xfs
[Top] [All Lists]

Re: process hang with xfs 1.0.1

To: linux-xfs@xxxxxxxxxxx
Subject: Re: process hang with xfs 1.0.1
From: Amit D Chaudhary <amitc@xxxxxxxxxxx>
Date: Mon, 29 Oct 2001 15:20:43 -0800
Cc: Robert Sander <robert.sander@xxxxxxxxxxxxxxx>
References: <3BC7950E.8030807@brocade.com> <news2mail-slrn9sghu1.f9u.gurubert-news@raman.epigenomics.epi> <200110131403.f9DE3QJ01897@jen.americas.sgi.com> <20011013195809.A1617@epigenomics.com> <3BC8CCBE.50903@brocade.com> <20011014153548.A6910@epigenomics.com>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4) Gecko/20010913
Hi,

Just an FYI as it might be useful to those using xfs 1.0.1.

The fix to the problem originally reported by me is to have the mmap_sem
as rw_semaphore and the new rw_semaphore implementation as in 2.4.7. So
though  xfs_read showed up in stack trace during the process hang, this
was more of a mainline kernel fix.
Robert's tip that 2.4.7 fixes it helped.

Thanks and Regards
Amit


Robert Sander wrote:

 > On Sat, Oct 13, 2001 at 04:22:38PM -0700, Amit D Chaudhary wrote:
 >
 >>
 >>Robert Sander wrote:
 >>
 >>
 >>>OK, using 2.4.12 or later was not an option because the aacraid
 >>>patch for the Dell RAID controller was not working.
 >>>
 >>>But with 2.4.7 this particular problem went away now.
 >>>
 >>Did you use xfs 1.0.1 with kernel 2.4.7?
 >>
 >
 > No, I took the patch for 2.4.7 from oss.sgi.com, I think
 > all these patches are "just" minor changes from 1.0.1 to
 > match the current kernel and have some bug fixes in them.
 >
 > Greetings
 >





<Prev in Thread] Current Thread [Next in Thread>