Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*Questions\s+about\s+pagebuf\s+code\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*Questions\s+about\s+pagebuf\s+code\s*$/: 30 ]

Total 30 documents matching your query.

1. Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Sat, 01 May 2004 12:24:16 -0600; 1) Are the versions of xfs in 2.4.26 and 2.6.x mostly the same (except for the change in kernel interface)? In particular are there any differences in the pagebuf code? 2) Are all memory allocations; /archives/xfs/2004-05/msg00001.html (8,172 bytes)
2. Re: Questions about pagebuf code (score: 1): Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>; Date: Sat, 1 May 2004 19:47:09 +0100; there's a bunch of differences, mostly in the handling of the worker threads and the I/O handling code. No. Everything dealing with xfs_buf_t (= mostly metadata, + O_DIRECT data I/O in 2.4) is handle; /archives/xfs/2004-05/msg00002.html (9,847 bytes)
3. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Sat, 01 May 2004 15:21:23 -0600; Thanks for the details. I am trying to debug a problem with file corruption writing to my xfs filesystem (over nfs) when the server is under heavy load (16+ clients writing simultaneously to differen; /archives/xfs/2004-05/msg00003.html (11,237 bytes)
4. Re: Questions about pagebuf code (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Mon, 3 May 2004 14:38:21 +1000; I'd guess your pagesize and fs blocksize are both 4K? If you read between the lines of Christophs mail, pagebuf isn't the right place to be looking for regular file data corruption, and probably the; /archives/xfs/2004-05/msg00006.html (10,219 bytes)
5. Re: Questions about pagebuf code (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Mon, 3 May 2004 16:24:15 +1000; Oh, missed this - setting MAX_SLAB_SIZE to 0 will make all allocations go the vmalloc route... under no circumstances is that what you want to do. By adding {BUG/WARN}_ON checks on irqs_disabled() in; /archives/xfs/2004-05/msg00008.html (9,573 bytes)
6. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Mon, 03 May 2004 11:08:52 -0600; Right. That isn't what I meant. I wanted to set MAX_SLAB_SIZE to something to ensoure only kmalloc is ever used, not vmalloc. I didn't see and references to irqs_disabled() in either 2.6.5 or 2.4.26.; /archives/xfs/2004-05/msg00018.html (10,008 bytes)
7. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Mon, 03 May 2004 11:25:13 -0600; Yes. ia32 page size is 4k, and I didn't change the blocksize (not that XFS would let me). I will look around there. That makes sense. Craig; /archives/xfs/2004-05/msg00020.html (10,710 bytes)
8. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Mon, 03 May 2004 16:01:25 -0600; I cannot say for certain it is, but I am not sure. My writes don't jump around in the file, it is just a series of reads, then writes, and then another series of reads then again writes. However, I d; /archives/xfs/2004-05/msg00023.html (12,914 bytes)
9. Re: Questions about pagebuf code (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Tue, 4 May 2004 13:20:27 +1000; It will -- use the -bsize= option to mkfs, can be any power of 2 between 512 bytes and your page size. cheers. -- Nathan; /archives/xfs/2004-05/msg00033.html (9,671 bytes)
10. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Tue, 04 May 2004 13:29:53 -0600; I ran doublewrite on my server. I bumped up the threads to 20 and was easy able to generate a problem. I wrote a script that submitted 1 doublewrite process to each of 20 nodes, to more mimic what I; /archives/xfs/2004-05/msg00038.html (9,889 bytes)
11. Re: Questions about pagebuf code (score: 1): Author: Russell Cattelan <cattelan@xxxxxxx>; Date: Tue, 04 May 2004 15:29:09 -0500; The test was designed to mimic the out of order writes that nfs often generates. So 20 nfs clients banging on the same file on a server would probably end up corrupting it. The problem still exists I; /archives/xfs/2004-05/msg00039.html (10,971 bytes)
12. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Tue, 04 May 2004 21:49:22 -0600; The doublewrite README indicated that corruption does not happen when a uni-processor kernel is used. I tried that with my code and I still got corruption. Your test is a bit better (smaller, runs l; /archives/xfs/2004-05/msg00044.html (9,995 bytes)
13. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Thu, 06 May 2004 13:05:35 -0600; It appears my corruption pattern is different (but not unrelated). I finally confirmed that when a file is corrupted, it is corrupted with data occurring from another write. For example, Temperature; /archives/xfs/2004-05/msg00057.html (10,594 bytes)
14. Re: Questions about pagebuf code (score: 1): Author: Russell Cattelan <cattelan@xxxxxxx>; Date: Thu, 06 May 2004 14:48:55 -0500; The problem is trying to figure out if the data is stale or really from a different process. If the data block was previously used and then freed either by removing the old file or truncating it and; /archives/xfs/2004-05/msg00060.html (12,646 bytes)
15. Re: Questions about pagebuf code (score: 1): Author: Craig Tierney <ctierney@xxxxxxxx>; Date: Mon, 10 May 2004 09:25:01 -0600; I ran several tests over the weekend where wrote a pattern to the disk (val=pos%256). My tests showed that the pages are just not being written. I see the original pattern, and not values that look; /archives/xfs/2004-05/msg00082.html (12,060 bytes)
16. Questions about pagebuf code (score: 1): Author:; Date: Sat, 01 May 2004 12:24:16 -0600; 1) Are the versions of xfs in 2.4.26 and 2.6.x mostly the same (except for the change in kernel interface)? In particular are there any differences in the pagebuf code? 2) Are all memory allocations; /archives/xfs/2004-05/msg00230.html (8,172 bytes)
17. Re: Questions about pagebuf code (score: 1): Author:; Date: Sat, 1 May 2004 19:47:09 +0100; there's a bunch of differences, mostly in the handling of the worker threads and the I/O handling code. No. Everything dealing with xfs_buf_t (= mostly metadata, + O_DIRECT data I/O in 2.4) is handle; /archives/xfs/2004-05/msg00231.html (9,847 bytes)
18. Re: Questions about pagebuf code (score: 1): Author:; Date: Sat, 01 May 2004 15:21:23 -0600; Thanks for the details. I am trying to debug a problem with file corruption writing to my xfs filesystem (over nfs) when the server is under heavy load (16+ clients writing simultaneously to differen; /archives/xfs/2004-05/msg00232.html (11,237 bytes)
19. Re: Questions about pagebuf code (score: 1): Author:; Date: Mon, 3 May 2004 14:38:21 +1000; I'd guess your pagesize and fs blocksize are both 4K? If you read between the lines of Christophs mail, pagebuf isn't the right place to be looking for regular file data corruption, and probably the; /archives/xfs/2004-05/msg00235.html (10,219 bytes)
20. Re: Questions about pagebuf code (score: 1): Author:; Date: Mon, 3 May 2004 16:24:15 +1000; Oh, missed this - setting MAX_SLAB_SIZE to 0 will make all allocations go the vmalloc route... under no circumstances is that what you want to do. By adding {BUG/WARN}_ON checks on irqs_disabled() in; /archives/xfs/2004-05/msg00237.html (9,573 bytes)

Current List: 1 - 20
Page: [1] [2]