00:34:49 Oops. Forgot to update NEWS. Oh well. 00:49:09 <_3b`> is "Unexpected success: dynamic-extent.impure.lisp / (NO-CONSING HASH-TABLES)" noteworthy? 1.0.42.21/linux/x8664 00:54:38 I... have no idea. 00:55:19 Though, that said, I'd expect the test suite to have -expected- results only for x86, x86-64, and ppc linux, in all build configurations. 00:57:23 <_3b`> i think possibly with-test doesn't support (and :ppc :sb-thread) for :fails-on 00:57:55 I find that possibility to be unlikely, TBH. 00:58:03 But perhaps. 00:58:15 <_3b`> i think it wants '(and ...) 00:58:27 ... Ah. That it might. 00:58:48 Running dynamic-extent.impure.lisp now. 00:58:50 um. get-bytes-consed doesn't actually work, does it? 00:59:11 oh, nvm, it runs it 10000 times 00:59:15 that's probably enough. :) 00:59:37 Not seeing any unexpected successes here. 00:59:52 x86-64/linux, 1.0.42.21. 00:59:58 Also not seeing any failures. 01:00:17 <_3b`> how do you run a single file of tests? 01:00:32 (cd tests && sh ./run-tests.sh dynamic-extent.impure.lisp) 01:01:01 Pass as many test files as you want on the command line. 01:02:03 <_3b`> yeah ' makes it go away, get it consistently without the ' though 01:02:19 How odd... 01:02:36 ... Ah. My ctf is killing threads. 01:03:42 And you are quite correct, it seems that a quote is required. 01:04:42 And that's the only test that needs fixing in this manner. 01:04:49 My fault, too. 01:06:09 That said, I'm not fixing it tonight. I'll deal with it tomorrow if nobody gets it before then. 01:19:17 -!- rbarraud [~rbarraud@202-180-88-252.callplus.net.nz] has quit [Ping timeout: 255 seconds] 01:54:09 The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has joined #sbcl 01:58:52 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 01:59:13 -!- nyef [~nyef@pool-71-255-129-229.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 02:11:27 rbarraud [~rbarraud@118-92-134-38.dsl.dyn.ihug.co.nz] has joined #sbcl 02:29:06 hmm... I don't think it's supposed to be sitting this long in: ::: Running (:HASH-TABLE-PARALLEL-READERS) 02:41:41 So, sbcl built on leopard will run on snow leopard, but not vice versa? 02:43:28 -!- rbarraud [~rbarraud@118-92-134-38.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 245 seconds] 02:43:46 rbarraud [~rbarraud@118.92.134.38] has joined #sbcl 02:46:02 deepfire: it should work both ways, given the right magic flags 02:49:01 pkhuong, also, how good is the fink way of getting sbcl? 03:23:09 this might be a dumb question, but why do i get so many nops in x86_64 disassembly for sbcl? 03:23:13 is it an alignment issue? 03:24:03 The_Jon_Smith: mostly, yes. 03:25:57 is just weird to see something like http://paste.lisp.org/+2G2G 03:26:02 with 8 nops or whatever 03:26:25 16* 03:28:37 you don't want to have data immediately following code. 03:31:50 ooh i see 03:32:21 -!- rbarraud [~rbarraud@118.92.134.38] has quit [Remote host closed the connection] 03:37:40 stassats [~stassats@wikipedia/stassats] has joined #sbcl 03:40:59 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 03:42:03 -!- stassats [~stassats@wikipedia/stassats] has quit [Remote host closed the connection] 03:42:10 What's the typical icache line size? 64/128 bytes? 16 1-byte nops might actually be not enough.. 04:05:18 -!- The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has left #sbcl 05:08:30 lnostdal [~quassel@56.84-48-233.nextgentel.com] has joined #sbcl 06:40:24 -!- gonzojive [~red@condi.Stanford.EDU] has quit [Quit: Leaving.] 06:58:30 -!- Krystof [~csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Ping timeout: 240 seconds] 07:06:36 hefner [~hefner@ppp-58-9-115-144.revip2.asianet.co.th] has joined #sbcl 07:11:51 -!- lnostdal [~quassel@56.84-48-233.nextgentel.com] has quit [Ping timeout: 258 seconds] 07:14:28 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 07:14:28 -!- ChanServ has set mode +o nikodemus 07:15:18 nikodemus: I should file an lp bug on this, but ::: Running (:HASH-TABLE-PARALLEL-READERS) hangs on x86-64/darwin/sb-threads 07:19:00 slyrus: can you get gdb backtraces on the threads? 07:19:29 tomorrow afternoon. it's late here and I'm going fishing in the morning :) 07:19:36 ok :) 07:29:14 slyrus, also, can you at some point check out https://bugs.launchpad.net/sbcl/+bug/545730 -- i'm still on leopard so can't check that 07:30:16 assigned to me 07:30:38 I've got the gdb up on that hash-table-parallel-readers hang 07:31:23 mostly in sem_wait_signal, one in mach_msg_trap (the exception handling thread probably) and three elsewhere 07:37:26 that sounds familiar 07:37:32 (unsolved, but familiar) 07:37:59 i suspect it's something being non-reentrant on darwin and reentrant on linux 07:38:49 any in malloc/free? 07:39:28 (your "tomorrow afternoon" arrived pretty quickly!) 07:42:35 heh 07:43:04 no, none in malloc/free. the other three look like they're in lisp land 07:49:34 rbarraud [~rbarraud@118-93-183-153.dsl.dyn.ihug.co.nz] has joined #sbcl 07:54:05 hm 07:54:05 can you get lisp backtraces for them from the gdb? 07:58:36 rbarraud_ [~rbarraud@118.93.183.153] has joined #sbcl 07:59:04 -!- rbarraud [~rbarraud@118-93-183-153.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 260 seconds] 08:05:48 attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has joined #sbcl 08:49:25 -!- attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has quit [Read error: Connection reset by peer] 08:49:33 attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has joined #sbcl 08:53:44 -!- hefner [~hefner@ppp-58-9-115-144.revip2.asianet.co.th] has quit [Quit: Arrrr!] 09:05:57 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 09:19:54 nikodemus [~nikodemus@cs181058025.pp.htv.fi] has joined #sbcl 09:19:54 -!- ChanServ has set mode +o nikodemus 09:30:24 deepfire: you get a full cache line if speed >>> space 10:08:49 minion: are you here? 10:09:01 oh well 10:26:55 I'm missing listpaste here 10:27:44 who's running minion and lisppaste these days? drewc and nyef? 10:28:56 http://paste.lisp.org/display/114139 10:29:29 I really would have expected the (t ) branch to be elided 10:34:45 tcr: (declare (optimize speed)) 10:35:54 how comes I need that? 10:36:29 comment from constrain-ref-type: 10:36:30 ;; Since massive symbol cases are an especially atrocious pattern 10:36:30 ;; and the (NOT (MEMBER ...ton of symbols...)) will never turn into 10:36:30 ;; a more useful type, don't propagate their negation except for NIL 10:36:30 ;; unless SPEED > COMPILATION-SPEED. 10:37:25 meaning constraint propagation ignores (NOT (MEMBER ...symbols...)) constraints for performance reasons under default policy 10:37:58 good to know, thank you! 10:45:45 -!- deepfire [~deepfire@80.92.100.69] has quit [Ping timeout: 258 seconds] 11:14:07 Krystof [~csr21@cpc1-bour2-0-0-cust414.15-1.cable.virginmedia.com] has joined #sbcl 11:14:07 -!- ChanServ has set mode +o Krystof 11:15:25 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #sbcl 11:49:44 -!- cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has quit [Ping timeout: 255 seconds] 11:50:27 cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has joined #sbcl 12:04:33 -!- rbarraud_ [~rbarraud@118.93.183.153] has quit [Ping timeout: 240 seconds] 12:07:31 -!- _3b` [foobar@cpe-72-179-19-4.austin.res.rr.com] has quit [Read error: Connection reset by peer] 12:07:45 _3b` [foobar@cpe-72-179-19-4.austin.res.rr.com] has joined #sbcl 12:11:51 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 12:21:47 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #sbcl 12:41:44 (sb-ext:defglobal **x** 42) 12:41:44 (defun quux () **x**) 12:41:57 deepfire [~deepfire@80.92.100.69] has joined #sbcl 12:42:13 Everytime I recompile quux, and look at the disassemble, the address in the beginning MOV changes 12:42:34 I'm intrigued by that, why is that? 12:43:55 i'm on x86-64 so the first MOV is a RIP-relative move here 12:44:08 can you paste the line from disassemble? 12:45:05 ; 0C0CED7E: 8B0550ED0C0C MOV EAX, [#xC0CED50] ; '**X** 12:45:28 Luke guessed that this address is actually a relative offset which makes sense 12:45:53 of constant table or however that thingie is called again 12:46:21 yep, that is correct 12:47:23 that is, it's not relative, but it's the address of **X** in the constant table of the function 12:47:33 GC fixes those references up when the function moves 12:48:14 the wacky x86, how we love you! 14:15:55 lnostdal [~quassel@56.84-48-233.nextgentel.com] has joined #sbcl 15:18:03 -!- attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has quit [Quit: Leaving.] 15:52:31 bootstrapping gurus sought 16:01:48 hello 16:08:21 yay! 16:09:16 i'm trying to wrap the %defun call in defun expansion with a macro 16:09:26 (with-foo (%defun ...)) 16:10:19 tackling source-location when just loading? 16:10:20 defining the macro with def!macro still ends up trying to call the WITH-FOO as a function at init time 16:10:36 no, fixing https://bugs.launchpad.net/sbcl/+bug/576637 16:19:05 -!- lnostdal [~quassel@56.84-48-233.nextgentel.com] has quit [Ping timeout: 276 seconds] 16:21:21 hm 16:30:26 *nikodemus* despairs, and tries sb!xc:defmacro, even though he is almost certain it can't be the right answer 16:36:23 it wasn't 16:36:28 time to go home 16:38:20 -!- nikodemus [~nikodemus@cs181058025.pp.htv.fi] has quit [Quit: Leaving] 16:51:56 oh, you've gone 16:51:59 tsk 16:52:04 gah, no minion 16:52:27 OK, well, I'll be around later too, but in case he pops in again -- it's probably to do with cold-fset 17:03:50 cold-fset is magical if and only if it happens at toplevel 17:04:17 it is the way that functions exist immediately at the start of cold-sbcl.core's initial function 18:15:09 attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has joined #sbcl 18:23:03 -!- attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has quit [Ping timeout: 245 seconds] 18:43:49 attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has joined #sbcl 18:48:04 lnostdal [~quassel@56.84-48-233.nextgentel.com] has joined #sbcl 18:59:09 i think my router is dead :< 18:59:16 i'm connected directly to the modem now 19:11:23 -!- attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has quit [Ping timeout: 245 seconds] 19:21:47 attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has joined #sbcl 19:43:46 lnostdal_ [~quassel@56.84-48-233.nextgentel.com] has joined #sbcl 19:44:18 -!- lnostdal [~quassel@56.84-48-233.nextgentel.com] has quit [Ping timeout: 240 seconds] 20:11:49 (uh, sorry .. wrong channel) 20:22:10 tcr: what's the desired code for lp#629058? 20:23:11 I can't write assembly, but something which doesn't calculate the offsets twice 20:24:24 the offets for inner? 20:24:32 yes 20:25:49 it would retain the pointer in a reg? 20:25:55 ideally I mean 20:29:52 yeah, I could see that optimization as being helpful 20:30:03 even just for structure constructors 20:39:35 rbarraud_ [~rbarraud@118-93-183-153.dsl.dyn.ihug.co.nz] has joined #sbcl 20:42:10 abiessu [1871244e@gateway/web/freenode/ip.24.113.36.78] has joined #sbcl 20:43:14 hello 20:43:28 i've noticed that sbcl is consistently marked as 'port in progress' for x86 windows... is there a chance that it will be caught up and fully ported soon? 20:43:52 (is there an easy way a semi-novice could contribute to the effort?) 20:44:07 financial? 20:44:32 are there significant financial hurdles? 20:45:49 given a sufficient amount of money some experienced SBCL developer could be sponsored for a full-time work 20:46:05 that makes sense 20:47:41 i guess i was hoping to hear about some obscure bug that nobody really wants to track down... 20:48:57 abiessu: it works, it's fairly stable, there are a few rough edges 20:49:41 abiessu: if you want bugs, https://bugs.launchpad.net/sbcl has in plenty 20:54:43 thanks! 21:01:19 nyef [~nyef@pool-71-255-129-229.cncdnh.east.myfairpoint.net] has joined #sbcl 21:01:46 Hello all. 21:03:50 hi nyef 21:05:44 -!- abiessu [1871244e@gateway/web/freenode/ip.24.113.36.78] has quit [Quit: Page closed] 21:08:23 tcr: you could definitely do it, but you might as well implement local value numbering or somesuch while you're at it 21:09:03 tcr: the easiest way would be to split the RAW-INSTANCE VOPs into compute-length/compute-offset/memop sequences 21:09:16 tcr: and then value number once that's done 21:09:49 Add that as a comment, I won't have the time :-) 21:09:49 heck, the splitting would make RAW-INSTANCE vop writing less tedious for new platforms anyway 21:13:54 Fare [~Fare@ita4fw1.itasoftware.com] has joined #sbcl 21:22:26 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 22:48:57 -!- Kae [b@c-cfcee253.97-16-64736c12.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer] 22:53:36 -!- attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has quit [Quit: Leaving.] 22:54:49 I have a trivial clisp 2.49 build fix in https://bugs.launchpad.net/sbcl/+bug/629144 if anyone wants an easy commit :) 22:57:49 -!- Fare [~Fare@ita4fw1.itasoftware.com] has quit [Quit: Leaving] 23:03:31 Tagged for review? 23:06:49 -!- tcr [~tcr@81-233-246-97-no37.tbcn.telia.com] has quit [Quit: Leaving.] 23:26:07 yes 23:49:29 tokenrove [~julian@miranda.org] has joined #sbcl