Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[RFC\s+0\/3\]\s+Convert\s+XFS\s+inode\s+hashes\s+to\s+radix\s+trees\s*$/: 20 ]

Total 20 documents matching your query.

1. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: Shailendra Tripathi <stripathi@xxxxxxxxx>
Date: Tue, 14 Nov 2006 17:09:03 -0800
Hi David, I regret for making comments and questions on this quite late (somehow I missed to email). It does appear to me that using this approach can potentially help in cluster hash list related ma
/archives/xfs/2006-11/msg00077.html (12,875 bytes)

2. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Mon, 20 Nov 2006 13:13:28 +1100
Yes, there is less parallelism in the radix tree approach, as I stated in the original description. Sure, but tuning for specsfs is not the problem we are trying to solve here. The problem we are sol
/archives/xfs/2006-11/msg00106.html (11,700 bytes)

3. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: >
Date: Tue, 14 Nov 2006 17:09:03 -0800
Hi David, I regret for making comments and questions on this quite late (somehow I missed to email). It does appear to me that using this approach can potentially help in cluster hash list related ma
/archives/xfs/2006-11/msg00285.html (12,875 bytes)

4. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: >
Date: Mon, 20 Nov 2006 13:13:28 +1100
Yes, there is less parallelism in the radix tree approach, as I stated in the original description. Sure, but tuning for specsfs is not the problem we are trying to solve here. The problem we are sol
/archives/xfs/2006-11/msg00314.html (11,700 bytes)

5. [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Tue, 3 Oct 2006 16:06:10 +1000
One of the long standing problems with XFS on large machines and filesystems is the sizing of the inode cache hashes used by XFS to index the xfs_inode_t structures. The mount option ihashsize became
/archives/xfs/2006-10/msg00015.html (10,636 bytes)

6. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: Chris Wedgwood <cw@xxxxxxxx>
Date: Tue, 3 Oct 2006 14:23:35 -0700
commit afefdbb28a0a2af689926c30b94a14aea6036719 tree 6ee500575cac928cd90045bcf5b691cf2b8daa09 parent 1d32849b14bc8792e6f35ab27dd990d74b16126c author David Howells <dhowells@xxxxxxxxxx> 1159863226 -07
/archives/xfs/2006-10/msg00026.html (9,602 bytes)

7. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Wed, 4 Oct 2006 08:22:56 +1000
That's a good question. In a recent thread on linux-fsdevel about these patches Christoph Hellwig pointed out that 32bit user space is not ready for 64 bit inodes, so it's probably going to be a whil
/archives/xfs/2006-10/msg00027.html (12,028 bytes)

8. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: Chris Wedgwood <cw@xxxxxxxx>
Date: Tue, 3 Oct 2006 17:47:08 -0700
yes a patch changing struct kstat and filldir* was merged... so that leaves NFS3+ is it really worth the pain then?
/archives/xfs/2006-10/msg00032.html (9,935 bytes)

9. Re: [RFC 0/3] Convert XFS inode hashes to radix trees (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Wed, 4 Oct 2006 11:43:34 +1000
Ahhh.... I think I misread what Chris wrote here - _32_ bit inodes on 32 bit platforms not working? I can't see how this would be the case with the mods I posted given that they are entirely internal
/archives/xfs/2006-10/msg00034.html (11,474 bytes)

10. failed (ScanMail has removed a file) (score: 1)
Author: xxxxxxxxxx>
Date: 04 Oct 2006 19:59:15 +0200
One issue is that people often still run a lot of 32bit userland even with 64bit kernels. The compat layer will just truncate the inodes I think. But so far I haven't heard of anybody complaining on
/archives/xfs/2006-10/msg00044.html (9,288 bytes)

11. XFS inode hashes to radix trees (score: 1)
Author: xxxxxxxxxx>
Date: Wed, 04 Oct 2006 15:22:19 -0400
Which completely ignored the fact that NFS systems are already having to truncate 64-bit inode numbers to 32-bits and pass these truncated values up to userspace. Collisions have been observed in the
/archives/xfs/2006-10/msg00046.html (11,099 bytes)

12. ABLE (score: 1)
Author: ter Grandi)
Date: Thu, 5 Oct 2006 10:37:12 +1000
Which is one of the reasons why XFS uses 32 bit inodes by default even on 64 bit kernels. XFS does not use 64 bit inodes unless you tell it to via the inode64 mount option.... Cheers, Dave. -- Dave C
/archives/xfs/2006-10/msg00050.html (9,697 bytes)

13. ion (score: 1)
Author: ner <dgc@xxxxxxx>
Date: Tue, 3 Oct 2006 16:06:10 +1000
l
/archives/xfs/2006-10/msg00392.html (10,636 bytes)

14. Re: Long sleep with i_mutex in xfs_flush_device(), affects NFS service, was: Re: several messages (score: 1)
Author: <melvinsamson@xxxxxxx>
Date: Tue, 3 Oct 2006 14:23:35 -0700
me
/archives/xfs/2006-10/msg00403.html (9,602 bytes)

15. ects NFS service, was: Re: several messages (score: 1)
Author: up6@xxxxxxxxxxx>
Date: Wed, 4 Oct 2006 08:22:56 +1000
od
/archives/xfs/2006-10/msg00404.html (12,028 bytes)

16. om.m (score: 1)
Author: @xxxxxxxxxx>
Date: Tue, 3 Oct 2006 17:47:08 -0700
lem in need of fixing: when writing a 5GB file over NFS to an XFS file system and hitting ENOSPC, it takes on the order of 22hours before m
/archives/xfs/2006-10/msg00409.html (9,935 bytes)

17. pin (score: 1)
Author: ll.com" <novedades-bolivia@xxxxxxxxxxxxxxxxx>
Date: Wed, 4 Oct 2006 11:43:34 +1000
...] -- Forwarded messa
/archives/xfs/2006-10/msg00411.html (11,474 bytes)

18. e: Error (score: 1)
Author: >
Date: 04 Oct 2006 19:59:15 +0200
T
/archives/xfs/2006-10/msg00421.html (9,288 bytes)

19. ¾Ð»Ð»ÐµÐºÑ‚ивно-договорное регулирование (score: 1)
Author: Kleen <ak@xxxxxxx>
Date: Wed, 04 Oct 2006 15:22:19 -0400
.6
/archives/xfs/2006-10/msg00423.html (11,099 bytes)

20. de hashes to radix trees (score: 1)
Author: king@xxxxxxxxxx>
Date: Thu, 5 Oct 2006 10:37:12 +1000
is
/archives/xfs/2006-10/msg00427.html (9,697 bytes)


This search system is powered by Namazu