public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
@ 2011-06-04 19:52 andi-gcc at firstfloor dot org
  2011-06-04 22:14 ` [Bug other/49284] " andi-gcc at firstfloor dot org
                   ` (6 more replies)
  0 siblings, 7 replies; 8+ messages in thread
From: andi-gcc at firstfloor dot org @ 2011-06-04 19:52 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

           Summary: -fdump-ipa-cgraph leads to ICE in
                    generate_canonical_option, at opts-common.c:303
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: other
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: andi-gcc@firstfloor.org


gcc version 4.7.0 20110604 (experimental) (GCC) 


Passing -fdump-ipa-cgraph to a IPA final build gives:

lto1: internal compiler error: in generate_canonical_option, at
opts-common.c:303


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
@ 2011-06-04 22:14 ` andi-gcc at firstfloor dot org
  2011-06-05 23:43 ` andi-gcc at firstfloor dot org
                   ` (5 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: andi-gcc at firstfloor dot org @ 2011-06-04 22:14 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

--- Comment #1 from Andi Kleen <andi-gcc at firstfloor dot org> 2011-06-04 22:13:34 UTC ---
Some investigation:

This depends heavily on the command line used.

A simple test with a hello world works.

On my kernel build when I strip the lto link command line down I get
the error with

gcc47 -fdump-cgraph-ipa -nostdlib -fuse-linker-plugin -flto=jobserver
-fwhole-program -Wl,-m,elf_x86_64 -Wl,--build-id -o .tmp_vmlinux1
-Wl,-T,arch/x86/kernel/vmlinux.lds arch/x86/kernel/head_64.o
arch/x86/kernel/head_64.o

and not with (last file deleted)

gcc47 -fdump-cgraph-ipa -nostdlib -fuse-linker-plugin -flto=jobserver
-fwhole-program -Wl,-m,elf_x86_64 -Wl,--build-id -o .tmp_vmlinux1
-Wl,-T,arch/x86/kernel/vmlinux.lds arch/x86/kernel/head_64.o


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
  2011-06-04 22:14 ` [Bug other/49284] " andi-gcc at firstfloor dot org
@ 2011-06-05 23:43 ` andi-gcc at firstfloor dot org
  2011-06-06  9:08 ` joseph at codesourcery dot com
                   ` (4 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: andi-gcc at firstfloor dot org @ 2011-06-05 23:43 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

--- Comment #2 from Andi Kleen <andi-gcc at firstfloor dot org> 2011-06-05 23:43:07 UTC ---
FWIW I just commented out the offending assert and the option works
now. 

Is that the correct fix?


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
  2011-06-04 22:14 ` [Bug other/49284] " andi-gcc at firstfloor dot org
  2011-06-05 23:43 ` andi-gcc at firstfloor dot org
@ 2011-06-06  9:08 ` joseph at codesourcery dot com
  2011-06-06 10:35 ` rguenth at gcc dot gnu.org
                   ` (3 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: joseph at codesourcery dot com @ 2011-06-06  9:08 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

--- Comment #3 from joseph at codesourcery dot com <joseph at codesourcery dot com> 2011-06-06 09:07:54 UTC ---
On Sun, 5 Jun 2011, andi-gcc at firstfloor dot org wrote:

> FWIW I just commented out the offending assert and the option works
> now. 
> 
> Is that the correct fix?

No, that's certainly wrong.  The caller of generate_canonical_option 
should never be passing an argument to an option that doesn't take one.


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
                   ` (2 preceding siblings ...)
  2011-06-06  9:08 ` joseph at codesourcery dot com
@ 2011-06-06 10:35 ` rguenth at gcc dot gnu.org
  2011-06-06 10:40 ` andi-gcc at firstfloor dot org
                   ` (2 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: rguenth at gcc dot gnu.org @ 2011-06-06 10:35 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2011.06.06 10:34:36
     Ever Confirmed|0                           |1

--- Comment #4 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-06-06 10:34:36 UTC ---
Waiting for sth like a testcase or more investigation.


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
                   ` (3 preceding siblings ...)
  2011-06-06 10:35 ` rguenth at gcc dot gnu.org
@ 2011-06-06 10:40 ` andi-gcc at firstfloor dot org
  2011-06-06 10:45 ` rguenth at gcc dot gnu.org
  2011-08-18 18:32 ` matt at use dot net
  6 siblings, 0 replies; 8+ messages in thread
From: andi-gcc at firstfloor dot org @ 2011-06-06 10:40 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

--- Comment #5 from Andi Kleen <andi-gcc at firstfloor dot org> 2011-06-06 10:39:32 UTC ---
If you use my command line and just supply any random object files
for the ones specified it should reproduce I think.

If not i'll upload my builddir, but it's large.


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
                   ` (4 preceding siblings ...)
  2011-06-06 10:40 ` andi-gcc at firstfloor dot org
@ 2011-06-06 10:45 ` rguenth at gcc dot gnu.org
  2011-08-18 18:32 ` matt at use dot net
  6 siblings, 0 replies; 8+ messages in thread
From: rguenth at gcc dot gnu.org @ 2011-06-06 10:45 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

--- Comment #6 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-06-06 10:44:43 UTC ---
Can you provide source for "random" object files?


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug other/49284] -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303
  2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
                   ` (5 preceding siblings ...)
  2011-06-06 10:45 ` rguenth at gcc dot gnu.org
@ 2011-08-18 18:32 ` matt at use dot net
  6 siblings, 0 replies; 8+ messages in thread
From: matt at use dot net @ 2011-08-18 18:32 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49284

Matt Hargett <matt at use dot net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |matt at use dot net

--- Comment #7 from Matt Hargett <matt at use dot net> 2011-08-18 18:28:04 UTC ---
I get this when trying to compile scummvm with LTO and whole-program:

$ /usr/lib/gcc-snapshot/bin/g++ --version

g++ (Ubuntu/Linaro 20110813-1ubuntu1) 4.7.0 20110813 (experimental) [trunk
revision 177733]


$ CC=/usr/lib/gcc-snapshot/bin/gcc CXX=/usr/lib/gcc-snapshot/bin/g++
CFLAGS="-Ofast -flto" CXXFLAGS="-Ofast -flto" LDFLAGS="-flto -fwhole-program"
./configure

[...]

$ make -j9

[...]


    RANLIB   base/libbase.a
    LINK     scummvm
lto1: internal compiler error: in generate_canonical_option, at
opts-common.c:303
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-snapshot/README.Bugs> for instructions.
lto-wrapper: /usr/lib/gcc-snapshot/bin/g++ returned 1 exit status
/usr/bin/ld.bfd.real: lto-wrapper failed
collect2: error: ld returned 1 exit status


Removing -fwhole-program and changing the CXXFLAGS to -O1 doesn't change the
behaviour. I can attach a tarball if downloading the latest scummvm tarball is
too bothersome.

Note that I'm not using that dump-ipa-graph option at all.


^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2011-08-18 18:29 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-06-04 19:52 [Bug other/49284] New: -fdump-ipa-cgraph leads to ICE in generate_canonical_option, at opts-common.c:303 andi-gcc at firstfloor dot org
2011-06-04 22:14 ` [Bug other/49284] " andi-gcc at firstfloor dot org
2011-06-05 23:43 ` andi-gcc at firstfloor dot org
2011-06-06  9:08 ` joseph at codesourcery dot com
2011-06-06 10:35 ` rguenth at gcc dot gnu.org
2011-06-06 10:40 ` andi-gcc at firstfloor dot org
2011-06-06 10:45 ` rguenth at gcc dot gnu.org
2011-08-18 18:32 ` matt at use dot net

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).