Re: Tested the flusher slightly

2013-06-24 Thread OGAWA Hirofumi
Daniel Phillips daniel.raymond.phill...@gmail.com writes: And FLUSH command makes slower slightly. blocksize 512 makes slower, and increase cpu systime, because probably buffer_head overhead, especially sort of buffer_head (perf says buffer_index_cmp() is top cpu user). The performance

Re: Draft to optimize wait_sb_indoes()

2013-06-24 Thread Raymond Jennings
Also I think it's spelled inodes, not indoes If you have this incorrect spelling anywhere in a script or document I suggest fixing it. On Mon, Jun 24, 2013 at 4:06 PM, Daniel Phillips daniel.raymond.phill...@gmail.com wrote: A beautifully clear explanation and big efficiency gain from a tiny

Re: Draft to optimize wait_sb_indoes()

2013-06-24 Thread OGAWA Hirofumi
Raymond Jennings shent...@gmail.com writes: Also I think it's spelled inodes, not indoes Oh, I'll fix. Thanks. -- OGAWA Hirofumi hirof...@mail.parknet.co.jp ___ Tux3 mailing list Tux3@phunq.net http://phunq.net/mailman/listinfo/tux3

Design note: Simplified implementation of free block tags

2013-06-24 Thread Daniel Phillips
Free tags Free tags in Tux3 will perforrm a similar function to Ext2's unallocated block counts. In Ext2, an unallocated block count is a 16 bit field in the group descriptor for each block group. Tux3 does not have group descriptors, but will use for a similar purpose a table of free tags, where