[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#277884: Bug#277699 isn't fixed in xfree86 4.3.0.dfsg.1-9 :(



On Sat, Dec 11, 2004 at 10:17:46PM -0400, Peter Cordes wrote:
> On Thu, Dec 09, 2004 at 01:20:18PM -0800, Debian Bug Tracking System wrote:
> > Changes: 
> >  xfree86 (4.3.0.dfsg.1-9) unstable; urgency=high
> >    Changes by Branden Robinson:
> >    * Apply some fixes to the XFree86 X server's ELF object loader:
> >      + Stop conditionally #including <sys/mman.h> since it is unconditionally
> >        imported earlier in the file.
> >      + Perform an mprotect() for AMD64 Linux systems as well as OpenBSD to
>                                   ^^^^^^
>  All 32bit systems that supports the nx feature (as listed in /proc/cpuinfo)
> need this.  I even _think_ some CPUs that support nx but not long-mode
> exist.  (Maybe the Sempron.)

A later patch removed architecture restrictions entirely.

In Debian, now, the XFree86 ELF loader uses mprotect(PROT_EXEC) on all
Linux and OpenBSD systems.

>  Unfortunately, it doesn't help on my Athlon 64 w/ a Radeon AIW vid card.
> This is with a 32bit 2.6.10-rc2-mm4 kernel, compiled with highmem64g (to
> enable third-level pagetable support).  I'm running Sarge, but I upgraded my
> X packages to test this.  (libc6 is still 2.3.2.ds1-18, though.)

I can't imagine why.  It does indeed look like you're SEGVing on the first
module loaded, but the code the way it is currently, I'm having difficulty
figuring out why.

>  My X server log looks like:
[...]
> XFree86 Version 4.3.0.1 (Debian 4.3.0.dfsg.1-9 20041209161604 fabbione@fabbione.net)
[...]
> Module Loader present
[...]
> (II) Loader running on linux
> (II) LoadModule: "bitmap"
> (II) Loading /usr/X11R6/lib/modules/fonts/libbitmap.a
> (II) Module bitmap: vendor="The XFree86 Project"
> 	compiled for 4.3.0.1, module version = 1.0.0
> 	Module class: XFree86 Font Renderer
> 	ABI class: XFree86 Font Renderer, version 0.4
> 
>    *** If unresolved symbols were reported above, they might not
>    *** be the reason for the server aborting.
> 
> Fatal server error:
> Caught signal 11.  Server aborting

Hrmph.

[addressing debian-x now] Is anyone else seeing this?

-- 
G. Branden Robinson                |    Freedom is kind of a hobby with me,
Debian GNU/Linux                   |    and I have disposable income that
branden@debian.org                 |    I'll spend to find out how to get
http://people.debian.org/~branden/ |    people more of it. -- Penn Jillette

Attachment: signature.asc
Description: Digital signature


Reply to: