| To: | Steve Lord <lord@xxxxxxx> |
|---|---|
| Subject: | Re: Q: Filesystem block sizes available? |
| From: | Roy Sigurd Karlsbakk <roy@xxxxxxxxxxxxx> |
| Date: | Tue, 13 Nov 2001 16:19:14 +0100 (CET) |
| Cc: | Seth Mos <knuffie@xxxxxxxxx>, XFS Mailing list <linux-xfs@xxxxxxxxxxx> |
| In-reply-to: | <1005664030.17854.6.camel@xxxxxxxxxxxxxxxxxxxx> |
| Sender: | owner-linux-xfs@xxxxxxxxxxx |
> Reading in Linux does readahead - but I do not think it asks for more > than 32 pages at once. We are using mostly the generic linux code paths > for read unless you use O_DIRECT. ok... I currently use the Tux web server, and cannot change the way it reads. Can I rather patch the kernel source to allow for 64-page readahead? That'll be 256kB chunks, something the SCSI devices will appreciate a lot, and thus making me a happy man :-) -- Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA Computers are like air conditioners. They stop working when you open Windows. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Q: Filesystem block sizes available?, Steve Lord |
|---|---|
| Next by Date: | Re: Q: Filesystem block sizes available?, Eric Sandeen |
| Previous by Thread: | Re: Q: Filesystem block sizes available?, Steve Lord |
| Next by Thread: | Re: Q: Filesystem block sizes available?, Steve Lord |
| Indexes: | [Date] [Thread] [Top] [All Lists] |