| To: | Martin Steigerwald <Martin@xxxxxxxxxxxx> |
|---|---|
| Subject: | Re: md raid1 passes barriers, but xfs doesn't use them? |
| From: | Eric Sandeen <sandeen@xxxxxxxxxxx> |
| Date: | Tue, 24 Jun 2008 08:34:28 -0500 |
| Cc: | linux-xfs@xxxxxxxxxxx |
| In-reply-to: | <200806241131.12112.Martin@lichtvoll.de> |
| References: | <48605A8E.9070903@sandeen.net> (sfid-20080624_084605_814814_B137A507) <200806241131.12112.Martin@lichtvoll.de> |
| Sender: | xfs-bounce@xxxxxxxxxxx |
| User-agent: | Thunderbird 2.0.0.14 (Macintosh/20080421) |
Martin Steigerwald wrote: > Hi Eric! > > Am Dienstag 24 Juni 2008 schrieb Eric Sandeen: >> So md raid1 is happy to pass down any barrier writes that it sees, but >> this bit in xfs_mountfs_check_barriers() at mount time: > > Interesting. Since when? Which part? :) XFS has done the flag check for a very long time... > On my last test also ext3 disabled barries over device mapper / > reiserfs[1]. well but I used it with LVM, maybe its different with RAID > 1. I thought this to be a generic device mapper issue. > > Did you test with ext3 too? This is xfs-specific; ext3 does not look for the queue ordered flag so won't have this problem on md raid1. > [1] http://bugzilla.kernel.org/show_bug.cgi?id=9554 but that's for device-mapper, which has never supported barriers .... I'm talking about md raid1 here (not dm). -Eric > Ciao, |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: md raid1 passes barriers, but xfs doesn't use them?, Martin Steigerwald |
|---|---|
| Next by Date: | Re: md raid1 passes barriers, but xfs doesn't use them?, Martin Steigerwald |
| Previous by Thread: | Re: md raid1 passes barriers, but xfs doesn't use them?, Martin Steigerwald |
| Next by Thread: | Re: md raid1 passes barriers, but xfs doesn't use them?, Martin Steigerwald |
| Indexes: | [Date] [Thread] [Top] [All Lists] |