00:09:42 -!- lambda-avenger [~roman@adsl-99-50-120-120.dsl.pltn13.sbcglobal.net] has left #sbcl 00:11:11 -!- hargettp [~anonymous@pool-71-184-191-100.bstnma.east.verizon.net] has quit [Quit: hargettp] 01:12:42 rpg [~rpg@216.243.156.16.real-time.com] has joined #sbcl 01:25:34 -!- homie [~levgue@xdsl-78-35-180-2.netcologne.de] has quit [Ping timeout: 255 seconds] 01:31:54 -!- tcr [~tcr@cpc1-bour5-2-0-cust921.15-1.cable.virginmedia.com] has quit [Quit: Leaving.] 01:41:16 homie [~levgue@xdsl-78-35-144-175.netcologne.de] has joined #sbcl 03:26:38 homie` [~levgue@xdsl-78-35-144-175.netcologne.de] has joined #sbcl 03:27:51 -!- homie` [~levgue@xdsl-78-35-144-175.netcologne.de] has quit [Remote host closed the connection] 03:29:44 homie` [~levgue@xdsl-78-35-144-175.netcologne.de] has joined #sbcl 03:32:09 -!- homie` [~levgue@xdsl-78-35-144-175.netcologne.de] has quit [Client Quit] 03:34:20 rpg_ [~rpg@216.243.156.16.real-time.com] has joined #sbcl 03:34:53 -!- homie [~levgue@xdsl-78-35-144-175.netcologne.de] has quit [*.net *.split] 03:34:54 -!- luis [~luis@nhop.r42.eu] has quit [*.net *.split] 03:34:55 -!- foom [~jknight@ita4fw1.itasoftware.com] has quit [*.net *.split] 03:35:26 homie [~levgue@xdsl-78-35-144-175.netcologne.de] has joined #sbcl 03:36:38 -!- rpg [~rpg@216.243.156.16.real-time.com] has quit [Ping timeout: 240 seconds] 03:37:24 luis [~luis@nhop.r42.eu] has joined #sbcl 03:37:24 foom [~jknight@ita4fw1.itasoftware.com] has joined #sbcl 04:34:26 slyrus_ [~chatzilla@adsl-75-36-217-249.dsl.pltn13.sbcglobal.net] has joined #sbcl 05:03:19 -!- rpg_ [~rpg@216.243.156.16.real-time.com] has quit [Quit: rpg_] 05:14:36 -!- redline6561 [~user@c-66-56-55-169.hsd1.ga.comcast.net] has quit [Remote host closed the connection] 05:20:02 redline6561 [~user@c-66-56-55-169.hsd1.ga.comcast.net] has joined #sbcl 07:02:47 -!- Krystof [~csr21@csrhodes.plus.com] has quit [Ping timeout: 240 seconds] 07:03:24 Krystof [~csr21@csrhodes.plus.com] has joined #sbcl 07:03:33 -!- ChanServ has set mode +o Krystof 07:20:15 flip214 [~marek@unaffiliated/flip214] has joined #sbcl 08:32:32 -!- fe[nl]ix [~lacedaemo@pdpc/supporter/professional/fenlix] has quit [Read error: Operation timed out] 08:33:00 -!- flip214 [~marek@unaffiliated/flip214] has quit [Ping timeout: 260 seconds] 08:33:15 -!- deepfire [~deepfire@80.92.100.69] has quit [Ping timeout: 272 seconds] 08:33:19 fe[nl]ix [~lacedaemo@pdpc/supporter/professional/fenlix] has joined #sbcl 08:34:17 deepfire [~deepfire@80.92.100.69] has joined #sbcl 08:46:19 flip214 [~marek@2001:858:107:1:baac:6fff:fe6b:9183] has joined #sbcl 08:46:19 -!- flip214 [~marek@2001:858:107:1:baac:6fff:fe6b:9183] has quit [Changing host] 08:46:19 flip214 [~marek@unaffiliated/flip214] has joined #sbcl 09:07:08 tcr [~tcr@cpc1-bour5-2-0-cust921.15-1.cable.virginmedia.com] has joined #sbcl 09:50:08 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 10:15:25 stassats [~stassats@wikipedia/stassats] has joined #sbcl 10:45:11 -!- cmm [~cmm@109.65.200.120] has quit [Ping timeout: 240 seconds] 10:46:15 cmm [~cmm@109.65.200.120] has joined #sbcl 11:36:31 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 240 seconds] 11:59:49 attila_lendvai [~attila_le@catv-89-133-171-67.catv.broadband.hu] has joined #sbcl 12:32:28 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 246 seconds] 12:47:38 stassats [~stassats@wikipedia/stassats] has joined #sbcl 12:51:22 The SSE intrinsics project seems to have stalled. I wonder what should I do to make some progress... 13:36:06 thorstadt [~user@nc-76-6-29-155.dhcp.embarqhsd.net] has joined #sbcl 13:36:15 -!- thorstadt [~user@nc-76-6-29-155.dhcp.embarqhsd.net] has left #sbcl 13:37:59 -!- deepfire [~deepfire@80.92.100.69] has quit [Ping timeout: 260 seconds] 13:38:35 deepfire [~deepfire@80.92.100.69] has joined #sbcl 14:22:45 -!- homie [~levgue@xdsl-78-35-144-175.netcologne.de] has quit [Read error: Connection reset by peer] 14:23:38 -!- attila_lendvai [~attila_le@catv-89-133-171-67.catv.broadband.hu] has left #sbcl 14:23:39 homie [~levgue@xdsl-78-35-144-41.netcologne.de] has joined #sbcl 14:42:37 hargettp [~hargettp@dhcp-162.mirrorimage.net] has joined #sbcl 14:51:06 -!- slyrus_ [~chatzilla@adsl-75-36-217-249.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 250 seconds] 14:51:38 -!- flip214 [~marek@unaffiliated/flip214] has quit [Remote host closed the connection] 14:52:15 pkhuong: Maybe I should send an email about SSE to sbcl-devel to get things moving somewhat?.. 14:53:18 angavrilov: I'm not sure if anyone has the brain cycle for that 15:01:49 pkhuong: Email is asynchronous, and thus easier than finding people (especially from different time zones) on IRC... 15:06:30 angavrilov: I meant generally, for the next couple weeks. 15:23:23 pkhuong: I think I'll send it anyway sometime this weekend, or maybe a week later. I haven't done any non-dayjob programming for the last 2 months and need some event to motivate myself to start again :) 15:55:16 attila_lendvai [~attila_le@catv-89-133-171-67.catv.broadband.hu] has joined #sbcl 16:33:06 -!- cmm [~cmm@109.65.200.120] has quit [Remote host closed the connection] 16:33:28 cmm [~cmm@bzq-109-65-200-120.red.bezeqint.net] has joined #sbcl 16:46:01 hulloo 16:56:00 derrotebaron [~derroteba@static.7.69.47.78.clients.your-server.de] has joined #sbcl 17:09:45 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 18:02:34 -!- attila_lendvai [~attila_le@catv-89-133-171-67.catv.broadband.hu] has left #sbcl 19:25:07 -!- cmm [~cmm@bzq-109-65-200-120.red.bezeqint.net] has quit [Ping timeout: 255 seconds] 19:26:10 cmm [~cmm@bzq-109-65-200-120.red.bezeqint.net] has joined #sbcl 19:41:35 i get 3 errors with maxima with sbcl backend, but none with clisp backend, and all those errors are due to some double precision to single precision conversion or so..... 19:42:29 this is on x86? 19:47:28 antgreen [~user@nat/redhat/x-enwzwphybcmmkful] has joined #sbcl 19:48:38 -!- cmm [~cmm@bzq-109-65-200-120.red.bezeqint.net] has quit [Ping timeout: 240 seconds] 19:49:53 cmm [~cmm@109.65.200.120] has joined #sbcl 19:50:01 yep 19:50:07 x86-32 19:50:23 core2duo, 1.8Ghz 19:50:50 that's probably the issue, then; some FP computations are getting rearranged, or rounding is not taking place at the right time 19:52:01 ok, now cmucl showed me no errors too 19:52:23 can't be that i get errors only in sbcl 19:52:53 it must be something in sbcl then, i yesterday wrote something in here, are there any logs of this channel ? 19:54:23 and maxima dev's won't fix it, they say it's someting about double to single precision truncation thing, which is due to how exp(y*log(x)) is computed in sbcl or so 19:54:35 they say that log(x) is single precision in sbcl 19:55:34 sbcl computes x^y using exp(y*log(x)). Since x here is an integer 19:55:34 log(x) is a single-float, and hence the result only has single-float 19:55:34 accuracy, even though the result is returned as a double-float. 19:55:34 19:55:37 I have heard from from sbcl developers that this will be fixed. 19:55:54 that's a quote now 19:55:59 I see that 19:56:09 I bet nobody remembered about actually doing that. :) 19:57:00 i woud fix it myself, if i knew where in the source it is located actually....gah 19:57:56 maxima could workaround by coercing the integer to a double-float itself, but I agree it shouldn't have to. 19:58:32 finding the source for log is not particularly hard 19:58:59 it's the source for expt you want, I think 20:00:00 float.lisp ? 20:01:06 irrat.lisp ? 20:02:00 slime M-. ? 20:02:12 or (describe 'expt) 20:04:26 i see, it's all in irrat.lisp 20:04:55 but changing one thing there would imply changing other things too... 20:05:10 such as? 20:05:53 defun log and defun expt and real-expt 20:06:26 -!- hargettp [~hargettp@dhcp-162.mirrorimage.net] has quit [Remote host closed the connection] 20:06:57 right, i'm gonna do that now, despite all just to test.... 20:11:52 -!- cmm [~cmm@109.65.200.120] has quit [Ping timeout: 255 seconds] 20:12:32 Is there a test suite for correct math function type coercion? 20:12:41 cmm [~cmm@109.65.200.120] has joined #sbcl 20:13:11 There may be other bugs in there too, afterall. 20:13:43 well, if cmucl works, you may be able to crib from there 20:21:01 well, compile-cold-sbcl gives now errors, failure-p was set for irrat.lisp 20:22:20 it may be easier to load your changes into a running sbcl and fix problems there first 20:28:27 i'm taking the gross approach of s/single-float/double-float/ in irrat.lisp, and continuing on failures, just to see... 20:29:02 that's a bit excessive, I think 20:30:07 yes....never mind, i don't know the sbcl infrastucture so well, as to conclude where to make changes in parallel to make it work... 20:34:05 failed lol 20:35:15 -- 20:35:17 20:35:53 hargettp [~hargettp@dhcp-162.mirrorimage.net] has joined #sbcl 20:36:45 20:36:48 9 20:36:48 20:36:49 20:36:49 20:37:21 froydnj: cat trouble? 20:37:58 bah 20:39:58 -!- stassats [~stassats@wikipedia/stassats] has quit [*.net *.split] 20:40:00 -!- ruediger [ruediger@pseudoterminal.org] has quit [*.net *.split] 20:42:51 stassats [~stassats@wikipedia/stassats] has joined #sbcl 20:42:51 ruediger [ruediger@pseudoterminal.org] has joined #sbcl 20:48:15 lol 21:23:16 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 276 seconds] 21:29:02 why do they talk of no precision loss there in irrat.lisp ? when there are several double to single float coercions actually ? 21:32:15 oh man, i need something like etags or cscope for lisp 21:33:59 Yes, if only slime had something like M-. 21:34:20 what does that do ? 21:34:49 go to the definition of something 21:35:01 M-? will visit all uses of something 21:35:59 oh ok, but no i meant something like etags or cscope outside of emacs i mean....for just lisp in general or common-lisp in special.... 22:01:20 -!- hargettp [~hargettp@dhcp-162.mirrorimage.net] has quit [Quit: hargettp] 22:07:28 ok i just get one failure with log-base-zero-return-type in tests/float-pure.lisp due to me getting rid of most double to single float coercions...where can i put irrat.lisp so that someone can have a look at it how to prevent the last error ? 22:09:44 send a diff of your copy with the version in cvs to sbcl-devel 22:09:55 include in your message as much information as you can 22:10:17 hmmm, ok 22:13:27 -!- antgreen [~user@nat/redhat/x-enwzwphybcmmkful] has quit [Remote host closed the connection] 22:20:20 haha, whooohooo, maxima gives no errors now 22:21:03 with the new sbcl compiled even, it's just one test fails cause it expects single-float return type, which i don't have anymore, it's all double float 22:22:47 maybe i should have preserved just one case of single-float return-type 22:25:28 Krystof: i don't need a subscription to the mailing list do i ? 22:26:04 i mean to post 22:26:34 homie: you do 22:26:44 oh 22:27:47 yeah, how tragic 22:41:06 homie: if you still have failing test cases, you should perhaps spend some time understanding why 22:41:22 for example, sbcl is a Common Lisp implementation, which means that it tries to follow the standard 22:41:43 sometimes, the standard requires a single-float return value 22:49:28 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds] 22:50:26 hmm, ok i send the diff out now, i hope somone can read it, i don't know the real fix, i just tried to get as much close to a real solution, but yes at least one single-float return-type has to remain... 22:51:18 and the fix as such just works, at least for the maxima case...those errors disappeared now in maxima 23:30:30 hargettp [~anonymous@pool-71-174-137-226.bstnma.east.verizon.net] has joined #sbcl 23:54:19 -!- tcr [~tcr@cpc1-bour5-2-0-cust921.15-1.cable.virginmedia.com] has quit [Quit: Leaving.]