00:21:14 cffi complains regarding cl+ssl using FOO instead of either (:pointer (:struct foo)) or (:struct foo). What's the legacy meaning that it defaults to? 00:21:47 LiamH [~none@pool-74-96-4-63.washdc.east.verizon.net] has joined #sbcl 00:21:53 that's a cffi warning, not sbcl 00:23:54 Fare: FOO is now (:pointer (:struct FOO)), but I think cl+ssl was fixed already 00:24:40 not the one on gitorious, or I did something wrong along the way 00:24:46 Fare: well, not in all cases 00:25:41 CFFI used to define implicit pointers in some cases, now it uses a more C-like semantics 00:25:59 the code in src/types.lisp suggests that foo is parsed as (:struct foo) not (:pointer (:struct foo)) 00:26:51 so there is no automatic translation from the previous behavior? 00:27:27 there is, a bare type reference should behave like before 00:27:38 you just get the warnings 00:28:19 ok, so replace foo by (:struct foo) 00:28:30 no, that will break things 00:29:13 in allocation forms, replace foo by (:struct foo) 00:29:31 if you replace foo with (:struct foo), the replace mem-aref with mem-aptr 00:29:45 in foreign-alloc, use (:struct foo), in foreign-slot-value, use (:pointer (:struct foo)) ? 00:29:51 in dereference forms, replace it by :pointer 00:30:20 Fare: yes, think of it as a funny C 00:30:48 I see foreign-alloc foo, should it be foreign-alloc (:struct foo) or foreign-alloc (:pointer (:struct foo)) ? 00:30:49 foreign-slot-value works with (:struct foo) 00:31:01 the former 00:32:15 ok. And if foo is passed by value to a lisp function, is it a :pointer automatically, or does cffi do the wrapping so it is a (:struct foo) object? 00:33:00 it's always a pointer 00:33:17 so I cannot use foreign-slot-value, but must use something else? 00:33:32 or case use foreign-slot-value, just using (:pointer (:struct foo)) ? 00:33:44 -!- davazp [~user@178.167.170.27.threembb.ie] has quit [Remote host closed the connection] 00:33:47 use foreign-slot-value with (:struct foo) 00:34:07 if you use (:struct foo) with mem-aref, you will get a plist 00:34:45 and shouldn't this be discussed in #lisp? 00:35:22 Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has joined #sbcl 00:36:54 oh sorry, wrong channel 00:40:51 -!- Bike [~Glossina@65-100-32-70.ptld.qwest.net] has quit [Quit: Lost terminal] 00:41:57 Bike [~Glossina@65-100-32-70.ptld.qwest.net] has joined #sbcl 00:48:30 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 01:02:29 -!- scymtym [~user@ip-178-202-214-83.unitymediagroup.de] has quit [Ping timeout: 256 seconds] 01:03:31 erikc [~erikc@CPE00222d53fe78-CM00222d53fe75.cpe.net.cable.rogers.com] has joined #sbcl 01:18:21 Bike_ [~Glossina@174-25-34-56.ptld.qwest.net] has joined #sbcl 01:20:43 -!- Bike [~Glossina@65-100-32-70.ptld.qwest.net] has quit [Ping timeout: 246 seconds] 01:20:56 Bike [~Glossina@174-25-38-57.ptld.qwest.net] has joined #sbcl 01:21:06 rpg [~rpg@216.243.156.16.real-time.com] has joined #sbcl 01:22:41 -!- Bike_ [~Glossina@174-25-34-56.ptld.qwest.net] has quit [Ping timeout: 252 seconds] 01:23:26 Bike_ [~Glossina@174-25-38-57.ptld.qwest.net] has joined #sbcl 01:24:31 -!- ebobby [~fms@199.21.86.106] has quit [Ping timeout: 264 seconds] 01:26:14 -!- Bike [~Glossina@174-25-38-57.ptld.qwest.net] has quit [Ping timeout: 240 seconds] 01:26:25 -!- Bike_ is now known as Bike 01:27:58 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Remote host closed the connection] 01:32:11 -!- Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has quit [Ping timeout: 252 seconds] 01:32:15 Bike_ [~Glossina@174-25-54-20.ptld.qwest.net] has joined #sbcl 01:35:10 -!- Bike [~Glossina@174-25-38-57.ptld.qwest.net] has quit [Ping timeout: 258 seconds] 01:42:49 -!- Bike_ is now known as Bike 01:50:01 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Remote host closed the connection] 01:54:34 -!- loke [~loke@203.127.16.194] has quit [Remote host closed the connection] 01:56:30 loke [~loke@203.127.16.194] has joined #sbcl 02:54:32 Bike_ [~Glossina@174-25-42-96.ptld.qwest.net] has joined #sbcl 02:54:41 -!- Bike [~Glossina@174-25-54-20.ptld.qwest.net] has quit [Disconnected by services] 02:54:54 -!- Bike_ is now known as Bike 03:20:25 jl_2 [~jl_2@174-16-222-121.hlrn.qwest.net] has joined #sbcl 03:27:20 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds] 03:44:30 -!- LiamH [~none@pool-74-96-4-63.washdc.east.verizon.net] has quit [Ping timeout: 264 seconds] 03:46:34 -!- erikc [~erikc@CPE00222d53fe78-CM00222d53fe75.cpe.net.cable.rogers.com] has quit [Quit: erikc] 04:12:13 Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has joined #sbcl 04:24:09 attila_lendvai [~attila_le@92.47.220.76] has joined #sbcl 04:24:09 -!- attila_lendvai [~attila_le@92.47.220.76] has quit [Changing host] 04:24:09 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 04:58:03 sdemarre [~serge@91.176.241.189] has joined #sbcl 05:00:38 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 05:19:10 pranavrc [~pranavrc@117.193.195.111] has joined #sbcl 05:19:10 -!- pranavrc [~pranavrc@117.193.195.111] has quit [Changing host] 05:19:10 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 05:35:42 -!- Fare [fare@nat/google/x-aomcdazzzscpvxom] has quit [Ping timeout: 252 seconds] 05:36:34 -!- slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has quit [Remote host closed the connection] 05:37:23 slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has joined #sbcl 05:52:02 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 05:56:49 attila_lendvai [~attila_le@92.47.220.76] has joined #sbcl 05:56:49 -!- attila_lendvai [~attila_le@92.47.220.76] has quit [Changing host] 05:56:49 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 06:08:30 -!- sdemarre [~serge@91.176.241.189] has quit [Ping timeout: 264 seconds] 06:15:24 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 06:21:31 tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has joined #sbcl 06:31:36 prxq [~mommer@mnhm-5f75e30b.pool.mediaWays.net] has joined #sbcl 06:55:22 -!- ASau` is now known as ASau 06:56:39 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: mental blackout] 06:58:45 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 07:03:49 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 07:13:39 Munksgaard [~philip@80-71-135-82.u.parknet.dk] has joined #sbcl 07:27:01 -!- tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has quit [Quit: Leaving.] 07:27:27 tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has joined #sbcl 07:28:04 -!- tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has quit [Client Quit] 07:30:25 foreignFunction [~niksaak@94.27.88.7] has joined #sbcl 07:38:23 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 07:40:27 -!- Bike [~Glossina@174-25-42-96.ptld.qwest.net] has quit [Quit: tire] 07:41:28 Odyessus [~odyessus@089144192006.atnat0001.highway.a1.net] has joined #sbcl 07:56:06 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Remote host closed the connection] 08:11:44 -!- Odyessus [~odyessus@089144192006.atnat0001.highway.a1.net] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 08:21:44 jl_3 [~jl_2@184-96-179-27.hlrn.qwest.net] has joined #sbcl 08:24:05 -!- jl_2 [~jl_2@174-16-222-121.hlrn.qwest.net] has quit [Ping timeout: 248 seconds] 08:31:54 -!- yacks [~py@180.151.36.168] has quit [Ping timeout: 264 seconds] 08:47:00 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 264 seconds] 08:47:08 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 08:47:16 kanru` [~kanru@226.Red-83-58-222.dynamicIP.rima-tde.net] has joined #sbcl 08:58:42 attila_lendvai1 [~attila_le@92.47.220.76] has joined #sbcl 08:58:42 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 09:08:21 easye [~user@213.33.70.157] has joined #sbcl 09:08:28 yacks [~py@180.151.36.168] has joined #sbcl 09:17:21 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 09:30:17 rudi [~rudi@1x-193-157-205-6.uio.no] has joined #sbcl 10:04:37 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: all hope lost] 10:06:04 davazp [~user@31.200.155.22] has joined #sbcl 10:32:00 -!- kanru` [~kanru@226.Red-83-58-222.dynamicIP.rima-tde.net] has quit [Ping timeout: 276 seconds] 10:43:42 stassats [~stassats@wikipedia/stassats] has joined #sbcl 11:08:09 -!- attila_lendvai1 [~attila_le@92.47.220.76] has quit [Read error: Operation timed out] 11:14:40 -!- Munksgaard [~philip@80-71-135-82.u.parknet.dk] has quit [Ping timeout: 246 seconds] 11:25:45 attila_lendvai [~attila_le@92.47.220.76] has joined #sbcl 11:25:45 -!- attila_lendvai [~attila_le@92.47.220.76] has quit [Changing host] 11:25:45 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 11:26:21 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 11:32:01 kanru` [~kanru@93.Red-79-147-88.dynamicIP.rima-tde.net] has joined #sbcl 11:37:35 Odyessus [~odyessus@guest-wlan.highway.telekom.at] has joined #sbcl 11:37:54 easye [~user@213.33.70.157] has joined #sbcl 12:06:08 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 12:07:31 -!- kanru` [~kanru@93.Red-79-147-88.dynamicIP.rima-tde.net] has quit [Read error: Connection reset by peer] 12:10:31 Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has joined #sbcl 12:14:51 LiamH [~none@129-2-129-147.wireless.umd.edu] has joined #sbcl 12:15:02 -!- davazp [~user@31.200.155.22] has quit [Ping timeout: 245 seconds] 12:22:45 davazp [~user@31.200.155.22] has joined #sbcl 12:29:23 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 12:42:22 -!- yacks [~py@180.151.36.168] has quit [Quit: Leaving] 12:45:08 kanru` [~kanru@54.Red-83-44-206.dynamicIP.rima-tde.net] has joined #sbcl 12:50:44 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 12:52:18 drmeister [~drmeister@mobile-166-147-109-052.mycingular.net] has joined #sbcl 12:55:32 davorb [~davor@hallonpaj.df.lth.se] has joined #sbcl 12:58:08 scoofy [~scoofy@catv-89-135-71-167.catv.broadband.hu] has joined #sbcl 13:05:51 -!- Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has quit [Ping timeout: 245 seconds] 13:11:02 -!- drmeister [~drmeister@mobile-166-147-109-052.mycingular.net] has quit [Remote host closed the connection] 13:11:37 yacks [~py@180.151.36.168] has joined #sbcl 13:24:23 *stassats* thinks about starting a blog "the things i realized during plowing through SBCL internals" 13:25:35 though most of the conclusions seem trivial in hindsight, many of them have taken other prerequisite knowledge, maybe that could save somebody else a bit of time 13:26:11 stassats: that would likely be a good thing. 13:26:22 Fare [fare@nat/google/x-dvvhcvyovunwtugd] has joined #sbcl 13:28:14 now that slime-tips is dead, time for sbcl-internals-tips 13:28:54 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 13:32:53 -!- LiamH [~none@129-2-129-147.wireless.umd.edu] has quit [Ping timeout: 248 seconds] 13:44:53 wbooze [~wbooze@xdsl-78-35-164-112.netcologne.de] has joined #sbcl 13:48:46 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 245 seconds] 13:55:23 -!- rudi [~rudi@1x-193-157-205-6.uio.no] has quit [Quit: Textual IRC Client: http://www.textualapp.com/] 13:59:57 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:03:49 -!- rpg [~rpg@216.243.156.16.real-time.com] has quit [Ping timeout: 248 seconds] 14:19:18 -!- kanru` [~kanru@54.Red-83-44-206.dynamicIP.rima-tde.net] has quit [Ping timeout: 245 seconds] 14:31:25 kanru` [~kanru@201.Red-88-12-18.staticIP.rima-tde.net] has joined #sbcl 14:33:07 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 14:35:07 dalecooper [~chris@xdsl-89-0-225-220.netcologne.de] has joined #sbcl 14:36:30 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:39:42 drmeiste_ [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:39:53 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 14:40:21 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 14:40:56 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:40:58 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Read error: Connection reset by peer] 14:41:06 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:41:51 -!- Fare [fare@nat/google/x-dvvhcvyovunwtugd] has quit [Ping timeout: 252 seconds] 14:45:35 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:50:28 Fare [fare@nat/google/x-umqguwrllncklsfk] has joined #sbcl 14:52:30 Munksgaard [~philip@80-71-135-82.u.parknet.dk] has joined #sbcl 14:53:37 -!- drmeiste_ [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 14:53:44 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:55:54 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Ping timeout: 264 seconds] 14:55:55 attila_lendvai [~attila_le@92.46.23.135] has joined #sbcl 14:55:55 -!- attila_lendvai [~attila_le@92.46.23.135] has quit [Changing host] 14:55:55 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 14:56:07 psilord [~psilord@23-25-144-220-static.hfc.comcastbusiness.net] has joined #sbcl 15:00:51 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 245 seconds] 15:06:04 rpg_ [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has joined #sbcl 15:06:41 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 15:06:55 Bike [~Glossina@174-25-42-96.ptld.qwest.net] has joined #sbcl 15:08:55 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 15:09:07 -!- psilord [~psilord@23-25-144-220-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 15:09:15 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 15:16:30 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Ping timeout: 252 seconds] 15:17:21 attila_lendvai [~attila_le@92.46.23.135] has joined #sbcl 15:17:21 -!- attila_lendvai [~attila_le@92.46.23.135] has quit [Changing host] 15:17:21 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 15:36:59 stassats: for that, or in general for internals, consider a wiki instead of a series of static posts... 15:41:25 -!- Munksgaard [~philip@80-71-135-82.u.parknet.dk] has quit [Ping timeout: 248 seconds] 15:43:30 Munksgaard [~philip@80-71-135-72.u.parknet.dk] has joined #sbcl 15:51:47 -!- asedeno [asedeno@nat/google/x-cwjftwvkibmhnxes] has quit [Remote host closed the connection] 16:00:00 TuckerD [~tucker@c-98-216-148-67.hsd1.nh.comcast.net] has joined #sbcl 16:00:27 asedeno [asedeno@nat/google/x-mrmpgvwswoznwuyk] has joined #sbcl 16:01:32 -!- dalecooper [~chris@xdsl-89-0-225-220.netcologne.de] has quit [Quit: dalecooper] 16:02:43 -!- asedeno [asedeno@nat/google/x-mrmpgvwswoznwuyk] has quit [Client Quit] 16:03:19 asedeno [asedeno@nat/google/x-moxlyuctlwwlrgoi] has joined #sbcl 16:05:56 -!- Odyessus [~odyessus@guest-wlan.highway.telekom.at] has quit [Excess Flood] 16:17:08 nyef [~nyef@c-76-119-183-159.hsd1.ma.comcast.net] has joined #sbcl 16:19:19 erikc [~erikc@209.20.28.194] has joined #sbcl 16:23:49 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 16:26:08 -!- rpg_ [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has quit [Quit: rpg_] 16:26:13 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 16:27:27 -!- bege [~bege@S0106001d7e5132b0.ed.shawcable.net] has quit [Quit: leaving] 16:29:13 bege [~bege@S0106001d7e5132b0.ed.shawcable.net] has joined #sbcl 16:32:07 Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has joined #sbcl 16:43:18 -!- TuckerD [~tucker@c-98-216-148-67.hsd1.nh.comcast.net] has quit [Ping timeout: 264 seconds] 16:46:35 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 16:47:42 -!- Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has quit [Ping timeout: 276 seconds] 16:50:35 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 16:52:38 -!- kanru` [~kanru@201.Red-88-12-18.staticIP.rima-tde.net] has quit [Ping timeout: 245 seconds] 16:52:44 -!- wbooze [~wbooze@xdsl-78-35-164-112.netcologne.de] has quit [Remote host closed the connection] 16:54:53 wbooze [~wbooze@xdsl-78-35-164-112.netcologne.de] has joined #sbcl 16:55:59 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 16:57:11 easye [~user@213.33.70.157] has joined #sbcl 17:05:57 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 17:06:53 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 17:23:56 attila_lendvai [~attila_le@92.46.23.135] has joined #sbcl 17:23:58 -!- attila_lendvai [~attila_le@92.46.23.135] has quit [Changing host] 17:23:58 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:28:56 -!- erikc [~erikc@209.20.28.194] has left #sbcl 17:33:26 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 17:35:01 -!- slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 256 seconds] 17:37:18 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 17:37:36 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 17:37:48 attila_lendvai1 [~attila_le@92.46.23.135] has joined #sbcl 17:37:48 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 17:40:24 easye [~user@213.33.70.157] has joined #sbcl 17:40:55 MrOrdinaire [~minh@155.69.163.245] has joined #sbcl 17:41:57 hi, I've just clone the code from the github repo and built it without problems. However, when running tests, i'm getting the following errors 17:41:59 Finished running tests. 17:42:02 Status: 17:42:04 Expected failure: packages.impure.lisp / USE-PACKAGE-CONFLICT-SET 17:42:06 Expected failure: packages.impure.lisp / IMPORT-SINGLE-CONFLICT 17:42:08 Unhandled Error run-program.impure.lisp 17:42:10 Expected failure: walk.impure.lisp / (WALK-LET* HAIRY-SPECIALS) 17:42:12 Expected failure: walk.impure.lisp / (WALK-LET* HAIRY-SPECIALS) 17:42:13 (4 tests skipped for this combination of platform and features) 17:42:16 test failed, expected 104 return code, got 1 17:42:21 is this expected from the current tip? 17:42:40 does it say expected? 17:42:44 Mmm. The one to investigate would be the run-program.impure.lisp failure, the others are definitely expected. 17:43:15 ... I might know of another bug in the package system, but I haven't taken the time to track it down and work up a test case. 17:43:48 Hasn't bitten me recently, either, but that might be because I haven't done the thing that causes it to so do. 17:45:55 <|3b|> are there still tests that expect uncommon binaries on the system? 17:46:12 *|3b|* would suspect that if so 17:47:56 <|3b|> MrOrdinaire: do you have /bin/cat and /bin/ed? (assuming linux or similar) 17:48:02 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 17:48:33 what about compiling special C programs, and then calling them? 17:48:39 or maybe even calling sbcl itself 17:48:49 |3b|: I do 17:48:52 <|3b|> and /bin/echo /usr/bin/perl /bin/sh 17:49:07 unhandled error means the test is outside of with-test 17:49:23 |3b|: yes, all of them 17:49:35 i'm running ArchLinux by the way 17:50:23 when I run the script with `sbcl --script run-program.impure.lisp`, I got 17:50:25 unhandled UNDEFINED-FUNCTION in thread # {10029E90D3}>: 17:50:28 <|3b|> looks like it runs at least /bin/ed outside with-test 17:50:30 The function :NAME is undefined. 17:50:35 MrOrdinaire: you would need to do sh ./run-tests.sh run-program.impure.lisp 17:50:59 beter yet, &> log and paste it to http://paste.lisp.org 17:52:00 it turns out I don't have `ed` 17:52:07 installing it and the test runs fine now 17:52:09 thanks 17:52:44 right, (defparameter *ed* (run-program "/bin/ed" (list *tmpfile*) :wait nil :pty t)) is outside of with-test 17:54:00 |3b|, stassats, nyef: thanks for the help 17:54:48 I will run the whole testsuit tomorrow 17:55:35 -!- MrOrdinaire [~minh@155.69.163.245] has left #sbcl 18:03:22 attila_lendvai [~attila_le@92.46.23.135] has joined #sbcl 18:03:22 -!- attila_lendvai [~attila_le@92.46.23.135] has quit [Changing host] 18:03:22 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 18:04:22 rpg_ [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has joined #sbcl 18:04:56 -!- rpg_ [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 18:05:02 -!- attila_lendvai1 [~attila_le@92.46.23.135] has quit [Read error: Connection reset by peer] 18:05:04 rpg_ [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 18:05:32 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Read error: Connection reset by peer] 18:10:57 TuckerD [84b1ec3b@gateway/web/freenode/ip.132.177.236.59] has joined #sbcl 18:12:10 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 18:12:26 -!- TuckerD [84b1ec3b@gateway/web/freenode/ip.132.177.236.59] has quit [Client Quit] 18:12:26 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 18:13:17 TuckerD [84b1ec3b@gateway/web/freenode/ip.132.177.236.59] has joined #sbcl 18:14:36 -!- wbooze [~wbooze@xdsl-78-35-164-112.netcologne.de] has quit [Read error: Connection reset by peer] 18:15:55 easye [~user@213.33.70.157] has joined #sbcl 18:16:04 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 18:19:44 wbooze [~wbooze@xdsl-78-35-135-176.netcologne.de] has joined #sbcl 18:22:20 scymtym [~user@ip-5-147-116-166.unitymediagroup.de] has joined #sbcl 18:22:38 -!- specbot [~specbot@50.7.166.114] has quit [Remote host closed the connection] 18:22:38 -!- minion [~minion@50.7.166.114] has quit [Remote host closed the connection] 18:25:21 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 18:34:41 I thought ed was required by POSIX ;) 18:36:47 I have enough trouble with the whole bit where "time" isn't always available. 18:36:55 ... I think it was "time", at least... 18:36:59 yeah 18:38:52 And trying to set up a new chroot for a development environment? Ugh! "Oh, wait, I need git as well... and why doesn't this thing have less?!?" 18:41:02 when you run a script as /bin/sh, bash disables all built-in commands 18:46:18 -!- TuckerD [84b1ec3b@gateway/web/freenode/ip.132.177.236.59] has quit [Ping timeout: 245 seconds] 18:46:19 better use /usr/bin/env bash and laugh when people complain they don't have bash 18:47:51 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 18:55:24 -!- davazp [~user@31.200.155.22] has quit [Ping timeout: 252 seconds] 18:59:13 Krystof: have you seen Tom Emerson's email on sbcl-devel? 19:00:32 minion [~minion@50.7.166.114] has joined #sbcl 19:00:54 specbot [~specbot@50.7.166.114] has joined #sbcl 19:01:00 -!- minion [~minion@50.7.166.114] has quit [Remote host closed the connection] 19:01:00 -!- specbot [~specbot@50.7.166.114] has quit [Read error: Connection reset by peer] 19:04:53 slyrus [~chatzilla@173-228-44-92.dsl.static.sonic.net] has joined #sbcl 19:07:25 specbot [~specbot@50.7.166.114] has joined #sbcl 19:07:25 minion [~minion@50.7.166.114] has joined #sbcl 19:08:52 yes; will reply 19:08:53 thanks 19:09:12 -!- slyrus [~chatzilla@173-228-44-92.dsl.static.sonic.net] has quit [Remote host closed the connection] 19:13:11 robgssp [~user@2620:8d:8000:e50:21c:c0ff:fea3:7cc5] has joined #sbcl 19:17:42 have there been mach-port-related fixes after 1.0.57? in https://code.cor-lab.org/issues/1462, a colleague of mine seemingly had mach-port-related issues disappearing after switching 1.0.57 -> 1.1.5 19:18:10 scymtym: yes 19:18:33 pkhuong: so its plausible that the update actually fixed the problem? 19:18:56 scymtym: mach doesn't like > 32 bit mach port names. It manifested as exactly that error message. 19:19:45 pkhuong: i'm happy to this :) 19:20:47 i was a bit worried because a quick search didn't turn up any mach-port-related commits after 1.0.47 19:20:48 I'm trying to understand another report on 10.8.0, if you have more issues. (There's a fix, I just don't see why it works) 19:21:08 where 10.8.0 is the macos version? 19:21:32 I'm still running 10.7.5, so I can't help you there. 19:21:46 It's 578362575fc2112b828597cc1025e3ead43d43ba (Fix threads on Darwin 10.8) 19:22:09 I'm pretty sure that's post 57 (: 19:22:31 pkhuong: thanks. my colleague uses the code on a regular basis and would probably be willing to help, if more issues show up 19:22:32 Right, made it in .58. 19:23:00 scymtym: right, darwin 10.8.0 (I'm not seeing it here, on 10.8.2, fwiw) 19:24:30 pkhuong: i missed that commit earlier because it didn't have "port" in the subject 19:24:48 sdemarre [~serge@91.176.241.189] has joined #sbcl 19:25:42 pkhuong: should i ask my colleague to run sbcl unit tests on his machine? 19:26:11 Always run the tests after a build. (-: 19:26:29 dknight [~amitav@114.143.157.70] has joined #sbcl 19:27:19 nyef: in this case, I'm distributing binaries built on a different machine 19:28:12 scymtym: it fails during the build in fact (: 19:30:16 pkhuong: ok :) 19:32:05 -!- dknight [~amitav@114.143.157.70] has left #sbcl 19:32:20 rpg [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has joined #sbcl 19:32:47 -!- rpg [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 19:32:48 -!- rpg_ [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Read error: Connection reset by peer] 19:32:58 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 19:36:35 -!- specbot [~specbot@50.7.166.114] has quit [Remote host closed the connection] 19:36:36 -!- minion [~minion@50.7.166.114] has quit [Remote host closed the connection] 19:37:06 tcr1 [~tcr@46-126-110-164.dynamic.hispeed.ch] has joined #sbcl 19:37:43 I'm trying a sparc/linux build right now. Am I likely to run into any problems? 19:39:36 dknight [~dknight@114.143.157.70] has joined #sbcl 19:39:39 minion [~minion@50.7.166.114] has joined #sbcl 19:40:01 specbot [~specbot@50.7.166.114] has joined #sbcl 19:41:30 -!- scoofy [~scoofy@catv-89-135-71-167.catv.broadband.hu] has quit [Ping timeout: 264 seconds] 20:02:36 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 20:02:41 pkhuong: hey 20:03:40 clhs list 20:03:40 http://www.lispworks.com/reference/HyperSpec/Body/a_list.htm 20:03:42 clist list/f 20:03:46 clhs list/f 20:03:46 http://www.lispworks.com/reference/HyperSpec/Body/f_list_.htm 20:03:48 clhs list/t 20:03:49 http://www.lispworks.com/reference/HyperSpec/Body/t_list.htm 20:04:00 clhs //v 20:04:00 http://www.lispworks.com/reference/HyperSpec/Body/v_sl_sls.htm 20:04:25 eh, that was supposed to #lisp! 20:08:32 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:10:39 dknight: yes? 20:12:26 -!- foreignFunction [~niksaak@94.27.88.7] has quit [Quit: Leaving.] 20:20:22 pkhuong: well I was just curious about the task I was talking about yesterday 20:21:26 pkhuong: I understand you might not have got time. I was just thinking if you can just point me to some links with more details about it. 20:22:43 davazp [~user@31.200.155.22] has joined #sbcl 20:23:22 Hashing, right? 20:24:35 pkhuong: yes 20:30:59 jsnell's first sbcl project was hashing 20:31:04 so, there's precedent :-) 20:31:46 Krystof: that is nice 20:33:48 Do we have a test case that specifically tests for with-pinned-objects not consing? 20:34:33 leuler [~user@p548FC465.dip.t-dialin.net] has joined #sbcl 20:34:52 TuckerD [~tucker@c-98-216-148-67.hsd1.nh.comcast.net] has joined #sbcl 20:37:52 dknight: from your introductory statement, I get the impression that you're not already a lisp hacker (let alone an sbcl one) -- so maybe the first step from your side is to start playing? 20:39:30 get sbcl built, hit M-. on sxhash, poke around the source? Maybe have a look at the sb-md5 contrib (including how it's implemented)? 20:43:21 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: rpg] 20:46:25 Krystof: :) yes already at it. well I have done some lisp coding. it was an edit to some emacs package. not much of course, so yeah I will play around with it more 20:49:04 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 20:49:32 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:50:52 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 245 seconds] 20:53:25 -!- Fare [fare@nat/google/x-umqguwrllncklsfk] has quit [Ping timeout: 256 seconds] 20:56:59 Krystof, dknight: Would it be sensible, afterwards, as a first useful change, to make the various constants in sxhash depend on the word size so that they are larger under 64 bit? (According to CLHS one should target the full fixnum range). This change would probably be superseded by later work, but may be a nice exercise? 20:58:03 leuler: sure. isn't there a launchpad ticket on that? 20:58:41 May be. Searching ... 20:59:47 There is lp #309443, but that is more about MIX, I think. 20:59:55 lp 309443 20:59:55 https://bugs.launchpad.net/bugs/309443 20:59:59 -!- prxq [~mommer@mnhm-5f75e30b.pool.mediaWays.net] has quit [Quit: Leaving] 21:00:16 309443 21:03:36 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 21:04:27 -!- davazp [~user@31.200.155.22] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 21:05:11 -!- Bike [~Glossina@174-25-42-96.ptld.qwest.net] has quit [Ping timeout: 252 seconds] 21:05:19 I was just thinking about improving the small fixnums, e.g. currently (sxhash 0) -> 361475658 (x86-64), which firmly is a 29-bit number. 21:05:42 Bike [~Glossina@174-25-42-96.ptld.qwest.net] has joined #sbcl 21:06:28 leuler: okay 21:06:35 I'm really not sure what that is supposed to be. 21:07:58 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 21:09:06 -!- sdemarre [~serge@91.176.241.189] has quit [Read error: Operation timed out] 21:10:20 Bike_ [~Glossina@174-25-42-96.ptld.qwest.net] has joined #sbcl 21:10:24 oh, what. "; to get sign bit into hash"... I don't think that code was ever tested. 21:10:40 -!- Bike [~Glossina@174-25-42-96.ptld.qwest.net] has quit [Disconnected by services] 21:10:44 -!- Bike_ is now known as Bike 21:10:52 ASau` [~user@p4FF9657F.dip.t-dialin.net] has joined #sbcl 21:11:53 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Read error: Connection reset by peer] 21:12:20 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 21:13:12 Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has joined #sbcl 21:13:35 pkhuong: What's that "that" that you don't know what it should be? 21:13:59 -!- ASau [~user@p5797EF08.dip.t-dialin.net] has quit [Ping timeout: 252 seconds] 21:15:50 the routine to sxhash fixnums. 21:16:02 looks like it was designed back in the days when multiplications were really slow. 21:16:38 OK. 21:22:13 So, change the formulas for sxhash on fixnums and other non-compound objects to something better and make their parameters word-size dependent in one go. 21:23:52 So, test failures. Worse than I had hoped, better than I had feared: http://paste.lisp.org/display/136749 21:24:16 nyef: but is it colored? 21:24:31 the colors should cheer you up 21:24:33 Yes, but that doesn't come through in the paste. 21:25:12 do you have /bin/ed? 21:25:40 Nope. But that shouldn't account for most of the errors. 21:25:50 just for Unhandled Error run-program.impure.lisp 21:26:16 That's so far down the list... 21:30:36 nyef: i built mine with --fancy and the only error i had was run-program.impure.lisp which was fixed by installing ed 21:30:56 Unhandled Error that is 21:33:28 nicdev: that wasn't on sparc. 21:33:34 nicdev: I bet you built yours on an x86oid system. 21:36:25 nyef: that's true, my bad, i did not check the architecture in question 21:37:41 It's been a while, so I figured I'd at least try to build it and run the test suite. 21:38:16 leuler: I wonder how often code is bottlenecked on computing hash values though (: 21:38:28 Of course, this leaves building a cheneygc version, and possibly also adding thread support. 21:42:19 pkhuong: I think a reasonably slowdown to achieve better results is completely fine (I still only comment on non-compound things here), so I would immediately approve for instance a three times slower fixnum hash if it's better distributed. 21:43:24 leuler: same here. I'm thinking we could have something much more robust if we only inlined enough code to avoid consing, and computed hash values out of line. 21:45:16 Do we currently inline every occurrence of sxhash? Or just too many of them? 21:46:19 Sorry, just looked it up again. Lots of deftransforms. 21:46:33 yeah. We inline pretty much all the immediate types 21:48:40 sxhash is pretty useless anyways. ;p 21:49:58 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 21:50:20 foom: if it could be almost useful for non-EQ types that'd already be soemthing 21:55:18 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 21:57:49 leuler: for a loop that only increments a double float and hashes it, tabulation hashing (http://arxiv.org/abs/1011.5200) is ~2x as slow as sxhash on my macbook. 21:58:07 Hrm. (:no-consing :hash-tables) is marked :fails-on '(and :ppc :sb-thread), which seems "iffy" to me. 21:58:24 I can likely bring that down a bit by calling out to an assembly routine instead of a full call. 22:00:50 Okay, looks like most of the dynamic-extent.impure.lisp failures just need updating for sparc, while one just looks iffy to begin with. 22:03:15 drmeister [~drmeister@mobile-166-147-109-052.mycingular.net] has joined #sbcl 22:06:31 Fare [fare@nat/google/x-rjlbryrovwowvmbn] has joined #sbcl 22:06:31 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Ping timeout: 246 seconds] 22:08:00 -!- jdz [~jdz@85.254.212.34] has quit [Excess Flood] 22:08:36 jdz [~jdz@85.254.212.34] has joined #sbcl 22:12:45 The interface.pure.lisp failure is easily repeatable, the disassembler is doing something that make-lisp-obj doesn't like. 22:13:51 pkhuong: Didn't we talk about that already some months ago? I fear table accesses would slow down other parts of the user's code by throwing out useful cache entries. 22:14:13 Okay, giving up on this for a bit. Looks like it should be a grab-bag of easy stuff and more-involved stuff. 22:15:36 leuler: might have been on logcount ;) 22:21:09 The way I see it, occasional sxhash won't hurt that much. If sxhash is called a lot though, the much improved distribution compared to stuff like multiply/mod must be worth it. 22:23:02 How large would these tables be all together? 22:24:52 On x86-64, something like 16KB for all boxed immediates, and 16 KB for doubles. 22:24:57 -!- nyef [~nyef@c-76-119-183-159.hsd1.ma.comcast.net] has quit [Quit: G'night all.] 22:26:35 And we can make sure there's some locality to accesses for the immediates. 22:26:52 eesh, that'd end up with not much L1 left for the working set, wouldn't it? 22:26:53 How that? 22:26:54 That seems like a bad idea. 22:27:11 the only point to having a good hash function is to avoid collisions in the hashtable. 22:27:16 the question is rather the size of the working set. 22:27:31 and basically the only point of avoiding collisions of the hashtable is to avoid extra memory accesses. 22:27:50 foom: Good point! 22:28:08 foom: bad hash functions catastrophically increase the number of memory accesses. 22:28:59 -!- drmeister [~drmeister@mobile-166-147-109-052.mycingular.net] has quit [Remote host closed the connection] 22:29:39 leuler: if the ith byte has value b, read table[b*8+i]. Zero bytes will all hit the same region. 22:30:16 but there are *so many* choices of reasonable hash functions out there. 22:30:19 that don't use memory 22:30:36 foom: reasonable? Only if you're willing to go to heroic lengths in your hash table. 22:32:22 is something like murmurhash too slow (or not good enough?) 22:32:47 I don't think mummurhash is geared toward hashing 64 bits into 64 bits. 22:33:39 I remember hearing FNV-1a was an okay thing too (?) 22:34:14 http://paste.lisp.org/display/136751 22:34:16 Fiora: same. You're thinking of string hash functions. 22:34:18 *stassats* came up with a way to optimize integer-length for fixnums with n-fixnum-tag-bits == 1: instead of untagging, doing bsr, checking if ZF is set and zeroing the result or incrementing the result, just doing OR RDX, 1 BSR RDX, RDX 22:34:43 ah... they don't work okay for single 64-bit chunks? 22:34:43 That's what I'm using, I'm sure it's from some literature's standard hash function, but I don't know which one. 22:34:56 not untagging will make the result 1 bigger, no increment needed, and OR 1 will not require checking if BSR sets ZF in case it's 0 22:35:09 (are they just, collision-tastic? or too slow?) 22:35:29 stassats: nice. 22:35:38 stassats: nice! 22:36:32 n-fixnum-tag-bits > 1 can be optimized too, just reducing the amount of shifting for untagging 22:36:35 foom: Does that thing have a name? 22:36:44 leuler: I'm sure it does, but I don't know it 22:36:57 foom: Where do you have it from? 22:40:00 stassats: seems simpler to me to OR in all the low bits and subtract 22:40:13 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 22:41:17 leuler: Carl deMarken. 22:41:29 Anyway, I need to quit. dknight: You were silent for a while. I hope the fraying discussion didn't alienate you. Bye. 22:42:16 Bike_ [~Glossina@67-5-198-42.ptld.qwest.net] has joined #sbcl 22:43:33 (but I know he didn't invent it; it's from some hash literature, I just don't know how to search for it) 22:44:05 foom: yeah, I've seen something like that before. 22:44:09 savings are -8 bytes, no branches, (loop for i to 200000000 do (test i)) => 700 ms to 600 ms 22:44:20 -!- leuler [~user@p548FC465.dip.t-dialin.net] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 22:44:29 pkhuong: works too, yeah 22:45:39 -!- Bike [~Glossina@174-25-42-96.ptld.qwest.net] has quit [Ping timeout: 256 seconds] 22:45:44 -!- Bike_ is now known as Bike 22:45:48 and that's for positive fixnums, of course 22:46:03 all fixnums can be improved too 22:46:04 here you go: http://web.archive.org/web/20110622054546/http://www.cris.com/~Ttwang/tech/inthash.htm 22:46:16 foom: thing is, we've only learned recently that simple hash tables like linear probing only works if you have really solid hash functions. The analysis only makes sense if we have a family of hash function, but the classic simple families (multiply-add-shift, for instance) all fail to guarantee good performance with high probability. 22:47:00 If we stick to more complicated overflow schemes like chaining, then it's ok, but we're probably not too worried about the number of memory accesses then. 22:47:25 -!- tcr1 [~tcr@46-126-110-164.dynamic.hispeed.ch] has quit [Quit: Leaving.] 22:47:40 ahhh, that makes sense. so the tradeoff is "good hash function + linear probing" vs "worse hash function + fancier things"... 22:48:24 I actually used that hash function with a linear chaining hashtable, but I didn't do the analysis on it to show that it was actually a good idea and not me being lazy. :P 22:49:11 foom: has some nice graphs at the end (page 71), for both timings and number of memory accesses. 22:53:01 also, I'm more than OK with a couple L1D misses to avoid TLB misses in the hash table (: 22:54:37 stassats: also, you can subtract and tag with a single LEA ;) 22:57:51 pkhuong: interesting, but it certainly doesn't look like an obvious win for a table-based hash. 23:01:40 foom: depends if you're worried about the average or near-worst case. 23:02:51 davazp [~user@212.129.66.115] has joined #sbcl 23:03:49 Well, I didn't read terribly closely, but the best one looked to be cwtrick which isn't table based? 23:04:31 anyways, the univ and univ2 things they tested were certainly terrible functions. 23:05:23 oh, sorry, read the wrong line. their table thing is faster than cwtrick. 23:05:32 it looks like it depends on the computer? I'm guessing cwtrick32 relies really heavily on the fast 32x32->64 multiply 23:08:02 foom: so terrible that they're still the classic universal hash function. I'm pretty sure Knuth says something like "this class of hash function is nice and simple, and the version you get by setting the multiplier to some Phi-related constant works well" 23:08:04 the gap between the table and CWtrick version gets huge for the 64-bit one though... 23:08:18 I guess that makes sense, the cwtrick64 is really complicated. 23:08:27 Fiora: one of the computers is in 32 bit mode. 23:08:42 Fiora is looking at the table on 69 now, right? 23:08:58 yea, "Computer A" is a 32-bit computer. 23:09:27 given the code, no wonder it's so slow on 32-bit. 23:09:31 Anyways, I don't find it terribly convincing; they tested two known-to-be-poor hash functions, one that worked well but is slow, and a table implementation. 23:09:46 MultAddPrime89 looks like a total nightmare on 32-bit 23:11:18 foom: until recently they were known to be good... 23:12:09 No way. A single multiplication has been known to be poor unless you used a prime-number-sized hashtable for ages. 23:14:06 wait, yeah, they use a power of two hash table size :\ 23:14:53 not that it's an issue for their test cases. 23:15:23 -!- dknight [~dknight@114.143.157.70] has left #sbcl 23:21:27 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 23:27:00 anyway. I'd be more comfortable if we chose from a known strong family of hash functions than with ad hoc functions that look random enough. Currently, our hash tables can cope with pretty weak hashes, but I'm not sure it's ideal to be stuck with hash tables that are OK with weak hash functions. 23:27:49 -!- Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has quit [Remote host closed the connection] 23:30:08 Hydan [~hydan@ip-89-103-110-116.net.upcbroadband.cz] has joined #sbcl 23:31:52 Also, the results I posted is older than the arxiv one, where they use a much simpler tabular hash and show that it's still good for open addressing. 23:32:38 it would be nice if disassemble could show from which VOP the code is coming from 23:36:18 stassats: trace the compilation :\ 23:36:34 that's too primitive 23:37:10 i want to be able click on a disassemble portion in slime 23:37:59 that may involve tracing the compilation process under the hood 23:38:19 -!- ASau` [~user@p4FF9657F.dip.t-dialin.net] has quit [Ping timeout: 264 seconds] 23:39:03 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 23:42:02 also... wait, why are we worrying about inlining sxhash: when is it ever called with a known type? 23:42:33 packages? 23:44:16 stassats: not inlined. 23:54:23 -!- redline6561 [~redline65@li69-162.members.linode.com] has quit [Ping timeout: 256 seconds] 23:55:41 redline6561 [~redline65@li69-162.members.linode.com] has joined #sbcl 23:59:54 hm, the new LOGBITP optimization seem to be slower for me