00:06:48 -!- kwmiebach [kwmiebach@vps-6813.united-hoster.de] has quit [Read error: No route to host] 00:06:57 kwmiebach [kwmiebach@vps-6813.united-hoster.de] has joined #sbcl 01:34:50 echo-area [~user@182.92.247.2] has joined #sbcl 01:58:33 pkhuong, I see DEFUN GCD in numbers.lisp, TWO-ARG-GCD VOP, and a source transform for GCD 01:59:22 Quadrescence: bignum.lisp, bignum-gcd. 01:59:47 pkhuong, can you explain the path of code to reach that? How does gcd in numbers.lisp "find" that? 02:03:32 there's a source transform to turn multi-arg calls into 2-arg calls to GCD 02:03:53 a VOP is defined to punt 2-arg calls to a static fun 02:04:25 I see 02:04:36 two-arg-gcd is called by the static function, and there, it goes through a type-based dispatch. 02:07:07 I see, thanks 02:11:41 scymtym__ [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has joined #sbcl 02:12:33 -!- scymtym_ [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has quit [Ping timeout: 245 seconds] 02:13:15 gko [~gko@220.228.255.202] has joined #sbcl 02:32:42 sbcl, i love you and hate you for constant folding during rudimentary tests 02:42:12 Quadrescence: eval (: 02:46:11 Can someone explain the results at the bottom? http://paste.lisp.org/display/128777 02:46:27 what is the fundamental slowdown in all functions but CL:GCD and GCD:REC-GCD? 02:46:41 Is SBCL being smart about recursion and data sharing in REC-GCD? 02:47:20 I just find these results mind boggling 02:49:00 rec-gcd punts to cl:gcd. 02:49:11 oh my 02:49:15 I refactored wrong 02:50:30 *Quadrescence* is feeling quite dumb today. :( 02:51:53 Well, changing that caused it to go to 14 seconds. CL:GCD is either doing some magic or is truly fast code 02:52:39 it's pretty fast (: 02:59:46 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 03:39:36 -!- whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has quit [Quit: Farewell] 03:39:55 whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has joined #sbcl 04:10:15 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 04:17:17 -!- jiacobucci [~jiacobucc@gw-asdl.ae.gatech.edu] has quit [Ping timeout: 246 seconds] 04:17:49 jiacobucci [~jiacobucc@gw-asdl.ae.gatech.edu] has joined #sbcl 04:18:10 -!- antifuchs [foobar@care.boinkor.net] has quit [Ping timeout: 272 seconds] 04:18:19 antifuchs [foobar@care.boinkor.net] has joined #sbcl 04:26:23 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 05:45:14 sdemarre [~serge@91.176.115.235] has joined #sbcl 06:06:31 slyrus [~chatzilla@99-28-161-110.lightspeed.miamfl.sbcglobal.net] has joined #sbcl 06:09:25 stassats [~stassats@wikipedia/stassats] has joined #sbcl 06:26:23 Kryztof [~user@81.174.155.115] has joined #sbcl 06:26:23 -!- ChanServ has set mode +o Kryztof 06:31:02 prxq [~mommer@mnhm-590c0143.pool.mediaWays.net] has joined #sbcl 06:38:25 -!- sdemarre [~serge@91.176.115.235] has quit [Ping timeout: 276 seconds] 07:10:20 attila_lendvai [~attila_le@87.247.12.222] has joined #sbcl 07:10:22 -!- attila_lendvai [~attila_le@87.247.12.222] has quit [Changing host] 07:10:22 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:14:51 -!- flip212 [~marek@unaffiliated/flip214] has quit [Remote host closed the connection] 07:15:30 flip214 [~marek@unaffiliated/flip214] has joined #sbcl 08:22:11 -!- christoph_debian [~user@oteiza.siccegge.de] has quit [Ping timeout: 260 seconds] 08:25:45 milanj [~milanj_@79-101-212-30.dynamic.isp.telekom.rs] has joined #sbcl 08:28:50 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 08:58:56 cmm [~cmm@bzq-79-182-232-253.red.bezeqint.net] has joined #sbcl 09:22:43 attila_lendvai1 [~attila_le@87.247.12.196] has joined #sbcl 09:22:44 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 09:31:35 nikodemus [~nikodemus@188-67-156-31.bb.dnainternet.fi] has joined #sbcl 09:31:35 -!- ChanServ has set mode +o nikodemus 09:33:47 christoph_debian [~user@DSL01.212.114.250.149.ip-pool.NEFkom.net] has joined #sbcl 09:42:41 -!- christoph_debian [~user@DSL01.212.114.250.149.ip-pool.NEFkom.net] has quit [Ping timeout: 260 seconds] 09:53:56 christoph_debian [~user@oteiza.siccegge.de] has joined #sbcl 10:09:20 -!- nikodemus [~nikodemus@188-67-156-31.bb.dnainternet.fi] has quit [Ping timeout: 265 seconds] 10:16:06 -!- milanj [~milanj_@79-101-212-30.dynamic.isp.telekom.rs] has quit [Ping timeout: 265 seconds] 10:46:45 -!- gko [~gko@220.228.255.202] has quit [Ping timeout: 248 seconds] 10:50:52 milanj [~milanj_@79-101-212-30.dynamic.isp.telekom.rs] has joined #sbcl 11:06:07 -!- echo-area [~user@182.92.247.2] has quit [Read error: Connection reset by peer] 11:16:10 -!- antifuchs [foobar@care.boinkor.net] has quit [Ping timeout: 272 seconds] 11:18:52 -!- whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has quit [Ping timeout: 250 seconds] 11:21:26 whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has joined #sbcl 11:23:27 antifuchs [foobar@care.boinkor.net] has joined #sbcl 11:28:12 -!- antifuchs [foobar@care.boinkor.net] has quit [Ping timeout: 272 seconds] 11:28:12 -!- whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has quit [Ping timeout: 272 seconds] 11:35:11 antifuchs [foobar@care.boinkor.net] has joined #sbcl 11:35:50 -!- attila_lendvai1 [~attila_le@87.247.12.196] has quit [Ping timeout: 246 seconds] 11:36:55 whoops [whoops@2600:3c01::f03c:91ff:fe93:da36] has joined #sbcl 11:48:55 -!- edgar-rft [~user@HSI-KBW-078-043-123-191.hsi4.kabel-badenwuerttemberg.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 12:07:53 gko [~gko@114-34-168-13.HINET-IP.hinet.net] has joined #sbcl 12:33:40 attila_lendvai [~attila_le@37.99.77.87] has joined #sbcl 12:33:40 -!- attila_lendvai [~attila_le@37.99.77.87] has quit [Changing host] 12:33:40 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 12:34:29 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Client Quit] 12:39:48 dtw [~dtw@pdpc/supporter/active/dtw] has joined #sbcl 12:54:20 <_8david> antifuchs: herep 12:54:51 <_8david> We just noticed that boinkmarks are code 503. Is that a known problem? 12:55:11 <_8david> Do you (or anyone else here) have plans to fix/resurrect them? 12:56:29 -!- psilord [~psilord@c-24-118-208-140.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 13:04:25 -!- scymtym__ [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has quit [Read error: Connection reset by peer] 13:04:53 scymtym [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has joined #sbcl 13:12:18 hey _8david. Dare I ask what your current plans on the windows fork are? 13:15:18 <_8david> Well, having done nothing about it for 4 months, I have just in the last few days talked to hlavaty about it. 13:16:20 <_8david> We need a testing environment to check that changes don't "break stuff". 13:16:35 <_8david> (I hope) we will have autobench running on linux/x86, wine/x86, linux/ppc, and opensolaris/sparc soonish. 13:18:44 cool 13:19:48 <_8david> An improved autobench, I should say, that runs tests in addition to benchmarks. 13:21:19 that would indeed be nice 13:47:49 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 13:48:43 -!- milanj [~milanj_@79-101-212-30.dynamic.isp.telekom.rs] has quit [Ping timeout: 276 seconds] 13:54:19 why isn't more things declared MAYBE-INLINE? 13:56:18 -!- hlavaty [~user@91-65-218-223-dynip.superkabel.de] has quit [Read error: Connection reset by peer] 14:01:52 milanj [~milanj_@93-86-54-102.dynamic.isp.telekom.rs] has joined #sbcl 14:15:54 saschakb [~saschakb@p4FEA0BC2.dip0.t-ipconnect.de] has joined #sbcl 14:31:12 stassats`` [~stassats@pppoe.178-66-90-12.dynamic.avangarddsl.ru] has joined #sbcl 14:33:12 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 14:33:30 -!- specbot [~specbot@pppoe.178-66-40-208.dynamic.avangarddsl.ru] has quit [Ping timeout: 265 seconds] 14:33:34 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 276 seconds] 14:34:18 -!- stassats`` is now known as stassats 14:35:05 -!- stassats [~stassats@pppoe.178-66-90-12.dynamic.avangarddsl.ru] has quit [Changing host] 14:35:05 stassats [~stassats@wikipedia/stassats] has joined #sbcl 14:35:48 specbot [~specbot@pppoe.178-66-90-12.dynamic.avangarddsl.ru] has joined #sbcl 14:49:52 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 15:22:51 nikodemus [~nikodemus@cs181241043.pp.htv.fi] has joined #sbcl 15:22:58 -!- ChanServ has set mode +o nikodemus 16:14:15 -!- slyrus [~chatzilla@99-28-161-110.lightspeed.miamfl.sbcglobal.net] has quit [Ping timeout: 265 seconds] 16:33:10 -!- nikodemus [~nikodemus@cs181241043.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 16:36:15 -!- gko [~gko@114-34-168-13.HINET-IP.hinet.net] has quit [Ping timeout: 260 seconds] 16:38:10 -!- saschakb [~saschakb@p4FEA0BC2.dip0.t-ipconnect.de] has quit [Remote host closed the connection] 17:02:27 sdemarre [~serge@91.176.115.235] has joined #sbcl 17:04:33 yay, my fix for pfdietz' modular arithmetic bug also fixes 903821, I think 17:04:49 lp 903821 17:04:49 https://bugs.launchpad.net/bugs/903821 17:05:36 writing a test case might be tricky 17:06:05 grep disassembly output? 17:06:14 ew 17:08:10 <|3b|> trace the full call and count the calls while running the test? 17:08:42 I like that, because it will break hard when someone renames it 17:10:23 *|3b|* wonders how hard it would be to instrument the compiler to report that sort of thing 17:14:11 (loop with header = (sb-kernel:fun-code-header #'lowmask) for i from sb-vm:code-constants-offset below (sb-kernel:get-header-data header) thereis (typep (sb-kernel:code-header-ref header i) 'sb-kernel:fdefn)) 17:16:08 let's compete for the most horrible solution :) 17:16:48 <|3b|> timing it vs a version compiled with notinline? 17:17:16 (block nil (sb-introspect::map-code-constants (sb-kernel:fun-code-header #'lowmask) (lambda (x) (when (typep x 'sb-kernel:fdefn) (return t))))) 17:18:03 (sb-introspect:find-function-callees #'lowmask) => (#) 17:19:15 add a hook in %compile to walk the IR2 graph? 17:23:19 are there are any drawbacks to adding MAYBE-INLINE for things like REDUCE? 17:23:51 saschakb [~saschakb@p4FEA0BC2.dip0.t-ipconnect.de] has joined #sbcl 17:24:41 stassats: eesh. I'd rather have a couple transforms for reduce of list or vectors. 17:25:21 i'm just seeing that SORT is MAYBE-INLINE 17:30:21 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 17:33:54 and why is sort-vector implemented as a define-source-transform? 17:34:06 and not just a function 17:34:19 antgreen [~user@bas3-toronto06-1177890637.dsl.bell.ca] has joined #sbcl 17:35:47 stassats: that way we get inlining. 17:35:58 why not declare the function inline? 17:36:48 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl 17:37:02 the compiler doesn't always inline inline functions. It sometimes gets bored halfway through ;) 17:37:50 and i'm fuzzy on maybe-inline conditions, speed > space? 17:38:43 and speed > 1, I think. 17:39:02 speed 3, space 1, not inlining, speed 1, space 0, inlining 17:39:13 (that's all about SORT) 17:39:17 space 0, then. 17:39:31 i just can't find where it's spelled out 17:40:31 ah, just space 0. 17:41:28 ok, now i see it 17:41:28 17:41:32 ir1opt, recognize-known-call. 17:41:41 yeah 17:43:01 and when can compiler get bored inlining? 17:43:58 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 17:44:49 *inline-expansion-limit*, or the limit on the number of optimisation rounds. 17:45:21 and surely, since SORT itself is declared maybe-inline, it can happen anyway 17:47:03 i know see why deftransform could be better for those things 17:49:33 stassats: the issue is that we'd have SORT inlined, but not the calls in seq-dispatch. 17:57:59 slyrus [~chatzilla@173-228-44-92.dsl.static.sonic.net] has joined #sbcl 18:18:08 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 18:57:12 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 252 seconds] 19:03:33 -!- prxq [~mommer@mnhm-590c0143.pool.mediaWays.net] has quit [Quit: Leaving] 19:15:14 jsnell: you said you were getting threads tests failures? So am I 19:18:08 Does SBCL do any optimizations with TYPECASE? E.g., can it recognize a TYPECASE in a function and specialize calls to this function elsewhere? 19:18:20 if you inline it, yes 19:18:36 if you don't, then no, because the function could be redefined at any time 19:18:58 Kryztof, how about recursive functions employing TYPECASE? 19:19:04 talk to flip214 19:19:10 okay :) 19:19:13 and read recent correspondence on sbcl-devel 19:40:24 Would there be any interest in someone mentoring me on SBCL for pay (provided the pay is somehow targeted toward SBCL in some way)? 19:41:39 like keeping an sbcl committer from starving? 19:41:51 yes 19:42:00 that is completely legitimate 19:50:07 -!- sdemarre [~serge@91.176.115.235] has quit [Ping timeout: 276 seconds] 19:55:26 -!- dtw [~dtw@pdpc/supporter/active/dtw] has quit [Quit: Zzzz] 20:28:31 -!- saschakb [~saschakb@p4FEA0BC2.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 20:33:20 Qworkescence: if you don't get any other offers, I could consider that -- dependng on your resources there are various possibilities. But I don't aim to make a living from sbcl work, whereas some people (notably nikodemus) do, so ask him 20:33:24 very good question, though 20:34:43 I probably couldn't provide someone with a living for this kind of stuff. 20:50:05 prxq [~mommer@mnhm-590c0143.pool.mediaWays.net] has joined #sbcl 20:56:48 Qworkescence: sure. Every little helps 20:57:09 nikodemus might be grateful for a few hours a week, or similar 21:04:58 saschakb [~saschakb@p4FEA0BC2.dip0.t-ipconnect.de] has joined #sbcl 21:06:30 -!- milanj [~milanj_@93-86-54-102.dynamic.isp.telekom.rs] has quit [Quit: Leaving] 21:21:53 -!- antgreen [~user@bas3-toronto06-1177890637.dsl.bell.ca] has quit [Remote host closed the connection] 21:32:48 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 22:01:29 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: rpg] 22:20:08 nikodemus [~nikodemus@cs181241043.pp.htv.fi] has joined #sbcl 22:20:09 -!- ChanServ has set mode +o nikodemus 22:33:56 -!- nikodemus [~nikodemus@cs181241043.pp.htv.fi] has quit [Ping timeout: 265 seconds] 22:47:33 homie [~levgue@xdsl-78-35-143-169.netcologne.de] has joined #sbcl 22:47:36 -!- kwmiebach [kwmiebach@vps-6813.united-hoster.de] has quit [Read error: Connection reset by peer] 22:47:49 kwmiebach [kwmiebach@vps-6813.united-hoster.de] has joined #sbcl 23:05:46 -!- |42| [user@2a01:7e00::f03c:91ff:fedf:3aab] has quit [Remote host closed the connection] 23:06:05 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 23:32:34 -!- prxq [~mommer@mnhm-590c0143.pool.mediaWays.net] has quit [Remote host closed the connection] 23:36:53 -!- Phoodus [~foo@wsip-68-107-217-139.ph.ph.cox.net] has quit [Ping timeout: 244 seconds] 23:37:24 -!- slyrus [~chatzilla@173-228-44-92.dsl.static.sonic.net] has quit [Ping timeout: 244 seconds]