Bug 112020 - segfaults when building package
Summary: segfaults when building package
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Red Hat Raw Hide
Classification: Retired
Component: rpm
Version: 1.0
Hardware: s390
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Jeff Johnson
QA Contact: Mike McLean
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2003-12-12 22:13 UTC by Bill Nottingham
Modified: 2014-03-17 02:40 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2004-03-02 03:28:58 UTC
Embargoed:

Attachments (Terms of Use)
source rpm that blows up on build (94.07 KB, application/octet-stream)
2003-12-12 22:14 UTC, Bill Nottingham 		no flags Details
View All

Description Bill Nottingham 2003-12-12 22:13:56 UTC 
Rebuild the attached, and you get:


Checking for unpackaged file(s): /usr/lib/rpm/check-files
/var/tmp/MAKEDEV-root
Wrote: /usr/src/redhat/RPMS/s390x/MAKEDEV-3.3.9-1.s390x.rpm
free(): invalid pointer 0x441288!
free(): invalid pointer 0x4270f0!
Segmentation fault
Comment 1 Bill Nottingham 2003-12-12 22:14:26 UTC 
Created attachment 96505 [details]
source rpm that blows up on build
Comment 2 Bill Nottingham 2003-12-12 22:26:32 UTC 
Wrote: /usr/src/redhat/RPMS/s390x/MAKEDEV-3.3.9-1.s390x.rpm
 
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1077715616 (LWP 10212)]
0x402f0f18 in _int_malloc () from /lib/tls/libc.so.6
(gdb) bt
#0  0x402f0f18 in _int_malloc () from /lib/tls/libc.so.6
#1  0x402f2ba0 in malloc_check () from /lib/tls/libc.so.6
#2  0x402f0872 in calloc () from /lib/tls/libc.so.6
#3  0x4007d706 in saveHardLink (fsm=0x43e8d8) at fsm.c:420
#4  0x4007f8b6 in fsmStage (fsm=0x43e8d8, stage=FSM_INIT) at fsm.c:1602
#5  0x4007fa64 in fsmStage (fsm=0x43e8d8, stage=FSM_PKGBUILD) at
fsm.c:1413
#6  0x4007dc00 in fsmSetup (fsm=0x43e8d8, goal=1077704208, ts=0x47dd60,
    fi=0x4466a0, cfd=0x1, archiveSize=0x7fffb5e8, failedFile=0x7fff9470)
    at fsm.c:559
#7  0x4003ffea in cpio_doio (fdo=0x47dfd0, h=0x403c7610, csa=0x47dfd0,
    fmodeMacro=0x419ed0 "") at pack.c:82
#8  0x400416e4 in writeRPM (hdrp=0x4151b8, pkgidp=0x0,
    fileName=0x4168b8 "/usr/src/redhat/RPMS/s390x/dev-3.3.9-1.s390x.rpm",
    type=0, csa=0x7fffb5e8, passPhrase=0x417758 "", cookie=0x0) at
pack.c:509
#9  0x40041c26 in packageBinaries (spec=0x41d238) at pack.c:833
#10 0x40038782 in buildSpec (ts=0x4, spec=0x41d238, what=2975, test=0)
    at build.c:330
#11 0x00402c74 in ?? ()
#12 0x7fff9110 in ?? ()
(gdb)
Comment 3 Jeff Johnson 2003-12-13 21:39:39 UTC 
Ick, s390x. Will look soonish.
Comment 4 Jeff Johnson 2004-01-16 18:46:08 UTC 
Hmmm, valgrind on i686 shows nothing unusual.

And -lefence works fine on s390. Starting to smell a s390 fish ...
Comment 5 Jeff Johnson 2004-03-01 13:51:36 UTC 
Attached srpm builds on s390x with rpm-4.2.2-0.12.
Comment 6 Bill Nottingham 2004-03-02 03:28:58 UTC 
Thanks.
Note You need to log in before you can comment on or make changes to this bug.