15:57:34 ccl-logbot [~ccl-logbo@setf.clozure.com] has joined #sbcl 15:57:34 15:57:34 -!- names: ccl-logbot antgreen akovalenko LiamH nyef milanj homie |3b| Blkt @nikodemus samebchase angavrilov attila_lendvai ASau Phoodus cmm spacebat peddie jiacobucci foom redline6561_ cow-orker deepfire blackwol` lichtblau tsuru` pchrist whoops antoszka slyrus Intensity @nikodemus_ sshirokov daimrod fe[nl]ix ljos danlarkin luis jsnell sbryant Quadrescence pkhuong natesm kanru ottergwc flip214 christoph_debian dsp_ reb```` MikeSeth 16:45:19 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl 16:45:23 drdo [~user@85.207.54.77.rev.vodafone.pt] has joined #sbcl 16:57:47 -!- blackwol` [~blackwolf@ool-45763eb0.dyn.optonline.net] has quit [Ping timeout: 244 seconds] 16:59:20 chp [~user@dyn-carl-202-105.dyn.columbia.edu] has joined #sbcl 17:09:43 Is there any point in testing both non-thread, futex and non-thread, non-futex? 17:11:50 in theory no... 17:11:57 But, in theory, there's no point in testing ;) 17:12:25 ... we must have a different body of theory. 17:15:53 It seems a shame to blow a full two-hour (or longer?) cycle to test a few code paths that are unlikely to have bugs, since I already did both futex and non-futex tests with threads enabled. 17:23:33 -!- Blkt [~user@89-96-199-46.ip13.fastwebnet.it] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 17:26:05 -!- slyrus [~chatzilla@adsl-99-49-14-228.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 260 seconds] 17:33:25 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 240 seconds] 17:34:11 nyef: do you have access to gcc compile farm? 17:34:23 gcc110 is a ppc box, and a rather snappy one 17:34:29 I don't believe so, no. 17:34:37 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:34:47 you can get one for the asking. directions are on the farm wiki 17:35:07 (well, it's ppc64, but now you can build sbcl on it) 17:35:36 I used to build SBCL on a ppc64, back when my G5 was in working order. 17:36:01 That was actually the machine I did the PPC threading work on. 17:36:43 was it linux? 17:36:53 Yeah. 17:37:03 Debian of some stripe. 17:37:26 I still have the disk, in case I get the machine working again, or come into another ppc64 system. 17:37:45 hm. funny bitrot. or did you just never commit whatever -m32 flags you added? 17:38:17 oh, or you had an actual 32 bit userland 17:38:21 a-ha 17:38:37 Yeah, debian tends towards multilib. 17:42:05 -!- chp [~user@dyn-carl-202-105.dyn.columbia.edu] has quit [Remote host closed the connection] 17:43:30 Is using the high 16 bits for tagging purposes worth it at all in x86_64? 17:44:23 Do you not run into trouble with the page mapping if you do that? 17:45:17 According to the amd manual, only 48bits are actually used for pointers 17:46:21 Does it define semantics for the upper bits during page translation? 17:46:38 It says they must always be kept at 0 17:46:53 so you would need to mask it out before dereferencing 17:47:22 I don't think you gain anything, then. 17:47:39 the inherited wisdom is that high tags aren't worth it -- of course things may always change, but... 17:50:08 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving] 17:50:26 Is there something i can read about how sbcl lays out memory? 17:50:56 Lays out in what sense? 17:51:24 There's some stuff in the internals manual, for example, but that's more of how objects are laid out in memory than about where the heap spaces are... 17:51:34 memory representation for objects, allocation/collection, etc 17:52:17 We don't have much about collection, and allocation varies a bit from backend to backend. 17:52:51 For the memory representation stuff, the internals manual and SYS:SRC;COMPILER;GENERIC;OBJDEF.LISP are probably your best bet. 17:56:02 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl 18:05:48 docs/internals/internal-design.txt in the /CMUCL/ tree is also a nice overview (somewhat stale, but still good for understanding) 18:06:29 updating that for SBCL would be worth many kudos 18:09:52 ... which one is that? 18:10:47 Heh. Only 20 years since its last revision. 18:13:43 ... This is still good, but spectacularly out of date for us, and probably for CMUCL as well. 18:15:23 Heh. Predates non-cheney GC. 18:21:22 Build failure, 1.0.53.102-41bc875. thread.c: In function 'new_thread_trampoline': thread.c:307: error: 'struct thread' has no member named 'alloc_region' 18:23:13 nyef: re. that doc. i think i never fully grokked sbcl's tagging scheme till i saw it 18:23:41 Oh, oops. My mistake. 18:24:19 Accidentally tried building a threaded cheneygc version. 18:24:24 heh 18:24:37 i'm calling this a day and looking for dinner. if new stuff that looks like my fault comes up, please encourage people to spam my email 18:25:03 I'm not expecting anything on my end at this point, TBH. 18:40:08 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 18:54:50 Kryztof [~user@81.174.155.115] has joined #sbcl 18:54:50 -!- ChanServ has set mode +o Kryztof 19:00:52 tcr1 [~tcr@user-0c9h4tj.cable.mindspring.com] has joined #sbcl 19:09:38 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 255 seconds] 19:13:32 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 19:13:32 -!- ChanServ has set mode +o nikodemus 19:53:43 nikodemus: The function SB-EXT:DYNAMIC-SPACE-SIZE is undefined. 19:55:01 I rather expect that that's on you. 20:02:14 chp [~user@c-68-45-180-238.hsd1.nj.comcast.net] has joined #sbcl 20:32:45 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Ping timeout: 260 seconds] 20:32:53 -!- angavrilov [~angavrilo@217.71.227.181] has quit [Ping timeout: 255 seconds] 20:34:38 pchrist [~spirit@gentoo/developer/pchrist] has joined #sbcl 20:41:04 ..seems like getrusage thing was actually a Linux bug with a known fix pending for 3.2: 20:43:54 borkman [~user@S0106001111de1fc8.cg.shawcable.net] has joined #sbcl 20:49:00 -!- borkman [~user@S0106001111de1fc8.cg.shawcable.net] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 21:06:13 sdemarre [~serge@91.176.42.187] has joined #sbcl 21:13:11 -!- akovalenko [~akovalenk@95.73.218.80] has quit [Quit: rcirc on GNU Emacs 24.0.91.1] 21:18:34 akovalenko [~akovalenk@95.73.49.241] has joined #sbcl 21:32:49 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 22:13:59 -!- sdemarre [~serge@91.176.42.187] has quit [Ping timeout: 244 seconds] 22:58:35 drdo` [~user@85.207.54.77.rev.vodafone.pt] has joined #sbcl 22:58:57 -!- drdo [~user@85.207.54.77.rev.vodafone.pt] has quit [Read error: Connection reset by peer] 23:02:36 -!- nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has quit [Quit: G'night all.] 23:18:40 -!- tcr1 [~tcr@user-0c9h4tj.cable.mindspring.com] has quit [Quit: Leaving.] 23:21:20 -!- milanj [~milanj_@77-46-169-35.dynamic.isp.telekom.rs] has quit [Quit: Leaving] 23:25:08 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 23:51:29 tcr1 [~tcr@user-0c9h4tj.cable.mindspring.com] has joined #sbcl