| To: | david@xxxxxxxxxxxxx |
|---|---|
| Subject: | protecting per-ag access in the sync path |
| From: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
| Date: | Tue, 24 Nov 2009 12:04:55 -0500 |
| Cc: | xfs@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.19 (2009-01-05) |
Since Linux 2.6.29 we use the perag structures a lot in the sync path, but we're not actually protecting against the re-allocation in growfs. I've recently received a new, fast SSD which allows me to hit those races. With it 104 reproducibly crashes the system, hitting the slab redzoning for the old per-ag structure. I experimentally put synchronization using m_peraglock into xfs_inode_ag_walk which fixes the issue at the cost of lock order reversals between the i_lock and the m_peraglock in various places. Any better idea how to protect the new sync path against growfs? |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | consistency checks, Rick Meertens |
|---|---|
| Next by Date: | [PATCH v2 2/7] xfs: simplify xfs_buf_get / xfs_buf_read interfaces, Christoph Hellwig |
| Previous by Thread: | consistency checks, Rick Meertens |
| Next by Thread: | Re: protecting per-ag access in the sync path, Dave Chinner |
| Indexes: | [Date] [Thread] [Top] [All Lists] |