Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 223375 - mail-client/mozilla-thunderbird Attachment failure with gcc-4.3
Summary: mail-client/mozilla-thunderbird Attachment failure with gcc-4.3
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] GCC Porting (show other bugs)
Hardware: AMD64 Linux
: High normal (vote)
Assignee: Mozilla Gentoo Team
URL: https://bugzilla.mozilla.org/show_bug...
Whiteboard:
Keywords:
: 217805 (view as bug list)
Depends on:
Blocks: gcc-4.3
  Show dependency tree
 
Reported: 2008-05-23 17:20 UTC by Mikko C.
Modified: 2008-07-21 17:30 UTC (History)
4 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Mikko C. 2008-05-23 17:20:06 UTC
I'll paste the description from: https://bugzilla.mozilla.org/show_bug.cgi?id=426228


Hi,

I noticed that my Thunderbird (x86_64) does not display many attachments (i.e.
all pdf attachments) properly when compiled with gcc-4.3.

When the mail arrives the paperclip icon is being displayed in the list. After
I click onto the the mail to get it displayed in the preview area the paperclip
disapears and no attachment is being shown. The problem also exists for old
mails.

The attachment afterwards is still existent on the mail server (I use IMAP).
Using mozilla-thunderbird-bin-2.0.0.12 (binary version from gentoo) everything
can be displayed properly.

I already tried to create a new profile and to recompile thunderbird with
minimal CFLAGS and LDFLAGS, but the problem was still there.

I recently recompiled my system (gentoo) with gcc 4.3, thus as a last step I
recompiled with gcc-4.2.3 and everything works as ususal.


Reproducible: Always

Steps to Reproduce:
1. Compile mozilla-thunderbird with Gcc-4.3 on amd64

Actual Results:  
Attachments are not showed, can't be saved either.

Expected Results:  
Attachments should be showed.

It seems it can be fixed:

Jan de Groot   2008-05-22 14:02:18 PDT

Not an issue anymore here: after removing almost all patches we have applied on
archlinux, this issue wasn't fixed yet. After unsetting CFLAGS and CXXFLAGS and
removing the --enable-optimize=$CFLAGS thing from .mozconfig, thunderbird works
reliable and no longer has problems with attachments when compiled with gcc
4.3.
Comment 1 Ginsu 2008-06-02 06:27:35 UTC
I can confirm I'm seeing this problem also, and disabling the optimize fixes it.  I put the following line in src_compile():

mozconfig_annotate "broken on amd64" --disable-optimize
Comment 2 Mikko C. 2008-06-02 07:14:30 UTC
Thanks Ginsu!
I've modified the ebuild and it works fine :)
Comment 3 Raúl Porcel (RETIRED) gentoo-dev 2008-06-03 15:55:15 UTC
*** Bug 217805 has been marked as a duplicate of this bug. ***
Comment 4 Raúl Porcel (RETIRED) gentoo-dev 2008-06-03 15:56:32 UTC
Fixed, thanks for the info
Comment 5 Michael Gaber 2008-06-03 16:06:35 UTC
here it happens also on x86, did you fix it for all versions?
Comment 6 Raúl Porcel (RETIRED) gentoo-dev 2008-06-03 18:02:59 UTC
(In reply to comment #5)
> here it happens also on x86, did you fix it for all versions?
> 

Yeah
Comment 7 Walter Meinl 2008-06-03 21:29:46 UTC
(In reply to comment #6)

Raul, sth went wrong, you've added an unconditional "die" in an extra line after econf || die

snippet from my build 
creating ldap/libraries/libssldap/Makefile
 * 
 * ERROR: mail-client/mozilla-thunderbird-2.0.0.14 failed.
 * Call stack:
 *               ebuild.sh, line   49:  Called src_compile
 *             environment, line 3259:  Called die
 * The specific snippet of code:
 *       die;
 *  The die message:
 *   (no error message)
you can see it also here
http://sources.gentoo.org/viewcvs.py/gentoo-x86/mail-client/mozilla-thunderbird/mozilla-thunderbird-2.0.0.14.ebuild?r1=1.5&r2=1.6
Comment 8 Raúl Porcel (RETIRED) gentoo-dev 2008-06-03 22:25:12 UTC
Fixed, sorry :P
Comment 9 Nikos Chantziaras 2008-07-21 17:30:24 UTC
I was wondering why Thunderbird is so slow. A temporary USE flag (force-broken-optimization?) would be useful, since upstream closed this with RESOLVED WORKSFORME.