| To: | Eric Sandeen <sandeen@xxxxxxx> |
|---|---|
| Subject: | Re: [Bug 324] New: oops with xfs_fsr on 2.6.5-1.326 |
| From: | Steve Lord <lord@xxxxxxx> |
| Date: | Wed, 21 Apr 2004 10:43:53 -0500 |
| Cc: | Matteo Centonza <matteo@xxxxxx>, Chris Wedgwood <cw@xxxxxxxx>, bugzilla-daemon@xxxxxxxxxxx, xfs-master@xxxxxxxxxxx |
| In-reply-to: | <1082560958.15446.1.camel@xxxxxxxxxxxxxxxxxxxxxx> |
| References: | <Pine.LNX.4.44.0404210956220.17402-100000@xxxxxxxxxxxxx> <1082560958.15446.1.camel@xxxxxxxxxxxxxxxxxxxxxx> |
| Sender: | linux-xfs-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6b) Gecko/20031205 Thunderbird/0.4 |
Eric Sandeen wrote: > On Wed, 2004-04-21 at 04:08, Matteo Centonza wrote: > > >>i haven't my laptop handy but seems not the case since RH backed it out >>starting from 2.6.5-1.309. >> >>Thanks, >> >>-m > > > it seems to be in as of 1.332.... > > -Eric > The problem here is that we are not just talking about xfs stack usage, we are talking about the complete I/O stack in some cases. So if you are using a combination of XFS, LVM, and some driver which is stack hungry, then all bets are off without lots of work - and not just in XFS. I have actually been running Linus's tree with 4K stacks on my XFS only laptop for a week or so without problems. I build kernels, I use bitkeeper etc. Probably the fact that I am just using a simple ide drive is the saving grace here. Steve |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [Bug 324] New: oops with xfs_fsr on 2.6.5-1.326, Eric Sandeen |
|---|---|
| Next by Date: | Problem with writing data to xfs+nfs server under heavy load, Craig Tierney |
| Previous by Thread: | Re: [Bug 324] New: oops with xfs_fsr on 2.6.5-1.326, Eric Sandeen |
| Next by Thread: | Re: [Bug 324] New: oops with xfs_fsr on 2.6.5-1.326, Matteo Centonza |
| Indexes: | [Date] [Thread] [Top] [All Lists] |