2008-09-22 01:48 <bh> yo
2008-09-22 04:13 <flipz>   7/3: 1000033 => 777; 1000044 => 888; 1000099 => 999;
2008-09-22 04:13 <flipz>   8/3: 3000044 => 444 666; 3000055 => 555; 3000000 => <corrupt>; <- result of first dwak_pack attempt
2008-09-22 04:13 <flipz> not bad, it actually added an entry
2008-09-22 04:13 <flipz> messed it up, but.
2008-09-22 04:14 <flipz> but bug hunt commences
2008-09-22 04:21 <flipz>   8/2: 3000044 => 444 666; 3000055 => 555 123 456; <- 3rd try = good
2008-09-22 04:21 <flipz> now some more broken cases
2008-09-22 04:25 <flipz> 3000044 => 444 666; 3000055 => 555; 3000056 =>; <- 4th attempt, hmm seem to be missing something
2008-09-22 04:33 <flipz>   8/3: 3000044 => 444 666; 3000055 => 555; 3000056 => 123; <- result of fixing an off by one
2008-09-22 04:33 <flipz> next boundary: add a new group
2008-09-22 04:37 <flipz>   9/1: 4000000 =>; <- not bad for a first attempt
2008-09-22 04:37 <flipz> also overwrote the 0th extent of the leaf with the new group descriptor ;)
2008-09-22 04:46 <flipz>   9/1: 4000123 => 123 0 0 0; <- closer
2008-09-22 04:46 <flipz> few extra zeros came from somewhere
2008-09-22 04:46 <flipz> hmm
2008-09-22 04:47 <flipz> ah, walk extend base needs to be bumpbed for the new group
2008-09-22 04:47 <flipz> extent base
2008-09-22 04:48 <flipz> ...by the group count of the current group
2008-09-22 04:49 <flipz> err, no
2008-09-22 04:49 <flipz> by the amount of the most recent entry limit
2008-09-22 04:50 <flipz>   9/1: 4000123 => 123; <- correct
2008-09-22 04:50 <flipz> this is too easy
2008-09-22 04:50 <flipz> you lamers who didn't rise to my sunday afternoon challenge should hang your heads ;)
2008-09-22 04:50 <flipz> going to be a funny lkml post about this
2008-09-22 04:52 <flipz> couple more boundaries to check
2008-09-22 04:52 <flipz> next one: group count overflow
2008-09-22 04:55 <flipz>   8/7: 3000044 => 444 666; 3000055 => 555; 3001001 => 1; 3001002 => 1; 3001003 => 1; 3001004 => 1; 3001005 => 1;
2008-09-22 04:55 <flipz>   9/1: 3001006 => 1; <- works great
2008-09-22 04:55 <flipz> first time
2008-09-22 04:55 <flipz> now what?
2008-09-22 04:55 <flipz> got to be more
2008-09-22 04:55 <flipz> dwalk_pack looks too simple
2008-09-22 04:57 <flipz> need to be able to add stuff in the middle of a dleaf, not just at the end I suppose
2008-09-22 04:57 <flipz> though for now we can just re-append everything after the add point
2008-09-22 04:57 <flipz> easy
2008-09-22 04:57 <flipz> will serve for some time
2008-09-22 05:07 <flipz> I know, I'll add some asserts on leaf full
2008-09-22 05:07 <flipz> though of course it will never happen ;)
2008-09-22 05:18 <flipz> ok, all properly and anally asserted
2008-09-22 05:18 <flipz> now just about time to write dwalk_mock
2008-09-22 05:18 <flipz> going to be the cute+funny subject of the lkml post
2008-09-22 05:19 <flipz> hmm, maybe time for a checkin
2008-09-22 05:24 <flipz> final score: 9 lines of the dwalk_pack prototype survived, 4 were changed, 12 were added not counting comments and asserts
2008-09-22 05:27 <flipz> ok, dwalk_mock
2008-09-22 05:32 <flipz> http://www.linuxtoday.com/infrastructure/2008092200135OSCY <- LOL
2008-09-22 05:32 <flipz> really
2008-09-22 05:33 <flipz> "Microsoft isn't the answer. Microsoft is the question. 'Linux' or 'No' is the answer." -- some wag
2008-09-22 05:33 <flipz> where is everybody?
2008-09-22 05:34 <flipz> it's only 5 in the morning
2008-09-22 05:34 <flipz> wimps
2008-09-22 06:10 <shapor> 
ACTION is alive

2008-09-22 07:09 -!- hirofumi(~hirofumi@210.171.168.39) has joined #tux3
2008-09-22 07:14 <flipz> ok, dwalk_mock is a fait accompli
2008-09-22 07:15 <flipz> almost within striking zone of putting this together in inode.c
2008-09-22 07:15 <flipz> need to think about extenty implications now
2008-09-22 07:15 <flipz> extents just about here :)
2008-09-22 07:15 <flipz> => let the benchmark wars begin
2008-09-22 07:35 -!- hirofumi(~hirofumi@210.171.168.39) has joined #tux3
2008-09-22 07:36 -!- ceatinge(~ceatinge@veryclever.net) has joined #tux3
2008-09-22 09:41 -!- RazvanM(~RazvanM@dazzler.isi.jhu.edu) has joined #tux3
2008-09-22 09:50 -!- nataliep_(~nataliep@207.47.98.129.static.nextweb.net) has joined #tux3
2008-09-22 10:10 -!- MaZe(~MaZe@216-239-45-4.google.com) has joined #tux3
2008-09-22 10:15 <flipz> ok, prolly about time to try dropping these gizmos into inode.c
2008-09-22 10:15 <flipz> and see if they make extents happen
2008-09-22 10:16 <flipz> I wonder if we need extent.c
2008-09-22 11:24 <flipz> we got filemap.c instead
2008-09-22 11:24 <flipz> extents are a detail
2008-09-22 12:30 -!- MaZe(~MaZe@216-239-45-4.google.com) has joined #tux3
2008-09-22 12:44 -!- nataliep_(~nataliep@207.47.98.129.static.nextweb.net) has joined #tux3
2008-09-22 13:29 <bh> hey flipz 
2008-09-22 13:30 <flipz> hi
2008-09-22 13:30 <bh> got extents working or something like that ?
2008-09-22 13:37 <flipz> check out the code and try it
2008-09-22 14:18 <bh> I'll do so a bit later. I've been reading a bit of the code online
2008-09-22 14:18 <bh> flipz: you know there's a #linuxfs channel, right ?
2008-09-22 14:24 <flipz> with clueful stuff happening?
2008-09-22 15:39 -!- MaZe(~MaZe@216-239-45-4.google.com) has joined #tux3
2008-09-22 15:40 -!- openblast(~quassel@static.230.173.47.78.clients.your-server.de) has left #tux3
2008-09-22 16:47 -!- ajonat(~ajonat@190.48.103.186) has joined #tux3
2008-09-22 17:29 <flipz> ok, here we go, final push for an extents prototype
2008-09-22 20:41 -!- RazvanM(~RazvanM@pool-151-196-118-156.balt.east.verizon.net) has joined #tux3
2008-09-22 21:12 -!- tim_dimm(~timothyhu@cpe-76-90-98-247.socal.res.rr.com) has joined #tux3
2008-09-22 21:39 -!- Bushman(~marcin@c-76-23-106-132.hsd1.sc.comcast.net) has joined #tux3
2008-09-22 21:50 -!- hirofumi(~hirofumi@210.171.168.39) has joined #tux3