00:07:50 has anyone looked at the file-position bug I recently filed? 00:10:32 hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has joined #sbcl 00:38:27 -!- hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has quit [Quit: hargettp] 00:41:50 hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has joined #sbcl 00:52:38 -!- drewc [~user@S01060013101b6ddb.vc.shawcable.net] has quit [Ping timeout: 245 seconds] 00:56:25 -!- hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has quit [Quit: hargettp] 01:08:07 attila_lendvai [~attila_le@adsl-89-135-206-229.monradsl.monornet.hu] has joined #sbcl 01:43:31 -!- attila_lendvai [~attila_le@adsl-89-135-206-229.monradsl.monornet.hu] has quit [Quit: Leaving.] 01:51:43 -!- Fare [~Fare@ita4fw1.itasoftware.com] has quit [Disconnected by services] 01:51:46 -!- FareWell is now known as Fare 01:52:10 FareWell [~Fare@ita4fw1.itasoftware.com] has joined #sbcl 01:53:41 rbarraud [~rbarraud@118-92-1-3.dsl.dyn.ihug.co.nz] has joined #sbcl 02:43:51 -!- nyef [~nyef@pool-70-109-147-129.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 02:48:39 -!- Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer] 02:48:50 Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has joined #sbcl 08:14:17 Blkt [~user@93-33-133-105.ip44.fastwebnet.it] has joined #sbcl 08:15:01 good morning everyone 08:15:31 -!- Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has quit [Ping timeout: 240 seconds] 08:15:56 Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has joined #sbcl 08:26:24 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 08:26:24 -!- ChanServ has set mode +o nikodemus 08:26:47 -!- `micro [~micro@www.bway.net] has left #sbcl 08:29:02 good morning 08:58:50 stassats [~stassats@wikipedia/stassats] has joined #sbcl 09:17:08 tcr [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has joined #sbcl 10:48:54 attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has joined #sbcl 10:51:18 -!- attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has quit [Client Quit] 11:00:06 Krystof [~csr21@nat65.mia.three.co.uk] has joined #sbcl 11:00:06 -!- ChanServ has set mode +o Krystof 11:44:39 hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has joined #sbcl 12:11:43 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 12:17:10 -!- hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has quit [Quit: hargettp] 12:19:16 hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has joined #sbcl 12:43:26 -!- Krystof [~csr21@nat65.mia.three.co.uk] has quit [Ping timeout: 240 seconds] 12:52:55 nyef [~nyef@pool-70-109-147-129.cncdnh.east.myfairpoint.net] has joined #sbcl 12:53:05 G'morning all. 12:53:38 hi nyef 12:55:39 Krystof [~csr21@nat66.mia.three.co.uk] has joined #sbcl 12:55:39 -!- ChanServ has set mode +o Krystof 13:10:26 -!- Krystof [~csr21@nat66.mia.three.co.uk] has quit [Ping timeout: 240 seconds] 13:11:50 mega1 [~quassel@pool-03811.externet.hu] has joined #sbcl 13:41:53 -!- Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer] 13:43:40 -!- hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has quit [Quit: hargettp] 13:46:52 Kaer [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has joined #sbcl 13:54:22 nikodemus [~nikodemus@cs181058025.pp.htv.fi] has joined #sbcl 13:54:23 -!- ChanServ has set mode +o nikodemus 13:57:59 attila_lendvai [~attila_le@adsl-89-135-206-229.monradsl.monornet.hu] has joined #sbcl 14:05:11 hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has joined #sbcl 14:42:56 -!- mega1 [~quassel@pool-03811.externet.hu] has quit [Ping timeout: 276 seconds] 14:43:01 mega1 [~quassel@pool-00093.externet.hu] has joined #sbcl 15:05:37 -!- froydnj [~froydnj@cpe-173-89-193-139.indy.res.rr.com] has quit [Quit: using sirc version 2.211+ssfe] 15:05:57 froydnj [~froydnj@gateway.codesourcery.com] has joined #sbcl 15:21:42 mega1_ [~quassel@pool-00e97.externet.hu] has joined #sbcl 15:23:00 -!- mega1 [~quassel@pool-00093.externet.hu] has quit [Ping timeout: 272 seconds] 16:01:49 -!- attila_lendvai [~attila_le@adsl-89-135-206-229.monradsl.monornet.hu] has quit [Quit: Leaving.] 16:42:42 -!- mega1_ [~quassel@pool-00e97.externet.hu] has quit [Read error: Operation timed out] 16:47:45 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 17:10:46 Krystof [~csr21@nat65.mia.three.co.uk] has joined #sbcl 17:10:46 -!- ChanServ has set mode +o Krystof 17:22:26 -!- Krystof [~csr21@nat65.mia.three.co.uk] has quit [Ping timeout: 240 seconds] 18:14:28 -!- nikodemus [~nikodemus@cs181058025.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 18:23:05 now to remember the magic incantations to commit stuff to CVS. 18:25:15 Krystof [~csr21@nat76.mia.three.co.uk] has joined #sbcl 18:25:15 -!- ChanServ has set mode +o Krystof 18:31:43 do we have any position on having some compilation units in C++? 18:33:07 in src/runtime? I don't think we've thought about it before 18:33:15 right, for the runtime. 18:37:50 I assume you have a specific thing to convert? 18:38:16 I wanted to import google's hash table code. 18:38:39 and expose it with unmangled symbols. 18:40:18 Rewriting GCish code in C is seriously tedious... OTOH, I'm always wary of introducing C++ in a project. 18:52:47 google's hash table? can't you just do that via alien (tediously)? 18:54:02 not in the GC. 18:54:22 ah 18:54:31 hm. what's the use of it in the gc? 18:55:24 for now, my EQ sxhash hash. 18:56:08 where we use an address-based hash table to map objects to hash values that are independent of GC. 19:07:51 mm... interesting bug. With more constant folding, I'm discovering functions that defknowned as foldable, but aren't defined at compile-time. 19:16:48 kclifton [~kclifton@s198-166-45-245.ab.hsia.telus.net] has joined #sbcl 19:20:02 -!- Fare [~Fare@64.119.159.126] has quit [Quit: Leaving] 19:21:26 -!- Krystof [~csr21@nat76.mia.three.co.uk] has quit [Ping timeout: 240 seconds] 19:39:03 nyef: can you cook a test case up for constant ash of 0? 20:20:34 Krystof [~csr21@nat65.mia.three.co.uk] has joined #sbcl 20:20:34 -!- ChanServ has set mode +o Krystof 20:29:06 -!- hargettp [~anonymous@pool-71-174-128-10.bstnma.east.verizon.net] has quit [Quit: hargettp] 20:30:28 -!- Krystof [~csr21@nat65.mia.three.co.uk] has quit [Ping timeout: 240 seconds] 20:45:46 -!- kclifton [~kclifton@s198-166-45-245.ab.hsia.telus.net] has quit [Quit: kclifton] 20:51:40 Krystof [~csr21@84.93.173.92] has joined #sbcl 20:51:40 -!- ChanServ has set mode +o Krystof 21:08:00 -!- tcr [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has quit [Quit: Leaving.] 21:22:50 -!- Blkt [~user@93-33-133-105.ip44.fastwebnet.it] has quit [Quit: Error: do not makunbound t please!] 21:28:15 pkhuong: It's in the bug report for PPC. Something like (defun foo (x) (declare (type (integer 0 0) x)) (ash x 57)) or so. 22:03:41 attila_lendvai [~attila_le@adsl-89-135-203-148.monradsl.monornet.hu] has joined #sbcl 22:13:55 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 240 seconds] 22:24:01 nyef: and if it compiles successfully, we're good? 22:24:09 guess I could do something with disassemble on x86oids 22:24:30 also, does anyone else think we should swap the order in which we perform rewrites in ir1-optimize-f? 22:24:33 *if 22:24:40 i.e. constant propagate *before* doing if/if conversion 22:25:03 Actually, I don't know if it would fail outright on x86oids. On PPC it's due to the instruction encoding not having space for such a large constant shift. 22:26:04 it just works on x86oids, but I can SEARCH on disassemble. 22:31:22 You could throw an explicit check in the VOP for a fixnum result with a constant shift count over word-width - n-fixnum-tag-bits or similar. 22:44:14 and thoughts on constant propagation before if/if? 22:48:16 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 264 seconds] 22:49:49 rbarraud_ [~rbarraud@118-92-1-3.dsl.dyn.ihug.co.nz] has joined #sbcl 22:50:10 -!- rbarraud [~rbarraud@118-92-1-3.dsl.dyn.ihug.co.nz] has quit [Read error: Connection reset by peer] 22:50:16 -!- rbarraud_ [~rbarraud@118-92-1-3.dsl.dyn.ihug.co.nz] has quit [Read error: Connection reset by peer] 23:04:06 I have no real idea about that part of the compiler. 23:11:23 -!- Krystof [~csr21@84.93.173.92] has quit [Ping timeout: 245 seconds] 23:12:38 seems right to me. if/if conversion duplicates the conditional branch to each node that sets the tested value. 23:13:02 seems to me simplifying ifs to remove the conditional branch first is even better. 23:15:37 I am so badly tempted to split the compiler up into little pieces in an attempt to reduce the surface area for understanding any given piece. 23:16:08 it would be good to disentangle rewriting and analysis. 23:16:58 Rewriting being all the transforms and optimizers, and analysis being the various "compilery" bits? 23:17:11 right. 23:17:40 as it is, we depend on rewriting to perform some analyses (e.g. value numbering or constant propagation) 23:18:09 Joy. 23:18:36 I think it's the intuitive way to do it for lispers 23:18:42 But I'm also thinking of those files which rely on various IR1 or IR2 structures, do something important, but have a /single/ function that is their entry point. 23:18:58 (it's like macros, but iteratively) 23:19:45 nyef: I don't think they're that bad, actually; they tend to have a nicely-defined interface and specific task. 23:19:55 Sure, but why aren't they in their own packages? 23:20:17 sure :) 23:32:16 ... And why don't we have a bindcomp.lisp or something instead of having the dynamic binding crap in ir2tran? 23:33:27 I think putting each IR2 phase or whatever in its own package is a bit of overkill 23:35:54 It might be, but it at the same time it's an explicit statement "this is self-contained". 23:36:04 this is true 23:36:43 One of my concerns here is trying to make it easier for new compiler hackers to find their way around. 23:37:39 (That includes me, as I'm still trying to find my way around IR1 and the various analysis passes.) 23:42:23 -!- attila_lendvai [~attila_le@adsl-89-135-203-148.monradsl.monornet.hu] has quit [Quit: Leaving.] 23:58:05 nyef: the singleton type thingy is evolving into ~4 patches.