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/100162
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/100162
>From 8e99d0743c69f3e5981deb24f501ea993567522e Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Tue, 23 Jul 2024 09:56:53 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Mark all SIMD regions as non-throwing
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/100162
>From 8e99d0743c69f3e5981deb24f501ea993567522e Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Tue, 23 Jul 2024 09:56:53 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Mark all SIMD regions as non-throwing
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/100162
[4.5:75:19], [5.0:114:3], [5.1:137:21], [5.2:235:30]
"No exception can be raised in the **simd** region."
>From 8e99d0743c69f3e5981deb24f501ea993567522e Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date:
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/99905
>From 4657c66f04cf5b603afcc356468e61f8805cb264 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Thu, 18 Jul 2024 15:21:10 -0500
Subject: [PATCH 1/4] [clang][OpenMP] Move "loop" directive mapping from sem
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/99905
>From 4657c66f04cf5b603afcc356468e61f8805cb264 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Thu, 18 Jul 2024 15:21:10 -0500
Subject: [PATCH 1/3] [clang][OpenMP] Move "loop" directive mapping from sem
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/99905
>From 4657c66f04cf5b603afcc356468e61f8805cb264 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Thu, 18 Jul 2024 15:21:10 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Move "loop" directive mapping from sem
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/99905
Given "loop" construct, clang will try to treat it as "for", "distribute" or
"simd", depending on either the implied binding, or the bind clause if present.
This patch moves the code that performs this construc
Author: Krzysztof Parzyszek
Date: 2024-07-22T10:22:20-05:00
New Revision: e9709899db7d4a8b1c36475e90e4a934335f3d95
URL:
https://github.com/llvm/llvm-project/commit/e9709899db7d4a8b1c36475e90e4a934335f3d95
DIFF:
https://github.com/llvm/llvm-project/commit/e9709899db7d4a8b1c36475e90e4a934335f3d95
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/98500
>From bc7e6962de9e1e44773eee89d0f552342b40 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Wed, 10 Jul 2024 14:38:57 -0500
Subject: [PATCH 1/4] [clang][OpenMP] Fix region nesting check for `scan`
d
Author: Krzysztof Parzyszek
Date: 2024-07-15T09:10:11-05:00
New Revision: 97ebc9794941d9e73792ab9deab7abafaf750a17
URL:
https://github.com/llvm/llvm-project/commit/97ebc9794941d9e73792ab9deab7abafaf750a17
DIFF:
https://github.com/llvm/llvm-project/commit/97ebc9794941d9e73792ab9deab7abafaf750a17
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/98633
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/98633
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/98500
>From bc7e6962de9e1e44773eee89d0f552342b40 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Wed, 10 Jul 2024 14:38:57 -0500
Subject: [PATCH 1/4] [clang][OpenMP] Fix region nesting check for `scan`
d
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/98500
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/98387
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/98387
>From bc7e6962de9e1e44773eee89d0f552342b40 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Wed, 10 Jul 2024 14:38:57 -0500
Subject: [PATCH 1/3] [clang][OpenMP] Fix region nesting check for `scan`
d
Author: Krzysztof Parzyszek
Date: 2024-07-12T08:18:28-05:00
New Revision: ba29e3a58dcdf0012df9fdaa3e7ed10cee56d5c5
URL:
https://github.com/llvm/llvm-project/commit/ba29e3a58dcdf0012df9fdaa3e7ed10cee56d5c5
DIFF:
https://github.com/llvm/llvm-project/commit/ba29e3a58dcdf0012df9fdaa3e7ed10cee56d5c5
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/98387
>From bc7e6962de9e1e44773eee89d0f552342b40 Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Wed, 10 Jul 2024 14:38:57 -0500
Subject: [PATCH 1/3] [clang][OpenMP] Fix region nesting check for `scan`
d
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/98387
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/98386
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/98386
The previous check was inconsistent. For example, it would allow
```
#pragma omp target
#pragma omp parallel for
for (...) {
#pragma omp scan
}
```
but not
```
#pragma omp target parallel for
for (...) {
#p
@@ -760,6 +760,9 @@ void clang::getOpenMPCaptureRegions(
case OMPD_parallel:
CaptureRegions.push_back(OMPD_parallel);
break;
+case OMPD_assume:
+ CaptureRegions.push_back(OMPD_unknown);
+ break;
kparzysz wrote:
The final output fr
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/97535
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -760,6 +760,9 @@ void clang::getOpenMPCaptureRegions(
case OMPD_parallel:
CaptureRegions.push_back(OMPD_parallel);
break;
+case OMPD_assume:
+ CaptureRegions.push_back(OMPD_unknown);
+ break;
kparzysz wrote:
This is what happen
@@ -760,6 +760,9 @@ void clang::getOpenMPCaptureRegions(
case OMPD_parallel:
CaptureRegions.push_back(OMPD_parallel);
break;
+case OMPD_assume:
+ CaptureRegions.push_back(OMPD_unknown);
+ break;
kparzysz wrote:
Also, please add OM
@@ -760,6 +760,9 @@ void clang::getOpenMPCaptureRegions(
case OMPD_parallel:
CaptureRegions.push_back(OMPD_parallel);
break;
+case OMPD_assume:
+ CaptureRegions.push_back(OMPD_unknown);
+ break;
kparzysz wrote:
Please don't push O
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/98180
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/98180
Rewrite `getOpenMPCaptureRegionForClause` with more generic code, relying on
directive composition instead of listing individual directives.
>From e137ccb9dcb74450f9615465e03db74c0a24040c Mon Sep 17 00:00:00 20
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/97445
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/97446
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97446
>From 316d5a11bd7f6afb99cd7d25baf43ec2679561dc Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 1 Jul 2024 15:18:07 -0500
Subject: [PATCH 1/3] [clang][OpenMP] Use leaf constructs in `mapLoopConstruc
@@ -6270,16 +6270,20 @@ bool SemaOpenMP::mapLoopConstruct(
if (BindKind == OMPC_BIND_unknown) {
// Setting the enclosing teams or parallel construct for the loop
// directive without bind clause.
+ // [5.0:129:25-28] If the bind clause is not present on the
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97446
>From 316d5a11bd7f6afb99cd7d25baf43ec2679561dc Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Mon, 1 Jul 2024 15:18:07 -0500
Subject: [PATCH 1/2] [clang][OpenMP] Use leaf constructs in `mapLoopConstruc
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/97446
This removes mentions of specific combined directives.
Also, add a quote from the OpenMP spec to explain the code dealing with the
`bind` clause.
>From 316d5a11bd7f6afb99cd7d25baf43ec2679561dc Mon Sep 17 00:00
https://github.com/kparzysz created
https://github.com/llvm/llvm-project/pull/97445
Instead of checking specific directives, this function now gets the list of
captured regions, and processes them individually. This makes this function
directive-agnostic (except a few cases of leaf constructs)
https://github.com/kparzysz closed
https://github.com/llvm/llvm-project/pull/97110
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97110
>From 789ec614d285e3fe632aae6280f6c4cf01746cdf Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 28 Jun 2024 12:27:10 -0500
Subject: [PATCH 01/11] [clang][OpenMP] Implement `isOpenMPExecutableDirecti
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97110
>From 789ec614d285e3fe632aae6280f6c4cf01746cdf Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 28 Jun 2024 12:27:10 -0500
Subject: [PATCH 01/10] [clang][OpenMP] Implement `isOpenMPExecutableDirecti
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97110
>From 789ec614d285e3fe632aae6280f6c4cf01746cdf Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 28 Jun 2024 12:27:10 -0500
Subject: [PATCH 1/9] [clang][OpenMP] Implement `isOpenMPExecutableDirective
https://github.com/kparzysz updated
https://github.com/llvm/llvm-project/pull/97110
>From 789ec614d285e3fe632aae6280f6c4cf01746cdf Mon Sep 17 00:00:00 2001
From: Krzysztof Parzyszek
Date: Fri, 28 Jun 2024 12:27:10 -0500
Subject: [PATCH 1/8] [clang][OpenMP] Implement `isOpenMPExecutableDirective
https://github.com/kparzysz edited
https://github.com/llvm/llvm-project/pull/97110
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
1 - 100 of 401 matches
Mail list logo
