SCIENTIFIC-LINUX-USERS Archives

January 2009

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Stephan Wiesand <[log in to unmask]>
Reply To:
Stephan Wiesand <[log in to unmask]>
Date:
Wed, 28 Jan 2009 16:51:36 +0100
Content-Type:
TEXT/PLAIN
Parts/Attachments:
TEXT/PLAIN (53 lines)
Hi Troy,

On Wed, 28 Jan 2009, Troy Dawson wrote:

> Dr Andrew C Aitchison wrote:
>> On Tue, 27 Jan 2009, Troy Dawson wrote:
[...]
>>> NOTE:  I haven't gotten XFS (the file system) to compile yet for it.

Removing the now conflicting definitions from the xfs header seems correct 
to me. Something like adding

grep -q 'BH_Unwritten' %{kernel_src_path}/include/linux/buffer_head.h && \
     sed -i '/^BUFFER_FNS.*unwritten/d' linux-2.6/xfs_linux.h

to %prep. It's hard to fix it with a patch without breaking builds for 
the older kernels. An SRPM with this hack is

   http://www-zeuthen.desy.de/~wiesand/SL5/xfs-0.4-2.sl5.src.rpm

It builds and survives light testing with the new kernel.

[...]
>>>        yum --enablerepo=sl-testing update kernel\*
>> 
>> --> Processing Conflict: kernel conflicts ecryptfs-utils < 44
>> 
>> (only a problem if ecryptfs-utils is installed ...)
>> 
> I now have the newer ecryptfs-utils (and ecryptfs-utils-devel) in the testing 
> area.

There's a similar issue with cpuspeed. The new conflict doesn't work 
because they missed that cpuspeed has epoch=1, but things may still break 
if that isn't updated along with the kernel. It probably only affects 
AMD CPUs only.

Any ETA for 5.3? ;-)

- Stephan

> You might have to do a "yum --enablerepo=sl-testing clean all" but it should 
> work now.
> Thanks for testing
> Troy
>

-- 
Stephan Wiesand
   DESY - DV -
   Platanenallee 6
   15738 Zeuthen, Germany

ATOM RSS1 RSS2