Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*File\s+system\s+block\s+reservation\s+mechanism\s+is\s+broken\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*File\s+system\s+block\s+reservation\s+mechanism\s+is\s+broken\s*$/: 10 ]

Total 10 documents matching your query.

1. File system block reservation mechanism is broken (score: 1): Author: Stephane Doyon <sdoyon@xxxxxxxxx>; Date: Fri, 15 Sep 2006 12:44:16 -0400 (EDT); [Resending. Seems my previous post did not make it somehow...] The mechanism allowing to reserve file system blocks, xfs_reserve_blocks() / XFS_IOC_SET_RESBLKS, appears to have been broken by the pat; /archives/xfs/2006-09/msg00067.html (14,206 bytes)
2. Re: File system block reservation mechanism is broken (score: 1): Author: Shailendra Tripathi <stripathi@xxxxxxxxx>; Date: Mon, 18 Sep 2006 20:32:24 +0530; Hi Stephane, This is the fast path. However, there is slow path where it actually falls back to the earlier mechanism where global lock (spin-lock) is held and then counters are guranteed to be consi; /archives/xfs/2006-09/msg00074.html (16,146 bytes)
3. Re: File system block reservation mechanism is broken (score: 1): Author: Stephane Doyon <sdoyon@xxxxxxxxx>; Date: Mon, 18 Sep 2006 11:24:44 -0400 (EDT); On Mon, 18 Sep 2006, Shailendra Tripathi wrote: Hi Stephane, The code in xfs_reserve_blocks() just locks the superblock and then consults and modifies mp->m_sb.sb_fdblocks. However, if the per-cpu co; /archives/xfs/2006-09/msg00075.html (17,582 bytes)
4. Re: File system block reservation mechanism is broken (score: 1): Author: Shailendra Tripathi <stripathi@xxxxxxxxx>; Date: Tue, 19 Sep 2006 02:17:48 +0530; Stephane Doyon wrote: It assumes the superblock counters are current and accurate, and that they are authoritative... it hasn't been converted to use the new fast path, it always uses the slow path.; /archives/xfs/2006-09/msg00081.html (17,692 bytes)
5. Re: File system block reservation mechanism is broken (score: 1): Author: David Chinner <dgc@xxxxxxx>; Date: Tue, 19 Sep 2006 08:39:18 +1000; Thanks for finding this, Stephane. It turns out our xfsqa test that is supposed to test this feature only tests whether the ioctl succeeds or fails - it doesn't check whether values have been set pro; /archives/xfs/2006-09/msg00084.html (8,917 bytes)
6. File system block reservation mechanism is broken (score: 1): Author: Stephane Doyon <sdoyon@xxxxxxxxx>; Date: Fri, 15 Sep 2006 12:44:16 -0400 (EDT); [Resending. Seems my previous post did not make it somehow...] The mechanism allowing to reserve file system blocks, xfs_reserve_blocks() / XFS_IOC_SET_RESBLKS, appears to have been broken by the pat; /archives/xfs/2006-09/msg00235.html (14,206 bytes)
7. Re: File system block reservation mechanism is broken (score: 1): Author: Shailendra Tripathi <stripathi@xxxxxxxxx>; Date: Mon, 18 Sep 2006 20:32:24 +0530; Hi Stephane, This is the fast path. However, there is slow path where it actually falls back to the earlier mechanism where global lock (spin-lock) is held and then counters are guranteed to be consi; /archives/xfs/2006-09/msg00242.html (16,146 bytes)
8. Re: File system block reservation mechanism is broken (score: 1): Author: Stephane Doyon <sdoyon@xxxxxxxxx>; Date: Mon, 18 Sep 2006 11:24:44 -0400 (EDT); On Mon, 18 Sep 2006, Shailendra Tripathi wrote: Hi Stephane, The code in xfs_reserve_blocks() just locks the superblock and then consults and modifies mp->m_sb.sb_fdblocks. However, if the per-cpu co; /archives/xfs/2006-09/msg00243.html (17,582 bytes)
9. Re: File system block reservation mechanism is broken (score: 1): Author: Shailendra Tripathi <stripathi@xxxxxxxxx>; Date: Tue, 19 Sep 2006 02:17:48 +0530; Stephane Doyon wrote: It assumes the superblock counters are current and accurate, and that they are authoritative... it hasn't been converted to use the new fast path, it always uses the slow path.; /archives/xfs/2006-09/msg00249.html (17,692 bytes)
10. Re: File system block reservation mechanism is broken (score: 1): Author: David Chinner <dgc@xxxxxxx>; Date: Tue, 19 Sep 2006 08:39:18 +1000; Thanks for finding this, Stephane. It turns out our xfsqa test that is supposed to test this feature only tests whether the ioctl succeeds or fails - it doesn't check whether values have been set pro; /archives/xfs/2006-09/msg00252.html (8,917 bytes)

Current List: 1 - 10
Page: [1]