From: Richard Biener <richard.guenther@gmail.com>
To: Hale Wang <Hale.Wang@arm.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] PR61123 : Fix the ABI mis-matching error caused by LTO
Date: Wed, 18 Jun 2014 10:21:00 -0000 [thread overview]
Message-ID: <CAFiYyc3KYx1p2AyY2CkBbR92qjuzE4pACaqiq-BdbgHo0OVUvQ@mail.gmail.com> (raw)
In-Reply-To: <001801cf8acd$512b1360$f3813a20$@arm.com>
On Wed, Jun 18, 2014 at 10:14 AM, Hale Wang <Hale.Wang@arm.com> wrote:
> Hi,
>
> With LTO, -fno-short-enums is ignored, resulting in ABI mis-matching in
> linking.
>
> Refer https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61123 for details.
>
> This patch add fshort-enums and fshout-wchar to LTO group.
>
> To check it, a new procedure object-readelf is added in
> testsuite/lib/lto.exp and new lto tests are added in gcc.target/arm/lto.
>
> Bootstrap and no make check regression on X86-64.
>
> Patch also attached for convenience. Is It ok for trunk?
>
> Thanks and Best Regards,
> Hale Wang
>
> c-family/ChangeLog
> 2014-06-18 Hale Wang <hale.wang@arm.com>
>
> PR lto/61123
> *c.opt (fshort-enums): Add to LTO.
> *c.opt (fshort-wchar): Likewise.
Space after the *.
I think you don't need to copy the LTO harness but you can simply
use dg.exp and sth similar to gcc.dg/20081223-1.c (there is an
effective target 'lto' to guard for lto support).
So simply place the testcase in gcc.target/arm/ (make sure to
put a dg-do compile on the 2nd file and use dg-additional-sources).
If that doesn't work I'd say put the testcase in gcc.dg/lto/ instead
and do a dg-skip-if for non-arm targets.
Ok with one of those changes.
Thanks,
Richard.
> testsuite/ChangeLog
> 2014-06-18 Hale Wang <hale.wang@arm.com>
>
> * gcc.target/arm/lto/: New folder to verify the LTO
> option for ARM specific.
> * gcc.target/arm/lto/pr61123-enum-size_0.c: New test
> case.
> * gcc.target/arm/lto/pr61123-enum-size_1.c: Likewise.
> * gcc.target/arm/lto/lto.exp: New exp file used to test
> LTO option for ARM specific.
> * lib/lto.exp (object-readelf): New procedure used to
> catch the enum size in the final executable.
>
> Index: gcc/c-family/c.opt
> ===================================================================
> --- gcc/c-family/c.opt (revision 211394)
> +++ gcc/c-family/c.opt (working copy)
> @@ -1189,11 +1189,11 @@
> Use the same size for double as for float
>
> fshort-enums
> -C ObjC C++ ObjC++ Optimization Var(flag_short_enums)
> +C ObjC C++ ObjC++ LTO Optimization Var(flag_short_enums)
> Use the narrowest integer type possible for enumeration types
>
> fshort-wchar
> -C ObjC C++ ObjC++ Optimization Var(flag_short_wchar)
> +C ObjC C++ ObjC++ LTO Optimization Var(flag_short_wchar)
> Force the underlying type for \"wchar_t\" to be \"unsigned short\"
>
> fsigned-bitfields
> Index: gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_0.c
> ===================================================================
> --- gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_0.c (revision
> 0)
> +++ gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_0.c (revision 0)
> @@ -0,0 +1,22 @@
> +/* { dg-lto-do link } */
> +/* { dg-lto-options { { -fno-short-enums -Wl,-Ur,--no-enum-size-warning -Os
> -nostdlib -flto } } } */
> +
> +#include <stdlib.h>
> +
> +enum enum_size_attribute
> +{
> + small_size, int_size
> +};
> +
> +struct debug_ABI_enum_size
> +{
> + enum enum_size_attribute es;
> +};
> +
> +int
> +foo1 (struct debug_ABI_enum_size *x)
> +{
> + return sizeof (x->es);
> +}
> +
> +/* { dg-final { object-readelf Tag_ABI_enum_size int { target arm_eabi } }
> } */
> Index: gcc/testsuite/gcc.target/arm/lto/lto.exp
> ===================================================================
> --- gcc/testsuite/gcc.target/arm/lto/lto.exp (revision 0)
> +++ gcc/testsuite/gcc.target/arm/lto/lto.exp (revision 0)
> @@ -0,0 +1,59 @@
> +# Copyright (C) 2009-2014 Free Software Foundation, Inc.
> +
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 3 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with GCC; see the file COPYING3. If not see
> +# <http://www.gnu.org/licenses/>.
> +#
> +# Contributed by Diego Novillo <dnovillo@google.com>
> +
> +
> +# Test link-time optimization across multiple files.
> +#
> +# Programs are broken into multiple files. Each one is compiled
> +# separately with LTO information. The final executable is generated
> +# by collecting all the generated object files using regular LTO or WHOPR.
> +
> +if $tracelevel then {
> + strace $tracelevel
> +}
> +
> +# Load procedures from common libraries.
> +load_lib standard.exp
> +load_lib gcc.exp
> +
> +# Load the language-independent compabibility support procedures.
> +load_lib lto.exp
> +
> +# If LTO has not been enabled, bail.
> +if { ![check_effective_target_lto] } {
> + return
> +}
> +
> +gcc_init
> +lto_init no-mathlib
> +
> +# Define an identifier for use with this suite to avoid name conflicts
> +# with other lto tests running at the same time.
> +set sid "c_lto"
> +
> +# Main loop.
> +foreach src [lsort [find $srcdir/$subdir *_0.c]] {
> + # If we're only testing specific files and this isn't one of them, skip
> it.
> + if ![runtest_file_p $runtests $src] then {
> + continue
> + }
> +
> + lto-execute $src $sid
> +}
> +
> +lto_finish
> Index: gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_1.c
> ===================================================================
> --- gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_1.c (revision
> 0)
> +++ gcc/testsuite/gcc.target/arm/lto/pr61123-enum-size_1.c (revision 0)
> @@ -0,0 +1,5 @@
> +int
> +foo2 (int y)
> +{
> + return y*10;
> +}
> Index: gcc/testsuite/lib/lto.exp
> ===================================================================
> --- gcc/testsuite/lib/lto.exp (revision 211394)
> +++ gcc/testsuite/lib/lto.exp (working copy)
> @@ -650,3 +650,82 @@
> fail "scan-symbol $args"
> }
> }
> +
> +# Call pass if object readelf is ok, otherwise fail.
> +# example: /* { dg-final { object-readelf Tag_ABI_enum_size int} } */
> +proc object-readelf { args } {
> + global readelf
> + global base_dir
> + upvar 2 execname execname
> +
> + if { [llength $args] < 2 } {
> + error "object-readelf: too few arguments"
> + return
> + }
> + if { [llength $args] > 3 } {
> + error "object-readelf: too many arguments"
> + return
> + }
> + if { [llength $args] >= 3 } {
> + switch [dg-process-target [lindex $args 2]] {
> + "S" { }
> + "N" { return }
> + "F" { setup_xfail "*-*-*" }
> + "P" { }
> + }
> + }
> +
> + # Find size like we find g++ in g++.exp.
> + if ![info exists readelf] {
> + set readelf [findfile $base_dir/../../../binutils/readelf \
> + $base_dir/../../../binutils/readelf \
> + [findfile $base_dir/../../readelf
> $base_dir/../../readelf \
> + [findfile $base_dir/readelf
> $base_dir/readelf \
> + [transform readelf]]]]
> + verbose -log "readelf is $readelf"
> + }
> +
> + set what [lindex $args 0]
> + set with [lindex $args 1]
> +
> + if ![file_on_host exists $execname] {
> + verbose -log "$execname does not exist"
> + unresolved "object-readelf $what "
> + return
> + }
> +
> + set output [remote_exec host "$readelf -A" "$execname"]
> + set status [lindex $output 0]
> + if { $status != 0 } {
> + verbose -log "object-readelf: $readelf failed"
> + unresolved "object-readelf $what $execname"
> + return
> + }
> +
> + set text [lindex $output 1]
> + set lines [split $text "\n"]
> +
> + set done 0
> + set i 0
> + while { !$done } {
> + set line_tex [lindex $lines $i]
> + if { [llength ${line_tex}] > 1} {
> + incr i
> + if [regexp -- $what $line_tex]
> {
> + set match
> [regexp -- $with $line_tex]
> + set done 1
> + }
> + } else {
> + set done 1
> + }
> + }
> +
> + verbose -log "$what size is $with;"
> + if { $match == 1 } {
> + pass "object-readelf $what size is correct."
> + } else {
> + fail "object-readelf $what size is incorrect."
> + }
> +}
> +
> +
next prev parent reply other threads:[~2014-06-18 10:21 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-18 8:14 Hale Wang
2014-06-18 10:21 ` Richard Biener [this message]
2014-06-18 10:22 ` Richard Biener
2014-06-18 17:41 ` Mike Stump
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=CAFiYyc3KYx1p2AyY2CkBbR92qjuzE4pACaqiq-BdbgHo0OVUvQ@mail.gmail.com \
--to=richard.guenther@gmail.com \
--cc=Hale.Wang@arm.com \
--cc=gcc-patches@gcc.gnu.org \
/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).