01:50:46 alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 02:03:01 -!- pjb [~t@AMontsouris-651-1-31-168.w82-123.abo.wanadoo.fr] has quit [Read error: Connection reset by peer] 02:08:44 -!- DataLinkDroid [~DataLinkD@1.147.59.31] has quit [Ping timeout: 256 seconds] 02:29:35 DataLinkDroid [~DataLinkD@1.144.169.91] has joined #ccl 02:31:41 -!- alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has quit [Quit: alms_] 02:46:10 -!- segv- [~mb@dslb-092-078-122-164.pools.arcor-ip.net] has quit [Quit: segv-] 03:33:10 alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 03:42:19 -!- alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has quit [Quit: alms_] 03:44:57 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #ccl 03:50:44 -!- DataLinkDroid [~DataLinkD@1.144.169.91] has quit [Quit: Bye] 03:50:59 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 04:02:59 Fare [fare@nat/google/x-bejmuycqghlbhggo] has joined #ccl 04:19:20 -!- PuffTheMagic [uid3325@gateway/web/irccloud.com/x-firleqmlyjcgwyjg] has quit [Ping timeout: 276 seconds] 04:51:56 -!- svs_ [~svs@104-252-AGAVEBB-NM.abq.nm.agavebb.net] has quit [Ping timeout: 252 seconds] 05:04:37 rme_ [~rme@50.43.156.16] has joined #ccl 05:05:43 -!- rme [~rme@50.43.160.36] has quit [Ping timeout: 260 seconds] 05:05:43 -!- rme_ is now known as rme 05:06:10 -!- rme [rme@5DA0AC28.9AB02A83.699BA7A6.IP] has quit [Ping timeout] 05:41:59 -!- Fare [fare@nat/google/x-bejmuycqghlbhggo] has quit [Read error: Operation timed out] 06:15:21 PuffTheMagic [uid3325@gateway/web/irccloud.com/x-dveriualjwsintfq] has joined #ccl 08:19:24 Fare [fare@nat/google/x-umfvysfndjmoncab] has joined #ccl 08:38:07 -!- |3b|` is now known as |3b| 09:09:21 -!- rme [rme@C384DED8.F26E1912.699BA7A6.IP] has quit [Quit: rme] 09:09:21 -!- rme [~rme@50.43.156.16] has quit [Quit: rme] 09:52:22 -!- jdz [~jdz@85.254.212.34] has quit [Quit: Byebye.] 09:54:45 jdz [~jdz@85.254.212.34] has joined #ccl 11:56:45 alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 12:19:11 wws [~billstcla@p-74-209-20-234.dsl1.rtr.chat.fpma.frpt.net] has joined #ccl 12:20:39 -!- wws [wws@clozure-B5CF9DDC.dsl1.rtr.chat.fpma.frpt.net] has quit [Ping timeout] 12:21:50 -!- billstclair [~billstcla@unaffiliated/billstclair] has quit [Ping timeout: 252 seconds] 14:17:16 segv- [~mb@dslb-088-075-117-083.pools.arcor-ip.net] has joined #ccl 14:25:51 -!- wws is now known as billstclair 14:25:58 -!- billstclair [~billstcla@p-74-209-20-234.dsl1.rtr.chat.fpma.frpt.net] has quit [Changing host] 14:25:58 billstclair [~billstcla@unaffiliated/billstclair] has joined #ccl 14:26:05 -!- billstclair is now known as wws 14:41:27 segv-_ [~mb@dslb-088-074-073-253.pools.arcor-ip.net] has joined #ccl 14:42:31 -!- segv- [~mb@dslb-088-075-117-083.pools.arcor-ip.net] has quit [Ping timeout: 245 seconds] 14:42:31 -!- segv-_ is now known as segv- 14:48:58 -!- alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has quit [Quit: alms_] 14:58:06 alms_ [~alms_@209-6-130-32.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 15:05:02 -!- Fare [fare@nat/google/x-umfvysfndjmoncab] has quit [Ping timeout: 245 seconds] 15:09:13 -!- segv- [~mb@dslb-088-074-073-253.pools.arcor-ip.net] has quit [Read error: Connection reset by peer] 15:09:16 segv-_ [~mb@dslb-088-074-073-253.pools.arcor-ip.net] has joined #ccl 16:05:08 gko [~user@114-34-168-13.HINET-IP.hinet.net] has joined #ccl 17:27:24 rme [~rme@50.43.156.16] has joined #ccl 17:39:46 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Remote host closed the connection] 18:38:05 pjb [~t@AMontsouris-651-1-31-168.w82-123.abo.wanadoo.fr] has joined #ccl 19:55:08 clop [~jared@moat3.centtech.com] has joined #ccl 19:56:03 hey guys, does anyone know how save-application works? it seems to be behaving weirdly on our nfs so i'm just wondering what kind of low-level operations it's using 19:59:54 mostly write(). 20:01:26 pnpuff [~pnpuff@unaffiliated/pnpuff] has joined #ccl 20:03:02 -!- pnpuff [~pnpuff@unaffiliated/pnpuff] has quit [Quit: .] 20:04:06 rme_ [~rme@50.43.150.18] has joined #ccl 20:04:20 -!- rme [rme@C384DED8.F26E1912.699BA7A6.IP] has quit [Ping timeout] 20:04:20 -!- rme_ is now known as rme 20:06:06 -!- rme [~rme@50.43.156.16] has quit [Ping timeout: 264 seconds] 20:06:06 -!- rme_ is now known as rme 20:07:34 if (save-application "some-file" :purify nil) behaves any differently, that'd be interesting, but most of what it does is to write (often large numbers of) bytes from memory to a file. I think that the code that does that is fairly good about checking for errors and fairly good about detecting partial writes (where fewer bytes are written than expected) and doing the right thing. 20:14:18 interesting 20:14:32 yeah i mean, the saved application seems fine, it works and runs well as you'd expect 20:14:45 but when we write it, for whatever reason, it's causing our nfs to hang for like 30-40 seconds 20:14:50 like... company wide :( 20:15:35 it's like a 600 mb image, which i guess is kind of big, but still pretty odd, normal copying, etc., doesn't have that kind of issue 20:18:42 I think that we try to write in large chunks, and that may confuse whatever version of NFS you're using. I don't know for sure. 20:20:21 rgr 20:20:34 hard to debug, too many things involved :( 20:21:21 Can you mock your NFS server vendor? Probably not if it's just a Linux box... 20:21:40 it's some oracle thing 20:21:51 so, i think we can all mock it 20:25:01 the function that does most of the writing is called writebuf(), in lisp-kernel/image.c. It tries to write at most INT_MAX bytes (because of a Darwin bug) and sets the variable 'this_size' accordingly. If you set 'this_size' to something relatively small - like 4096 - does NFS like that better ? 20:27:35 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #ccl 20:27:46 thanks -- i'll try it 20:28:14 (but i guess i shouldn't try it until tomorrow :) 20:28:40 Blame me! 20:29:16 fortunately everyone is blaming sol right now :D 20:30:16 Also: is the image that you're saving -from- on a local filesystem, or is it out in NFS land as well ? 20:32:14 ASDF 2.27 is out -- considering the extent of changes, please test extensively before including in CCL 20:32:36 (that said, CCL is one of my main platforms) 20:33:09 on the other hand, your recent bootstrap binaries require a recent libc and won't work on my old ubuntu :-/ 20:33:33 Just recompile the lisp kernel locally. cd lisp-kernel/linuxx8664 && make clean && make 20:33:41 oh, thanks 20:34:09 what are the steps to compile the kernel and the lisp code separately? 20:34:13 We'll pass the request to test on to people that use ASDF ... I'll try to build the release candidate/release from old OS versions, but we can't win. 20:34:25 at Google, we have our own thing instead of make 20:34:25 I don't know ... I'd have to LOOK IN THE MANUAL. 20:35:20 http://ccl.clozure.com/ccl-documentation.html#building-ccl-from-source 20:35:50 thanks 20:35:56 Good cop/bad cop. 20:37:26 :) 20:40:11 Documentation is worth it just to be able to answer all your mail with 'RTFM'. — Alan Cox 20:42:02 it would be great if ASDF 2.27 or later could make it to your 1.9 release. 20:42:28 what is your expected schedule? 20:43:01 I should have the "release candidate" out later today; we usually ask people to smoke-test that for a week or two before declaring it "final". 20:43:43 So if the new ASDF is ready to go, I'll try to put it in today. 20:48:59 pnpuff [~dioxyrane@unaffiliated/pnpuff] has joined #ccl 20:50:32 Fare: So, just git pull, make, and check in build/asdf.lisp? 20:50:49 yup 20:51:23 hopefully in two weeks we have time to smooth out any kink. 20:51:39 I'll try to compile qres with the new ccl. 20:52:34 -!- pnpuff [~dioxyrane@unaffiliated/pnpuff] has left #ccl 20:53:21 rme: if you can "make t u l=ccl" just to check everything is alright, then that's great. 20:56:23 Fare: test-run-program.script failed, but I assume that's because on OS X, there's no /bin/grep (it's /usr/bin/grep there) 20:56:41 oops 20:57:05 I'll remove the absolute path 20:57:15 (does it work if you do?) 20:57:43 make t l=ccl t=test-run-program.script 20:58:21 btw, what's the structure of a pathname under windows on ccl? (just curious) 20:58:46 It passes when I replace "/bin/grep" with "grep". 20:59:09 if you have asdf users on windows, it would be good to have them run tests 20:59:14 excellent 20:59:59 a physical pathname often has a non-nil pathname-device (drive letter); most/all other things are the same. 21:01:55 what is \\host\mount\bar\baz ? 21:02:06 Unrecognized. 21:02:31 interesting. 21:03:30 It could be that people actually use that sort of thing. I've never heard of anyone doing so. 21:04:48 I don't use windows much if at all. But when I do, I often copy stuff across machines this way. 21:07:32 I admit that I only use Windows under mild protest; I usually ssh into windows boxes and try to deal with windows networking as little as possible. 21:15:36 nice 21:30:22 already a 2.27.1 -- yikes. 21:38:34 when I try to recompile CCL, I get this: Read error between positions 49454 and 50814 in /home/tunes/src/ccl/tools/defsystem.lisp. 21:39:36 apparently, mk-defsystem believes you're a CLtL1 with the package "LISP". 21:59:10 Odd. It gets recompiled every time one does (REBUILD-CCL) (at least the version that we distribute does.) 22:00:00 I'll be releasing 2.27.2 as 2.28 in a moment 22:00:19 embarrassing bug that doesn't prevent compilation, but will break some tests in many systems 22:00:42 gbyers: does mk-defsystem compilation work for you? 22:00:48 it breaks the ccl build for me 22:00:49 in trunk 22:01:02 Yes, and it works for everyone else, too, I assume. 22:01:15 interesting. 22:01:48 it's not like asdf is loaded, so nothing differs on my machine, that I know of 22:02:40 file position 49454 is in the middle of something that tries to ensure that :CLTL2 is on *FEATURES*. 22:03:04 apparently I had merge conflicts in tools 22:04:01 Something like that seems likely; I'm not sure how many things use mk:defsystem, but it gets compiled all the time and that's usually uneventful. 22:04:38 I'm surprised that anyone still uses mk-defsystem. 22:04:51 but lispers are conservative. 22:11:57 OK, sorry about that -- please use ASDF 2.28 instead of 2.27. 22:25:34 -!- rme [~rme@50.43.150.18] has quit [Quit: rme] 22:29:21 rme [~rme@50.43.150.18] has joined #ccl 23:47:32 -!- segv-_ [~mb@dslb-088-074-073-253.pools.arcor-ip.net] has quit [Ping timeout: 248 seconds]