C/C++ support for OpenACC init,set,shutdown directives. #225

Closed

galindos wants to merge 1 commit from galindos/gcc-fork-galindos:talkysafe/wip/init_set_shutdown/c_cpp into trunk

pull from: galindos/gcc-fork-galindos:talkysafe/wip/init_set_shutdown/c_cpp

merge into: gcc:trunk

gcc:trunk

gcc:releases/gcc-16

gcc:releases/gcc-15

gcc:releases/gcc-14

gcc:releases/gcc-13

gcc:releases/gcc-12

gcc:releases/gcc-11

gcc:releases/gcc-10

gcc:releases/gcc-9

gcc:releases/gcc-8

gcc:releases/gcc-7

gcc:releases/gcc-6

gcc:releases/gcc-5

gcc:releases/gcc-4.9

gcc:releases/gcc-4.8

gcc:releases/gcc-4.7

gcc:releases/gcc-4.6

gcc:releases/gcc-4.5

gcc:releases/gcc-4.4

gcc:releases/gcc-4.3

gcc:releases/gcc-3.0

gcc:releases/gcc-3.2

gcc:releases/gcc-3.3

gcc:releases/gcc-3.4

gcc:releases/gcc-4.2

gcc:releases/gcc-4.1

gcc:releases/gcc-4.0

gcc:releases/gcc-3.1

gcc:releases/gcc-2.95

gcc:releases/gcc-2.95.2.1-branch

gcc:releases/libgcj-2.95

gcc:releases/egcs-1.1

gcc:releases/egcs-1.0

Conversation 10 Commits 1 Files changed 25 +939 -26

galindos commented 2026-08-25 14:50:28 +00:00

Member

Copy link

This patch add OpenACC init,set,shutdown directives support for the C
and C++ front-ends.

An unexpected runtime error was fixed: When the device_type clause had
the argument as *, the set directive was passing an invalid argument to
the runtime library.

The bug 126507 was resolved (dtype alias for device_type).

gcc/c-family/ChangeLog:

* c-common.h (c_finish_oacc_executable_directive): Declare.
* c-omp.cc (oacc_eval_device_type): New.
(c_finish_oacc_executable_directive): New.  Lower init, set and
shutdown to GOACC builtins; map set's device_type(*) to
GOMP_DEVICE_DEFAULT.
* c-pragma.cc (oacc_pragmas): Register init, set and shutdown.
* c-pragma.h (enum pragma_kind): Add PRAGMA_OACC_INIT,
PRAGMA_OACC_SET and PRAGMA_OACC_SHUTDOWN.
(enum pragma_omp_clause): Add PRAGMA_OACC_CLAUSE_DEVICE_NUM and
PRAGMA_OACC_CLAUSE_DEVICE_TYPE alias.

gcc/c/ChangeLog:

* c-parser.cc (c_parser_omp_clause_name): Recognize device_num and
dtype as an alias of device_type.
(c_parser_oacc_single_int_clause): Allow device_num 0; clamp
non-positive device_num to 0.
(c_parser_oacc_clause_device_type): New.  Parse OpenACC
device_type/dtype arguments.
(c_parser_oacc_all_clauses): Handle DEVICE_NUM and OpenACC
DEVICE_TYPE.
(OACC_INIT_CLAUSE_MASK, OACC_SHUTDOWN_CLAUSE_MASK,
OACC_SET_CLAUSE_MASK): New.
(c_parser_oacc_executable_directive): New.
(c_parser_omp_construct): Handle init, set and shutdown.
* c-typeck.cc (c_finish_omp_clauses): Accept OMP_CLAUSE_DEVICE_NUM.

gcc/cp/ChangeLog:

* parser.cc (cp_parser_omp_clause_name): Recognize device_num and
dtype as an alias of device_type.
(cp_parser_oacc_single_int_clause): Allow device_num 0; clamp
non-positive device_num to 0.
(cp_parser_oacc_clause_device_type): New.  Parse OpenACC
device_type/dtype arguments.
(cp_parser_oacc_all_clauses): Handle DEVICE_NUM and OpenACC
DEVICE_TYPE.
(OACC_INIT_CLAUSE_MASK, OACC_SHUTDOWN_CLAUSE_MASK,
OACC_SET_CLAUSE_MASK): New.
(cp_parser_oacc_executable_directive): New.
(cp_parser_omp_construct): Handle init, set and shutdown.
(cp_parser_pragma): Allow init, set and shutdown in statement
context.
* semantics.cc (finish_omp_clauses): Handle OMP_CLAUSE_DEVICE_NUM
like other integral OpenACC clauses, including template
instantiation; require a non-negative value.

gcc/fortran/ChangeLog:

* openmp.cc (gfc_match_omp_clauses): Accept dtype as an alias of
OpenACC device_type.
* trans-openmp.cc (gfc_trans_oacc_executable_directive): For set,
map device_type(*) to GOMP_DEVICE_DEFAULT.

gcc/ChangeLog:

* tree-core.h (enum omp_clause_code): Add OMP_CLAUSE_DEVICE_NUM.
(enum oacc_clause_device_type_kind): New.
(struct tree_omp_clause): Add oacc_device_type_kind.
* tree-pretty-print.cc (dump_omp_clause): Dump DEVICE_NUM and
OpenACC DEVICE_TYPE kinds.
* tree.cc (omp_clause_num_ops, omp_clause_code_name): Handle
DEVICE_NUM.
* tree.h (OMP_CLAUSE_OACC_DEVICE_TYPE_KIND): Define.

gcc/testsuite/ChangeLog:

* c-c++-common/goacc/uninit-if-clause.c: Test uninitialized if on
init.
* c-c++-common/goacc/update-if_present-2.c: Expect errors for
if_present on init and shutdown.
* gfortran.dg/goacc/acc-set-1.f90: Test set with device_type(*).
* c-c++-common/goacc/acc-dtype-1.c: New test.
* c-c++-common/goacc/acc-init-1.c: New test.
* c-c++-common/goacc/acc-init-clauses-1.c: New test.
* c-c++-common/goacc/acc-set-1.c: New test.
* c-c++-common/goacc/acc-set-clauses-1.c: New test.
* c-c++-common/goacc/acc-shutdown-1.c: New test.
* c-c++-common/goacc/acc-shutdown-clauses-1.c: New test.
* gfortran.dg/goacc/acc-dtype-1.f90: New test.

Signed-off-by: Sebastian Galindo sebastian.galindo143@gmail.com
CC: tburnus@baylibre.com, tschwinge@baylibre.com

This patch add OpenACC init,set,shutdown directives support for the C and C++ front-ends. An unexpected runtime error was fixed: When the device_type clause had the argument as *, the set directive was passing an invalid argument to the runtime library. The bug 126507 was resolved (dtype alias for device_type). gcc/c-family/ChangeLog: * c-common.h (c_finish_oacc_executable_directive): Declare. * c-omp.cc (oacc_eval_device_type): New. (c_finish_oacc_executable_directive): New. Lower init, set and shutdown to GOACC builtins; map set's device_type(*) to GOMP_DEVICE_DEFAULT. * c-pragma.cc (oacc_pragmas): Register init, set and shutdown. * c-pragma.h (enum pragma_kind): Add PRAGMA_OACC_INIT, PRAGMA_OACC_SET and PRAGMA_OACC_SHUTDOWN. (enum pragma_omp_clause): Add PRAGMA_OACC_CLAUSE_DEVICE_NUM and PRAGMA_OACC_CLAUSE_DEVICE_TYPE alias. gcc/c/ChangeLog: * c-parser.cc (c_parser_omp_clause_name): Recognize device_num and dtype as an alias of device_type. (c_parser_oacc_single_int_clause): Allow device_num 0; clamp non-positive device_num to 0. (c_parser_oacc_clause_device_type): New. Parse OpenACC device_type/dtype arguments. (c_parser_oacc_all_clauses): Handle DEVICE_NUM and OpenACC DEVICE_TYPE. (OACC_INIT_CLAUSE_MASK, OACC_SHUTDOWN_CLAUSE_MASK, OACC_SET_CLAUSE_MASK): New. (c_parser_oacc_executable_directive): New. (c_parser_omp_construct): Handle init, set and shutdown. * c-typeck.cc (c_finish_omp_clauses): Accept OMP_CLAUSE_DEVICE_NUM. gcc/cp/ChangeLog: * parser.cc (cp_parser_omp_clause_name): Recognize device_num and dtype as an alias of device_type. (cp_parser_oacc_single_int_clause): Allow device_num 0; clamp non-positive device_num to 0. (cp_parser_oacc_clause_device_type): New. Parse OpenACC device_type/dtype arguments. (cp_parser_oacc_all_clauses): Handle DEVICE_NUM and OpenACC DEVICE_TYPE. (OACC_INIT_CLAUSE_MASK, OACC_SHUTDOWN_CLAUSE_MASK, OACC_SET_CLAUSE_MASK): New. (cp_parser_oacc_executable_directive): New. (cp_parser_omp_construct): Handle init, set and shutdown. (cp_parser_pragma): Allow init, set and shutdown in statement context. * semantics.cc (finish_omp_clauses): Handle OMP_CLAUSE_DEVICE_NUM like other integral OpenACC clauses, including template instantiation; require a non-negative value. gcc/fortran/ChangeLog: * openmp.cc (gfc_match_omp_clauses): Accept dtype as an alias of OpenACC device_type. * trans-openmp.cc (gfc_trans_oacc_executable_directive): For set, map device_type(*) to GOMP_DEVICE_DEFAULT. gcc/ChangeLog: * tree-core.h (enum omp_clause_code): Add OMP_CLAUSE_DEVICE_NUM. (enum oacc_clause_device_type_kind): New. (struct tree_omp_clause): Add oacc_device_type_kind. * tree-pretty-print.cc (dump_omp_clause): Dump DEVICE_NUM and OpenACC DEVICE_TYPE kinds. * tree.cc (omp_clause_num_ops, omp_clause_code_name): Handle DEVICE_NUM. * tree.h (OMP_CLAUSE_OACC_DEVICE_TYPE_KIND): Define. gcc/testsuite/ChangeLog: * c-c++-common/goacc/uninit-if-clause.c: Test uninitialized if on init. * c-c++-common/goacc/update-if_present-2.c: Expect errors for if_present on init and shutdown. * gfortran.dg/goacc/acc-set-1.f90: Test set with device_type(*). * c-c++-common/goacc/acc-dtype-1.c: New test. * c-c++-common/goacc/acc-init-1.c: New test. * c-c++-common/goacc/acc-init-clauses-1.c: New test. * c-c++-common/goacc/acc-set-1.c: New test. * c-c++-common/goacc/acc-set-clauses-1.c: New test. * c-c++-common/goacc/acc-shutdown-1.c: New test. * c-c++-common/goacc/acc-shutdown-clauses-1.c: New test. * gfortran.dg/goacc/acc-dtype-1.f90: New test. Signed-off-by: Sebastian Galindo <sebastian.galindo143@gmail.com> CC: tburnus@baylibre.com, tschwinge@baylibre.com

forge-bot commented 2026-08-25 14:50:32 +00:00

Member

Copy link

Hi @galindos. This bot helps send your PR as a patch series to the mailing list.

Next steps

1. Write a clear PR description. It will be used as the cover letter.
2. To CC extra people, add lines like this at the end of the PR description: CC: Reviewer1 <revi.ewer@example.com>, Reviewer2 <an.other@example.com>.
3. To preview the emails, comment with /preview.
4. When ready, request review in the Reviewers section. Or, if you just want to send the series, comment with /submit.
5. Use the cover-letter links below to follow replies.
6. For a new patch series, push your changes, update the PR description with a Changes since vX section, and comment with /submit again.

See Submitting Patches and UsingForge.

Consider joining the gcc and gcc-patches mailing lists..
For real time communication, check the gcc irc channels.

Or join #overseers on Libera Chat, particularly if this automation is not working (stay online to get replies, IRC does not save messages if people are not online).

<!-- pr-welcome --> Hi @galindos. This bot helps send your PR as a patch series to the mailing list. ## Next steps 1. Write a clear PR description. It will be used as the cover letter. 2. To CC extra people, add lines like this **at the end** of the PR description: `CC: Reviewer1 <revi.ewer@example.com>, Reviewer2 <an.other@example.com>`. 3. To preview the emails, comment with `/preview`. 4. When ready, request review in the Reviewers section. Or, if you just want to send the series, comment with `/submit`. 5. Use the cover-letter links below to follow replies. 6. For a new patch series, push your changes, update the PR description with a `Changes since vX` section, and comment with `/submit` again. See [Submitting Patches](https://gcc.gnu.org/contribute.html#patches) and [UsingForge](https://gcc.gnu.org/wiki/UsingForge). Consider joining the [gcc and gcc-patches mailing lists.](https://gcc.gnu.org/lists.html). For real time communication, check the [gcc irc channels](https://gcc.gnu.org/wiki/GCConIRC). Or join [`#overseers`](https://web.libera.chat/#overseers) on Libera Chat, particularly if this automation is not working (stay online to get replies, IRC does not save messages if people are not online).

galindos commented 2026-08-25 14:58:08 +00:00

Author

Member

Copy link

/preview

/preview

forge-bot commented 2026-08-25 14:58:20 +00:00

Member

Copy link

Sent patch version 1 as a single email to @galindos's email address.
Message ID is <bmm.hloc7kwmka.gcc.gcc.galindos.225.1.0@forge-stage.sourceware.org>.

Sent patch version 1 as a single email to @galindos's email address. Message ID is `<bmm.hloc7kwmka.gcc.gcc.galindos.225.1.0@forge-stage.sourceware.org>`.

galindos force-pushed talkysafe/wip/init_set_shutdown/c_cpp from d5b0ae1ee1 to 3128ae9fbb 2026-08-27 03:04:45 +00:00 Compare

galindos force-pushed talkysafe/wip/init_set_shutdown/c_cpp from 3128ae9fbb to 4d1ae1abe7 2026-08-27 03:07:56 +00:00 Compare

galindos commented 2026-08-27 03:10:17 +00:00

Author

Member

Copy link

/preview

/preview

forge-bot commented 2026-08-27 03:10:29 +00:00

Member

Copy link

Sent patch version 1 as a single email to @galindos's email address.
Message ID is <bmm.hlq02zt3o8.gcc.gcc.galindos.225.1.0@forge-stage.sourceware.org>.

Sent patch version 1 as a single email to @galindos's email address. Message ID is `<bmm.hlq02zt3o8.gcc.gcc.galindos.225.1.0@forge-stage.sourceware.org>`.

galindos commented 2026-08-27 03:11:30 +00:00

Author

Member

Copy link

/submit

/submit

forge-bot commented 2026-08-27 03:12:09 +00:00

Member

Copy link

Version 1 of this pull request has been stored. It includes the following commits:

• C/C++ support for OpenACC init,set,shutdown directives. - 4d1ae1abe7

<!-- pr-new-version --> Version 1 of this pull request has been stored. It includes the following commits: - C/C++ support for OpenACC init,set,shutdown directives. - 4d1ae1abe734ff7cd607af9a1d5b4141b7b906f4

forge-bot commented 2026-08-27 03:12:09 +00:00

Member

Copy link

Pull Request versions:

#	Base	Head
1	b085ae4b49	4d1ae1abe7	(diff)

In order to compare , clone this repository and run

PR=225
git fetch origin "refs/versioned_pull/${PR}/*:refs/versioned_pull/${PR}/*"
PRV1=1
PRV2=2
git range-diff "refs/versioned_pull/${PR}/${PRV1}/base..refs/versioned_pull/${PR}/${PRV1}/head"  "refs/versioned_pull/${PR}/${PRV2}/base..refs/versioned_pull/${PR}/${PRV2}/head"

<!-- pr-versions --> Pull Request versions: | # | Base | Head | | | - | ---- | ---- | - | | 1 | b085ae4b49b41e146c0a6a6c16c8c588438b9528 | 4d1ae1abe734ff7cd607af9a1d5b4141b7b906f4 | [(diff)](https://forge.sourceware.org/gcc/gcc.git/compare/b085ae4b49b41e146c0a6a6c16c8c588438b9528...4d1ae1abe734ff7cd607af9a1d5b4141b7b906f4) | In order to compare , clone this repository and run ``` PR=225 git fetch origin "refs/versioned_pull/${PR}/*:refs/versioned_pull/${PR}/*" PRV1=1 PRV2=2 git range-diff "refs/versioned_pull/${PR}/${PRV1}/base..refs/versioned_pull/${PR}/${PRV1}/head" "refs/versioned_pull/${PR}/${PRV2}/base..refs/versioned_pull/${PR}/${PRV2}/head" ```

forge-bot commented 2026-08-27 03:12:42 +00:00

Member

Copy link

Sent patch version 1 as a single email to gcc-patches mailing list gcc-patches@gcc.gnu.org and cc'd tburnus@baylibre.com, tschwinge@baylibre.com.
Cover letter

Sent patch version 1 as a single email to gcc-patches mailing list <gcc-patches@gcc.gnu.org> and cc'd tburnus@baylibre.com, tschwinge@baylibre.com. [Cover letter](https://inbox.sourceware.org/gcc-patches/bmm.hlq046v70c.gcc.gcc.galindos.225.1.0@forge-stage.sourceware.org)

tschwinge commented 2026-09-20 13:50:45 +00:00

Owner

Copy link

An unexpected runtime error was fixed: When the device_type clause had the argument as *, the set directive was passing an invalid argument to the runtime library.

Please make that a separate patch submission (including the corresponding test case updates).

The bug 126507 was resolved (dtype alias for device_type).

..., and that one as well, separately. Put [PR126507] at the end of the Git commit summary line, and PR fortran/126507 in the ChangeLog updates as part of the Git commit log.

(I'm now reviewing the C, C++ changes here.)

> An unexpected runtime error was fixed: When the device_type clause had the argument as *, the set directive was passing an invalid argument to the runtime library. Please make that a separate patch submission (including the corresponding test case updates). > The bug 126507 was resolved (dtype alias for device_type). ..., and that one as well, separately. Put `[PR126507]` at the end of the Git commit summary line, and `PR fortran/126507` in the ChangeLog updates as part of the Git commit log. (I'm now reviewing the C, C++ changes here.)

galindos closed this pull request 2026-10-05 03:03:57 +00:00

All checks were successful

Hide checks

/ format-checks (pull_request) Successful in 1m28s

Details

Pull request closed

Please reopen this pull request to perform a merge.

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

Compat/Breaking

Breaking change that won't be backward compatible

  

Frontend/ada

  

Frontend/c

  

Frontend/c++

  

Frontend/rust

Code affecting the Rust front end

  

General/forge

Forgejo infrastructure

  

Kind/Bug

Something is not working

  

Kind/Documentation

Documentation changes

  

Kind/Enhancement

Improve existing functionality

  

Kind/Feature

New functionality

  

Kind/Security

This is security issue

  

Kind/Testing

Issue or pull request related to testing

  

Library/libgcc

  

Library/libstdc++

  

Midend/gimple

  

Midend/rtl

  

Midend/tree

  

Priority

Critical

The priority is critical

  

Priority

High

The priority is high

  

Priority

Low

The priority is low

  

Priority

Medium

The priority is medium

  

Reviewed

Confirmed

Issue has been confirmed

  

Reviewed

Duplicate

This issue or pull request already exists

  

Reviewed

Invalid

Invalid issue

  

Reviewed

Won't Fix

This issue won't be fixed

  

Status

Abandoned

Somebody has started to work on this but abandoned work

  

Status

Blocked

Something is blocking this issue or pull request

  

Status

Need More Info

Feedback is required to reproduce issue or to continue work

  

Target/aarch64

  

Target/arm

  

Target/i386

  

bug

Something is not working

  

duplicate

This issue or pull request already exists

  

enhancement

New feature

  

help wanted

Need some help

  

invalid

Something is wrong

  

question

More information is needed

  

wontfix

This won't be fixed

No labels

Compat/Breaking

Frontend/ada

Frontend/c

Frontend/c++

Frontend/rust

General/forge

Kind/Bug

Kind/Documentation

Kind/Enhancement

Kind/Feature

Kind/Security

Kind/Testing

Library/libgcc

Library/libstdc++

Midend/gimple

Midend/rtl

Midend/tree

Priority

Critical

Priority

High

Priority

Low

Priority

Medium

Reviewed

Confirmed

Reviewed

Duplicate

Reviewed

Invalid

Reviewed

Won't Fix

Status

Abandoned

Status

Blocked

Status

Need More Info

Target/aarch64

Target/arm

Target/i386

bug

duplicate

enhancement

help wanted

invalid

question

wontfix

Milestone

Clear milestone

No items

No milestone

Assignees

Clear assignees

No assignees

3 participants

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

gcc/gcc!225

No description provided.