00:41:39 -!- nikodemus_ [~nikodemus@cs181058025.pp.htv.fi] has quit [Ping timeout: 258 seconds] 00:51:08 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 00:51:08 -!- ChanServ has set mode +o nikodemus 00:56:25 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 01:17:41 hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has joined #sbcl 01:33:33 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 01:36:46 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 246 seconds] 01:39:01 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 02:07:22 -!- hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has quit [Quit: Leaving...] 03:49:54 mm. how hard do you think it would be to compile each function twice, the second time with a specialised calling convention (e.g. closed over values as arguments as well and everything pre-parsed)? 04:09:24 -!- antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has quit [Remote host closed the connection] 05:04:24 slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has joined #sbcl 07:45:48 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 08:07:14 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 08:07:14 -!- ChanServ has set mode +o nikodemus 08:45:46 each function? Dunno. It would also be fun to have some specialized entry points for passing unboxed (numerical) arguments 08:46:04 of course, we're in danger of reinventing block compilation 09:22:21 pkhuong: istr that jsnell was looking at unboxed arguments at the time of sbcl10 09:23:21 i should probably take another look at fixing block compilation some day 09:41:50 i have a suggestion for a low-overhead way of increasing their sbcl commit count for anyone who has been slacking :) 09:43:12 i can make a please-review branch on github, and push things which i think are OK but would rather someone else took look at as well 09:52:26 any takers? 10:03:32 -!- homie` [~levent.gu@xdsl-78-35-139-201.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 10:11:08 homie [~levent.gu@xdsl-78-35-139-201.netcologne.de] has joined #sbcl 10:28:22 hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has joined #sbcl 10:38:30 -!- homie [~levent.gu@xdsl-78-35-139-201.netcologne.de] has quit [Read error: Connection reset by peer] 10:39:46 homie [~levent.gu@xdsl-78-35-158-176.netcologne.de] has joined #sbcl 11:17:50 -!- cmm [~cmm@109.67.199.173] has quit [Ping timeout: 276 seconds] 11:18:18 cmm [~cmm@109.67.199.173] has joined #sbcl 11:57:49 -!- hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has quit [Quit: Leaving...] 12:03:45 hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has joined #sbcl 12:42:49 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Ping timeout: 248 seconds] 12:57:55 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 13:02:57 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 13:29:55 tcr1: https://github.com/nikodemus/sbcl/tree/end-of-the-world 13:30:02 does that work for you? 13:48:44 loke [~elias@bb116-14-102-156.singnet.com.sg] has joined #sbcl 13:57:17 specialised entry points are actually doable as is 13:57:27 I'd like to have exit points as well ;) 14:08:05 -!- pchrist_ [~spirit@gentoo/developer/pchrist] has quit [Quit: leaving] 14:08:58 pchrist [~spirit@gentoo/developer/pchrist] has joined #sbcl 14:29:49 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 14:53:20 -!- cmm [~cmm@109.67.199.173] has quit [Ping timeout: 260 seconds] 14:53:53 cmm [~cmm@109.67.199.173] has joined #sbcl 15:14:47 pkhuong: your type widening stuff has been committed, right? 15:14:52 no. 15:14:59 oh, ok 15:15:12 (trying to clean stale branches from my tree) 15:15:33 git branch | wc -l says ... 110 15:16:22 -!- stassats` [~stassats@wikipedia/stassats] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 15:34:04 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 16:21:54 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has left #sbcl 16:27:15 -!- slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has quit [Ping timeout: 260 seconds] 16:34:11 slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has joined #sbcl 16:41:21 -!- slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has quit [Ping timeout: 252 seconds] 16:46:36 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 16:46:36 -!- ChanServ has set mode +o nikodemus 17:46:48 slyrus_ [~chatzilla@adsl-99-173-27-219.dsl.pltn13.sbcglobal.net] has joined #sbcl 18:03:40 -!- slyrus_ [~chatzilla@adsl-99-173-27-219.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 240 seconds] 18:12:00 -!- hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has quit [Quit: Leaving...] 18:53:37 hargettp [~hargettp@pool-71-184-185-93.bstnma.east.verizon.net] has joined #sbcl 18:58:07 -!- cmm [~cmm@109.67.199.173] has quit [Ping timeout: 252 seconds] 18:59:12 cmm [~cmm@bzq-109-67-199-173.red.bezeqint.net] has joined #sbcl 19:17:22 rmarynch [~roman@88.135.194.233] has joined #sbcl 19:23:05 What is the extra verbosity that is missing in ASSERT? It prints the failed assertion currently, right? 19:25:01 flip214 [~marek@h081217084238.dyn.cm.kabsi.at] has joined #sbcl 19:25:01 -!- flip214 [~marek@h081217084238.dyn.cm.kabsi.at] has quit [Changing host] 19:25:01 flip214 [~marek@unaffiliated/flip214] has joined #sbcl 19:25:19 nikodemus: thank you very much for 1.0.48.28: make TRULY-THE macroexpandable 19:25:31 although the commit message is slightly incorrect 19:25:33 you're welcome 19:25:44 what's the issue? 19:26:20 "Add %MACROEXPAND and %MACROEXPAND" ?? should be "Add %MACROEXPAND and %MACROEXPAND-1" I think 19:26:41 but it doesn't really matter 19:28:15 oh, that 19:28:28 i noticed it as soon as i saw the commit email 19:28:48 as far as my commit message typos go, it's on the less confusing end of the spectrum :) 19:29:23 yes, right 19:47:47 ah, I see now. ASSERT does not show the values in the failed forms. But the current behavior has an advantage - it is easier to find the failed assertion. (= sum expected-sum) vs (= 6 4), especially when there are many assertions in a single function 19:49:03 rmarynch: it should ideally show both 19:49:34 the ticket does not say that :) 19:49:45 but I agree 19:50:06 it doesn't say the current behaviour is bad. it asks for more :) 19:50:20 I'm not sure I agree for a general purpose assert, as opposed to a test framework one 19:50:58 it's one thing for (assert-eql (foo) 1) to show the value of foo 19:51:05 the cost is trivial, the convenience is great 19:51:52 I wasn't even thinking of explicit costs there, more about how in the world it would work without being confusing 19:52:21 i know i've wanted something like that often 19:53:09 random assert somewhere breaks, and i need to edit it into (unless ... (error ...)) to figure out what was what 19:53:37 so you'd do what? code-walk the whole assert form and find all free variables and expand them only? 19:53:48 or code-walk it and expand all variables? 19:54:19 i would punt on macros, and for functions just report the subforms 19:54:21 or just do something hacky for the simple case of (predicate values) 19:55:04 something like: Assertion (FOO (X Y) B C) failed. (X Y) = "OK", B = 10, C = NIL 19:56:07 and (assert (eql a 1)) -> A = 2, 1 = 1, :-D 19:56:14 but fair enoguh 19:56:35 or possibly Assertion (FOO "OK" 10 NIL) failed. Original form: (FOO (X Y) B C) 19:58:44 <|3b|> any way they could be put somewhere that would end up inspectable in slime debugger? 19:59:33 ohoh 20:00:14 i like that 20:01:44 nikodemus: the unrelated question, about that simple make-thread +args patch (#727384). Is it wrong in some way? If not, could you please merge it? These merging requests are something boring, I realize, but I really need it to move further with the parallel tests runner 20:02:00 rmarynch: paul said he's going to take care of it 20:02:11 we /do/ have other committers than me :) 20:02:40 yeah. I'm off in academia land until ~ wednesday 20:03:56 pkhuong: do you plan to change the patch, or it is fine already? 20:04:50 I moved some checks to avoid dying in apply 20:05:03 and a couple extra unit tests, but that's it. 20:05:35 thank you 20:05:37 I wanted to allow a argument list designator for a while, but I don't know what I think of that. 20:05:46 s/args/arguments/? 20:06:20 as well. 20:09:54 -!- rmarynch [~roman@88.135.194.233] has quit [Quit: Leaving] 20:16:26 nikodemus: re hash table and EQ, it seems that most of it is our own. 20:17:04 obviously symbols/keywords can be fixed. 20:17:05 pkhuong: where was that paste you had 20:17:09 http://paste.lisp.org/display/122252 20:17:25 i have a system that might be interesting, just a sec 20:17:53 (oh, wait now I remember the fuck up with symbols: we must hash on symbol name for sxhash, so we need a second slot) 20:18:10 for the rest, we can specify our hash functions 20:18:21 the only remaining badness is that MOP likes to use raw CONSes as keys 20:18:28 (effective method stuff) 20:19:51 why a second slot? 20:19:57 one for sxhash 20:20:00 one for eq-hash 20:20:14 why can't we use sxhash for eq-hash? 20:20:38 (eql (sxhash :and) (sxhash 'cl:and)) 20:21:10 we used to have a good hash function there, but then we realized that sxhash mandated hashing on the name for symbols 20:21:37 so we _can_, but we get better distribution with a second slot. ok 20:23:22 I think anyone stuffing more than a few symbols of a same name but with different / undefined packages into the same hash table deserves what they're getting 20:23:42 sure ;) 20:40:58 http://paste.lisp.org/display/122252#10 20:52:56 slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has joined #sbcl 21:00:30 -!- flip214 [~marek@unaffiliated/flip214] has quit [Read error: Operation timed out] 22:05:43 nikodemus` [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 22:07:44 -!- nikodemus` [~nikodemus@cs181063174.pp.htv.fi] has quit [Client Quit] 22:31:53 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Ping timeout: 258 seconds] 23:10:08 -!- homie [~levent.gu@xdsl-78-35-158-176.netcologne.de] has quit [Read error: Connection reset by peer] 23:11:16 homie [~levent.gu@xdsl-78-35-176-121.netcologne.de] has joined #sbcl 23:35:52 slyrus__ [~chatzilla@adsl-108-80-231-224.dsl.pltn13.sbcglobal.net] has joined #sbcl 23:36:20 -!- slyrus__ is now known as slyrus 23:38:00 -!- slyrus_ [~chatzilla@99-27-205-152.lightspeed.irvnca.sbcglobal.net] has quit [Ping timeout: 255 seconds] 23:39:43 nikodemus` [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl