00:01:45 ASau [~user@p5797EB14.dip0.t-ipconnect.de] has joined #sbcl 00:15:04 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 00:30:32 Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has joined #sbcl 01:03:19 -!- Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has quit [Ping timeout: 260 seconds] 01:06:40 prxq_ [~mommer@mnhm-5f75ca32.pool.mediaWays.net] has joined #sbcl 01:10:06 -!- prxq [~mommer@mnhm-590c0d5d.pool.mediaWays.net] has quit [Ping timeout: 258 seconds] 01:15:04 Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has joined #sbcl 01:55:08 -!- prxq_ [~mommer@mnhm-5f75ca32.pool.mediaWays.net] has quit [Ping timeout: 245 seconds] 02:06:54 prxq_ [~mommer@mnhm-5f75ca32.pool.mediaWays.net] has joined #sbcl 02:18:22 -!- nyef [~nyef@pool-71-161-84-192.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 02:28:40 les [moreorles@lesharris.com] has joined #sbcl 02:28:40 -!- les [moreorles@lesharris.com] has quit [Changing host] 02:28:40 les [moreorles@fsf/member/les] has joined #sbcl 02:29:12 -!- les [moreorles@fsf/member/les] has quit [Client Quit] 02:29:32 les [moreorles@fsf/member/les] has joined #sbcl 02:29:34 -!- les [moreorles@fsf/member/les] has quit [Client Quit] 02:31:27 les [moreorles@lesharris.com] has joined #sbcl 02:31:28 -!- les [moreorles@lesharris.com] has quit [Changing host] 02:31:28 les [moreorles@fsf/member/les] has joined #sbcl 02:35:25 echo-area [~user@182.92.247.2] has joined #sbcl 02:37:28 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 02:43:11 -!- Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has quit [Ping timeout: 260 seconds] 02:44:44 Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has joined #sbcl 02:49:35 -!- ASau [~user@p5797EB14.dip0.t-ipconnect.de] has quit [Quit: fsdb.] 03:04:08 ASau [~user@p5797EB14.dip0.t-ipconnect.de] has joined #sbcl 03:30:10 -!- Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has quit [Quit: leaving] 03:30:42 -!- Bike [~Glossina@67-5-212-240.ptld.qwest.net] has quit [Ping timeout: 264 seconds] 03:41:32 Bike [~Glossina@67-5-212-240.ptld.qwest.net] has joined #sbcl 03:44:52 tsuru```` [~charlie@adsl-98-87-43-116.bna.bellsouth.net] has joined #sbcl 03:46:23 -!- tsuru``` [~charlie@adsl-74-179-250-226.bna.bellsouth.net] has quit [Ping timeout: 245 seconds] 03:53:31 -!- slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has quit [Remote host closed the connection] 03:54:04 slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has joined #sbcl 04:05:45 -!- les [moreorles@fsf/member/les] has quit [Quit: leaving] 04:06:23 -!- ASau [~user@p5797EB14.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 04:15:02 -!- yacks [~py@180.151.36.168] has quit [Read error: Connection reset by peer] 04:23:41 yacks [~py@180.151.36.168] has joined #sbcl 04:26:09 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 04:30:50 -!- yacks [~py@180.151.36.168] has quit [Quit: Leaving] 04:44:17 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 04:59:20 yacks [~py@180.151.36.168] has joined #sbcl 05:47:41 sdemarre [~serge@198.146-66-87.adsl-dyn.isp.belgacom.be] has joined #sbcl 06:37:13 -!- echo-area [~user@182.92.247.2] has quit [Ping timeout: 245 seconds] 06:40:20 echo-area [~user@182.92.247.2] has joined #sbcl 06:50:55 -!- wbooze [~wbooze@xdsl-78-35-187-173.netcologne.de] has quit [Quit: none] 06:57:22 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 07:10:09 -!- Bike [~Glossina@67-5-212-240.ptld.qwest.net] has quit [Quit: leaving] 07:18:43 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 07:33:08 woops, pkhuong, around? what did you have in mind for hygienic macros? 07:34:05 wbooze [~wbooze@xdsl-78-35-187-173.netcologne.de] has joined #sbcl 07:39:19 Quadrescence: for SBCL? take hygienic defmacro from scheme, adapt it for IR1. 07:39:31 for conversion from sexps to ir1, that is. 07:39:55 pkhuong, what do you mean "adapt it for IR1"? 07:40:16 -!- sdemarre [~serge@198.146-66-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 272 seconds] 07:40:27 our lexenv already maps from bindings to complex IR1 structure. 07:43:44 and adapt it as in implement and find out where there's a mismatch. 07:51:32 Quadrescence: probably something like hygienic expanders are associated with their parent lexenv (child for macroletrec), and the expansion further IR1 translated in the parent lexenv. 08:02:53 Quadrescence: and pass pairs of sexp + lexenv to expanders, rather than only sexps. 08:04:12 i see 08:12:30 gabnet [~gabnet@ACaen-652-1-281-41.w92-154.abo.wanadoo.fr] has joined #sbcl 08:18:29 -!- wbooze [~wbooze@xdsl-78-35-187-173.netcologne.de] has quit [Ping timeout: 255 seconds] 08:37:49 attila_lendvai [~attila_le@95.56.77.199] has joined #sbcl 08:37:49 -!- attila_lendvai [~attila_le@95.56.77.199] has quit [Changing host] 08:37:49 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 08:43:43 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 08:43:44 attila_lendvai1 [~attila_le@95.56.77.199] has joined #sbcl 08:44:04 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 08:47:40 stassats [~stassats@wikipedia/stassats] has joined #sbcl 08:49:18 -!- attila_lendvai1 [~attila_le@95.56.77.199] has quit [Ping timeout: 272 seconds] 08:51:34 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 08:52:05 attila_lendvai [~attila_le@95.56.77.199] has joined #sbcl 08:52:05 -!- attila_lendvai [~attila_le@95.56.77.199] has quit [Changing host] 08:52:05 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 09:08:36 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 264 seconds] 09:10:01 attila_lendvai [~attila_le@95.56.77.199] has joined #sbcl 09:10:01 -!- attila_lendvai [~attila_le@95.56.77.199] has quit [Changing host] 09:10:01 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 09:17:19 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 09:17:19 attila_lendvai1 [~attila_le@95.56.77.199] has joined #sbcl 09:24:44 -!- attila_lendvai1 [~attila_le@95.56.77.199] has quit [Ping timeout: 268 seconds] 09:25:56 attila_lendvai [~attila_le@95.56.77.199] has joined #sbcl 09:25:56 -!- attila_lendvai [~attila_le@95.56.77.199] has quit [Changing host] 09:25:56 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 09:28:53 Ralt_ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has joined #sbcl 09:28:56 hi 09:29:13 well, you can 09:29:20 :) 09:29:32 for context 09:29:39 I'm trying to compile sbcl with zlib support 09:29:42 and you can reproduce it again? 09:29:45 but sbcl doesn't compile 09:29:47 http://pastebin.com/g2RgPTDe 09:29:54 stassats: every time I run sh make.sh 09:30:08 how did you enable zlib support? 09:30:30 I didn't try for now 09:30:34 I'm just trying to compile 09:30:49 have you compiled it before successfully? 09:30:51 I don't even know how to do it, but I can't even compile so zlib support isn't there yet... 09:30:57 nope, just downloaded the binaries 09:31:03 Ralt_: looks like you have bad RAM. 09:31:04 maybe I'm missing some dependencies? 09:31:07 as you can see, this is very strange and shouldn't happen 09:31:20 pkhuong: yeah looks like so 09:31:28 but... should I memtest? 09:32:45 pnpuff [~pnpuff@unaffiliated/pnpuff] has joined #sbcl 09:32:48 just for my information, can one of you git clone https://github.com/Ralt/mdtransform && cd mdtransform && make && du -hs dist/mdtransform and tell me the output? :) 09:33:14 will it do rm -rf ~ ? 09:33:26 nope 09:33:34 but you'd have to check yourself 09:33:43 although the code isn't that big. 09:34:16 it just compiles some code than saves an image 09:34:41 and tells me whether a compressed image is smaller than 50mb 09:34:42 have you checked that the downloaded sbcl sources/binaries are not corrupted? 09:34:52 re-downloaded, yes 09:35:03 yes, it'd be around 10MB 09:35:05 although it's an archive, so corrupted download would be pretty easy to see 09:35:43 cool 09:36:03 10-15, if you want even lower, use 32-bit sbcl 09:36:26 nah it's ok 09:36:31 I'll just reboot to memtest right now 09:36:36 cya 09:40:53 -!- Ralt_ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has quit [Ping timeout: 245 seconds] 09:41:34 hi... I find a strange `. output/prefix.def' at line 13 of the install.sh script in the sbcl-1.1.6 dir obtained untarring the sbcl-1.1.6-source.tar.bz2, but there is not a prefix.def file or directory. What is the meaning of that line? thx 09:42:48 The script interrupts after "INSTALL_ROOT=/usr/local sh install.sh" 09:44:45 pnpuff: if you download the source package, you have to build before installing. 09:52:39 capisce [~srodal@cm-84.215.35.251.getinternet.no] has joined #sbcl 10:06:21 pkhuong: thx, so make.sh script performs a build and an installation. But, sorry, what's the purpose of the install.sh script in the source package? 10:06:41 make.sh doesn't perform that 10:06:59 what made the impression that it does? 10:07:31 -!- gabnet [~gabnet@ACaen-652-1-281-41.w92-154.abo.wanadoo.fr] has quit [Quit: Ex-Chat] 10:10:32 stassats: --xc-host and --prefix options? 10:13:14 stassats: anyway you're right, to install sbcl there is the install.sh script 10:13:35 Strigoides [~owen@114-134-0-86.lightwire.co.nz] has joined #sbcl 10:17:36 -!- astertronistic [~astertron@ip70-181-235-122.sd.sd.cox.net] has quit [Quit: Leaving] 10:17:45 wbooze [~wbooze@xdsl-84-44-177-63.netcologne.de] has joined #sbcl 10:20:23 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 10:30:08 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 10:38:58 rahul [~chatzilla@14.139.60.220] has joined #sbcl 10:39:22 -!- rahul is now known as Guest75054 10:41:03 -!- Guest75054 [~chatzilla@14.139.60.220] has quit [Client Quit] 10:42:41 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 256 seconds] 10:44:32 ASau [~user@p5797EB14.dip0.t-ipconnect.de] has joined #sbcl 10:46:07 -!- pnpuff [~pnpuff@unaffiliated/pnpuff] has left #sbcl 10:53:23 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 11:04:35 foreignFunction [~niksaak@ip-4761.sunline.net.ua] has joined #sbcl 11:14:40 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 260 seconds] 11:27:48 attila_lendvai [~attila_le@95.56.69.231] has joined #sbcl 11:27:48 -!- attila_lendvai [~attila_le@95.56.69.231] has quit [Changing host] 11:27:48 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 11:38:55 leuler [~user@p548FDD96.dip0.t-ipconnect.de] has joined #sbcl 11:44:17 drmeiste_ [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 11:45:56 -!- Strigoides [~owen@114-134-0-86.lightwire.co.nz] has quit [Quit: leaving] 11:46:47 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Ping timeout: 258 seconds] 11:50:50 Strigoides [~owen@114-134-0-121.lightwire.co.nz] has joined #sbcl 11:57:51 -!- milosn_ is now known as milosn 12:32:04 LiamH [~none@96.231.225.69] has joined #sbcl 12:36:02 nyef [~nyef@pool-71-161-84-192.cncdnh.east.myfairpoint.net] has joined #sbcl 12:44:18 stassats [~stassats@wikipedia/stassats] has joined #sbcl 12:46:39 -!- Strigoides [~owen@114-134-0-121.lightwire.co.nz] has quit [Quit: leaving] 13:20:36 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 13:25:46 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 13:35:58 -!- p_l [~pl@2002:5965:ded2::1] has quit [Remote host closed the connection] 13:56:33 Ralt_ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has joined #sbcl 13:56:36 alright 13:56:39 did a full memset 13:56:40 no error 13:57:02 but still this error to compile sbcl http://pastebin.com/g2RgPTDe 13:57:34 scymtym_ [~user@ip-5-147-116-166.unitymediagroup.de] has joined #sbcl 13:57:53 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: supernova explosion] 13:58:41 -!- drmeiste_ [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 14:03:02 do you use anything special? 14:03:09 xen? some kind of security? 14:03:11 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 14:04:31 what happens when you run something else in SBCL? 14:06:29 stassats: I don't think so 14:06:35 pkhuong: nothing wrong 14:06:46 I can run my app perfectly fine 14:07:47 What happens if you build in a directory with no non-ASCII characters in the pathname? 14:07:55 hm 14:07:59 trying that 14:08:02 ah, that's it! 14:08:20 what's LANG and all that then? 14:08:27 *nyef* has been known to have problems due to his paths containing an extra "src/" component. 14:08:59 nyef: that early in the bootstrap? 14:09:02 http://pastebin.com/TJrRBSL3 14:09:04 that's the host keeling over 14:09:41 hehe 14:09:44 works perfectly fine 14:09:47 thanks nyef ! 14:09:54 sbcl is compiling now 14:09:55 lp 601576 14:09:56 https://bugs.launchpad.net/bugs/601576 14:10:33 finally 14:10:40 good to see something running :-) 14:10:53 now I have to see how I compile sbcl with zlib support 14:10:54 to enable compression, use --with-sb-compression 14:10:58 heh. 14:11:13 err, sb-CORE-compression 14:11:29 running :) 14:11:33 sbcl-צה lol 14:12:03 I like the view of compilations. 14:12:10 mostly because it means it's working. 14:12:22 then you should have settled with gentoo, not ubuntu 14:12:27 :P 14:12:34 it's my work pc 14:12:38 everybody's using ubuntu 14:12:43 i'm not 14:12:48 and only ubuntu is supported 14:12:58 there are tons of company-related packages 14:13:09 also, I like ubuntu because I don't have to configure anything, it just works 14:13:21 honestly, it's mostly about the window manager anyway 14:13:36 (and I use awesome wm) 14:15:03 *stassats* puts that bug into the queue to see what can be done 14:15:09 because failing in that way is just bonkers 14:15:18 can you reproduce? 14:15:27 see the link above 14:15:33 I see 14:15:48 it's known for quite some itme 14:16:04 I never would've thought of looking for this though... 14:16:31 stassats: are you an sbcl dev? 14:16:45 yes 14:17:06 k :) 14:18:05 most vocal people on #sbcl are 14:18:36 No, not that early in the bootstrap, but it's an example of a pathname-related problem with a build. 14:18:58 there is lot of style warnings 14:19:04 don't mind them 14:19:12 unless you want to fix them 14:19:19 heh 14:19:23 there really is a lot of them :P 14:19:40 I'm not good enough at lisp yet to start working on SBCL 14:19:51 although it'd be pretty interesting 14:19:58 There's a place or two in the build where full warnings get ignored on certain platforms. 14:20:54 sbcl is so picky about style, it can't follow it itself 14:22:15 You want spectacularly poor style in SBCL? There's a place where EVAL is used to define functions at macroexpand-time. 14:22:54 why? 14:23:15 Good question. 14:24:10 I know there is some kind of "tell me what are sbcl options" function, but I can't find it anymore 14:24:40 --help 14:24:48 I think that it's one of those things that very few people know about, and that everyone who's tried to figure out how to clean it up has given up and suppressed their memory of the damage. 14:24:56 doesn't tell me about this command 14:25:04 which command? 14:25:05 nyef: I see. 14:25:15 stassats: --help doesn't show me which option should be used 14:25:22 for what? 14:25:43 I'd like to know if zlib support is correctly built-in before installing the sbcl I just compiled 14:25:52 *features* 14:26:04 ah, thanks 14:26:32 doesn't look like it's here 14:26:44 (or #+sb-core-compression t) 14:27:11 definitely isn't there :) 14:27:15 alright 14:27:21 I'll look at the doc to see what I did wrong 14:27:27 did you do --with-sb-core-compression? 14:27:29 Okay, time to build and try to test my current tree, even if the control logic for M-A-O is still outright missing when it comes to dynamic space. 14:27:40 not like i told you wrongly --with-sb-compression the first time 14:27:52 stassats: yeah, I did sh make.sh --with-sb-core-compression 14:28:01 ah no damnit 14:28:21 I did it, then cancelled it because xterm couldn't handle so much output then redid it without 14:28:51 rerunning compilation :-) 14:30:42 not xterm actually, gnome-terminal 14:32:14 sbcl can't also start in deleted directories 14:33:10 ? 14:34:20 attila_lendvai [~attila_le@95.56.69.231] has joined #sbcl 14:34:20 -!- attila_lendvai [~attila_le@95.56.69.231] has quit [Changing host] 14:34:20 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 14:34:58 yay 14:35:05 12mb image instead of 50mb 14:35:29 ... And I just got to first genesis. 14:35:42 Maybe I should get a faster machine. 14:35:59 or make sbcl compiler faster 14:36:02 what's easier? 14:36:32 I'm having enough trouble with functionality shortfalls, improved compilation speed is beyond me. 14:37:11 DAMNIT. Forgot to enable :trace-file for the file I've been hacking. )-: 14:37:43 *nyef* restarts his build. 14:38:23 that calls for think-hard debugging and overnight/overlunch builds 14:38:36 segv- [~mb@95-91-241-75-dynip.superkabel.de] has joined #sbcl 14:39:06 Also for SLAMmable changes. 14:39:38 Not that I'm convinced that I cat get SLAM working right now. 14:39:56 Or that it's entirely appropriate for what I'm doing. 14:53:27 *stassats* has a potential fix for the CWD business, will commit after the freeze 14:54:20 i need to actually test it first 14:55:50 it basically catches the error, turns it into a warning and defaults *default-pathname-defaults* to #p"" instead 14:58:47 *stassats* likes (ERROR "argh! can't happen") ===> ;note: deleting unreachable code 14:58:50 it truly can't 15:00:05 Hrm. 10 ERROR conditions, 13 WARNINGs, 12 STYLE-WARNINGs, and 25 notes. Clearly not yet ready for testing. /-: 15:06:04 WARNING: Error while initializing *DEFAULT-PATHNAME-DEFAULTS*with the current directory: #. Using #P"" instead. 15:06:07 looks better 15:06:47 strange, it only works with my image, not with the default sbcl.core 15:10:59 seems like the problem with printing 15:11:50 (stream-reinit t) is called after (os-cold-init-or-reinit) 15:17:28 and putting (stream-reinit t) first didn't help 15:18:36 *stassats* puts (thread-init-or-reinit) just in case first too 15:19:10 though os-cold-init-or-reinit being third seems a bit strange, although despite its name all it does is initializing pathnames and *posix-argv* 15:26:05 (os-cold-init-or-reinit) doesn't even need to run under without-gcing 15:28:40 There are a few "interesting" dependencies involved in cold-init, you might be running into those. 15:29:20 it builds fine, i just can't print from reinit 15:30:27 and can print from when i save a new core 15:33:41 even at the very end of reinit 15:37:27 this is not funny, i'm editing the code one machine, and compiling via ssh on another 15:37:34 no wonder it fails no matter what i do 15:40:00 my laptop tends to heat too much when compiling sbcl, but i want to lay on the couch 15:42:23 and of course, putting (stream-reinit t) first works as expected 15:52:02 -!- segv- [~mb@95-91-241-75-dynip.superkabel.de] has quit [Remote host closed the connection] 15:56:34 ok, decoding errors and missing directories now work nicely, now to test cores and run-times in weird places 16:23:22 Bike [~Glossina@67-5-212-240.ptld.qwest.net] has joined #sbcl 16:28:06 -!- Bike [~Glossina@67-5-212-240.ptld.qwest.net] has quit [Ping timeout: 252 seconds] 16:29:46 Bike [~Glossina@67-5-212-240.ptld.qwest.net] has joined #sbcl 16:33:10 *lichtblau* *boggle* 16:33:32 is lp#1173996 for real? 16:34:15 lp 1173996 16:34:16 https://bugs.launchpad.net/bugs/1173996 16:35:02 That's... impressive, if true. 16:35:03 that's weird, i'll check it 16:36:17 -!- stassats [~stassats@wikipedia/stassats] has quit [Remote host closed the connection] 16:36:41 stassats [~stassats@wikipedia/stassats] has joined #sbcl 16:37:58 *stassats* transformed memory faults and WARNING: Starting a poll(2) without a timeout while interrupts are disabled. into nice 16:38:01 "WARNING: Error initializing *RUNTIME-PATHNAME*: :ASCII c-string decoding error: the octet sequence #(209) cannot be decoded. Using NIL instead." 16:39:13 (POSIX-GETENV "SBCL_HOME") for SB-INT:SBCL-HOMEDIR-PATHNAME still ends up in a debugger, but not in ldb and with interrupts enabled 16:41:03 arguably, all these *runtime-pathname* initialization can be run in such context too, though ending up in the debugger is probably not necessary in any case 16:44:36 and i can't reproduce this windows thing on 1.1.4.27 16:46:58 what i get sometimes, is 0 printed twice 16:47:37 something's with flushing 16:47:57 Multiple threads competing over the same stream? 16:48:12 yeah, something like that 16:48:58 need to build 1.1.6, but i'm sure it'll be the same, will mark invalid after the build 16:50:13 Ahh, yet another stupid bug in my little project stomped. Maybe this time I'll get something close to a working build. 16:58:53 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 245 seconds] 16:59:01 *stassats* likes the condescending "Do you guys plan to fix this:" 16:59:04 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 16:59:06 attila_lendvai1 [~attila_le@95.56.69.231] has joined #sbcl 16:59:06 -!- attila_lendvai1 is now known as attila_lendvai 16:59:06 -!- attila_lendvai [~attila_le@95.56.69.231] has quit [Changing host] 16:59:07 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:03:09 With all the commits already announced we will be at 1.1.7.50 in no time after the freeze is over. 17:03:41 *stassats* wishes to polish up his half-done changes 17:04:00 to get it up to .55 17:04:59 Go stassats! 17:05:11 go free time 17:06:01 ok, 1.1.6.current works as expected on windows with variables 17:06:27 (and i marked it invalid without waiting) 17:06:41 Not marked it as a streams bug? 17:07:13 nope, you bring it on yourself by printing concurrently 17:07:46 Fair enough. 17:07:57 somebody else can open a bug/wishlist 17:08:16 Though I rather imagine that having thread-safe streams for *standard-output* and whatnot would be A Good Thing. 17:08:30 <|3b|> is that printing from more than 1 thread? 17:08:35 i think they are thread-safe, just the result may be not expected 17:08:53 So, current bug in my new ROOM hack: "Objects of type SPECIALIZED-ARRAY-ELEMENT-TYPE-PROPERTIES can't be dumped into fasl files." 17:10:08 and now that i'm on windows, i can run tests 17:10:30 attila_lendvai1 [~attila_le@95.56.69.231] has joined #sbcl 17:10:31 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 17:10:34 *stassats* wages that they will hang at some poiint 17:13:54 -!- LiamH [~none@96.231.225.69] has quit [Ping timeout: 264 seconds] 17:18:06 and it did 17:25:56 nyef: l-t-v that up? 17:28:17 I might have to, yes. 17:28:51 Unfortunately, it's *ROOM-INFO* that's affected. 17:29:00 Still, it's doable. 17:34:11 Okay, vm-array comes before room in the build order, which means that its toplevels should be run first. That's workable. 17:45:58 Ugh. make-array :initial-contents #.`(list ,@(map 'list ...)). Ugh, ugh, ugh. 18:05:27 huh, run-program output redirection to string streams does not work? 18:12:18 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 18:18:55 all the fd-handler stuff is left out on windows 18:19:45 TuckerD [~user@c-98-216-148-67.hsd1.nh.comcast.net] has joined #sbcl 18:20:39 hmm, did I merge all of akovalenko's run-program fixes? 18:21:07 my recollection is "no", but it's a rather hazy memory 18:21:15 arrdem [~arrdem@dhcp-53-132.ece.utexas.edu] has joined #sbcl 18:28:26 -!- Posterdati [~antani@host183-239-dynamic.16-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection] 18:28:35 LiamH [~none@96.231.225.69] has joined #sbcl 18:30:28 Posterdati [~antani@host183-239-dynamic.16-87-r.retail.telecomitalia.it] has joined #sbcl 18:31:12 -!- attila_lendvai1 [~attila_le@95.56.69.231] has quit [Quit: Leaving.] 18:32:14 I'm writing a proposal for the google summer of code for work on vectorizing sequence operations and I'm wondering what the sequence operations people find most useful are, and if you use integers or floats more often. 18:32:28 that's enough windows for today 18:32:58 TuckerD: i would find a vectorized utf-8 length functions useful 18:33:28 7-bit ascii tests, for that matter, too 18:34:01 though, i can't say that simply applying masks is too slow for me 18:34:27 i should finish at least that and then worry about sse tricks 18:35:56 stassats: I hadn't actually thought much about char operations, I mostly work on scientific stuff but that probably isn't super useful to most people, so thanks for the idea. 18:36:32 simple things like (count 1 bit-vector) or (position 0 bit-vector) 18:36:38 and not only bit-vectors 18:45:19 sorry folks: release delayed until tomorrow; if I tried to do it now I would get it all wrong and/or fall asleep half-way through it 18:46:26 at least i had an opportunity to test the windows build 18:50:53 Ralt__ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has joined #sbcl 18:52:08 -!- Ralt_ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has quit [Ping timeout: 245 seconds] 18:52:29 -!- leuler [~user@p548FDD96.dip0.t-ipconnect.de] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 19:01:33 Well, I now have a "working" version of M-A-O, provided that it doesn't need to map over dynamic space on gencgc. 19:02:06 astertronistic [~astertron@ip70-181-235-122.sd.sd.cox.net] has joined #sbcl 19:02:16 Only lightly tested, of course, and there's a case where it will claim 64 bytes for 1 cons object, but said cons is NIL, which actually does take up that much space. 19:02:35 (Should I change my mind, and have NIL be a symbol for these purposes?) 19:20:56 conversion and comparison between simple-base-string and (simple-array character 1) would be nice. 19:22:07 nyef: treating NIL as a cons seems to violate least surprise. 19:26:04 Strigoides [~owen@114-134-0-67.lightwire.co.nz] has joined #sbcl 19:29:36 -!- daimrod [daimrod@sbrk.org] has quit [Ping timeout: 264 seconds] 19:30:07 That's true. At least it's an easy enough thing to change, I just switch out the type tag. 19:31:16 opportunistic type tagging ....lol 19:31:35 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 19:32:34 wbooze: Not exactly, it's more that this particular part of the heap doesn't follow the rules, so I need to arrange for the first object to be handled by dead reckoning before passing control on to the more normal logic. 19:33:20 daimrod [daimrod@sbrk.org] has joined #sbcl 19:41:43 -!- wbooze [~wbooze@xdsl-84-44-177-63.netcologne.de] has quit [Read error: Connection reset by peer] 19:57:11 sdemarre [~serge@198.146-66-87.adsl-dyn.isp.belgacom.be] has joined #sbcl 20:07:24 -!- Ralt__ [5d083c57@gateway/web/freenode/ip.93.8.60.87] has quit [Quit: Page closed] 20:11:31 wbooze [~wbooze@xdsl-84-44-177-63.netcologne.de] has joined #sbcl 20:23:01 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 248 seconds] 20:28:39 -!- wbooze [~wbooze@xdsl-84-44-177-63.netcologne.de] has quit [Read error: Connection reset by peer] 20:29:26 aba` [~user@ip-5-147-116-166.unitymediagroup.de] has joined #sbcl 20:30:24 wbooze [~wbooze@xdsl-78-35-152-231.netcologne.de] has joined #sbcl 20:31:32 -!- sdemarre [~serge@198.146-66-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 258 seconds] 20:33:49 -!- antifuchs [~foobar@boots.boinkor.net] has quit [Ping timeout: 246 seconds] 20:34:10 -!- whoops [~whoops@2a01:4f8:161:41e1::2] has quit [Ping timeout: 246 seconds] 20:34:46 Hi, i am applying for the register allocation project for GSoC and i have worked on bug #789497. shall i put the patch into paste lisp for review or attach it to the launchpad bug? 20:35:13 -!- fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has quit [Ping timeout: 246 seconds] 20:36:01 fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has joined #sbcl 20:37:44 antifuchs [~foobar@boots.boinkor.net] has joined #sbcl 20:38:23 whoops [~whoops@2a01:4f8:161:41e1::2] has joined #sbcl 20:45:39 lp 789497 20:45:39 https://bugs.launchpad.net/bugs/789497 20:46:23 aba`: launchpad will be easier for follow up, and people will be notified automatically. 20:47:04 pkhuong: ok, i'll try! 20:48:08 -!- LiamH [~none@96.231.225.69] has quit [Ping timeout: 272 seconds] 21:22:41 pkhuong: What do you think about this paper by Briggs "Improvements to Graph Coloring Register Allocation"? 21:28:17 -!- prxq_ [~mommer@mnhm-5f75ca32.pool.mediaWays.net] has quit [Quit: Leaving] 21:29:30 aba`: I remember it being pretty nice. 21:35:12 aba`: a simple hack might be to find good points in the CFG to break some/all live ranges that cross them... but I don't think that's a reasonable project without first getting familiar with the current register allocator. 21:41:03 pkhuong: sure, i would like to start with simple exploration 21:42:24 pkhuong: would you have time to read my proposal if i send it to you per email? 21:42:59 aba`: sure. 21:43:25 I'll be crossing the atlantic in 9 hours, so expect some latency. 21:43:58 pkhuong: thanks! 21:44:43 I'm applying for Vectorization of sequence operations for GSoC and am wondering what would be a good target for the number of functions to work on, I'm thinking 4-5, those being length,count find(and its variations), map(-into) and reduce. I'm not sure if this is enough, but I don't want to over commit myself. 21:46:18 there's nothing to vectorize in length. 21:47:46 map-into and reduce will be a bit more challenging, particularly reduce, but they'll probably make for nice demos. 21:48:36 and reduce of FP arithmetic might be a no go, for strict FP compliance (even though parallel sul=ms are usually more accurate...). 21:53:37 equal/mismatch might be easier two-argument operations than map-into. replace of differently-typed vectors would probably be "interesting." 21:54:31 You can always add more goals later... it seems best to me to start small, and try and construct a progression in complexity, so that each new step only adds one new challenge. 22:03:07 pkhoung:What about reduce would violate FP compliance? And I was thinking that it was best to start small but I wasn't sure what would be best, thinking about it a lot of code for functions like find/remove/substitute will be pretty similar, so I think I'll start with the those, and maybe work on map and reduce later if time permits. Also I would think that map and map-into would be about the same difficultly, would map-into be much 22:03:08 harder? 22:05:59 could there be an equivalent of -ffast-math for SBCL to allow non-compliant FP operations? 22:06:12 reduce of + performs the sum serially. If you want to vectorise that, you'll compute sums in parallel and then combine them. 22:06:24 Fiora: sure. 22:08:13 the standard's strict enough to disallow kahan sums and stuff, huh... 22:09:48 TuckerD: I'd think it'd be easiest to have map expand into map-into. You might want to handle misalignment in the result and in the input. The problem I see with map-into is that you'll likely want to have multiple inputs in addition to the result sequence (and to test for the mapped function's identity, but that's not too hard once you see how it's done). 22:12:15 Remember that we can massage our allocation granularity if we absolutely must. And that there's a design that takes advantage of an increased allocation granularity for non-conses to produce 31-bit fixnums on 32-bit platforms. 22:12:53 nyef: ohh, interesting (: but we're only 64 bit so far. 22:13:16 and I'm more interested in making sure we always allocate even numbers of stack slots... 22:13:39 Heh. Did you forget us kicking around the possibility of 31-bit fixnums as a follow-on project from the 63-bit fixnum project? (-: 22:14:34 I vaguely remember you saying it looked really hairy. 22:15:34 Sure, because we'd need a different allocation granularity for CONSes vs. everything else, we're STILL short a tag, the allocation granularity stuff affects the GC implementation... 22:16:20 right... and you don't feel like doubling the size of conses (: 22:16:46 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 22:16:49 Umm. I wouldn't double the size of conses. 22:17:06 I'd give them two out of sixteen lowtags instead. 22:17:43 During tests of my assert patch, i encountered an error in float.pure.lisp in the expression (assert (typep (nth-value 1 (ignore-errors (float-radix "notfloat"))) 'type-error)): the subexpression (nth-value ...) seems to return NIL. My macro expansion is here: http://paste.lisp.org/display/136940 . Am I doing something wrong? or is it a bug in IGNORE-ERRORS? 22:17:43 GC isn't too hard though. Our heuristic traversal of cons lists means we can just align when allocating non conses... conses are already mostly copied back to back. 22:18:45 aba`: You might double-check to make sure that (float-radix "notfloat") still errors. 22:19:24 pkhuong: Yeah, I still worry. Especially since I've just spent about a day and a half wading through the guts of ROOM and its interaction with the GC. 22:23:57 nyef: in isolation the individual expressions work, but inside the DO the (nth-value ..) seems to get evaluation to NIL 22:24:42 aba`: How odd. Possible differences between the interpreter and the compiler, or compiler policy effects maybe? 22:25:03 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving] 22:25:29 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 22:25:32 Ugh. heap_base is declared as static, but I need access to it. 22:26:00 At which point I should rename it to something less ambiguous. Such as gencgc_heap_base or something. 22:26:07 Anyway, food time. Back in a while. 22:37:15 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving] 22:37:34 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 22:40:11 seems like an inling bug. 22:40:17 * inlining. good night 22:41:49 the first version of the patch is attached to the launchpad bug. 22:42:20 thanks for help, ciao 22:45:46 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 22:48:13 ok. Don't know if it's a bug or not... I'm inclined to say no. 22:48:54 we don't check type argument type of float-radix before returning 2. In most cases, that's fine because the ftype will kick in, but sometimes, the function is inlined too early. 22:49:16 good news is, there's nothing broken in our inlining logic. 22:50:05 that is, no to a compiler bug (: we should just add a (type float x) declaration to float-radix. 22:51:50 LiamH [~none@96.231.225.69] has joined #sbcl 22:52:22 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 22:57:43 The first page in any gencgc allocation region (open or closed) has a region_start_offset of zero, and this is the only page with such a region_start_offset in the allocation region, right? 22:59:19 pkhuong: adding (TYPE FLOAT X) to FLOAT-RADIX fixed the problem! 23:03:04 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 23:24:23 milosn_ [~milosn@user-5af507c7.broadband.tesco.net] has joined #sbcl 23:26:52 -!- milosn [~milosn@user-5af50474.broadband.tesco.net] has quit [Ping timeout: 260 seconds] 23:27:33 -!- slyrus [~chatzilla@adsl-99-183-240-66.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 248 seconds] 23:34:29 -!- aba` [~user@ip-5-147-116-166.unitymediagroup.de] has quit [Ping timeout: 256 seconds] 23:37:43 -!- lichtblau [~user@port-212-202-134-139.static.qsc.de] has quit [Ping timeout: 256 seconds] 23:48:44 ASau` [~user@p5797F422.dip0.t-ipconnect.de] has joined #sbcl 23:51:43 -!- ASau [~user@p5797EB14.dip0.t-ipconnect.de] has quit [Ping timeout: 252 seconds]