--- Log opened Wed Aug 19 00:00:46 2009
00:27 -!- tgrabiec [n=tomekg@ala59.neoplus.adsl.tpnet.pl] has quit ["Leaving"]
--- Log closed Wed Aug 19 01:25:41 2009
--- Log opened Wed Aug 19 18:48:32 2009
18:48 -!- jato-irc-logger [n=vegard@cm-84.209.125.101.getinternet.no] has joined #jato
18:48 -!- Irssi: #jato: Total of 7 nicks [1 ops, 0 halfops, 0 voices, 6 normal]
18:48 -!- Irssi: Join to #jato was synced in 5 secs
19:03 -!- ahuillet_ [n=ahuillet@14.152.72-86.rev.gaoland.net] has joined #jato
19:19 -!- ahuillet [n=ahuillet@163.169.66-86.rev.gaoland.net] has quit [Read error: 110 (Connection timed out)]
19:19 -!- ahuillet_ is now known as ahuillet
19:32 -!- ahuillet [n=ahuillet@14.152.72-86.rev.gaoland.net] has quit ["ZNC by prozac - http://znc.sourceforge.net"]
19:34 -!- ahuillet [n=ahuillet@14.152.72-86.rev.gaoland.net] has joined #jato
19:40 -!- ahuillet [n=ahuillet@14.152.72-86.rev.gaoland.net] has quit ["ZNC by prozac - http://znc.sourceforge.net"]
19:41 -!- ahuillet [n=ahuillet@14.152.72-86.rev.gaoland.net] has joined #jato
20:48 < vegard> t_grabiec: can I please have again your test-case which shows the broken method resolution?
21:04 < t_grabiec> vegard: sure
21:06 < t_grabiec> vegard: http://pastebin.com/d4dc5cb28
21:15 < vegard> thanks
21:49 < t_grabiec> penberg_home, ahuillet: I've just sent some solution for the empty basic block bug.
21:49 < t_grabiec> If you enjoy this solution, you can merge it
21:49 < t_grabiec> or rework it as you like.
21:49 < t_grabiec> I'll be offline until saturday
21:52 < ahuillet> Data flow resolution code assumed that basic blocks always contain
21:52 < ahuillet> some struct insn, before which spill/reload instructions can be
21:52 < ahuillet> put. When data flow resolution was run on empty basic block, it worked
21:52 < ahuillet> on invalid struct insn pointer and entered infinite loop.
21:52 < ahuillet> reading this makes me feel so stupid
21:52 < ahuillet> it's always my code that's broken
21:53 < ahuillet> ok, sometimes it's penberg_home's too, but that's fair game since he wrote the whole thing
21:53 < t_grabiec> ahuillet: well, regalloc is the most complex part
21:54 < ahuillet> yeah...
21:54 < ahuillet> though it's perfectly normal to write bugs, I think I have a pretty bad bugs/LoC ratio in this project :)
21:54 < penberg_home> t_grabiec: I'll look at the patch in more detail tomorrow
21:54 < penberg_home> t_grabiec: have fun at the mountains! :)
21:55 < t_grabiec> penberg_home: thanks :)
22:40 < vegard> well, I think it's also just that we get a lot of cases which we didn't foresee
23:44 -!- t_grabiec [n=tomekg@apo171.neoplus.adsl.tpnet.pl] has quit ["Leaving"]
--- Log closed Thu Aug 20 00:00:21 2009