2015-01-15T00:06:32Z Bicyclidine quit (Ping timeout: 245 seconds)
2015-01-15T00:08:45Z Bicyclidine joined #sbcl
2015-01-15T00:13:12Z Bicyclidine quit (Ping timeout: 245 seconds)
2015-01-15T00:14:02Z Bicyclidine joined #sbcl
2015-01-15T00:20:43Z cmack` joined #sbcl
2015-01-15T00:22:33Z cmack quit (Ping timeout: 256 seconds)
2015-01-15T00:22:55Z Bicyclidine quit (Quit: no recitation today, i guess)
2015-01-15T00:27:05Z slyrus quit (Ping timeout: 256 seconds)
2015-01-15T00:27:11Z attila_lendvai quit (Quit: Leaving.)
2015-01-15T00:30:53Z stassats joined #sbcl
2015-01-15T00:34:01Z edgar-rft quit (Quit: no meaningful message available)
2015-01-15T00:41:07Z stassats quit (Ping timeout: 245 seconds)
2015-01-15T00:50:43Z slyrus joined #sbcl
2015-01-15T00:53:10Z Adlai quit (Quit: WeeChat 1.0.1)
2015-01-15T00:58:18Z slyrus quit (Ping timeout: 245 seconds)
2015-01-15T01:00:57Z gingerale joined #sbcl
2015-01-15T01:02:15Z Bike quit (Read error: Connection reset by peer)
2015-01-15T01:02:32Z Bike joined #sbcl
2015-01-15T01:12:50Z Adlai joined #sbcl
2015-01-15T01:12:51Z echo-area quit (Read error: Connection reset by peer)
2015-01-15T01:20:23Z gingerale quit (Read error: Connection reset by peer)
2015-01-15T01:38:52Z Adlai quit (Quit: WeeChat 1.0.1)
2015-01-15T02:22:02Z akkad: has sbcl always had this performance lead over all other CLs? quite impressive on large loads compared to our 64bit LW
2015-01-15T03:09:51Z Quadrescence quit (Quit: This computer has gone to sleep)
2015-01-15T03:23:46Z echo-area joined #sbcl
2015-01-15T03:26:53Z cmack` quit (Ping timeout: 252 seconds)
2015-01-15T03:32:17Z scymtym quit (Ping timeout: 272 seconds)
2015-01-15T03:38:47Z christoph_debian quit (Ping timeout: 264 seconds)
2015-01-15T03:40:38Z snuglas joined #sbcl
2015-01-15T03:44:27Z LiamH joined #sbcl
2015-01-15T03:46:27Z snuglas: i've a question about trying to improve deftransform find-position, if anyone's got a moment to help out
2015-01-15T03:51:52Z christoph_debian joined #sbcl
2015-01-15T03:56:04Z LiamH quit (Quit: Leaving.)
2015-01-15T04:18:44Z pkhuong: snuglas?
2015-01-15T04:21:17Z joshe joined #sbcl
2015-01-15T04:42:35Z slyrus joined #sbcl
2015-01-15T05:03:03Z slyrus quit (Ping timeout: 264 seconds)
2015-01-15T05:08:33Z Bike quit (Read error: Connection reset by peer)
2015-01-15T05:09:08Z Bike joined #sbcl
2015-01-15T05:58:49Z snuglas quit (Ping timeout: 246 seconds)
2015-01-15T06:24:39Z oleo is now known as Guest90965
2015-01-15T06:25:28Z oleo__ joined #sbcl
2015-01-15T06:25:44Z oleo__ quit (Read error: Connection reset by peer)
2015-01-15T06:27:17Z Guest90965 quit (Ping timeout: 252 seconds)
2015-01-15T06:39:23Z chu quit (Ping timeout: 264 seconds)
2015-01-15T06:40:41Z chu joined #sbcl
2015-01-15T07:15:35Z gabriel_laddel joined #sbcl
2015-01-15T07:15:38Z gabriel_laddel quit (Changing host)
2015-01-15T07:15:38Z gabriel_laddel joined #sbcl
2015-01-15T07:27:42Z gingerale joined #sbcl
2015-01-15T07:44:20Z Shinmera joined #sbcl
2015-01-15T08:04:31Z edgar-rft joined #sbcl
2015-01-15T08:15:33Z pacon joined #sbcl
2015-01-15T08:50:18Z fridim_ joined #sbcl
2015-01-15T08:59:33Z gabriel_laddel quit (Ping timeout: 245 seconds)
2015-01-15T09:14:38Z angavrilov joined #sbcl
2015-01-15T09:30:42Z gabriel_laddel joined #sbcl
2015-01-15T09:30:44Z gabriel_laddel quit (Changing host)
2015-01-15T09:30:44Z gabriel_laddel joined #sbcl
2015-01-15T10:08:21Z nikki93 joined #sbcl
2015-01-15T10:10:10Z nikki93 quit (Remote host closed the connection)
2015-01-15T10:36:23Z karswell` joined #sbcl
2015-01-15T10:36:44Z karswell quit (Remote host closed the connection)
2015-01-15T10:51:32Z echo-area quit (Remote host closed the connection)
2015-01-15T10:59:54Z cmack joined #sbcl
2015-01-15T11:03:08Z stassats joined #sbcl
2015-01-15T11:32:49Z pacon quit (Read error: Connection reset by peer)
2015-01-15T11:35:45Z attila_lendvai joined #sbcl
2015-01-15T11:56:24Z eudoxia joined #sbcl
2015-01-15T12:05:52Z joshe quit (Ping timeout: 265 seconds)
2015-01-15T12:35:14Z soggybread quit (Quit: leaving)
2015-01-15T12:37:09Z soggybread joined #sbcl
2015-01-15T12:54:31Z psy_ quit (Ping timeout: 256 seconds)
2015-01-15T12:59:11Z Cymew joined #sbcl
2015-01-15T13:01:36Z gabriel_laddel quit (Remote host closed the connection)
2015-01-15T13:18:17Z psy_ joined #sbcl
2015-01-15T13:21:30Z Hache_ joined #sbcl
2015-01-15T13:21:36Z gabriel_laddel joined #sbcl
2015-01-15T13:23:16Z gabriel_laddel quit (Remote host closed the connection)
2015-01-15T13:23:33Z gabriel_laddel joined #sbcl
2015-01-15T13:23:36Z gabriel_laddel quit (Remote host closed the connection)
2015-01-15T13:32:58Z LiamH joined #sbcl
2015-01-15T13:34:51Z Cymew quit (Ping timeout: 264 seconds)
2015-01-15T13:35:21Z Cymew joined #sbcl
2015-01-15T13:40:39Z snuglas joined #sbcl
2015-01-15T13:44:22Z snuglas:  i'm removing the circularity detection in find/position for unsafe inlined code. CLHS defines "safe" as safety 3 but I'm wondering if we should be more forgiving and says that non-zero = safe for this particular use
2015-01-15T13:45:31Z stassats: find isn't even have to detect circularity, does it?
2015-01-15T13:50:26Z snuglas: CLHS says "should be prepared to signal an error" which is defined to mean "should in safe code"
2015-01-15T13:50:32Z snuglas: see also https://bugs.launchpad.net/sbcl/+bug/452008
2015-01-15T13:50:49Z |3b|: clhs 1.4.2
2015-01-15T13:50:50Z specbot: Error Terminology: http://www.lispworks.com/reference/HyperSpec/Body/01_db.htm
2015-01-15T13:50:55Z |3b|: it is allowed to not return even in safe code
2015-01-15T13:51:17Z |3b| would prefer it to do better than that though, as a user
2015-01-15T13:51:56Z snuglas: right, but I'm not certain I want to go from "we were always maximally safe" to "we hold you to the strictures of CLHS in that you can't even possibly guarantee safety without (SAFETY 3)"
2015-01-15T13:52:06Z stassats: i do not expect there to be errors when i write (loop), why would there be when i try to run things on '#1=(loop . #1#)?
2015-01-15T13:52:41Z snuglas: personally I have no problem with it, as QPX uses either 3 or 0, so there's no middle ground... but I'm wondering if we should detect circularity in, for example SAFETY 2 in inlined code.
2015-01-15T13:53:22Z stassats: safety 2 sounds like a compromise
2015-01-15T13:53:29Z snuglas: stassats: i'm not sure what the example of syntactic circularity is intended to point out.
2015-01-15T13:53:39Z snuglas: "a compromise" that you agree with or disagree with ?
2015-01-15T13:54:00Z stassats: i agree with it
2015-01-15T13:54:11Z stassats: as i only run things with safety 1
2015-01-15T13:54:18Z snuglas: ok, thanks. i'll add a NEWS entry as such.
2015-01-15T14:09:22Z stassats: lp 1390544
2015-01-15T14:09:22Z specbot: https://bugs.launchpad.net/bugs/1390544
2015-01-15T14:09:53Z stassats: that happens because the node of #'list is to be deleted, so it's not let converted, but why is mv-combination optimized if it's would be deleted too?
2015-01-15T14:10:26Z stassats: i can just insert a condition to check that the #'list is not deleted, but maybe it can be avoided earlier
2015-01-15T14:11:04Z LiamH quit (Ping timeout: 255 seconds)
2015-01-15T14:14:07Z LiamH joined #sbcl
2015-01-15T14:14:55Z DeadTrickster joined #sbcl
2015-01-15T14:16:44Z fikusz quit (Quit: Leaving)
2015-01-15T14:16:54Z cmack quit (Ping timeout: 245 seconds)
2015-01-15T14:20:24Z fikusz joined #sbcl
2015-01-15T14:23:52Z snuglas quit (Ping timeout: 246 seconds)
2015-01-15T14:36:08Z LiamH quit (Quit: Leaving.)
2015-01-15T14:36:56Z snuglas joined #sbcl
2015-01-15T14:37:13Z snuglas: stas, did you mean to touch ir1 in your FORMAT change? and what about strings with escapes in them? that's not working
2015-01-15T14:38:06Z stassats: i meant
2015-01-15T14:38:32Z stassats: what about escapes?
2015-01-15T14:40:08Z Cymew quit (Ping timeout: 246 seconds)
2015-01-15T14:44:32Z stassats: so, ~s wouldn't work then
2015-01-15T14:44:50Z stassats: wasn't really a useful thing anyway
2015-01-15T14:50:51Z Cymew joined #sbcl
2015-01-15T14:55:31Z loke_ joined #sbcl
2015-01-15T14:56:46Z snuglas quit (Ping timeout: 246 seconds)
2015-01-15T15:00:35Z stassats: and ir1opt i didn't mean to commit
2015-01-15T15:00:42Z stassats: now i need to decide whether it's the right fix after all
2015-01-15T15:01:29Z oleo joined #sbcl
2015-01-15T15:01:41Z cmack joined #sbcl
2015-01-15T15:08:17Z LiamH joined #sbcl
2015-01-15T15:19:30Z LiamH quit (Quit: Leaving.)
2015-01-15T15:22:48Z LiamH joined #sbcl
2015-01-15T15:30:46Z stassats: turns out, mv-combination and the function it calls are in different blocks
2015-01-15T15:30:56Z stassats: and the function block is deleted before mv-combination
2015-01-15T15:42:19Z loke_ quit (Quit: Leaving)
2015-01-15T15:44:59Z slyrus joined #sbcl
2015-01-15T16:07:43Z Cymew quit (Ping timeout: 252 seconds)
2015-01-15T16:12:32Z LiamH quit (Quit: Leaving.)
2015-01-15T16:28:16Z attila_lendvai quit (Quit: Leaving.)
2015-01-15T16:28:52Z joshe joined #sbcl
2015-01-15T16:32:54Z psy_ quit (Remote host closed the connection)
2015-01-15T16:50:14Z eudoxia quit (Quit: Leaving)
2015-01-15T17:11:04Z attila_lendvai joined #sbcl
2015-01-15T17:11:04Z attila_lendvai quit (Changing host)
2015-01-15T17:11:04Z attila_lendvai joined #sbcl
2015-01-15T17:19:14Z cmack` joined #sbcl
2015-01-15T17:22:09Z cmack quit (Ping timeout: 252 seconds)
2015-01-15T17:44:56Z slyrus quit (Ping timeout: 246 seconds)
2015-01-15T18:03:51Z slyrus joined #sbcl
2015-01-15T18:11:57Z slyrus quit (Ping timeout: 245 seconds)
2015-01-15T18:12:52Z Hache_ quit (Ping timeout: 240 seconds)
2015-01-15T18:42:08Z oleo is now known as Guest73034
2015-01-15T18:43:00Z oleo__ joined #sbcl
2015-01-15T18:44:22Z Guest73034 quit (Ping timeout: 240 seconds)
2015-01-15T19:34:09Z fridim_ quit (Ping timeout: 252 seconds)
2015-01-15T19:37:47Z psy_ joined #sbcl
2015-01-15T19:44:38Z flip214: Hmmm, I successfully built sbcl on arm, but make-target-contrib.sh fails with "Failed to find the TRUENAME of SYS:CONTRIB;ASDF;UIOP.LISP.NEWEST: No such file or directory"
2015-01-15T19:44:51Z flip214: although all the files seem to be here - although in lowercase.
2015-01-15T19:45:15Z flip214: "bash run-sbcl.sh" works, and says 1.2.7.38-5f888f1
2015-01-15T19:46:49Z flip214: hmmm, "make fasl" in contrib/asdf fails, too
2015-01-15T19:49:43Z oleo__ quit (Quit: Verlassend)
2015-01-15T19:51:49Z flip214: If I change the compile-file argument to #p"uiop.lisp" it works. Is contrib/asdf/Makefile generated?
2015-01-15T19:52:21Z stassats: you shouldn't touch makefiles
2015-01-15T19:54:28Z flip214: the "wrong" path is from d7598560e4d105b4f9b37a9b04280048e25d4b92
2015-01-15T19:54:47Z stassats: it's not wrong
2015-01-15T19:55:38Z flip214: well, it's not working for me. what could be wrong that #p"asdf.lisp" works but #p"SYS:CONTRIB;ASDF;ASDF.LISP" doesn't?
2015-01-15T19:56:30Z flip214: because when "make" is run, the PWD is the ASDF directory; so the additional contrib/asdf path seems to be wrong.
2015-01-15T19:56:53Z stassats: it's only wrong for you, it has been right for everybody else
2015-01-15T19:57:13Z flip214: well, where is the path of SYS stored? in which file?
2015-01-15T19:57:57Z stassats: it's your build directory
2015-01-15T19:58:21Z stassats: if you compiled it one directory and then moved, then of course nothing will work
2015-01-15T19:58:39Z flip214: hmmm, that might be a reason... the NFS mount is now a different directory.
2015-01-15T19:59:27Z flip214: but why use the full path if the relative works more often?
2015-01-15T19:59:50Z stassats: because people don't move their directories in the middle of compilation
2015-01-15T20:00:56Z flip214: still, the relative path "just works", even if there's a remount between compilation and contrib build
2015-01-15T20:01:04Z oleo joined #sbcl
2015-01-15T20:02:42Z stassats: does it now?
2015-01-15T20:03:40Z flip214: I stopped the contrib build, remounted the NFS in the previous location, and restarted it... so now it should run without problems anyway.
2015-01-15T20:04:16Z flip214: but it's still a small change in two lines, and one more failure cause is gone...
2015-01-15T20:26:40Z Quadrescence joined #sbcl
2015-01-15T20:35:33Z Quadrescence: akkad, SBCL has consistently been more performant for me over LW.
2015-01-15T20:59:53Z flip214: has anyone build sb-bsd-sockets on a raspberry?
2015-01-15T21:03:24Z LiamH joined #sbcl
2015-01-15T21:07:55Z pacon joined #sbcl
2015-01-15T21:17:29Z eudoxia joined #sbcl
2015-01-15T21:21:52Z pacon quit (Read error: Connection reset by peer)
2015-01-15T21:23:09Z irsol quit (Ping timeout: 245 seconds)
2015-01-15T21:29:17Z LiamH quit (Ping timeout: 252 seconds)
2015-01-15T21:41:35Z Krystof: flip214: yes
2015-01-15T21:42:18Z Krystof: flip214: you may need to enable ipv6 (not enabled by default on pi)
2015-01-15T21:42:35Z flip214: ah, that's possible.
2015-01-15T21:42:50Z flip214: well, I just faked the tests passing and could then install and use it.
2015-01-15T21:43:23Z stassats: so, CASTs on IFs somehow interfere with DX stack analysis
2015-01-15T21:44:10Z stassats: it gets pushed into two different locations
2015-01-15T21:44:53Z slyrus joined #sbcl
2015-01-15T21:45:42Z stassats: http://paste.lisp.org/display/145253
2015-01-15T21:45:53Z angavrilov quit (Remote host closed the connection)
2015-01-15T21:46:57Z stassats: without casts, both legs of CIF push into the same location
2015-01-15T21:47:14Z stassats: but with casts, there's it's different locations
2015-01-15T21:47:32Z stassats: and i keep typing "cats" instead of "casts"
2015-01-15T21:52:42Z oleo: lol
2015-01-15T21:52:52Z LiamH joined #sbcl
2015-01-15T22:00:39Z scymtym joined #sbcl
2015-01-15T22:00:47Z attila_lendvai quit (Quit: Leaving.)
2015-01-15T22:02:50Z attila_lendvai joined #sbcl
2015-01-15T22:03:47Z eudoxia quit (Quit: Leaving)
2015-01-15T22:15:01Z stassats: so, there's such a thing as nested dxes, and apparently casts somehow interfere with that
2015-01-15T22:15:43Z pkhuong: yeah
2015-01-15T22:15:54Z pkhuong: we try to flow through casts, but it not always obvious
2015-01-15T22:17:30Z stassats: interestingly, (cons (vector x x) (vector x x)) shows one lvar from handle-nested-dynamic-extent-lvars, but (cons (cons  x x) nil) returns two
2015-01-15T22:17:45Z stassats: (if x (vector) (vector)) gives 3
2015-01-15T22:20:24Z stassats: the first lvar is always the original one, the rest are children
2015-01-15T22:21:22Z LiamH quit (Ping timeout: 240 seconds)
2015-01-15T22:21:40Z stassats: so, the actual problem is multiple dx lvars in cif
2015-01-15T22:22:06Z stassats: (defun foo (x) (let ((a (if x (list (list x)) (list (list x))))) (declare (dynamic-extent a)) (prin1 a)))
2015-01-15T22:22:10Z stassats: the most minimal test case
2015-01-15T22:22:26Z stassats: works with just (if x (list x) (list x))
2015-01-15T22:23:28Z pkhuong: DX and if could get messy.
2015-01-15T22:23:30Z stassats: that's a good conclusion to take a break for the day on it
2015-01-15T22:23:49Z pkhuong: it's not clear that there's always a clear location for the cleanup?
2015-01-15T22:24:50Z stassats: i think they all target the same cleanup?
2015-01-15T22:25:12Z stassats: i'm not sure yet what's going on in stack.lisp so that it gets upset
2015-01-15T22:25:39Z stassats: but there's some inconsistency between preceding and succeeding blocks wrt stack values
2015-01-15T22:27:32Z Quadrescence quit (Quit: This computer has gone to sleep)
2015-01-15T22:28:48Z stassats: walking IR1 in the inspector, hard to see the big picture
2015-01-15T22:29:45Z stassats: maybe some simple graphviz visualizer will help
2015-01-15T22:29:59Z gingerale quit (Ping timeout: 244 seconds)
2015-01-15T22:34:21Z Quadrescence joined #sbcl
2015-01-15T22:34:33Z irsol joined #sbcl
2015-01-15T22:41:39Z LiamH joined #sbcl
2015-01-15T22:44:15Z edgar-rf_ joined #sbcl
2015-01-15T22:44:23Z scymtym_ joined #sbcl
2015-01-15T22:44:45Z minion quit (Disconnected by services)
2015-01-15T22:44:47Z minion joined #sbcl
2015-01-15T22:45:36Z Shinmera- joined #sbcl
2015-01-15T22:46:47Z specbot quit (Disconnected by services)
2015-01-15T22:46:51Z specbot joined #sbcl
2015-01-15T22:47:00Z flip214_ joined #sbcl
2015-01-15T22:49:15Z carvite_ joined #sbcl
2015-01-15T22:50:26Z ivan4th` joined #sbcl
2015-01-15T22:50:59Z p_l_ joined #sbcl
2015-01-15T22:51:07Z mood_ joined #sbcl
2015-01-15T22:51:49Z scymtym quit (*.net *.split)
2015-01-15T22:51:50Z edgar-rft quit (*.net *.split)
2015-01-15T22:51:50Z Bike quit (*.net *.split)
2015-01-15T22:51:50Z Intensity quit (*.net *.split)
2015-01-15T22:51:50Z foom quit (*.net *.split)
2015-01-15T22:51:51Z mood quit (*.net *.split)
2015-01-15T22:51:51Z alchemis7 quit (*.net *.split)
2015-01-15T22:51:51Z Quadrescence quit (*.net *.split)
2015-01-15T22:51:52Z hlavaty quit (*.net *.split)
2015-01-15T22:51:52Z jasom quit (*.net *.split)
2015-01-15T22:51:53Z attila_lendvai quit (*.net *.split)
2015-01-15T22:51:53Z slyrus quit (*.net *.split)
2015-01-15T22:51:54Z drmeister quit (*.net *.split)
2015-01-15T22:51:54Z milosn quit (*.net *.split)
2015-01-15T22:51:55Z les quit (*.net *.split)
2015-01-15T22:51:55Z White_Flame quit (*.net *.split)
2015-01-15T22:51:55Z snafuchs quit (*.net *.split)
2015-01-15T22:51:56Z kanru quit (*.net *.split)
2015-01-15T22:51:56Z sobel quit (*.net *.split)
2015-01-15T22:51:56Z brucem quit (*.net *.split)
2015-01-15T22:52:21Z chu quit (*.net *.split)
2015-01-15T22:52:21Z ivan4th quit (*.net *.split)
2015-01-15T22:52:22Z psilord quit (*.net *.split)
2015-01-15T22:52:22Z pkhuong quit (*.net *.split)
2015-01-15T22:52:22Z flip214 quit (*.net *.split)
2015-01-15T22:52:40Z Shinmera quit (*.net *.split)
2015-01-15T22:52:40Z nicdev`` quit (*.net *.split)
2015-01-15T22:53:21Z jsnell_ joined #sbcl
2015-01-15T22:53:21Z Bike_ joined #sbcl
2015-01-15T22:53:21Z attila_lendvai joined #sbcl
2015-01-15T22:53:21Z slyrus joined #sbcl
2015-01-15T22:53:21Z drmeister joined #sbcl
2015-01-15T22:53:21Z milosn joined #sbcl
2015-01-15T22:53:21Z les joined #sbcl
2015-01-15T22:53:21Z White_Flame joined #sbcl
2015-01-15T22:53:21Z snafuchs joined #sbcl
2015-01-15T22:53:21Z kanru joined #sbcl
2015-01-15T22:53:21Z sobel joined #sbcl
2015-01-15T22:53:21Z brucem joined #sbcl
2015-01-15T22:54:03Z PuercoPop quit (*.net *.split)
2015-01-15T22:54:03Z Hydan quit (*.net *.split)
2015-01-15T22:54:03Z jdz quit (*.net *.split)
2015-01-15T22:54:03Z fitzsim quit (*.net *.split)
2015-01-15T22:54:03Z |3b| quit (*.net *.split)
2015-01-15T22:54:04Z pchrist quit (*.net *.split)
2015-01-15T22:54:05Z ams quit (*.net *.split)
2015-01-15T22:54:05Z carvite quit (*.net *.split)
2015-01-15T22:54:06Z antoszka quit (*.net *.split)
2015-01-15T22:54:06Z pkhuong joined #sbcl
2015-01-15T22:54:25Z antoszka_ joined #sbcl
2015-01-15T22:54:25Z Quadrescence joined #sbcl
2015-01-15T22:54:25Z hlavaty joined #sbcl
2015-01-15T22:54:25Z jasom joined #sbcl
2015-01-15T22:54:46Z pkhuong is now known as Guest63524