From: Michael Collison <Michael.Collison@arm.com>
To: Richard Earnshaw <Richard.Earnshaw@arm.com>,
"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Cc: nd <nd@arm.com>
Subject: RE: [PATCH][Arm] Test suite failures resulting from deprecation of -mstructure-size-boundary
Date: Wed, 09 Aug 2017 05:25:00 -0000 [thread overview]
Message-ID: <HE1PR0802MB23775593A29B4A00E8BC114B958B0@HE1PR0802MB2377.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <363725a3-b677-85ac-d4d2-95277dbbc524@arm.com>
Because the comment (for example) in g+=.dg/ext/packed8.C says
// NOTE: This test assumes packed structure layout differs from unpacked
// structure layout. This isn't true, e.g., with the default
// arm-none-elf options.
If we could just delete the -mstructure-size-boundary=8 option why was it added in the first place?
-----Original Message-----
From: Richard Earnshaw (lists) [mailto:Richard.Earnshaw@arm.com]
Sent: Monday, August 7, 2017 5:32 AM
To: Michael Collison <Michael.Collison@arm.com>; gcc-patches@gcc.gnu.org
Cc: nd <nd@arm.com>
Subject: Re: [PATCH][Arm] Test suite failures resulting from deprecation of -mstructure-size-boundary
On 06/08/17 00:25, Michael Collison wrote:
> This patch fixes test case failures on arm targets due to '-mstructure-size-boundary' being deprecated. The test cases were failing because a warning was being issued and due to the fact that the size of packed and unpacked structures is the same after deprecating '-mstructure-size-boundary'
>
> Okay for trunk?
>
> 2017-08-04 Michael Collison <michael.collison@arm.com>
>
> * testsuite/g++.dg/ext/packed8.C: Skip test for arm_eabi.
> * testsuite/g++.dg/init/array16.C: Skip test for arm_eabi.
> * testsuite/g++.dg/other/crash-4.C: Skip test for arm_eabi.
> * testsuite/gcc.dg/builtin-stringop-chk-1.c: Skip test for arm_eabi.
>
Why would we want to skip the test? If you delete the -mstructure-size-boundary option then the test should execute correctly on all arm-eabi based platforms and most other ARM platforms where 8 was the default anyway.
R.
>
> pr7519v1.patch
>
>
> diff --git a/gcc/testsuite/g++.dg/ext/packed8.C
> b/gcc/testsuite/g++.dg/ext/packed8.C
> index 91ee8b3..4f38670 100644
> --- a/gcc/testsuite/g++.dg/ext/packed8.C
> +++ b/gcc/testsuite/g++.dg/ext/packed8.C
> @@ -2,7 +2,7 @@
> // NOTE: This test assumes packed structure layout differs from unpacked
> // structure layout. This isn't true, e.g., with the default
> // arm-none-elf options.
> -// { dg-options "-mstructure-size-boundary=8" { target arm*-*-* } }
> +// { dg-skip-if "packed structure layout does not differ from
> +unpacked layout" { { arm*-*-* } && { arm_eabi } } }
>
> class A
> {
> diff --git a/gcc/testsuite/g++.dg/init/array16.C
> b/gcc/testsuite/g++.dg/init/array16.C
> index 188d1a8..3334e25 100644
> --- a/gcc/testsuite/g++.dg/init/array16.C
> +++ b/gcc/testsuite/g++.dg/init/array16.C
> @@ -1,7 +1,7 @@
> // Causes timeout for the MMIX simulator on a 3GHz P4 and we can't
> // have "compile" for some targets and "run" for others.
> // { dg-do run { target { ! mmix-*-* } } } -// { dg-options
> "-mstructure-size-boundary=8" { target arm*-*-* } }
> +// { dg-skip-if "packed structure layout does not differ from
> +unpacked layout" { { arm*-*-* } && { arm_eabi } } }
>
> // Copyright (C) 2004 Free Software Foundation, Inc.
> // Contributed by Nathan Sidwell 8 Dec 2004 <nathan@codesourcery.com>
> diff --git a/gcc/testsuite/g++.dg/other/crash-4.C
> b/gcc/testsuite/g++.dg/other/crash-4.C
> index a77fe05..8530f44 100644
> --- a/gcc/testsuite/g++.dg/other/crash-4.C
> +++ b/gcc/testsuite/g++.dg/other/crash-4.C
> @@ -7,7 +7,7 @@
> // NOTE: This test assumes packed structure layout differs from unpacked
> // structure layout. This isn't true, e.g., with the default
> // arm-none-elf options.
> -// { dg-options "-mstructure-size-boundary=8" { target arm*-*-* } }
> +// { dg-skip-if "packed structure layout does not differ" { {
> +arm*-*-* } && { arm_eabi } } }
>
> struct a
> {
> diff --git a/gcc/testsuite/gcc.dg/builtin-stringop-chk-1.c
> b/gcc/testsuite/gcc.dg/builtin-stringop-chk-1.c
> index e265578..d839097 100644
> --- a/gcc/testsuite/gcc.dg/builtin-stringop-chk-1.c
> +++ b/gcc/testsuite/gcc.dg/builtin-stringop-chk-1.c
> @@ -2,7 +2,7 @@
> are emitted properly. */
> /* { dg-do compile } */
> /* { dg-options "-O2 -Wno-format -std=gnu99
> -ftrack-macro-expansion=0" } */
> -/* { dg-additional-options "-mstructure-size-boundary=8" { target
> arm*-*-* } } */
> +// { dg-skip-if "packed structure layout does not differ from
> +unpacked layout" { { arm*-*-* } && { arm_eabi } } }
> // { dg-skip-if "packed attribute missing for t" { "epiphany-*-*" } }
>
> extern void abort (void);
>
next prev parent reply other threads:[~2017-08-09 5:25 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-05 23:25 Michael Collison
2017-08-07 12:32 ` Richard Earnshaw (lists)
2017-08-09 5:25 ` Michael Collison [this message]
2017-08-09 8:21 ` Richard Earnshaw (lists)
2017-08-09 15:06 ` Michael Collison
2017-08-09 16:11 ` Richard Earnshaw (lists)
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=HE1PR0802MB23775593A29B4A00E8BC114B958B0@HE1PR0802MB2377.eurprd08.prod.outlook.com \
--to=michael.collison@arm.com \
--cc=Richard.Earnshaw@arm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=nd@arm.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).