| To: | Alex Elder <aelder@xxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 02/11] xfs: remove the unlock argument to xfs_buf_delwri_queue |
| From: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
| Date: | Thu, 1 Sep 2011 17:39:57 -0400 |
| Cc: | Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx |
| In-reply-to: | <1314904924.2903.60.camel@doink> |
| References: | <20110823082802.335389799@xxxxxxxxxxxxxxxxxxxxxx> <20110823082912.127871770@xxxxxxxxxxxxxxxxxxxxxx> <1314904924.2903.60.camel@doink> |
| User-agent: | Mutt/1.5.21 (2010-09-15) |
On Thu, Sep 01, 2011 at 02:22:04PM -0500, Alex Elder wrote: > On Tue, 2011-08-23 at 04:28 -0400, Christoph Hellwig wrote: > > We can just unlock the buffer in the caller, and the decrement of b_hold > > would also be needed in the !unlock, we just never hit that case currently > > given that the caller handles that case. > > More specifically, the only way we'd hit that case would > involve an unqueued buffer (in xfs_buf_unlock()) getting > queued before bt_delwrite_lock could be acquired (in > xfs_buf_delwri_queue()). But that can't happen because > the buffer is locked the entire time between the check > in xfs_buf_unlock() and the one in xfs_buf_delwri_queue(). > (Right?) Exactly. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH 11/11] xfs: remove i_iocount, Alex Elder |
|---|---|
| Next by Date: | Re: [PATCH 04/11] xfs: call xfs_buf_delwri_queue directly, Christoph Hellwig |
| Previous by Thread: | Re: [PATCH 02/11] xfs: remove the unlock argument to xfs_buf_delwri_queue, Alex Elder |
| Next by Thread: | Re: [PATCH 03/11] xfs: move more delwri setup into xfs_buf_delwri_queue, Alex Elder |
| Indexes: | [Date] [Thread] [Top] [All Lists] |