internal error in cc1 with -O3 or -O4 and -funroll-loops

Bug #582844 reported by canavan
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
gcc-4.4 (Ubuntu)
Expired
Undecided
Unassigned

Bug Description

Binary package hint: gcc-4.4

After using about 7:25 minutes of CPU-Time, gcc/cc1 crashes with an internal error when compiling a file of Imagemagick:

usr/bin/gcc-4.4 -DHAVE_CONFIG_H -I. -O3 -funroll-loops -DNO_DBM_REWRITEMAP -c -o magick/magick_libMagickCore_la-decorate.lo magick/decorate.c -save-temps
gcc-4.4: Internal error: Killed (program cc1)
Please submit a full bug report.
See <file:///usr/share/doc/gcc-4.4/README.Bugs> for instructions

gcc-4.4:
  Installed: 4.4.3-4ubuntu5
  Candidate: 4.4.3-4ubuntu5
  Version table:
 *** 4.4.3-4ubuntu5 0
        500 http://de.archive.ubuntu.com/ubuntu/ lucid/main Packages
        100 /var/lib/dpkg/status

Architecture: amd64
DistroRelease: Ubuntu 10.04
kernel: 2.6.32-22-server

Revision history for this message
canavan (rainer-ubuntu) wrote :
Revision history for this message
canavan (rainer-ubuntu) wrote :

cc1 is executed as

/usr/lib/gcc/x86_64-linux-gnu/4.4.3/cc1 -fpreprocessed decorate.i -quiet -dumpbase decorate.c -mtune=generic -auxbase-strip magick/magick_libMagickCore_la-decorate.lo -O3 -funroll-loops -fstack-protector -o decorate.s

Revision history for this message
Matthias Klose (doko) wrote :

not reproducible in lucid and maverick (4GB RAM). Is this reproducible on another machine?

Changed in gcc-4.4 (Ubuntu):
status: New → Incomplete
Revision history for this message
canavan (rainer-ubuntu) wrote :

It's reproducible on a server with 4GB RAM and no swap, on a similar box, it finishes successfully after about 9:30 minutes and a peak memory usage of almost 6GB, as compared to 1.8s and 56MB with -O2.

Revision history for this message
canavan (rainer-ubuntu) wrote :

I should mention that the "similar box" has 8GM RAM.

Revision history for this message
Matthias Klose (doko) wrote : Re: [Bug 582844] Re: internal error in cc1 with -O3 or -O4 and -funroll-loops

On 17.06.2010 11:28, canavan wrote:
> It's reproducible on a server with 4GB RAM and no swap, on a similar
> box, it finishes successfully after about 9:30 minutes and a peak memory
> usage of almost 6GB,

so the ICE isn't reproducible for you either?

Revision history for this message
canavan (rainer-ubuntu) wrote :

It is reproducible with "only" 4GB, with 8GB RAM, it isn't. You may close this bug as invalid or Won't fix, whichever you think is more appropriate.

Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for gcc-4.4 (Ubuntu) because there has been no activity for 60 days.]

Changed in gcc-4.4 (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.