xfs
[Top] [All Lists]

XSF on Xscale

To: "Eric Sandeen" <sandeen@xxxxxxx>
Subject: XSF on Xscale
From: "Ranslam, Robert E" <robert.e.ranslam@xxxxxxxxx>
Date: Thu, 11 Mar 2004 10:38:44 -0800
Cc: "Steve Lord" <lord@xxxxxxx>, <linux-xfs@xxxxxxxxxxx>, "Nathan Scott" <nathans@xxxxxxx>, "Vinesh Christopher" <vineshc@xxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
Thread-index: AcQHkdq3W32UAZPiQ5u+6drSuBLlowAACzWg
Thread-topic: XSF on Xscale
Changing the Subject to something more relevant

Just sent the patch directly to you. Also saw Vinesh post it

We do have customers that are interested in XFS it seems like all of a
sudden it's the "hot thing". Greg Ungerer ( gerg@xxxxxxxxxxxx) did some
initial testing to help out an ODM that was hacking (weakly) at it.

MontaVista 'claims' to support it.  But they could see the same problem
Greg saw on the IXDP425 - I do not think that it is a "GCC problem" as
the patch adds a variable that was missing in the original code.  But
then Greg deferred "Judgment" to the XFS experts on the calculation.
There could be other issues that ARE tool chain related. I can't really
rule anything out yet as I have personally not done the testing. 

From Vinesh's reply, it looks as if it is not a BE/LE thing - I'm always
concerned about this though as I have had to track down enough issues
related to drivers/firmware assumed LE always.

Also comments from Steve Lord  make me feel better about the LE/BE stuff
I have seem in other systems.

RR




> -----Original Message-----
> From: Eric Sandeen [mailto:sandeen@xxxxxxx] 
> Sent: Thursday, March 11, 2004 9:54 AM
> To: Ranslam, Robert E
> Subject: RE: Bug : XFS - XSCALE "Directory Not Empty"
> 
> 
> Hi Robert - 
> 
> Where was the patch posted?  I have not seen it yet.
> 
> Sorry I was being dense about xscale, I was not familiar with the
> technology.  If Intel is keen to have xfs run properly on the board,
> perhaps we should look into getting some hardware to test on. 
>  While SGI
> probably doesn't have a clear business interest in XFS on 
> XSCALE, I have
> a personal linux-hacker interest in making sure that xfs runs properly
> on all platforms that Linux supports... :)
> 
> Thanks,
> 
> -Eric
> 
> On Thu, 2004-03-11 at 11:39, Ranslam, Robert E wrote:
> > FYI:
> > Its also in the IXDP425 as well as the IQ80321.  These are two
> > completely different boards the only thing in common really is the
> > Xscale core.
> > 
> > Greg Ungerer posted a patch that I echoed.  One problem is that is
> > appears to be and issue with a calculation.  The comment seems to
> > indicate that the variable used should be 'namelen' but instead is
> > 'count'
> > 
> > One thing to consider here - the x86 is Little endian. We 
> are BE on the
> > IXP425
> > 
> > RR
> > 
> > > -----Original Message-----
> > > From: linux-xfs-bounce@xxxxxxxxxxx 
> > > [mailto:linux-xfs-bounce@xxxxxxxxxxx] On Behalf Of Vinesh 
> Christopher
> > > Sent: Thursday, March 11, 2004 7:40 AM
> > > To: 'Nathan Scott'; Vinesh Christopher
> > > Cc: 'linux-xfs@xxxxxxxxxxx'
> > > Subject: RE: Bug : XFS - XSCALE "Directory Not Empty"
> > > 
> > > 
> > > It works on x86 platforms for me. The problem is 
> > > With the XSCALE board.
> > > 
> > > -----Original Message-----
> > > From: Nathan Scott [mailto:nathans@xxxxxxx] 
> > > Sent: Wednesday, March 10, 2004 8:11 PM
> > > To: Vinesh Christopher
> > > Cc: 'linux-xfs@xxxxxxxxxxx'
> > > Subject: Re: Bug : XFS - XSCALE "Directory Not Empty"
> > > 
> > > On Wed, Mar 10, 2004 at 02:14:58PM -0500, Vinesh 
> Christopher wrote:
> > > >  
> > > > Intel IQ80321 (Xscale)  Evaluation Board  
> > > > Running Linux 2.6.0 with -rmk2 patches
> > > > xfsprogs_2.6.3-1  is taken from debian
> > > >  
> > > > I did the following
> > > >  
> > > > # mkfs.xfs /dev/sda1
> > > > # mount /dev/sda1 /mnt
> > > > # cd /mnt
> > > > # mkdir t
> > > > # cp /lib/* t
> > > > # rm -r -f t
> > > > rm: cannot remove directory 't' : Directory not empty
> > > > #
> > > 
> > > Any errors on your console or in your system logs?
> > > 
> > > I tried this locally and (not really surprisingly) it works
> > > just fine on my system.
> > > 
> > > > I tried with 2.4.21 and XFS 1.3.1 patches -> same problem
> > > 
> > > Uhrm, odd - there are lots of people using XFS from that kernel.
> > > I'm inclined to suspect your hardware at this stage.
> > > 
> > > cheers.
> > > 
> > > -- 
> > > Nathan
> > > 
> > > 
> > > 
> -- 
> Eric Sandeen      [C]XFS for Linux   http://oss.sgi.com/projects/xfs
> sandeen@xxxxxxx   SGI, Inc.          651-683-3102
> 
> 


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