Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 413865 - app-arch/libarchive-3.0.4 breaks tar on amd64-fbsd
Summary: app-arch/libarchive-3.0.4 breaks tar on amd64-fbsd
Status: RESOLVED FIXED
Alias: None
Product: Gentoo/Alt
Classification: Unclassified
Component: FreeBSD (show other bugs)
Hardware: AMD64 FreeBSD
: Normal minor (vote)
Assignee: Samuli Suominen (RETIRED)
URL:
Whiteboard:
Keywords: Bug, PATCH, REGRESSION, REVIEWED
Depends on:
Blocks:
 
Reported: 2012-04-28 12:52 UTC by Richard Yao (RETIRED)
Modified: 2012-06-14 16:45 UTC (History)
2 users (show)

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

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Richard Yao (RETIRED) gentoo-dev 2012-04-28 12:52:27 UTC 
app-arch/libarchive-3.0.4 causes an infinite loop when running `tar -cf - .`. That command is part of sys-devel/gcc's build system. Other packages, such as dev-libs/glib, also break.

Please mask app-arch/libarchive-3.0.4 and keyword app-arch/libarchive-3.0.3.
Comment 1 Alexis Ballier gentoo-dev 2012-04-28 14:11:14 UTC 
works here so you'll have to give more details and try to see if you havent broken your system in some way:

# tar -cf - . > /tmp/toto.tar
# file /tmp/toto.tar 
/tmp/toto.tar: POSIX tar archive
Comment 2 Richard Yao (RETIRED) gentoo-dev 2012-04-28 14:21:28 UTC 
ping me when you are on IRC.
Comment 3 Alexis Ballier gentoo-dev 2012-04-28 14:52:46 UTC 
(In reply to comment #2)
> ping me when you are on IRC.

i'm not so you'll have to put details here :)
Comment 4 Richard Yao (RETIRED) gentoo-dev 2012-05-03 15:35:38 UTC 
nigoro and I were talking about this in IRC. It seems that it only happens in x86 chroots on amd64 kernels.
Comment 5 Alexis Ballier gentoo-dev 2012-05-04 12:16:10 UTC 
(In reply to comment #4)
> nigoro and I were talking about this in IRC. It seems that it only happens
> in x86 chroots on amd64 kernels.

if you only provide details like that, without backtrace and an actual analysis i'm afraid there's absolutely nothing we can do about this...
Comment 6 Alexis Ballier gentoo-dev 2012-05-05 13:18:31 UTC 
and this for sure doesnt prevent making stages here
Comment 7 Naohiro Aota gentoo-dev 2012-05-15 00:09:59 UTC 
try gdb and paste backtrace here if possible.

FYI I got libarchive-3.0.4 test_fuzz hang with this backtrace:

# gdb  ./libarchive_test -p 27027
...
(gdb) bt
#0  0x286ea316 in ?? () from /lib/libc.so.7
#1  0x286e8221 in ?? () from /lib/libc.so.7
#2  0x286e84ef in ?? () from /lib/libc.so.7
#3  0x286e8574 in ?? () from /lib/libc.so.7
#4  0x286ea8a5 in mktime () from /lib/libc.so.7
#5  0x0806ae55 in archive_read_format_cab_read_header ()
#6  0x00000004 in ?? ()
#7  0x2888af00 in ?? ()
Backtrace stopped: previous frame inner to this frame (corrupt stack?)

Something similar to bug #324445.
Comment 8 Naohiro Aota gentoo-dev 2012-05-25 04:17:26 UTC 
Fix for bug #324445 has been commited.

I think this is DUPLICATE.. but there's no info to check if it is so.
Comment 9 Richard Yao (RETIRED) gentoo-dev 2012-06-14 10:49:18 UTC 
(In reply to comment #8)
> Fix for bug #324445 has been commited.
> 
> I think this is DUPLICATE.. but there's no info to check if it is so.

uaie in #gentoo-bsd reported that this is still an issue. I am masking app-arch/libarchive-3.0.4.
Comment 10 Richard Yao (RETIRED) gentoo-dev 2012-06-14 11:16:22 UTC 
This is fixed by the following patch by naota:

https://github.com/libarchive/libarchive/pull/13

Upstream has accepted it. Are there any objections against committing it to the tree?
Comment 11 Richard Yao (RETIRED) gentoo-dev 2012-06-14 16:45:01 UTC 
Committed to CVS as app-arch/libarchive-3.0.4-r1.