Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*Introduce\s+SEEK_DATA\/SEEK_HOLE\s+to\s+XFS\s+V5\s*$/: 20 ]

Total 20 documents matching your query.

1. Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Fri, 06 Jan 2012 21:28:58 +0800
Hello, This is a revised patch according to Christoph's comments at V4. Changes to V5: -- * Revise xfs_has_unwritten_buffer() to lookup pages match tag. * For unwritten extents, in both xfs_seek_data
/archives/xfs/2012-01/msg00067.html (28,863 bytes)

2. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Ben Myers <bpm@xxxxxxx>
Date: Tue, 10 Jan 2012 11:18:55 -0600
Hey Jeff, I got caught up on versions 1-4, and am looking at this now. Thanks for your excellent contribution! I have some inital comments here, and then I'll get into it a little more deeply. You sh
/archives/xfs/2012-01/msg00093.html (35,738 bytes)

3. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Wed, 11 Jan 2012 13:45:09 +0800
Hi Ben, Thanks for your timely response and pointing this out! I am suffering with a data loss when improving seek copy tester to produce a file with around thousands of extents these days, looks the
/archives/xfs/2012-01/msg00099.html (39,151 bytes)

4. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Ben Myers <bpm@xxxxxxx>
Date: Wed, 11 Jan 2012 09:43:18 -0600
Hey Jeff, Here are a few additional minor comments from yesterday. I'm looking forward to seeing your next version, and I'm still working through this one. I would like to suggest that you split this
/archives/xfs/2012-01/msg00119.html (36,611 bytes)

5. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Mark Tinguely <tinguely@xxxxxxx>
Date: Wed, 11 Jan 2012 15:06:09 -0600
Good work. I know this is an important addition. The test in the for loop of the xfs_seek_data() and xfs_seek_hole() routines, are independent of each other except in order of them being called. One
/archives/xfs/2012-01/msg00126.html (11,664 bytes)

6. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Mark Tinguely <tinguely@xxxxxxx>
Date: Wed, 11 Jan 2012 15:07:29 -0600
xfs_bmapi_read() returns the br_state == XFS_EXT_NORM for a hole. There are a couple places that a hole can trigger a data test. BTW, I could not generate a large enough hole that xfs_bmapi_read() wo
/archives/xfs/2012-01/msg00127.html (11,472 bytes)

7. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Mark Tinguely <tinguely@xxxxxxx>
Date: Wed, 11 Jan 2012 15:12:00 -0600
xfs_has_unwritten_buffer() always returns the offset of the first dirty unwritten page. This can cause xfs_seek_data() and xfs_seek_hole() to give the wrong results in certain circumstances. In xfs_s
/archives/xfs/2012-01/msg00129.html (11,717 bytes)

8. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Ben Myers <bpm@xxxxxxx>
Date: Wed, 11 Jan 2012 16:28:16 -0600
Hey Jeff, Ok, since I'm the jackass who is asking you to do the extra work I'll try to be of assistance. Understand that at this point I'm trying to make sure that I understand your code fully. I'm n
/archives/xfs/2012-01/msg00135.html (14,717 bytes)

9. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Thu, 12 Jan 2012 20:53:29 +0800
Hi Ben, Thank you for pointing this out! ok. I will take care the same issues below too. ok. :) Definitely! I have gone through other file systems have SEEK_XXX stuff support, OCFS2 returns -EINVAL i
/archives/xfs/2012-01/msg00149.html (40,745 bytes)

10. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Thu, 12 Jan 2012 21:21:49 +0800
Hi Ben, Thanks a lot for your so much detailed info! Do you means I only need to post a patch to treat unwritten extents as data next time, and then try to work out another patch for probing unwritte
/archives/xfs/2012-01/msg00151.html (18,530 bytes)

11. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Thu, 12 Jan 2012 21:29:33 +0800
Hi Mark, Thanks for your comments! Yes, this is key point I have missed before. Ok, I'll add !isnullstartblock() test for normal extents test. looks this issue is caused by missing hole test for exte
/archives/xfs/2012-01/msg00152.html (13,825 bytes)

12. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Thu, 12 Jan 2012 21:52:54 +0800
Hi Mark, Sorry, am was well understood your opinions in this point for now. IMHO, we can only find and return the data buffer offset at a dirty or unwritten page once the first page was probed. Hmm,
/archives/xfs/2012-01/msg00153.html (14,999 bytes)

13. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Mark Tinguely <tinguely@xxxxxxx>
Date: Thu, 12 Jan 2012 09:01:48 -0600
xfs_has_unwritten_buffer() always returns the offset of the first dirty unwritten page. This can cause xfs_seek_data() and xfs_seek_hole() to give the wrong results in certain circumstances. Sorry,
/archives/xfs/2012-01/msg00156.html (14,751 bytes)

14. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 12 Jan 2012 11:22:10 -0500
With all the complications that we got compared to the initial version, namely multiple hole extents, dirty unwritten extent detection and so on I think it's time to stop using xfs_bmapi_read against
/archives/xfs/2012-01/msg00160.html (10,492 bytes)

15. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 12 Jan 2012 11:39:43 -0500
We've been through this before, you need to overflow the 32-bit extent length counter to get there. Jeff, did you manage to create a test case for that particular scenario?
/archives/xfs/2012-01/msg00161.html (10,396 bytes)

16. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 12 Jan 2012 11:41:48 -0500
Are you testing on ia64 with 64k blocks? :) xfs_bmapi_read will find holes down to block granularity, that's how it's implemented. However recent XFS does fairly aggressive preallocation, so you prob
/archives/xfs/2012-01/msg00162.html (10,722 bytes)

17. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Ben Myers <bpm@xxxxxxx>
Date: Thu, 12 Jan 2012 11:39:59 -0600
^^^^ it is speculative delay in this case, I think. -Ben
/archives/xfs/2012-01/msg00168.html (11,013 bytes)

18. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Ben Myers <bpm@xxxxxxx>
Date: Thu, 12 Jan 2012 11:50:16 -0600
Hey, c) having a cursor here means that Jeff can always get the job done with a single btree search, which could be an important optimisation for heavily preallocated workloads. I agree that this is
/archives/xfs/2012-01/msg00169.html (11,354 bytes)

19. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Fri, 13 Jan 2012 10:14:57 +0800
Ok, I'll try it out. Thanks, -Jeff
/archives/xfs/2012-01/msg00188.html (11,483 bytes)

20. Re: Introduce SEEK_DATA/SEEK_HOLE to XFS V5 (score: 1)
Author: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Fri, 13 Jan 2012 10:41:52 +0800
Good catch! Looks we need to examine the max value between seekoff and For SEEK_DATA, if the seekoff is less than map->br_startoff, IMHO, we need to pass the map->br_startoff to it. By combining with
/archives/xfs/2012-01/msg00189.html (15,941 bytes)


This search system is powered by Namazu