00:11:06 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 00:30:27 -!- tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has quit [Quit: Leaving.] 01:01:18 -!- wbooze [~wbooze@xdsl-84-44-179-55.netcologne.de] has quit [Ping timeout: 276 seconds] 01:04:21 wbooze [~wbooze@xdsl-84-44-179-55.netcologne.de] has joined #sbcl 01:42:02 yacks [~yacks@180.151.36.169] has joined #sbcl 02:44:40 Thra11_ [~thrall@87.115.108.182] has joined #sbcl 02:46:58 -!- Thra11 [~thrall@31.185.188.80] has quit [Ping timeout: 252 seconds] 02:50:08 -!- wbooze [~wbooze@xdsl-84-44-179-55.netcologne.de] has quit [Quit: none] 02:54:27 wbooze [~wbooze@xdsl-84-44-179-55.netcologne.de] has joined #sbcl 02:58:52 -!- Thra11_ [~thrall@87.115.108.182] has quit [Ping timeout: 244 seconds] 04:43:11 -!- danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Quit: Dan Lentz ... Out.] 04:45:50 danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 04:54:05 -!- danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Quit: Dan Lentz ... Out.] 05:39:39 dioxirane [~OXO@unaffiliated/dioxirane] has joined #sbcl 05:53:09 akovalen` [~user@77.51.6.108] has joined #sbcl 05:55:03 -!- akovalenko [~user@77.51.0.89] has quit [Ping timeout: 252 seconds] 06:19:20 -!- wbooze [~wbooze@xdsl-84-44-179-55.netcologne.de] has quit [Read error: Connection reset by peer] 06:52:55 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 07:20:46 sdemarre [~serge@91.176.241.92] has joined #sbcl 07:30:17 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 07:32:50 attila_lendvai [~attila_le@95.56.78.11] has joined #sbcl 07:32:50 -!- attila_lendvai [~attila_le@95.56.78.11] has quit [Changing host] 07:32:50 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:35:24 danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 07:43:22 -!- danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Quit: Dan Lentz ... Out.] 08:23:16 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 08:47:28 danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 08:47:33 -!- danlentz0 [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Client Quit] 09:11:19 prxq [~mommer@mnhm-590c0ece.pool.mediaWays.net] has joined #sbcl 09:56:50 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 10:17:59 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 10:18:34 anyone knows anything about an issue with asdf and sb-bsd-sockets? (illegal function call...) 10:19:17 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 10:19:28 even (require :sb-posix) .? 10:19:55 hi 10:20:04 hello Fare 10:21:01 attila_lendvai: try first (require :sb-posix), after (require :sb-bsd-sockets) 10:21:19 *attila_lendvai* tries 10:34:44 fare managed to reproduce it. waiting for the open source fairies... :) 10:34:50 -!- sdemarre [~serge@91.176.241.92] has quit [Ping timeout: 252 seconds] 10:35:13 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 10:36:44 attila_lendvai: issue solved or not? 10:37:14 dioxirane: knowing fare it will be... 10:38:09 Was the SB-EXT:CONSTANT-FUNCTION declaration removed? I don't see it in the (PDF) manual. 10:39:01 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 10:41:05 And I don't seem to see it when grepping the source. :( 10:42:24 attila_lendvai: no issue anyway here (at least at the moment) 10:43:59 OK, the issue was that in the dumped image without an exported SBCL_HOME, the (asdf::lisp-implementation-directory) was NIL and so we no longer had (builtin-system-p (find-system :sb-posix)) 10:44:51 Fare: you fixed it in asdf? 10:45:55 I didn't touch asdf. 10:46:07 I mean, since you just contacted me 10:48:42 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 10:58:28 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 11:12:42 -!- dioxirane [~OXO@unaffiliated/dioxirane] has quit [Remote host closed the connection] 11:18:25 dioxirane [~OXO@unaffiliated/dioxirane] has joined #sbcl 11:21:29 -!- ASau [~user@46.115.100.212] has quit [Remote host closed the connection] 11:22:20 ASau [~user@46.115.100.212] has joined #sbcl 11:24:25 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving...] 11:42:57 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 12:02:51 luis [~luis@nhop.r42.eu] has joined #sbcl 12:10:58 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 12:15:42 sdemarre [~serge@91.176.241.92] has joined #sbcl 12:20:26 -!- yacks [~yacks@180.151.36.169] has quit [Quit: Leaving] 12:22:07 -!- tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has quit [Quit: Leaving.] 12:58:45 if contribs are each a single .fasl file without source directory, 12:58:54 should they go in $SBCL_HOME 12:58:59 or $SBCL_HOME/contrib 13:00:22 Fare: and if I do not export SBCL_HOME at all? 13:03:14 either SBCL has one builtin, or you lose. 13:03:29 better, if I build SBCL from sources using "sh make.sh --prefix=/my-path/ ..." maybe I have not need to create a new SBCL_HOME variable 13:03:47 I'm not touching that part of the code 13:03:58 I'm just changing the contrib to not need asdf and source code at runtime 13:04:03 only at compile-time 13:05:03 Fare: another question..what is the safest way to use a modified asdf in SBCL, maybe not included in the contrib directory? -thanks- 13:07:27 have a git checkout in your registry 13:07:40 works not just for sbcl but for any implementation 13:07:59 on asdf3, it will be automatically loaded, unless it's a downgrade, in which case you get a warning. 13:08:18 automatically = after you (require :asdf) and first try to use it 13:08:44 just curious - does asdf still not support freezing libraries loaded in an image without hacks? 13:08:48 on asdf2, after you (require "asdf"), you still need to (asdf:load-system :asdf) to load it. 13:09:02 and you need to do it FIRST THING before anything else. 13:09:06 i.e. making it assume that a system is loaded and up to date without accessing file system in any way 13:09:19 "freezing libraries" ?? what do you mean? 13:09:29 that's called (asdf:require-system ....) 13:09:30 https://github.com/angavrilov/cl-linux-debug/blob/master/mk-runtime/load-libs.lisp#L31 13:09:35 as opposed to (asdf:load-system ...) 13:10:05 that's what I wrote a year ago to do what I wanted 13:10:22 it might do read-only accesses to the file-system, I don't remember how much. 13:10:43 the idea is to build an image that can then be moved to a completely different computer without any of the source code 13:10:49 angavrilov: do-traverse doesn't exist anymore 13:11:30 and then tell it to load the final application from source, and have asdf properly assume that all of it's third-party dependences preloaded into the image are there 13:12:21 https://github.com/angavrilov/df-structures/blob/master/start.sh-core 13:12:27 angavrilov, clever enough. But require-system is the supported way, these days. 13:12:53 in ASDF3, I have register-preloaded-system 13:13:32 it will let you provide your systems so that *if* something is available in the filesystem, it takes precedence, but your thing is otherwise available from the image. 13:14:58 sounds like that's probably what I want - didn't really watch recent developments 13:14:59 as opposed to the invasive way your code prevents any future upgrade of these systems. 13:19:28 leoc [~leoc.git@p57B9B606.dip.t-dialin.net] has joined #sbcl 13:25:16 OK, since there's another directory site-systems/ I propose contribs go to contrib/ 13:25:31 it makes for a cleaner $SBCL_HOME/ too 13:34:08 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 252 seconds] 13:35:29 wbooze [~wbooze@xdsl-78-35-166-231.netcologne.de] has joined #sbcl 13:42:14 yacks [~yacks@180.151.36.169] has joined #sbcl 13:45:54 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 13:59:31 of course, if I move the contrib .fasls to $SBCL_HOME/contrib/ then run-sbcl.sh is confused 14:00:14 should I create the fasls under contrib/contrib? 14:00:19 create them under contrib/ 14:00:31 and have the SBCL_HOME be the sbcl dir? 14:18:37 installing / looking for contribs under $SBCL_HOME/contrib makes a lot of sense to me, because it makes the installation structure more similar to the sources. 14:19:22 hence it's a good step in the right direction (the direction being towards the fiery death of the SBCL_HOME variable ultimately) 14:21:34 next step: have the makefile put sbcl/sbcl.exe into a bin directory, not into src/runtime 14:22:07 (and years later, we'll almost be as usable again as CMUCL has been for a long time now) 14:22:44 Fare: sorry for now applying those cxml patches yet 14:23:41 Fare: btw, is there an easy way these days to tell asdf to not treat warnings as failures (no matter which implementation), or do I still have to define my own methods? 14:23:49 s/now/not/ 14:24:54 asdf:*compile-file-failure-behaviour* 14:26:13 same as always: (setf asdf:*compile-file-failure-behaviour* :warn) 14:26:23 or :ignore 14:26:42 you never had to define your own methods 14:26:52 or maybe in the early days of asdf 14:27:33 since 2.21 or so, you can use :around-compile to handle warnings 14:27:55 and since 2.27, you can use *uninteresting-compiler-conditions* 14:52:55 gko [~user@114-34-168-13.HINET-IP.hinet.net] has joined #sbcl 14:53:06 Fare: good, just boostrapped asdf version 2.30.6 and loaded the asdf.lisp file from the build dir 14:55:42 Fare: is " make bump v=3.0 " used for the upgrading to the asdf3? 14:56:05 for the releasing, you mean 14:56:20 when everyone is happy with asdf3, I'll release it. 14:57:45 -!- lichtblau [~user@port-212-202-134-139.static.qsc.de] has left #sbcl 15:12:24 -!- sdemarre [~serge@91.176.241.92] has quit [Ping timeout: 252 seconds] 15:21:55 Fare: with "gmake bump v=3.0" command .bin/asdf.builder do not find sbcl even if I set ASDF_TEST_LISPS="/my-path-to-sbcl/bin/sbcl --core /my-path-to-sbcl/lib/sbcl/sbcl.core" 15:22:02 sdemarre [~serge@91.176.241.92] has joined #sbcl 15:24:07 dioxirane, what are you trying to do??? 15:24:24 are you trying to release a new version of asdf???? 15:24:41 ..no.. just I want try asdf3 15:24:42 unless you're going to take over from me, don't!!!!! 15:24:46 then don't do that 15:24:47 make 15:24:52 and that's all 15:36:06 LiamH [~none@96.231.218.61] has joined #sbcl 15:37:15 ok.. Fare, now I'm reading the manual on common-lisp.net to clarify my ideas. Thanks 15:49:15 how do I quickly rebuild just one code file in sbcl? 15:49:38 what's a code file? 15:56:51 FareWell [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 15:57:29 grrrrr, laptop died again compiling sbcl 15:57:44 maybe this cold vodka bottle can keep it cool 15:57:55 sbcl/src/code/module.lisp 15:58:01 that's what I want to recompile 15:58:12 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 15:59:20 -!- sdemarre [~serge@91.176.241.92] has quit [Ping timeout: 252 seconds] 16:01:39 -!- akovalen` is now known as akovalenko 16:06:08 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 16:09:19 FareWell: maybe just compile name.lisp or compile-file-pathname to recompile all the content of a directory 16:10:18 well, sbcl internals require special package magic 16:12:24 FareWell: slam.sh should be ok there. You need to have built with a core saved after XC. 16:13:43 ok, I can get most contribs running in my new setup, but sb-sprof looks like it needs more love 16:13:47 rebuilding the whole thing usually turns out to be faster 16:14:02 unless you're expecting to rebuild it N times 16:16:17 well, I'm debugging it 16:16:34 hopefully, now only debugging contribs, so just running make-target-contrib 16:16:34 C-c C-k in slime should work 16:16:42 but previously debugging src/code/module.lisp 16:18:49 wait, something made it work, and I have no idea what 16:22:02 found it -- needed to edit make-target-contrib.sh so its SBCL_HOME doesn't append contrib/ anymore. 16:22:31 alternatively... shouldn't the contrib output their files in obj/ too ? 16:25:06 is better to use in .sbclrc, after (require "asdf") and (asdf:load-system :asdf), (pushnew #p"my-path" asdf:*central-registry* :test #'equal) or (push #p"my-path" asdf:*central-registry*) ? 16:25:07 of maybe obj/from-host/ ? 16:25:20 I mean, contribs in obj/from-host/contrib/ or just obj/contrib/ ? 16:26:33 FareWell: what do you think about my question ? 16:27:50 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 16:29:27 where is obj/ configured? 16:29:45 what do you mean, configured? 16:29:47 dioxirane, what questoin? 16:30:09 I mean, where is it written that fasls go to $SBCL_PWD/obj 16:30:32 no single place 16:30:45 why do you want to change it, though? 16:31:19 not changing it 16:31:32 I'd like to put contrib output in there, too 16:31:57 why not leave it as it is? 16:35:44 -!- tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has quit [Quit: Leaving.] 16:36:01 cleaner? 16:44:03 sdemarre [~serge@91.176.241.92] has joined #sbcl 16:46:04 are fasls always .fasl under sbcl? 16:46:59 can be modified with sb-fasl:*fasl-file-type* 16:54:07 may I assume it's always "fasl" during the build of asdf itself? 16:55:15 of asdf? 16:55:35 or you could just use sb-fasl:*fasl-file-type* 16:55:42 whatever is it that you're doing 16:56:26 clever, very clever 17:09:29 milosn_ [~milosn@user-5AF501D3.broadband.tesco.net] has joined #sbcl 17:11:30 -!- milosn [~milosn@user-5AF507FD.broadband.tesco.net] has quit [Ping timeout: 256 seconds] 17:12:50 attila_lendvai [~attila_le@95.56.126.233] has joined #sbcl 17:12:50 -!- attila_lendvai [~attila_le@95.56.126.233] has quit [Changing host] 17:12:50 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:14:30 -!- cmm [~cmm@109.65.218.23] has quit [Ping timeout: 240 seconds] 17:15:31 cmm [~cmm@109.65.218.23] has joined #sbcl 17:16:41 leuler [~user@p548FC277.dip.t-dialin.net] has joined #sbcl 17:18:31 I have just tested asdf 2.30.6 with respect to running SBCL's tests while having a systemwide old asdf installed. I report success. 17:20:55 milosn [~milosn@user-5AF501E4.broadband.tesco.net] has joined #sbcl 17:23:04 Namely: In a fresh checkout of sbcl-1.1.4-36-g506af58 I upgraded contrib/asdf to 2.30.6 (via git clone, make), then deleted the subdirectory asdf-upstream, made SBCL and ran the tests. I saw the warning message and the test ran to completion. 17:23:35 -!- milosn_ [~milosn@user-5AF501D3.broadband.tesco.net] has quit [Ping timeout: 260 seconds] 17:24:34 -!- wbooze [~wbooze@xdsl-78-35-166-231.netcologne.de] has quit [Quit: none] 17:26:11 When asdf-upstream is not deleted, no warning is shown; it seems this directory is pushed into the source registry in front of the systemwide one so asdf finds the same version that is already running first: (asdf::locate-system :asdf) -> #P"/home/lutz/lisp/sbcl/sbcl-git/contrib/asdf/asdf-upstream/asdf.asd" 17:27:08 But SBCL is distributed without the asdf-upstream directory so all is fine. 17:35:49 wbooze [~wbooze@xdsl-78-35-166-231.netcologne.de] has joined #sbcl 17:39:28 where is the logical pathname SYS setup? 17:40:00 rgrep it? 17:46:46 -!- sdemarre [~serge@91.176.241.92] has quit [Ping timeout: 252 seconds] 17:48:34 src/code/target-pathname.lisp 17:53:18 bah, internal error #26 (An attempt was made to use an undefined SYMBOL-VALUE.) 17:53:33 which symbol-value? 17:57:50 sb-show should help 18:00:42 "/interning ~A", that's really helpful... 18:02:18 -!- dioxirane [~OXO@unaffiliated/dioxirane] has quit [Remote host closed the connection] 18:04:23 * (require :sb-sprof) 18:04:23 ("SB-SPROF") 18:04:23 * (find-package :asdf) 18:04:23 NIL 18:04:26 YES! 18:04:53 isn't that already the case? 18:05:48 it is, sb-sprof doesn't use asdf 18:08:59 dioxirane [~OXO@unaffiliated/dioxirane] has joined #sbcl 18:11:23 -!- ASau [~user@46.115.100.212] has quit [Remote host closed the connection] 18:12:15 ASau [~user@46.115.100.212] has joined #sbcl 18:15:06 oh sorry 18:15:15 let's try with sb-md5 instead 18:16:41 works 18:21:02 what the hell, ; caught WARNING: ; The variable #:THING29 is defined but never used, why a full warning? 18:22:57 i see ;; There's no reason to accept this kind of equivocation when compiling our own code, though. 18:26:13 *stassats`* guessed right who added it 18:33:40 -!- dioxirane [~OXO@unaffiliated/dioxirane] has quit [Quit: leaving] 18:35:27 -!- yacks [~yacks@180.151.36.169] has quit [Remote host closed the connection] 18:39:48 -!- wbooze [~wbooze@xdsl-78-35-166-231.netcologne.de] has quit [Ping timeout: 256 seconds] 18:46:43 -!- leuler [~user@p548FC277.dip.t-dialin.net] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 18:52:40 who was it? 18:53:37 dioxirane [~OXO@unaffiliated/dioxirane] has joined #sbcl 18:53:45 FareWell: Krystof 19:03:46 -!- FareWell is now known as Fare 19:04:04 dioxriane [~OXO@unaffiliated/dioxirane] has joined #sbcl 19:05:36 -!- dioxirane [~OXO@unaffiliated/dioxirane] has quit [Ping timeout: 264 seconds] 19:08:15 whoa, I got sbcl confused 19:08:53 it's in /opt/local -- I export no $SBCL_HOME, but it thinks it's in /usr/local and still has SYS: pointing to the source tree. WTF? 19:10:09 SBCL_HOME doesn't affect SYS 19:12:43 so SYS stays forever at the place the thing was compiled? 19:12:54 no, until it's changed 19:15:48 -!- dioxriane [~OXO@unaffiliated/dioxirane] has quit [Ping timeout: 264 seconds] 19:47:19 wbooze [~wbooze@xdsl-84-44-210-220.netcologne.de] has joined #sbcl 20:21:23 stassats`: annoying sbcl developers since 2003 or so 20:29:27 dioxirane [~OXO@unaffiliated/dioxirane] has joined #sbcl 20:31:54 *stassats`* tries to get add :load-toplevel to all eval-when (:compile-toplevel :execute), stumbles somewhere in cold-init 20:32:35 looks like insts.lisp is loaded before disassem.lisp, or something of that sort 20:33:09 and cold-init is an exceedingly cold place for debugging... 20:44:36 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 248 seconds] 21:07:12 Thra11 [~thrall@87.115.108.182] has joined #sbcl 21:24:58 -!- leoc [~leoc.git@p57B9B606.dip.t-dialin.net] has quit [Read error: Connection reset by peer] 21:38:17 sdemarre [~serge@91.176.241.92] has joined #sbcl 21:44:31 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 260 seconds] 21:50:48 It's been so long since I did the SSE work that I don't even remember what's missing anymore. 21:52:31 Right... something about tracking int and float SSE values, in order to avoid ping-ponging between the FP and int pipes... and then issues with negated types and what happens when the user doesn't specify whether it's a float or int SIMD packet. 21:55:38 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 22:00:24 is it better to have trivial .asd files in the contrib directory, or no .asd files and instead a magic function that tells asdf about the contribs? 22:03:26 I'm all for Less Magic; is there any interaction with the goal of not leaving any stray pathname references in dumped cores? 22:04:45 maybe a small one. magic functions requires more magic to not leave stray pathname 22:05:48 but I never go around that part of the source, and I don't really exercise it either. 22:27:50 trivial one-liner defsystem in a .asd file per contrib that is. 22:27:59 it's more "backward compatible", too. 22:28:35 and while I can see how it could be cleaner, that would require refactoring asdf to implement a feature stelian asked for, first (deferred parsing of defsystem) 22:33:21 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: all hope lost] 22:33:58 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 22:38:16 -!- sdemarre [~serge@91.176.241.92] has quit [Ping timeout: 252 seconds] 22:41:15 -!- prxq [~mommer@mnhm-590c0ece.pool.mediaWays.net] has quit [Quit: Leaving] 22:45:25 -!- Thra11 [~thrall@87.115.108.182] has quit [Read error: Operation timed out] 22:47:44 Thra11 [~thrall@46.208.89.201] has joined #sbcl 22:55:35 -!- dioxirane [~OXO@unaffiliated/dioxirane] has quit [Remote host closed the connection]