2015-02-20T00:09:13Z nyef: stassats: That'd be an interesting new trend. Maybe we can do that for MIPS64 as well?
2015-02-20T00:09:26Z nyef: Well, for MIPS IV, at least.
2015-02-20T00:09:44Z stassats: files have a lot of (unsigned-byte 32) and other assumptions, those would have to be audited
2015-02-20T00:10:51Z stassats: even x86 and x86-64 could be somewhat unified
2015-02-20T00:14:29Z nikki93 joined #sbcl
2015-02-20T00:26:19Z attila_lendvai quit (Ping timeout: 256 seconds)
2015-02-20T00:31:59Z nyef: Can I have dibs on MIPS?
2015-02-20T00:32:11Z edgar-rft quit (Quit: execution ended into paranoid error)
2015-02-20T00:32:34Z stassats: i wasn't actually planing on ppc64, just researching
2015-02-20T00:32:37Z nyef: So close to getting my 64-bit MIPS system workingish, at least in single-CPU mode.
2015-02-20T00:44:12Z nikki93 quit (Remote host closed the connection)
2015-02-20T00:48:56Z nyef: Damnit, now you've actually got me thinking about what would be involved in a 64-bit mips backend.
2015-02-20T01:03:07Z oleo quit (Ping timeout: 255 seconds)
2015-02-20T03:14:33Z nikki93 joined #sbcl
2015-02-20T03:19:49Z oleo joined #sbcl
2015-02-20T03:30:58Z stassats` joined #sbcl
2015-02-20T03:34:30Z stassats` quit (Client Quit)
2015-02-20T03:38:50Z christoph_debian quit (Ping timeout: 252 seconds)
2015-02-20T03:52:13Z christoph_debian joined #sbcl
2015-02-20T04:06:48Z Bicyclidine quit (Ping timeout: 265 seconds)
2015-02-20T04:07:36Z ASau joined #sbcl
2015-02-20T04:13:14Z Bicyclidine joined #sbcl
2015-02-20T04:13:16Z nikki93 quit (Remote host closed the connection)
2015-02-20T04:20:40Z Bicyclidine quit (Ping timeout: 255 seconds)
2015-02-20T04:24:34Z nikki93 joined #sbcl
2015-02-20T04:24:36Z nikki93 quit (Remote host closed the connection)
2015-02-20T04:24:49Z nikki93 joined #sbcl
2015-02-20T04:27:34Z Bicyclidine joined #sbcl
2015-02-20T04:32:03Z Bicyclidine quit (Ping timeout: 246 seconds)
2015-02-20T04:54:37Z nyef quit (Quit: G'night all.)
2015-02-20T05:03:49Z Bicyclidine joined #sbcl
2015-02-20T05:04:13Z Bicyclidine quit (Client Quit)
2015-02-20T05:09:29Z psy_ quit (Ping timeout: 246 seconds)
2015-02-20T05:12:09Z gingerale joined #sbcl
2015-02-20T05:19:57Z nikki93 quit (Remote host closed the connection)
2015-02-20T05:20:15Z nikki93 joined #sbcl
2015-02-20T05:20:50Z psilord joined #sbcl
2015-02-20T05:26:02Z psy_ joined #sbcl
2015-02-20T05:30:17Z csziacobus quit (Quit: csziacobus)
2015-02-20T05:34:07Z milosn quit (Read error: Connection reset by peer)
2015-02-20T05:34:28Z nikki93 quit (Remote host closed the connection)
2015-02-20T05:39:30Z milosn joined #sbcl
2015-02-20T05:44:10Z nikki93 joined #sbcl
2015-02-20T05:54:22Z rszeno quit (Ping timeout: 240 seconds)
2015-02-20T05:54:58Z nikki93 quit (Remote host closed the connection)
2015-02-20T05:59:53Z pranavrc joined #sbcl
2015-02-20T06:10:25Z gabriel_laddel joined #sbcl
2015-02-20T06:10:56Z rszeno joined #sbcl
2015-02-20T06:14:55Z karswell joined #sbcl
2015-02-20T06:22:39Z gingerale quit (Ping timeout: 246 seconds)
2015-02-20T06:50:20Z rszeno quit (Quit: Leaving.)
2015-02-20T06:51:29Z gabriel_laddel quit (Ping timeout: 245 seconds)
2015-02-20T07:08:00Z angavrilov joined #sbcl
2015-02-20T07:14:11Z loke_ quit (Ping timeout: 250 seconds)
2015-02-20T07:15:18Z loke_ joined #sbcl
2015-02-20T07:18:20Z psy_ quit (Read error: No route to host)
2015-02-20T08:04:00Z ASau quit (Ping timeout: 264 seconds)
2015-02-20T08:07:50Z Shinmera joined #sbcl
2015-02-20T08:18:29Z edgar-rft joined #sbcl
2015-02-20T08:30:34Z fridim_ joined #sbcl
2015-02-20T08:55:02Z psy_ joined #sbcl
2015-02-20T08:55:34Z psy_ quit (Max SendQ exceeded)
2015-02-20T08:56:14Z psy_ joined #sbcl
2015-02-20T09:04:15Z Shinmera quit (Ping timeout: 250 seconds)
2015-02-20T09:06:53Z Shinmera joined #sbcl
2015-02-20T09:14:41Z fridim_ quit (Ping timeout: 265 seconds)
2015-02-20T09:23:45Z joshe quit (Ping timeout: 250 seconds)
2015-02-20T09:23:46Z fridim_ joined #sbcl
2015-02-20T09:36:12Z sdemarre joined #sbcl
2015-02-20T09:56:12Z sdemarre quit (Ping timeout: 264 seconds)
2015-02-20T09:57:48Z davazp joined #sbcl
2015-02-20T10:12:04Z fridim_ quit (Quit: Leaving)
2015-02-20T10:17:17Z stassats` joined #sbcl
2015-02-20T10:21:07Z stassats quit (Ping timeout: 255 seconds)
2015-02-20T10:28:59Z attila_lendvai joined #sbcl
2015-02-20T10:34:29Z selat joined #sbcl
2015-02-20T10:45:07Z attila_lendvai quit (Ping timeout: 256 seconds)
2015-02-20T10:54:38Z Shinmera quit (Quit: しつれいしなければならないんです。)
2015-02-20T11:32:28Z Shinmera joined #sbcl
2015-02-20T12:03:31Z rjmacready joined #sbcl
2015-02-20T12:05:19Z Shinmera quit (Quit: しつれいしなければならないんです。)
2015-02-20T12:12:01Z eudoxia joined #sbcl
2015-02-20T12:16:09Z Shinmera joined #sbcl
2015-02-20T12:22:03Z ASau joined #sbcl
2015-02-20T12:47:37Z davazp quit (Remote host closed the connection)
2015-02-20T12:48:03Z davazp joined #sbcl
2015-02-20T12:54:42Z pranavrc quit
2015-02-20T13:12:14Z psy_ quit (Quit: Leaving)
2015-02-20T13:28:06Z Shinmera quit (Quit: しつれいしなければならないんです。)
2015-02-20T13:54:51Z stassats` quit (Remote host closed the connection)
2015-02-20T13:56:21Z stassats joined #sbcl
2015-02-20T13:57:25Z Shinmera joined #sbcl
2015-02-20T14:18:40Z attila_lendvai joined #sbcl
2015-02-20T14:18:40Z attila_lendvai quit (Changing host)
2015-02-20T14:18:40Z attila_lendvai joined #sbcl
2015-02-20T14:19:39Z stassats quit (Remote host closed the connection)
2015-02-20T14:20:02Z stassats joined #sbcl
2015-02-20T14:20:23Z davazp quit (Ping timeout: 246 seconds)
2015-02-20T14:21:48Z stassats` joined #sbcl
2015-02-20T14:34:17Z nyef joined #sbcl
2015-02-20T14:34:30Z nyef: G'morning all.
2015-02-20T14:45:18Z loke_ quit (Remote host closed the connection)
2015-02-20T14:52:55Z psy_ joined #sbcl
2015-02-20T14:53:25Z psy_ quit (Max SendQ exceeded)
2015-02-20T14:53:55Z psy_ joined #sbcl
2015-02-20T15:02:07Z oleo: morning
2015-02-20T15:09:14Z oleo is now known as Guest1639
2015-02-20T15:10:08Z oleo__ joined #sbcl
2015-02-20T15:11:49Z Guest1639 quit (Ping timeout: 255 seconds)
2015-02-20T15:24:03Z oleo__ quit (Quit: Verlassend)
2015-02-20T15:27:33Z oleo__ joined #sbcl
2015-02-20T15:27:48Z oleo__ is now known as oleo
2015-02-20T15:41:41Z rjmacready quit (Quit: Page closed)
2015-02-20T15:57:46Z cojy_ quit (Remote host closed the connection)
2015-02-20T16:04:56Z scymtym joined #sbcl
2015-02-20T16:16:03Z cojy_ joined #sbcl
2015-02-20T16:40:14Z stassats: mega1: your named-readtables documentation doesn't look right in firefox
2015-02-20T16:40:22Z mega1: err
2015-02-20T16:41:13Z mega1: looks fabulous here
2015-02-20T16:41:28Z stassats: the contents table is too wide
2015-02-20T16:42:18Z mega1: "can't reproduce. closing" :-)
2015-02-20T16:42:37Z mega1: can you describe it in more detail?
2015-02-20T16:42:45Z stassats: too wide!
2015-02-20T16:42:57Z mega1: oh, too damn wide. got it
2015-02-20T16:43:47Z mega1: relevant part of the CSS: width: 50ex; max-width: 33%;
2015-02-20T16:43:47Z mega1:  
2015-02-20T16:44:03Z mega1: it's designed to resize like a champ
2015-02-20T16:44:38Z mega1: is it taking up more than 1/3 of the width of the window?
2015-02-20T16:44:48Z stassats: it encroaches on the content
2015-02-20T16:44:57Z mega1: aha
2015-02-20T16:45:01Z mega1: by a bit?
2015-02-20T16:45:23Z stassats: by enough, but not by a lost
2015-02-20T16:45:24Z stassats: lot
2015-02-20T16:46:28Z mega1: ah, it must be the minimum font size you have set up
2015-02-20T16:46:41Z stassats: right
2015-02-20T16:46:50Z mega1: does it fix itself if you zoom out?
2015-02-20T16:47:00Z stassats: no
2015-02-20T16:47:12Z stassats: only when i zoom in
2015-02-20T16:47:31Z mega1: right, in
2015-02-20T16:49:22Z stassats: i think the problem is in content having margin-left: 40ex;
2015-02-20T16:50:11Z stassats: instead of being relative to the content page
2015-02-20T16:50:19Z stassats: content table
2015-02-20T16:51:25Z mega1: test it now
2015-02-20T16:51:47Z stassats: that works
2015-02-20T16:52:30Z mega1: thanks for the heads up
2015-02-20T16:52:42Z mega1: and apologies for the innocent bystanders here
2015-02-20T16:53:32Z mega1: s/for/to/
2015-02-20T17:42:44Z specbot quit (Remote host closed the connection)
2015-02-20T17:43:02Z specbot joined #sbcl
2015-02-20T17:43:34Z selat quit (Ping timeout: 245 seconds)
2015-02-20T17:43:42Z minion joined #sbcl
2015-02-20T17:45:27Z selat joined #sbcl
2015-02-20T18:07:20Z Bicyclidine joined #sbcl
2015-02-20T18:25:18Z rszeno joined #sbcl
2015-02-20T18:30:01Z rszeno quit (Read error: Connection reset by peer)
2015-02-20T18:50:31Z mega1: Tracing FOO fails in this: (let ((*print-readably* t)) (foo (make-instance 'aaa)))
2015-02-20T18:50:40Z mega1: because AAA cannot be printed readably.
2015-02-20T18:50:54Z ehaliewicz joined #sbcl
2015-02-20T18:51:11Z mega1: I can't find anyhing in the standard related to this.
2015-02-20T18:51:20Z mega1: Is it something to be fixed?
2015-02-20T18:52:10Z stassats: it's unlikely that anybody would want to read trace output, so, yes
2015-02-20T18:54:20Z mega1: good, this has been driving me up the wall
2015-02-20T19:06:29Z Bicyclidine quit (Ping timeout: 245 seconds)
2015-02-20T19:08:27Z Bicyclidine joined #sbcl
2015-02-20T19:08:38Z gingerale joined #sbcl
2015-02-20T19:15:45Z mega1: on the other hand, binding *PRINT-READABLY* to NIL can change how stuff is printed
2015-02-20T19:16:20Z stassats`: trace has options to print things the way you want
2015-02-20T19:19:30Z mega1: how? I can't find anything relevant
2015-02-20T19:19:49Z mega1: :PRINT?
2015-02-20T19:21:14Z stassats`: yes
2015-02-20T19:21:50Z mega1: ok, this would be a user visible change in behaviour
2015-02-20T19:22:18Z stassats`: people rarely set *print-readably* globally
2015-02-20T19:22:36Z stassats`: it can't even print simple-base-strings
2015-02-20T19:23:23Z stassats`: so, it's a detectable but not noticeable change
2015-02-20T19:24:19Z ASau quit (Ping timeout: 252 seconds)
2015-02-20T19:27:21Z minion quit (Remote host closed the connection)
2015-02-20T19:27:21Z specbot quit (Remote host closed the connection)
2015-02-20T19:28:39Z minion joined #sbcl
2015-02-20T19:28:39Z specbot joined #sbcl
2015-02-20T19:30:03Z edgar-rft quit (Quit: you are not expected to understand this message)
2015-02-20T19:35:21Z Bicyclidine quit (Ping timeout: 250 seconds)
2015-02-20T19:37:05Z Bicyclidine joined #sbcl
2015-02-20T20:04:50Z Bicyclidine quit (Ping timeout: 246 seconds)
2015-02-20T20:06:43Z Bicyclidine joined #sbcl
2015-02-20T20:10:44Z Bicyclidine quit (Ping timeout: 246 seconds)
2015-02-20T20:23:25Z sdemarre joined #sbcl
2015-02-20T20:33:25Z sdemarre quit (Ping timeout: 250 seconds)
2015-02-20T20:37:02Z Bicyclidine joined #sbcl
2015-02-20T21:00:47Z csziacobus joined #sbcl
2015-02-20T21:01:48Z rszeno joined #sbcl
2015-02-20T21:24:25Z Bicyclidine quit (Ping timeout: 264 seconds)
2015-02-20T21:29:44Z eudoxia quit (Quit: Leaving)
2015-02-20T21:31:20Z Bicyclidine joined #sbcl
2015-02-20T21:36:04Z Bicyclidine quit (Ping timeout: 245 seconds)
2015-02-20T21:36:32Z edgar-rf_ joined #sbcl
2015-02-20T21:36:41Z edgar-rf_ quit (Read error: Connection reset by peer)
2015-02-20T21:37:50Z edgar-rft joined #sbcl
2015-02-20T21:42:37Z Bicyclidine joined #sbcl
2015-02-20T21:51:54Z stassats`: there's a deadlock on safepoints indeed
2015-02-20T21:56:07Z Bicyclidine quit (Ping timeout: 252 seconds)
2015-02-20T21:56:12Z edgar-rft quit (Quit: Ex-Chat)
2015-02-20T21:56:44Z rszeno quit (Quit: Leaving.)
2015-02-20T21:56:49Z edgar-rft joined #sbcl
2015-02-20T22:03:00Z Bicyclidine joined #sbcl
2015-02-20T22:04:22Z edgar-rft quit (Quit: Ex-Chat)
2015-02-20T22:04:58Z edgar-rft joined #sbcl
2015-02-20T22:29:11Z karswell quit (Remote host closed the connection)
2015-02-20T22:29:28Z karswell joined #sbcl
2015-02-20T22:34:36Z attila_lendvai quit (Ping timeout: 246 seconds)
2015-02-20T22:43:52Z nyef: stassats`: Would anything break if we changed EVERY to SOME in LVAR-GOOD-FOR-DX-P ?
2015-02-20T22:44:39Z stassats`: the new changes should handle that, shouldn't they?
2015-02-20T22:45:12Z Shinmera quit (Quit: しつれいしなければならないんです。)
2015-02-20T22:45:19Z angavrilov quit (Remote host closed the connection)
2015-02-20T22:46:01Z stassats`: and i'm always confused between uses and dests
2015-02-20T22:46:24Z nyef: Yeah, USEs are weird.
2015-02-20T22:47:14Z nyef: ... No-go. There's something else going on.
2015-02-20T22:49:26Z ehaliewicz quit (Remote host closed the connection)
2015-02-20T22:49:35Z stassats`: looks like the deadlock has something to do with printing initial thread trampoline frames
2015-02-20T22:51:11Z stassats`: deadlock notwithstanding i can see an optimization or two, niling the threads list so that it doesn't linger around
2015-02-20T22:52:04Z nyef: ... Looks like there's something odd about inline struct constructors?
2015-02-20T22:52:37Z stassats`: inlining the constructors is the only way to get structures dxed
2015-02-20T22:53:05Z stassats`: so, the problem is actually in printing the thread list
2015-02-20T22:56:03Z nyef: Definitely something weird here. 1044465 triggers with an inlined struct constructor, but not with CONS.
2015-02-20T22:56:59Z nyef: We'd still have the bug if struct constructors didn't do this, but it wouldn't have been reported as quickly.
2015-02-20T22:57:25Z Bicyclidine quit (Ping timeout: 264 seconds)
2015-02-20T22:58:11Z stassats`: stassats` on #sbcl (+cnt,lag:1) SBCL developer hangout, more dev than help -- but if #lisp can't solve it and it's SBCL specific, we'll try (nikodemu
2015-02-20T22:58:11Z stassats`: <nyef> Yeah, USEs are weird.
2015-02-20T22:58:11Z stassats`: <nyef> ... No-go. There's something else going on.                     01:47
2015-02-20T22:58:11Z stassats`: <stassats`> looks like the deadlock has something to do with printing
2015-02-20T22:58:14Z stassats`:             initial thread trampoline frames                           01:49
2015-02-20T22:58:16Z stassats`: <stassats`> deadlock notwithstanding i can see an optimization or two,
2015-02-20T22:58:19Z stassats`:             niling the threads list so that it doesn't linger around   01:51
2015-02-20T22:58:21Z stassats` quit (Remote host closed the connection)
2015-02-20T22:58:21Z stassats quit (Remote host closed the connection)
2015-02-20T22:58:53Z stassats joined #sbcl
2015-02-20T22:59:01Z stassats: stupid console emacs and chrome os pasting
2015-02-20T22:59:30Z gingerale quit (Ping timeout: 246 seconds)
2015-02-20T22:59:35Z stassats: http://paste.lisp.org/display/145847 happily deadlocks
2015-02-20T22:59:52Z stassats: (on sb-safepoint)
2015-02-20T23:01:10Z stassats: but why
2015-02-20T23:03:45Z stassats: and the list is actually the copy of the original list
2015-02-20T23:04:42Z stassats: when printing a thread it tries to access its status which is under a lock, maybe that's what's going on
2015-02-20T23:07:42Z stassats: it actually tries to join the thread
2015-02-20T23:10:29Z nyef: Preliminary idea: There's a CAST node on the allocated structure, and the DX decision is being made... Yeah, is being made on the LVAR going into the cast, not the one coming out of it.
2015-02-20T23:13:26Z nyef: Hrm. Actually, the CAST may be what trips it up.
2015-02-20T23:14:06Z Bicyclidine joined #sbcl
2015-02-20T23:14:09Z nyef: It hides the actual DX operation, like the outer CONS hides the inner one in the other examples.
2015-02-20T23:16:44Z nyef: Okay, this is cute: (defun test-2 (f a) (let ((x (when a (cons (cons a a) a)))) (declare (dynamic-extent x)) (funcall f x))) DX-allocates the inner CONS and heap-allocates the outer.
2015-02-20T23:16:51Z stassats: seems like the problem is accessing waiting-for
2015-02-20T23:17:27Z nyef: I smell dinner. It's about time for a break.
2015-02-20T23:26:50Z nyef: So, the problem with simply declaring SOME instead of EVERY is that it doesn't "hide" the LVAR, so a REF (or whatever) that USEs the LVAR doesn't know to set up the stack pointer, but it's still a DX LVAR, so the stack analysis thinks that the stack pointer has been set up.
2015-02-20T23:34:12Z stassats: quite weird http://paste.lisp.org/display/145847#1
2015-02-20T23:35:49Z stassats: so, nothing is being printed but somehow circularit detection messes things up
2015-02-20T23:36:50Z stassats: same thing with (set-pprint-dispatch 'thread (constantly nil))
2015-02-20T23:38:28Z stassats: but does it have to be threads
2015-02-20T23:47:53Z stassats: seems to be, and they have to be the ones running
2015-02-20T23:48:42Z ASau joined #sbcl
2015-02-20T23:56:52Z Bicyclidine quit (Ping timeout: 240 seconds)
2015-02-20T23:59:00Z Bicyclidine joined #sbcl