2015-10-22T03:08:30Z ccl-logbot joined #sbcl
2015-10-22T03:08:30Z 
2015-10-22T03:08:30Z names: ccl-logbot abbe hydan angavrilov ASau abeaumont reb Quadrescence zacts brucem |3b| hydan` antoszka DeadTrickster Posterdati salva yauz p_l dougk_ DGASAU kanru jlarocco_ foom ym PuercoPop sigjuice mood Bike scymtym_ fikusz specbot minion jdz alchemis7 christoph_debian jackdaniel drmeister chris2 jackc-_ pkhuong psilord pootler jozip dim les` hzp Ober ferada joshe @Xof White_Flame frankS2 fe[nl]ix Blkt flip214 carvite jsnell dustinm` jlarocco irsol loke dsp_
2015-10-22T03:16:54Z momo-reina joined #sbcl
2015-10-22T03:29:51Z nyef joined #sbcl
2015-10-22T03:49:57Z nikki93 joined #sbcl
2015-10-22T04:36:15Z nikki93_ joined #sbcl
2015-10-22T04:37:10Z nikki93 quit (Ping timeout: 240 seconds)
2015-10-22T04:41:52Z nikki93_ quit (Read error: Connection reset by peer)
2015-10-22T04:42:07Z nikki93 joined #sbcl
2015-10-22T05:01:20Z nikki93 quit (Ping timeout: 246 seconds)
2015-10-22T05:04:58Z nikki93 joined #sbcl
2015-10-22T05:58:45Z Shinmera joined #sbcl
2015-10-22T06:15:00Z Cymew joined #sbcl
2015-10-22T06:45:10Z phoe_krk joined #sbcl
2015-10-22T06:45:47Z phoe_krk left #sbcl
2015-10-22T07:06:29Z fridim_ joined #sbcl
2015-10-22T07:13:15Z zacts quit (Ping timeout: 256 seconds)
2015-10-22T07:24:50Z Quadrescence quit (Ping timeout: 246 seconds)
2015-10-22T07:27:21Z attila_lendvai joined #sbcl
2015-10-22T07:27:21Z attila_lendvai quit (Changing host)
2015-10-22T07:27:21Z attila_lendvai joined #sbcl
2015-10-22T07:27:55Z Quadrescence joined #sbcl
2015-10-22T07:51:10Z Cymew quit (Ping timeout: 240 seconds)
2015-10-22T07:55:07Z anthracite joined #sbcl
2015-10-22T08:15:15Z stassats` joined #sbcl
2015-10-22T08:18:14Z anthracite quit (Remote host closed the connection)
2015-10-22T08:23:27Z ASau quit (Ping timeout: 260 seconds)
2015-10-22T08:41:05Z nikki93 quit (Remote host closed the connection)
2015-10-22T08:42:51Z nikki93 joined #sbcl
2015-10-22T08:44:26Z stassats` quit (Ping timeout: 268 seconds)
2015-10-22T08:47:41Z nikki93 quit (Remote host closed the connection)
2015-10-22T08:55:40Z nikki93 joined #sbcl
2015-10-22T09:13:55Z nikki93 quit (Remote host closed the connection)
2015-10-22T10:02:28Z Cymew joined #sbcl
2015-10-22T10:07:09Z whiteline joined #sbcl
2015-10-22T10:46:52Z Cymew quit (Read error: Connection reset by peer)
2015-10-22T11:20:42Z Cymew joined #sbcl
2015-10-22T11:21:33Z pchrist joined #sbcl
2015-10-22T11:36:45Z stassats` joined #sbcl
2015-10-22T13:33:09Z mood quit (Quit: Gone.)
2015-10-22T13:41:12Z mood joined #sbcl
2015-10-22T13:47:50Z eudoxia joined #sbcl
2015-10-22T14:00:31Z nzambe joined #sbcl
2015-10-22T14:15:46Z attila_lendvai quit (Ping timeout: 240 seconds)
2015-10-22T14:50:06Z nikki93 joined #sbcl
2015-10-22T14:57:22Z scymtym__ joined #sbcl
2015-10-22T14:58:03Z sjl joined #sbcl
2015-10-22T15:00:26Z momo-reina quit (Ping timeout: 240 seconds)
2015-10-22T15:00:57Z White__Flame joined #sbcl
2015-10-22T15:04:08Z PuercoPope joined #sbcl
2015-10-22T15:06:46Z Quadrescence quit (*.net *.split)
2015-10-22T15:06:47Z brucem quit (*.net *.split)
2015-10-22T15:06:47Z dougk_ quit (*.net *.split)
2015-10-22T15:06:47Z foom quit (*.net *.split)
2015-10-22T15:06:47Z PuercoPop quit (*.net *.split)
2015-10-22T15:06:47Z scymtym_ quit (*.net *.split)
2015-10-22T15:06:48Z frankS2 quit (*.net *.split)
2015-10-22T15:06:48Z White_Flame quit (*.net *.split)
2015-10-22T15:06:48Z Xof quit (*.net *.split)
2015-10-22T15:06:52Z PuercoPope is now known as PuercoPop
2015-10-22T15:08:01Z eudoxia quit (Quit: Leaving)
2015-10-22T15:09:15Z nikki93 quit (Remote host closed the connection)
2015-10-22T15:14:58Z frankS2 joined #sbcl
2015-10-22T15:15:26Z foom joined #sbcl
2015-10-22T15:15:26Z Cymew quit (Ping timeout: 250 seconds)
2015-10-22T15:24:53Z brucem joined #sbcl
2015-10-22T15:26:08Z Quadrescence joined #sbcl
2015-10-22T15:27:00Z gingerale joined #sbcl
2015-10-22T15:30:21Z dougk_ joined #sbcl
2015-10-22T15:30:36Z brucem quit (Changing host)
2015-10-22T15:30:36Z brucem joined #sbcl
2015-10-22T16:47:55Z psy_ joined #sbcl
2015-10-22T16:50:02Z whiteline_ joined #sbcl
2015-10-22T16:50:52Z Guest23951 joined #sbcl
2015-10-22T16:51:06Z sjl quit (Ping timeout: 240 seconds)
2015-10-22T16:58:04Z nikki93 joined #sbcl
2015-10-22T17:03:33Z nikki93 quit (Remote host closed the connection)
2015-10-22T17:55:24Z ASau joined #sbcl
2015-10-22T18:25:37Z scymtym joined #sbcl
2015-10-22T18:54:50Z fikusz quit (Ping timeout: 240 seconds)
2015-10-22T18:58:21Z dra joined #sbcl
2015-10-22T19:00:02Z dra: Hi all. Are the abbreviations used throughout the SBCL code, e.g. DFO, VOP and so on, documented anywhere?
2015-10-22T19:01:42Z Guest23951 is now known as attila_lendvai
2015-10-22T19:04:09Z Bicyclidine joined #sbcl
2015-10-22T19:07:27Z nyef: dra: There's a partial explanation in the file "TLA" in the toplevel directory.
2015-10-22T19:08:29Z nyef: dra: Some of them might also have been documented in the (no longer available) sbcl-internals cliki.
2015-10-22T19:09:03Z nyef: For the two that you specifically asked after, DFO would be Depth First Order, and VOP would be Virtual OPeration.
2015-10-22T19:09:42Z fikusz joined #sbcl
2015-10-22T19:10:19Z clique joined #sbcl
2015-10-22T19:13:42Z dra: nyef: Thanks so far! archive.org seems to have a copy of the CLiki pages.
2015-10-22T19:20:51Z nikki93 joined #sbcl
2015-10-22T19:25:06Z nikki93 quit (Ping timeout: 240 seconds)
2015-10-22T19:28:02Z nyef: dra: Are you just poking around, or do you have an actual goal in mind?
2015-10-22T19:32:26Z attila_lendvai: is there a wiki somewhere where a lone hacker (looking at dra) could document his findings? :)
2015-10-22T19:32:26Z dra: nyef: Just poking around. For now. I'd like to implement a "simple" compiler for the lisp I wrote in Ada.
2015-10-22T19:32:54Z nyef: attila_lendvai: There was, once.
2015-10-22T19:33:17Z dra: nyef, attila_lendvai: Hmm. Looks like I have a mission now. ;)
2015-10-22T19:33:27Z nyef: We could try to get the old data files, convert them to a modern format, and upload them to a new wiki.
2015-10-22T19:33:49Z nyef: At one point, IIRC, they were on clnet, so the archives should still be there if we can get a clnet admin to release them to us.
2015-10-22T19:40:31Z gingerale quit (Remote host closed the connection)
2015-10-22T19:43:38Z clique left #sbcl
2015-10-22T19:51:42Z clique joined #sbcl
2015-10-22T19:52:18Z dra: I found documentation for Nightshade, which is also a CMUL derivative. Seems to contain quite a bit of information about the internals: http://www.mundell.ukfsn.org/nightshade/doc/Documentation.html
2015-10-22T20:01:22Z psy_ quit (Read error: No route to host)
2015-10-22T20:02:51Z psy_ joined #sbcl
2015-10-22T20:03:25Z psy_ quit (Max SendQ exceeded)
2015-10-22T20:03:53Z psy_ joined #sbcl
2015-10-22T20:04:31Z clique quit (Quit: Page closed)
2015-10-22T20:10:15Z stassats`: minion knows all the acronyms
2015-10-22T20:10:19Z stassats`: minion: what does DFO stand for?
2015-10-22T20:10:19Z minion: Dabble Fordy Ornithon
2015-10-22T20:10:30Z dra: Sweet.
2015-10-22T20:10:56Z dra: minion: What does USA stand for?
2015-10-22T20:10:56Z minion: Uncircularized Sloppage Aralkyl
2015-10-22T20:11:18Z stassats`: sounds about right
2015-10-22T20:11:26Z dra: Yep, definitely right.
2015-10-22T20:11:32Z nikki93 joined #sbcl
2015-10-22T20:12:05Z nyef: On the one hand, wow, another CMUCL fork. On the other hand... why?
2015-10-22T20:12:24Z stassats`: which year is that?
2015-10-22T20:12:36Z nyef: Forked in 2006, last release in 2009.
2015-10-22T20:12:53Z stassats`: there was even sbcl 1.0 in 2006
2015-10-22T20:13:17Z nyef: Mmm. Buggiest SBCL release, ever.
2015-10-22T20:13:36Z stassats`: i think it was the first sbcl i installed
2015-10-22T20:13:42Z stassats`: good times
2015-10-22T20:14:06Z nyef: I had a large chunk of code merged in 1.0 or shortly before, but that was before I had commit access.
2015-10-22T20:14:07Z Shinmera: Looks like Nightshade fulfils Zawinski's law.
2015-10-22T20:14:18Z stassats`: http://www.mundell.ukfsn.org/nightshade/public.html
2015-10-22T20:14:20Z dra: Haha. Well, I found the CMUCL documentation now so I guess that's as good as it gets.
2015-10-22T20:14:31Z nyef: dra: The EncyCMUCLopedia?
2015-10-22T20:14:32Z stassats`: seems like it's doing what RMS is doing about GPL but in reverse
2015-10-22T20:14:49Z stassats`: "everything must be public domain"
2015-10-22T20:14:50Z stassats`: meh
2015-10-22T20:15:07Z dra: nyef: This: https://common-lisp.net/project/cmucl/doc/CMUCL-design.pdf
2015-10-22T20:15:33Z stassats`: nyef: so, the answer to "why" is "it must be completely public domain"
2015-10-22T20:17:50Z stassats`: dra: i didn't find that to be particularly illuminating
2015-10-22T20:17:56Z nyef: Weird, weird motivation. It's almost surprising that it lasted three years, then.
2015-10-22T20:18:29Z dra: stassats`: But it's long so it must be good. Right? Right?
2015-10-22T20:18:33Z Shinmera: People go to great lengths about licensing.
2015-10-22T20:19:09Z stassats`: descriptions like "Revert Pierre Mai changes." and "Revert isqrt implementation to 1993-02-27 version." to be pretty amusing
2015-10-22T20:19:27Z stassats`: also considering that public domain has different legal standings in different countries
2015-10-22T20:19:43Z stassats`: so, this endeavour appears to be pretty pointless, oh well
2015-10-22T20:20:47Z p_l: stassats`: well, for his purposes, the most commonly cited differences aren't real ones, afaik
2015-10-22T20:20:50Z dra: I can't "release" anything into the public domain. Well, I could, but that would require killing myself and waiting 75 years.
2015-10-22T20:21:19Z Shinmera: dra: In some countries it's less than 75 years for software.
2015-10-22T20:21:28Z stassats`: gotta go do a real productive thing, rewatching Back to the future
2015-10-22T20:21:34Z p_l: dra: dunno about you, I can't release anything into american-style public domain till the laws are abolished ;)
2015-10-22T20:21:52Z p_l: however, in practical terms, I can give (nearly) the same rights immediately to anyone
2015-10-22T20:22:01Z dra: Beerware for the win.
2015-10-22T20:23:02Z p_l: pretty much the same thing the author of nightshade wanted is what polish law calls "waiving material rights to the creation"
2015-10-22T20:25:55Z stassats`: but why wouldn't the BSD license accomplish the same thing?
2015-10-22T20:26:19Z stassats`: i wonder if it includes the mit loop and pcl
2015-10-22T20:26:39Z stassats`: with the "mail it to us" license
2015-10-22T20:28:42Z stassats`: or was it clx?
2015-10-22T20:30:08Z dra: Some time ago someone over in #ada said: If you're worried about licenses you might as well watch frogs mate.
2015-10-22T20:30:39Z stassats`: well, there's one thing to worry, gpl vs non-gpl, everything else is murky
2015-10-22T20:30:51Z dra: Aye.
2015-10-22T20:30:59Z p_l: stassats`: well, MIT license is nearly the same thing as public domain in most EU countries
2015-10-22T20:31:08Z p_l: BSD used to include advertising clause
2015-10-22T20:35:26Z fridim_ quit (Ping timeout: 240 seconds)
2015-10-22T21:03:26Z dra quit (Ping timeout: 240 seconds)
2015-10-22T21:33:03Z psy_ quit (Remote host closed the connection)
2015-10-22T21:34:46Z nyef: Am I supposed to see a message about leftover cold-init symbols "Found -1:" "NIL" at the end of target-2?
2015-10-22T21:36:37Z dra joined #sbcl
2015-10-22T21:45:19Z nikki93 quit (Remote host closed the connection)
2015-10-22T22:15:25Z stassats`: nyef: i'm ignoring those things
2015-10-22T22:16:24Z nyef: I also caught an "Unhandled FLOATING-POINT-OVERFLOW" on a self-hosted build.
2015-10-22T22:16:43Z stassats`: that would be quite weird, did i break something?
2015-10-22T22:16:53Z stassats`: is that mips or hppa?
2015-10-22T22:16:56Z nyef: MIPS.
2015-10-22T22:17:01Z nyef: I don't know if you broke anything or not.
2015-10-22T22:17:17Z stassats`: i removed a sigfpe handler, but my qemu self-build succeeded
2015-10-22T22:17:21Z nyef: At least it's a plausible error, as it's on (DEFCONSTANT MOST-POSITIVE-DOUBLE-FLOAT).
2015-10-22T22:17:50Z stassats`: an install-irrevocably handler, the lisp handler is still supposed to wokr
2015-10-22T22:18:49Z nyef: Haven't started to try and diagnose yet, really.
2015-10-22T22:20:15Z stassats`: but, if it's "Unhandled FLOATING-POINT-OVERFLOW", then it means that it reached lisp
2015-10-22T22:20:29Z stassats`: besides, it shouldn't overflow at all
2015-10-22T22:20:58Z stassats`: unless denormals are disabled?
2015-10-22T22:21:39Z stassats`: but only the LEAST- are denormals
2015-10-22T22:25:24Z chris2 quit (Ping timeout: 252 seconds)
2015-10-22T22:27:43Z stassats`: and besides, denormals are underflowing
2015-10-22T22:28:06Z chris2 joined #sbcl
2015-10-22T22:33:41Z Shinmera quit (Quit: しつれいしなければならないんです。)
2015-10-22T22:38:50Z dra quit (Remote host closed the connection)
2015-10-22T22:39:11Z sjl joined #sbcl
2015-10-22T22:45:58Z sjl__ joined #sbcl
2015-10-22T22:48:11Z sjl quit (Ping timeout: 264 seconds)
2015-10-22T23:05:17Z Bicyclidine quit (Ping timeout: 250 seconds)
2015-10-22T23:15:16Z nyef: Okay, most-positive-double-float => (double-from-bits 0 double-float-normal-exponent-max (ldb (byte double-float-digits 0) -1)).
2015-10-22T23:15:41Z nyef: And double-from-bits refers to double-float-exponent-byte and double-float-significand-byte.
2015-10-22T23:17:18Z nyef: The normal-exponent-max is #x7fe, the exponent-byte is (byte 11 20), and the significand byte is (byte 20 0).
2015-10-22T23:19:30Z nyef: Heh. And there's a note from 2001-04-05 in src/code/cross-float.lisp about a workaround for a bug in CMUCL 18c that can go away once the CMUCL folks have made a bugfix release. (-:
2015-10-22T23:21:29Z nyef: The cross-float version of MAKE-SINGLE-FLOAT and MAKE-DOUBLE-FLOAT use several magic numbers and byte descriptions, rather than using the SB!VM constants, while the functions that produce the inputs for these cross-float functions use the sb!vm constants.
2015-10-22T23:21:48Z nyef: It looks like they agree, but... Eesh.
2015-10-22T23:31:37Z rszeno joined #sbcl
2015-10-22T23:46:19Z nyef: Hrm.
2015-10-22T23:46:37Z nyef: What's the correct value for (expt 0.5d0 52) ?
2015-10-22T23:50:04Z nyef: Yeah, definitely found something wrong.
2015-10-22T23:50:41Z nyef: On MIPS, the double-float-high-bits for that result is coming up as 3FF00000, while on x86-64 it's coming up as 3CB00000.