perl 5.20.2 fails to compile with gcc 5.1.0 -flto -O2 #14689
Comments
From steffen@hauihau.deCreated by steffen@hauihau.deCompiling perl 5.20.2 with GCC 5.1.0 reintroduced the issue reported in bug #113022. I was told, that you should use "__attribute__ ((used))" to check for the availability of functions. This small patch fixes the issue: Inline Patch--- a/Configure.orig 2015-05-04 11:51:34.170517546 +0200
+++ b/Configure 2015-05-04 11:51:58.025516693 +0200
@@ -7954,13 +7954,13 @@
if $contains $tlook $tf >/dev/null 2>&1; then
tval=true;
elif $test "$mistrustnm" = compile -o "$mistrustnm" = run; then
- echo "$extern_C void *$1$tdc; void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
+ echo "$extern_C void *$1$tdc; __attribute__ ((used)) void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
$cc -o try $optimize $ccflags $ldflags try.c >/dev/null 2>&1 $libs && tval=true;
$test "$mistrustnm" = run -a -x try && { $run ./try$_exe >/dev/null 2>&1 || tval=false; };
$rm_try;
fi;
else
- echo "$extern_C void *$1$tdc; void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
+ echo "$extern_C void *$1$tdc; __attribute__ ((used)) void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
$cc -o try $optimize $ccflags $ldflags try.c $libs >/dev/null 2>&1 && tval=true;
$rm_try;
fi;Perl Info |
From @tonycozOn Mon May 04 03:02:16 2015, steffen@hauihau.de wrote:
Not every compiler supports gcc's extensions, so we need to probe for The attached patch is against bleadperl, detected __attribute__((used)), and if If you want to test this againt 5.20 you should be able to copy Configure from blead and apply that patch to it, or copy Configure from: http://perl5.git.perl.org/perl.git/blob_plain/refs/heads/tonyc/attribute-used:/Configure I couldn't reproduce the problem you were having however - with your options with gcc 5.1.0 malloc_size wasn't being detected (Debian oldstable.) Tony |
From @tonycoz0001-perl-125096-detect-attribute-used-and-use-it.patchFrom f65b96ad9dca9dcda19fe9a85ffd5d27811d2961 Mon Sep 17 00:00:00 2001
From: Tony Cook <tony@develop-help.com>
Date: Thu, 7 May 2015 10:44:39 +1000
Subject: [PATCH] [perl #125096] detect attribute used and use it
gcc with -O2 -flto builds can optimize out all the references to
the function we're probing for, mark the function as used to
avoid that.
---
Configure | 36 +++++++++++++++++++++++++++++++++++-
1 file changed, 35 insertions(+), 1 deletion(-)
diff --git a/Configure b/Configure
index 0a405d3..98cb7fe 100755
--- a/Configure
+++ b/Configure
@@ -388,6 +388,7 @@ d_attribute_nonnull=''
d_attribute_noreturn=''
d_attribute_pure=''
d_attribute_unused=''
+d_attribute_used=''
d_attribute_warn_unused_result=''
d_printf_format_null=''
d_backtrace=''
@@ -6737,6 +6738,10 @@ case "$3" in
-a) tf=libc.tmp; tdc="[]";;
*) tlook="^$1\$"; tf=libc.list; tdc="()";;
esac;
+case "$d_attribute_used" in
+ define) attrused="__attribute__ ((used))" ;;
+ *) attrused= ;;
+esac;
case "$d_cplusplus" in
$define) extern_C="extern \"C\"" ;;
*) extern_C="extern" ;;
@@ -6753,7 +6758,7 @@ yes)
if $contains $tlook $tf >/dev/null 2>&1; then
tval=true;
elif $test "$mistrustnm" = compile -o "$mistrustnm" = run; then
- echo "$extern_C void *$1$tdc; void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
+ echo "$extern_C void *$1$tdc; $attrused void *(*(p()))$tdc { return &$1; } int main() { if(p() && p() != (void *)main) return(0); else return(1); }"> try.c;
$cc -o try $optimize $ccflags $ldflags try.c >/dev/null 2>&1 $libs && tval=true;
$test "$mistrustnm" = run -a -x try && { $run ./try$_exe >/dev/null 2>&1 || tval=false; };
$rm_try;
@@ -6852,6 +6857,34 @@ $rm_try
set d_longdbl
eval $setvar
+: Look for GCC-style attribute used
+case "$d_attribute_used" in
+'')
+echo " "
+echo "Checking whether your compiler can handle __attribute__((used)) ..." >&4
+$cat >attrib.c <<'EOCP'
+#include <stdio.h>
+int used_somewhere(void) __attribute__((used));
+EOCP
+if $cc $ccflags -c attrib.c >attrib.out 2>&1 ; then
+ if $contains 'warning' attrib.out >/dev/null 2>&1; then
+ echo "Your C compiler doesn't support __attribute__((used))."
+ val="$undef"
+ else
+ echo "Your C compiler supports __attribute__((used))."
+ val="$define"
+ fi
+else
+ echo "Your C compiler doesn't seem to understand __attribute__ at all."
+ val="$undef"
+fi
+;;
+*) val="$d_attribute_used" ;;
+esac
+set d_attribute_used
+eval $setvar
+$rm -f attrib*
+
: see if ldexpl exists
set ldexpl d_ldexpl
eval $inlibc
@@ -23793,6 +23826,7 @@ d_attribute_nonnull='$d_attribute_nonnull'
d_attribute_noreturn='$d_attribute_noreturn'
d_attribute_pure='$d_attribute_pure'
d_attribute_unused='$d_attribute_unused'
+d_attribute_used='$d_attribute_used'
d_attribute_warn_unused_result='$d_attribute_warn_unused_result'
d_backtrace='$d_backtrace'
d_bcmp='$d_bcmp'
--
1.7.10.4
|
|
The RT System itself - Status changed from 'new' to 'open' |
From @TuxOn Wed, 6 May 2015 17:49:27 -0700, "Tony Cook via RT"
[ Not high priority, as this involves *A LOT* of work ] There are plans to move the meta/dist framework towards a newer version That said, here's the feedback from the future :) --8<--- In particular, the current intsize.U unit uses "static asserts" Even IEEE float endianness is now determined without running a program. The rationale for moving out of run-tests is that it prevents As I do not have unlimited resources, and if I had they were to be In summary, Yes, moving to compile-only is a good plan and certainly an -- |
From steffen@hauihau.deAm 07.05.2015 um 02:49 schrieb Tony Cook via RT:
Hi Tony, your patch does not cleanly apply (hunk #4 must be after #1). After |
From steffen@hauihau.deAm 07.05.2015 um 08:35 schrieb H. Merijn Brand via RT:
The fix is a trivial change, so please include it in the next release. |
From @jkeenanOn Thu May 07 09:01:57 2015, steffen@hauihau.de wrote:
Tony: Could we get an update on the status of this ticket? Thank you very much. -- |
From @tonycozOn Wed, 09 Dec 2015 15:53:32 -0800, jkeenan wrote:
I believe this was fixed in a different way by adec5bb. Tony |
From @tonycozOn Tue, 09 Jul 2019 21:26:06 -0700, tonyc wrote:
I've looked over it again and I still think so, so closing. Tony |
|
@tonycoz - Status changed from 'open' to 'resolved' |
Migrated from rt.perl.org#125096 (status was 'resolved')
Searchable as RT125096$
The text was updated successfully, but these errors were encountered: