From: Thomas Schwinge <thomas@codesourcery.com>
To: Tobias Burnus <tobias@codesourcery.com>, <gcc-patches@gcc.gnu.org>
Cc: Jakub Jelinek <jakub@redhat.com>, <fortran@gcc.gnu.org>
Subject: Re: [Patch, Fortran] OpenMP/OpenACC – fix more issues with OPTIONAL
Date: Thu, 18 Jun 2020 00:22:20 +0200 [thread overview]
Message-ID: <878sgl2uzn.fsf@euler.schwinge.homeip.net> (raw)
In-Reply-To: <85ca99af-f009-c843-59e0-54cef7906c18@codesourcery.com>
[-- Attachment #1: Type: text/plain, Size: 2511 bytes --]
Hi!
On 2020-04-29T16:01:56+0200, Tobias Burnus <tobias@codesourcery.com> wrote:
> See: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94848
>
> The problem is a generated static array variable in the
> device function:
> static integer(kind=4) A.12[3] = {1, 2, 3};
> used as
> _26 = A.12[S.13_67];
>
> [...]
..., and that PR was recently fixed, and so...
> On 4/29/20 12:00 PM, Thomas Schwinge wrote:
>> On 2020-01-08T09:55:06+0100, Tobias Burnus <tobias@codesourcery.com> wrote:
>>> On 1/8/20 9:33 AM, Thomas Schwinge wrote:
>>>> With 'dg-do run' added, on [...] x86_64-pc-linux-gnu with offloading I'm seeing:
>> | PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 (test for excess errors)
>> | PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 execution test
>> | PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 (test for excess errors)
>> | PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 execution test
>> | FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 (test for excess errors)
>> | UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 compilation failed to produce executable
>> | FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions (test for excess errors)
>> | UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions compilation failed to produce executable
>> | FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g (test for excess errors)
>> | UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g compilation failed to produce executable
>> | FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -Os (test for excess errors)
>> | UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -Os compilation failed to produce executable
... that problem now is gone, too. I've thus pushed "Add 'dg-do run' to
'libgomp.fortran/use_device_ptr-optional-3.f90' [PR94848]" to master
branch in commit 5864930754f63e2dcef9606f2514ae20e80f436e, and
releases/gcc-10 branch in commit
61c896d84bdefbfffa7573a8af89119d4db7b3de, see attached.
Grüße
Thomas
-----------------
Mentor Graphics (Deutschland) GmbH, Arnulfstraße 201, 80634 München / Germany
Registergericht München HRB 106955, Geschäftsführer: Thomas Heurung, Alexander Walter
[-- Attachment #2: 0001-Add-dg-do-run-to-libgomp.fortran-use_device_ptr-opti.patch --]
[-- Type: text/x-diff, Size: 2721 bytes --]
From 5864930754f63e2dcef9606f2514ae20e80f436e Mon Sep 17 00:00:00 2001
From: Thomas Schwinge <thomas@codesourcery.com>
Date: Tue, 7 Jan 2020 17:40:14 +0100
Subject: [PATCH] Add 'dg-do run' to
'libgomp.fortran/use_device_ptr-optional-3.f90' [PR94848]
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Fix-up for r279858/commit f760c0c77fe350616da9dbeaea16442b0acfb09c "Fortran]
OpenMP/OpenACC – fix more issues with OPTIONAL".
With offloading enabled, we then saw:
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 (test for excess errors)
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 execution test
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 (test for excess errors)
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 execution test
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -Os (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -Os compilation failed to produce executable
... due to:
/tmp/cciVc43I.o:(.gnu.offload_vars+0x10): undefined reference to `A.12.4064'
[...]
..., but after the recent PR94848, PR95551 changes, that problem is now gone.
libgomp/
PR lto/94848
* testsuite/libgomp.fortran/use_device_ptr-optional-3.f90: Add
'dg-do run'.
---
libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90 | 1 +
1 file changed, 1 insertion(+)
diff --git a/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90 b/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
index f2e1a60757f9..b06a88415b47 100644
--- a/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
+++ b/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
@@ -1,3 +1,4 @@
+! { dg-do run }
! Check whether absent optional arguments are properly
! handled with use_device_{addr,ptr}.
program main
--
2.27.0
[-- Attachment #3: 0001-Add-dg-do-run-to-libgomp.fortran-use_device_ptr-.g10.patch --]
[-- Type: text/x-diff, Size: 2793 bytes --]
From 61c896d84bdefbfffa7573a8af89119d4db7b3de Mon Sep 17 00:00:00 2001
From: Thomas Schwinge <thomas@codesourcery.com>
Date: Tue, 7 Jan 2020 17:40:14 +0100
Subject: [PATCH] Add 'dg-do run' to
'libgomp.fortran/use_device_ptr-optional-3.f90' [PR94848]
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Fix-up for r279858/commit f760c0c77fe350616da9dbeaea16442b0acfb09c "Fortran]
OpenMP/OpenACC – fix more issues with OPTIONAL".
With offloading enabled, we then saw:
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 (test for excess errors)
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O0 execution test
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 (test for excess errors)
PASS: libgomp.fortran/use_device_ptr-optional-3.f90 -O1 execution test
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O2 compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -O3 -g compilation failed to produce executable
FAIL: libgomp.fortran/use_device_ptr-optional-3.f90 -Os (test for excess errors)
UNRESOLVED: libgomp.fortran/use_device_ptr-optional-3.f90 -Os compilation failed to produce executable
... due to:
/tmp/cciVc43I.o:(.gnu.offload_vars+0x10): undefined reference to `A.12.4064'
[...]
..., but after the recent PR94848, PR95551 changes, that problem is now gone.
libgomp/
PR lto/94848
* testsuite/libgomp.fortran/use_device_ptr-optional-3.f90: Add
'dg-do run'.
(cherry picked from commit 5864930754f63e2dcef9606f2514ae20e80f436e)
---
libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90 | 1 +
1 file changed, 1 insertion(+)
diff --git a/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90 b/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
index f2e1a60757f9..b06a88415b47 100644
--- a/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
+++ b/libgomp/testsuite/libgomp.fortran/use_device_ptr-optional-3.f90
@@ -1,3 +1,4 @@
+! { dg-do run }
! Check whether absent optional arguments are properly
! handled with use_device_{addr,ptr}.
program main
--
2.27.0
next prev parent reply other threads:[~2020-06-17 22:22 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <75222f0b-b55d-dcd9-b4f4-8e218675e8d7@mentor.com>
2019-07-12 11:35 ` [PATCH 0/5, OpenACC] Add support for Fortran optional arguments in OpenACC Kwok Cheung Yeung
2019-07-12 11:36 ` [PATCH 1/5, OpenACC] Allow NULL as an argument to OpenACC 2.6 directives Kwok Cheung Yeung
2019-11-29 14:42 ` [PR92726] OpenACC: 'NULL'-in -> no-op, and/or 'NULL'-out (was: [PATCH 1/5, OpenACC] Allow NULL as an argument to OpenACC 2.6 directives) Thomas Schwinge
2019-11-20 13:11 ` [Patch][OpenMP/OpenACC/Fortran] Fix mapping of optional (present|absent) arguments Tobias Burnus
2019-11-29 12:17 ` Tobias Burnus
2019-12-05 15:16 ` Jakub Jelinek
2019-12-05 15:47 ` [RFC] Characters per line: from punch card (80) to line printer (132) (was: [Patch][OpenMP/OpenACC/Fortran] Fix mapping of optional (present|absent) arguments) Thomas Schwinge
2019-12-05 16:04 ` Jakub Jelinek
2019-12-05 20:21 ` Segher Boessenkool
2019-12-05 16:17 ` Joseph Myers
2019-12-05 16:24 ` Paul Koning
2019-12-05 16:40 ` Jeff Law
2019-12-05 16:55 ` [RFC] Characters per line: from punch card (80) to line printer (132) Florian Weimer
2019-12-05 17:55 ` Andrew Stubbs
2019-12-05 18:12 ` Eric Gallager
2019-12-05 18:22 ` Robin Curtis
2019-12-05 19:16 ` James Secan
2019-12-06 9:22 ` Andrew Stubbs
2019-12-05 16:44 ` [RFC] Characters per line: from punch card (80) to line printer (132) (was: [Patch][OpenMP/OpenACC/Fortran] Fix mapping of optional (present|absent) arguments) Michael Matz
2019-12-05 17:03 ` Jonathan Wakely
2019-12-05 18:07 ` Marek Polacek
2019-12-05 20:06 ` Segher Boessenkool
2019-12-05 20:38 ` Marek Polacek
2019-12-05 22:02 ` Segher Boessenkool
2019-12-05 20:56 ` Jonathan Wakely
2019-12-05 22:19 ` Segher Boessenkool
2019-12-05 22:34 ` Jonathan Wakely
2019-12-05 22:37 ` Jonathan Wakely
2019-12-05 23:02 ` Segher Boessenkool
2019-12-05 17:29 ` N.M. Maclaren
2019-12-05 20:12 ` Segher Boessenkool
2019-12-05 20:41 ` Jason Merrill
2019-12-05 18:54 ` [RFC] Characters per line: from punch card (80) to line printer (132) Martin Sebor
2019-12-05 20:32 ` Segher Boessenkool
2019-12-07 14:49 ` [Patch][OpenMP/OpenACC/Fortran] Fix mapping of optional (present|absent) arguments Thomas Schwinge
2019-12-11 10:52 ` Tobias Burnus
2019-12-10 17:54 ` [Patch, Fortran] OpenMP/OpenACC – fix more issues with OPTIONAL Tobias Burnus
2019-12-16 9:25 ` *ping* " Tobias Burnus
2019-12-29 23:46 ` *ping**2 " Tobias Burnus
2019-12-30 4:33 ` Jerry
2020-01-03 11:29 ` Jakub Jelinek
2020-01-08 8:33 ` Thomas Schwinge
2020-01-08 8:55 ` Tobias Burnus
2020-04-29 10:00 ` Thomas Schwinge
2020-04-29 14:01 ` Tobias Burnus
2020-04-30 6:24 ` Richard Biener
2020-04-30 7:17 ` Jakub Jelinek
2020-05-05 9:08 ` Tobias Burnus
2020-06-17 22:22 ` Thomas Schwinge [this message]
2019-12-02 16:59 ` [PR92726] OpenACC: 'NULL'-in -> no-op, and/or 'NULL'-out Tobias Burnus
2019-07-12 11:37 ` [PATCH 2/5, OpenACC] Support Fortran optional arguments in the firstprivate clause Kwok Cheung Yeung
2019-07-12 11:42 ` Jakub Jelinek
2019-07-17 18:08 ` Kwok Cheung Yeung
2019-07-17 20:49 ` Tobias Burnus
2019-07-18 9:30 ` Tobias Burnus
2019-07-18 11:45 ` Kwok Cheung Yeung
2019-07-29 21:01 ` Kwok Cheung Yeung
2019-07-31 12:54 ` Jakub Jelinek
2019-10-07 9:26 ` Thomas Schwinge
2019-10-07 13:16 ` Kwok Cheung Yeung
2019-07-12 11:39 ` [PATCH 3/5, OpenACC] Add support for allocatable arrays as optional arguments Kwok Cheung Yeung
2019-07-12 11:39 ` [PATCH 4/5, OpenACC] Allow optional arguments to be used in the use_device OpenACC clause Kwok Cheung Yeung
2019-07-29 22:42 ` Kwok Cheung Yeung
2019-07-31 13:13 ` Jakub Jelinek
2019-07-12 11:41 ` [PATCH 5/5, OpenACC] Add tests for Fortran optional arguments in OpenACC 2.6 Kwok Cheung Yeung
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=878sgl2uzn.fsf@euler.schwinge.homeip.net \
--to=thomas@codesourcery.com \
--cc=fortran@gcc.gnu.org \
--cc=gcc-patches@gcc.gnu.org \
--cc=jakub@redhat.com \
--cc=tobias@codesourcery.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).