01:02:02 obviously multiple output streams are harder, but our memory allocator introduces much worse slowdowns 01:13:49 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 01:22:02 maybe not *much* worse, but almost on the same order 02:05:10 -!- stassats [~stassats@wikipedia/stassats] has quit [Remote host closed the connection] 02:06:23 stassats [~stassats@wikipedia/stassats] has joined #sbcl 02:15:35 tcr1 [~tcr@155-dom-3.acn.waw.pl] has joined #sbcl 02:42:57 -!- tcr1 [~tcr@155-dom-3.acn.waw.pl] has quit [Quit: Leaving.] 02:59:01 -!- stassats [~stassats@wikipedia/stassats] has quit [Remote host closed the connection] 02:59:35 stassats [~stassats@pppoe.178-66-54-180.dynamic.avangarddsl.ru] has joined #sbcl 02:59:40 -!- stassats [~stassats@pppoe.178-66-54-180.dynamic.avangarddsl.ru] has quit [Changing host] 02:59:40 stassats [~stassats@wikipedia/stassats] has joined #sbcl 04:15:39 -!- cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has quit [Ping timeout: 255 seconds] 04:16:29 cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has joined #sbcl 06:01:28 -!- christoph_debian [~user@oteiza.siccegge.de] has quit [Remote host closed the connection] 06:23:02 flip214 [~marek@2001:858:107:1:7a2b:cbff:fed0:c11c] has joined #sbcl 06:23:02 -!- flip214 [~marek@2001:858:107:1:7a2b:cbff:fed0:c11c] has quit [Changing host] 06:23:02 flip214 [~marek@unaffiliated/flip214] has joined #sbcl 06:38:12 tcr1 [~tcr@155-dom-3.acn.waw.pl] has joined #sbcl 07:51:00 -!- homie [~levgue@xdsl-78-35-137-156.netcologne.de] has quit [Read error: Operation timed out] 07:51:26 homie` [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 07:52:08 -!- homie` [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Remote host closed the connection] 08:04:42 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 08:11:06 daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has joined #sbcl 08:17:06 misterncw [~misterncw@82.71.241.25] has joined #sbcl 08:37:53 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 08:39:34 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 08:46:39 -!- tcr1 [~tcr@155-dom-3.acn.waw.pl] has quit [Quit: Leaving.] 08:57:35 hlavaty [~user@91-65-223-81-dynip.superkabel.de] has joined #sbcl 09:04:48 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Ping timeout: 276 seconds] 09:44:57 angavrilov [~angavrilo@217.71.227.181] has joined #sbcl 10:00:32 -!- misterncw [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 10:02:11 misterncw [~misterncw@82.71.241.25] has joined #sbcl 10:12:21 misternc_ [~misterncw@82.71.241.25] has joined #sbcl 10:12:26 -!- misterncw [~misterncw@82.71.241.25] has quit [Read error: Connection reset by peer] 10:46:04 attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has joined #sbcl 10:46:04 -!- attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has quit [Changing host] 10:46:04 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 10:47:45 -!- daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has quit [Quit: WeeChat 0.3.5] 11:05:25 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 11:05:25 -!- ChanServ has set mode +o nikodemus 11:31:30 -!- misternc_ [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 11:34:41 misterncw [~misterncw@82.71.241.25] has joined #sbcl 11:51:35 attila_lendvai1 [~attila_le@catv-89-132-189-7.catv.broadband.hu] has joined #sbcl 11:51:35 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 11:51:36 -!- attila_lendvai1 is now known as attila_lendvai 11:51:36 -!- attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has quit [Changing host] 11:51:36 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 12:18:10 daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has joined #sbcl 12:20:14 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 12:20:25 -!- daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has quit [Client Quit] 12:21:16 daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has joined #sbcl 12:25:50 -!- misterncw [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 12:29:01 misterncw [~misterncw@82.71.241.25] has joined #sbcl 12:40:21 -!- cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has quit [Ping timeout: 264 seconds] 12:41:12 cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has joined #sbcl 12:46:44 -!- flip214 [~marek@unaffiliated/flip214] has quit [Remote host closed the connection] 12:56:50 -!- Krystof [~csr21@csrhodes.plus.com] has quit [Ping timeout: 246 seconds] 12:58:20 -!- misterncw [~misterncw@82.71.241.25] has quit [Read error: Connection reset by peer] 12:58:56 misterncw [~misterncw@82.71.241.25] has joined #sbcl 13:10:42 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Ping timeout: 258 seconds] 13:15:52 -!- misterncw [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 13:24:55 misterncw [~misterncw@82.71.241.25] has joined #sbcl 13:37:53 christoph_debian [~user@oteiza.siccegge.de] has joined #sbcl 13:42:21 -!- cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has quit [Ping timeout: 276 seconds] 13:42:40 cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has joined #sbcl 14:20:02 attila_lendvai1 [~attila_le@catv-89-132-189-7.catv.broadband.hu] has joined #sbcl 14:20:03 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 14:20:03 -!- attila_lendvai1 is now known as attila_lendvai 14:20:04 -!- attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has quit [Changing host] 14:20:04 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 14:24:47 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 14:25:04 attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has joined #sbcl 14:25:04 -!- attila_lendvai [~attila_le@catv-89-132-189-7.catv.broadband.hu] has quit [Changing host] 14:25:04 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 15:31:34 -!- misterncw [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 15:34:17 misterncw [~misterncw@82.71.241.25] has joined #sbcl 15:43:14 -!- stassats` [~stassats@wikipedia/stassats] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 16:06:39 -!- cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has quit [Ping timeout: 258 seconds] 16:07:12 -!- misterncw [~misterncw@82.71.241.25] has quit [Remote host closed the connection] 16:07:33 cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has joined #sbcl 16:12:13 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 16:22:01 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Remote host closed the connection] 16:26:30 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 16:28:24 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Read error: Connection reset by peer] 16:29:51 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 16:31:17 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Client Quit] 16:38:07 -!- hlavaty [~user@91-65-223-81-dynip.superkabel.de] has quit [Read error: Operation timed out] 16:43:14 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 16:46:24 gor[e] [~svr@gw2.masterhost.ru] has joined #sbcl 17:05:06 -!- gor[e] [~svr@gw2.masterhost.ru] has quit [Read error: Operation timed out] 17:34:38 cow-orker [~foobar@pogostick.net] has joined #sbcl 17:36:16 -!- angavrilov [~angavrilo@217.71.227.181] has quit [Remote host closed the connection] 17:38:10 pkhuong: wait, it went from 6.5Mcycles to 1.8Mcycles by adding the xor/cmp? 17:39:57 pkhuong: you know, this work should combine nicely with gc safepoints: you can omit multiple write log calls to provably the same object entirely, between gc safepoints. 17:51:14 tcr1 [~tcr@80-218-247-218.dclient.hispeed.ch] has joined #sbcl 17:57:33 yeah. 17:57:43 the problem isn't reads, it 17:57:46 s with writes 17:57:51 writes eat up bandwidth :) 17:58:20 no need for safe points. Just keep the last logged object across GCs 17:59:51 -!- daimrod [~daimrod@ANantes-556-1-326-23.w90-59.abo.wanadoo.fr] has quit [Quit: WeeChat 0.3.5] 18:00:31 I've got a really kludgey plan to allow for some static analysis during VOP emission, that'd take care of common cases like writes to slots in the same object, especially, if we allow the log to choose among a few SSBs at compile-time. 18:01:50 doesn't the static analysis become a lot easier with a safepoint? 18:02:43 no. GCs don't have to flush the SSB completely. 18:02:53 although, yeah, interrupts. 18:02:56 fsck. 18:03:25 interrupts just make a new SSB? 18:03:37 right. That'd work. 18:08:41 (that, or go pseudo-atomic in the slow path) 18:12:02 that means (likely) memory allocation in interrupt handlers. sigh. 18:53:51 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 18:58:16 homie [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 19:15:18 homie` [~levgue@xdsl-84-44-209-111.netcologne.de] has joined #sbcl 19:22:33 -!- homie` [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 19:34:04 frito [~user@cpc2-sotn4-0-0-cust13.15-1.cable.virginmedia.com] has joined #sbcl 19:44:54 -!- frito [~user@cpc2-sotn4-0-0-cust13.15-1.cable.virginmedia.com] has quit [Ping timeout: 255 seconds] 19:45:37 -!- cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has quit [Ping timeout: 246 seconds] 19:46:37 cmm [~cmm@bzq-79-177-212-194.red.bezeqint.net] has joined #sbcl 20:03:00 <_8david> Apropos safepoints: I'm not going to push or announce anything publically before doing the actual Windows changes, but safepoints for (GC and INTERRUPT-THREAD) on POSIX are pretty much ready; patch review appreciated. 20:03:16 <_8david> https://gitorious.org/~lichteblau/sbcl/david-sbcl-misc/commits/stw-linux-1 20:07:30 -!- slyrus_ [~chatzilla@173-228-44-88.dsl.static.sonic.net] has quit [Ping timeout: 260 seconds] 20:09:24 _8david: can you enable the CIA hook on github ? 20:10:28 homie` [~levgue@xdsl-78-35-172-202.netcologne.de] has joined #sbcl 20:12:02 -!- homie [~levgue@xdsl-84-44-209-111.netcologne.de] has quit [Ping timeout: 240 seconds] 20:12:55 <_8david> I see the CIA hook, with several paramaters. Which configuration do you have in mind? 20:13:02 prxq [~mommer@mnhm-4d0135a7.pool.mediaWays.net] has joined #sbcl 20:13:24 -!- homie` [~levgue@xdsl-78-35-172-202.netcologne.de] has quit [Client Quit] 20:13:55 address: http://cia.vc 20:13:59 project: sbcl 20:14:02 branch: master 20:14:08 long orl: on 20:14:11 *url 20:15:34 then update, go back to the menu entry and press "test hook" 20:15:54 at which point the last 3 commits should appear at https://cia.vc/stats/project/sbcl/ 20:16:17 <_8david> Done. 20:16:18 homie [~levgue@xdsl-78-35-172-202.netcologne.de] has joined #sbcl 20:16:42 yay :) 20:18:12 and anybody with a CIA account can add a bot to this channel that advertises commits 20:19:22 cia? 20:20:11 prxq: https://cia.vc/stats/project/sbcl/ 20:20:13 not the most inspiring of all names. Looks cool nonetheless 20:20:43 .vc is pretty neat 20:20:56 I can see version control and venture capitalists fighting over it (: 20:21:22 the last ones are reputed to get .angel 20:21:27 prxq: you should read "Snow crash" :D 20:23:37 fe[nl]ix: looks interesting :) 20:36:29 if you read it you'll get the joke about having a "CIA account" :) 20:37:36 ah... I see. 20:38:19 lol 21:24:28 _8david: would it be less kludgey in handle_safepoint_violation to skip the safepoint check? 21:24:58 fiddle with RIP a bit, instead of hoping for a timing window? 21:25:29 bah, I'll see what the final commit does with it. 21:33:03 _8david: how does that work with callbacks? 21:34:01 ISTM we'd need some logic to stash the CSP/IP pair somewhere and clear the fields on callbacks? 21:45:55 <_8david> re the second question: yes, all callbacks go through callback_wrapper_trampoline, which does that. 21:46:02 k. 21:46:28 <_8david> hidden in the WITH_GC_AT_SAFEPOINTS_ONLY macro 21:46:51 re safepoint, we could actually overwrite the safepoint instruction with a jump to a handler or something equally awful. 21:47:16 even on !x86, we'd "just" have to call some instruction to clear the i$ 21:47:52 *re safepoint_violation 21:51:40 <_8david> with "fiddle with RIP" you mean what I was referring to as "arrange_return_to_c_function"? 21:56:23 -!- tcr1 [~tcr@80-218-247-218.dclient.hispeed.ch] has quit [Quit: Leaving.] 22:20:58 -!- prxq [~mommer@mnhm-4d0135a7.pool.mediaWays.net] has quit [Quit: Leaving]