Bug 59490

Summary: SIGILL during Xserver startup
Product: [Retired] Red Hat Linux Reporter: Richard Henderson <rth>
Component: XFree86Assignee: Mike A. Harris <mharris>
Status: CLOSED CURRENTRELEASE QA Contact: David Lawrence <dkl>
Severity: high Docs Contact:
Priority: medium    
Version: 7.1CC: mharris
Target Milestone: ---   
Target Release: ---   
Hardware: alpha   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2002-12-19 10:00:42 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Xserver startup log
none
gdb session none

Description Richard Henderson 2002-02-08 21:37:08 UTC 
Description of problem:
On PC164 with Elsa Permedia2, the X server crashes during startup.
Logs to be attached.

Version-Release number of selected component (if applicable):
XFree86-4.1.0-15

How reproducible:
Always

Steps to Reproduce:
X

Additional info:
Twice I've seen /usr/X11R6/lib/modules/libramdac.a get corrupted
in the process.  This aspect I don't understand, since strace shows
that the file was indeed opened O_RDONLY.  Perhaps it's related to
the kernel version -- not all bugfixes for PC164 were included before
2.4.15, so I cannot run a stock 2.4.9-21 kernel.  Presently running
2.4.18-pre9.
Comment 1 Richard Henderson 2002-02-08 21:38:15 UTC 
Created attachment 45017 [details]
Xserver startup log
Comment 2 Richard Henderson 2002-02-08 21:40:19 UTC 
Created attachment 45018 [details]
gdb session
Comment 3 Mike A. Harris 2002-02-08 23:10:01 UTC 
Reassigning to alpha development group.
Comment 4 Mike A. Harris 2002-02-09 22:32:46 UTC 
no, really this time.  promise.
Comment 5 Beth Uptagrafft 2002-02-14 17:45:12 UTC 
This case was filed against 7.1, is this correct?  Does this happen on 7.1 or 
7.2?
Comment 6 Richard Henderson 2002-02-14 17:56:41 UTC 
It's really 7.1 with all updates installed.
Comment 7 Mike A. Harris 2002-11-08 15:16:13 UTC 
Reassigning...
Comment 8 Mike A. Harris 2002-11-08 15:17:37 UTC 
>(II) GLINT(0): Not using default mode "1920x1440" (bad mode
>clock/interlace/doublescan)
>
>Fatal server error:
>Caught signal 4.  Server aborting

SIGILL is generally a case of bad memory, overheating or some other
hardware issue.  Does this problem persist still?

If so, have you also tried RHL 7.2/Alpha?

Can you test the memory in this machine for failure?
Comment 9 Richard Henderson 2002-11-08 19:08:09 UTC 
SIGILL on Alpha is generally a case of compiling code for a newer ISA
and running it on a machine that only supports an older ISA.

RHL 7.2/Alpha isn't.  There are no generally available ISOs, there is
no published security errata.  Despite the fact that we gave a cd to
Compaq, IMO it isn't a real release.

All that said, the problems went away with XFree86-4.1.0-25.
Comment 10 Mike A. Harris 2002-12-19 10:00:42 UTC 
Ok, cool, closing CURRENTRELEASE.