00:10:02 -!- foreignFunction [~niksaak@ip-4761.sunline.net.ua] has quit [Quit: Leaving.] 00:21:09 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 00:23:27 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 00:49:35 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: version mismatch] 01:51:06 attila_lendvai [~attila_le@92.46.9.198] has joined #sbcl 01:51:06 -!- attila_lendvai [~attila_le@92.46.9.198] has quit [Changing host] 01:51:06 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 02:08:12 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 02:19:20 kanru [~kanru@118-163-10-190.HINET-IP.hinet.net] has joined #sbcl 02:39:42 -!- christoph4 [~christoph@ppp-188-174-138-184.dynamic.mnet-online.de] has quit [Ping timeout: 264 seconds] 02:41:25 kanru` [~kanru@118-163-10-190.HINET-IP.hinet.net] has joined #sbcl 02:52:41 christoph4 [~christoph@ppp-188-174-105-2.dynamic.mnet-online.de] has joined #sbcl 03:07:32 LiamH [~none@pool-173-73-133-61.washdc.east.verizon.net] has joined #sbcl 03:07:36 -!- LiamH [~none@pool-173-73-133-61.washdc.east.verizon.net] has quit [Client Quit] 03:47:42 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 03:53:36 attila_lendvai [~attila_le@92.46.9.198] has joined #sbcl 03:53:39 -!- attila_lendvai [~attila_le@92.46.9.198] has quit [Changing host] 03:53:39 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 04:11:28 scymtym [~user@89.31.118.161] has joined #sbcl 04:30:34 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: lights off] 05:20:49 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 05:26:47 -!- wbooze [~wbooze@xdsl-78-35-191-76.netcologne.de] has quit [Ping timeout: 252 seconds] 05:47:43 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 05:56:54 -!- jarod_ch_ [~jarod_che@115.205.239.230] has quit [Quit: Textual IRC Client: http://www.textualapp.com/] 05:57:36 echo-area [~user@182.92.247.2] has joined #sbcl 06:00:33 -!- yacks [~py@180.151.36.168] has quit [Ping timeout: 248 seconds] 06:03:09 yacks [~py@180.151.36.168] has joined #sbcl 06:05:28 sdemarre [~serge@91.176.221.3] has joined #sbcl 06:23:28 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 06:29:10 -!- milosn [~milosn@user-5af50b7a.broadband.tesco.net] has quit [Ping timeout: 246 seconds] 06:31:12 milosn [~milosn@user-5af5051f.broadband.tesco.net] has joined #sbcl 06:38:40 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 06:53:33 milosn_ [~milosn@user-5af5056b.broadband.tesco.net] has joined #sbcl 06:55:53 -!- milosn [~milosn@user-5af5051f.broadband.tesco.net] has quit [Ping timeout: 252 seconds] 07:14:47 -!- Bike [~Glossina@67-5-223-234.ptld.qwest.net] has quit [Ping timeout: 256 seconds] 07:16:16 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 07:48:49 attila_lendvai [~attila_le@87.247.13.235] has joined #sbcl 07:48:49 -!- attila_lendvai [~attila_le@87.247.13.235] has quit [Changing host] 07:48:49 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:49:08 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 07:54:41 -!- kanru` [~kanru@118-163-10-190.HINET-IP.hinet.net] has quit [Ping timeout: 248 seconds] 07:56:44 -!- Posterdati [~antani@host183-239-dynamic.16-87-r.retail.telecomitalia.it] has quit [Read error: Operation timed out] 08:05:35 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Ping timeout: 256 seconds] 08:12:07 Posterdati [~antani@host89-93-dynamic.19-79-r.retail.telecomitalia.it] has joined #sbcl 08:13:18 teggi [~teggi@123.20.50.245] has joined #sbcl 08:34:10 Odyessus [~odyessus@089144192127.atnat0001.highway.a1.net] has joined #sbcl 09:07:28 -!- Odyessus [~odyessus@089144192127.atnat0001.highway.a1.net] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 09:09:31 Odyessus [~odyessus@213.47.71.36] has joined #sbcl 09:20:15 wbooze [~wbooze@xdsl-78-35-133-246.netcologne.de] has joined #sbcl 09:28:03 -!- Odyessus [~odyessus@213.47.71.36] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 09:31:08 tcr1 [~tcr@77-56-40-229.dclient.hispeed.ch] has joined #sbcl 09:33:38 kanru` [~kanru@118-163-10-190.HINET-IP.hinet.net] has joined #sbcl 09:34:15 -!- yacks [~py@180.151.36.168] has quit [Quit: Leaving] 10:16:28 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 10:23:02 yacks [~py@180.151.36.168] has joined #sbcl 10:28:27 davazp [~user@92.251.174.142.threembb.ie] has joined #sbcl 10:29:58 prxq [~mommer@mnhm-590c35e0.pool.mediaWays.net] has joined #sbcl 10:37:09 -!- yacks [~py@180.151.36.168] has quit [Ping timeout: 245 seconds] 10:45:03 -!- kanru` [~kanru@118-163-10-190.HINET-IP.hinet.net] has quit [Ping timeout: 260 seconds] 10:52:02 yacks [~py@180.151.36.168] has joined #sbcl 11:28:25 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 246 seconds] 11:34:55 Odyessus [~odyessus@213.47.71.36] has joined #sbcl 11:44:58 -!- Odyessus [~odyessus@213.47.71.36] has quit [Remote host closed the connection] 11:47:55 -!- sdemarre [~serge@91.176.221.3] has quit [Ping timeout: 256 seconds] 11:50:07 Odyessus [~odyessus@213.47.71.36] has joined #sbcl 12:02:24 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 12:23:31 -!- Odyessus [~odyessus@213.47.71.36] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 12:27:03 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 12:35:16 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 12:50:55 -!- scymtym [~user@89.31.118.161] has quit [Ping timeout: 264 seconds] 13:11:55 -!- davazp [~user@92.251.174.142.threembb.ie] has quit [Ping timeout: 264 seconds] 13:16:13 nyef [~nyef@c-50-157-244-41.hsd1.ma.comcast.net] has joined #sbcl 13:16:22 G'morning all. 13:19:29 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 13:25:33 segv- [~mb@95-91-241-76-dynip.superkabel.de] has joined #sbcl 13:26:43 pranavrc [~pranavrc@122.174.5.22] has joined #sbcl 13:26:44 -!- pranavrc [~pranavrc@122.174.5.22] has quit [Changing host] 13:26:44 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 13:39:32 davazp [~user@92.251.243.192.threembb.ie] has joined #sbcl 13:44:17 to what extent should we exploit associativity and commutativity to enable more constant folding? (e.g. stuff like (logior (logior x constant) constant)) 13:45:01 I'm thinking bitwise ops and rational arithmetic. 13:48:07 sdemarre [~serge@91.176.221.3] has joined #sbcl 13:48:40 -!- davazp [~user@92.251.243.192.threembb.ie] has quit [Ping timeout: 252 seconds] 13:50:50 we do have a float-accuracy policy 13:51:28 so under suitable versions of that we could also do float arithmetic 13:52:17 (and in fact if we do that properly we could remove float constant-folding altogether when float-accuracy is high so that Fateman can be happy) 13:53:27 float-lossiness seems like a better policy setting ;) 13:54:36 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 13:55:24 no objection; it's still sb-c:: internal 13:56:01 drmeiste_ [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 13:59:30 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Ping timeout: 264 seconds] 14:00:44 re float constant folding, would you create a new attribute, or just look at the arguments when constant folding? 14:02:59 foreignFunction [~niksaak@ip-4761.sunline.net.ua] has joined #sbcl 14:04:06 I'm thinking to commit this branch (all eight commits) either this afternoon or tomorrow: http://repo.or.cz/w/sbcl/nyef.git/shortlog/refs/heads/gencgc-hacking-2 14:05:15 pkhuong: I think I'd just look at the arguments 14:05:25 but I could be persuaded otherwise 14:06:24 I clearly still have more work to do in this direction, but I'd like to know that I'm not completely off-base, if I broke or missed anything, and if this sort of work deserves a NEWS entry. (-: 14:07:07 Krystof: looking at the arguments is easier for me. 14:08:01 -!- drmeiste_ is now known as drmeister 14:11:12 scymtym [~user@89.31.118.161] has joined #sbcl 14:14:11 davazp [~user@92.251.169.68.threembb.ie] has joined #sbcl 14:15:58 on x86, is this expected with current master: sh ./run-sbcl.sh --noinform --no-userinit --load ~/.local/share/common-lisp/quicklisp/setup.lisp --eval '(room)' --quit => Unrecognized widetag #x4A in reconstitute-object? 14:16:25 No, that shouldn't be happening. 14:16:38 it does not without the --load 14:16:48 can i investigate further? 14:17:34 Can you look at src/runtime/genesis/constants.h and tell me which _WIDETAG constant has a value of 0x4a? 14:18:13 (I'm on an x86-64 build right now, so my widetag constants are different.) 14:18:46 #define UNBOUND_MARKER_WIDETAG 74 /* 0x4A */ 14:19:16 Okay, who managed to put an unbound marker into a cons cell? 14:19:28 ... Is it a cons cell? 14:19:30 Hrm. 14:20:29 nyef: should i try to find out if any specific part of the loaded file causes the problem? 14:21:46 You could try, though I'm not sure what it would tell us. I'm going to try to reproduce it here as I have time (I'm supposed to be getting a document written for work right now). 14:21:49 nyef: hm, look at fixup-code-object in x86-vm? 14:22:47 nyef: ok 14:22:55 psilord [~pkeller@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:25:01 nyef: the commits sound good to me. 14:25:02 Krystof: What am I looking for? 14:25:45 Krystof: Other than evidence that the whole fixup thing could do with a bit of, if you'll pardon the pun, "fixing up"? 14:26:34 Yes, it's directly manipulating what could be an unbound-marker, but it's doing it entirely in the context of a local variable, which shouldn't get consed to a list, should it? 14:27:29 nyef: it seems to suggest that there could be unbound markers in code-constants 14:30:34 Yeah, hang on, unbound-marker is valid as an immediate for cons cells. 14:30:41 At least, it is according to the runtime. 14:30:54 it could happen in low-safety code, I think. 14:31:30 Oh, ouch. 14:31:32 i can trigger the problem by load asdf.fasl via (require 'asdf) 14:31:36 Okay, this is my bug, clearly. 14:31:40 loading the source does not trigger it 14:32:43 scymtym: Have a look at src/code/room.lisp, search for "cons-info". You'll see exactly how I broke it... And the fix becomes obvious as well, so I'll put together a commit for that today. 14:33:06 -!- davazp [~user@92.251.169.68.threembb.ie] has quit [Ping timeout: 264 seconds] 14:33:36 nyef: ok, having a look 14:34:21 why are you so certain that it's to do with conses? 14:34:50 Because reconstitute-object is called on valid heap objects, and what OTHER valid heap object would start with unbound-marker-widetag? 14:35:05 The other angle is that I somehow screwed up computing the size of an object. 14:35:12 But that's less likely. 14:35:21 if the code-fixup slot can contain the unbound marker, what happens? 14:35:38 ... Nothing, because ROOM doesn't look at it? 14:36:11 ok, taking your word for it. 14:36:35 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Ping timeout: 256 seconds] 14:37:20 I'm slightly surprised that the unbound marker ends up in a cons 14:37:50 As am I, but it's legal according to gc-common.c and runtime.h. 14:38:30 The only part of a code object that ROOM looks at is the header and the code-size slot. 15:28:42 how do you feel about transforming (f (if foo k1 k2) k3) => (if foo (f k1 k3) (f k2 k3)), where k1, k2, k3 are constants? 15:28:53 (and f is foldable) 15:47:03 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 16:10:43 -!- scymtym [~user@89.31.118.161] has quit [Ping timeout: 276 seconds] 16:16:00 Bike [~Glossina@67-5-227-206.ptld.qwest.net] has joined #sbcl 17:06:35 -!- wbooze [~wbooze@xdsl-78-35-133-246.netcologne.de] has quit [Remote host closed the connection] 17:12:50 attila_lendvai [~attila_le@92.46.9.198] has joined #sbcl 17:12:50 -!- attila_lendvai [~attila_le@92.46.9.198] has quit [Changing host] 17:12:50 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:32:38 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 17:42:11 wbooze [~wbooze@xdsl-78-35-133-246.netcologne.de] has joined #sbcl 17:48:22 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 17:51:54 -!- sdemarre [~serge@91.176.221.3] has quit [Quit: Leaving.] 17:55:41 -!- wbooze [~wbooze@xdsl-78-35-133-246.netcologne.de] has quit [Ping timeout: 256 seconds] 17:58:15 -!- joshe [~joshe@opal.elsasser.org] has quit [Remote host closed the connection] 18:00:01 joshe [~joshe@opal.elsasser.org] has joined #sbcl 18:13:34 -!- psilord [~pkeller@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 18:18:07 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Ping timeout: 276 seconds] 18:42:24 davazp [~user@178.167.143.120.threembb.ie] has joined #sbcl 18:54:13 wbooze [~wbooze@xdsl-78-35-149-171.netcologne.de] has joined #sbcl 19:04:50 Okay, I can confirm the "unrecognized widetag #x4a" thing. 19:05:16 And that it's unbound-marker-widetag. 19:08:10 mm... I have patch that pads code objects to 64 bytes. If we always allocate code objects on code pages, that should result in 64 byte alignment, right? 19:08:28 Turns out that we don't always allocate code objects on code pages. 19:08:44 argh. 19:09:01 The exception that I know of is PPC, and I suspect that SPARC also doesn't. 19:12:51 And this definitely feels like a series of two CONSes (# . (98 . NIL)), followed in memory by #. 19:19:11 -!- teggi [~teggi@123.20.50.245] has quit [Remote host closed the connection] 19:21:44 -!- wbooze [~wbooze@xdsl-78-35-149-171.netcologne.de] has quit [Quit: none] 19:28:17 And the "obvious" fix certainly seems to work. Just need to wait for the test suite to finish running and I can commit it locally. 19:28:39 wbooze [~wbooze@xdsl-78-35-149-171.netcologne.de] has joined #sbcl 19:28:58 -!- Bike [~Glossina@67-5-227-206.ptld.qwest.net] has quit [Ping timeout: 246 seconds] 19:33:02 Bike [~Glossina@67-5-227-206.ptld.qwest.net] has joined #sbcl 19:34:04 -!- davazp [~user@178.167.143.120.threembb.ie] has quit [Ping timeout: 252 seconds] 19:35:51 I have a DEFTYPE, do compile-and-load (or use ASDF for that), and get an error. I change the type, re-compile and re-load, and still get the error. 19:35:56 testcase is at http://paste.lisp.org/display/137226 19:36:21 Does that mean that the accessors are not recompiled every time? 19:37:44 this is 1.1.6.debian, BTW 19:40:49 I'm not entirely surprised, TBH, but the whole PCL bit is beyond my ken. /-: 20:12:51 stassats [~stassats@wikipedia/stassats] has joined #sbcl 20:14:33 *stassats* wonders what made this month so conducive to sbcl activities 20:14:51 GSoC! Positive feedback 20:15:09 let's keep on going! Happy happy fun times. 20:30:45 In my case, it's also a matter of having spare bandwidth, though some of the GSoC project ideas were definitely an inspiration. 20:31:20 I'm flushing my queue before focusing on the thesis. 20:37:38 ASau` [~user@p5797E512.dip0.t-ipconnect.de] has joined #sbcl 20:41:09 -!- ASau [~user@p4FF963D1.dip0.t-ipconnect.de] has quit [Ping timeout: 248 seconds] 20:42:05 pkhuong: boooo! 20:42:38 (does your queue include division-by-multiply code? Or paper?) 20:42:41 even today is a bonus for victoria day. 20:42:55 I'll try to do div by mul tonight. 20:43:49 ok. Freeze soon, though -- Wednesday OK? 20:50:17 Fine by me. I'm busier this week anyway, and while I've still got things to do with gencgc, I'd rather work up another queue and land it post-freeze anyway, in case I do manage to screw things up and not notice, in which case having things a little more separated should make it easier to track down any bugs. 20:51:01 Bike_ [~Glossina@67-5-226-252.ptld.qwest.net] has joined #sbcl 20:51:19 -!- Bike [~Glossina@67-5-227-206.ptld.qwest.net] has quit [Disconnected by services] 20:51:24 -!- Bike_ is now known as Bike 20:52:09 so, this clos typecheck thing, generate-slotd-typecheck caches its output 20:52:19 which goes awry when the type expansion changes 20:54:57 Bike_ [~Glossina@67-5-234-210.ptld.qwest.net] has joined #sbcl 20:55:25 -!- Bike [~Glossina@67-5-226-252.ptld.qwest.net] has quit [Disconnected by services] 20:55:28 -!- Bike_ is now known as Bike 20:56:06 can just use an expanded type for caching and compiling 20:56:35 Now you have more overhead on probing the cache, though. 20:57:00 when will that be a problem? it's only during class finalization 20:57:21 Who cares if it's a problem if the alternative is outright incorrect behavior? 21:01:29 And now I have to run. 21:01:31 -!- nyef [~nyef@c-50-157-244-41.hsd1.ma.comcast.net] has quit [Quit: G'night all.] 21:04:05 is sb-kernel:values-specifier-type guaranteed to always return EQ objects for EQ types? 21:12:59 ok, something to sort out tomorrow 21:13:15 (it also doesn't call generate-slotd-typecheck if it thinks it's not necessary) 21:17:35 davazp [~user@178.167.143.120.threembb.ie] has joined #sbcl 21:18:34 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 21:19:04 -!- ASau` is now known as ASau 21:53:09 -!- prxq [~mommer@mnhm-590c35e0.pool.mediaWays.net] has quit [Ping timeout: 248 seconds] 21:54:23 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 21:56:01 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 22:08:58 drmeister [~drmeister@166.137.107.155] has joined #sbcl 22:19:59 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 22:30:04 -!- drmeister [~drmeister@166.137.107.155] has quit [Remote host closed the connection] 22:40:21 milanj [~milanj@cable-94-189-145-213.dynamic.sbb.rs] has joined #sbcl 22:54:08 -!- davazp [~user@178.167.143.120.threembb.ie] has quit [Ping timeout: 256 seconds] 23:12:15 -!- segv- [~mb@95-91-241-76-dynip.superkabel.de] has quit [Remote host closed the connection] 23:32:34 -!- Bike [~Glossina@67-5-234-210.ptld.qwest.net] has quit [Quit: gone] 23:37:36 Bike [~Glossina@67-5-234-210.ptld.qwest.net] has joined #sbcl 23:37:40 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl