2017-09-01T00:14:05Z stassats joined #sbcl
2017-09-01T00:30:52Z whoman quit (Remote host closed the connection)
2017-09-01T00:31:19Z whoman joined #sbcl
2017-09-01T00:35:43Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-01T01:00:45Z stassats quit (Ping timeout: 240 seconds)
2017-09-01T01:29:21Z stassats joined #sbcl
2017-09-01T01:47:31Z karswell quit (Remote host closed the connection)
2017-09-01T01:52:03Z attila_lendvai quit (Quit: Leaving.)
2017-09-01T03:00:36Z whoman quit (Remote host closed the connection)
2017-09-01T03:02:41Z whoman joined #sbcl
2017-09-01T03:04:03Z shka_ joined #sbcl
2017-09-01T03:19:00Z oleo quit (Ping timeout: 246 seconds)
2017-09-01T03:19:24Z Bike quit (Ping timeout: 240 seconds)
2017-09-01T04:45:57Z oleo joined #sbcl
2017-09-01T05:04:49Z oleo quit (Quit: irc client terminated!)
2017-09-01T05:11:23Z stassats: dug into bad register allocation and i'm coming up to my earlier conclusion that the moves should be more transparent
2017-09-01T05:11:59Z stassats: since moves between scs even if the are just a simple copy get retained
2017-09-01T05:12:58Z stassats: while some moves have to perform coercion
2017-09-01T05:19:51Z stassats: or maybe why a vop that accepts a fixnum needs a move from t
2017-09-01T05:20:15Z stassats: i can see fixnum->t being problematic, but not t->fixnum
2017-09-01T05:37:48Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T05:38:13Z DeadTrickster joined #sbcl
2017-09-01T05:41:27Z stassats quit (Ping timeout: 248 seconds)
2017-09-01T05:47:51Z scymtym quit (Ping timeout: 248 seconds)
2017-09-01T06:04:03Z angavrilov joined #sbcl
2017-09-01T06:07:17Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-01T06:08:44Z BitPuffin|osx joined #sbcl
2017-09-01T06:15:33Z shka_ quit (Ping timeout: 248 seconds)
2017-09-01T07:17:52Z scymtym joined #sbcl
2017-09-01T08:05:42Z stassats joined #sbcl
2017-09-01T08:33:41Z stassats quit (Ping timeout: 248 seconds)
2017-09-01T08:35:25Z stassats joined #sbcl
2017-09-01T08:57:35Z stassats quit (Ping timeout: 240 seconds)
2017-09-01T08:57:42Z DeadTrickster_ joined #sbcl
2017-09-01T09:09:44Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T09:10:07Z DeadTrickster joined #sbcl
2017-09-01T09:20:11Z stassats joined #sbcl
2017-09-01T09:30:15Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T09:30:38Z DeadTrickster joined #sbcl
2017-09-01T10:07:39Z stassats quit (Ping timeout: 240 seconds)
2017-09-01T10:27:26Z leah2: hm, sbcl 1.3.21 ftbfs with glibc 2.26 http://sprunge.us/LibX should fix it
2017-09-01T10:28:37Z m00natic joined #sbcl
2017-09-01T10:32:06Z Bike joined #sbcl
2017-09-01T10:40:00Z Xof: ah, the bleeding edge.  Can you send to the mailing list so it doesn't get lost?  (Or file a launchpad bug?)
2017-09-01T11:10:40Z leah2: will send a fix
2017-09-01T11:11:16Z leah2: -bugs or -devel?
2017-09-01T11:11:50Z Xof: -devel please
2017-09-01T11:12:44Z leah2: does that work w/o subscription?
2017-09-01T11:13:05Z Xof: not sure
2017-09-01T11:19:44Z DeadTrickster_ quit (Ping timeout: 240 seconds)
2017-09-01T11:24:19Z DeadTrickster_ joined #sbcl
2017-09-01T12:00:02Z rpg joined #sbcl
2017-09-01T12:04:13Z stassats joined #sbcl
2017-09-01T12:04:14Z stassats quit (Changing host)
2017-09-01T12:04:14Z stassats joined #sbcl
2017-09-01T12:04:23Z rpg quit (Ping timeout: 248 seconds)
2017-09-01T12:08:09Z stassats quit (Ping timeout: 240 seconds)
2017-09-01T12:11:39Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T12:12:22Z DeadTrickster joined #sbcl
2017-09-01T12:23:15Z Bike quit (Ping timeout: 252 seconds)
2017-09-01T12:28:28Z attila_lendvai joined #sbcl
2017-09-01T12:57:00Z Bike joined #sbcl
2017-09-01T13:00:45Z oleo joined #sbcl
2017-09-01T14:04:26Z stassats joined #sbcl
2017-09-01T14:08:38Z stassats quit (Ping timeout: 240 seconds)
2017-09-01T14:33:31Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T14:33:56Z DeadTrickster joined #sbcl
2017-09-01T14:54:01Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T14:54:22Z DeadTrickster joined #sbcl
2017-09-01T15:09:29Z rpg joined #sbcl
2017-09-01T15:32:01Z stassats joined #sbcl
2017-09-01T15:51:59Z stassats quit (Ping timeout: 255 seconds)
2017-09-01T16:38:17Z m00natic quit (Read error: Connection reset by peer)
2017-09-01T17:14:24Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T17:14:50Z DeadTrickster joined #sbcl
2017-09-01T17:16:57Z BitPuffin|osx quit (Ping timeout: 252 seconds)
2017-09-01T17:25:03Z scymtym quit (Ping timeout: 255 seconds)
2017-09-01T17:37:11Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-01T17:39:45Z shka_ joined #sbcl
2017-09-01T17:44:30Z rpg joined #sbcl
2017-09-01T17:51:03Z Bike quit (Ping timeout: 252 seconds)
2017-09-01T17:57:41Z dougk_: stassats: mips build is currently broken due to "Reduce conditionals in define-vop expansion."
2017-09-01T17:57:48Z Bike joined #sbcl
2017-09-01T18:01:05Z Bicyclidine joined #sbcl
2017-09-01T18:02:03Z Bike quit (Ping timeout: 252 seconds)
2017-09-01T18:05:25Z DeadTrickster_ quit (Ping timeout: 248 seconds)
2017-09-01T18:05:58Z DeadTrickster quit (Ping timeout: 248 seconds)
2017-09-01T18:16:00Z DeadTrickster joined #sbcl
2017-09-01T18:16:13Z DeadTrickster_ joined #sbcl
2017-09-01T18:16:55Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-01T18:17:08Z DeadTrickster joined #sbcl
2017-09-01T18:20:43Z DeadTrickster_ quit (Ping timeout: 240 seconds)
2017-09-01T19:14:24Z scymtym joined #sbcl
2017-09-01T19:32:30Z shka_ quit (Ping timeout: 260 seconds)
2017-09-01T19:39:14Z oleo quit (Read error: Connection reset by peer)
2017-09-01T19:58:40Z oleo joined #sbcl
2017-09-01T20:36:23Z aeth_ joined #sbcl
2017-09-01T20:38:48Z aeth quit (Ping timeout: 252 seconds)
2017-09-01T20:39:01Z aeth_ is now known as aeth
2017-09-01T20:46:30Z angavrilov quit (Remote host closed the connection)
2017-09-01T21:17:55Z Bicyclidine quit (Ping timeout: 264 seconds)
2017-09-01T21:44:26Z attila_lendvai quit (Quit: Leaving.)
2017-09-01T22:29:31Z Bike joined #sbcl
2017-09-01T22:45:30Z whoman quit (Remote host closed the connection)
2017-09-01T22:45:52Z whoman joined #sbcl
2017-09-01T22:56:31Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-01T23:05:24Z rpg joined #sbcl
2017-09-01T23:35:33Z whoman quit (Read error: Connection reset by peer)
2017-09-01T23:36:03Z whoman joined #sbcl
2017-09-01T23:52:30Z Bicyclidine joined #sbcl
2017-09-01T23:56:16Z Bike quit (Ping timeout: 246 seconds)
2017-09-02T00:19:48Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-02T00:41:58Z zulu_inuoe_ joined #sbcl
2017-09-02T00:44:34Z zulu_inuoe quit (Ping timeout: 240 seconds)
2017-09-02T00:55:22Z stassats joined #sbcl
2017-09-02T01:39:50Z fiddlerwoaroof quit (Ping timeout: 248 seconds)
2017-09-02T01:46:42Z Bike joined #sbcl
2017-09-02T01:49:07Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-02T02:10:14Z rpg joined #sbcl
2017-09-02T02:23:10Z stassats quit (Ping timeout: 260 seconds)
2017-09-02T02:32:48Z fiddlerwoaroof joined #sbcl
2017-09-02T02:42:59Z stassats joined #sbcl
2017-09-02T02:49:10Z fiddlerwoaroof quit (Ping timeout: 248 seconds)
2017-09-02T02:56:58Z stassats: huh, something's happened to core size on ARM64, 1.3.16 has 50MB and 1.3.21.42 is 60MB
2017-09-02T02:57:12Z stassats: all the while x86-64 core size has been decreasing
2017-09-02T03:01:02Z Bike quit (Ping timeout: 252 seconds)
2017-09-02T03:04:01Z Bike joined #sbcl
2017-09-02T03:19:00Z slyrus: stassats: law of conversation of core sizes in effect?
2017-09-02T03:20:41Z stassats: slyrus: the cores talking to each other?
2017-09-02T03:30:10Z stassats: i'll bisect when i get home
2017-09-02T03:35:08Z stassats quit (Ping timeout: 240 seconds)
2017-09-02T03:54:12Z stassats joined #sbcl
2017-09-02T03:58:07Z Bike quit (Ping timeout: 246 seconds)
2017-09-02T04:13:54Z whoman: hmm. amd64 here. sbcl 1.3.20 core: 41mb, sbcl 1.3.21 core: 43mb
2017-09-02T04:24:53Z stassats quit (Ping timeout: 240 seconds)
2017-09-02T05:09:26Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-02T05:39:35Z shka_ joined #sbcl
2017-09-02T06:48:22Z leah2 quit (Ping timeout: 255 seconds)
2017-09-02T07:15:01Z angavrilov joined #sbcl
2017-09-02T07:50:08Z whoman quit (Remote host closed the connection)
2017-09-02T07:53:04Z whoman joined #sbcl
2017-09-02T09:19:01Z stassats joined #sbcl
2017-09-02T09:19:12Z stassats quit (Changing host)
2017-09-02T09:19:13Z stassats joined #sbcl
2017-09-02T09:19:25Z stassats: whoman: it's actually 39MB
2017-09-02T09:23:11Z whoman: i reported what it says for me
2017-09-02T09:24:02Z stassats: but it doesn't represent reality
2017-09-02T09:25:03Z whoman: the reality is that my core is 43mb
2017-09-02T09:25:32Z stassats: that's your reality
2017-09-02T09:26:02Z stassats: both 1.3.20 and 1.3.21 from https://sourceforge.net/projects/sbcl/files/sbcl/ are 39MB
2017-09-02T09:26:12Z stassats: (and the current HEAD is 38MB)
2017-09-02T09:26:13Z whoman: yeah. thats why i said "amd64 here" as well.
2017-09-02T09:27:12Z whoman: sorry dude.
2017-09-02T09:28:14Z whoman: i dont understand what you are actually talking about. i just wanted to be helpful because i was curious and checked the core of newly installed 1.3.21 sbcl. back to work
2017-09-02T09:28:34Z stassats: well, you're doing something to the cores
2017-09-02T09:30:28Z whoman: nothing. what is so surprising about the sizes?? i just configure, make install.
2017-09-02T09:31:03Z whoman: -rw-r--r-- 1 root staff 43188272 Sep  1 16:36 sbcl.core
2017-09-02T09:31:24Z stassats: there's no configure or make install in sbcl...
2017-09-02T09:31:53Z whoman: isnt it normal for core to be different size on different system, or no?
2017-09-02T09:32:32Z whoman: ./make.sh --fancy ?
2017-09-02T09:32:45Z stassats: yours not only are different but have increased in size
2017-09-02T09:33:10Z whoman: install.sh - i meant, basic plain configure method without any customizing.
2017-09-02T09:33:22Z whoman: hmm
2017-09-02T09:33:56Z scymtym quit (Ping timeout: 248 seconds)
2017-09-02T09:34:54Z whoman: https://pastebin.com/R5eiJFkP
2017-09-02T09:38:32Z stassats: that's irrelevant to the core
2017-09-02T09:39:16Z stassats: it's probable that you built 1.3.20 with different options
2017-09-02T09:39:44Z stassats: and 43188272 is actually 41MB
2017-09-02T09:39:48Z stassats: not 43
2017-09-02T09:40:41Z stassats: so, 1.3.20 you built without --fancy, and 1.3.21 with --fancy, hence the 2MB difference, and 41-2 is 39MB, which is the default size
2017-09-02T09:58:44Z DeadTrickster quit (Ping timeout: 260 seconds)
2017-09-02T10:02:54Z stassats: poking around the arm64 core, there's a 15MB region of just zeroes
2017-09-02T10:03:01Z stassats: while max zero length should be page-sized
2017-09-02T10:04:39Z stassats: on 1.3.16-arm64 the max zero chunk is 1MB
2017-09-02T10:04:59Z stassats: still too much, but less than the current 15MB
2017-09-02T10:05:40Z whoman: how small could we make it?
2017-09-02T10:06:35Z whoman: http://xach.livejournal.com/295584.html
2017-09-02T10:06:42Z stassats: i'm not in the mood for roundabout questions
2017-09-02T10:08:41Z whoman: i thought you were talking about the size of the core being too much, sorry just trying to help.
2017-09-02T10:10:41Z stassats: that's not helping
2017-09-02T10:12:52Z whoman: 9mb is small
2017-09-02T10:14:26Z stassats: are these zeros from a large-object region? or some other region? why would so many pages be preserved
2017-09-02T10:14:33Z stassats: dougk_: any ideas?
2017-09-02T10:15:46Z scymtym joined #sbcl
2017-09-02T10:19:11Z stylewarning: (WHICH-REGION-IS-THIS address)
2017-09-02T10:19:41Z stassats: stylewarning: ok, you're writing WHICH-REGION-IS-THIS
2017-09-02T10:21:14Z stassats: and this doesn't seem to be unique to ARM64, ppc has 3MB of zeroes
2017-09-02T10:23:44Z leah2 joined #sbcl
2017-09-02T10:27:40Z stassats: ok, it looks like it doesn't compact the heap enough leaving the zeros in the middle
2017-09-02T10:30:59Z stylewarning: Does that mean it doesn't move all/enough objects to the earliest address
2017-09-02T10:31:04Z stylewarning: es
2017-09-02T10:47:35Z stassats: gc_and_save doesn't seem to be doing a proper job
2017-09-02T11:21:02Z Bike joined #sbcl
2017-09-02T11:38:49Z oleo quit (Remote host closed the connection)
2017-09-02T11:39:26Z oleo joined #sbcl
2017-09-02T11:48:32Z DeadTrickster_ joined #sbcl
2017-09-02T11:59:57Z rpg joined #sbcl
2017-09-02T12:15:36Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-02T13:16:16Z stassats: ok, there seem to be some !move pages
2017-09-02T13:23:54Z rpg joined #sbcl
2017-09-02T13:24:56Z stassats: figured it out, yay
2017-09-02T13:28:36Z rpg quit (Ping timeout: 248 seconds)
2017-09-02T13:45:06Z dougk joined #sbcl
2017-09-02T13:46:13Z dougk: stassats: I think we should unwind the binding stacks and not collect them if 'conservative_stack' is 0.  Probably test for that at the comment "/* Scavenge the binding stacks. */"
2017-09-02T13:46:37Z stassats: dougk: it was *pinned-objects*
2017-09-02T13:47:06Z dougk: i know, but i also think the bindings stacks, which aren't restored into the new image anyway, should never enliven anything
2017-09-02T13:47:22Z dougk: and TLS isn't retained across save, so why does TLS enliven anything at that point?
2017-09-02T13:47:49Z stassats: true, but probably not going to have a big impact
2017-09-02T13:48:50Z stassats: shouldn't unwind, just ignore
2017-09-02T13:50:54Z stassats: ok, i guess unwinding is better since it doesn't have to conditionalize the normal path
2017-09-02T13:56:20Z dougk quit (Ping timeout: 248 seconds)
2017-09-02T13:57:30Z dougk joined #sbcl
2017-09-02T14:01:40Z stassats: dougk: the values pointed to by bindings around slad don't seem to be preserved
2017-09-02T14:05:12Z stassats: ok, they are, incorrectly tested
2017-09-02T14:14:14Z stassats: on sb-thread doing SetSymbolValue(FREE_TLS_INDEX, 0, 0); is enough, but for -sb-thread actual unwinding would have to be performed
2017-09-02T14:17:47Z stassats: hm, a special variable around slad appears to be unwound on -sb-thread
2017-09-02T14:23:23Z stassats: oh, save_to_filehandle does unbind_to_here
2017-09-02T14:32:13Z ryan_vw left #sbcl
2017-09-02T14:37:40Z ryan_vw joined #sbcl
2017-09-02T14:49:51Z stassats: for some reason, doing unbind_to_here before gcing crashes on arm64
2017-09-02T14:49:56Z stassats: oh well, for another time
2017-09-02T14:51:31Z dougk: oh dear, I think fe[nl]ix wanted to actually move that out of save_to_filehandle. now he'll have to figure out why
2017-09-02T14:51:49Z stassats: works on x86-64
2017-09-02T14:52:16Z stassats: and it crashes in a really weird place
2017-09-02T14:53:52Z stassats: maybe all the SymbolValue in gencgc? yet it works on x86-64..
2017-09-02T14:58:37Z stassats quit (Ping timeout: 240 seconds)
2017-09-02T15:03:52Z DeadTrickster_ quit (Read error: Connection reset by peer)
2017-09-02T15:04:21Z DeadTrickster_ joined #sbcl
2017-09-02T15:13:36Z stassats joined #sbcl
2017-09-02T15:43:20Z stassats quit (Ping timeout: 252 seconds)
2017-09-02T15:48:30Z slyrus quit (Remote host closed the connection)
2017-09-02T15:57:21Z scymtym quit (Ping timeout: 246 seconds)
2017-09-02T16:00:26Z slyrus joined #sbcl
2017-09-02T16:13:18Z whoman quit (Remote host closed the connection)
2017-09-02T16:13:37Z whoman joined #sbcl
2017-09-02T16:36:32Z dougk: stassats: if you send me the exact diff you made I'd like to take a look at the failure
2017-09-02T16:55:48Z fe[nl]ix: dougk: I wanted to move it to its own function so I can reuse it in save_elf_to_filehandle()
2017-09-02T16:56:04Z dougk: right, i know. and stas says that breaks arm64
2017-09-02T16:56:21Z dougk: i wasn't saying "figure out why he wants to do that", I'm saying "figure out why it breaks"
2017-09-02T16:57:04Z fe[nl]ix: ok
2017-09-02T17:03:22Z scymtym joined #sbcl
2017-09-02T18:31:07Z oleo quit (Read error: Connection reset by peer)
2017-09-02T18:38:06Z oleo joined #sbcl
2017-09-02T18:39:09Z oleo quit (Client Quit)
2017-09-02T18:56:41Z karswell joined #sbcl
2017-09-02T19:35:19Z oleo joined #sbcl
2017-09-02T19:43:41Z dougk quit (Ping timeout: 240 seconds)
2017-09-02T20:00:09Z fiddlerwoaroof joined #sbcl
2017-09-02T20:00:19Z Bicyclidine joined #sbcl
2017-09-02T20:02:21Z Bike quit (Ping timeout: 246 seconds)
2017-09-02T20:02:43Z shka_ quit (Ping timeout: 248 seconds)
2017-09-02T20:39:11Z stassats joined #sbcl
2017-09-02T20:43:17Z stassats quit (Ping timeout: 248 seconds)
2017-09-02T20:45:24Z attila_lendvai joined #sbcl
2017-09-02T20:45:24Z attila_lendvai quit (Changing host)
2017-09-02T20:45:24Z attila_lendvai joined #sbcl
2017-09-02T20:48:16Z dougk joined #sbcl
2017-09-02T20:54:26Z angavrilov quit (Remote host closed the connection)
2017-09-02T21:20:24Z dougk: stassats: sparc and mips both don't build.  Sparc is getting "**CURRENT-SEGMENT** is unbound." and mips is getting an unused variable warning in move-single-int-reg.  Your refactor of define-vop made it sensitive to that
2017-09-02T21:46:45Z dougk quit (Ping timeout: 248 seconds)
2017-09-02T22:10:37Z dougk joined #sbcl
2017-09-02T22:11:35Z oleo: why is the condition system so distributed.....
2017-09-02T22:17:46Z oleo: http://paste.lisp.org/+7LVZ
2017-09-02T22:29:13Z dougk quit (Ping timeout: 252 seconds)
2017-09-02T23:01:42Z slyrus quit (Ping timeout: 255 seconds)
2017-09-02T23:07:25Z attila_lendvai quit (Quit: Leaving.)
2017-09-03T00:13:38Z oleo quit (Read error: Connection reset by peer)
2017-09-03T00:20:54Z stassats joined #sbcl
2017-09-03T00:23:10Z oleo joined #sbcl
2017-09-03T00:32:36Z stassats quit (Ping timeout: 240 seconds)
2017-09-03T00:49:07Z stassats joined #sbcl
2017-09-03T00:49:54Z stassats: dougk_: in which functions?
2017-09-03T00:55:22Z stassats: the mips one, it should ignore FP
2017-09-03T01:13:50Z stassats: the sparc one is the scheduler
2017-09-03T01:27:59Z stassats quit (Ping timeout: 260 seconds)
2017-09-03T01:45:57Z Bike joined #sbcl
2017-09-03T01:46:00Z Bike quit (Remote host closed the connection)
2017-09-03T01:48:24Z Bicyclidine quit (Ping timeout: 260 seconds)
2017-09-03T01:50:03Z Bike joined #sbcl
2017-09-03T02:13:23Z nicdev joined #sbcl
2017-09-03T02:19:45Z dougk joined #sbcl
2017-09-03T02:23:27Z slyrus joined #sbcl
2017-09-03T02:24:04Z dougk quit (Ping timeout: 248 seconds)
2017-09-03T02:26:43Z dougk joined #sbcl
2017-09-03T02:37:16Z dougk quit (Ping timeout: 252 seconds)
2017-09-03T02:39:36Z dougk joined #sbcl
2017-09-03T02:40:36Z stassats joined #sbcl
2017-09-03T02:47:50Z DeadTrickster_ quit (Read error: Connection reset by peer)
2017-09-03T02:48:13Z DeadTrickster_ joined #sbcl
2017-09-03T02:50:06Z stassats quit (Ping timeout: 240 seconds)
2017-09-03T03:26:18Z dougk__ joined #sbcl
2017-09-03T03:29:06Z dougk quit (Ping timeout: 240 seconds)
2017-09-03T04:47:08Z stassats joined #sbcl
2017-09-03T04:49:05Z Bike quit (Ping timeout: 246 seconds)
2017-09-03T04:51:16Z stassats quit (Ping timeout: 248 seconds)
2017-09-03T05:41:06Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-03T06:29:17Z DeadTrickster_ quit (Read error: Connection reset by peer)
2017-09-03T06:29:51Z DeadTrickster_ joined #sbcl
2017-09-03T06:47:20Z stassats joined #sbcl
2017-09-03T06:52:06Z stassats quit (Ping timeout: 240 seconds)
2017-09-03T07:16:18Z shka_ joined #sbcl
2017-09-03T07:43:35Z angavrilov joined #sbcl
2017-09-03T08:48:43Z stassats joined #sbcl
2017-09-03T08:53:02Z stassats quit (Ping timeout: 246 seconds)
2017-09-03T09:21:06Z stassats joined #sbcl
2017-09-03T09:22:20Z oleo quit (Remote host closed the connection)
2017-09-03T09:25:14Z stassats quit (Ping timeout: 246 seconds)
2017-09-03T09:28:26Z stassats joined #sbcl
2017-09-03T09:31:30Z oleo joined #sbcl
2017-09-03T09:36:32Z stassats: The value #<invalid object #x10078973AF> is not of type SIMPLE-VECTOR, interesting
2017-09-03T09:37:56Z stassats: seems to have disappeared some time between then and HEAD
2017-09-03T09:50:44Z DeadTrickster_ quit (Read error: Connection reset by peer)
2017-09-03T09:52:03Z DeadTrickster joined #sbcl
2017-09-03T10:02:17Z stassats quit (Ping timeout: 255 seconds)
2017-09-03T11:13:07Z Posterdati quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2017-09-03T11:14:19Z Posterdati joined #sbcl
2017-09-03T11:58:40Z stassats joined #sbcl
2017-09-03T12:03:13Z stassats quit (Ping timeout: 252 seconds)
2017-09-03T12:08:46Z Bike joined #sbcl
2017-09-03T12:15:26Z Bicyclidine joined #sbcl
2017-09-03T12:18:27Z Bike quit (Ping timeout: 246 seconds)
2017-09-03T12:35:36Z Posterdati quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2017-09-03T12:36:16Z Posterdati joined #sbcl
2017-09-03T13:16:54Z oleo quit (Ping timeout: 246 seconds)
2017-09-03T13:18:36Z oleo joined #sbcl
2017-09-03T13:20:29Z Bike joined #sbcl
2017-09-03T13:21:27Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-03T14:51:00Z dougk joined #sbcl
2017-09-03T14:57:27Z dougk quit (Read error: Connection reset by peer)
2017-09-03T14:57:46Z dougk joined #sbcl
2017-09-03T15:10:28Z dougk quit (Ping timeout: 248 seconds)
2017-09-03T15:48:53Z slyrus quit (Quit: Client Quit)
2017-09-03T15:58:17Z whoman quit (Remote host closed the connection)
2017-09-03T15:58:41Z whoman joined #sbcl
2017-09-03T15:58:58Z DeadTrickster_ joined #sbcl
2017-09-03T16:03:36Z DeadTrickster_ quit (Ping timeout: 240 seconds)
2017-09-03T16:43:51Z karswell_ joined #sbcl
2017-09-03T16:44:40Z karswell quit (Read error: Connection reset by peer)
2017-09-03T17:37:45Z sindan joined #sbcl
2017-09-03T17:42:33Z sindan: I don't see a slime channel, so maybe someone can help here? It's slightly annoying that the lines in the REPL buffer in slime get soft-wrapped. My buffer is maybe 180 chars long but lines get broken up at about 80 columns. Is there a quick setting to let lines extend to the right without being wrapped?
2017-09-03T17:43:30Z Bike: #lisp is fine for asking slime questions
2017-09-03T17:43:51Z sindan: thx!
2017-09-03T18:20:22Z karswell_ quit (Remote host closed the connection)
2017-09-03T18:21:05Z karswell_ joined #sbcl
2017-09-03T18:54:30Z attila_lendvai joined #sbcl
2017-09-03T18:54:30Z attila_lendvai quit (Changing host)
2017-09-03T18:54:30Z attila_lendvai joined #sbcl
2017-09-03T19:08:20Z shka_ quit (Ping timeout: 240 seconds)
2017-09-03T19:24:34Z whoman: sindan, try #emacs, i think it is an Emacs setting for changing word-wrap style.
2017-09-03T19:31:45Z sindan: whoman: it's already solved: *print-right-margin*
2017-09-03T19:31:54Z dougk joined #sbcl
2017-09-03T19:33:23Z sindan: it's not an emacs setting, it's in the clhs
2017-09-03T19:36:11Z whoman: i dont believe you
2017-09-03T19:44:08Z karswell_ quit (Ping timeout: 240 seconds)
2017-09-03T19:44:19Z sindan: you don't have to; google it
2017-09-03T19:49:25Z whoman: not that i seriously dont believe you =) but that i think especially that its an emacs setting; im stubborn
2017-09-03T19:53:44Z sindan: emacs has a setting for everything, so you are in a way stating the obvious :)
2017-09-03T19:55:33Z whoman: true =)
2017-09-03T19:57:00Z whoman: i cant find any info about it. have you figured it out yet?
2017-09-03T20:43:37Z rpg joined #sbcl
2017-09-03T21:12:45Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-03T21:29:08Z dougk quit (Ping timeout: 240 seconds)
2017-09-03T21:30:25Z dougk joined #sbcl
2017-09-03T21:46:30Z dougk__ joined #sbcl
2017-09-03T21:49:08Z dougk quit (Ping timeout: 240 seconds)
2017-09-03T22:39:03Z Bicyclidine joined #sbcl
2017-09-03T22:41:47Z Bike quit (Ping timeout: 246 seconds)
2017-09-03T23:13:06Z libre-man quit (Ping timeout: 252 seconds)
2017-09-03T23:27:44Z libre-man joined #sbcl
2017-09-03T23:55:31Z attila_lendvai quit (Quit: Leaving.)
2017-09-03T23:59:38Z dougk__ quit (Read error: Connection reset by peer)
2017-09-03T23:59:53Z dougk joined #sbcl
2017-09-04T00:33:03Z dougk__ joined #sbcl
2017-09-04T00:35:46Z dougk quit (Ping timeout: 248 seconds)
2017-09-04T00:48:14Z sindan quit (Quit: Leaving)
2017-09-04T00:49:09Z intrigue left #sbcl
2017-09-04T01:09:39Z Bike joined #sbcl
2017-09-04T01:12:19Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-04T01:16:08Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-04T01:30:15Z dougk joined #sbcl
2017-09-04T01:59:45Z Bicyclidine joined #sbcl
2017-09-04T02:00:51Z Bike quit (Ping timeout: 240 seconds)
2017-09-04T02:09:23Z dougk quit (Read error: Connection reset by peer)
2017-09-04T02:09:38Z dougk joined #sbcl
2017-09-04T02:25:49Z Bike joined #sbcl
2017-09-04T02:28:31Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-04T03:29:36Z dougk__ joined #sbcl
2017-09-04T03:32:07Z dougk quit (Ping timeout: 255 seconds)
2017-09-04T04:19:37Z shka_ joined #sbcl
2017-09-04T05:32:26Z oleo quit (Quit: irc client terminated!)
2017-09-04T05:43:13Z Bike quit (Ping timeout: 246 seconds)
2017-09-04T05:46:37Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-04T05:48:28Z scymtym quit (Ping timeout: 255 seconds)
2017-09-04T05:52:33Z dougk joined #sbcl
2017-09-04T06:10:42Z galdor quit (Read error: Connection reset by peer)
2017-09-04T06:11:01Z galdor joined #sbcl
2017-09-04T06:46:50Z dougk quit (Ping timeout: 252 seconds)
2017-09-04T07:11:30Z scymtym joined #sbcl
2017-09-04T08:12:27Z aeth_ joined #sbcl
2017-09-04T08:14:07Z aeth quit (Ping timeout: 240 seconds)
2017-09-04T09:04:40Z BitPuffin|osx joined #sbcl
2017-09-04T09:18:14Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-04T09:25:05Z scymtym quit (Remote host closed the connection)
2017-09-04T09:25:17Z scymtym_ joined #sbcl
2017-09-04T09:29:16Z BitPuffin|osx joined #sbcl
2017-09-04T09:43:32Z attila_lendvai joined #sbcl
2017-09-04T10:17:40Z m00natic joined #sbcl
2017-09-04T10:52:47Z attila_lendvai quit (Quit: Leaving.)
2017-09-04T11:17:43Z aeth joined #sbcl
2017-09-04T11:17:48Z aeth_ quit (Ping timeout: 246 seconds)
2017-09-04T12:22:57Z Bike joined #sbcl
2017-09-04T13:05:14Z register_ joined #sbcl
2017-09-04T13:05:24Z register_: Hi
2017-09-04T13:05:43Z register_: how does sbcl manages to support debugging live stack frames?
2017-09-04T13:06:04Z register_: I can change a single form and resume debugging when the optimize level is 3
2017-09-04T13:06:08Z register_: how is that done?
2017-09-04T13:20:05Z rumbler31 joined #sbcl
2017-09-04T13:29:27Z register_ quit (Quit: Page closed)
2017-09-04T13:41:39Z rumbler31 quit (Remote host closed the connection)
2017-09-04T14:01:51Z dougk joined #sbcl
2017-09-04T14:04:06Z whoman quit (Ping timeout: 240 seconds)
2017-09-04T14:12:10Z dougk quit (Read error: Connection reset by peer)
2017-09-04T14:12:24Z dougk joined #sbcl
2017-09-04T14:17:55Z oleo joined #sbcl
2017-09-04T14:24:18Z rpg joined #sbcl
2017-09-04T14:25:00Z rpg quit (Read error: Connection reset by peer)
2017-09-04T14:26:11Z rpg joined #sbcl
2017-09-04T14:27:02Z rpg quit (Read error: Connection reset by peer)
2017-09-04T14:27:44Z rpg joined #sbcl
2017-09-04T14:28:49Z rpg quit (Read error: Connection reset by peer)
2017-09-04T14:30:41Z rpg joined #sbcl
2017-09-04T14:31:40Z rpg quit (Read error: Connection reset by peer)
2017-09-04T15:23:02Z foom joined #sbcl
2017-09-04T15:24:54Z foom2 quit (Ping timeout: 246 seconds)
2017-09-04T15:25:37Z whoman joined #sbcl
2017-09-04T15:50:06Z irsol quit (Ping timeout: 240 seconds)
2017-09-04T15:51:23Z irsol joined #sbcl
2017-09-04T15:51:53Z rumbler31 joined #sbcl
2017-09-04T16:00:23Z rpg joined #sbcl
2017-09-04T16:02:01Z sjl quit (Ping timeout: 260 seconds)
2017-09-04T16:18:47Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-04T16:19:17Z rpg joined #sbcl
2017-09-04T16:22:53Z igajsin joined #sbcl
2017-09-04T16:36:08Z rumbler31 quit (Remote host closed the connection)
2017-09-04T16:36:35Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-04T16:46:20Z rumbler31 joined #sbcl
2017-09-04T16:59:10Z rumbler31 quit (Remote host closed the connection)
2017-09-04T17:02:42Z m00natic quit (Remote host closed the connection)
2017-09-04T17:59:41Z rumbler31 joined #sbcl
2017-09-04T18:03:56Z rumbler31 quit (Ping timeout: 260 seconds)
2017-09-04T18:28:30Z abbe_ joined #sbcl
2017-09-04T18:28:31Z galdor1 joined #sbcl
2017-09-04T18:30:09Z jyc_ joined #sbcl
2017-09-04T18:32:33Z pkhuong_ joined #sbcl
2017-09-04T18:32:59Z abbe quit (Disconnected by services)
2017-09-04T18:33:04Z abbe_ is now known as abbe
2017-09-04T18:35:41Z malm quit (Killed (adams.freenode.net (Nickname regained by services)))
2017-09-04T18:35:44Z malm joined #sbcl
2017-09-04T18:37:45Z sigjuice_ joined #sbcl
2017-09-04T18:37:45Z galdor quit (*.net *.split)
2017-09-04T18:37:45Z jack_rabbit quit (*.net *.split)
2017-09-04T18:37:45Z jdz quit (*.net *.split)
2017-09-04T18:37:45Z pkhuong quit (*.net *.split)
2017-09-04T18:37:45Z sigjuice quit (*.net *.split)
2017-09-04T18:37:45Z jyc quit (*.net *.split)
2017-09-04T18:37:45Z gingerale quit (*.net *.split)
2017-09-04T18:37:45Z jyc_ is now known as jyc
2017-09-04T18:41:34Z shka_ quit (Write error: Broken pipe)
2017-09-04T18:42:15Z shka_ joined #sbcl
2017-09-04T18:43:30Z gingerale joined #sbcl
2017-09-04T18:43:43Z jack_rabbit joined #sbcl
2017-09-04T19:00:29Z rpg joined #sbcl
2017-09-04T19:38:58Z scymtym_ quit (Ping timeout: 246 seconds)
2017-09-04T19:45:51Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-04T19:51:31Z rpg joined #sbcl
2017-09-04T19:59:14Z jdz joined #sbcl
2017-09-04T20:04:53Z shka_ quit (Ping timeout: 252 seconds)
2017-09-04T20:08:52Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-04T20:28:26Z scymtym joined #sbcl
2017-09-04T20:48:00Z dougk quit (Ping timeout: 255 seconds)
2017-09-04T20:49:31Z dougk joined #sbcl
2017-09-04T20:51:41Z attila_lendvai joined #sbcl
2017-09-04T20:51:41Z attila_lendvai quit (Changing host)
2017-09-04T20:51:41Z attila_lendvai joined #sbcl
2017-09-04T20:55:12Z oleo quit (Ping timeout: 255 seconds)
2017-09-04T21:00:36Z igajsin quit (Ping timeout: 255 seconds)
2017-09-04T21:04:52Z rumbler31 joined #sbcl
2017-09-04T21:13:54Z rumbler31 quit (Remote host closed the connection)
2017-09-04T21:18:37Z igajsin joined #sbcl
2017-09-04T21:21:04Z dougk quit (Ping timeout: 240 seconds)
2017-09-04T21:22:43Z angavrilov quit (Remote host closed the connection)
2017-09-04T21:25:39Z igajsin quit (Ping timeout: 264 seconds)
2017-09-04T21:40:14Z dougk joined #sbcl
2017-09-04T21:45:39Z prxq joined #sbcl
2017-09-04T22:27:36Z dougk quit (Ping timeout: 260 seconds)
2017-09-04T22:59:53Z dougk joined #sbcl
2017-09-04T23:00:30Z scymtym quit (Ping timeout: 240 seconds)
2017-09-04T23:16:15Z cromachina joined #sbcl
2017-09-04T23:27:11Z dougk__ joined #sbcl
2017-09-04T23:30:35Z dougk quit (Ping timeout: 240 seconds)
2017-09-04T23:32:50Z dougk__ quit (Ping timeout: 248 seconds)
2017-09-04T23:33:04Z BitPuffin|osx quit (Ping timeout: 255 seconds)
2017-09-04T23:53:02Z dougk joined #sbcl
2017-09-05T02:33:09Z attila_lendvai quit (Quit: Leaving.)
2017-09-05T02:57:48Z prxq_ joined #sbcl
2017-09-05T03:02:05Z prxq quit (Ping timeout: 240 seconds)
2017-09-05T03:57:20Z dougk quit (Ping timeout: 248 seconds)
2017-09-05T04:01:26Z shka_ joined #sbcl
2017-09-05T04:07:07Z Bike quit (Ping timeout: 255 seconds)
2017-09-05T04:17:47Z dougk joined #sbcl
2017-09-05T04:27:19Z dougk quit (Read error: Connection reset by peer)
2017-09-05T04:27:33Z dougk joined #sbcl
2017-09-05T04:45:51Z oleo joined #sbcl
2017-09-05T05:06:13Z oleo quit (Quit: irc client terminated!)
2017-09-05T05:21:27Z igajsin joined #sbcl
2017-09-05T05:34:40Z igajsin quit (Ping timeout: 252 seconds)
2017-09-05T05:38:34Z igajsin joined #sbcl
2017-09-05T05:43:38Z igajsin quit (Ping timeout: 264 seconds)
2017-09-05T07:19:42Z scymtym joined #sbcl
2017-09-05T07:29:37Z shka_ quit (Ping timeout: 252 seconds)
2017-09-05T07:44:30Z prxq_ quit (Remote host closed the connection)
2017-09-05T07:45:03Z dougk quit (Ping timeout: 240 seconds)
2017-09-05T09:13:34Z tonton quit (Ping timeout: 255 seconds)
2017-09-05T09:15:12Z tonton joined #sbcl
2017-09-05T09:42:38Z BitPuffin|osx joined #sbcl
2017-09-05T10:37:41Z attila_lendvai joined #sbcl
2017-09-05T10:37:41Z attila_lendvai quit (Changing host)
2017-09-05T10:37:41Z attila_lendvai joined #sbcl
2017-09-05T10:43:24Z attila_lendvai quit (Quit: Leaving.)
2017-09-05T10:55:41Z Bike joined #sbcl
2017-09-05T10:57:28Z angavrilov joined #sbcl
2017-09-05T11:19:54Z Bicyclidine joined #sbcl
2017-09-05T11:21:53Z Bike quit (Ping timeout: 246 seconds)
2017-09-05T12:11:34Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-05T12:28:19Z m00natic joined #sbcl
2017-09-05T12:45:16Z BitPuffin|osx quit (Ping timeout: 240 seconds)
2017-09-05T12:46:55Z Bike joined #sbcl
2017-09-05T13:53:54Z oleo joined #sbcl
2017-09-05T14:33:39Z scymtym quit (Ping timeout: 246 seconds)
2017-09-05T15:09:21Z milanj joined #sbcl
2017-09-05T16:19:10Z dougk joined #sbcl
2017-09-05T16:46:31Z stassats joined #sbcl
2017-09-05T16:49:44Z m00natic quit (Remote host closed the connection)
2017-09-05T17:26:17Z shka_ joined #sbcl
2017-09-05T17:35:50Z attila_lendvai joined #sbcl
2017-09-05T17:39:45Z stassats quit (Ping timeout: 248 seconds)
2017-09-05T17:43:11Z stassats joined #sbcl
2017-09-05T17:59:50Z dougk quit (Ping timeout: 264 seconds)
2017-09-05T18:05:50Z dougk joined #sbcl
2017-09-05T18:33:24Z dougk quit (Ping timeout: 246 seconds)
2017-09-05T18:40:43Z dougk joined #sbcl
2017-09-05T18:44:47Z foom quit (Quit: Leaving)
2017-09-05T19:02:08Z dougk quit (Ping timeout: 246 seconds)
2017-09-05T19:03:27Z scymtym joined #sbcl
2017-09-05T19:04:58Z dougk joined #sbcl
2017-09-05T19:14:21Z dougk quit (Ping timeout: 246 seconds)
2017-09-05T19:16:03Z shka_ quit (Ping timeout: 240 seconds)
2017-09-05T19:24:55Z dougk joined #sbcl
2017-09-05T19:32:18Z dougk quit (Ping timeout: 255 seconds)
2017-09-05T19:34:10Z dougk joined #sbcl
2017-09-05T19:54:00Z stassats: new: (handler-bind ((unbound-variable (lambda (c) (use-value 123 c)))) (symbol-value (gensym))) => 123
2017-09-05T20:01:51Z irsol quit (Remote host closed the connection)
2017-09-05T20:06:03Z foom joined #sbcl
2017-09-05T20:20:30Z Shinmera: Nifty
2017-09-05T20:22:55Z stassats: i was planning on providing restarts for type errors but the values may get copied into multiple locations
2017-09-05T20:23:42Z jackdaniel: stassats: is previous behavior non-conformant? or it's not specified by the standard?
2017-09-05T20:24:02Z stassats: which behavior?
2017-09-05T20:24:57Z jackdaniel: (handler-bind ((unbound-variable (lambda (c) (use-value 123 c)))) (symbol-value (gensym)))
2017-09-05T20:25:01Z jackdaniel: -> unbound variable
2017-09-05T20:25:10Z stassats: there was no previous behavior
2017-09-05T20:25:50Z jackdaniel: I've put it in sbcl 1.3.20 repl, that's what I call "previous", maybe mistakenly
2017-09-05T20:26:03Z stassats: a new feature can have no previous behavior
2017-09-05T20:27:06Z jackdaniel: OK, then let me rephrase my question: "The variable #:G445 is unbound." on this expression is conformant?
2017-09-05T20:27:39Z stassats: you know where to look for the answer
2017-09-05T20:29:50Z jsnell: jackdaniel: if you think there's something wrong with that, you might be misunderstanding how handler-bind works in general
2017-09-05T20:30:43Z irsol joined #sbcl
2017-09-05T20:30:53Z stassats: i think i can also make INVALID-ARRAY-INDEX-ERROR restartable
2017-09-05T20:31:44Z jackdaniel: jsnell: I don't think anything about it, I just saw difference between behavior of a few implementations and wonder if it's unspecified behavior, or if I should add a test to ansi-test, that's all
2017-09-05T20:36:06Z jsnell: there aren't a ton of mandatory restarts in the clhs. the only one I can think of is the continue restart for cerror
2017-09-05T20:37:56Z stassats: though i don't think INVALID-ARRAY-INDEX-ERROR would be all that useful, just showing off
2017-09-05T21:19:00Z Bike quit (Ping timeout: 252 seconds)
2017-09-05T21:23:27Z dougk quit (Ping timeout: 255 seconds)
2017-09-05T21:30:33Z leah2 quit (Ping timeout: 240 seconds)
2017-09-05T21:36:35Z Xof: showing off is good
2017-09-05T21:36:48Z Xof: if we're going to delete warnings one release, let's have some actual features the next!!
2017-09-05T21:37:11Z stassats: but i did add some new warnings in that release
2017-09-05T21:37:24Z Xof: jsnell: would you like to do a PhD in software archaeology?
2017-09-05T21:37:57Z Xof: low low prices at bone fide UK universities
2017-09-05T21:37:58Z stassats: too bad nothing in quicklisp seems to be affected by them
2017-09-05T21:38:15Z Xof: register now to preserve your Home/EU fees
2017-09-05T21:41:06Z dougk joined #sbcl
2017-09-05T21:42:12Z jsnell: I'm just a dilettante of software archeology
2017-09-05T21:44:46Z Xof: you could get a recognized qualification, to allow you to claim professionalism
2017-09-05T21:45:13Z Xof: you must surely be bored by now of all this working for The Man
2017-09-05T21:45:22Z Xof: do something useless for a change!
2017-09-05T21:45:39Z angavrilov quit (Remote host closed the connection)
2017-09-05T21:46:23Z jsnell: after 6 years you finally admit our software wasn't actually useless :-P
2017-09-05T21:51:43Z Xof: look, I give out placebo degrees
2017-09-05T21:51:57Z Xof: I have to be positive about the "smart wire" we sold
2017-09-05T21:52:48Z Xof: (7 years!)
2017-09-05T21:55:08Z Xof: *weeps silently*
2017-09-05T21:59:27Z dougk quit (Ping timeout: 248 seconds)
2017-09-05T21:59:51Z Bike joined #sbcl
2017-09-05T22:01:03Z dougk joined #sbcl
2017-09-05T22:14:46Z karswell joined #sbcl
2017-09-05T22:22:03Z dougk quit (Ping timeout: 240 seconds)
2017-09-05T22:24:20Z stassats quit (Ping timeout: 260 seconds)
2017-09-05T22:50:11Z cromachina_ joined #sbcl
2017-09-05T22:53:03Z cromachina quit (Ping timeout: 240 seconds)
2017-09-05T23:01:15Z dougk joined #sbcl
2017-09-05T23:03:03Z rpg joined #sbcl
2017-09-05T23:03:58Z attila_lendvai quit (Quit: Leaving.)
2017-09-05T23:05:47Z dougk__ joined #sbcl
2017-09-05T23:06:32Z dougk quit (Ping timeout: 240 seconds)
2017-09-05T23:15:45Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-05T23:27:17Z flip214 quit (Ping timeout: 255 seconds)
2017-09-05T23:27:24Z flip214 joined #sbcl
2017-09-05T23:27:24Z flip214 quit (Changing host)
2017-09-05T23:27:24Z flip214 joined #sbcl
2017-09-05T23:37:33Z flip214 quit (Ping timeout: 240 seconds)
2017-09-05T23:37:40Z flip214 joined #sbcl
2017-09-05T23:37:40Z flip214 quit (Changing host)
2017-09-05T23:37:40Z flip214 joined #sbcl
2017-09-05T23:55:42Z Bicyclidine joined #sbcl
2017-09-05T23:59:09Z Bike quit (Ping timeout: 255 seconds)
2017-09-06T00:02:25Z rumbler31 joined #sbcl
2017-09-06T00:10:39Z dougk__ quit (Ping timeout: 248 seconds)
2017-09-06T00:14:02Z Bike joined #sbcl
2017-09-06T00:14:50Z dougk joined #sbcl
2017-09-06T00:16:55Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-06T00:22:36Z Bicyclidine joined #sbcl
2017-09-06T00:23:15Z Bike quit (Ping timeout: 252 seconds)
2017-09-06T00:34:56Z rpg joined #sbcl
2017-09-06T00:38:45Z Bike joined #sbcl
2017-09-06T00:39:56Z milanj quit (Quit: This computer has gone to sleep)
2017-09-06T00:42:17Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-06T01:03:14Z anticrisis joined #sbcl
2017-09-06T01:16:34Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-06T01:33:19Z dougk quit (Ping timeout: 260 seconds)
2017-09-06T01:57:04Z rpg joined #sbcl
2017-09-06T01:58:08Z anticrisis quit (Quit: be excellent to each other)
2017-09-06T02:30:08Z Bike quit (Ping timeout: 240 seconds)
2017-09-06T02:51:34Z karswell quit (Remote host closed the connection)
2017-09-06T02:56:38Z Bike joined #sbcl
2017-09-06T03:14:50Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-06T03:20:58Z anticrisis joined #sbcl
2017-09-06T03:50:54Z Bike quit (Ping timeout: 255 seconds)
2017-09-06T04:04:13Z zulu_inuoe joined #sbcl
2017-09-06T04:06:33Z zulu_inuoe_ quit (Ping timeout: 252 seconds)
2017-09-06T05:03:51Z shka_ joined #sbcl
2017-09-06T05:13:59Z leah2 joined #sbcl
2017-09-06T05:54:08Z scymtym quit (Ping timeout: 240 seconds)
2017-09-06T05:57:25Z oleo quit (Quit: irc client terminated!)
2017-09-06T06:17:06Z zulu_inuoe_ joined #sbcl
2017-09-06T06:18:33Z shka_ quit (Ping timeout: 252 seconds)
2017-09-06T06:19:02Z zulu_inuoe quit (Ping timeout: 240 seconds)
2017-09-06T06:47:58Z milanj joined #sbcl
2017-09-06T07:01:07Z angavrilov joined #sbcl
2017-09-06T07:21:19Z joshe quit (Ping timeout: 246 seconds)
2017-09-06T07:22:18Z joshe joined #sbcl
2017-09-06T07:22:32Z scymtym joined #sbcl
2017-09-06T08:19:26Z attila_lendvai joined #sbcl
2017-09-06T08:19:26Z attila_lendvai quit (Changing host)
2017-09-06T08:19:26Z attila_lendvai joined #sbcl
2017-09-06T08:40:20Z stassats joined #sbcl
2017-09-06T08:45:36Z Shinmera: SBCL 1.3.21 seems to fail to build on OS X 10.11: http://plaster.tymoon.eu/view/664
2017-09-06T09:09:30Z BitPuffin|osx joined #sbcl
2017-09-06T09:12:48Z anticrisis quit (Quit: be excellent to each other)
2017-09-06T09:19:49Z Shinmera: The above seems to happen since 1.3.18.
2017-09-06T09:26:08Z Shinmera: As in, 1.3.17 builds successfully.
2017-09-06T09:45:55Z Bike joined #sbcl
2017-09-06T10:12:20Z scymtym: Shinmera: i can't tell form the build log. are you building without sb-ldb?
2017-09-06T10:13:42Z scymtym: i mean, i can see that sb-ldb is not in target features, but are you actively disabling it?
2017-09-06T10:22:08Z eschatologist quit (Ping timeout: 240 seconds)
2017-09-06T10:22:47Z eschatologist joined #sbcl
2017-09-06T10:28:08Z scymtym: ok, the problem seems to be that building with sb-traceroot but without sb-ldb is currently broken
2017-09-06T10:41:00Z scymtym: partial fix: http://paste.lisp.org/display/355168
2017-09-06T10:52:42Z cro__ joined #sbcl
2017-09-06T10:53:21Z zulu_inuoe joined #sbcl
2017-09-06T10:55:58Z zulu_inuoe_ quit (Ping timeout: 248 seconds)
2017-09-06T10:55:58Z cromachina_ quit (Ping timeout: 248 seconds)
2017-09-06T11:58:48Z dougk joined #sbcl
2017-09-06T12:00:02Z rpg joined #sbcl
2017-09-06T12:04:29Z rpg quit (Ping timeout: 260 seconds)
2017-09-06T12:09:44Z Shinmera: scymtym: yes.
2017-09-06T12:10:12Z Shinmera: scymtym: I typically disable ldb since it gets my processes stuck in it if it severely crashes and there's nothing I can do with it anyway.
2017-09-06T12:13:51Z scymtym: Shinmera: could you test whether the patch helps on your system? after we fix this, i suggest --without-sb-traceroot --without-sb-ldb since both are advanced and/or mainly for developers.
2017-09-06T12:14:49Z Shinmera: scymtym: Sure. I'll run a test in a minute.
2017-09-06T12:24:57Z ralt joined #sbcl
2017-09-06T12:25:04Z dougk quit (Read error: Connection reset by peer)
2017-09-06T12:25:19Z dougk joined #sbcl
2017-09-06T12:30:58Z ralt left #sbcl
2017-09-06T12:31:47Z ralt joined #sbcl
2017-09-06T12:34:40Z Shinmera: scymtym: patch seems to fix things.
2017-09-06T12:34:51Z Shinmera: I'll test --without-sb-traceroot without the patch real quick too.
2017-09-06T12:37:23Z Bike quit (Ping timeout: 255 seconds)
2017-09-06T12:38:01Z scymtym: that will probably not work
2017-09-06T12:38:10Z scymtym: at least when i tried
2017-09-06T12:38:13Z Shinmera: Yeah. Failure occurs even if I disable both traceroot and ldb.
2017-09-06T12:38:39Z Shinmera: I'll still include the traceroot disable in my build script for now and keep SBCL at 1.3.17 until 1.3.22 is out.
2017-09-06T12:39:40Z scymtym: traceroot wasn't present in older SBCLs but --without-sb-traceroot should do no harm
2017-09-06T12:41:21Z scymtym: dougk: stassats: should i try harder to avoid the odd ifdef around throw_to_monitor before committing http://paste.lisp.org/display/355168 ?
2017-09-06T12:44:21Z dougk: shouldn't that test be based on whether ldb is enabled versus whether the feature is builtin? I never don't build it in, but I do disable it.  And why would we ever not want to support traceroot?
2017-09-06T12:50:06Z stassats quit (Ping timeout: 248 seconds)
2017-09-06T12:52:54Z scymtym: dougk: by ldb being enabled or not, do you mean the --disable-ldb runtime option? the runtime code tests for LISP_FEATURE_SB_LDB in several places to exclude certain code already, so i thought that was the way to go
2017-09-06T12:56:39Z dougk: imo, the right way to do this is that if LISP_FEATURE_SB_LDB is not present, then 'disable_lossage_handler_p' should be a "#define" to 0. Then don't use any #ifdefs  for it; instead test 'disable_lossage_handler_p' whose scope would have to be enlarged out of 'main'
2017-09-06T12:56:56Z dougk: Then the C compiler can statically detect what code it can elide in one case; or test it in the other
2017-09-06T12:57:59Z dougk: the only #ifdef will be one guarding the assignment to 'disable_p'. If the so-called variable is actually a #define, don't try to assign to it
2017-09-06T12:58:50Z dougk: still keep the guard around the 'monitor.c' file as a whole using the LISP_FEATURE_SB_LDB though
2017-09-06T12:58:52Z scymtym looks at the code
2017-09-06T13:00:43Z scymtym: print.c needs monitor.c, though and with-sb-traceroot without-sb-ldb needs print
2017-09-06T13:00:46Z dougk: actually i don't really care too much though. it looks like all you're trying to do is have continuep not die if ldb_monitor doesn't exist, but if the feature exists, you should be willing to call it in continue_p ?
2017-09-06T13:00:56Z scymtym: that's part of what caused Shinmera's build failure
2017-09-06T13:01:19Z dougk: ok, what's the issue? i'm not familiar with what you're trying to solve
2017-09-06T13:02:31Z scymtym: without-sb-ldb disables (via defined(LISP_FEATURE_SB_LDB)) some code in print.c that is used by traceroot
2017-09-06T13:04:05Z scymtym: (and without-sb-traceroot still builds traceroot.c)
2017-09-06T13:07:10Z rpg joined #sbcl
2017-09-06T13:07:18Z Bike joined #sbcl
2017-09-06T13:30:55Z stassats joined #sbcl
2017-09-06T13:31:01Z attila_lendvai quit (Quit: Leaving.)
2017-09-06T13:34:56Z dougk quit (Ping timeout: 248 seconds)
2017-09-06T13:42:47Z oleo joined #sbcl
2017-09-06T13:47:53Z dougk joined #sbcl
2017-09-06T13:52:15Z whoma1 joined #sbcl
2017-09-06T13:53:22Z dougk quit (Read error: Connection reset by peer)
2017-09-06T13:55:00Z whoman quit (Ping timeout: 264 seconds)
2017-09-06T13:57:51Z attila_lendvai joined #sbcl
2017-09-06T13:57:51Z attila_lendvai quit (Changing host)
2017-09-06T13:57:51Z attila_lendvai joined #sbcl
2017-09-06T13:58:22Z dougk joined #sbcl
2017-09-06T13:58:34Z dougk: what problem does it solve to build without-ldb given that it can be disabled by a flag?
2017-09-06T13:59:08Z dougk: fixing whatever problem that is seems more pertinent that removing ldb.
2017-09-06T14:00:32Z Shinmera: Not having to change everything that launches sbcl to disable ldb is nice.
2017-09-06T14:01:10Z dougk: still seems like a way to set the default value for the 'enabled' flag is more clean than removing C code
2017-09-06T14:01:39Z dougk: removing the C code was, as far as I can tell from comments, solely to reduce the size of the runtime, as if 1 or 2 K matters any more
2017-09-06T14:02:56Z Shinmera: I don't personally gain anything from having it compiled out, so I'd be fine with changing it to setting the default for the flag.
2017-09-06T14:05:06Z jdz quit (*.net *.split)
2017-09-06T14:20:14Z stassats: scymtym: i'd put ifdef LISP_FEATURE_SB_TRACEROOT in GNUMakefile
2017-09-06T14:22:16Z stassats: ldb can be disabled at runtime
2017-09-06T14:22:52Z stassats: (sb-alien:alien-funcall (sb-alien:extern-alien "disable_lossage_handler" (function sb-alien:void)))
2017-09-06T14:27:21Z scymtym: dougk: stassats: thanks for the comments. i will try to put together a new patch after dayjob
2017-09-06T14:30:14Z dougk quit (Ping timeout: 252 seconds)
2017-09-06T14:33:20Z dougk joined #sbcl
2017-09-06T14:40:30Z dougk quit (Ping timeout: 248 seconds)
2017-09-06T14:43:03Z dougk joined #sbcl
2017-09-06T14:44:47Z jdz joined #sbcl
2017-09-06T14:48:47Z dougk quit (Ping timeout: 255 seconds)
2017-09-06T15:06:29Z ralt` joined #sbcl
2017-09-06T15:10:55Z ralt quit (Ping timeout: 255 seconds)
2017-09-06T15:17:45Z slyrus joined #sbcl
2017-09-06T15:28:50Z edgar-rft quit (Quit: edgar-rft)
2017-09-06T15:44:59Z slyrus quit (Quit: Client Quit)
2017-09-06T15:58:50Z slyrus joined #sbcl
2017-09-06T16:03:37Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-06T16:04:11Z shka_ joined #sbcl
2017-09-06T16:08:57Z rpg joined #sbcl
2017-09-06T16:16:25Z dougk joined #sbcl
2017-09-06T17:07:54Z stassats quit (Ping timeout: 246 seconds)
2017-09-06T18:19:50Z swflint quit (Ping timeout: 255 seconds)
2017-09-06T18:25:52Z edgar-rft joined #sbcl
2017-09-06T18:25:54Z swflint joined #sbcl
2017-09-06T18:29:45Z ralt`` joined #sbcl
2017-09-06T18:33:34Z ralt` quit (Ping timeout: 240 seconds)
2017-09-06T18:57:25Z dougk quit (Ping timeout: 246 seconds)
2017-09-06T18:59:41Z dougk joined #sbcl
2017-09-06T19:05:57Z ralt`` is now known as ralt
2017-09-06T19:07:59Z Xof quit (Ping timeout: 255 seconds)
2017-09-06T19:13:36Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-06T19:15:55Z whoma1 is now known as whoman
2017-09-06T19:19:59Z shka_ quit (Ping timeout: 248 seconds)
2017-09-06T19:23:23Z rpg joined #sbcl
2017-09-06T19:31:50Z scymtym quit (Ping timeout: 255 seconds)
2017-09-06T19:51:19Z Xof joined #sbcl
2017-09-06T20:01:49Z dougk quit (Ping timeout: 246 seconds)
2017-09-06T20:03:14Z dougk joined #sbcl
2017-09-06T20:08:46Z angavrilov quit (Remote host closed the connection)
2017-09-06T20:29:51Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-06T21:01:10Z scymtym joined #sbcl
2017-09-06T21:03:27Z dougk quit (Ping timeout: 246 seconds)
2017-09-06T21:08:37Z dougk joined #sbcl
2017-09-06T21:25:14Z dougk quit (Ping timeout: 255 seconds)
2017-09-06T21:37:19Z prxq joined #sbcl
2017-09-06T21:43:39Z ralt left #sbcl
2017-09-06T21:46:38Z Bike quit (Ping timeout: 260 seconds)
2017-09-06T22:10:30Z dougk joined #sbcl
2017-09-06T22:19:32Z rpg joined #sbcl
2017-09-06T22:22:39Z rpg quit (Client Quit)
2017-09-06T22:23:35Z rpg joined #sbcl
2017-09-06T22:29:07Z dougk quit (Ping timeout: 255 seconds)
2017-09-06T22:31:19Z prxq quit (Remote host closed the connection)
2017-09-06T22:37:28Z dougk joined #sbcl
2017-09-06T22:39:21Z Bike joined #sbcl
2017-09-06T22:39:59Z MetaYan joined #sbcl
2017-09-06T22:42:52Z MetaYan: Hey, fellow SBCL-ers, I've got 1.3.21 stuck in an infinite loop, probably because of an accidental ) instead of ( in the SLIME REPL. Would some sort of dump be helpful to sort out what caused it?
2017-09-06T22:49:12Z Bike: like, you found a bug in the reader?
2017-09-06T22:51:54Z MetaYan: Don't know if it's caused by Emacs, SLIME, or SBCL, but it's SBCL that's gone into a loop, and the message I got was "Missing closing delimiter: )", when that was what I accidentally entered, all by itself in the REPL.
2017-09-06T22:52:44Z MetaYan: On OS X 10.11.6.
2017-09-06T22:59:02Z Bike: you just typed (?
2017-09-06T22:59:05Z Bike: just (
2017-09-06T22:59:19Z MetaYan: Just )
2017-09-06T22:59:35Z Bike: and it's an actual hang, not just an error or something? you said you got a message
2017-09-06T22:59:43Z MetaYan: (Had the wrong keyboard layout, and the places are swithed.)
2017-09-06T22:59:52Z MetaYan: *switched
2017-09-06T23:00:16Z MetaYan: SBCL is running at 100% CPU.
2017-09-06T23:01:45Z MetaYan: I think the "Missing closing delimiter: )" might be from paredit, but it's strange that SBCL went into a spin.
2017-09-06T23:02:46Z dougk quit (Ping timeout: 246 seconds)
2017-09-06T23:07:41Z dougk joined #sbcl
2017-09-06T23:15:18Z MetaYan: Bike: I just found that it's not SBCL that's stuck. It's Emacs. Sorry.
2017-09-06T23:19:13Z dougk quit (Ping timeout: 246 seconds)
2017-09-06T23:32:55Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-06T23:34:58Z dougk joined #sbcl
2017-09-06T23:44:48Z attila_lendvai quit (Quit: Leaving.)
2017-09-06T23:54:01Z dougk quit (Ping timeout: 240 seconds)
2017-09-06T23:57:41Z dougk joined #sbcl
2017-09-07T00:23:40Z MetaYan left #sbcl
2017-09-07T00:36:15Z dougk quit (Ping timeout: 248 seconds)
2017-09-07T00:42:25Z Bicyclidine joined #sbcl
2017-09-07T00:43:13Z Bike quit (Ping timeout: 246 seconds)
2017-09-07T00:51:08Z dougk joined #sbcl
2017-09-07T01:02:52Z Bike joined #sbcl
2017-09-07T01:05:29Z Bicyclidine quit (Ping timeout: 252 seconds)
2017-09-07T01:36:19Z Bicyclidine joined #sbcl
2017-09-07T01:37:03Z Bike quit (Ping timeout: 260 seconds)
2017-09-07T01:42:44Z rumbler31 quit (Remote host closed the connection)
2017-09-07T02:00:20Z karswell joined #sbcl
2017-09-07T02:19:37Z milanj quit (Quit: This computer has gone to sleep)
2017-09-07T02:24:39Z milanj joined #sbcl
2017-09-07T03:02:09Z rpg joined #sbcl
2017-09-07T03:16:04Z Bike joined #sbcl
2017-09-07T03:17:57Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-07T03:44:59Z BitPuffin|osx quit (Ping timeout: 252 seconds)
2017-09-07T03:49:33Z shka_ joined #sbcl
2017-09-07T03:54:34Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-07T03:55:24Z rpg joined #sbcl
2017-09-07T03:55:29Z rpg quit (Client Quit)
2017-09-07T04:10:06Z Bike quit (Ping timeout: 246 seconds)
2017-09-07T04:18:37Z whoma1 joined #sbcl
2017-09-07T04:20:52Z myrkraverk quit (Ping timeout: 248 seconds)
2017-09-07T04:20:52Z sigjuice_ quit (Ping timeout: 248 seconds)
2017-09-07T04:20:52Z myrkraverk_ joined #sbcl
2017-09-07T04:20:52Z fiddlerwoaroof quit (Ping timeout: 248 seconds)
2017-09-07T04:21:11Z myrkraverk_ is now known as myrkraverk
2017-09-07T04:21:17Z whoman quit (Ping timeout: 248 seconds)
2017-09-07T04:21:35Z fiddlerwoaroof joined #sbcl
2017-09-07T04:21:56Z sigjuice joined #sbcl
2017-09-07T05:14:53Z angavrilov joined #sbcl
2017-09-07T05:25:53Z oleo quit (Quit: irc client terminated!)
2017-09-07T06:06:19Z scymtym quit (Ping timeout: 252 seconds)
2017-09-07T06:55:36Z epipping joined #sbcl
2017-09-07T06:56:00Z dougk quit (Ping timeout: 240 seconds)
2017-09-07T07:15:30Z shka_ quit (Ping timeout: 240 seconds)
2017-09-07T07:27:04Z scymtym joined #sbcl
2017-09-07T07:34:27Z BitPuffin|osx joined #sbcl
2017-09-07T08:19:25Z dougk joined #sbcl
2017-09-07T08:23:49Z dougk quit (Ping timeout: 252 seconds)
2017-09-07T10:25:47Z Bike joined #sbcl
2017-09-07T10:50:22Z milanj quit (Quit: This computer has gone to sleep)
2017-09-07T10:50:49Z attila_lendvai joined #sbcl
2017-09-07T11:02:43Z Bike quit (Remote host closed the connection)
2017-09-07T11:03:42Z Bike joined #sbcl
2017-09-07T11:11:28Z Bicyclidine joined #sbcl
2017-09-07T11:12:18Z dougk joined #sbcl
2017-09-07T11:12:53Z Bike quit (Ping timeout: 246 seconds)
2017-09-07T11:16:13Z Bicyclidine quit (Ping timeout: 248 seconds)
2017-09-07T11:16:26Z m00natic joined #sbcl
2017-09-07T11:44:53Z milanj joined #sbcl
2017-09-07T11:59:32Z rumbler31 joined #sbcl
2017-09-07T11:59:49Z rpg joined #sbcl
2017-09-07T12:03:30Z dougk quit (Ping timeout: 240 seconds)
2017-09-07T12:04:18Z rpg quit (Ping timeout: 246 seconds)
2017-09-07T12:16:15Z Bike joined #sbcl
2017-09-07T12:29:16Z Bike quit (Ping timeout: 255 seconds)
2017-09-07T12:34:58Z rumbler31 quit (Remote host closed the connection)
2017-09-07T12:39:27Z dougk joined #sbcl
2017-09-07T12:39:41Z rumbler31 joined #sbcl
2017-09-07T12:46:13Z rumbler31 quit (Remote host closed the connection)
2017-09-07T13:05:24Z Bike joined #sbcl
2017-09-07T13:05:27Z Bike quit (Remote host closed the connection)
2017-09-07T13:05:49Z Bike joined #sbcl
2017-09-07T14:10:39Z cro__ quit (Read error: Connection reset by peer)
2017-09-07T14:20:29Z dougk_ quit (Ping timeout: 246 seconds)
2017-09-07T14:32:35Z dougk_ joined #sbcl
2017-09-07T14:36:15Z oleo joined #sbcl
2017-09-07T14:37:49Z scymtym quit (Ping timeout: 252 seconds)
2017-09-07T14:38:42Z rpg joined #sbcl
2017-09-07T14:42:13Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-07T14:45:02Z attila_lendvai joined #sbcl
2017-09-07T14:45:02Z attila_lendvai quit (Changing host)
2017-09-07T14:45:02Z attila_lendvai joined #sbcl
2017-09-07T15:16:14Z whoma1 is now known as whoman
2017-09-07T15:29:49Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-07T15:31:42Z rpg joined #sbcl
2017-09-07T16:05:37Z gko quit (Quit: ZNC - http://znc.in)
2017-09-07T16:19:37Z BitPuffin|osx quit (Ping timeout: 260 seconds)
2017-09-07T16:26:53Z gko joined #sbcl
2017-09-07T16:36:59Z BitPuffin|osx joined #sbcl
2017-09-07T16:39:39Z DGASAU quit (Read error: Connection reset by peer)
2017-09-07T16:40:55Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-07T16:42:36Z rpg joined #sbcl
2017-09-07T16:47:59Z dougk quit (Ping timeout: 240 seconds)
2017-09-07T16:58:10Z scymtym joined #sbcl
2017-09-07T16:58:13Z shka_ joined #sbcl
2017-09-07T17:06:07Z DGASAU joined #sbcl
2017-09-07T17:11:49Z stassats joined #sbcl
2017-09-07T17:17:27Z m00natic quit (Remote host closed the connection)
2017-09-07T17:19:11Z epipping quit
2017-09-07T17:24:06Z karswell quit (Remote host closed the connection)
2017-09-07T17:24:44Z karswell joined #sbcl
2017-09-07T19:16:06Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-07T19:20:10Z sjl joined #sbcl
2017-09-07T19:49:12Z dougk joined #sbcl
2017-09-07T19:52:41Z dougk quit (Read error: Connection reset by peer)
2017-09-07T19:52:57Z dougk joined #sbcl
2017-09-07T20:21:36Z scymtym: dougk: stassats: after our discussion, i think we can always build ldb, but disable it if requested and just support to build without traceroot: http://paste.lisp.org/display/355278 . does that seem good?
2017-09-07T20:22:14Z scymtym: s/just support/just not support/
2017-09-07T20:23:07Z stassats: do people check for #+sb-ldb?
2017-09-07T20:24:48Z scymtym: no idea. how would that be a problem?
2017-09-07T20:25:03Z dougk: this is not an answer, but ... We should document the features keywords people can check for. I'd really like to hide all the ones that control how to build the internals in ways that nobody should care about
2017-09-07T20:25:57Z Shinmera: Is there even any way to interact with ldb from the lisp side that would make people want to feature test for it?
2017-09-07T20:26:13Z stassats: yes, you can call it
2017-09-07T20:26:26Z stassats: "people"
2017-09-07T20:28:15Z scymtym: with the change, we would falsely claim that ldb wasn't available. that shouldn't cause problems
2017-09-07T20:29:05Z stassats: but nobody tests for #+sb-ldb
2017-09-07T20:29:46Z stassats: have you ever launched ldb from lisp?
2017-09-07T20:32:02Z scymtym: no, why did you bring up testing for :sb-ldb from lisp in this context? are you suggesting removing the target feature?
2017-09-07T20:32:21Z stassats: if it's no longer doing anything
2017-09-07T20:33:20Z scymtym: do we have a mechanism for --with[out]-foo build options that do not turn into target features?
2017-09-07T20:34:57Z stassats: when the debugger is disabled, a backtrace is printed, the same should happen when ldb is disabled
2017-09-07T20:36:04Z stassats: and shouldn't --disable-debugger imply --disable-ldb as well?
2017-09-07T20:36:58Z scymtym: is there a build-time option for disabling the debugger by default? i think Shinmera's use-case is making an sbcl binary that does not require commandline options to achieve that behavior
2017-09-07T20:37:58Z stassats: i only ever do --disable-ldb together with --disable-debugger and --lose-on-corruption
2017-09-07T20:38:35Z Shinmera: scymtym: Yes.
2017-09-07T20:39:20Z stassats: sb-ext:disable-debugger doesn't require runtime options
2017-09-07T20:42:44Z scymtym: stassats: same here, i use all of those as well. i think what Shinmera wants is users typing sbcl without options or init files and getting an sbcl with ldb disabled
2017-09-07T20:43:16Z stassats: sb-ext:disable-debugger doesn't require either
2017-09-07T20:44:00Z Shinmera: My issue with ldb is that it's never been useful for me in the cases I've landed in it, but it has prevented the process from just dying, which would have been more useful.
2017-09-07T20:44:29Z scymtym: in any case, the fact that make.sh --without-sb-ldb is accepted but then the runtime fails to build should be fixed
2017-09-07T20:44:42Z stassats: it is extremely useful
2017-09-07T20:45:59Z fe[nl]ix: it's mostly useful for SBCL developers
2017-09-07T20:46:38Z fe[nl]ix: end users would benefit from a crash handler that tries to print a stack trace of all threads then quits
2017-09-07T20:46:42Z stassats: sbcl crashes that reach ldb are not restricted to SBCL developers, though
2017-09-07T20:47:14Z fe[nl]ix: yes but if you're not familiar with the internals you won't know what to do with it
2017-09-07T20:47:33Z stassats: that's when you ask your local sbcl developer for "how do i debug this"
2017-09-07T20:49:12Z fe[nl]ix: I would definitely think compile sb-ldb always
2017-09-07T20:49:31Z fe[nl]ix: but I think it should be disabled by default at runtime
2017-09-07T20:49:55Z Shinmera: I agree that having it in always is no issue. I'd just like to be able to configure a default switch.
2017-09-07T20:50:12Z stassats: no to default disablement
2017-09-07T20:50:27Z fe[nl]ix: so if you have a repeatable crash at runtime, instead of logging a crash one could contact SBCL devs and then use ldb
2017-09-07T20:54:42Z scymtym: so consensus seems to be 1) always build ldb 2) --without-sb-ldb disables it by default 3) the --disable-ldb runtime option disables ldb if it is not already disabled through the build-time option 4) the :sb-ldb feature becomes useless and can be removed once we have a way to do that
2017-09-07T20:55:30Z fe[nl]ix: sounds right
2017-09-07T20:56:29Z fe[nl]ix: as much as feasible, I think it's a good idea to have code always compiled in and switch its functionality on/off with a flag
2017-09-07T20:57:24Z fe[nl]ix: so you don't have to worry that your change broke under #+sb-foo or #-sb-bar
2017-09-07T20:57:28Z fe[nl]ix: sb-unicode could be the next
2017-09-07T20:58:31Z stassats: it couldn't
2017-09-07T20:59:03Z fe[nl]ix: why not ?
2017-09-07T20:59:12Z stassats: it's not a switch
2017-09-07T21:14:06Z Bike quit (Ping timeout: 248 seconds)
2017-09-07T21:17:30Z dougk quit (Ping timeout: 240 seconds)
2017-09-07T21:21:56Z pkhuong_: yeah, sb-unicode is a pretty deep change.
2017-09-07T21:22:15Z pkhuong_: and code compiled without unicode shouldn't be mixed with code compiled with unicode
2017-09-07T21:35:15Z cromachina joined #sbcl
2017-09-07T21:43:38Z rpg joined #sbcl
2017-09-07T21:44:28Z Bike joined #sbcl
2017-09-07T21:56:55Z sjl_ joined #sbcl
2017-09-07T21:59:06Z sjl quit (Ping timeout: 240 seconds)
2017-09-07T22:09:18Z dougk joined #sbcl
2017-09-07T22:10:35Z angavrilov quit (Remote host closed the connection)
2017-09-07T22:13:44Z rpg quit (Read error: Connection reset by peer)
2017-09-07T22:14:10Z rpg joined #sbcl
2017-09-07T22:46:22Z rumbler31 joined #sbcl
2017-09-07T22:47:58Z dougk__ joined #sbcl
2017-09-07T22:48:16Z whoman quit (Remote host closed the connection)
2017-09-07T22:48:37Z whoman joined #sbcl
2017-09-07T22:48:50Z rumbler31 quit (Remote host closed the connection)
2017-09-07T22:49:58Z dougk quit (Ping timeout: 240 seconds)
2017-09-07T22:56:11Z attila_lendvai quit (Quit: Leaving.)
2017-09-07T22:57:14Z rumbler31 joined #sbcl
2017-09-07T22:59:10Z shka_ quit (Ping timeout: 264 seconds)
2017-09-07T23:02:42Z dougk__ quit (Ping timeout: 260 seconds)
2017-09-07T23:03:41Z dougk joined #sbcl
2017-09-07T23:06:49Z dougk quit (Read error: Connection reset by peer)
2017-09-07T23:07:08Z dougk joined #sbcl
2017-09-07T23:20:03Z rumbler31 quit (Remote host closed the connection)
2017-09-07T23:37:12Z dougk__ joined #sbcl
2017-09-07T23:39:50Z dougk quit (Read error: Connection reset by peer)
2017-09-07T23:44:29Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-07T23:46:13Z Bike quit (Remote host closed the connection)
2017-09-07T23:46:35Z Bike joined #sbcl
2017-09-08T00:27:34Z rpg joined #sbcl
2017-09-08T00:36:35Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-08T01:05:32Z stassats quit (Ping timeout: 248 seconds)
2017-09-08T01:18:27Z Bicyclidine joined #sbcl
2017-09-08T01:20:33Z Bike quit (Ping timeout: 246 seconds)
2017-09-08T01:47:30Z milanj quit (Quit: This computer has gone to sleep)
2017-09-08T02:48:44Z sjl joined #sbcl
2017-09-08T02:50:12Z sjl_ quit (Ping timeout: 260 seconds)
2017-09-08T02:56:30Z rumbler31 joined #sbcl
2017-09-08T03:37:32Z Bicyclidine quit (Ping timeout: 248 seconds)
2017-09-08T03:48:40Z whoman quit (Read error: Connection reset by peer)
2017-09-08T03:49:31Z whoman joined #sbcl
2017-09-08T05:22:35Z rumbler31 quit (Remote host closed the connection)
2017-09-08T05:46:04Z dougk__ quit (Ping timeout: 248 seconds)
2017-09-08T05:51:33Z oleo quit (Quit: irc client terminated!)
2017-09-08T06:02:50Z shka_ joined #sbcl
2017-09-08T06:04:29Z scymtym quit (Ping timeout: 240 seconds)
2017-09-08T06:16:31Z sjl_ joined #sbcl
2017-09-08T06:18:29Z sjl quit (Ping timeout: 240 seconds)
2017-09-08T06:20:08Z angavrilov joined #sbcl
2017-09-08T06:26:15Z carenz joined #sbcl
2017-09-08T06:27:42Z scymtym joined #sbcl
2017-09-08T06:32:08Z milanj joined #sbcl
2017-09-08T06:42:48Z dougk joined #sbcl
2017-09-08T06:47:36Z dougk quit (Ping timeout: 260 seconds)
2017-09-08T07:16:35Z shka_ quit (Ping timeout: 255 seconds)
2017-09-08T08:08:57Z Intensity quit (Remote host closed the connection)
2017-09-08T08:15:24Z milanj quit (Quit: This computer has gone to sleep)
2017-09-08T08:24:02Z Intensity joined #sbcl
2017-09-08T08:41:56Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-08T08:43:23Z BitPuffin|osx joined #sbcl
2017-09-08T08:55:24Z milanj joined #sbcl
2017-09-08T09:06:43Z attila_lendvai joined #sbcl
2017-09-08T10:13:46Z m00natic joined #sbcl
2017-09-08T10:15:54Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-08T10:18:44Z cromachina_ joined #sbcl
2017-09-08T10:21:29Z cromachina quit (Ping timeout: 240 seconds)
2017-09-08T10:22:08Z attila_lendvai joined #sbcl
2017-09-08T10:31:13Z gargaml joined #sbcl
2017-09-08T10:32:04Z gargaml quit (Client Quit)
2017-09-08T10:58:04Z nimiux quit (Ping timeout: 276 seconds)
2017-09-08T10:58:17Z nimiux joined #sbcl
2017-09-08T10:58:48Z dougk joined #sbcl
2017-09-08T11:02:52Z dougk quit (Ping timeout: 248 seconds)
2017-09-08T11:03:04Z Bike joined #sbcl
2017-09-08T11:18:06Z aeth_ joined #sbcl
2017-09-08T11:18:22Z aeth quit (Ping timeout: 246 seconds)
2017-09-08T11:52:13Z dougk joined #sbcl
2017-09-08T11:59:50Z rpg joined #sbcl
2017-09-08T12:04:29Z rpg quit (Ping timeout: 255 seconds)
2017-09-08T12:22:17Z stassats joined #sbcl
2017-09-08T12:24:27Z Bike quit (Ping timeout: 240 seconds)
2017-09-08T12:33:35Z rpg joined #sbcl
2017-09-08T12:39:31Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-08T12:41:45Z dougk quit (Ping timeout: 240 seconds)
2017-09-08T12:46:31Z attila_lendvai quit (Quit: Leaving.)
2017-09-08T12:48:04Z Bike joined #sbcl
2017-09-08T13:26:14Z oleo joined #sbcl
2017-09-08T13:49:11Z rumbler31 joined #sbcl
2017-09-08T14:09:08Z rumbler31 quit (Remote host closed the connection)
2017-09-08T14:21:01Z rpg joined #sbcl
2017-09-08T14:45:55Z dougk joined #sbcl
2017-09-08T14:49:56Z myrkraverk quit (Ping timeout: 240 seconds)
2017-09-08T14:53:26Z myrkraverk joined #sbcl
2017-09-08T15:20:11Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-08T15:35:25Z carenz quit (Ping timeout: 248 seconds)
2017-09-08T15:45:46Z rpg joined #sbcl
2017-09-08T15:46:59Z dougk quit (Ping timeout: 252 seconds)
2017-09-08T15:57:59Z Xof quit (Ping timeout: 240 seconds)
2017-09-08T16:04:20Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-08T16:06:57Z dougk joined #sbcl
2017-09-08T16:15:08Z rpg joined #sbcl
2017-09-08T16:31:42Z dougk__ joined #sbcl
2017-09-08T16:33:35Z dougk quit (Ping timeout: 255 seconds)
2017-09-08T16:37:30Z attila_lendvai joined #sbcl
2017-09-08T16:59:58Z sjl_ is now known as sjl
2017-09-08T17:07:17Z shka_ joined #sbcl
2017-09-08T17:08:20Z gendl: Hi, are :utf-16le and :utf-16be the only external-formats for utf-16? Apparently they will retain the BOM in parsed input?
2017-09-08T17:10:13Z stassats: check sb-impl::*external-formats*
2017-09-08T17:10:32Z dougk_: stassats: is it intended that static functions in immobile code are called through their fdefn now until the core becomes statically linked? it seems like what happens
2017-09-08T17:10:35Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-08T17:11:24Z stassats: dougk_: nothing should've changed on immobile-code
2017-09-08T17:11:32Z stassats: "should"
2017-09-08T17:12:27Z dougk_: hmm. i just built at that revision and I get "JMP #x203186B8 ; #<FDEFN LENGTH>" in immobile code
2017-09-08T17:13:11Z dougk_: i thought that I used to link those directly to the function. so while I agree with you that static-function are redundant, it seems like something isn't doing what it used to. Or else i remember wrong
2017-09-08T17:13:32Z stassats: maybe different fixups
2017-09-08T17:16:48Z m00natic quit (Remote host closed the connection)
2017-09-08T17:16:53Z stassats: using :static-call seems to do the job
2017-09-08T17:18:16Z stassats: although, what's really the difference in the end?
2017-09-08T17:19:26Z dougk_: probably nothing any more. there was a difference - static did not save the fixup into %code-fixups
2017-09-08T17:20:13Z dougk_: if i place the actual functions at known addresses in immobile code space, then space relocation does not need the stored fixup
2017-09-08T17:20:50Z dougk_: i think that ended up not working, or something. But I do notice now that there are 40,000 bignums in the core, which is kind of sad. They the packed fixup locs
2017-09-08T17:20:59Z dougk_: there used to be a lot less of them
2017-09-08T17:22:34Z dougk_: in fact, yes, i should try to get us back to the point where we do not need to store every location of the caller of a static function. Put them right after the assembly tramps and then all relative JMP or CALL magically is relocatable without re-fixup
2017-09-08T17:27:06Z dougk_: but hold on, one other question: the test you touched with FDEFN SUB-GC. It seems like there were two aspects to that test: (1) that tracing turns a static link back into an fdefn call, (2) that static fdefns can be found by the disassembler. Do both those aspects still work?
2017-09-08T17:29:11Z stassats: i rolled back my changes and it still was #<FUNCTION SUB-GC>
2017-09-08T17:29:24Z stassats: so i shrugged and snipped the FDEFN bit
2017-09-08T17:29:38Z stassats: if it's redefined at runtime it turns into a FDEFN
2017-09-08T17:29:54Z pkhuong_: we could also do the fixup at runtime?
2017-09-08T17:30:29Z pkhuong_: point to a trampoline, and overwrite the caller.
2017-09-08T17:31:33Z stassats: but it can be disassembled
2017-09-08T17:31:40Z stassats: retrying without my changes again
2017-09-08T17:31:42Z dougk_: pkhuong_: not sure what problem you're trying to solve. redefinition turns "call #<function foo>" back into "call #<fdefn foo>" and the assumption is that from then on you never want to turn it back to "call #<function foo>" again
2017-09-08T17:32:36Z stassats: i guess i didn't understand the test
2017-09-08T17:32:47Z dougk_: i'm not trying to optimize for the case where you want static call and re-defineable.  I also don't think static-funs (LENGTH et al) should "work" if they're redefined. I don't even know why I cared about sub-gc, other than that the disassembler failed to notice a static fdefn
2017-09-08T17:33:15Z dougk_: it disassembles fine, minus the annotation at the end. It didn't know who was at the address of #<fdefn sub-gc> as it never looked in static space
2017-09-08T17:33:24Z pkhuong_: dougk_: if you don't want to track all the locations you have to fix up. have them point to a trampoline and the trampoline can fixup its caller lazily.
2017-09-08T17:33:38Z stassats: ok, looks like using :named-call instead of :static-call makes TRACE not turn a FUNCTION into a FDEFN
2017-09-08T17:34:11Z stassats: a change to :static-call is trivial, but is it really needed?
2017-09-08T17:34:26Z dougk_: ok, sub-gc wasn't a "static-fun" that's why. we have two notions of fdefns in static space. 1 for C to call, and 1 for Lisp to call
2017-09-08T17:34:28Z dougk_: that's the difference
2017-09-08T17:35:05Z stassats: or can the :static-call just be removed?
2017-09-08T17:35:09Z stassats: fixup
2017-09-08T17:35:10Z dougk_: all i cared about, not performance,  in the disassembly test was that if an fdefn is in static space, it can be found
2017-09-08T17:35:40Z dougk_: :static-call can probably be removed, it wouldn't harm my sense of self.
2017-09-08T17:36:11Z stassats: it's presently not used, so it either should be used again on static functions or removed
2017-09-08T17:36:24Z stassats: (removing would not increase complexity)
2017-09-08T17:36:55Z dougk_: the argument in favor of use again on static-funs is that compilation to memory should avoid the fdefn and go straight for the function
2017-09-08T17:37:26Z dougk_: otherwise only a save-lisp-and-die will turn things from fdefn to the function
2017-09-08T17:38:12Z stassats: is it important to go from an fdefn to a function?
2017-09-08T17:38:28Z stassats: i see no performance differences
2017-09-08T17:39:19Z dougk_: dynamic linking ('.so' files, as in) is known to slow down an application by 1 to 3 %.  This is the same thing
2017-09-08T17:40:05Z stassats: and what about the relocatable static space prospects?
2017-09-08T17:41:04Z dougk_: very difficult. Still working on relocatable immobile space. Basically we need that for ELF-style core but (whether fe[nl]ix agrees or not) we don't need static space to be relocatable
2017-09-08T17:42:19Z stassats: even for sbcl.so?
2017-09-08T17:45:40Z dougk_: if it doesn't put static space where you want it, i'm just going to mmap it there myself. If *that* fails, then you lose
2017-09-08T17:46:05Z dougk_: i.e. i'm always going to force static space to be where it needs to be regardless of where any loader put it
2017-09-08T17:46:23Z stassats: but that reduces the usefulness of sbcl.so
2017-09-08T17:46:36Z stassats: (if it ever materializes)
2017-09-08T17:46:58Z stassats: so, are fdefns on immobile-code using some sort of trampolines?
2017-09-08T17:47:28Z stassats: "yes"
2017-09-08T17:47:43Z dougk_: raw-addr is the encoding of a jmp instruction
2017-09-08T17:47:48Z stassats: virtually no difference here in performance, at least calling the same function
2017-09-08T17:49:59Z stassats: ok, i'm changing it to use :static-call again
2017-09-08T17:52:56Z stassats: doesn't fix the sub-gc thing
2017-09-08T18:20:14Z aeth_ is now known as aeth
2017-09-08T18:30:21Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-08T18:40:31Z rpg joined #sbcl
2017-09-08T18:41:07Z attila_lendvai joined #sbcl
2017-09-08T18:44:51Z dougk joined #sbcl
2017-09-08T18:46:25Z dougk__ quit (Ping timeout: 255 seconds)
2017-09-08T19:00:22Z slyrus quit (Quit: Client Quit)
2017-09-08T19:09:24Z slyrus joined #sbcl
2017-09-08T19:40:16Z dougk quit (Ping timeout: 246 seconds)
2017-09-08T19:46:37Z myrkraverk quit (Ping timeout: 248 seconds)
2017-09-08T19:47:36Z dougk joined #sbcl
2017-09-08T19:54:51Z dougk quit (Ping timeout: 246 seconds)
2017-09-08T20:07:25Z aeth quit (Ping timeout: 255 seconds)
2017-09-08T20:09:01Z aeth joined #sbcl
2017-09-08T20:10:42Z dougk joined #sbcl
2017-09-08T20:14:19Z aeth quit (Ping timeout: 248 seconds)
2017-09-08T20:14:47Z dougk__ joined #sbcl
2017-09-08T20:15:27Z aeth joined #sbcl
2017-09-08T20:16:20Z dougk quit (Ping timeout: 255 seconds)
2017-09-08T20:25:20Z dougk__ quit (Ping timeout: 255 seconds)
2017-09-08T20:37:57Z dougk joined #sbcl
2017-09-08T20:40:58Z dougk_: just pushed some GC changes. wonder how many "FTB" emails from Eric Marsden I'll have to reply to
2017-09-08T20:48:11Z aeth quit (Ping timeout: 260 seconds)
2017-09-08T20:50:06Z aeth joined #sbcl
2017-09-08T20:53:53Z dougk quit (Ping timeout: 252 seconds)
2017-09-08T21:04:59Z slyrus: Franchise Tax Board?
2017-09-08T21:05:05Z slyrus: minion: FTB?
2017-09-08T21:05:05Z minion: Sorry, I couldn't find anything in the database for ``FTB''.
2017-09-08T21:09:50Z angavrilov quit (Remote host closed the connection)
2017-09-08T21:10:53Z stassats: minion: what does FTB mean?
2017-09-08T21:10:54Z minion: Footpath Towel Blackbreast
2017-09-08T21:11:32Z edgar-rft: Feed the birds
2017-09-08T21:28:59Z shka_ quit (Ping timeout: 248 seconds)
2017-09-08T21:31:44Z Bike quit (Ping timeout: 240 seconds)
2017-09-08T21:51:54Z dougk joined #sbcl
2017-09-08T22:00:50Z Bike joined #sbcl
2017-09-08T22:01:11Z dougk__ joined #sbcl
2017-09-08T22:01:12Z dougk quit (Read error: Connection reset by peer)
2017-09-08T22:20:28Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-08T22:22:04Z attila_lendvai quit (Quit: Leaving.)
2017-09-08T23:26:54Z rumbler31 joined #sbcl
2017-09-08T23:44:57Z dougk joined #sbcl
2017-09-08T23:47:52Z sjl_ joined #sbcl
2017-09-08T23:49:27Z sjl quit (Ping timeout: 246 seconds)
2017-09-09T00:35:50Z rumbler31 quit (Remote host closed the connection)
2017-09-09T00:44:20Z BitPuffin|osx quit (Ping timeout: 252 seconds)
2017-09-09T00:56:44Z Bicyclidine joined #sbcl
2017-09-09T00:59:04Z Bike quit (Ping timeout: 246 seconds)
2017-09-09T01:00:32Z whoman quit (Read error: Connection reset by peer)
2017-09-09T01:00:58Z whoman joined #sbcl
2017-09-09T01:05:13Z Bike joined #sbcl
2017-09-09T01:05:58Z stassats quit (Ping timeout: 240 seconds)
2017-09-09T01:08:33Z dougk quit (Read error: Connection reset by peer)
2017-09-09T01:08:33Z dougk__ joined #sbcl
2017-09-09T01:08:44Z Bicyclidine quit (Ping timeout: 248 seconds)
2017-09-09T01:10:56Z Bicyclidine joined #sbcl
2017-09-09T01:11:59Z Bike quit (Ping timeout: 255 seconds)
2017-09-09T01:18:46Z dougk joined #sbcl
2017-09-09T01:22:28Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-09T01:46:04Z dougk quit (Ping timeout: 240 seconds)
2017-09-09T02:24:41Z Bike joined #sbcl
2017-09-09T02:24:51Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-09T03:54:58Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-09T04:13:52Z dougk joined #sbcl
2017-09-09T04:17:58Z dougk quit (Ping timeout: 240 seconds)
2017-09-09T05:09:25Z slyrus quit (Ping timeout: 240 seconds)
2017-09-09T05:38:46Z Bicyclidine joined #sbcl
2017-09-09T05:42:15Z Bike quit (Ping timeout: 246 seconds)
2017-09-09T05:45:30Z Bicyclidine quit (Ping timeout: 248 seconds)
2017-09-09T05:59:38Z angavrilov joined #sbcl
2017-09-09T06:08:09Z DeadTrickster quit (Remote host closed the connection)
2017-09-09T07:31:17Z jrm quit (Quit: ciao)
2017-09-09T07:31:51Z jrm joined #sbcl
2017-09-09T09:00:02Z shka_ joined #sbcl
2017-09-09T09:08:53Z myrkraverk joined #sbcl
2017-09-09T09:20:18Z DeadTrickster joined #sbcl
2017-09-09T09:25:44Z scymtym quit (Ping timeout: 252 seconds)
2017-09-09T09:30:05Z BitPuffin|osx joined #sbcl
2017-09-09T09:37:27Z myrkraverk quit (Ping timeout: 246 seconds)
2017-09-09T09:50:15Z stassats joined #sbcl
2017-09-09T09:56:20Z scymtym joined #sbcl
2017-09-09T11:01:26Z Bike joined #sbcl
2017-09-09T11:17:44Z aeth_ joined #sbcl
2017-09-09T11:17:46Z aeth quit (Ping timeout: 248 seconds)
2017-09-09T12:09:25Z shka_ quit (Quit: Konversation terminated!)
2017-09-09T12:13:40Z shka joined #sbcl
2017-09-09T12:17:28Z myrkraverk joined #sbcl
2017-09-09T12:28:44Z myrkraverk quit (Ping timeout: 248 seconds)
2017-09-09T13:00:16Z milanj quit (Quit: This computer has gone to sleep)
2017-09-09T13:04:03Z stassats quit (Ping timeout: 240 seconds)
2017-09-09T13:19:54Z flip214 quit (Ping timeout: 240 seconds)
2017-09-09T13:38:37Z rpg joined #sbcl
2017-09-09T14:14:41Z dougk joined #sbcl
2017-09-09T14:15:27Z BitPuffin|osx quit (Ping timeout: 240 seconds)
2017-09-09T14:39:22Z myrkraverk joined #sbcl
2017-09-09T14:48:03Z rumbler31 joined #sbcl
2017-09-09T14:53:26Z rumbler31 quit (Remote host closed the connection)
2017-09-09T14:56:19Z dougk quit (Read error: Connection reset by peer)
2017-09-09T14:56:35Z dougk joined #sbcl
2017-09-09T15:12:57Z rumbler31 joined #sbcl
2017-09-09T15:22:33Z rumbler31 quit (Remote host closed the connection)
2017-09-09T15:48:05Z scymtym quit (Ping timeout: 246 seconds)
2017-09-09T16:05:10Z scymtym joined #sbcl
2017-09-09T16:34:40Z Bicyclidine joined #sbcl
2017-09-09T16:35:49Z Bike quit (Ping timeout: 252 seconds)
2017-09-09T16:59:24Z aeth_ is now known as aeth
2017-09-09T17:00:00Z milanj joined #sbcl
2017-09-09T17:05:04Z myrkraverk quit (Ping timeout: 255 seconds)
2017-09-09T17:05:56Z stassats joined #sbcl
2017-09-09T17:20:19Z flip214 joined #sbcl
2017-09-09T17:23:05Z rumbler31 joined #sbcl
2017-09-09T17:27:06Z rumbler31 quit (Ping timeout: 246 seconds)
2017-09-09T18:00:14Z whoman quit (Remote host closed the connection)
2017-09-09T18:00:35Z whoman joined #sbcl
2017-09-09T18:19:58Z AndreasO joined #sbcl
2017-09-09T18:26:10Z AndreasO quit (Quit: Found more important stuff than irc!)
2017-09-09T18:35:23Z whoman quit (Ping timeout: 240 seconds)
2017-09-09T18:44:25Z Bike joined #sbcl
2017-09-09T18:46:21Z Gorgias joined #sbcl
2017-09-09T18:47:57Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-09T18:48:39Z Bike quit (Ping timeout: 246 seconds)
2017-09-09T18:51:06Z Xof joined #sbcl
2017-09-09T18:51:57Z angavrilov quit (Remote host closed the connection)
2017-09-09T19:45:31Z attila_lendvai joined #sbcl
2017-09-09T20:02:47Z hel-io joined #sbcl
2017-09-09T20:04:12Z AndreasO joined #sbcl
2017-09-09T20:43:23Z attila_lendvai quit (Quit: Leaving.)
2017-09-09T20:43:36Z whoman joined #sbcl
2017-09-09T20:46:20Z easye quit (Remote host closed the connection)
2017-09-09T20:50:18Z stassats: huh, x86 breaks without a static two-arg-ior
2017-09-09T20:50:28Z stassats: inexplicably
2017-09-09T20:50:35Z minion quit (Ping timeout: 248 seconds)
2017-09-09T20:51:20Z specbot quit (Ping timeout: 264 seconds)
2017-09-09T20:51:34Z stassats: and of course the broken function is no longer broken when i redefine it at runtime
2017-09-09T20:52:31Z slyrus joined #sbcl
2017-09-09T20:53:43Z AndreasO quit (Quit: Found more important stuff than irc!)
2017-09-09T20:54:04Z Gorgias is now known as Bike
2017-09-09T21:01:56Z stassats: doesn't make any sense...
2017-09-09T21:02:17Z stassats: works fine on ppc
2017-09-09T21:12:24Z emaczen joined #sbcl
2017-09-09T21:12:36Z emaczen: how do I find the amount of memory an object is taking up?
2017-09-09T21:15:41Z stassats: but it's actually the logior vop that gets miscompiled, not a static function call
2017-09-09T21:17:02Z stassats: bad lifetimes? why would static function vops affect that?
2017-09-09T21:24:32Z stassats: yeah, just somehow got exposed and wasted my time by misdirection
2017-09-09T21:48:45Z emaczen quit (Quit: ERC (IRC client for Emacs 24.5.1))
2017-09-09T22:10:13Z shka quit (Ping timeout: 252 seconds)
2017-09-09T22:48:21Z hel-io quit (Remote host closed the connection)
2017-09-09T22:55:44Z hel-io joined #sbcl
2017-09-10T00:11:40Z karswell_ joined #sbcl
2017-09-10T00:11:58Z karswell quit (Remote host closed the connection)
2017-09-10T00:24:15Z jamtho joined #sbcl
2017-09-10T00:46:15Z karswell_ quit (Remote host closed the connection)
2017-09-10T00:46:59Z karswell_ joined #sbcl
2017-09-10T00:49:32Z karswell_ quit (Remote host closed the connection)
2017-09-10T00:50:15Z karswell_ joined #sbcl
2017-09-10T00:52:34Z karswell_ quit (Remote host closed the connection)
2017-09-10T00:53:21Z karswell_ joined #sbcl
2017-09-10T01:04:27Z stassats quit (Ping timeout: 248 seconds)
2017-09-10T01:17:15Z jamtho quit (Ping timeout: 248 seconds)
2017-09-10T01:20:04Z milanj quit (Quit: This computer has gone to sleep)
2017-09-10T03:08:48Z nimiux quit (Ping timeout: 246 seconds)
2017-09-10T03:09:08Z nimiux joined #sbcl
2017-09-10T03:12:56Z dougk quit (Read error: Connection reset by peer)
2017-09-10T03:13:11Z dougk joined #sbcl
2017-09-10T03:15:43Z dougk_ quit (Ping timeout: 246 seconds)
2017-09-10T03:27:43Z dougk_ joined #sbcl
2017-09-10T03:29:15Z hel-io quit (Remote host closed the connection)
2017-09-10T03:30:57Z dougk quit (Ping timeout: 240 seconds)
2017-09-10T04:07:36Z Bike quit (Ping timeout: 246 seconds)
2017-09-10T04:29:42Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-10T04:30:11Z hel-io joined #sbcl
2017-09-10T04:34:36Z hel-io quit (Remote host closed the connection)
2017-09-10T04:36:57Z sjl_ quit (Ping timeout: 240 seconds)
2017-09-10T05:44:35Z easye joined #sbcl
2017-09-10T06:02:24Z milanj joined #sbcl
2017-09-10T07:01:30Z shka joined #sbcl
2017-09-10T07:06:08Z angavrilov joined #sbcl
2017-09-10T07:10:36Z oleo quit (Read error: Connection reset by peer)
2017-09-10T07:22:34Z specbot joined #sbcl
2017-09-10T07:22:39Z minion joined #sbcl
2017-09-10T08:15:10Z milanj quit (Quit: Leaving)
2017-09-10T08:32:25Z whoma1 joined #sbcl
2017-09-10T08:33:31Z whoman quit (Ping timeout: 248 seconds)
2017-09-10T08:40:39Z Posterdati quit (Ping timeout: 260 seconds)
2017-09-10T08:49:46Z oleo joined #sbcl
2017-09-10T08:54:09Z Posterdati joined #sbcl
2017-09-10T08:56:33Z jamtho joined #sbcl
2017-09-10T09:14:54Z dougk_ quit (Ping timeout: 246 seconds)
2017-09-10T09:26:57Z dougk_ joined #sbcl
2017-09-10T09:30:41Z stassats joined #sbcl
2017-09-10T09:32:58Z jack_rabbit quit (Remote host closed the connection)
2017-09-10T09:38:02Z jamtho quit (Ping timeout: 240 seconds)
2017-09-10T09:41:46Z jack_rabbit joined #sbcl
2017-09-10T09:43:39Z scymtym quit (Ping timeout: 260 seconds)
2017-09-10T09:46:41Z jamtho joined #sbcl
2017-09-10T09:59:01Z scymtym joined #sbcl
2017-09-10T10:10:40Z scymtym_ joined #sbcl
2017-09-10T10:14:39Z scymtym quit (Ping timeout: 255 seconds)
2017-09-10T11:09:30Z scymtym_: sb-introspect test allocation-information.5 started to sometimes fail on x86_64 because the large vector can end up in :generation 0 even after the gc
2017-09-10T11:39:22Z scymtym_: dougk_: i would like to commit http://paste.lisp.org/display/355491 to unbreak the build unless you have a better solution
2017-09-10T11:40:04Z stassats: an sb-introspect test is also broken on x86-64 without immobile-code
2017-09-10T11:47:32Z Bike joined #sbcl
2017-09-10T11:47:42Z stassats quit (Ping timeout: 240 seconds)
2017-09-10T11:55:22Z Bicyclidine joined #sbcl
2017-09-10T11:56:41Z Bike quit (Ping timeout: 248 seconds)
2017-09-10T11:59:31Z myrkraverk joined #sbcl
2017-09-10T12:00:10Z rpg joined #sbcl
2017-09-10T12:04:48Z rpg quit (Ping timeout: 260 seconds)
2017-09-10T12:10:02Z scymtym_: i'm preparing commits for https://sourceforge.net/p/sbcl/sbcl/merge-requests/1/ . any opinions regarding the questions at the end of the page?
2017-09-10T12:10:12Z Bike joined #sbcl
2017-09-10T12:13:59Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-10T12:30:45Z rpg joined #sbcl
2017-09-10T12:38:03Z rpg quit (Ping timeout: 260 seconds)
2017-09-10T13:31:36Z dougk joined #sbcl
2017-09-10T13:35:42Z rpg joined #sbcl
2017-09-10T13:36:42Z epipping joined #sbcl
2017-09-10T13:42:17Z jamtho quit (Ping timeout: 248 seconds)
2017-09-10T13:50:41Z dougk__ joined #sbcl
2017-09-10T13:52:49Z dougk quit (Read error: Connection reset by peer)
2017-09-10T13:54:56Z whoma1 is now known as whoman
2017-09-10T14:02:26Z rpg quit (Ping timeout: 240 seconds)
2017-09-10T14:19:59Z oleo quit (Ping timeout: 255 seconds)
2017-09-10T14:33:23Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-10T14:34:11Z stassats joined #sbcl
2017-09-10T14:35:33Z dougk joined #sbcl
2017-09-10T14:54:51Z stassats: scymtym_: my opinion "whoa, sourceforge has pull requests"
2017-09-10T14:55:51Z oleo joined #sbcl
2017-09-10T14:56:23Z scymtym_: haha, yeah. well, saying it like that evokes images of a useful and usable feature
2017-09-10T14:57:35Z scymtym_: i don't even know whether i will be able to close the thing
2017-09-10T15:00:12Z rpg joined #sbcl
2017-09-10T15:02:10Z epipping: (I certainly can't close it; I can only "reject" it).
2017-09-10T15:02:27Z epipping: scymtym_: Thanks for improving and merging it!
2017-09-10T15:03:33Z scymtym_: epipping: you're welcome. i hope it doesn't break too badly on platforms i couldn't test
2017-09-10T15:05:27Z scymtym_: ok, i changed the status to "merged"
2017-09-10T15:06:28Z scymtym_: this is like "what github would have been like if linus made both linux and git in the nineties"
2017-09-10T15:13:47Z rpg quit (Read error: Connection reset by peer)
2017-09-10T15:13:52Z rpg_ joined #sbcl
2017-09-10T15:20:06Z oleo quit (Read error: Connection reset by peer)
2017-09-10T15:35:09Z slyrus quit (Quit: Client Quit)
2017-09-10T15:39:07Z scymtym_: isn't SB-IMPL:OUTPUT-OBJECT inconsistent w.r.t. to the supposed meaning of *SUPPRESS-PRINT-ERRORS*? if there is a random applicable condition handler somewhere on the stack that doesn't happen to invoke the CONTINUE restart, the print error isn't suppressed
2017-09-10T15:43:10Z Brucio-92 joined #sbcl
2017-09-10T15:43:20Z Brucio-92 is now known as slyrus
2017-09-10T15:47:06Z attila_lendvai joined #sbcl
2017-09-10T15:51:05Z scymtym_ quit (Ping timeout: 246 seconds)
2017-09-10T16:06:32Z scymtym joined #sbcl
2017-09-10T16:07:24Z oleo joined #sbcl
2017-09-10T16:55:10Z ASau joined #sbcl
2017-09-10T17:05:03Z ASau quit (Quit: ERC (IRC client for Emacs 25.2.1))
2017-09-10T17:30:37Z mateuszb joined #sbcl
2017-09-10T17:31:02Z mateuszb quit (Client Quit)
2017-09-10T17:37:37Z jamtho joined #sbcl
2017-09-10T17:44:56Z jamtho quit (Ping timeout: 248 seconds)
2017-09-10T17:54:59Z epipping quit
2017-09-10T18:11:06Z jamtho joined #sbcl
2017-09-10T18:11:48Z DeadTrickster quit (Remote host closed the connection)
2017-09-10T18:26:12Z DeadTrickster joined #sbcl
2017-09-10T19:07:36Z whoman quit (Ping timeout: 248 seconds)
2017-09-10T19:15:04Z jamtho quit (Ping timeout: 248 seconds)
2017-09-10T19:19:21Z dougk__ joined #sbcl
2017-09-10T19:22:26Z dougk quit (Ping timeout: 240 seconds)
2017-09-10T19:26:27Z angavrilov quit (Remote host closed the connection)
2017-09-10T19:26:45Z whoman joined #sbcl
2017-09-10T19:46:57Z karswell_ quit (Remote host closed the connection)
2017-09-10T19:47:35Z karswell_ joined #sbcl
2017-09-10T19:58:47Z oleo quit (Read error: Connection reset by peer)
2017-09-10T20:26:45Z hel-io joined #sbcl
2017-09-10T20:28:37Z rpg_ quit (Remote host closed the connection)
2017-09-10T20:29:13Z rpg joined #sbcl
2017-09-10T20:35:59Z hel-io quit (Remote host closed the connection)
2017-09-10T20:50:22Z slyrus quit (Ping timeout: 240 seconds)
2017-09-10T21:02:27Z aeth_ joined #sbcl
2017-09-10T21:04:56Z aeth quit (Ping timeout: 240 seconds)
2017-09-10T21:05:38Z aeth_ is now known as aeth
2017-09-10T21:13:26Z Bike quit (Ping timeout: 246 seconds)
2017-09-10T21:15:25Z Bike joined #sbcl
2017-09-10T21:53:51Z shka quit (Ping timeout: 240 seconds)
2017-09-10T21:55:47Z Bike quit (Ping timeout: 246 seconds)
2017-09-10T21:56:16Z hel-io joined #sbcl
2017-09-10T22:14:43Z Bike joined #sbcl
2017-09-10T22:31:42Z stassats quit (Ping timeout: 240 seconds)
2017-09-10T23:10:22Z hel-io quit (Remote host closed the connection)
2017-09-10T23:10:59Z stassats joined #sbcl
2017-09-10T23:29:03Z Bicyclidine joined #sbcl
2017-09-10T23:29:51Z attila_lendvai quit (Quit: Leaving.)
2017-09-10T23:31:35Z Bike quit (Ping timeout: 255 seconds)
2017-09-10T23:33:28Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-10T23:45:30Z stassats quit (Ping timeout: 248 seconds)
2017-09-10T23:47:14Z Bike joined #sbcl
2017-09-10T23:48:47Z hel-io joined #sbcl
2017-09-10T23:56:23Z hel-io quit (Remote host closed the connection)
2017-09-11T00:26:01Z slyrus joined #sbcl
2017-09-11T00:33:04Z karswell_ quit (Read error: Connection reset by peer)
2017-09-11T01:00:42Z zulu_inuoe_ joined #sbcl
2017-09-11T01:01:08Z zulu_inuoe quit (Ping timeout: 252 seconds)
2017-09-11T01:02:36Z zulu_inuoe_ quit (Remote host closed the connection)
2017-09-11T01:03:30Z galdor2 joined #sbcl
2017-09-11T01:03:55Z galdor1 quit (Ping timeout: 255 seconds)
2017-09-11T01:14:03Z Bike quit (Remote host closed the connection)
2017-09-11T01:22:42Z Bike joined #sbcl
2017-09-11T01:24:49Z jamtho joined #sbcl
2017-09-11T01:42:20Z emaczen joined #sbcl
2017-09-11T01:54:29Z emaczen left #sbcl
2017-09-11T02:10:21Z jamtho quit (Ping timeout: 240 seconds)
2017-09-11T02:38:57Z Bike quit (Ping timeout: 246 seconds)
2017-09-11T02:41:38Z Bike joined #sbcl
2017-09-11T02:58:02Z Bicyclidine joined #sbcl
2017-09-11T03:00:39Z Bike quit (Ping timeout: 246 seconds)
2017-09-11T03:27:38Z Bicyclidine quit (Ping timeout: 260 seconds)
2017-09-11T04:07:53Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-11T04:08:40Z cromachina joined #sbcl
2017-09-11T04:10:50Z cromachina_ quit (Ping timeout: 240 seconds)
2017-09-11T05:00:25Z cromachina quit (Ping timeout: 240 seconds)
2017-09-11T05:17:45Z slyrus quit (Ping timeout: 248 seconds)
2017-09-11T05:27:02Z angavrilov joined #sbcl
2017-09-11T06:35:50Z shka joined #sbcl
2017-09-11T07:23:04Z shka quit (Ping timeout: 248 seconds)
2017-09-11T07:40:36Z carenz joined #sbcl
2017-09-11T08:41:28Z dougk__ quit (Ping timeout: 248 seconds)
2017-09-11T09:01:01Z attila_lendvai joined #sbcl
2017-09-11T09:01:01Z attila_lendvai quit (Changing host)
2017-09-11T09:01:01Z attila_lendvai joined #sbcl
2017-09-11T09:26:05Z stassats joined #sbcl
2017-09-11T09:50:25Z jamtho joined #sbcl
2017-09-11T10:02:01Z scymtym quit (Ping timeout: 248 seconds)
2017-09-11T10:05:44Z m00natic joined #sbcl
2017-09-11T10:42:10Z scymtym joined #sbcl
2017-09-11T10:52:57Z Bike joined #sbcl
2017-09-11T10:54:17Z jamtho quit (Ping timeout: 248 seconds)
2017-09-11T10:57:57Z Bicyclidine joined #sbcl
2017-09-11T10:59:27Z Bike quit (Ping timeout: 246 seconds)
2017-09-11T11:41:11Z dougk joined #sbcl
2017-09-11T11:51:21Z stassats quit (Ping timeout: 248 seconds)
2017-09-11T11:59:41Z rpg joined #sbcl
2017-09-11T12:00:53Z rpg quit (Client Quit)
2017-09-11T12:27:15Z Bicyclidine is now known as Bike
2017-09-11T12:35:00Z Bike quit (Ping timeout: 246 seconds)
2017-09-11T12:45:03Z attila_lendvai quit (Quit: Leaving.)
2017-09-11T12:56:19Z Bike joined #sbcl
2017-09-11T13:07:26Z christoph_debian quit (Ping timeout: 246 seconds)
2017-09-11T13:34:55Z rpg joined #sbcl
2017-09-11T13:34:56Z christoph_debian joined #sbcl
2017-09-11T13:44:09Z jamtho joined #sbcl
2017-09-11T13:47:23Z attila_lendvai joined #sbcl
2017-09-11T13:47:23Z attila_lendvai quit (Changing host)
2017-09-11T13:47:23Z attila_lendvai joined #sbcl
2017-09-11T14:00:08Z jamtho quit (Ping timeout: 240 seconds)
2017-09-11T14:45:47Z scymtym quit (Ping timeout: 246 seconds)
2017-09-11T14:52:37Z jamtho joined #sbcl
2017-09-11T15:10:20Z rpg quit (Remote host closed the connection)
2017-09-11T15:10:51Z rpg joined #sbcl
2017-09-11T15:11:06Z dougk quit (Ping timeout: 246 seconds)
2017-09-11T15:11:40Z oleo joined #sbcl
2017-09-11T15:13:49Z scymtym joined #sbcl
2017-09-11T15:14:08Z slyrus joined #sbcl
2017-09-11T15:37:42Z dougk joined #sbcl
2017-09-11T15:57:11Z jamtho quit (Ping timeout: 248 seconds)
2017-09-11T16:10:25Z dougk quit (Ping timeout: 240 seconds)
2017-09-11T16:29:11Z dougk joined #sbcl
2017-09-11T16:33:42Z attila_lendvai quit (Quit: Leaving.)
2017-09-11T17:02:14Z m00natic quit (Read error: Connection reset by peer)
2017-09-11T17:04:43Z carenz quit (Ping timeout: 252 seconds)
2017-09-11T17:19:30Z dougk quit (Ping timeout: 255 seconds)
2017-09-11T17:21:40Z dougk joined #sbcl
2017-09-11T17:24:40Z shka_ joined #sbcl
2017-09-11T18:12:25Z AndreasO joined #sbcl
2017-09-11T19:28:16Z milanj joined #sbcl
2017-09-11T19:32:49Z shka_ quit (Ping timeout: 240 seconds)
2017-09-11T19:48:49Z dougk__ joined #sbcl
2017-09-11T19:49:24Z attila_lendvai joined #sbcl
2017-09-11T19:50:23Z dougk quit (Ping timeout: 246 seconds)
2017-09-11T20:07:03Z AndreasO quit (Remote host closed the connection)
2017-09-11T20:15:02Z foom2 joined #sbcl
2017-09-11T20:15:11Z whoman quit (Remote host closed the connection)
2017-09-11T20:18:53Z angular_mike___ joined #sbcl
2017-09-11T20:19:39Z pchrist_ joined #sbcl
2017-09-11T20:20:50Z |3b|` joined #sbcl
2017-09-11T20:21:56Z les` joined #sbcl
2017-09-11T20:23:28Z Bike quit (*.net *.split)
2017-09-11T20:23:28Z dougk_ quit (*.net *.split)
2017-09-11T20:23:29Z foom quit (*.net *.split)
2017-09-11T20:23:29Z gingerale quit (*.net *.split)
2017-09-11T20:23:30Z nicdev quit (*.net *.split)
2017-09-11T20:23:31Z |3b| quit (*.net *.split)
2017-09-11T20:23:32Z angular_mike_ quit (*.net *.split)
2017-09-11T20:23:33Z les quit (*.net *.split)
2017-09-11T20:23:33Z p_l quit (*.net *.split)
2017-09-11T20:23:34Z pchrist quit (*.net *.split)
2017-09-11T20:29:11Z attila_lendvai quit (Quit: Leaving.)
2017-09-11T20:30:11Z dougk_ joined #sbcl
2017-09-11T20:31:29Z gingerale joined #sbcl
2017-09-11T20:32:42Z Bike joined #sbcl
2017-09-11T20:38:41Z leah2 quit (Ping timeout: 264 seconds)
2017-09-11T20:41:22Z leah2 joined #sbcl
2017-09-11T21:05:19Z angavrilov quit (Remote host closed the connection)
2017-09-11T21:13:17Z stassats joined #sbcl
2017-09-11T21:16:00Z Bike quit (Ping timeout: 240 seconds)
2017-09-11T21:30:39Z whoman joined #sbcl
2017-09-11T21:53:02Z stassats quit (Ping timeout: 260 seconds)
2017-09-11T22:01:43Z Bike joined #sbcl
2017-09-11T22:13:25Z p_l joined #sbcl
2017-09-11T22:22:27Z stassats joined #sbcl
2017-09-11T22:42:02Z whoman quit (Ping timeout: 260 seconds)
2017-09-11T22:46:36Z bugrum joined #sbcl
2017-09-11T22:52:07Z whoman joined #sbcl
2017-09-11T23:09:18Z stassats quit (Ping timeout: 240 seconds)
2017-09-11T23:46:39Z jamtho joined #sbcl
2017-09-12T00:13:08Z emaczen joined #sbcl
2017-09-12T00:13:26Z emaczen: can i expect similar output from (room) on different linux machines/versions?
2017-09-12T00:14:35Z emaczen: It appears that Ubuntu is using 800 more megabtyes
2017-09-12T00:30:37Z whoman: thats.. much. whats in the image? same architecture from what it was 800mb less from?
2017-09-12T00:39:17Z luis quit (Ping timeout: 264 seconds)
2017-09-12T00:43:16Z luis joined #sbcl
2017-09-12T00:51:35Z jamtho quit (Ping timeout: 248 seconds)
2017-09-12T01:19:05Z rumbler31 joined #sbcl
2017-09-12T01:26:48Z Bike quit (Ping timeout: 248 seconds)
2017-09-12T01:32:39Z edgar-rft: emaczen: depends... I assume for example that Lisp on a linux version from the mid-90s used less RAM than today.
2017-09-12T01:37:23Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-12T01:50:24Z bugrum quit (Remote host closed the connection)
2017-09-12T01:54:04Z bugrum joined #sbcl
2017-09-12T02:01:25Z dougk__ quit (Ping timeout: 255 seconds)
2017-09-12T02:05:10Z bugrum quit (Remote host closed the connection)
2017-09-12T02:21:19Z sjl joined #sbcl
2017-09-12T02:23:04Z sjl quit (Client Quit)
2017-09-12T02:26:20Z sjl joined #sbcl
2017-09-12T02:27:11Z sjl quit (Client Quit)
2017-09-12T02:27:44Z sjl joined #sbcl
2017-09-12T02:54:48Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T03:10:01Z emaczen left #sbcl
2017-09-12T03:17:13Z sjl joined #sbcl
2017-09-12T03:21:40Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T03:25:41Z dougk joined #sbcl
2017-09-12T03:26:09Z Bike joined #sbcl
2017-09-12T03:29:29Z rumbler31 quit (Remote host closed the connection)
2017-09-12T03:33:25Z sjl joined #sbcl
2017-09-12T03:37:54Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T03:55:30Z dougk quit (Read error: Connection reset by peer)
2017-09-12T03:55:45Z dougk joined #sbcl
2017-09-12T03:58:57Z Bike quit (Ping timeout: 246 seconds)
2017-09-12T04:24:23Z slyrus quit (Ping timeout: 248 seconds)
2017-09-12T04:34:15Z sjl joined #sbcl
2017-09-12T04:38:48Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T04:40:24Z oleo quit (Quit: irc client terminated!)
2017-09-12T04:42:07Z malm quit (Ping timeout: 252 seconds)
2017-09-12T04:42:58Z malm joined #sbcl
2017-09-12T04:51:57Z sjl joined #sbcl
2017-09-12T04:52:00Z rumbler31 joined #sbcl
2017-09-12T04:56:20Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T04:58:00Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-12T05:10:33Z sjl joined #sbcl
2017-09-12T05:12:15Z sjl_ joined #sbcl
2017-09-12T05:14:48Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T05:27:48Z shka_ joined #sbcl
2017-09-12T05:55:26Z scymtym quit (Ping timeout: 260 seconds)
2017-09-12T05:58:24Z shka_ quit (Ping timeout: 240 seconds)
2017-09-12T06:10:40Z angavrilov joined #sbcl
2017-09-12T06:54:29Z rumbler31 joined #sbcl
2017-09-12T06:58:40Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-12T07:27:17Z scymtym joined #sbcl
2017-09-12T07:28:07Z joshe quit (Ping timeout: 240 seconds)
2017-09-12T07:28:24Z joshe joined #sbcl
2017-09-12T07:31:42Z HTTP_____GK1wmSU joined #sbcl
2017-09-12T07:34:32Z HTTP_____GK1wmSU left #sbcl
2017-09-12T07:41:42Z smokeink_ joined #sbcl
2017-09-12T07:43:00Z smokeink_: bug? (ql:quickload :log4cl) fails on sbcl 1.3.20: "failed AVER: (eq (sb-c::functional-kind sb-c::functional) :toplevel-xep) This is probably a bug in SBCL itself. [....]"
2017-09-12T07:49:37Z carenz joined #sbcl
2017-09-12T07:55:16Z rumbler31 joined #sbcl
2017-09-12T07:59:47Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-12T08:26:17Z DeadTrickster_ joined #sbcl
2017-09-12T08:35:06Z scymtym: smokeink_: can you paste a backtrace?
2017-09-12T08:55:47Z smokeink_ quit (Ping timeout: 240 seconds)
2017-09-12T09:04:34Z scymtym: macos build failure since bf2e122e: http://paste.lisp.org/display/355643
2017-09-12T09:09:29Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-12T09:12:31Z DeadTrickster_ joined #sbcl
2017-09-12T09:14:24Z tonton quit (Ping timeout: 240 seconds)
2017-09-12T09:16:05Z tonton joined #sbcl
2017-09-12T09:18:19Z 7GHAAK86A joined #sbcl
2017-09-12T09:55:57Z rumbler31 joined #sbcl
2017-09-12T09:56:45Z 7GHAAK86A: scymtym: http://pastecode.ru/83a9c8/
2017-09-12T10:00:17Z rumbler31 quit (Ping timeout: 255 seconds)
2017-09-12T10:06:05Z scymtym: 7GHAAK86A: please type "backtrace RET" into the debugger prompt. also, does this happen with more recent SBCL versions?
2017-09-12T10:09:18Z attila_lendvai joined #sbcl
2017-09-12T10:16:22Z m00natic joined #sbcl
2017-09-12T10:23:10Z jamtho joined #sbcl
2017-09-12T10:26:39Z dougk quit (Ping timeout: 240 seconds)
2017-09-12T10:29:42Z jamtho quit (Ping timeout: 248 seconds)
2017-09-12T10:51:28Z attila_lendvai quit (Quit: Leaving.)
2017-09-12T10:54:04Z stassats joined #sbcl
2017-09-12T11:00:54Z Bike joined #sbcl
2017-09-12T12:29:54Z Bike quit (Ping timeout: 240 seconds)
2017-09-12T12:33:29Z francogrex joined #sbcl
2017-09-12T12:35:27Z francogrex: hi is it possible to reconnect to a running thread: debugger invoked on a SB-INT:SIMPLE-READER-PACKAGE-ERROR in thread #<THREAD "main thread" RUNNING {23EE7E89}>
2017-09-12T12:35:48Z francogrex: sbcl hangs and i would like to open another sbcl and connect it to that hanging running thread
2017-09-12T13:05:19Z Bike joined #sbcl
2017-09-12T13:07:07Z 7GHAAK86A: scymtym: yes it happens with more recent versions http://pastecode.ru/294008/
2017-09-12T13:15:49Z francogrex quit (Quit: ERC (IRC client for Emacs 25.1.1))
2017-09-12T13:16:20Z francogrex joined #sbcl
2017-09-12T13:22:45Z rumbler31 joined #sbcl
2017-09-12T13:23:11Z 7GHAAK86A: hm the backtrace is incomplete on that paste link, lemme paste again
2017-09-12T13:23:27Z scymtym: 7GHAAK86A: thanks
2017-09-12T13:23:35Z dougk joined #sbcl
2017-09-12T13:24:10Z 7GHAAK86A: oh it's fine i was looking at the wrong window
2017-09-12T13:24:36Z 7GHAAK86A: the last link is fine
2017-09-12T13:30:50Z rumbler31 quit (Remote host closed the connection)
2017-09-12T13:36:17Z francogrex quit (Quit: ERC (IRC client for Emacs 25.1.1))
2017-09-12T14:01:21Z scymtym: reduced test case: http://paste.lisp.org/display/355653
2017-09-12T14:18:23Z whoman quit (Ping timeout: 240 seconds)
2017-09-12T14:23:51Z oleo joined #sbcl
2017-09-12T14:31:22Z rumbler3_ joined #sbcl
2017-09-12T14:35:57Z rumbler3_ quit (Ping timeout: 246 seconds)
2017-09-12T15:04:43Z slyrus joined #sbcl
2017-09-12T15:15:26Z scymtym: introduced by b4f3834e84f6946e99d37744718e528f4fa94dd4 if i bisected correctly
2017-09-12T15:44:18Z nimbus7 joined #sbcl
2017-09-12T15:46:30Z carenz quit (Ping timeout: 248 seconds)
2017-09-12T16:01:03Z foom2 is now known as foom
2017-09-12T16:06:28Z shka_ joined #sbcl
2017-09-12T16:21:21Z sjl_ quit (Ping timeout: 260 seconds)
2017-09-12T16:23:24Z rumbler3_ joined #sbcl
2017-09-12T16:29:04Z rumbler3_ quit (Ping timeout: 264 seconds)
2017-09-12T16:42:35Z stassats quit (Ping timeout: 255 seconds)
2017-09-12T17:00:00Z sjl_ joined #sbcl
2017-09-12T17:17:19Z m00natic quit (Remote host closed the connection)
2017-09-12T17:19:29Z nimbus7 quit (Quit: Leaving)
2017-09-12T17:19:36Z nimbus7 joined #sbcl
2017-09-12T17:23:43Z sjl_ quit (Quit: WeeChat 1.9)
2017-09-12T17:51:44Z sjl joined #sbcl
2017-09-12T17:55:00Z attila_lendvai joined #sbcl
2017-09-12T17:55:00Z attila_lendvai quit (Changing host)
2017-09-12T17:55:01Z attila_lendvai joined #sbcl
2017-09-12T17:57:50Z stassats joined #sbcl
2017-09-12T18:25:17Z rumbler3_ joined #sbcl
2017-09-12T18:29:19Z rumbler3_ quit (Ping timeout: 240 seconds)
2017-09-12T18:44:53Z sjl quit (Ping timeout: 240 seconds)
2017-09-12T18:55:26Z sjl joined #sbcl
2017-09-12T18:59:50Z sjl quit (Ping timeout: 255 seconds)
2017-09-12T19:05:22Z sjl joined #sbcl
2017-09-12T19:14:41Z sjl quit (Ping timeout: 255 seconds)
2017-09-12T19:16:57Z ym joined #sbcl
2017-09-12T19:33:30Z scymtym quit (Ping timeout: 255 seconds)
2017-09-12T19:41:43Z sjl joined #sbcl
2017-09-12T19:47:52Z sjl quit (Quit: WeeChat 1.9)
2017-09-12T20:05:42Z shka_ quit (Ping timeout: 248 seconds)
2017-09-12T20:10:30Z 7GHAAK86A quit (Ping timeout: 248 seconds)
2017-09-12T20:11:11Z smokeink_ joined #sbcl
2017-09-12T20:33:38Z nimbus7 quit (Ping timeout: 246 seconds)
2017-09-12T20:33:50Z nimbus7 joined #sbcl
2017-09-12T20:42:10Z scymtym joined #sbcl
2017-09-12T21:06:57Z oleo quit (Remote host closed the connection)
2017-09-12T21:09:49Z milanj quit (Quit: This computer has gone to sleep)
2017-09-12T21:15:19Z angavrilov quit (Remote host closed the connection)
2017-09-12T21:32:04Z Bike quit (Ping timeout: 264 seconds)
2017-09-12T22:05:24Z scymtym: attila_lendvai: i was going to fix a problem in INSPECT and stumbled upon an apparent problem with *SUPPRESS-PRINT-ERRORS*. do you have a minute to discuss this?
2017-09-12T22:07:40Z attila_lendvai: scymtym: yep, I'm here
2017-09-12T22:09:16Z attila_lendvai greps around to remind himself
2017-09-12T22:10:06Z scymtym: i basically wanted to use *SUPPRESS-PRINT-ERRORS* in INSPECT when printing slot values since that can signal errors (e.g. in case of user-written PRINT-OBJECT methods)
2017-09-12T22:10:43Z scymtym: this seemed to work initially, but failed in the context of the test harness
2017-09-12T22:11:33Z scymtym: the reason is that the test harness establishes a condition handler that handles error conditions (or maybe serious conditions, i don't remember)
2017-09-12T22:12:51Z scymtym: this is because the code in OUTPUT-OBJECT SIGNALs the condition to ;; Give outer handlers a chance.
2017-09-12T22:13:41Z scymtym: one second, child
2017-09-12T22:15:26Z attila_lendvai: I must note that I merely used/mimic'd this machinery in my backtrace code, but I'm trying to decypher the situation
2017-09-12T22:16:56Z attila_lendvai: that "Give outer handlers a chance." doesn't make sense to me. if print errors are suppressed, then, well, I expect them suppressed.
2017-09-12T22:17:54Z attila_lendvai: it's a neat trick to SIGNAL them, as opposed to ERROR them, but it still doesn't make sense to me
2017-09-12T22:19:17Z scymtym: a handler would have to invoke the CONTINUE restart or not handle the condition for the suppression to work
2017-09-12T22:19:33Z scymtym: that cannot be expected of arbitrary outer handlers
2017-09-12T22:20:02Z attila_lendvai: FWIW, I have a construct for such situation (possibility of errors happening while dealing with errors): https://hub.darcs.net/hu.dwim/hu.dwim.util/browse/source/error-handling.lisp#17
2017-09-12T22:20:22Z attila_lendvai: yep
2017-09-12T22:21:04Z scymtym: this behavior is probably ok for BREAK because the condition is not an error in that case
2017-09-12T22:21:41Z scymtym: i'm going to study your macro later, it looks interesting
2017-09-12T22:22:29Z scymtym: for the sbcl code, do you agree that SIGNALling the condition can be removed?
2017-09-12T22:22:50Z attila_lendvai: it's somewhat convoluted because of other stuff like logging, and thread state tracking, but it made a lot of code behave much better. many headaches involved logging and custom print-object methods triggering a nested error
2017-09-12T22:23:27Z attila_lendvai: yep, I don't see any use-case where it would be useful
2017-09-12T22:23:30Z scymtym: yeah, i think log4cl has problems with this to this day
2017-09-12T22:23:45Z attila_lendvai: and on the other hand, I see many where it's an unwelcome surprise
2017-09-12T22:24:31Z scymtym: ok, i'm going to remove that bit and fix INSPECT (closing an eight year old bug)
2017-09-12T22:24:41Z scymtym: thank you for your time
2017-09-12T22:26:43Z rumbler31 joined #sbcl
2017-09-12T22:27:17Z attila_lendvai: you're welcome, and thanks for polishing SBCL!
2017-09-12T22:31:19Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-12T22:37:06Z rpg joined #sbcl
2017-09-12T22:37:07Z attila_lendvai: scymtym: here's a relatively straightforward example of how it's used (to deal with errors coming from a transaction rollback): https://hub.darcs.net/hu.dwim/hu.dwim.rdbms/browse/source/generic/transaction.lisp#175
2017-09-12T22:39:54Z attila_lendvai has pushed some cleanups to the comments in the w-l-e-h macro
2017-09-12T22:42:37Z scymtym: i see. looks convenient. not having to syntactically layer the handlers
2017-09-12T22:48:35Z attila_lendvai: plus there are a lot of subtleties that can go wrong if you write out your own handler-bind's every time. and the default level 2+ handlers are suitable most of the time. I remember messing around with it for quite a while until it settled, but then it just worked.
2017-09-12T23:00:30Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-12T23:05:46Z scymtym: wow. the tests concerned with suppressing print errors were pretty broken
2017-09-12T23:12:23Z dougk quit (Ping timeout: 248 seconds)
2017-09-12T23:13:37Z DeadTrickster_ joined #sbcl
2017-09-12T23:15:31Z dougk joined #sbcl
2017-09-12T23:15:54Z DeadTrickster quit (Quit: Leaving)
2017-09-12T23:16:15Z DeadTrickster joined #sbcl
2017-09-12T23:17:28Z DeadTrickster quit (Client Quit)
2017-09-12T23:17:48Z DeadTrickster joined #sbcl
2017-09-12T23:20:10Z libre-man quit (Ping timeout: 260 seconds)
2017-09-12T23:21:26Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-12T23:31:33Z smokeink_ quit (Ping timeout: 248 seconds)
2017-09-12T23:32:44Z libre-man joined #sbcl
2017-09-12T23:34:45Z rumbler31 joined #sbcl
2017-09-12T23:34:48Z Bike joined #sbcl
2017-09-12T23:40:28Z Xof quit (Ping timeout: 264 seconds)
2017-09-12T23:48:39Z fiddlerwoaroof quit (Ping timeout: 240 seconds)
2017-09-12T23:53:10Z fiddlerwoaroof joined #sbcl
2017-09-12T23:55:08Z rumbler31 quit (Remote host closed the connection)
2017-09-12T23:59:49Z jibanes quit (Ping timeout: 248 seconds)
2017-09-13T00:01:42Z jibanes joined #sbcl
2017-09-13T00:04:45Z attila_lendvai quit (Quit: Leaving.)
2017-09-13T00:11:39Z libre-man quit (Ping timeout: 240 seconds)
2017-09-13T00:16:22Z smokeink joined #sbcl
2017-09-13T00:19:33Z Jesin joined #sbcl
2017-09-13T00:22:02Z Jesin: To get sbcl 1.3.21 to build, I had to do this: sed -Ei 's/\bstruct\s+ucontext\b/ucontext_t/g' src/runtime/*-os.[ch]
2017-09-13T00:23:16Z Jesin: so far I have found 0 documentation of "struct ucontext" ever existing; it's always been "ucontext_t"
2017-09-13T00:25:54Z libre-man joined #sbcl
2017-09-13T00:26:51Z milanj joined #sbcl
2017-09-13T00:27:16Z nimbus7 quit (Ping timeout: 240 seconds)
2017-09-13T00:54:30Z smokeink quit (Quit: leaving)
2017-09-13T00:55:41Z rumbler31 joined #sbcl
2017-09-13T01:00:30Z rumbler31 quit (Ping timeout: 260 seconds)
2017-09-13T01:00:30Z stassats quit (Ping timeout: 260 seconds)
2017-09-13T01:05:59Z libre-man quit (Ping timeout: 240 seconds)
2017-09-13T01:21:31Z libre-man joined #sbcl
2017-09-13T01:41:25Z dougk:  Jesin: of course it existed. See https://patchwork.kernel.org/patch/9815493/
2017-09-13T01:43:09Z cromachina joined #sbcl
2017-09-13T01:43:22Z Jesin: dougk: oh ok, thanks. Still, it seems it was standardized with the name "ucontext_t" a while back, and my system does not have "struct ucontext"
2017-09-13T01:45:39Z dougk: sure, that's https://bugs.launchpad.net/sbcl/+bug/1714549.
2017-09-13T01:47:01Z libre-man quit (Ping timeout: 248 seconds)
2017-09-13T02:16:14Z Bike quit (Ping timeout: 255 seconds)
2017-09-13T02:35:11Z rumbler31 joined #sbcl
2017-09-13T02:40:39Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-13T02:43:28Z jdz quit (Ping timeout: 264 seconds)
2017-09-13T02:46:07Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-13T02:48:38Z jdz joined #sbcl
2017-09-13T03:01:53Z dougk quit (Ping timeout: 240 seconds)
2017-09-13T03:07:09Z dougk joined #sbcl
2017-09-13T03:31:25Z rumbler31 joined #sbcl
2017-09-13T03:46:26Z milanj quit (Quit: This computer has gone to sleep)
2017-09-13T04:38:27Z rumbler31 quit (Remote host closed the connection)
2017-09-13T04:39:10Z rumbler31 joined #sbcl
2017-09-13T04:39:10Z rumbler31 quit (Remote host closed the connection)
2017-09-13T04:56:56Z shka_ joined #sbcl
2017-09-13T05:09:35Z cromachina quit (Ping timeout: 260 seconds)
2017-09-13T05:21:45Z cromachina joined #sbcl
2017-09-13T05:31:41Z milanj joined #sbcl
2017-09-13T05:35:24Z milanj quit (Client Quit)
2017-09-13T05:43:48Z cromachina quit (Read error: Connection reset by peer)
2017-09-13T05:44:53Z cromachina joined #sbcl
2017-09-13T05:45:39Z angavrilov joined #sbcl
2017-09-13T06:07:53Z scymtym quit (Ping timeout: 240 seconds)
2017-09-13T06:39:39Z rumbler31 joined #sbcl
2017-09-13T06:45:38Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-13T07:16:10Z shka_ quit (Ping timeout: 260 seconds)
2017-09-13T07:19:13Z DeadTrickster_ joined #sbcl
2017-09-13T07:26:26Z scymtym joined #sbcl
2017-09-13T07:29:34Z malm quit (Quit: Bye bye)
2017-09-13T07:29:58Z malm joined #sbcl
2017-09-13T08:03:50Z carenz joined #sbcl
2017-09-13T08:04:37Z libre-man joined #sbcl
2017-09-13T08:48:36Z edgar-rft quit (Quit: edgar-rft)
2017-09-13T08:58:11Z scymtym: dougk: lots of crashes "GC invariant lost, file "marknsweepgc.c", line 2259" on x86_64 when dumping an image. seems to have started with commit "x86-64: Save absolute assembler routine fixups"
2017-09-13T08:59:01Z dougk quit (Ping timeout: 248 seconds)
2017-09-13T09:18:20Z pchrist_ is now known as pchrist
2017-09-13T09:53:37Z milanj joined #sbcl
2017-09-13T09:55:37Z dougk joined #sbcl
2017-09-13T09:59:49Z dougk quit (Ping timeout: 255 seconds)
2017-09-13T10:28:01Z m00natic joined #sbcl
2017-09-13T10:40:24Z Bike joined #sbcl
2017-09-13T10:43:42Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-13T10:56:33Z dougk joined #sbcl
2017-09-13T11:01:01Z dougk quit (Ping timeout: 255 seconds)
2017-09-13T11:07:33Z stassats joined #sbcl
2017-09-13T11:12:06Z stassats: GC invariant lost, file "marknsweepgc.c", line 2259
2017-09-13T11:18:06Z Bicyclidine joined #sbcl
2017-09-13T11:19:59Z stassats: but gdb paints a different picture
2017-09-13T11:21:26Z stassats: oh, that's from the ldb backtrace
2017-09-13T11:21:34Z Bike quit (Ping timeout: 246 seconds)
2017-09-13T11:25:43Z stassats: it encounters a code header, not a simple fun
2017-09-13T11:31:01Z carenz quit (Ping timeout: 248 seconds)
2017-09-13T11:37:19Z scymtym: there is also this macos build failure since bf2e122e: http://paste.lisp.org/display/355643
2017-09-13T11:45:03Z stassats: the address is 0x21b00000
2017-09-13T11:45:22Z stassats: so, that's the assembly routines
2017-09-13T11:45:57Z stassats: they are all packed into a single code object
2017-09-13T11:46:21Z carenz joined #sbcl
2017-09-13T11:48:47Z stassats: they do not really have a function header, do they?
2017-09-13T11:54:01Z rumbler31 joined #sbcl
2017-09-13T11:57:54Z rumbler31 quit (Remote host closed the connection)
2017-09-13T11:59:22Z dougk joined #sbcl
2017-09-13T12:11:20Z DeadTrickster quit (Ping timeout: 252 seconds)
2017-09-13T12:17:18Z m00natic quit (Remote host closed the connection)
2017-09-13T12:28:30Z m00natic joined #sbcl
2017-09-13T12:37:27Z Bicyclidine quit (Ping timeout: 264 seconds)
2017-09-13T12:46:17Z m00natic quit (Remote host closed the connection)
2017-09-13T12:48:02Z dougk quit (Ping timeout: 255 seconds)
2017-09-13T12:49:05Z rumbler31 joined #sbcl
2017-09-13T12:59:23Z rgrau joined #sbcl
2017-09-13T12:59:34Z carenz quit (Ping timeout: 248 seconds)
2017-09-13T13:01:24Z m00natic joined #sbcl
2017-09-13T13:04:23Z Bike joined #sbcl
2017-09-13T13:09:48Z carenz joined #sbcl
2017-09-13T13:12:48Z dougk joined #sbcl
2017-09-13T13:21:26Z rumbler31 quit (Remote host closed the connection)
2017-09-13T13:25:03Z josemanuel[m] quit (Ping timeout: 246 seconds)
2017-09-13T13:32:05Z stassats quit (Ping timeout: 248 seconds)
2017-09-13T13:33:02Z cromachina quit (Read error: Connection reset by peer)
2017-09-13T13:42:52Z dougk quit (Ping timeout: 240 seconds)
2017-09-13T13:45:47Z josemanuel[m] joined #sbcl
2017-09-13T14:02:21Z DGASAU quit (Read error: Connection reset by peer)
2017-09-13T14:04:59Z DGASAU joined #sbcl
2017-09-13T14:31:07Z oleo joined #sbcl
2017-09-13T14:33:47Z scymtym quit (Ping timeout: 255 seconds)
2017-09-13T14:34:48Z rumbler3_ joined #sbcl
2017-09-13T14:35:43Z dougk joined #sbcl
2017-09-13T14:38:39Z dougk_: ok, i'll take a look at the GC lossage. I'm not having any build failure on macOS, either 32 or 64-bit
2017-09-13T14:39:18Z rumbler3_ quit (Ping timeout: 240 seconds)
2017-09-13T14:41:29Z dougk_: scymtym: tell me your build features for the failure; stassats: you're right, i messed up that thing. I might just add simple-fun headers to assembly code. But not necessary since I do know the range of assembly code components.
2017-09-13T15:36:29Z carenz quit (Ping timeout: 252 seconds)
2017-09-13T16:24:45Z dougk__ joined #sbcl
2017-09-13T16:26:25Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T16:26:47Z scymtym joined #sbcl
2017-09-13T16:29:55Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-13T16:37:29Z dougk joined #sbcl
2017-09-13T17:00:31Z dougk__ joined #sbcl
2017-09-13T17:01:04Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T17:03:14Z m00natic quit (Remote host closed the connection)
2017-09-13T17:07:08Z dougk_: fwiw I'm thrilled that I wrote that failing gc_assert where I did.
2017-09-13T17:08:44Z shka_ joined #sbcl
2017-09-13T17:14:20Z igajsin joined #sbcl
2017-09-13T17:32:52Z dougk__ quit (Ping timeout: 255 seconds)
2017-09-13T17:40:42Z dougk joined #sbcl
2017-09-13T17:49:08Z |3b|`: windows build failure (x8664): http://paste.lisp.org/+7MI6
2017-09-13T17:54:32Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T18:09:40Z |3b|`: looks like 27f36d4decfe336b95c8ae63c12ac4c5959671f3 "Impart consistency to accessing thread-local Lisp vars from C"
2017-09-13T18:10:43Z dougk joined #sbcl
2017-09-13T18:11:55Z |3b|`: possibly due to SetSymbolValue that didn't get renamed to write_TLS inside #ifdef LISP_FEATURE_WIN32
2017-09-13T18:12:25Z dougk: |3b|`: correct, can you try changing to "write_TLS(THRUPTION_PENDING, T, p)" and let me know ?
2017-09-13T18:12:47Z |3b|`: already building :)
2017-09-13T18:14:27Z |3b|`: got to contribs, so looks good
2017-09-13T18:14:47Z |3b|`: (locked up there, but that isn't too unusual)
2017-09-13T18:15:45Z dougk: ah, see, so my commit message was right - sometimes people used "SetSymbolValue" when they meant "SetTlSymbolValue" and vv, but they're really the same thing; and I got confused trying to unconfuse it
2017-09-13T18:17:52Z |3b|`: yeah, finished build that time
2017-09-13T18:27:23Z dougk: I attributed the fix to you. It didn't fail for me as I only test on linux, and by "test" I mean compile, never actually run w/safepoint
2017-09-13T18:41:51Z attila_lendvai joined #sbcl
2017-09-13T18:41:51Z attila_lendvai quit (Changing host)
2017-09-13T18:41:51Z attila_lendvai joined #sbcl
2017-09-13T18:53:46Z dougk_: cold-init is able to succeed on x86-64 with both immobile and dynamic spaces mapped at "wrong" addresses now.
2017-09-13T18:55:02Z shka_ quit (Ping timeout: 252 seconds)
2017-09-13T19:07:49Z dougk quit (Ping timeout: 255 seconds)
2017-09-13T19:18:19Z dougk joined #sbcl
2017-09-13T19:19:17Z Jesin quit (Quit: Leaving)
2017-09-13T19:23:06Z Jesin joined #sbcl
2017-09-13T19:23:56Z attila_lendvai quit (Quit: Leaving.)
2017-09-13T19:27:59Z scymtym: dougk_: the macos build fails for all configurations: e.g. https://ci.cor-lab.org/job/sbcl-master/2444/ . maybe this is just our slave running old software again
2017-09-13T19:32:33Z dougk_: there's no (extern-alien ) in make-target-2-load. Any chance you can see what happens if you #+nil out the call to shake-packages?
2017-09-13T19:33:49Z scymtym: i can't easily get to the machine from here but i can try that tomorrow. alternatively, if that helps at all, you should be able to download the partially build workspace at https://ci.cor-lab.org/job/sbcl-master/featureset=1,label=MAC_OS_mavericks_64bit/ws/*zip*/MAC_OS_mavericks_64bit.zip
2017-09-13T19:35:20Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T19:36:10Z dougk joined #sbcl
2017-09-13T19:48:04Z dim quit (Ping timeout: 248 seconds)
2017-09-13T20:04:52Z swflint quit (Ping timeout: 240 seconds)
2017-09-13T20:13:13Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T20:27:48Z attila_lendvai joined #sbcl
2017-09-13T20:28:25Z attila_lendvai quit (Client Quit)
2017-09-13T20:39:00Z angavrilov quit (Remote host closed the connection)
2017-09-13T20:40:20Z dougk joined #sbcl
2017-09-13T20:47:09Z |3b|`: (let ((a (delete-duplicates b))) (declare (ignore a))) -> ; caught STYLE-WARNING: The return value of DELETE-DUPLICATES should not be discarded.
2017-09-13T20:47:39Z |3b|` isn't sure that really applies there
2017-09-13T20:48:45Z Shinmera: Why should it not apply? It is effectively discarded.
2017-09-13T20:49:00Z |3b|`: sure, but code-wise it isn't
2017-09-13T20:49:24Z dougk_: yes it is, you are nowhere capture the effect of what DELETE did
2017-09-13T20:49:47Z dougk_: it's saying that you need to *actually* put its return value somewhere, not just pretend to
2017-09-13T20:49:48Z |3b|`: actually, i guess better with ignorable or no declatation
2017-09-13T20:50:13Z |3b|` just added that to get rid of the extra warning, real code just had a dead variable
2017-09-13T20:50:54Z |3b|`: and the warning for that was appropriate, and covered both problems
2017-09-13T20:50:57Z swflint joined #sbcl
2017-09-13T20:51:25Z |3b|` reads it as saying "don't call delete-duplicates for side effects" rather than "store the result and use it"
2017-09-13T20:51:46Z |3b|`: or rather that's my expectation of what it should mean
2017-09-13T20:52:01Z |3b|` supposes it actually reads as the latter
2017-09-13T20:52:49Z |3b|`: discarding the result is fine per spec, as long as you also don't have any expectations of the contents of input list
2017-09-13T20:53:09Z |3b|`: contents/structure
2017-09-13T20:54:15Z Shinmera: Well, it's a style-warning.
2017-09-13T20:56:07Z |3b|`: right, not non-conformant, just seems less than obvious what it means in that case, since it might not be obviously discarded in more complex code (aside from the possible unused variable warning without IGNORABLE)
2017-09-13T20:59:52Z dougk quit (Ping timeout: 255 seconds)
2017-09-13T21:01:49Z dougk joined #sbcl
2017-09-13T21:17:39Z dougk quit (Ping timeout: 264 seconds)
2017-09-13T21:27:03Z dougk joined #sbcl
2017-09-13T21:33:27Z oleo quit (Quit: irc client terminated!)
2017-09-13T21:34:25Z dougk quit (Ping timeout: 246 seconds)
2017-09-13T21:46:22Z Bike quit (Ping timeout: 240 seconds)
2017-09-13T21:51:15Z igajsin quit (Ping timeout: 264 seconds)
2017-09-13T21:58:59Z dim joined #sbcl
2017-09-13T22:20:40Z Bike joined #sbcl
2017-09-13T23:10:55Z DeadTrickster joined #sbcl
2017-09-13T23:24:51Z jamtho joined #sbcl
2017-09-13T23:55:12Z Bicyclidine joined #sbcl
2017-09-13T23:57:22Z Bike quit (Ping timeout: 240 seconds)
2017-09-13T23:58:38Z dougk joined #sbcl
2017-09-14T00:03:38Z jamtho quit (Ping timeout: 240 seconds)
2017-09-14T00:09:37Z cromachina joined #sbcl
2017-09-14T00:10:05Z rgrau quit (Ping timeout: 246 seconds)
2017-09-14T00:21:22Z rumbler31 joined #sbcl
2017-09-14T00:25:57Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-14T01:23:42Z stassats joined #sbcl
2017-09-14T01:43:17Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T01:55:32Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-14T01:57:44Z stassats: dougk_: i think i would want fun headers for assembly routines anyway, for tail calling assembly routines
2017-09-14T01:59:15Z stassats: and fewer vops that just perform function calls
2017-09-14T02:20:58Z dougk joined #sbcl
2017-09-14T02:22:46Z rumbler31 joined #sbcl
2017-09-14T02:28:06Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-14T02:55:47Z Jesin quit (Quit: Leaving)
2017-09-14T02:59:21Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-14T03:01:16Z Jesin joined #sbcl
2017-09-14T03:06:42Z stassats: so, sweep() is sweeping the linkage table?
2017-09-14T03:13:30Z jack_rabbit joined #sbcl
2017-09-14T03:35:28Z stassats: that's just error reporting, apparently
2017-09-14T03:35:39Z stassats: so, in sweep, NOTE_GARBAGE(page_table[find_page_index(where)].gen, where, nwords, zeroed);
2017-09-14T03:35:44Z stassats: find_page_index(where) returns -1
2017-09-14T03:38:53Z stylewarning: If I have foreign pointers wrapped in a structure, is there a way I can recreate these objects if I save an image with SLAD? Is there an extensible way to do this without adding additional logic?
2017-09-14T03:39:35Z stassats: no
2017-09-14T03:40:10Z stylewarning: Basically I want to ship a library that uses foreign libs for linear algebra, and I want to make the use of pointers underneath opaque. Users have said this has caused issues when they would DEFPARAMETER these and SLAD. It makes sense why, and I'm trying to think of ways to ameliorate the situation.
2017-09-14T03:49:11Z stylewarning: Any ideas that don't require modifying the DEFPARAMETERs would be appreciated.
2017-09-14T03:49:18Z stassats: there is no way
2017-09-14T03:50:26Z stassats: restructure your program in a way that does not save pointers before dumping
2017-09-14T03:51:28Z stylewarning: Basically you seem to be saying "don't make Lisp structures hide foreign data opaquely"
2017-09-14T03:51:39Z stassats: no
2017-09-14T03:52:05Z stassats: what does "opaquely" even mean in this context?
2017-09-14T03:52:53Z stylewarning: It means the library has a structure which, as far as the user is concerned, is a plain old Lisp object
2017-09-14T03:53:18Z stylewarning: The pointer is never accessed outside of the confines of the library. It's existence isn't even published.
2017-09-14T03:53:24Z stassats: that's not a technical term
2017-09-14T03:53:33Z stassats: that's your library term
2017-09-14T03:53:57Z stylewarning: Transparent and opaque are pretty well understood technical terms when describing abstractions.
2017-09-14T03:54:17Z stassats: everything is transparent in lisp
2017-09-14T03:55:10Z stassats: it doesn't matter how you hide your pointer, you can't save an image with it
2017-09-14T03:56:49Z stylewarning: Like, one unsatisfactory solution is to save all objects in a weak table, do the equivalent of a load form, and reconstruct them upon next access.
2017-09-14T03:57:07Z stylewarning: all objects under consideration, of course.
2017-09-14T03:57:50Z stassats: just don't bother
2017-09-14T03:58:58Z stylewarning: The only other alternative is to get these pointers out entirely and attempt to use Lisp objects and pass pointers into those to the foreign libs.
2017-09-14T03:59:50Z stassats: it's well know that saving images doesn't work with foreign code, your users should understand that
2017-09-14T04:00:19Z stassats: and any foreign initialization should happen after start up, not before saving
2017-09-14T04:01:10Z stylewarning: The user just sees (DEFPARAMETER *id* (make-identity-matrix))
2017-09-14T04:02:06Z stylewarning: Maybe I'm just in some sort of la-la land to imagine the possibility of providing a way to save and load that object. Maybe that goes against the principle of SLAD too much.
2017-09-14T04:02:10Z stassats: and knows that he can't do that. cause he's a good user and reads documentation
2017-09-14T04:05:47Z stylewarning: I'll get creative I guess. That's a very unacceptable answer. Perhaps I'll eventually find your answer is wise, and that SBCL's facilities are just unfortunately occasionally hostile.
2017-09-14T04:06:21Z stassats: uhm, no lisp is going to save your foreign memory
2017-09-14T04:07:12Z stylewarning: I'm not asking to save foreign memory. Is that not clear?
2017-09-14T04:07:17Z stassats: dougk: if make-target-2 survives sweep, (sb-ext:gc :gen 7) breaks in slime
2017-09-14T04:07:30Z stassats: stylewarning: then what are you asking?
2017-09-14T04:08:31Z stylewarning: I am asking for a way to save these objects. I can take any of these objects and make a load form for it. Why can't that type of logic exist with image saving?
2017-09-14T04:08:57Z stassats: so, you are asking to save foreign memory
2017-09-14T04:09:46Z stylewarning: I am not asking for Lisp to understand what's located at the pointer automatically, and attempt to recover it automatically.
2017-09-14T04:10:07Z stassats: then what are you asking?
2017-09-14T04:11:45Z stylewarning: I am asking for a way for me to tell Lisp how to do that, and to execute those actions—as specified by me—upon saving and/or loading the image.
2017-09-14T04:12:39Z stassats: to do what exactly?
2017-09-14T04:13:27Z stassats: anything? knock yourself out: sb-ext:*save-hooks* and sb-ext:*init-hooks*
2017-09-14T04:14:26Z stylewarning: To shift the responsibility of saving objects from the library used to the library vendor.
2017-09-14T04:14:42Z stylewarning: Saving and subsequent reinitializing *
2017-09-14T04:15:10Z stassats: well, you're talking about some abstract end goals, can't help you with those
2017-09-14T04:15:12Z stylewarning: The save and init hooks might be good enough. I'll look into those.
2017-09-14T04:16:44Z stassats: but i advise against doing that anyway
2017-09-14T04:16:57Z stylewarning: Any reason why?
2017-09-14T04:17:14Z stassats: simplicity
2017-09-14T04:17:36Z stassats: it's a library, the user should deal with that
2017-09-14T04:19:31Z stylewarning: I'd rather not have some folks who are trying to do a little linear algebra have to figure out how to fix their program breaking when they try to make a "simple executable". Unfortunately the world isn't filled with advanced users of Lisp.
2017-09-14T04:19:49Z shka_ joined #sbcl
2017-09-14T04:19:52Z stassats: well, there's documentation for a reason
2017-09-14T04:20:32Z stylewarning: Programming the human is usually more expensive and error prone than programming the computer. :)
2017-09-14T04:21:00Z stylewarning: I'd rather document to the advanced user "you may disable these hooks at your peril"
2017-09-14T04:21:02Z stassats: then they are in the wrong field
2017-09-14T04:21:20Z stylewarning: Some folks have to program as a means to an end for their main objective
2017-09-14T04:21:37Z stassats: and now your objects are more expensive for everyone
2017-09-14T04:22:03Z stylewarning: Why? Expensive to save and load from an image?
2017-09-14T04:22:13Z stassats: instantiate
2017-09-14T04:22:49Z stylewarning: Why can't I just look for them in the image later? Can't I just map across all objects and look for the ones I care about?
2017-09-14T04:23:38Z stylewarning: They're saving an image. Who cares if that process is not the pinnacle of efficiency?
2017-09-14T04:24:28Z rumbler31 joined #sbcl
2017-09-14T04:24:32Z stassats: well, now you're deep into implementation dependent territory
2017-09-14T04:25:13Z stylewarning: I'm using init-hook. That's already dependent.
2017-09-14T04:25:57Z stassats: it's a supported extension
2017-09-14T04:28:37Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-14T04:34:57Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T04:35:52Z dougk joined #sbcl
2017-09-14T04:49:24Z dougk quit (Ping timeout: 248 seconds)
2017-09-14T05:26:21Z luis quit (Ping timeout: 240 seconds)
2017-09-14T05:29:32Z luis` joined #sbcl
2017-09-14T05:46:16Z dougk joined #sbcl
2017-09-14T05:50:52Z dougk quit (Ping timeout: 255 seconds)
2017-09-14T05:57:40Z stassats quit (Ping timeout: 248 seconds)
2017-09-14T05:59:23Z scymtym quit (Ping timeout: 260 seconds)
2017-09-14T06:01:10Z angavrilov joined #sbcl
2017-09-14T06:16:14Z igajsin joined #sbcl
2017-09-14T06:23:03Z cromachina quit (Read error: Connection reset by peer)
2017-09-14T06:25:14Z rumbler31 joined #sbcl
2017-09-14T06:25:15Z galdor2 is now known as galdor
2017-09-14T06:27:55Z cromachina joined #sbcl
2017-09-14T06:29:37Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-14T06:50:32Z carenz joined #sbcl
2017-09-14T07:23:37Z shka_ quit (Ping timeout: 240 seconds)
2017-09-14T07:31:07Z Autcrock joined #sbcl
2017-09-14T07:36:50Z scymtym joined #sbcl
2017-09-14T07:39:53Z Autcrock quit (Remote host closed the connection)
2017-09-14T08:06:47Z scymtym: dougk_: the macos build on our slaves succeeds with SHAKE-PACKAGES disabled. anything else i should try?
2017-09-14T08:26:00Z rumbler31 joined #sbcl
2017-09-14T08:28:36Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-14T08:30:13Z rumbler31 quit (Ping timeout: 246 seconds)
2017-09-14T08:39:13Z tonton quit (Ping timeout: 260 seconds)
2017-09-14T08:41:58Z jack_rabbit joined #sbcl
2017-09-14T09:19:14Z jamtho joined #sbcl
2017-09-14T09:28:21Z flip214: with current git HEAD, a DEFSTRUCT and a corresponding (DEFMETHOD PRINT-OBJECT) tells me that some calls couldn't be inlined because the structure was not yet defined.
2017-09-14T09:28:41Z flip214: should I use some EVAL-WHEN, or should that work?
2017-09-14T09:33:49Z tonton joined #sbcl
2017-09-14T09:56:18Z scymtym: flip214: i couldn't reproduce this. do you have a recipe for reproduction?
2017-09-14T09:57:10Z scymtym compiled a file containing (defstruct foo) (defmethod print-object ((object foo) stream))
2017-09-14T10:04:22Z flip214: well, I've got a file with these and some more, and load via asdf.
2017-09-14T10:05:57Z scymtym: flip214: can you paste the file?
2017-09-14T10:15:36Z flip214: scymtym: not necessary anymore.... my error. sorry about the noise.
2017-09-14T10:15:47Z scymtym: flip214: no worries
2017-09-14T10:18:38Z flip214: thank you very much, though!
2017-09-14T10:19:58Z milanj quit (Quit: This computer has gone to sleep)
2017-09-14T10:21:11Z m00natic joined #sbcl
2017-09-14T10:22:37Z jamtho quit (Ping timeout: 240 seconds)
2017-09-14T10:25:42Z milanj joined #sbcl
2017-09-14T10:26:42Z rumbler31 joined #sbcl
2017-09-14T10:30:57Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-14T10:57:27Z Bike joined #sbcl
2017-09-14T11:14:54Z jamtho joined #sbcl
2017-09-14T11:19:31Z jamtho quit (Client Quit)
2017-09-14T11:31:36Z ryan_vw_ joined #sbcl
2017-09-14T11:31:53Z ryan_vw quit (Ping timeout: 260 seconds)
2017-09-14T11:51:44Z dougk joined #sbcl
2017-09-14T12:27:32Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-14T12:27:34Z rumbler31 joined #sbcl
2017-09-14T12:28:01Z DeadTrickster joined #sbcl
2017-09-14T12:29:28Z epipping joined #sbcl
2017-09-14T12:31:37Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-14T12:32:20Z igajsin quit (Ping timeout: 255 seconds)
2017-09-14T12:33:36Z Bike quit (Ping timeout: 255 seconds)
2017-09-14T12:37:07Z epipping quit
2017-09-14T12:44:10Z stassats joined #sbcl
2017-09-14T12:46:21Z stassats: when NOTE_GARBAGE is taken out both make-target-2 and (gc :gen 7) in slime succeed
2017-09-14T12:56:08Z Bike joined #sbcl
2017-09-14T13:12:13Z myrkraverk quit (Ping timeout: 260 seconds)
2017-09-14T13:35:01Z cromachina quit (Quit: Leaving)
2017-09-14T14:17:39Z dougk: how is note_garbage killing it?
2017-09-14T14:17:55Z stassats: -1 from find_page_index
2017-09-14T14:20:28Z carenz quit (Ping timeout: 260 seconds)
2017-09-14T14:20:40Z dougk: that drops out of log_sweep isn't on though
2017-09-14T14:20:47Z dougk: i mean yes, it shouldn't access an array with a negative index
2017-09-14T14:20:54Z stassats: it doesn't drop out
2017-09-14T14:21:07Z dougk: ah, so why it it scanning a page not in dynamic space ?
2017-09-14T14:21:27Z stassats: not updated page tables?
2017-09-14T14:21:38Z dougk: ah, because varyobj space is in there
2017-09-14T14:21:47Z dougk: ok, that'a a bug, i'll fix that
2017-09-14T14:22:31Z stassats: why do you go through zeroed and then update totals? can't you just increase totals directly?
2017-09-14T14:23:47Z dougk: yes, could. I think it used to do something differently for which that made sense
2017-09-14T14:23:56Z dougk: will change that as well
2017-09-14T14:25:17Z stassats: interestingly, to diagnose it i disabled -O2, that helped, but in the cases when it did not crash it did crash later during slad
2017-09-14T14:25:24Z stassats: does slad not work without -O2?
2017-09-14T14:25:34Z stassats: that'd be pretty strange
2017-09-14T14:26:46Z dougk: pretty. and another thing, sweep seems to clobber way too much on precise platforms. totally unusable
2017-09-14T14:27:11Z dougk: i wanted to diagnose it using 'traceroot' but that doesn't work on precise platforms either
2017-09-14T14:27:22Z stassats: what is the fullcgc doing exactly?
2017-09-14T14:27:45Z dougk: just bzeroing garbage. generation 6 is the problem. whether it bzeros younger is irrelevant
2017-09-14T14:27:59Z dougk: but it uses only static space and stacks as the root of the scan
2017-09-14T14:29:13Z stassats: isn't gencgc zeroing enough garbage?
2017-09-14T14:30:06Z dougk: it never zeros garbage in generation 6, ever.
2017-09-14T14:30:46Z stassats: even during slad?
2017-09-14T14:31:00Z stassats: so, fullcgc just zeroes without compacting?
2017-09-14T14:31:01Z dougk: generation 6 is demoted to 5 prior to slad
2017-09-14T14:31:25Z dougk: yes, it just zeros without compacting so that subsequent passes have fewer spurious roots
2017-09-14T14:32:20Z foom: Seems like a good start for a future more general non-moving GC, too. :)
2017-09-14T14:33:15Z stassats: another define-static-fun gone
2017-09-14T14:33:37Z stassats: i think i can do mips, then i'm out of testing machines
2017-09-14T14:33:51Z stassats: i'd really like to get rid of it universally
2017-09-14T14:37:27Z stassats: and sparc too
2017-09-14T14:37:33Z stassats: that leaves alpha and hppa...
2017-09-14T14:39:33Z stassats: ./make.sh: time: not found
2017-09-14T14:39:35Z stassats: annoying
2017-09-14T14:43:01Z dougk quit (Ping timeout: 248 seconds)
2017-09-14T14:43:41Z dougk joined #sbcl
2017-09-14T14:50:50Z Shinmera: Why does make.sh invoke time anyway? Shouldn't that be a responsibility of the user if they want timing data?
2017-09-14T14:51:17Z stassats: useful for build logs
2017-09-14T14:52:20Z stassats: anyway, i finally fixed it
2017-09-14T14:56:53Z stassats: ugh, the runtime doesn't build on sparc
2017-09-14T14:57:02Z stassats: error: 'os_context_t {aka struct sigcontext}' has no member named 'si_mask'; did you mean 'sigc_mask'?
2017-09-14T15:00:07Z stassats: nobody uses sparc anyway, so i'll forgo #ifdefs and just make it work on my machine (ha-ha)
2017-09-14T15:01:36Z Jesin quit (Quit: Leaving)
2017-09-14T15:03:50Z stassats: /usr/bin/ld: sparc architecture of input file `sparc-assem.o' is incompatible with sparc:v9 output
2017-09-14T15:03:52Z stassats: hell
2017-09-14T15:08:19Z foom: Time to just delete the sparc build?
2017-09-14T15:08:20Z Jesin joined #sbcl
2017-09-14T15:09:00Z stassats: no
2017-09-14T15:09:43Z salva quit (Remote host closed the connection)
2017-09-14T15:12:12Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T15:17:32Z stassats: and sparc-assem.S is not compatible with v9
2017-09-14T15:21:07Z stassats: no more v8plus support in the runtime (oh no)
2017-09-14T15:21:48Z foom: I mean, there's really no point in building sbcl for a 32-bit v8 sparc.
2017-09-14T15:22:15Z stassats: and there is for alpha?
2017-09-14T15:29:44Z stassats: i wonder if i just remove static functions from hppa and alpha altogether, will it work?
2017-09-14T15:36:13Z stassats: invalid magic number in core: 0x5342434c00000f14 should have been 0x5342434c.
2017-09-14T15:36:17Z stassats: did i build a 32-bit runtime?
2017-09-14T15:39:24Z tonton quit (Ping timeout: 246 seconds)
2017-09-14T15:53:02Z Shinmera: foom: You'll probably have to fight nyef if you want to get rid of any architecture
2017-09-14T16:07:58Z stassats: getting a nice Segmentation fault
2017-09-14T16:08:06Z stassats: which means it's quite early
2017-09-14T16:09:04Z stassats: call_into_lisp () at sparc-assem.S:59 59              st      reg_ZERO, [reg_NL0+%lo(foreign_function_call_active)]
2017-09-14T16:09:05Z stassats: right
2017-09-14T16:11:19Z stassats: the problem is from sethi   %hi(foreign_function_call_active), reg_NL0
2017-09-14T16:11:26Z stassats: foreign_function_call_active is 0 for some reason
2017-09-14T16:16:59Z carenz joined #sbcl
2017-09-14T16:21:53Z stassats: ugh, but why
2017-09-14T16:48:18Z attila_lendvai joined #sbcl
2017-09-14T16:49:41Z tonton joined #sbcl
2017-09-14T16:51:26Z stassats: -fno-PIC...
2017-09-14T17:00:31Z m00natic quit (Remote host closed the connection)
2017-09-14T17:02:44Z shka_ joined #sbcl
2017-09-14T17:06:31Z Jesin quit (Quit: Leaving)
2017-09-14T17:12:14Z dougk joined #sbcl
2017-09-14T17:17:50Z attila_lendvai quit (Quit: Leaving.)
2017-09-14T17:25:33Z oleo joined #sbcl
2017-09-14T18:08:02Z oleo quit (Quit: Leaving)
2017-09-14T18:13:09Z flip214: I've given my SBCL process ~6GB of RAM, but after an RSS of 2.4GB it drops into LDB, saying "No more immobile pages available"
2017-09-14T18:13:43Z stassats: those numbers you listed have no effect on immobile pages
2017-09-14T18:14:25Z flip214: SBCL 1.3.21.139-8408fe5a7
2017-09-14T18:14:33Z flip214: okay... so, what should I tune?
2017-09-14T18:15:08Z flip214: now I tried with 10GB of dynamic space -- but as you said, stopped again at the same size.
2017-09-14T18:15:44Z flip214: --control-stack-size wouldn't help too, I guess.
2017-09-14T18:16:23Z stassats: it's fixed
2017-09-14T18:17:11Z flip214: but what would allocate in there? closures?
2017-09-14T18:17:25Z stassats: not closures
2017-09-14T18:17:27Z flip214: I'm allocating quite some data, but what should I change?
2017-09-14T18:17:32Z oleo joined #sbcl
2017-09-14T18:17:35Z stassats: code, symbols
2017-09-14T18:18:35Z flip214: ah, too many symbols? okay, that's an idea...
2017-09-14T18:19:17Z flip214: I'm currently using a package as a quick de-duplication storage, I can use a hash-table instead.
2017-09-14T18:19:49Z stassats: well, regardless of whether symbols are immobile, never use abuse packages like that
2017-09-14T18:23:56Z flip214: thanks for the hint
2017-09-14T18:24:24Z igajsin joined #sbcl
2017-09-14T18:49:44Z oleo quit (Remote host closed the connection)
2017-09-14T18:58:12Z dougk quit (Ping timeout: 246 seconds)
2017-09-14T19:07:29Z dougk joined #sbcl
2017-09-14T19:09:41Z attila_lendvai joined #sbcl
2017-09-14T19:09:41Z attila_lendvai quit (Changing host)
2017-09-14T19:09:41Z attila_lendvai joined #sbcl
2017-09-14T19:14:38Z flip214: hrmpf.
2017-09-14T19:14:50Z flip214: I get "There is no applicable method for the generic function XXX when called with arguments ( ... )"
2017-09-14T19:14:58Z flip214: but there IS such a method
2017-09-14T19:15:09Z stassats: would it lie to you?
2017-09-14T19:15:42Z flip214: well, if I try to inspect the functions symbol, I get
2017-09-14T19:15:44Z flip214: unhandled DEBUG-CONDITION:
2017-09-14T19:16:03Z flip214: #<SB-DI::BOGUS-DEBUG-FUN "foreign function: new_thread_trampoline"> has no debug variable information.
2017-09-14T19:16:26Z flip214: so I guess something is broken
2017-09-14T19:17:06Z flip214: I had that already a few hours ago... but it vanished after I removed all traces of a class (including its methods), and did a restart of my sbcl (to really clean up)
2017-09-14T19:18:41Z flip214: and (sb-mop:generic-function-methods #'...) tells me that there IS a method for that (single) argument's type
2017-09-14T19:19:04Z stassats: class, not type
2017-09-14T19:19:21Z flip214: yeah, sorry
2017-09-14T19:19:33Z flip214: it did work with less data, ie. smaller inputs
2017-09-14T19:19:54Z flip214: how would I keep the source line information for the sbcl internal function when rebuilding?
2017-09-14T19:20:46Z flip214: I already do "make.sh --fancy", so :sb-xref-for-internals should be active
2017-09-14T19:20:58Z stassats: that's not xref
2017-09-14T19:21:45Z flip214: what would I need, then?
2017-09-14T19:21:59Z stassats: nothing?
2017-09-14T19:22:39Z dougk quit (Ping timeout: 255 seconds)
2017-09-14T19:22:51Z flip214: but I'd like to keep the information about source location of eg. SB-PCL::CALL-NO-APPLICABLE-METHOD
2017-09-14T19:23:48Z stassats: it's already there?
2017-09-14T19:28:55Z edgar-rft joined #sbcl
2017-09-14T19:29:06Z flip214: sb-pcl::flush-effective-method-cache doesn't help
2017-09-14T19:30:18Z scymtym quit (Ping timeout: 255 seconds)
2017-09-14T19:32:33Z carenz quit (Ping timeout: 260 seconds)
2017-09-14T19:32:53Z dougk joined #sbcl
2017-09-14T19:34:15Z flip214: freshly loaded it works...
2017-09-14T19:35:10Z flip214: but after some heavy load (11G of 14G RAM used) sbcl doesn't find the method any more
2017-09-14T19:35:20Z flip214: although the generic function still has it listed
2017-09-14T19:41:46Z flip214: hmmm, this time sbcl complains about /proc/sys/vm/max_map_count...
2017-09-14T19:41:54Z Jesin joined #sbcl
2017-09-14T20:01:23Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T20:07:26Z oleo joined #sbcl
2017-09-14T20:15:29Z Jesin quit (Quit: Leaving)
2017-09-14T20:16:30Z scymtym joined #sbcl
2017-09-14T20:19:57Z dougk joined #sbcl
2017-09-14T20:27:16Z shka_ quit (Ping timeout: 240 seconds)
2017-09-14T20:29:12Z dougk quit (Ping timeout: 246 seconds)
2017-09-14T20:29:39Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-14T20:32:48Z dougk joined #sbcl
2017-09-14T20:39:16Z dougk quit (Ping timeout: 246 seconds)
2017-09-14T20:45:55Z dougk joined #sbcl
2017-09-14T20:52:18Z dougk quit (Ping timeout: 246 seconds)
2017-09-14T20:52:28Z dougk joined #sbcl
2017-09-14T21:00:25Z Autcrock joined #sbcl
2017-09-14T21:12:56Z dougk quit (Ping timeout: 246 seconds)
2017-09-14T21:19:56Z Bike quit (Ping timeout: 240 seconds)
2017-09-14T21:23:06Z angavrilov quit (Remote host closed the connection)
2017-09-14T21:32:54Z milanj quit (Quit: This computer has gone to sleep)
2017-09-14T21:46:38Z Autcrock quit (Remote host closed the connection)
2017-09-14T21:49:13Z dougk joined #sbcl
2017-09-14T21:50:16Z Bike joined #sbcl
2017-09-14T21:52:03Z milanj joined #sbcl
2017-09-14T22:08:50Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T22:24:42Z dougk joined #sbcl
2017-09-14T22:25:53Z abbe quit (Ping timeout: 255 seconds)
2017-09-14T22:28:16Z abbe joined #sbcl
2017-09-14T22:35:41Z dougk quit (Ping timeout: 240 seconds)
2017-09-14T22:44:17Z attila_lendvai quit (Read error: No route to host)
2017-09-14T22:46:33Z dougk joined #sbcl
2017-09-14T22:58:54Z cromachina joined #sbcl
2017-09-15T01:00:57Z stassats quit (Ping timeout: 260 seconds)
2017-09-15T01:38:28Z milanj quit (Quit: This computer has gone to sleep)
2017-09-15T03:04:25Z Bike quit (Ping timeout: 255 seconds)
2017-09-15T03:33:02Z rumbler31 joined #sbcl
2017-09-15T03:37:56Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-15T04:20:50Z dougk quit (Ping timeout: 240 seconds)
2017-09-15T04:24:32Z jack_rabbit joined #sbcl
2017-09-15T04:53:37Z shka_ joined #sbcl
2017-09-15T05:15:42Z oleo quit (Quit: irc client terminated!)
2017-09-15T05:34:13Z rumbler31 joined #sbcl
2017-09-15T05:38:28Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-15T05:49:28Z angavrilov joined #sbcl
2017-09-15T05:57:49Z scymtym quit (Ping timeout: 264 seconds)
2017-09-15T06:09:57Z carenz joined #sbcl
2017-09-15T07:26:18Z milanj joined #sbcl
2017-09-15T07:34:56Z rumbler31 joined #sbcl
2017-09-15T07:39:16Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-15T07:49:10Z scymtym joined #sbcl
2017-09-15T09:35:49Z rumbler31 joined #sbcl
2017-09-15T09:39:56Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-15T10:25:28Z attila_lendvai joined #sbcl
2017-09-15T10:25:54Z Bike joined #sbcl
2017-09-15T10:37:44Z shka_ quit (Quit: Konversation terminated!)
2017-09-15T11:10:44Z stassats joined #sbcl
2017-09-15T11:14:29Z hlavaty joined #sbcl
2017-09-15T11:15:58Z hlavaty: hi on windoes, is there a way to get pid for run-program process, which is suitable for taskkill?  sb-ext:process-pid gives a number which doesn't seem to be windows pid
2017-09-15T11:18:04Z dougk joined #sbcl
2017-09-15T11:19:22Z jackdaniel: hlavaty: I believe that it is meant to be killed with sb-ext:process-close called on the return value of sb-ext:run-program
2017-09-15T11:19:35Z jackdaniel: from inside sbcl
2017-09-15T11:21:17Z hlavaty: jackdaniel: thanks
2017-09-15T11:22:33Z hlavaty: i hope it does kill the whole process tree
2017-09-15T11:23:49Z attila_lendvai quit (Read error: Connection reset by peer)
2017-09-15T11:24:06Z hlavaty: actually it looks like sb-ext:process-pid returns handle for which GetProcessID returns the actual pid
2017-09-15T11:25:33Z attila_lendvai joined #sbcl
2017-09-15T11:28:01Z scymtym thought PROCESS-CLOSE detaches the child process without killing it
2017-09-15T11:28:04Z hlavaty: sb-ext:process-close doesnt say anything about killing the process
2017-09-15T11:28:26Z jackdaniel: OK, sorry for confusing things then
2017-09-15T11:28:54Z scymtym: on non-win32 platforms there is PROCESS-KILL
2017-09-15T11:30:39Z hlavaty: scymtym: yeah, we are using that, but the issue is on windows
2017-09-15T11:31:08Z Bike quit (Ping timeout: 246 seconds)
2017-09-15T11:31:16Z jackdaniel: then maybe (defun terminate-win-process (pid) (sb-ext:run-program "taskkill" '(…))) ;? if pid is correct
2017-09-15T11:32:00Z hlavaty: jackdaniel: yes, but as i said initially, sb-ext:process-pid doesn't return the pid
2017-09-15T11:32:21Z Bike joined #sbcl
2017-09-15T11:32:38Z hlavaty: we'll try sb-ext:process-pid -> handle -> GetProcessId -> pid -> taskkill and see if it works
2017-09-15T11:33:15Z scymtym: hlavaty: maybe try this via sb-alien: https://msdn.microsoft.com/en-us/library/ms686714(VS.85).aspx
2017-09-15T11:34:04Z attila_lendvai quit (Read error: Connection reset by peer)
2017-09-15T11:34:31Z hlavaty: scymtym: yeah that is an option, although it doesnt seem to advertise killing the whole process subtree
2017-09-15T11:34:37Z BitPuffin|osx joined #sbcl
2017-09-15T11:35:18Z attila_lendvai joined #sbcl
2017-09-15T11:37:42Z scymtym: neither does PROCESS-KILL
2017-09-15T11:38:30Z scymtym: unless the signal is SIGTERM or SIGINT and the child process takes care of it itself
2017-09-15T11:39:07Z hlavaty: sb-ext:process-pid -> GetProcessId -> taskkill seems to work
2017-09-15T11:47:17Z hlavaty quit (Ping timeout: 260 seconds)
2017-09-15T12:15:22Z stassats: the hppa build is broken because it has an ldb instruction, the new instruction format defines functions => conflict
2017-09-15T12:15:34Z stassats: maybe %ldb plus an instruction macro named ldb
2017-09-15T12:16:51Z stassats: or there's some other way
2017-09-15T12:18:24Z stassats: there is
2017-09-15T12:18:26Z dougk: op-encoder-name can add an exception for LDB
2017-09-15T12:18:34Z stassats: yes
2017-09-15T12:19:22Z stassats: now onto   unknown type specifier: SYSTEM-AREA-POINTER
2017-09-15T12:20:18Z attila_lendvai quit (Quit: Leaving.)
2017-09-15T12:20:57Z stassats: from IMMEDIATE-CONSTANT-SC
2017-09-15T12:21:08Z stassats: bad ordering?
2017-09-15T12:23:10Z stassats: no other backend has sap as immediate
2017-09-15T12:23:33Z Bike quit (Ping timeout: 246 seconds)
2017-09-15T12:38:00Z m00natic joined #sbcl
2017-09-15T12:41:48Z stassats: failed AVER: (= (LOGAND (SB!FASL::DESCRIPTOR-BITS #) SB!VM:WIDETAG-MASK) SB!VM:SIMPLE-FUN-WIDETAG)
2017-09-15T12:41:54Z stassats: strange
2017-09-15T12:46:44Z stassats: and The variable SB!FASL::DESCRIPTOR-BITS is unbound.
2017-09-15T12:46:45Z stassats: huh
2017-09-15T12:47:23Z stassats: oh, it's a function, # confused me
2017-09-15T12:49:42Z stassats: and the widetag is 0
2017-09-15T12:53:35Z Bike joined #sbcl
2017-09-15T12:57:14Z jsnell quit (Ping timeout: 246 seconds)
2017-09-15T13:03:30Z stassats: no idea there
2017-09-15T13:09:13Z oleo joined #sbcl
2017-09-15T13:12:19Z shka joined #sbcl
2017-09-15T13:28:22Z oleo quit (Remote host closed the connection)
2017-09-15T13:29:03Z oleo joined #sbcl
2017-09-15T13:33:27Z dougk quit (Ping timeout: 260 seconds)
2017-09-15T13:38:11Z jrm quit (Quit: ciao)
2017-09-15T13:38:40Z jrm joined #sbcl
2017-09-15T13:46:16Z dougk joined #sbcl
2017-09-15T13:50:30Z dougk quit (Remote host closed the connection)
2017-09-15T13:50:44Z dougk joined #sbcl
2017-09-15T13:51:31Z cromachina quit (Read error: Connection reset by peer)
2017-09-15T13:52:29Z stassats: finally, no trace of define-static-fun
2017-09-15T13:52:47Z stassats: now, making assembly routines into functions
2017-09-15T14:08:03Z stassats: on darwin, compiling without -O2 breaks SLAD
2017-09-15T14:08:09Z carenz quit (Ping timeout: 240 seconds)
2017-09-15T14:08:09Z stassats: how on earth
2017-09-15T14:09:08Z stassats: probably worth investigating
2017-09-15T14:11:05Z stassats: during defrag_immobile_space=>set_forwarding_pointer
2017-09-15T14:15:32Z stassats: and only on darwin, works on linux even with clang
2017-09-15T14:18:55Z stassats: (i didn't want to deal with lldb)
2017-09-15T14:22:14Z stassats: Memory fault at 0x38087e8 (pc=0x38087ef
2017-09-15T14:22:38Z BitPuffin|osx quit (Ping timeout: 246 seconds)
2017-09-15T14:24:39Z stassats: frame #15: 0x0000000009aaa84f sbcl`set_forwarding_pointer(pointer=0x0000000009aaa84f, newspace_copy=165909120) at forwarding-ptr.h:55
2017-09-15T14:25:04Z stassats: why would the pointer be at the same place as PC
2017-09-15T14:32:21Z scymtym quit (Ping timeout: 246 seconds)
2017-09-15T14:37:03Z |3b|` is now known as |3b|
2017-09-15T14:37:39Z dougk quit (Ping timeout: 248 seconds)
2017-09-15T14:43:44Z oleo quit (Remote host closed the connection)
2017-09-15T14:57:24Z oleo joined #sbcl
2017-09-15T14:57:35Z stassats: i guess it's just misinterpreting the address or something
2017-09-15T15:01:18Z Jesin joined #sbcl
2017-09-15T15:10:36Z attila_lendvai joined #sbcl
2017-09-15T15:10:36Z attila_lendvai quit (Changing host)
2017-09-15T15:10:36Z attila_lendvai joined #sbcl
2017-09-15T15:25:07Z stassats: lldb can't catch sigsegv?
2017-09-15T15:38:47Z |3b| thought i'd seen it discussed before, but can't find it in the logs... how hard would it be to make sbcl generate/maintain windows-style unwind info? (doesn't have to be 100%, since i just want it for profiling rather than actual real unwinding)
2017-09-15T15:39:38Z |3b|: from what i could see, main parts would be dealing with things moving during GC (assuming functions move?) and generating the stuff for interpreting function entry/exit code... might be missing things though
2017-09-15T15:40:39Z |3b| is finding profile info with no caller info annoying (though still much better than no profile info)
2017-09-15T15:41:48Z |3b|: "2    "foreign function NtQueryInformationThread"             36.04  36.04  85.01" for example could use some more context :)
2017-09-15T15:43:58Z rpg joined #sbcl
2017-09-15T15:49:56Z Jesin quit (Quit: Leaving)
2017-09-15T15:50:47Z |3b|: also, is there any way to get the windows thread id from an sbcl thread? so far i've just been interrupting other thread and getting the thread id there, but that's pretty ugly.
2017-09-15T15:54:45Z Intensity quit (Ping timeout: 246 seconds)
2017-09-15T15:56:46Z Intensity joined #sbcl
2017-09-15T16:11:56Z stassats: ok, set_forwarding_pointer is writing into a write protected page
2017-09-15T16:14:43Z stassats: but the debugger is really confused about it
2017-09-15T16:21:06Z stassats: but why is the fault reported as being around PC
2017-09-15T16:24:02Z stassats: although -O0 is not a default but i'm sure there's an actual problem and it'll come back to hunt us
2017-09-15T16:42:36Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-15T16:46:35Z Jesin joined #sbcl
2017-09-15T16:47:27Z dougk joined #sbcl
2017-09-15T16:48:06Z stassats: forgot about SBCL_DYNDEBUG=all
2017-09-15T16:48:33Z stassats: so, it looks like the page is unprotected, the WP violation is handled but it clobbers a register or two
2017-09-15T16:48:55Z stassats: no wonder all the debuggers are as confused as i am
2017-09-15T16:53:29Z rpg joined #sbcl
2017-09-15T16:58:39Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T16:59:33Z oleo quit (Quit: Leaving)
2017-09-15T17:02:36Z BitPuffin|osx joined #sbcl
2017-09-15T17:04:45Z scymtym joined #sbcl
2017-09-15T17:07:18Z attila_lendvai quit (Quit: Leaving.)
2017-09-15T17:14:47Z dougk joined #sbcl
2017-09-15T17:24:55Z m00natic quit (Remote host closed the connection)
2017-09-15T17:37:30Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T17:50:06Z dougk_: probably just need to make everything read/write prior to defrag?
2017-09-15T17:50:55Z dougk_: and would you believe we can't enable read_protect_free_pages (in a debugging scenario) because we actually do read them? I just pushed a fix for one such bug
2017-09-15T17:51:43Z stassats: dougk_: read/write is ok, just the darwin signal emulation is probably borked
2017-09-15T17:51:59Z dougk joined #sbcl
2017-09-15T17:52:19Z stassats: it returns from a WP fault with a clobbered register, now to find out where that happens
2017-09-15T17:52:51Z dougk_: stassats: I have long suspected that darwin has exactly that situation happen
2017-09-15T17:53:55Z dougk_: not to mention it's the only one I routinely get the "Feh" lossage in the concurrency tests
2017-09-15T17:57:37Z oleo joined #sbcl
2017-09-15T18:01:55Z mfiano joined #sbcl
2017-09-15T18:03:43Z mfiano: Hello. I'm not sure if this is a bug, but I cannot reproduce on other implementation. The following is allowed on SBCL: (defclass foo () ((x :initarg :x))) (make-instance 'foo :x 1 :y 2 :allow-other-keys nil)
2017-09-15T18:04:27Z stassats: not here
2017-09-15T18:05:04Z mfiano: My distribution is behind one release it seems. This is on 1.3.20 with a default core image.
2017-09-15T18:08:43Z mfiano left #sbcl
2017-09-15T18:10:53Z attila_lendvai joined #sbcl
2017-09-15T18:13:33Z Jesin quit (Quit: Leaving)
2017-09-15T18:14:40Z oleo quit (Remote host closed the connection)
2017-09-15T18:15:12Z oleo joined #sbcl
2017-09-15T18:18:15Z stassats: doesn't seem to be anything obviously wrong with catch_exception_raise
2017-09-15T18:18:21Z stassats: could be a darwin problem
2017-09-15T18:18:27Z stassats: (hate those)
2017-09-15T18:22:34Z stassats: it's more likely that it doesn't clobber a register but a stack location
2017-09-15T18:22:39Z stassats: that would explain -O0
2017-09-15T18:24:30Z stassats: wait
2017-09-15T18:25:05Z stassats: clang doesn't extend RSP before using stack
2017-09-15T18:25:10Z stassats: well, that's not interrupt safe, duh
2017-09-15T18:27:34Z stassats: for leaf functions, i assume
2017-09-15T18:28:27Z stassats: neither does gcc?
2017-09-15T18:29:34Z stassats: and without -O2 set_forwarding_pointer doesn't get inlined, a leaf function, boom
2017-09-15T18:31:16Z stassats: altstack?
2017-09-15T18:31:51Z dougk: i thought AMD specifies a red zone of 128 bytes for free without decrementing RSP
2017-09-15T18:32:10Z stassats: ok, let's check
2017-09-15T18:35:25Z stassats: right
2017-09-15T18:40:51Z stassats: i'm so lucky even emacs crashes
2017-09-15T18:40:54Z stassats: losing all my context
2017-09-15T18:41:32Z oleo: x86-64-arch.c: In function 'context_eflags_addr':
2017-09-15T18:41:32Z oleo: x86-64-arch.c:141:44: error: dereferencing pointer to incomplete type 'os_context_t {aka struct ucontext}'
2017-09-15T18:41:48Z oleo: tried to recompile my 1.3.20 sources and got this
2017-09-15T18:41:50Z stassats: your libc is too new
2017-09-15T18:41:55Z oleo: on a new libc ya
2017-09-15T18:42:20Z oleo: now i'm getting my sources via sbopkg and it crosscompiles via clisp
2017-09-15T18:42:35Z oleo: maybe it will work.....
2017-09-15T18:42:48Z oleo: but this time it is 1.3.21
2017-09-15T18:42:54Z oleo: formerly that one failed too here
2017-09-15T18:43:22Z oleo: but then my platfrom installation was botched too.... i set it all new now
2017-09-15T18:43:30Z oleo: hope it works
2017-09-15T18:43:45Z stassats: 1.3.21 wouldn't compile either
2017-09-15T18:43:56Z oleo: oh
2017-09-15T18:44:52Z oleo: then someone has to rewrite x86-64-arch.c
2017-09-15T18:45:09Z stassats: someone has to
2017-09-15T18:45:15Z oleo: according to the needs of a newer glibc but with old compat or so ?
2017-09-15T18:47:09Z stassats: dougk: a 128-byte red zone and it works
2017-09-15T18:47:23Z stassats: half a day lost to this
2017-09-15T18:50:38Z stassats: but good thing i didn't ignore it
2017-09-15T18:52:44Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T18:54:48Z stassats: does x86 have a red zone?
2017-09-15T18:55:15Z stassats: let's see what linux does
2017-09-15T18:56:51Z foom: no, only x86-64
2017-09-15T18:56:58Z stassats: only x86-64's setup_signal_stack_si adds 128 bytes
2017-09-15T18:57:13Z stassats: ok, we're good then
2017-09-15T18:57:23Z stassats: (subtracts)
2017-09-15T18:59:21Z dougk joined #sbcl
2017-09-15T19:08:51Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T19:09:11Z dougk joined #sbcl
2017-09-15T19:18:50Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-15T19:22:29Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T19:35:08Z dougk joined #sbcl
2017-09-15T19:45:35Z dougk quit (Ping timeout: 255 seconds)
2017-09-15T19:52:04Z fiddlerwoaroof quit (Read error: Connection reset by peer)
2017-09-15T20:07:23Z dougk joined #sbcl
2017-09-15T20:08:32Z Jesin joined #sbcl
2017-09-15T20:13:56Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T20:47:25Z attila_lendvai quit (Quit: Leaving.)
2017-09-15T20:48:40Z milanj quit (Quit: This computer has gone to sleep)
2017-09-15T20:51:37Z attila_lendvai joined #sbcl
2017-09-15T21:08:48Z dougk joined #sbcl
2017-09-15T21:16:28Z dougk quit (Ping timeout: 255 seconds)
2017-09-15T21:17:01Z shka quit (Ping timeout: 246 seconds)
2017-09-15T21:20:23Z dougk joined #sbcl
2017-09-15T21:23:03Z milanj joined #sbcl
2017-09-15T21:26:49Z dougk quit (Ping timeout: 255 seconds)
2017-09-15T21:27:48Z Bike quit (Ping timeout: 255 seconds)
2017-09-15T21:27:50Z dougk joined #sbcl
2017-09-15T21:38:13Z angavrilov quit (Remote host closed the connection)
2017-09-15T21:48:40Z rpg joined #sbcl
2017-09-15T21:56:20Z stylewarning: I'm having an issue. I'm overwriting a struct slot from a foreign pointer to a Lisp object in *save-hooks*. All seemingly works well (no errors; I can inspect the changed struct). But when I do SLAD and run the resulting executable, I get an error: test-reify(10971,0x2ec3c0) malloc: *** error for object 0x7000d0: pointer being freed was not allocated
2017-09-15T21:56:49Z stylewarning: However, when I call the hook manually before SLAD (and disable it in *save-hooks*), things work fine
2017-09-15T21:58:02Z stylewarning: This is apparently before the execution of the entry point function (:TOPLEVEL)
2017-09-15T21:59:54Z stylewarning: (by "struct slot" i mean the slot of a Lisp structure object)
2017-09-15T22:06:07Z stylewarning: I think this has to do with finalizers somewhere. I do a full GC in my hook and it works.
2017-09-15T22:06:20Z dougk__ joined #sbcl
2017-09-15T22:08:02Z stylewarning: Are finalizers called / is GC done before they're deinit'd? Are finalizers called in GC-AND-SAVE?
2017-09-15T22:08:07Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T22:16:51Z Jesin quit (Quit: Leaving)
2017-09-15T22:24:57Z Bike joined #sbcl
2017-09-15T22:27:22Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-15T22:33:11Z dougk joined #sbcl
2017-09-15T22:39:55Z dougk__ joined #sbcl
2017-09-15T22:40:40Z dougk quit (Ping timeout: 246 seconds)
2017-09-15T22:44:25Z pkhuong_: finalisers
2017-09-15T22:44:40Z pkhuong_: are not called before dumping a core
2017-09-15T22:49:45Z rumbler31 joined #sbcl
2017-09-15T22:50:07Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-15T23:18:51Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-15T23:27:13Z stylewarning: the spooky thing is that they're attempted to be called after the core is saved
2017-09-15T23:27:48Z stylewarning: (at least that's what i hypothesize)
2017-09-15T23:37:23Z slyrus quit (Ping timeout: 248 seconds)
2017-09-15T23:45:23Z Jesin joined #sbcl
2017-09-16T00:24:19Z stassats quit (Ping timeout: 240 seconds)
2017-09-16T00:36:34Z karswell joined #sbcl
2017-09-16T00:43:41Z attila_lendvai quit (Quit: Leaving.)
2017-09-16T00:45:02Z Jesin quit (Quit: Leaving)
2017-09-16T00:59:21Z dougk joined #sbcl
2017-09-16T01:14:08Z dougk quit (Read error: Connection reset by peer)
2017-09-16T01:14:27Z dougk joined #sbcl
2017-09-16T01:16:01Z milanj quit (Quit: This computer has gone to sleep)
2017-09-16T01:17:39Z aeth quit (Ping timeout: 248 seconds)
2017-09-16T01:19:32Z aeth joined #sbcl
2017-09-16T01:38:15Z BitPuffin|osx quit (Ping timeout: 240 seconds)
2017-09-16T02:10:15Z cromachina joined #sbcl
2017-09-16T02:14:09Z dougk quit (Ping timeout: 248 seconds)
2017-09-16T02:16:05Z dougk joined #sbcl
2017-09-16T02:25:46Z Jesin joined #sbcl
2017-09-16T02:48:14Z karswell quit (Remote host closed the connection)
2017-09-16T03:16:38Z Bike quit (Ping timeout: 240 seconds)
2017-09-16T03:34:23Z karswell joined #sbcl
2017-09-16T03:37:21Z slyrus joined #sbcl
2017-09-16T04:18:36Z DeadTrickster_ joined #sbcl
2017-09-16T04:21:48Z DeadTrickster quit (Ping timeout: 264 seconds)
2017-09-16T05:03:31Z dougk quit (Read error: Connection reset by peer)
2017-09-16T05:03:46Z dougk joined #sbcl
2017-09-16T05:15:47Z dougk quit (Ping timeout: 264 seconds)
2017-09-16T05:39:38Z fiddlerwoaroof joined #sbcl
2017-09-16T05:51:20Z angavrilov joined #sbcl
2017-09-16T06:45:52Z shka joined #sbcl
2017-09-16T07:13:15Z attila_lendvai joined #sbcl
2017-09-16T07:13:15Z attila_lendvai quit (Changing host)
2017-09-16T07:13:15Z attila_lendvai joined #sbcl
2017-09-16T07:26:46Z milanj joined #sbcl
2017-09-16T07:43:40Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-16T07:48:17Z oleo quit (Quit: Leaving)
2017-09-16T07:57:19Z attila_lendvai joined #sbcl
2017-09-16T08:07:00Z oleo joined #sbcl
2017-09-16T08:15:01Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-16T08:34:01Z attila_lendvai joined #sbcl
2017-09-16T08:34:01Z attila_lendvai quit (Changing host)
2017-09-16T08:34:01Z attila_lendvai joined #sbcl
2017-09-16T08:43:37Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-16T09:55:39Z stassats joined #sbcl
2017-09-16T10:43:23Z stassats quit (Ping timeout: 264 seconds)
2017-09-16T10:44:11Z Bike joined #sbcl
2017-09-16T11:12:54Z scymtym quit (Ping timeout: 240 seconds)
2017-09-16T11:30:10Z scymtym joined #sbcl
2017-09-16T12:48:21Z karswell quit (Ping timeout: 240 seconds)
2017-09-16T12:48:23Z jsnell joined #sbcl
2017-09-16T12:54:06Z ym quit (Quit: Leaving)
2017-09-16T13:15:28Z dougk joined #sbcl
2017-09-16T13:16:11Z milanj quit (Quit: This computer has gone to sleep)
2017-09-16T14:13:56Z dougk quit (Read error: Connection reset by peer)
2017-09-16T14:14:11Z dougk joined #sbcl
2017-09-16T14:15:51Z whoman joined #sbcl
2017-09-16T14:18:47Z dougk quit (Ping timeout: 264 seconds)
2017-09-16T14:21:01Z dougk joined #sbcl
2017-09-16T15:41:10Z dougk quit (Read error: Connection reset by peer)
2017-09-16T15:42:10Z dougk joined #sbcl
2017-09-16T15:58:59Z whoman quit (Ping timeout: 264 seconds)
2017-09-16T16:26:09Z sjl_ joined #sbcl
2017-09-16T16:38:21Z karswell joined #sbcl
2017-09-16T16:45:34Z slyrus: so I have a rather annoying memory corruption (or other crash) that only seems to happen when running inside a docker container
2017-09-16T16:46:29Z slyrus: it's relatively reproducible and I can get an SLDB prompt when it crashes, but I'm somewhat a loss to use the debugging tools to actually figure out _why_ things go south
2017-09-16T16:46:52Z pkhuong_: slyrus: can you try and uncontainerise as many things as possible?
2017-09-16T16:47:37Z slyrus: Yes, uncontainerizing the SBCL process seems to make the problem go away
2017-09-16T16:47:40Z pkhuong_: also is sbcl running as pid 1 in the container?
2017-09-16T16:47:53Z slyrus: happens either way
2017-09-16T16:47:55Z pkhuong_: slyrus: I mean turn off namespaces one by one, in docker.
2017-09-16T16:48:08Z slyrus: turn off namespaces?
2017-09-16T16:48:29Z slyrus: the error is usually of the form "Unahndled memory fault at #xD."
2017-09-16T16:49:23Z pkhuong_: you can opt out of containerisation for features at a time, this isn't a VM, it's a process + special sauce. if you can figure out what subset of the special sauce causes the crash, finding the problem might be easier.
2017-09-16T16:49:56Z slyrus: ah
2017-09-16T16:58:06Z slyrus: running the container with --privileged makes the warning message on startup go away, which is nice, but, alas, doesn't fix the problem
2017-09-16T17:01:44Z pkhuong_: something like "--ipc=host --net=host --pid=host --uts=host --privileged --security-opt seccomp=unconfined"
2017-09-16T17:08:40Z scymtym quit (Ping timeout: 246 seconds)
2017-09-16T17:11:44Z igajsin quit (Ping timeout: 255 seconds)
2017-09-16T17:14:09Z oleo quit (Quit: Leaving)
2017-09-16T17:17:53Z oleo joined #sbcl
2017-09-16T17:19:04Z stassats joined #sbcl
2017-09-16T17:26:35Z slyrus: still crashes :(
2017-09-16T17:26:52Z oleo: slyrus ?
2017-09-16T17:26:56Z slyrus: howdy
2017-09-16T17:27:01Z oleo: hullo
2017-09-16T17:27:25Z oleo: my sbcl was cross compiled thru clisp on my new system
2017-09-16T17:27:29Z oleo: and it works now
2017-09-16T17:27:41Z oleo: tho i had to update several libs in quicklisp too
2017-09-16T17:28:06Z oleo: recursive mutex issues.....
2017-09-16T17:30:02Z slyrus: and did you try not cross-compiling to see if you could bring back the problem?!?
2017-09-16T17:32:04Z pkhuong_: slyrus: I'm pretty sure we containerised an SBCL app here.
2017-09-16T17:32:34Z pkhuong_: I'll ask and see if we hit any weirdness
2017-09-16T17:32:41Z slyrus: ok, thanks!
2017-09-16T17:35:01Z Jesin quit (Quit: Leaving)
2017-09-16T17:36:41Z Jesin joined #sbcl
2017-09-16T17:47:49Z cromachina_ joined #sbcl
2017-09-16T17:48:01Z cromachina quit (Ping timeout: 240 seconds)
2017-09-16T17:48:55Z oleo quit (Quit: Leaving)
2017-09-16T17:53:11Z igajsin joined #sbcl
2017-09-16T17:59:48Z flip214: slyrus: I remember having seen strange issues with SBCL within LXR; perhaps there is a connection.
2017-09-16T18:00:21Z flip214: please take a look at the ulimits for processes in the container, and/or cgroups settings... pasting them somewhere might help diagnose the problem.
2017-09-16T18:11:41Z oleo joined #sbcl
2017-09-16T18:13:02Z slyrus: nothing to surprising in there
2017-09-16T18:13:07Z cromachina_ quit (Read error: Connection reset by peer)
2017-09-16T18:13:29Z slyrus: max locked memory is only 64kb
2017-09-16T18:14:14Z cromachina joined #sbcl
2017-09-16T18:18:51Z milanj joined #sbcl
2017-09-16T18:28:11Z dougk__ joined #sbcl
2017-09-16T18:28:25Z dougk quit (Remote host closed the connection)
2017-09-16T18:33:23Z oleo quit (Quit: Leaving)
2017-09-16T18:45:47Z igajsin quit (Ping timeout: 264 seconds)
2017-09-16T18:49:06Z sjl_ quit (Ping timeout: 240 seconds)
2017-09-16T18:55:10Z sjl_ joined #sbcl
2017-09-16T19:36:46Z oleo joined #sbcl
2017-09-16T19:48:04Z scymtym joined #sbcl
2017-09-16T20:10:40Z adlai quit (Ping timeout: 246 seconds)
2017-09-16T20:17:06Z adlai joined #sbcl
2017-09-16T20:33:01Z adlai quit (Ping timeout: 240 seconds)
2017-09-16T20:57:50Z Jesin quit (Quit: Leaving)
2017-09-16T21:07:14Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-16T21:08:30Z dougk joined #sbcl
2017-09-16T21:15:40Z dougk quit (Read error: Connection reset by peer)
2017-09-16T21:15:55Z dougk joined #sbcl
2017-09-16T21:19:09Z adlai joined #sbcl
2017-09-16T21:28:58Z dougk quit (Ping timeout: 264 seconds)
2017-09-16T21:32:18Z Bicyclidine joined #sbcl
2017-09-16T21:33:46Z Bike quit (Ping timeout: 264 seconds)
2017-09-16T22:29:49Z carvite quit (Read error: Connection reset by peer)
2017-09-16T22:31:13Z Jesin joined #sbcl
2017-09-16T22:31:52Z angavrilov quit (Remote host closed the connection)
2017-09-17T00:17:53Z stassats quit (Ping timeout: 240 seconds)
2017-09-17T00:46:34Z dougk joined #sbcl
2017-09-17T00:52:00Z dougk quit (Ping timeout: 248 seconds)
2017-09-17T01:14:56Z dougk joined #sbcl
2017-09-17T01:19:52Z Bicyclidine quit (Remote host closed the connection)
2017-09-17T01:20:40Z Bike joined #sbcl
2017-09-17T01:44:54Z slyrus: oleo: After blaming fukamachi's web stack, I'm beginning to think the problem might be down in Djula
2017-09-17T02:19:46Z dougk__ joined #sbcl
2017-09-17T02:22:40Z dougk quit (Ping timeout: 248 seconds)
2017-09-17T02:27:53Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-17T02:57:46Z sjl_ quit (Ping timeout: 264 seconds)
2017-09-17T03:22:22Z Bike quit (Ping timeout: 264 seconds)
2017-09-17T03:28:59Z Jesin quit (Quit: Leaving)
2017-09-17T03:40:45Z milanj quit (Quit: This computer has gone to sleep)
2017-09-17T05:07:22Z shka quit (Ping timeout: 264 seconds)
2017-09-17T05:10:45Z angavrilov joined #sbcl
2017-09-17T05:13:29Z sjl_ joined #sbcl
2017-09-17T05:28:59Z slyrus: pkhuong_:  so it's a bit harder to get the crash to occur when not dockerized, but it still happens, with the same failure mode
2017-09-17T05:29:11Z slyrus: I'm thinking the docker thing is a red herring
2017-09-17T05:30:28Z slyrus: seems to happen near the djula template rendering. It's possible there's some bugs in the various libraries, but the unhandled memory fault is starting to make me suspect SBCL
2017-09-17T05:51:17Z |3b|: can you get a backtrace from debugger?
2017-09-17T05:57:18Z |3b|: if you get a "ldb>" prompt, backtrace command is (any prefix of) backtrace
2017-09-17T06:07:41Z |3b|: might also try clearing fasl cache and doing (sb-ext:restrict-compiler-policy :safety 1) before loading
2017-09-17T06:08:48Z |3b|: which may slow things down a bit, but should make it easier to detect one of the "not sbcl's fault" cases of that sort of error
2017-09-17T06:14:38Z slyrus: |3b| I can reliably get the slime debugger to catch the error, which is nice.
2017-09-17T06:15:04Z |3b|: ah, slime debugger is nicer :)
2017-09-17T06:15:10Z slyrus: and it's always (almost anyway) unhandled memory fault at #xD
2017-09-17T06:15:16Z |3b|: what is the backtrace?
2017-09-17T06:17:04Z slyrus: https://git.io/v5dSq
2017-09-17T06:18:33Z slyrus: and the last non-offending (or first offending maybe) function is my web page rendering code which calls djula:render-template* and I think that's where things go south
2017-09-17T06:19:49Z |3b|: ok, looks like it is in a function with safety 0, so turning that off (with restrict-compiler-policy or looks like you could set a var during compilation) will probably tell you what assumption it makes that isn't true
2017-09-17T06:19:57Z |3b|: in fast-http
2017-09-17T06:20:57Z slyrus: OK, trying now.
2017-09-17T06:22:26Z |3b|: actually maybe it isn't there exactly, but probably next stepis still forcing safety
2017-09-17T06:23:15Z slyrus: I thought it was there for a while (and I removed some declarations there) and fixed at least one other thing along the way, but this one is still problematic...
2017-09-17T06:23:25Z |3b|: is tumak.view your code?
2017-09-17T06:24:21Z slyrus: yes
2017-09-17T06:25:13Z slyrus: and, sure, there could be bugs in my code, but with safety 3 and speed 0, it shouldn't cause a memory fault
2017-09-17T06:26:51Z slyrus: Ok, even with restrict-compiler-policy 'safety 1 I get the same error
2017-09-17T06:26:59Z |3b|: same backtrace?
2017-09-17T06:27:23Z slyrus: one interesting thing is that I have to hit the server reasonably hard to trigger the error. maybe there's some bogus interrupt handling stuff going on?
2017-09-17T06:27:48Z slyrus: or maybe it's the sb-mop funcallable instance that djual relies on
2017-09-17T06:28:29Z slyrus: or ... who knows. if my disassembly foo was better, perhaps, I could look at my render function and see where things fall apart
2017-09-17T06:29:06Z |3b|: probably not worth looking at asm at this point
2017-09-17T06:29:39Z slyrus: why would we get a memory fault at #xD?
2017-09-17T06:30:14Z |3b|: do you get a source location for the tumak.view::render frame? (hit v on that line in sldb)
2017-09-17T06:30:41Z slyrus: yeah, it's where it calls apply #'djula:render-template*
2017-09-17T06:31:20Z slyrus: interestingly, I can't drill down into that frame without getting another memory fault
2017-09-17T06:32:31Z |3b|: the the frame in your code?
2017-09-17T06:33:53Z slyrus: yes, I just pasted that function to the gist
2017-09-17T06:36:01Z slyrus: OK, whatever is in my *template-registry* can't be printed out without triggering a memory fault.
2017-09-17T06:37:11Z |3b|: does (mapcar 'type-of (alexandria:hash-table-values *template-registry*)) fault?
2017-09-17T06:37:45Z slyrus: yes
2017-09-17T06:38:08Z |3b|: hmm
2017-09-17T06:38:57Z |3b|: is *template-registry* your code?
2017-09-17T06:39:07Z |3b|: and maintained by your code
2017-09-17T06:40:07Z |3b|: if it were only printing i'd guess print-object methods or pprint diapatch, but if type-of doesn't work, sounds more like corrupted data somewhere
2017-09-17T06:40:12Z slyrus: yes, but it's basically https://git.io/v5dSh
2017-09-17T06:40:25Z slyrus: which is the caveman2 skeleton view.lisp
2017-09-17T06:41:19Z |3b|: ah, i guess most of the interesting parts ofthat was in the code you pasted before :)
2017-09-17T06:41:49Z slyrus: sort of. why should the hash value not allow taking the type-of?
2017-09-17T06:42:25Z milanj joined #sbcl
2017-09-17T06:42:30Z |3b|: it shouldn't, which is why it sounds like bad data (as in memory corruption or something using safety 0 in bad ways
2017-09-17T06:44:56Z slyrus: OK. but we've restricted safety to 1 and the problem still occurs. also, the fact that the error doesn't always occur, but does so more frequently under heavier load makes me think that this isn't just some bug in my code somewhere.
2017-09-17T06:45:07Z |3b|: yeah
2017-09-17T06:46:23Z slyrus: oh, nice. room gives me an aver. not too surprising I guess.
2017-09-17T06:46:44Z |3b|: what does it say?
2017-09-17T06:47:52Z slyrus: failed AVER: (<= (SB-KERNEL:GET-LISP-OBJ-ADDRESS SB-VM::START) (SB-KERNEL:GET-LISP-OBJ-ADDRESS SB-VM::END))
2017-09-17T06:52:22Z slyrus: as much as a bug in SBCL seems possible, there's also the whole stray pointers from alien code possibility. and this stuff definitely uses alien code.
2017-09-17T06:53:21Z |3b|: seems a bit predictable for alien code overwriting things, but could be
2017-09-17T06:53:38Z slyrus: true
2017-09-17T06:54:03Z |3b|: is :djula-prod in *features* ?
2017-09-17T06:55:19Z slyrus: have tried it both ways, but not now
2017-09-17T06:55:59Z slyrus: and without touching the templates, that shouidn't affect things too much
2017-09-17T06:59:04Z |3b|: if the contents of *template-registry* normal before it crashes?
2017-09-17T06:59:20Z igajsin joined #sbcl
2017-09-17T06:59:22Z |3b|: and has an extry for #P"blog-entry.html" ?
2017-09-17T06:59:38Z slyrus: yes to the second part
2017-09-17T07:01:50Z slyrus: and for the first part, yes, it's a djula::compiled-template
2017-09-17T07:02:44Z slyrus: which has a metaclass of funcallable-standard-class
2017-09-17T07:07:13Z |3b|: seems odd that something about it changes
2017-09-17T07:07:25Z slyrus: indeed
2017-09-17T07:08:09Z |3b|: what sbcl version?
2017-09-17T07:08:15Z slyrus: HEAD
2017-09-17T07:08:24Z slyrus: but has happened for some time
2017-09-17T07:19:27Z Xof joined #sbcl
2017-09-17T07:20:49Z slyrus: morning Xof
2017-09-17T07:20:51Z |3b|: *template-registry* is always the global binding, right? not bound locally anywhere?
2017-09-17T07:21:27Z slyrus: correct
2017-09-17T07:31:27Z slyrus: Hmm... turning off compact-instance-header seems to fix things
2017-09-17T07:31:40Z slyrus: or at least it hasn't crashed yet, and I've been pouinding on it pretty hard
2017-09-17T07:36:23Z slyrus: OK, after heavy pounding, no crashes. Time to turn compact-instance-header back on and see if the problem reappears.
2017-09-17T07:47:19Z slyrus: bingo
2017-09-17T07:48:15Z slyrus: OK, time for bed. but it seems pretty clear that the problem only happens with :compact-instance-header
2017-09-17T07:48:24Z slyrus: thanks for helping me debug this |3b|!
2017-09-17T09:52:42Z Xof: morning slyrus
2017-09-17T09:54:32Z shka joined #sbcl
2017-09-17T10:36:07Z stassats joined #sbcl
2017-09-17T10:39:48Z Bike joined #sbcl
2017-09-17T11:12:42Z scymtym quit (Remote host closed the connection)
2017-09-17T11:19:58Z Bicyclidine joined #sbcl
2017-09-17T11:23:25Z Bike quit (Ping timeout: 255 seconds)
2017-09-17T11:36:22Z scymtym joined #sbcl
2017-09-17T11:51:50Z milanj quit (Quit: This computer has gone to sleep)
2017-09-17T12:25:38Z rumbler31 quit (Remote host closed the connection)
2017-09-17T13:16:59Z ASau joined #sbcl
2017-09-17T13:26:07Z rumbler31 joined #sbcl
2017-09-17T13:26:49Z oleo quit (Quit: Leaving)
2017-09-17T13:30:57Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-17T13:44:56Z oleo joined #sbcl
2017-09-17T13:53:31Z dougk joined #sbcl
2017-09-17T14:08:11Z rumbler31 joined #sbcl
2017-09-17T15:04:38Z dougk quit (Ping timeout: 252 seconds)
2017-09-17T15:46:49Z slyrus: dougk_: around?
2017-09-17T16:09:50Z oleo quit (Quit: Leaving)
2017-09-17T16:35:54Z slyrus: so if I add back :compact-instance-header boom, things crash. without it I can pound on the webserver indefinitely, both dockerized and not, without it crashing.
2017-09-17T16:36:17Z panji joined #sbcl
2017-09-17T16:40:09Z sjl_ quit (Quit: WeeChat 1.3)
2017-09-17T16:41:16Z sjl joined #sbcl
2017-09-17T16:55:48Z stassats: compact-instance-header may not survive the same memory corruption you're heaving without crashing
2017-09-17T16:57:52Z slyrus: possible, but it's also possible that the compact-instance-header stuff is the source of the problem
2017-09-17T16:59:37Z slyrus: in any event, the pathological behavior is always the same. A corrupt funcallable-instance.
2017-09-17T17:00:19Z slyrus: seems unlikely that some arbitrary memory corruption would manifest itself the same way everytime
2017-09-17T17:13:49Z oleo joined #sbcl
2017-09-17T17:17:18Z dougk joined #sbcl
2017-09-17T17:18:17Z panji quit (Ping timeout: 240 seconds)
2017-09-17T17:18:37Z stassats: slyrus: is the code multi threaded?
2017-09-17T17:20:06Z scymtym quit (Ping timeout: 246 seconds)
2017-09-17T17:25:19Z stassats: dougk: can't the layout-of vop lose the pointer?
2017-09-17T17:26:37Z stassats: the lifetimes should be disjoint, though
2017-09-17T17:30:40Z panji joined #sbcl
2017-09-17T17:31:04Z panji quit (Client Quit)
2017-09-17T17:32:14Z oleo quit (Quit: Leaving)
2017-09-17T17:35:03Z dougk quit (Read error: Connection reset by peer)
2017-09-17T17:35:22Z dougk joined #sbcl
2017-09-17T17:35:46Z oleo joined #sbcl
2017-09-17T17:48:26Z dougk: stassats: are you saying that it's ok or not ok?
2017-09-17T17:55:56Z stassats: ok
2017-09-17T17:56:58Z stassats: though the result target is also not packed due to lifetimes
2017-09-17T17:58:53Z dougk: right, it's specified to conflict with everything
2017-09-17T17:59:13Z stassats: so,  :target result :to :result would help, but i'd just go with :to :result
2017-09-17T18:00:04Z stassats: but LOAD-FROM-VECTOR wouldn't work
2017-09-17T18:00:56Z stassats: so, :target result ought to be removed (it's not causing anything, though)
2017-09-17T18:01:12Z dougk: right
2017-09-17T18:02:16Z ASau quit (Remote host closed the connection)
2017-09-17T18:02:40Z dougk: my original plan was to have the layout vector be an immobile object, allowing "mov result, [rax*8+address_of_constant_vector]". Never got there
2017-09-17T18:02:43Z ASau joined #sbcl
2017-09-17T18:03:45Z scymtym joined #sbcl
2017-09-17T18:03:48Z dougk: either way, this is worth less that than the issue i'm currently looking at which is the GC failing to pin some objects referenced from the stack in a very obscure case that I can reproduce but can't explain why it's a reproduction
2017-09-17T18:04:05Z dougk: "worth less fixing" i mean
2017-09-17T18:04:08Z stassats: well, something's wrong for slyrus
2017-09-17T18:04:18Z stassats: the reason i was looking at layout-of
2017-09-17T18:05:23Z dougk: i wonder if this is related to what i'm looking at.
2017-09-17T18:06:11Z dougk: i have some memory that the GC does not believe is any GC-managed space or a stack space, which points to objects that clearly would merit pinning thereby, because one of the addresses shows up in a register some time later. So it must have been a stack.  But it's not getting scanned in the stack scan
2017-09-17T18:06:54Z dougk: sounds eerily suspiciously similar to whatever unexplained crash slyrus gets, though i don't know enough of those details to say. I know only what I see, which is exactly as I said
2017-09-17T18:08:15Z dougk: anyway, 100% reproducible for me, but only by enabling my mmap fuzzer, which nobody does except me, i presume
2017-09-17T18:08:48Z dougk: however the effect of the container plus automatic heap relocation *could* mean that people are getting the effect of that
2017-09-17T18:09:08Z dougk: the test would be for slyrus to disable relocatable heap and see what happens in the container
2017-09-17T18:09:32Z Bike joined #sbcl
2017-09-17T18:11:22Z stassats: slyrus: it's a container, can you share it?
2017-09-17T18:11:29Z ASau quit (Remote host closed the connection)
2017-09-17T18:11:38Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-17T18:12:11Z ASau joined #sbcl
2017-09-17T18:19:51Z slyrus: stassats: it'
2017-09-17T18:20:05Z slyrus: s multihtreaded, but I think only one thread is handling requests
2017-09-17T18:20:28Z slyrus: stassats: sure, let me figure out how to share it
2017-09-17T18:22:30Z slyrus: dougk: happy to disable the relocatable heap. I suppose I should turn compact-instance-header back on though?
2017-09-17T18:23:03Z dougk: i'm suspicious that heap relocation is causing a problem, and not that compact-instance is
2017-09-17T18:23:45Z slyrus: ok, so just disable :relocatable-heap? rebuilding...
2017-09-17T18:23:47Z dougk: compact-instance has been in production for us for years, relocatable heap not at all. only to enable llvm sanitizers to work; and they don't, so ....
2017-09-17T18:24:11Z dougk: yes, either --without-relocatable-heap if using the command-line based option, or in customize-target-features if that
2017-09-17T18:25:08Z slyrus: oh, didn't know about the command line option. that would be easier.
2017-09-17T18:25:21Z stassats: can't work together with customize-target-features, thoguh
2017-09-17T18:25:46Z slyrus: plus I need to rebuild to get compact-instance-header back on, so, rebuildng anyway.
2017-09-17T18:26:18Z stassats: --without-relocatable-heap is a make-config option, not sbcl
2017-09-17T18:26:30Z slyrus: ah, ok.
2017-09-17T18:26:43Z igajsin quit (Remote host closed the connection)
2017-09-17T18:26:47Z stassats: or make.sh
2017-09-17T18:29:03Z Bicyclidine joined #sbcl
2017-09-17T18:29:37Z milanj joined #sbcl
2017-09-17T18:30:55Z Bike quit (Ping timeout: 255 seconds)
2017-09-17T18:49:16Z slyrus: dougk: without relocatable-heap and with compact-instance-header on I'm able to trigger a crash right away :(
2017-09-17T18:49:41Z slyrus: same failure mode. Unhandled memory fault at #xD.
2017-09-17T18:50:20Z slyrus: same unprintable funcallable-instance in the stack trace.
2017-09-17T18:50:43Z slyrus: and room gives me the same aver
2017-09-17T18:50:52Z stassats: i need a way to reproduce it
2017-09-17T18:51:36Z slyrus: OK. I'll have to figure out how to package all of this up.
2017-09-17T18:53:38Z stassats: or what code to load
2017-09-17T18:55:24Z slyrus quit (Remote host closed the connection)
2017-09-17T18:57:01Z slyrus joined #sbcl
2017-09-17T19:01:34Z irsol quit (Read error: Connection reset by peer)
2017-09-17T19:02:52Z slyrus quit (Remote host closed the connection)
2017-09-17T19:03:29Z dougk quit (Ping timeout: 260 seconds)
2017-09-17T19:03:30Z slyrus joined #sbcl
2017-09-17T19:03:40Z slyrus: argh... no PMs from beirc today.
2017-09-17T19:04:10Z stassats: did you receive mine?
2017-09-17T19:04:13Z slyrus: no :(
2017-09-17T19:04:53Z stassats: it was "the latter sounds faster, as long as it has gdb"
2017-09-17T19:05:27Z stassats: basically, i need to be able to launch sbcl with --lose-on-corruption, have it crash in some manner and attach gdb to it
2017-09-17T19:05:44Z slyrus: OK. give me a few minutes.
2017-09-17T19:06:01Z karswell_ joined #sbcl
2017-09-17T19:07:21Z karswell quit (Ping timeout: 264 seconds)
2017-09-17T19:13:37Z slyrus: ok, with --lose-on-corruption I get LDB instead of the swank debugger
2017-09-17T19:13:48Z slyrus: but in the same spot
2017-09-17T19:15:19Z slyrus: https://git.io/v5dSq
2017-09-17T19:15:31Z slyrus: stassats: it will take me an hour or so to set that up
2017-09-17T19:18:02Z stassats: that's kinda interesting
2017-09-17T19:18:26Z stassats: but i'll wait for gdb
2017-09-17T19:29:50Z irsol joined #sbcl
2017-09-17T19:51:56Z ASau quit (Remote host closed the connection)
2017-09-17T19:55:39Z ASau joined #sbcl
2017-09-17T20:00:14Z dougk joined #sbcl
2017-09-17T20:04:34Z dougk quit (Ping timeout: 240 seconds)
2017-09-17T20:07:05Z stassats: dougk: one theory: X is a funcallable instance, (funcall x) is mov rax, X then call rax, enter your trampoline, which has MOV RAX,[RIP-23] JMP [RAX-3]
2017-09-17T20:07:39Z stassats: if the GC hits after MOV RAX,[RIP-23] nothings pins the trampoline, PC doesn't think it's inside a code object
2017-09-17T20:07:58Z stassats: dougk_:
2017-09-17T20:11:28Z stassats: although pair_interior_pointer may work on funcallable instances
2017-09-17T20:12:33Z stassats: or is that gengc-only?
2017-09-17T20:13:43Z stassats: right, so, where does PC pin the code object?
2017-09-17T20:25:26Z cromachina_ joined #sbcl
2017-09-17T20:26:51Z dougk joined #sbcl
2017-09-17T20:26:51Z dougk: the aver in room happens before or after the crash?  Because room's aver is actually unsafe and wrong, but most of the time works by accident
2017-09-17T20:27:35Z dougk: the thing room is trying to assert can't safely be asserted if the allocator is using a page in which a region was opened, then closed, then opened again because there were more available bytes on the page
2017-09-17T20:28:04Z cromachina quit (Ping timeout: 260 seconds)
2017-09-17T20:28:12Z dougk: or maybe i'm thinking of map-allocated-objects
2017-09-17T20:29:27Z stassats: ok, conservative_root_p doesn't seem to be pinning trampolines in funcallable-instance
2017-09-17T20:29:37Z stassats: or fdefn, but that's presumably immobile
2017-09-17T20:30:38Z dougk: yes, those things are immobile, but i wonder if liveness is the issue
2017-09-17T20:31:01Z stassats: are funcallable-instance immobile?
2017-09-17T20:31:09Z dougk: some are, some aren't
2017-09-17T20:31:27Z angavrilov quit (Remote host closed the connection)
2017-09-17T20:31:29Z dougk: fdefn's are definitely ok because they are only referenced from jumps, which means you're in a code component whose header necessarily referenced the fdefn that you're calling through
2017-09-17T20:32:07Z dougk: the code that did the call had to be itself live.  As to funcallable-instances, the only ones with builtin trampolines should be immobile, but i believe they are live because again the code component header references the GF directly
2017-09-17T20:32:39Z stassats: i was thinking moveness, not liveness
2017-09-17T20:33:20Z shka quit (Ping timeout: 248 seconds)
2017-09-17T20:33:35Z slyrus: dougk: the aver happens after for sure, not sure about before, but I think it's clean before
2017-09-17T20:37:05Z cromachina_ quit (Read error: Connection reset by peer)
2017-09-17T20:38:09Z cromachina joined #sbcl
2017-09-17T20:38:44Z dougk: i suppose the safe thing to do, if we thought there was a liveness issue with the immobile GFs, is to consider an untagged pointer (the program counter) as a root; and I can actually do that check after everything else as evidence that it was necessary to do the check at all. so perhaps random testing can prove that it's necessary; but it would be nicer to construct an argument that it isn't
2017-09-17T20:40:50Z dougk: as to funcallable-instance-tramp, that's actually an interesting question - we're reloading RAX with the thing in the funcallable-instance, are we losing the instance itself in between the two instructions ?
2017-09-17T20:43:07Z stassats: normally it's not important, unless the funcallable-instance is holding the next instruction
2017-09-17T20:47:03Z stassats: doesn't really explain the room thing, or why PC is at zero
2017-09-17T20:47:18Z stassats: icache incoherence?
2017-09-17T20:48:17Z stassats: or the #xD fault
2017-09-17T20:48:47Z stassats: should clear up once i can reproduce the thing
2017-09-17T21:25:58Z dougk quit (Ping timeout: 255 seconds)
2017-09-17T21:56:04Z dougk joined #sbcl
2017-09-17T22:00:16Z dougk quit (Ping timeout: 240 seconds)
2017-09-17T22:14:17Z dougk joined #sbcl
2017-09-17T22:16:14Z dougk quit (Read error: Connection reset by peer)
2017-09-17T22:16:32Z dougk joined #sbcl
2017-09-17T22:59:46Z Jesin joined #sbcl
2017-09-17T23:33:03Z slyrus quit (Ping timeout: 240 seconds)
2017-09-17T23:58:04Z milanj quit (Quit: This computer has gone to sleep)
2017-09-18T00:22:47Z dougk quit (Read error: Connection reset by peer)
2017-09-18T00:23:01Z dougk joined #sbcl
2017-09-18T00:51:12Z stassats quit (Ping timeout: 240 seconds)
2017-09-18T01:13:18Z Bicyclidine quit (Ping timeout: 260 seconds)
2017-09-18T01:29:51Z Bike joined #sbcl
2017-09-18T02:20:00Z slyrus joined #sbcl
2017-09-18T02:22:12Z ASau quit (Ping timeout: 240 seconds)
2017-09-18T03:59:23Z stylewarning: What hardware are the MIPS/ARM/Alpha machines tested on?
2017-09-18T04:08:21Z |3b| 's impression is "whatever someone can get to run" for mips/alpha
2017-09-18T04:57:22Z shka joined #sbcl
2017-09-18T05:28:39Z oleo quit (Quit: Leaving)
2017-09-18T05:40:10Z slyrus quit (Remote host closed the connection)
2017-09-18T05:40:47Z slyrus joined #sbcl
2017-09-18T05:41:20Z angavrilov joined #sbcl
2017-09-18T06:07:01Z scymtym quit (Ping timeout: 246 seconds)
2017-09-18T07:13:20Z shka quit (Ping timeout: 248 seconds)
2017-09-18T07:27:34Z scymtym joined #sbcl
2017-09-18T07:39:17Z scymtym_ joined #sbcl
2017-09-18T07:40:46Z carenz joined #sbcl
2017-09-18T07:43:21Z scymtym quit (Ping timeout: 255 seconds)
2017-09-18T08:00:58Z milanj joined #sbcl
2017-09-18T08:13:46Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-18T08:46:59Z DeadTrickster joined #sbcl
2017-09-18T09:04:44Z dougk_ quit (Ping timeout: 246 seconds)
2017-09-18T09:16:34Z dougk_ joined #sbcl
2017-09-18T10:12:36Z m00natic joined #sbcl
2017-09-18T10:33:05Z stassats joined #sbcl
2017-09-18T10:34:35Z stassats: stylewarning: i find your grouping of mips/arm together with alpha amusing
2017-09-18T10:34:43Z stassats: mips and arm are everywhere
2017-09-18T10:40:52Z Bicyclidine joined #sbcl
2017-09-18T10:42:28Z Bike quit (Ping timeout: 246 seconds)
2017-09-18T12:18:52Z BitPuffin|osx joined #sbcl
2017-09-18T12:20:01Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-18T12:48:43Z slyrus: morning
2017-09-18T12:54:46Z Bike joined #sbcl
2017-09-18T13:31:26Z dougk quit (Ping timeout: 248 seconds)
2017-09-18T13:42:14Z cromachina quit (Read error: Connection reset by peer)
2017-09-18T13:46:54Z foom2 joined #sbcl
2017-09-18T13:49:56Z foom quit (Ping timeout: 264 seconds)
2017-09-18T14:12:52Z stassats: so, looking at that corrupted funcallable-instance
2017-09-18T14:13:12Z stassats: the whole header is 48.
2017-09-18T14:14:07Z stassats: that does look like a liveness problem, but it's being referenced from the heap
2017-09-18T14:15:55Z foom2 is now known as foom
2017-09-18T14:28:24Z stassats: 0x205c9d5b: fun pointer header: 0x00000545: symbol
2017-09-18T14:28:34Z stassats: something's already squatting the location
2017-09-18T14:31:47Z stassats: the header has 48 in it, but all slots are 0, so that would explain the PC at 0
2017-09-18T14:39:15Z oleo joined #sbcl
2017-09-18T14:44:04Z erratum joined #sbcl
2017-09-18T14:47:57Z dougk joined #sbcl
2017-09-18T14:54:24Z stassats: incidentally, (sb-ext:gc :gen t) is bad
2017-09-18T14:55:19Z Jesin quit (Quit: Leaving)
2017-09-18T15:04:43Z Jesin joined #sbcl
2017-09-18T15:21:12Z stassats: after i do (defvar *x* funcallable-instance) stops crashing
2017-09-18T15:21:36Z stassats: dougk: is it possible that hash-table values are not enliven funcallable-instances?
2017-09-18T15:22:46Z dougk_: unlikely. If there were a bug there, hash-tables should fail to enliven lots of things
2017-09-18T15:22:59Z dougk_: weak hash table?
2017-09-18T15:23:16Z stassats: no
2017-09-18T15:23:30Z stassats: a normal EQUAL hash-table
2017-09-18T15:23:50Z dougk_: then its storage is just a regular vector that doesn't even need to call the weakness tests; should be fine.
2017-09-18T15:24:03Z stassats: "should"
2017-09-18T15:24:04Z dougk_: i would bet the bug is that a program counter can't enliven a funcallable instance
2017-09-18T15:24:22Z stassats: but it's being referenced from the hash-table too
2017-09-18T15:24:31Z stassats: and the value in the hash-table gets corrupted too
2017-09-18T15:24:38Z stassats: well, it's the same pointer
2017-09-18T15:25:20Z dougk_: so the hash-table has a plain old Lisp descriptor to the object, not its address as a fixnum or sap, and it fails to enliven ?
2017-09-18T15:25:49Z stassats: yes, nothing internal is going on
2017-09-18T15:26:29Z stassats: putting it into a simple vector also prevents the crash
2017-09-18T15:26:43Z stassats: will try putting into another ht
2017-09-18T15:28:02Z dougk: i hope I didn't bork the subtype_VectorValidHashing bit
2017-09-18T15:28:35Z dougk: vectors have other bits in their header now, but they are supposed to be in a different byte, and I don't think setting the other byte is racy with GC
2017-09-18T15:28:48Z dougk: there's the "coalescible" bit which is set for strings, bit-vectors
2017-09-18T15:28:55Z stassats: no crash either
2017-09-18T15:30:24Z stassats: just printing the value doesn't prevent the crash
2017-09-18T15:32:42Z stassats: out of ideas to try
2017-09-18T15:39:36Z stassats: and i don't need to use ab
2017-09-18T15:39:43Z stassats: or call the function
2017-09-18T15:39:47Z stassats: (time (loop repeat  10000000 do (eval '(cons 1 2))))
2017-09-18T15:39:56Z stassats: (sb-kernel:layout-of (car (alexandria:hash-table-values tumak.view::*template-registry*))) => 0
2017-09-18T15:51:42Z dougk: statsats: is this the container thing?  and you have a reliable repro?  I want to take a look.  I have all sorts of tools I wrote to debug GC
2017-09-18T15:52:23Z stassats: dougk: it is, i'm now working on building a reproducible test case
2017-09-18T15:52:31Z slyrus: it's not just a container thing, FWIW
2017-09-18T15:54:01Z stassats: there are foreign callbacks involved
2017-09-18T15:55:02Z stassats: if i instantiate the fin by hand, (tumak.view:render  "/src/lisp/tumak/templates/blog.html")
2017-09-18T15:55:05Z stassats: nothing crashes
2017-09-18T15:55:21Z stassats: but not if the first init is from an http request
2017-09-18T15:55:50Z slyrus: interesting
2017-09-18T15:57:03Z stassats: (there's your workaround)
2017-09-18T15:57:04Z stassats: (ha-ha)
2017-09-18T15:57:24Z Jesin quit (Quit: Leaving)
2017-09-18T15:57:40Z slyrus: and I couldn't get it to crash with :compact-instance-header turned off either, so that's two workarounds
2017-09-18T15:59:08Z Jesin joined #sbcl
2017-09-18T16:04:07Z blackwolf joined #sbcl
2017-09-18T16:13:46Z stassats: ok, i'm out of ideas again
2017-09-18T16:24:20Z stassats: not necessary to cons, just (gc) will do
2017-09-18T16:35:53Z slyrus: but the fin has to be created from a foreign callback?
2017-09-18T16:36:16Z stassats: not clear
2017-09-18T16:38:14Z stassats: if it were clear i would have been able to reproduce it locally
2017-09-18T16:38:30Z stassats: unlikely to be cause by the callback
2017-09-18T16:38:31Z shka_ joined #sbcl
2017-09-18T16:38:36Z stassats: maybe by its environment
2017-09-18T17:31:36Z carvite joined #sbcl
2017-09-18T17:34:57Z attila_lendvai joined #sbcl
2017-09-18T17:34:57Z attila_lendvai quit (Changing host)
2017-09-18T17:34:57Z attila_lendvai joined #sbcl
2017-09-18T18:02:41Z dougk quit (Ping timeout: 246 seconds)
2017-09-18T18:03:28Z dougk joined #sbcl
2017-09-18T18:13:51Z scymtym_: http://paste.lisp.org/display/356224 for lp#1717971 ?
2017-09-18T18:26:45Z carenz quit (Ping timeout: 240 seconds)
2017-09-18T18:27:15Z dougk: slyris: if I implement pinning of FINs from unboxed pointers on the stack and mail you a patch, can you test it?
2017-09-18T18:27:20Z dougk: slyrus:
2017-09-18T18:27:31Z slyrus: please
2017-09-18T18:34:24Z stassats: dougk: but there's no stack with the pointer
2017-09-18T18:35:24Z dougk: well, this is beginning so sound incredibly similar to the issue I'm seeing. There is *allegedly* no stack with the pointer. But if you do a full scan of memory based on reading /proc/self/maps, and peek at literally every word, there is a pointer that "coincidentally" is the object that ought to have been pinned. Too spooky for mere coincidence
2017-09-18T18:35:46Z dougk: it has literally come to that, that I'm looking into all memory to track down whence this pointer came
2017-09-18T18:36:58Z dougk: this stray word is in my case, 2000 bytes below RSP, so the stack scan wouldn't have seen it. So how did anyone else "concoct" that thing later given that the code in question doesn't fabricate pointers out of nothing
2017-09-18T18:37:02Z stassats: using an alist instead of a hash-table => no crash
2017-09-18T18:38:20Z stassats: just storing it in a variable => no crash
2017-09-18T18:38:42Z dougk: slyrus: do you have a breaking revision number perhaps ?
2017-09-18T18:39:28Z dougk: but btw, i don't think my issue has anything to do with FINs, and I these objects are in alists
2017-09-18T18:39:41Z stassats: inside a simple-vector => no crash
2017-09-18T18:41:53Z stassats: now, an equalp hash-table => a crash
2017-09-18T18:41:57Z stassats: err
2017-09-18T18:42:05Z stassats: equal crashes, equalp doesn't crash
2017-09-18T18:43:42Z stassats: that's pretty weird...
2017-09-18T18:49:52Z stassats: i start with an equalp hash-table, run it for a while, then switch to equal => doesn't crash
2017-09-18T18:50:14Z stassats: with a new funcallable-instance
2017-09-18T18:57:40Z stassats: i do (setf (gethash "zz" table) 1) before there's anything in the table  => no crash
2017-09-18T18:59:24Z stassats: i do that after the instance is in the table => no crash
2017-09-18T19:02:29Z slyrus: dougk: I think this has been around for a while. Are you asking when the crash first appeared?
2017-09-18T19:03:26Z dougk: slyrus: yes, I touched hash-tables recently. within the month. How long is "a while" ?
2017-09-18T19:04:48Z slyrus: good question. I had some similar crashes with fukamachi's web stack some months back (3? 6?) but it may have been a different problem. I can certainly try with .19 or so and see if I can reproduce it.
2017-09-18T19:20:08Z m00natic quit (Remote host closed the connection)
2017-09-18T19:27:30Z slyrus: roughly the same problem with 1.3.19. memory fault at #x5.
2017-09-18T19:27:48Z slyrus: shall I keep going backwards?
2017-09-18T19:28:09Z stassats: wouldn't hurt
2017-09-18T19:35:25Z scymtym_ quit (Ping timeout: 246 seconds)
2017-09-18T19:39:57Z stassats: looking at the mark bit
2017-09-18T19:40:36Z stassats: i conses a bunch funcallable-instances, and (gc) doesn't trigger the problem anymore, and the object gets a mark bit, but after i use ab it gets triggered anyway
2017-09-18T19:42:04Z stassats: but (gc :gen 2) triggers it
2017-09-18T19:43:58Z slyrus: 1.3.14 looks good (hasn't crashed yet)
2017-09-18T19:44:33Z slyrus: bisection time. Too bad each turn takes 10 minutes or so...
2017-09-18T19:45:11Z stassats: you could use the sbcl.org binaries
2017-09-18T19:45:18Z stassats: to reduce the search space a bit
2017-09-18T19:45:56Z dougk quit (Ping timeout: 246 seconds)
2017-09-18T19:48:41Z stassats: dougk_: when did the immobile space objects start being collected?
2017-09-18T19:57:40Z stassats: back to having no testable ideas
2017-09-18T19:59:51Z dougk joined #sbcl
2017-09-18T20:01:53Z slyrus: it's somewhat reassuring to know that 1.3.14 works
2017-09-18T20:02:21Z stassats: 1.3.14 didn't really have the immobile space stuff, did it?
2017-09-18T20:07:57Z dougk_: immobile space always had a collector. Whether it was scrubbing aggressively enough ... that is the question
2017-09-18T20:08:49Z dougk_: anyway, i'm pretty 100% convinced that the GC is technically in need of pinning FINs based on the program counter, *but* the compiler is never, repeat NEVER, smart enough to actually store the FIN in exactly one register
2017-09-18T20:09:03Z slyrus: OK, two interesting things. 1) 1.3.17 is broken.
2017-09-18T20:09:34Z dougk_: in other words, the call is done through RAX, and we then load the FIN's closure from RAX, while trying to execute the next instruction in the FIN, which has always been pinned by another register sort of by accident
2017-09-18T20:09:46Z stassats: dougk_: that may be true, but won't help this particular problem
2017-09-18T20:09:51Z dougk_: I'm concocting a test to show that if the compiler were in fact smart enough to have only RAX holding the FIN, then we'd die.
2017-09-18T20:09:52Z slyrus: 2) if inside the djula template code I throw a break, things get messed up. This wasn't the case in 1.3.14. Perhaps the error handling stuff is hosed?
2017-09-18T20:13:38Z dougk_: the last time this was known good is end of January ?
2017-09-18T20:14:58Z stassats: why is it so fickle though, the shape of the hash-table affects it
2017-09-18T20:15:06Z slyrus: trying 1.3.15 now. 1.3.17 is broken.
2017-09-18T20:15:11Z stassats: especially the equal/equalp difference
2017-09-18T20:22:44Z slyrus: ironclad takes a loooong time to compile...
2017-09-18T20:23:00Z stassats: yes
2017-09-18T20:25:21Z slyrus: .16 is broken
2017-09-18T20:28:12Z scymtym joined #sbcl
2017-09-18T20:28:34Z slyrus: now trying .15
2017-09-18T20:33:19Z stassats: dougk_: sweep_fixedobj_pages is called from sweep_immobile_space not with a pointer
2017-09-18T20:33:48Z stassats: ok, there's two sweep_fixedobj_pages
2017-09-18T20:34:24Z dougk_: yah, probably a bad idea to overload the name
2017-09-18T20:34:57Z dougk_: also probably a bad idea that 'fullcgc' isn't named 'marknsweep', but the other got there first
2017-09-18T20:36:24Z stassats: what's the difference between fixedobj and varyobj?
2017-09-18T20:36:47Z stassats: instances, i presume, are varyobj?
2017-09-18T20:38:44Z dougk_: they're fixedobj actually because there are only layouts in immobile space
2017-09-18T20:38:47Z dougk_: varyobj is code components
2017-09-18T20:39:07Z stassats: and funcallable-instances?
2017-09-18T20:39:14Z dougk_: funcallable are fixedobj
2017-09-18T20:39:42Z dougk_: actually, i slightly misspoke - packages are also in fixedobj
2017-09-18T20:40:45Z stassats: and the header actually turns into #x30
2017-09-18T20:40:58Z stassats: the following slots are zero
2017-09-18T20:44:01Z stassats: the preceding bytes do not change
2017-09-18T20:44:52Z stassats: i could see it being zero, but why #x30?
2017-09-18T20:45:21Z stassats: the size of zeros?
2017-09-18T20:45:53Z stassats: object+48 is non zero
2017-09-18T20:46:59Z stassats: looks like it's something called a hole
2017-09-18T20:50:20Z dougk_: fixnums in immobile fixedobj space are free sparc markers
2017-09-18T20:50:24Z dougk_: free space markers
2017-09-18T20:51:01Z dougk_: #x30 is 6 words, sounds right
2017-09-18T20:51:11Z dougk_: it's saying the next free space is 6 words ahead
2017-09-18T20:51:16Z dougk_: so the object was trashed
2017-09-18T20:51:33Z stassats: so it's zeroing the instance alright
2017-09-18T20:51:44Z stassats: but nobody asked it to
2017-09-18T20:52:03Z slyrus: .15 looks good
2017-09-18T20:52:13Z slyrus: so something between .15 and .16
2017-09-18T20:52:18Z stassats: unclear how immobile_obj_gen_bits works
2017-09-18T20:52:21Z dougk_: the thing to do is turn on verify_gens to see whether GC at least believes that it did the right thing
2017-09-18T20:52:37Z dougk_: if it didn't believe that it did the right thing, the problem is far easier than if it did believe that it did the right thing
2017-09-18T20:53:28Z dougk_: gen_bits accesses byte 3 (least-significant is byte 0) of the header. simple-funs don't have a gen byte, they have to read the code-component's gen byte
2017-09-18T20:53:57Z stassats: fatal error encountered in SBCL pid 1418(tid 0x7fd90460f700):
2017-09-18T20:53:58Z dougk_: bits 0 through 2 give 3 bits which can convey gen 0 through 7. bit 4 is the visited bit
2017-09-18T20:53:58Z stassats: Ptr 0x205c9d5b @ 0x1005232678 sees trashed object.
2017-09-18T20:54:06Z dougk_: there we go
2017-09-18T20:55:28Z slyrus: so 2b8bfc is right before fin's in immobile fixedobj space. the next two commits look like likely suspects, no?
2017-09-18T20:55:46Z francogrex joined #sbcl
2017-09-18T20:55:53Z stassats: checked, it's the object in question
2017-09-18T20:56:20Z francogrex: hi this here was fixed: https://bugs.launchpad.net/sbcl/+bug/981106 but i get the behavior again with v 1.2.7
2017-09-18T20:56:21Z dougk quit (Ping timeout: 246 seconds)
2017-09-18T20:58:04Z stassats: would you also report a bug against cmucl?
2017-09-18T20:59:37Z |3b|: francogrex: test case in bug works here on 1.3.21
2017-09-18T20:59:51Z stassats: what's a three year old version when you can try a 20 year old version
2017-09-18T21:01:06Z francogrex: those are the v of the release binaries on the main page
2017-09-18T21:01:37Z rumbler31 quit (Remote host closed the connection)
2017-09-18T21:02:06Z Bike quit (Ping timeout: 248 seconds)
2017-09-18T21:02:15Z francogrex: regardless, it's after 2012
2017-09-18T21:02:18Z |3b|: those tend to be out of date for less popular platforms, but give you something with which to build current versions
2017-09-18T21:02:44Z dougk_: stassats: or slyrus: there is a brute-force way to capture the heap that i sometimes resort to as follows: add a counter in collect_garbage, just increment it on entry and print it as "entering GC cycle <n>". Determine the last good N prior to bad N. Put a gdb breakpoint on that line at the good N. Then use gdb to get all memory spaces to a file (http://www.delorie.com/gnu/docs/gdb/gdb_69.html).  I can work with that
2017-09-18T21:02:52Z francogrex: ok
2017-09-18T21:03:01Z stassats: francogrex: well, i see that you're saving on typing "versions", i guess you're also saving the energy it takes to compile the latest sbcl
2017-09-18T21:03:37Z francogrex: v is for vendetta
2017-09-18T21:03:40Z stassats: dougk_: i just need two calls to (gc)
2017-09-18T21:04:18Z dougk_: that's fine, I'm willing to do the debugging, but i need something. a heap dump or a repro
2017-09-18T21:04:48Z dougk_: or, if SLAD works after the first GC, send me that core?
2017-09-18T21:05:06Z stassats: considering that changing from an equal to an equalp table makes it go away, really hard to reproduce locally
2017-09-18T21:07:16Z stassats: dougk_: that works
2017-09-18T21:07:17Z dougk_: the funny thing about standard-generic-function is that there's almost no way for it to lack a root.  As soon as it has one method, which is on a symbol probably, then it's in the list of methods attached to that specializer. That method points to the GF
2017-09-18T21:07:31Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-18T21:07:59Z dougk_: as such, lack of a root is almost inconceivable, but it seems you two have found a way to cause that, and i've like to know how. Does this involve multi-threads adding methods?
2017-09-18T21:08:01Z stassats: now to exfiltrate the core
2017-09-18T21:08:14Z stassats: dougk_: it's not a gf
2017-09-18T21:08:26Z dougk_: oh, what kind of funcallable instance?
2017-09-18T21:08:32Z stassats: it's a funcallable-instances inside a hash-table
2017-09-18T21:08:35Z stassats: that's it
2017-09-18T21:08:36Z slyrus: dougk_: but is this even a standard-generic-function? the object in question has a funcallable-standard-class metaclass
2017-09-18T21:08:40Z stassats: it's attached to nothing else
2017-09-18T21:08:41Z stassats: no name
2017-09-18T21:08:49Z slyrus: ritght, what stassats said
2017-09-18T21:09:12Z stassats: slyrus: how do i get a file from a container?
2017-09-18T21:09:14Z dougk_: so it's a funcallable-instance that is not a GF? this should not be in immobile space
2017-09-18T21:09:33Z dougk_: the GFs in immobile-space should all be standard-GF.
2017-09-18T21:09:43Z dougk_: rather the FINs in immobile-space should all be standard-GF
2017-09-18T21:09:51Z stassats: dougk_: just (defclass fin () () (:metaclass sb-mop:funcallable-standard-class)) will be immobile
2017-09-18T21:09:57Z slyrus: you can cheat and copy it to /root/quicklisp/host-local-projects which will put it wherever you set that to on the docker run cmd
2017-09-18T21:09:59Z dougk_: oh?
2017-09-18T21:10:20Z dougk_: maybe i did that.  ok, so it's a GF that lacks a symbol as a root or something, interesting
2017-09-18T21:11:12Z stassats: slyrus: docker cp
2017-09-18T21:12:16Z attila_lendvai joined #sbcl
2017-09-18T21:12:23Z dougk joined #sbcl
2017-09-18T21:14:59Z stassats: well, the core has an already corrupted object
2017-09-18T21:15:05Z stassats: i guess slad does that
2017-09-18T21:16:12Z stassats: dougk_: do you still want it?
2017-09-18T21:16:29Z dougk_: argh, i was hoping to get pre-corruption
2017-09-18T21:16:44Z dougk_: ah, because slad can't avoid GC
2017-09-18T21:19:55Z stassats: i wonder if i can remove the dependence on postgres
2017-09-18T21:20:02Z dougk: is it possible to get a map-referencing-objects to the object? possibly not because map-referencing-objects was massively broken until recently
2017-09-18T21:20:04Z stassats: save the core before the object is initialized
2017-09-18T21:20:30Z dougk: i wouldn't expect traceroot to work, because GC would trash the object before traceroot got a chance to show the path
2017-09-18T21:23:51Z stassats: dougk: http://paste.lisp.org/display/356236
2017-09-18T21:23:53Z slyrus: as expected dougk's commits from 2/23 are the culprits
2017-09-18T21:23:57Z stassats: that's (sb-vm::map-referencing-objects #'print :all (gethash #p"blog.html"  tumak.view::*template-registry*))
2017-09-18T21:25:03Z dougk: the #<compiled-template> is the FIN that's getting trashed?
2017-09-18T21:25:17Z slyrus: yes
2017-09-18T21:32:11Z dougk: when i load djula from quicklisp i get: misplaced &AUX in lambda list
2017-09-18T21:33:05Z slyrus: what's the date on that one?
2017-09-18T21:33:13Z dougk: trying again after update-all-dists
2017-09-18T21:33:17Z dougk: worked
2017-09-18T21:33:18Z slyrus: from quicklisp/dist/quicklisp/software/
2017-09-18T21:33:19Z slyrus: ok
2017-09-18T21:33:51Z jibanes quit (Ping timeout: 240 seconds)
2017-09-18T21:34:52Z jibanes joined #sbcl
2017-09-18T21:37:49Z Bike joined #sbcl
2017-09-18T21:40:48Z slyrus: thanks fellas. gotta run.
2017-09-18T21:42:40Z francogrex quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2017-09-18T22:24:40Z ASau joined #sbcl
2017-09-18T22:29:17Z blackwolf quit (Quit: ERC (IRC client for Emacs 25.2.1))
2017-09-18T22:37:30Z dougk quit (Ping timeout: 246 seconds)
2017-09-18T23:00:15Z shka_ quit (Ping timeout: 240 seconds)
2017-09-18T23:02:02Z rumbler31 joined #sbcl
2017-09-18T23:03:12Z cromachina joined #sbcl
2017-09-18T23:08:31Z rumbler31 quit (Ping timeout: 248 seconds)
2017-09-18T23:24:29Z rumbler31 joined #sbcl
2017-09-18T23:54:17Z dougk joined #sbcl
2017-09-19T00:06:03Z dougk quit (Ping timeout: 255 seconds)
2017-09-19T00:47:30Z milanj quit (Quit: This computer has gone to sleep)
2017-09-19T01:13:31Z stassats quit (Ping timeout: 240 seconds)
2017-09-19T01:21:30Z stassats joined #sbcl
2017-09-19T01:25:12Z stassats: dougk_: and how are code objects preserved?
2017-09-19T01:26:24Z attila_lendvai quit (Quit: Leaving.)
2017-09-19T01:26:36Z stassats: or what about the fdefn trampolines?
2017-09-19T01:32:59Z stassats quit (Ping timeout: 255 seconds)
2017-09-19T01:41:59Z dougk joined #sbcl
2017-09-19T01:45:30Z dougk quit (Read error: Connection reset by peer)
2017-09-19T01:45:53Z dougk joined #sbcl
2017-09-19T03:03:31Z karswell_ quit (Remote host closed the connection)
2017-09-19T03:04:16Z karswell_ joined #sbcl
2017-09-19T03:14:08Z malm quit (Quit: Bye bye)
2017-09-19T03:30:02Z malm joined #sbcl
2017-09-19T03:47:05Z Bike quit (Ping timeout: 255 seconds)
2017-09-19T04:25:27Z dougk quit (Ping timeout: 260 seconds)
2017-09-19T04:59:03Z shka_ joined #sbcl
2017-09-19T06:01:15Z scymtym quit (Ping timeout: 240 seconds)
2017-09-19T06:04:39Z malm quit (*.net *.split)
2017-09-19T06:04:39Z dougk_ quit (*.net *.split)
2017-09-19T06:04:39Z angavrilov quit (*.net *.split)
2017-09-19T06:04:50Z DGASAU quit (*.net *.split)
2017-09-19T06:04:50Z les` quit (*.net *.split)
2017-09-19T06:04:50Z gko quit (*.net *.split)
2017-09-19T06:04:50Z Lord_Nightmare quit (*.net *.split)
2017-09-19T06:04:50Z stylewarning quit (*.net *.split)
2017-09-19T06:04:50Z lonjil quit (*.net *.split)
2017-09-19T06:04:50Z hzp quit (*.net *.split)
2017-09-19T06:04:50Z PuercoPop quit (*.net *.split)
2017-09-19T06:04:52Z Lord_Nightmare joined #sbcl
2017-09-19T06:05:13Z PuercoPop joined #sbcl
2017-09-19T06:05:44Z les joined #sbcl
2017-09-19T06:05:44Z gko joined #sbcl
2017-09-19T06:05:58Z stylewarning joined #sbcl
2017-09-19T06:08:14Z josemanuel[m] quit (Ping timeout: 246 seconds)
2017-09-19T06:10:01Z malm joined #sbcl
2017-09-19T06:10:01Z dougk_ joined #sbcl
2017-09-19T06:10:01Z angavrilov joined #sbcl
2017-09-19T06:10:01Z DGASAU joined #sbcl
2017-09-19T06:10:01Z lonjil joined #sbcl
2017-09-19T06:10:01Z hzp joined #sbcl
2017-09-19T07:01:33Z shka_ quit (Ping timeout: 248 seconds)
2017-09-19T07:04:28Z rumbler3_ joined #sbcl
2017-09-19T07:05:50Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-19T07:07:02Z rumbler31 joined #sbcl
2017-09-19T07:09:07Z rumbler3_ quit (Ping timeout: 246 seconds)
2017-09-19T07:11:17Z milanj joined #sbcl
2017-09-19T07:15:35Z josemanuel[m] joined #sbcl
2017-09-19T07:23:14Z scymtym joined #sbcl
2017-09-19T08:27:41Z carenz joined #sbcl
2017-09-19T10:25:29Z m00natic joined #sbcl
2017-09-19T10:37:56Z reg_ joined #sbcl
2017-09-19T10:38:32Z reg_: Hi
2017-09-19T10:38:51Z reg_: can somebody tell me why debugging support in windows is different with respect to linux under sbcl?
2017-09-19T10:39:10Z reg_: step doesn't seem to work ( under slime )
2017-09-19T10:45:25Z Bike joined #sbcl
2017-09-19T10:54:00Z dougk joined #sbcl
2017-09-19T11:16:20Z stassats joined #sbcl
2017-09-19T11:40:27Z rumbler31 quit (Remote host closed the connection)
2017-09-19T11:55:35Z stassats: getting back to that funcallable-instance problem
2017-09-19T11:55:55Z stassats: if i have two different instances in the table, both get wiped
2017-09-19T11:57:53Z stassats: but only if both are allocated from http requests, when i do one by hand they remain
2017-09-19T11:59:56Z reg_ quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
2017-09-19T12:07:10Z attila_lendvai joined #sbcl
2017-09-19T12:18:00Z stassats: huh, (sxhash (make-instance 'sb-mop:funcallable-standard-object )) => 42
2017-09-19T12:20:49Z stassats: that's easy to correct
2017-09-19T12:33:57Z Bike quit (Ping timeout: 255 seconds)
2017-09-19T12:41:04Z rumbler31 joined #sbcl
2017-09-19T12:45:30Z rumbler31 quit (Ping timeout: 240 seconds)
2017-09-19T12:59:33Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-19T13:06:05Z attila_lendvai joined #sbcl
2017-09-19T13:06:05Z attila_lendvai quit (Changing host)
2017-09-19T13:06:05Z attila_lendvai joined #sbcl
2017-09-19T13:07:50Z Bike joined #sbcl
2017-09-19T13:09:03Z attila_lendvai quit (Read error: Connection reset by peer)
2017-09-19T13:10:25Z attila_lendvai joined #sbcl
2017-09-19T13:26:18Z slyrus__ joined #sbcl
2017-09-19T13:27:33Z slyrus__: morning
2017-09-19T13:29:31Z rumbler31 joined #sbcl
2017-09-19T13:30:59Z attila_lendvai quit (Quit: Leaving.)
2017-09-19T13:53:21Z cromachina quit (Read error: Connection reset by peer)
2017-09-19T14:16:21Z slyrus__ quit (Ping timeout: 260 seconds)
2017-09-19T14:24:01Z nimbus7 joined #sbcl
2017-09-19T14:25:53Z stassats: more bizarre stuff: i renamed the variable to *x* => no crash
2017-09-19T14:26:49Z stassats: but the hash-table itself doesn't go away, why would it depend on the variable
2017-09-19T14:31:28Z stassats: i do (defparameter *x* *old-var*) => it crashes
2017-09-19T14:33:15Z stassats: (defparameter *x* (shiftf *old-var* nil)) => crashes as well
2017-09-19T14:33:52Z stassats: (defparameter *x* (make-hash-table :test #'equal)) doesn't crash, neither does doing (defparameter *old-var* (make-hash-table :test #'equal)) twice
2017-09-19T14:38:09Z scymtym quit (Ping timeout: 246 seconds)
2017-09-19T14:45:24Z stassats: using (load-time-value *old-var*) also crashes
2017-09-19T14:46:02Z stassats: so the variable somehow matters
2017-09-19T14:51:06Z stassats: i've been modifying the code in slime before that, now i do that to the original code
2017-09-19T14:51:15Z stassats: and now all eq/equal/equalp crash
2017-09-19T14:53:27Z _rumbler31 joined #sbcl
2017-09-19T14:54:23Z slyrus___ joined #sbcl
2017-09-19T14:54:26Z slyrus___: stassats: around?
2017-09-19T14:54:31Z stassats: yes
2017-09-19T14:54:36Z _rumbler31: I haven't built sbcl from git in a while.  If I've already built it in the past with options like choosing the lisp compiler and the build output directoy, is it enough to simply git pull and run sh make.sh again?  or are those options not preserved between builds and I have to specify them every time I rebuild?
2017-09-19T14:55:04Z stassats: _rumbler31: not preserved
2017-09-19T14:55:13Z _rumbler31: stassats: thank you
2017-09-19T14:55:29Z stassats: _rumbler31: customize-target-features.lisp is preserved
2017-09-19T14:56:36Z _rumbler31: do I need to clean before I run sh make.sh again if all I'm doing is a git pull?
2017-09-19T14:56:38Z slyrus___: You've probably got this but I have a reproducible #xD error that doesn't require ab and all that
2017-09-19T14:57:31Z _rumbler31: ahh I can get the old command line from build.log
2017-09-19T14:57:44Z stassats: slyrus___: yes, i don't need ab
2017-09-19T14:57:52Z stassats: _rumbler31: no clean needed
2017-09-19T14:58:03Z stassats: (it is needed, make.sh runs it)
2017-09-19T14:58:32Z stassats: slyrus___: do you still need postgres?
2017-09-19T14:58:41Z stassats: i still need a single curl invocation
2017-09-19T14:58:51Z slyrus___: no
2017-09-19T14:59:20Z _rumbler31: ty
2017-09-19T14:59:24Z stassats: without postgres, it can be packed up and shipped to dougk
2017-09-19T14:59:36Z slyrus___: you need three things, some consing, gc, and sb-alien
2017-09-19T14:59:57Z stassats: ok, show me
2017-09-19T15:00:02Z slyrus___: working on it...
2017-09-19T15:00:09Z slyrus___: 2 minutes
2017-09-19T15:03:21Z stassats: and i don't need hash-tables
2017-09-19T15:03:40Z stassats: (defstruct abc (z (vector nil nil nil nil))) is doing the same thing
2017-09-19T15:04:11Z stassats: that's a good finding
2017-09-19T15:05:25Z stassats: (list (vector nil)) too
2017-09-19T15:05:28Z stassats: even better
2017-09-19T15:06:45Z stassats: just (vector nil) is not enough
2017-09-19T15:06:59Z stassats: (the funcallable-instance is written instead of the NIL)
2017-09-19T15:06:59Z slyrus___ quit (Ping timeout: 240 seconds)
2017-09-19T15:11:02Z slyrus2 joined #sbcl
2017-09-19T15:13:49Z slyrus2: stassats: https://git.io/v5AA5
2017-09-19T15:14:23Z slyrus2: you can probably elide the djula stuff, but I've run out of time.
2017-09-19T15:14:31Z stassats: ok
2017-09-19T15:14:34Z slyrus2: taking off. thanks!
2017-09-19T15:14:39Z slyrus2 quit (Client Quit)
2017-09-19T15:14:50Z stassats: if i can reproduce it here, you'll be able to dismantle my account
2017-09-19T15:23:56Z stassats: but i can't reproduce even in the container
2017-09-19T15:24:34Z stassats: ah, a few times
2017-09-19T15:26:23Z oleo quit (Quit: Leaving)
2017-09-19T15:26:50Z stassats: ok, i can do it locally
2017-09-19T15:41:00Z stylewarning: This Slyrus bug saga has been fun to follow
2017-09-19T15:43:14Z stassats: would be more fun to get it fixed
2017-09-19T15:53:56Z scymtym joined #sbcl
2017-09-19T15:54:16Z stylewarning: I just want to know why
2017-09-19T15:54:35Z stassats: don't we all
2017-09-19T16:19:21Z m00natic quit (Remote host closed the connection)
2017-09-19T16:19:38Z stassats: should we just not advertise sbcl-bugs@ anywhere?
2017-09-19T16:19:46Z stassats: nobody's administering it
2017-09-19T16:20:01Z carenz quit (Ping timeout: 260 seconds)
2017-09-19T16:20:11Z stassats: not that i don't want to check sbcl-bugs@, i just don't know how
2017-09-19T16:23:51Z BitPuffin|osx quit (Ping timeout: 246 seconds)
2017-09-19T16:31:04Z oleo joined #sbcl
2017-09-19T16:51:53Z Jesin quit (Quit: Leaving)
2017-09-19T16:58:04Z stassats: dougk: https://gist.github.com/slyrus/dec38bf17d1f639c4de6960df9a01442#gistcomment-2207488 loaded in slime reproduces the crash
2017-09-19T16:58:21Z stassats: but not from the repl
2017-09-19T17:07:25Z stassats: https://gist.github.com/slyrus/dec38bf17d1f639c4de6960df9a01442#gistcomment-2207533
2017-09-19T17:07:28Z stassats: shorter and fails at the repl
2017-09-19T17:09:29Z slyrus_ joined #sbcl
2017-09-19T17:16:57Z Jesin joined #sbcl
2017-09-19T17:26:13Z shka_ joined #sbcl
2017-09-19T17:42:13Z oleo quit (Remote host closed the connection)
2017-09-19T17:48:03Z nimbus7 quit (Quit: Leaving)
2017-09-19T17:55:44Z slyrus_: stassats: your reduced test case works for me (no crash). clearly there's some stochasticity in play here (alignment? gc)?
2017-09-19T17:56:10Z stassats: slyrus_: it crashed on your container
2017-09-19T17:56:27Z slyrus_: heh
2017-09-19T17:56:45Z slyrus_: my other (reduced) test case is a reliably crasher on MacOS
2017-09-19T17:56:48Z slyrus_: reliable
2017-09-19T17:57:00Z stassats: yeah, this doesn't crash on macos
2017-09-19T17:57:04Z stassats: but does on linux
2017-09-19T17:57:30Z stassats: both yours and mine
2017-09-19T17:57:39Z slyrus_: ok. look at them funny for a long enough time and they'll all crash eventually :)
2017-09-19T17:58:09Z stassats: i had some iteration of a test case crash on macos too
2017-09-19T17:59:26Z slyrus_: well, it's nice that we can crash without postgres/djula/woo/etc...
2017-09-19T17:59:59Z slyrus_: or even the alien callbacks
2017-09-19T18:04:47Z nimbus7 joined #sbcl
2017-09-19T18:30:17Z shka_ quit (Ping timeout: 240 seconds)
2017-09-19T18:47:01Z dougk quit (Ping timeout: 260 seconds)
2017-09-19T19:35:28Z oleo joined #sbcl
2017-09-19T19:46:31Z ASau` joined #sbcl
2017-09-19T19:47:30Z ASau quit (Ping timeout: 240 seconds)
2017-09-19T19:49:28Z ASau` is now known as ASau
2017-09-19T19:58:59Z stassats: first gc => works, the vector is promoted to gen6, second gc => scav_vector gets it at garbage_collect_generation(5)
2017-09-19T19:59:08Z stassats: and the object is already thrashed
2017-09-19T20:07:59Z stassats: if i put something else in that hash vector, it gets scavenged at 4 as well
2017-09-19T20:12:49Z stassats: the fin gets thrashed at 1,4, and 5
2017-09-19T20:16:46Z slyrus_: hmm...
2017-09-19T20:18:13Z rpg joined #sbcl
2017-09-19T20:18:34Z rpg quit (Client Quit)
2017-09-19T20:18:37Z luis` quit (Quit: ZNC - http://znc.sourceforge.net)
2017-09-19T20:18:54Z rpg joined #sbcl
2017-09-19T20:22:50Z stassats: scav1 gets called on it at gen 0
2017-09-19T20:30:42Z luis joined #sbcl
2017-09-19T20:53:13Z nimbus7 quit (Quit: Leaving)
2017-09-19T20:54:41Z stassats: ok, scav_vector calls scav1 on it at gen0, but it gets thrashed at gen1
2017-09-19T20:59:50Z stassats: the vector is at gen5, a fin is written to it, triggering wp, so it gets scavenged once and reprotected
2017-09-19T21:00:16Z stassats: but the generation bit for the fin is not updated on the next gen collection, so it gets thrashed
2017-09-19T21:16:48Z Bike quit (Ping timeout: 246 seconds)
2017-09-19T21:18:53Z sjl quit (Quit: WeeChat 1.3)
2017-09-19T21:19:24Z sjl joined #sbcl
2017-09-19T21:19:24Z sjl quit (Client Quit)
2017-09-19T21:19:52Z sjl joined #sbcl
2017-09-19T21:19:57Z sjl quit (Client Quit)
2017-09-19T21:31:48Z angavrilov quit (Remote host closed the connection)
2017-09-19T21:55:29Z _rumbler31 quit (Ping timeout: 240 seconds)
2017-09-19T22:04:54Z DeadTrickster quit (Ping timeout: 264 seconds)
2017-09-19T22:06:46Z DeadTrickster joined #sbcl
2017-09-19T22:09:31Z scymtym: stassats: re lp#1717971 did you see the http://paste.lisp.org/display/356224 ? probably needs cross-typep extension, though
2017-09-19T22:10:29Z stassats: i thought about that solution, i guess it's the easiest
2017-09-19T22:13:11Z stassats: maybe only do that REOPTIMIZE-LVAR is set
2017-09-19T22:13:14Z Shinmera quit (Quit: しつれいしなければならないんです。)
2017-09-19T22:13:15Z stassats: lvar-reoptimize
2017-09-19T22:17:10Z scymtym: i will try that
2017-09-19T22:24:48Z stassats: something like http://paste.lisp.org/display/356224#1
2017-09-19T22:28:02Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-19T22:33:47Z Jesin quit (Quit: Leaving)
2017-09-19T22:37:40Z stassats: i thought ctypep would be bad on unknown types, but i guess it's ok
2017-09-19T22:37:44Z stassats: and is better on satisfies
2017-09-19T22:38:57Z Jesin joined #sbcl
2017-09-19T22:40:59Z scymtym: with lvar-reoptimize, self-build works even with unmodified cross-typep
2017-09-19T22:41:30Z stassats: but for how long?
2017-09-19T22:42:09Z stassats: what did you have to modify?
2017-09-19T22:43:18Z scymtym: with my version of the paste, cross-typep was called with (complex (double-float 0.0 0.0)), i think
2017-09-19T22:43:48Z scymtym: it can only do (complex double-float) as it stands
2017-09-19T22:45:43Z stassats: http://paste.lisp.org/display/356224#2
2017-09-19T22:46:04Z attila_lendvai joined #sbcl
2017-09-19T22:47:06Z stassats: scymtym: did you already modify cross-typep?
2017-09-19T22:47:30Z stassats: if so, commit it anyway, otherwise let the future someone deal with it
2017-09-19T22:49:06Z scymtym: stassats: no, i would like to avoid that. making it handle more cases seems like a can of worms
2017-09-19T22:50:06Z scymtym: your last annotation is basically the version i tested. should we check lvar-reoptimize for all uses if the principal lvar has multiple?
2017-09-19T22:50:57Z stassats: only the final really matters here
2017-09-19T22:51:05Z stassats: since that's where the constant is
2017-09-19T22:52:08Z stassats: let me see if it should actually use type-intersection
2017-09-19T22:54:35Z stassats: can't decide
2017-09-19T22:55:03Z stassats: i guess since we do have the value, ctypep should be the most accurate
2017-09-19T23:00:25Z rumbler31 left #sbcl
2017-09-19T23:03:42Z scymtym: stassats: http://paste.lisp.org/display/356224#3
2017-09-19T23:04:58Z stassats: is modifying lvar-value necessary?
2017-09-19T23:05:55Z scymtym: no, unrelated
2017-09-19T23:06:04Z scymtym: i can pull that out of the commit
2017-09-19T23:08:40Z stassats: and the test, i like to call the resulting functions
2017-09-19T23:09:12Z stassats: (and check their result)
2017-09-19T23:10:06Z stassats: maybe replace (throw 'ct5 0) with (throw 'ct5 123) for that
2017-09-19T23:10:11Z stassats: better testing
2017-09-19T23:17:21Z scymtym: ok
2017-09-19T23:17:25Z scymtym: come to think of it, i have checked-compile-and-call or something like that somewhere
2017-09-19T23:20:56Z Bike joined #sbcl
2017-09-20T00:05:24Z dougk joined #sbcl
2017-09-20T00:17:36Z cromachina joined #sbcl
2017-09-20T00:39:10Z stassats quit (Ping timeout: 248 seconds)
2017-09-20T01:00:28Z dougk quit (Ping timeout: 240 seconds)
2017-09-20T01:08:10Z milanj quit (Quit: This computer has gone to sleep)
2017-09-20T01:12:15Z attila_lendvai quit (Quit: Leaving.)
2017-09-20T01:32:04Z jdz quit (Ping timeout: 255 seconds)
2017-09-20T01:36:36Z milanj joined #sbcl
2017-09-20T01:38:27Z jdz joined #sbcl
2017-09-20T01:44:01Z nicdev joined #sbcl
2017-09-20T02:32:03Z slyrus_: dougk_: did you see the reduced test case for the fin bug?
2017-09-20T02:35:11Z Bike quit (Ping timeout: 252 seconds)
2017-09-20T02:38:35Z Bike joined #sbcl
2017-09-20T02:38:38Z Bike quit (Remote host closed the connection)
2017-09-20T02:39:18Z Bike joined #sbcl
2017-09-20T02:50:58Z Bicyclidine joined #sbcl
2017-09-20T02:53:26Z Bike quit (Ping timeout: 255 seconds)
2017-09-20T04:10:13Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-20T04:28:09Z shka_ joined #sbcl
2017-09-20T04:45:21Z angavrilov joined #sbcl
2017-09-20T05:30:06Z Lord_Nightmare quit (Ping timeout: 264 seconds)
2017-09-20T05:56:32Z Lord_Nightmare joined #sbcl
2017-09-20T05:58:36Z scymtym quit (Ping timeout: 248 seconds)
2017-09-20T06:10:52Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-20T06:11:57Z milanj quit (Quit: This computer has gone to sleep)
2017-09-20T06:18:46Z dougk joined #sbcl
2017-09-20T06:25:23Z jack_rabbit joined #sbcl
2017-09-20T06:38:14Z dougk__ joined #sbcl
2017-09-20T06:40:45Z dougk quit (Ping timeout: 248 seconds)
2017-09-20T06:43:24Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-20T06:57:53Z jack_rabbit joined #sbcl
2017-09-20T07:16:58Z shka_ quit (Ping timeout: 240 seconds)
2017-09-20T07:30:39Z scymtym joined #sbcl
2017-09-20T07:33:37Z milanj joined #sbcl
2017-09-20T08:19:18Z fiddlerwoaroof quit (Ping timeout: 264 seconds)
2017-09-20T08:49:58Z carenz joined #sbcl
2017-09-20T08:52:49Z fiddlerwoaroof joined #sbcl
2017-09-20T09:35:33Z m00natic joined #sbcl
2017-09-20T09:41:31Z ASau quit (Ping timeout: 248 seconds)
2017-09-20T10:06:55Z akovalenko joined #sbcl
2017-09-20T10:35:20Z BitPuffin|osx joined #sbcl
2017-09-20T10:44:01Z slyrus_ quit (Quit: Client Quit)
2017-09-20T11:13:09Z Bike joined #sbcl
2017-09-20T11:18:01Z slyrus_ joined #sbcl
2017-09-20T11:21:33Z dougk__ quit (Ping timeout: 255 seconds)
2017-09-20T11:22:41Z dougk joined #sbcl
2017-09-20T11:33:21Z stassats joined #sbcl
2017-09-20T11:34:06Z stassats: any luck with the funcallable-instance stuff?
2017-09-20T11:34:13Z slyrus_: morning
2017-09-20T11:40:21Z Bicyclidine joined #sbcl
2017-09-20T11:42:35Z Bike quit (Ping timeout: 248 seconds)
2017-09-20T12:31:20Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-20T12:54:46Z Bike joined #sbcl
2017-09-20T12:56:59Z stassats: it appears that if i unprotect a vector with an immobile object then it gets scavenged only at gen0 and gen5
2017-09-20T12:57:31Z stassats: be it functions or symbols
2017-09-20T12:57:52Z stassats: same thing with fixnums
2017-09-20T13:01:00Z stassats: or anything from younger generations
2017-09-20T13:01:55Z stassats: so the immobile stuff doesn't get enlivened
2017-09-20T13:08:28Z stassats: and because it does get enlivened at gen0 its gen bit is equal to discard_gen
2017-09-20T13:12:37Z stassats: i see a potential solution
2017-09-20T13:15:02Z stassats: oh, a trivial one, actually, let me test it
2017-09-20T13:17:24Z stassats: it works
2017-09-20T13:19:28Z stassats: and the logic checks out, testing some more and pushing
2017-09-20T13:29:08Z stassats: and the constant-lvar changes broke something
2017-09-20T13:29:36Z dougk quit (Ping timeout: 240 seconds)
2017-09-20T13:30:22Z stassats: (LET ((X 'A)) (CTYPECASE X (NUMBER 'BAD) (#.(find-class 'symbol) 'GOOD))) => failed AVER: (SB-C::CONSTANT-LVAR-P TYPE)
2017-09-20T13:36:28Z stassats: slyrus slyrus_ : fixed
2017-09-20T13:40:22Z stassats: huh
2017-09-20T13:40:34Z stassats: there's actually #<INTERSECTION-TYPE (AND BUILT-IN-CLASS CLASS)>
2017-09-20T13:40:43Z stassats: but (#<STANDARD-CLASSOID BUILT-IN-CLASS> #<UNKNOWN-TYPE CLASS>)
2017-09-20T13:41:07Z stassats: what
2017-09-20T13:44:23Z stassats: reduced to (let () (typep * #.(find-class 'symbol)))
2017-09-20T13:53:51Z dougk joined #sbcl
2017-09-20T14:00:46Z scymtym: can we rename the restart for undefined functions whose report is "Retry calling FOO." from CONTINUE to RETRY?
2017-09-20T14:00:57Z stassats: clhs retry
2017-09-20T14:00:57Z specbot: Couldn't find anything for retry.
2017-09-20T14:01:24Z scymtym: i think only continue and abort are standard
2017-09-20T14:01:37Z stassats: use-value, store-value
2017-09-20T14:01:45Z scymtym: right
2017-09-20T14:02:13Z dougk: stassats: thanks for the fix. my daughter's birthday was yesterday and i wasn't around
2017-09-20T14:02:55Z stassats: so, at fndb.lisp time CLASS is not defined
2017-09-20T14:07:38Z stassats: not sure how to reconcile that
2017-09-20T14:08:22Z slyrus_: stassats: very nice!
2017-09-20T14:08:58Z scymtym: for similar situations involving generic functions (such as applicable method), we establish a restart called RETRY. my reason for asking is that in my experience, CONTINUE is mostly chosen for situations in which the execution can continue without additional actions being taken (ASSERT being the exception, of course)
2017-09-20T14:10:01Z slyrus_: thanks stassats and dougk for the forensic work and the fix!
2017-09-20T14:10:30Z stassats: slyrus_: you can now remove my account
2017-09-20T14:10:44Z slyrus_: will do
2017-09-20T14:11:44Z stassats: scymtym: i like continue (obviously)
2017-09-20T14:12:05Z stassats: besides, it's too late
2017-09-20T14:12:39Z cromachina quit (Read error: Connection reset by peer)
2017-09-20T14:13:07Z scymtym: stassats: ok
2017-09-20T14:15:48Z stassats: CLASS seems to be defined in pcl
2017-09-20T14:15:53Z stassats: too late
2017-09-20T14:18:20Z stassats: could go through all function types and fix up CLASS
2017-09-20T14:20:14Z dougk: do you need the real class metaobject for CLASS or would just the classoid be enough? condition-boot simulates condition classes, could do something similar
2017-09-20T14:20:44Z stassats: dougk: dunno, it's just the TYPE-SPECIFIER type in fndb
2017-09-20T14:21:05Z stassats: or anything with CLASS there
2017-09-20T14:23:03Z stassats: interestingly, when i look at (info :function :type 'typep) it has the right CLASS
2017-09-20T14:23:17Z dougk: those get fixed up
2017-09-20T14:23:33Z stassats: why do they end up in the compiler?
2017-09-20T14:23:44Z dougk: what is "they" ?
2017-09-20T14:23:51Z stassats: bogus types
2017-09-20T14:23:57Z stassats: unknown
2017-09-20T14:24:20Z stassats: ok, (info :function :type 'sb-c::%typep-wrapper) has an unknown-type
2017-09-20T14:24:35Z stassats: that's where the actual problem stems from
2017-09-20T14:27:01Z stassats: dougk: i assume not enough fixing up, where does it happen?
2017-09-20T14:27:51Z dougk: some in src/pcl/fixup, some .. elsewhere let me see
2017-09-20T14:29:53Z dougk: code/setf-funs, for some reason
2017-09-20T14:30:25Z stassats: not doing enough, apparently
2017-09-20T14:30:47Z stassats: not fboundp
2017-09-20T14:31:12Z stassats: ok, an easy fix then
2017-09-20T14:35:10Z stassats: and that fixes (let () (typep * #.(find-class 'symbol)))
2017-09-20T14:39:55Z stassats: ok, no more bugs to fix, time to conjure some new ones
2017-09-20T14:54:12Z scymtym: stassats: if you are bored, there is still: http://paste.lisp.org/display/355653 which was introduced by b4f3834e84f6946e99d37744718e528f4fa94dd4 if i bisected correctly
2017-09-20T14:54:42Z stassats: interesting
2017-09-20T14:56:00Z stassats: failed AVER: (EQ (SB-C::FUNCTIONAL-KIND SB-C::FUNCTIONAL) :TOPLEVEL-XEP) is not something i like seeing
2017-09-20T15:13:45Z carenz quit (Ping timeout: 260 seconds)
2017-09-20T15:23:40Z edgar-rft quit (Quit: edgar-rft)
2017-09-20T15:25:10Z stassats: doesn't make much sense
2017-09-20T15:25:24Z stassats: i can just revert b4f3834e84f6946e99d37744718e528f4fa94dd4
2017-09-20T15:25:38Z stassats: but it'll probably uncovers something silly
2017-09-20T15:25:52Z stassats: and will come back in the future
2017-09-20T15:29:27Z stassats: ok, i can reproduce it if i do (proclaim '(optimize (speed 0) (debug 0) (safety 0) (compilation-speed 0) (space 0) (sb-c::preserve-single-use-debug-variables 0)))
2017-09-20T15:29:29Z stassats: without the changes
2017-09-20T15:29:32Z stassats: so, yeah
2017-09-20T15:38:12Z stassats: scymtym: i think i have a fix, but i'm out of desire to write a test
2017-09-20T15:41:49Z dougk quit (Ping timeout: 246 seconds)
2017-09-20T15:45:25Z scymtym: stassats: i can try to write a test later
2017-09-20T15:49:47Z slyrus_: speaking of tests... should we add the fin tests?
2017-09-20T15:50:31Z stassats: slyrus_: too unreliable
2017-09-20T15:52:05Z stassats: but i don't mind if anyone adds it
2017-09-20T15:52:19Z stassats: just not me (i've had enough of the issue)
2017-09-20T15:52:47Z rgrau joined #sbcl
2017-09-20T16:39:48Z dougk joined #sbcl
2017-09-20T16:52:42Z shka_ joined #sbcl
2017-09-20T17:12:02Z m00natic quit (Remote host closed the connection)
2017-09-20T17:29:47Z rgrau quit (Ping timeout: 248 seconds)
2017-09-20T17:41:29Z eschatologist quit (Ping timeout: 240 seconds)
2017-09-20T17:45:17Z eschatologist joined #sbcl
2017-09-20T17:47:56Z akovalenko quit (Ping timeout: 240 seconds)
2017-09-20T17:54:43Z akovalenko joined #sbcl
2017-09-20T18:13:43Z Jesin quit (Quit: Leaving)
2017-09-20T18:17:17Z BitPuffin|osx quit (Ping timeout: 248 seconds)
2017-09-20T18:21:05Z slyrus_ quit (Ping timeout: 264 seconds)
2017-09-20T18:35:16Z Jesin joined #sbcl
2017-09-20T19:02:58Z attila_lendvai joined #sbcl
2017-09-20T19:02:58Z attila_lendvai quit (Changing host)
2017-09-20T19:02:58Z attila_lendvai joined #sbcl
2017-09-20T19:24:20Z scymtym quit (Ping timeout: 246 seconds)
2017-09-20T19:25:47Z BitPuffin|osx joined #sbcl
2017-09-20T19:37:12Z dougk quit (Ping timeout: 240 seconds)
2017-09-20T19:45:48Z shka_ quit (Ping timeout: 240 seconds)
2017-09-20T19:50:03Z akovalenko quit (Ping timeout: 248 seconds)
2017-09-20T20:05:03Z sebboh joined #sbcl
2017-09-20T20:05:46Z akovalenko joined #sbcl
2017-09-20T20:09:51Z scymtym joined #sbcl
2017-09-20T20:10:25Z akovalenko quit (Client Quit)
2017-09-20T20:10:43Z akovalenko joined #sbcl
2017-09-20T20:11:26Z carenz joined #sbcl
2017-09-20T20:18:58Z akovalen` joined #sbcl
2017-09-20T20:21:05Z akovalenko quit (Ping timeout: 264 seconds)
2017-09-20T20:23:12Z carenz quit (Ping timeout: 240 seconds)
2017-09-20T20:38:44Z angavrilov quit (Remote host closed the connection)
2017-09-20T21:19:36Z Bike quit (Ping timeout: 255 seconds)
2017-09-20T21:28:46Z sjl joined #sbcl
2017-09-20T21:38:57Z dougk joined #sbcl
2017-09-20T21:46:08Z dougk quit (Ping timeout: 240 seconds)
2017-09-20T21:48:25Z Jesin quit (Quit: Leaving)
2017-09-20T21:50:45Z attila_lendvai quit (Quit: Leaving.)
2017-09-20T21:51:13Z attila_lendvai joined #sbcl
2017-09-20T21:51:52Z attila_lendvai quit (Client Quit)
2017-09-20T22:01:50Z Bike joined #sbcl
2017-09-20T22:20:59Z kilfer joined #sbcl
2017-09-20T22:21:16Z kilfer left #sbcl
2017-09-20T22:21:55Z Bicyclidine joined #sbcl
2017-09-20T22:25:00Z Bike quit (Ping timeout: 246 seconds)
2017-09-20T22:48:56Z Jesin joined #sbcl
2017-09-20T22:59:32Z m3tti joined #sbcl
2017-09-20T23:00:35Z m3tti: hi there anyone knows why on my musl based distro sbcl doesn't find sb-bsd-sockets
2017-09-20T23:01:01Z stassats: ask your distro
2017-09-20T23:02:13Z m3tti: is there a setting to enable it on build?
2017-09-20T23:02:33Z stassats: no
2017-09-20T23:16:46Z cromachina joined #sbcl
2017-09-20T23:38:20Z m3tti quit (Quit: leaving)
2017-09-20T23:38:24Z edgar-rft joined #sbcl
2017-09-20T23:51:44Z Bike joined #sbcl
2017-09-20T23:53:22Z Bicyclidine quit (Ping timeout: 252 seconds)
2017-09-21T00:22:36Z BitPuffin|osx quit (Ping timeout: 248 seconds)
2017-09-21T00:22:47Z itPuffin|osxB joined #sbcl
2017-09-21T00:25:56Z Bike quit (Ping timeout: 240 seconds)
2017-09-21T00:28:44Z Bike joined #sbcl
2017-09-21T01:05:14Z stassats quit (Ping timeout: 260 seconds)
2017-09-21T01:19:54Z Jesin quit (Quit: Leaving)
2017-09-21T01:40:56Z dougk joined #sbcl
2017-09-21T01:56:01Z dim quit (Ping timeout: 252 seconds)
2017-09-21T01:56:41Z dim joined #sbcl
2017-09-21T01:57:28Z slyrus_ joined #sbcl
2017-09-21T02:12:56Z milanj quit (Quit: This computer has gone to sleep)
2017-09-21T02:54:17Z shka_ joined #sbcl
2017-09-21T05:18:24Z dougk quit (Ping timeout: 260 seconds)
2017-09-21T06:02:12Z scymtym quit (Ping timeout: 240 seconds)
2017-09-21T06:14:29Z dougk joined #sbcl
2017-09-21T06:18:02Z carenz joined #sbcl
2017-09-21T06:18:42Z dougk quit (Ping timeout: 240 seconds)
2017-09-21T06:26:54Z milanj joined #sbcl
2017-09-21T07:14:46Z dougk joined #sbcl
2017-09-21T07:16:07Z shka_ quit (Read error: Connection reset by peer)
2017-09-21T07:16:18Z shka_ joined #sbcl
2017-09-21T07:18:42Z dougk quit (Ping timeout: 240 seconds)
2017-09-21T07:22:50Z shka_ quit (Ping timeout: 248 seconds)
2017-09-21T07:30:15Z Bike quit (Ping timeout: 240 seconds)
2017-09-21T07:39:29Z scymtym joined #sbcl
2017-09-21T08:04:07Z Bike joined #sbcl
2017-09-21T08:09:18Z DeadTrickster quit (Remote host closed the connection)
2017-09-21T08:15:50Z dougk joined #sbcl
2017-09-21T08:18:20Z Bike quit (Ping timeout: 248 seconds)
2017-09-21T08:18:50Z DeadTrickster joined #sbcl
2017-09-21T08:21:27Z dougk quit (Ping timeout: 255 seconds)
2017-09-21T09:10:29Z knobo joined #sbcl
2017-09-21T09:10:38Z knobo: What is this: Unexpected error 9 on netlink descriptor 6fatal error encountered in SBCL pid 17146(tid 0x7fffe70bf700):
2017-09-21T09:15:44Z |3b|: supposedly that error is usually due to closing wrong FDs in another thread or something like that
2017-09-21T09:17:00Z |3b|: 4th * in https://www.sourceware.org/ml/libc-alpha/2016-02/msg00502.html
2017-09-21T09:17:55Z dougk joined #sbcl
2017-09-21T09:18:18Z knobo: |3b|: sounds like something that could be happening here.
2017-09-21T09:22:03Z dougk quit (Ping timeout: 252 seconds)
2017-09-21T09:29:10Z knobo: |3b|: is it because close is called two times?
2017-09-21T09:30:39Z knobo: If the other end calls shutdown(2), and I call close. It should be ok, right?
2017-09-21T09:32:48Z |3b|: calling twice sounds bad (particularly if another thread might have opened it again in between), or closing an fd that is in use by another thread
2017-09-21T09:33:04Z |3b|: not sure about shutdown/close, haven't done networking in a while
2017-09-21T09:41:48Z m00natic joined #sbcl
2017-09-21T10:42:56Z angavrilov joined #sbcl
2017-09-21T10:47:40Z Bike joined #sbcl
2017-09-21T11:03:08Z DeadTrickster_ joined #sbcl
2017-09-21T11:06:12Z DeadTrickster quit (Ping timeout: 240 seconds)
2017-09-21T11:18:58Z dougk joined #sbcl
2017-09-21T11:23:34Z dougk quit (Ping timeout: 260 seconds)
2017-09-21T11:23:39Z edgar-rft quit (Quit: edgar-rft)
2017-09-21T11:48:07Z itPuffin|osxB quit (Ping timeout: 240 seconds)
2017-09-21T11:51:53Z stassats joined #sbcl
2017-09-21T11:54:12Z Xof quit (Ping timeout: 240 seconds)
2017-09-21T12:02:14Z |3b|: hmm, C-c C-c when cursor is on a lambda form in slime repl kills sbcl with nested errors
2017-09-21T12:04:44Z Bike quit (Remote host closed the connection)
2017-09-21T12:05:32Z |3b| wonders if there is something wrong with my slime, seems to have been having trouble with source locations lately too
2017-09-21T12:05:43Z |3b|: (or if i broke my sbcl somehow)
2017-09-21T12:19:26Z stassats: reproducibly?
2017-09-21T12:19:41Z |3b|: yeah
2017-09-21T12:20:13Z |3b|: same for defun and defvar forms but not (+ 1 2 3)
2017-09-21T12:20:24Z stassats: delete slime fasls
2017-09-21T12:20:27Z |3b| got bored of restarting sbcl before trying other things
2017-09-21T12:20:46Z |3b|: are they in the usual asdf cache?
2017-09-21T12:20:53Z stassats: ~/.slime/fasl
2017-09-21T12:21:42Z |3b|: still breaks
2017-09-21T12:22:08Z stassats: what's the first error?
2017-09-21T12:22:39Z stassats: and any local changes to slime or sbcl?
2017-09-21T12:22:58Z |3b|: local changes to sbcl, so could be my fault... will try a clean build in a bit
2017-09-21T12:25:30Z |3b| can't find a first error, building clean sbcl now
2017-09-21T12:27:37Z |3b|: test case is start slime, type (lambda()) move left 2 chars, hit C-c C-c, exit debugger if it gives normal interrupt, hit C-c C-c again, -> spam about nested errors -> crash
2017-09-21T12:28:57Z |3b|: and still breaks with "1.3.21.134-fcae5b0a6" without local changes
2017-09-21T12:29:19Z |3b|: this is on windows, slime from quicklisp
2017-09-21T12:29:37Z |3b| supposes i should verify i haven't made any local changes to that too
2017-09-21T12:34:19Z |3b|: yeah, same with clean slime too
2017-09-21T12:47:12Z |3b| also wonders if it is intended that (cffi:foreign-funcall-pointer p nil :void) gives optimization notes at debug 3, speed 1
2017-09-21T12:49:24Z |3b|: hmm, just noticed last attempt left a meaningful error message in debugger: Recursive lock attempt #<SB-THREAD:MUTEX owner: #<SB-THREAD:THREAD "repl-thread" RUNNING {100AA00033}> {100AA08783}>.
2017-09-21T12:50:55Z rgrau joined #sbcl
2017-09-21T12:52:37Z |3b|: and same on "1.3.21.196-9beff641b"
2017-09-21T12:53:46Z stassats: |3b|: sb alien is utterly inefficient
2017-09-21T12:54:05Z stassats: despite trying to be
2017-09-21T12:54:50Z |3b| is more wondering about getting the note at all at speed 1 than about its contents
2017-09-21T12:55:09Z stassats: sap consing always does this
2017-09-21T12:55:42Z stassats: cause apparently "oh it's so bad better fix it" yet you can't fix it because it's sb-alien
2017-09-21T12:57:20Z |3b|: ok, if it is expected i won't worry about it, and will just try to remember not to use C-u C-c C-k on that file :)
2017-09-21T13:02:02Z Bike joined #sbcl
2017-09-21T13:10:35Z |3b|: and broke on (list 1 2 3) that time, so probably not specific to form, and just luck that it worked on (+ 1 2 3) before
2017-09-21T13:22:59Z prxq joined #sbcl
2017-09-21T13:31:00Z prxq quit (Remote host closed the connection)
2017-09-21T13:32:59Z dougk joined #sbcl
2017-09-21T13:48:13Z carenz quit (Ping timeout: 246 seconds)
2017-09-21T13:49:46Z |3b|: is ((FLET "CLEANUP-FUN-33" :IN SB-THREAD::%CONDITION-WAIT)) from unwind-protect?
2017-09-21T13:51:40Z stassats: |3b|: yeah
2017-09-21T14:11:26Z cromachina quit (Read error: Connection reset by peer)
2017-09-21T14:27:33Z DeadTrickster_ quit (Remote host closed the connection)
2017-09-21T14:31:38Z knobo quit (Ping timeout: 248 seconds)
2017-09-21T14:32:55Z dougk quit (Ping timeout: 240 seconds)
2017-09-21T14:41:08Z scymtym quit (Ping timeout: 246 seconds)
2017-09-21T14:53:43Z _death joined #sbcl
2017-09-21T14:54:58Z _death: https://github.com/sbcl/sbcl/blob/master/src/code/defpackage.lisp#L72 <- missing optname as first argument.. (defpackage :foo (:documentation "bar" "quux"))
2017-09-21T14:55:20Z stassats: test case?
2017-09-21T14:55:51Z oleo quit (Quit: Leaving)
2017-09-21T14:56:19Z _death: the defpackage form I gave results in an error formatting the error
2017-09-21T14:56:46Z stassats: ok
2017-09-21T15:06:10Z |3b|: ok, looks like receive-if is in condition-wait, and when it grabs the mutex and leaves the without-interrupts, there is an interrupt-thread waiting from the C-c C-c so before it returns from condition-wait the interrupt calls wake-thread, which tries to grab the mutex, which errors on recursive lock, then debugger loop does receive-if which tries to grab the mutex, etc
2017-09-21T15:07:47Z |3b|: and there is another place it breaks sometimes, haven't looked at it but guessing it is similar problem
2017-09-21T15:08:33Z stassats: _death: fixed
2017-09-21T15:11:19Z _death: stassats: thanks
2017-09-21T15:13:51Z |3b|: and somewhere in there it tries to open the debugger for the C-c C-c, but same problem there anyway
2017-09-21T15:14:19Z stassats: |3b|: what's the test case, again?
2017-09-21T15:14:31Z |3b|: just C-c C-c in slime repl :/
2017-09-21T15:14:45Z stassats: obviously works for me
2017-09-21T15:14:50Z |3b|: usually breaks on first or 2nd try
2017-09-21T15:15:02Z stassats: os?
2017-09-21T15:15:06Z |3b|: windows
2017-09-21T15:15:08Z stassats: figures
2017-09-21T15:15:12Z |3b|: yeah :/
2017-09-21T15:15:33Z stassats: interrupt-thread is approximately totally broken on windows
2017-09-21T15:15:55Z oleo joined #sbcl
2017-09-21T15:16:36Z |3b| leans towards slime problem at this point, but still not quite sure what should be happening well enough to tell if it is or not
2017-09-21T15:22:01Z |3b|: how do i build with sb-thruption on linux?
2017-09-21T15:23:04Z |3b| tried make.sh --with-sb-thruption, but that says "sbcl/src/runtime/interrupt.c:1054: undefined reference to `check_pending_thruptions'"
2017-09-21T15:23:36Z |3b|: and undefined reference to wake_thread_posix in thread.c
2017-09-21T15:24:36Z |3b| tries with sb-safepoint too
2017-09-21T15:25:07Z |3b|: yeah, looks like it got past .c stuff with that
2017-09-21T15:29:31Z dougk joined #sbcl
2017-09-21T15:29:45Z Jesin joined #sbcl
2017-09-21T15:30:58Z |3b|: and can't reproduce the problem on sbcl/linux with safetpoint+thruption
2017-09-21T15:31:09Z |3b|: (running through ssh, though not sure that matters)
2017-09-21T15:33:54Z dougk quit (Ping timeout: 240 seconds)
2017-09-21T15:44:58Z Jesin quit (Quit: Leaving)
2017-09-21T15:46:52Z Jesin joined #sbcl
2017-09-21T16:00:14Z |3b|: stassats: i think the problem is in slime, in https://github.com/slime/slime/blob/master/swank.lisp#L445
2017-09-21T16:00:51Z |3b|: or else the sbcl implementation of wake-thread
2017-09-21T16:01:11Z stassats: what problem is in there?
2017-09-21T16:01:52Z |3b|: https://github.com/slime/slime/blob/master/swank/sbcl.lisp#L1753
2017-09-21T16:02:46Z |3b|: it interrupts the thread while the mutex is held, then wake-thread tries to get the mutex too and it breaks
2017-09-21T16:03:17Z |3b|: i guess 3rd option is that that mutex should allow recursive locks
2017-09-21T16:05:19Z stassats: what holds the mutex?
2017-09-21T16:05:46Z |3b|: swank/backend:receive-if
2017-09-21T16:06:05Z stassats: does it hold it, though?
2017-09-21T16:07:33Z |3b|: it just exited the without-interrupts in call-with-mutex
2017-09-21T16:07:59Z |3b| isn't sure if the backtrace i'm looking at is the case i was investigating earlier or the other one
2017-09-21T16:09:34Z |3b|: yeah, i think this is the other one, so might be slightly different
2017-09-21T16:12:20Z stassats: i think invoke-or-queue-interrupt doesn't return nil enough
2017-09-21T16:12:43Z |3b|: that's also possible
2017-09-21T16:13:40Z stassats: can you put NIL in all branches but the last one?
2017-09-21T16:13:44Z stassats: the T one
2017-09-21T16:14:56Z |3b|: still breaks i think
2017-09-21T16:15:13Z |3b|: i think it was going through the T path
2017-09-21T16:15:25Z stassats: probably, since it does work normally
2017-09-21T16:16:39Z |3b|: and looks like C-c C-b in .lisp buffers is broken too, so not specific to repl
2017-09-21T16:16:41Z tonton quit (Ping timeout: 240 seconds)
2017-09-21T16:16:55Z stassats: it'd be the same
2017-09-21T16:16:57Z stassats: most of the time
2017-09-21T16:18:26Z tonton joined #sbcl
2017-09-21T16:19:19Z scymtym joined #sbcl
2017-09-21T16:20:58Z stassats: |3b|: does it work without https://github.com/slime/slime/commit/5f23adaa88c034449d935b7a0bc26ed936d5ed0e
2017-09-21T16:23:47Z Jesin quit (Quit: Leaving)
2017-09-21T16:24:08Z |3b|: without the wake-thread it works (aside from being slow to notice the interrupt sometimes)
2017-09-21T16:24:32Z |3b|: testing without the whole change will take a bit, need to check it out from git...just a sec
2017-09-21T16:24:37Z stassats: sb-thread:condition-wait maybe just broken on windows thne
2017-09-21T16:24:50Z stassats: may be
2017-09-21T16:25:43Z |3b|: i don't think it is only condition-wait, breaks in with-mutex some of the time, and in condition-wait some of the time
2017-09-21T16:26:19Z |3b|: i guess it is possible both are broken though
2017-09-21T16:26:25Z stassats: or just interrupts
2017-09-21T16:27:04Z |3b|: well, seems like the interrupts are happening at places they are allowed to
2017-09-21T16:27:40Z |3b|: and if they should be allowed there, the problem is grabbing the mutex from the interruption
2017-09-21T16:27:55Z |3b|: no idea if they should be allowed there or not :)
2017-09-21T16:30:03Z Bike quit (Ping timeout: 248 seconds)
2017-09-21T16:30:42Z Bike joined #sbcl
2017-09-21T16:30:42Z dougk joined #sbcl
2017-09-21T16:35:06Z dougk quit (Ping timeout: 255 seconds)
2017-09-21T16:37:05Z stassats: well, it works on linux and macos, so somethings must be not right on windows
2017-09-21T16:37:54Z |3b|: yeah, seems OK with that change reverted
2017-09-21T16:48:42Z slyrus_ quit (Ping timeout: 248 seconds)
2017-09-21T16:53:56Z stassats: i guess i see, it's precisely because it interrupts in the safe places
2017-09-21T16:54:14Z stassats: but the safe place on safepoint makes it exit condition-wait and grab the mutex again
2017-09-21T16:58:39Z erratum left #sbcl
2017-09-21T16:59:53Z stassats: |3b|: a fix i pushed works here
2017-09-21T17:00:12Z |3b|: to slime or sbcl?
2017-09-21T17:00:19Z stassats: slime
2017-09-21T17:02:25Z |3b|: seems to work here
2017-09-21T17:12:07Z akovalen` quit (Ping timeout: 240 seconds)
2017-09-21T17:21:36Z m00natic quit (Remote host closed the connection)
2017-09-21T17:23:09Z shka_ joined #sbcl
2017-09-21T17:33:40Z dougk joined #sbcl
2017-09-21T17:47:27Z eschatologist quit (Ping timeout: 240 seconds)
2017-09-21T17:47:40Z eschatologist joined #sbcl
2017-09-21T18:21:13Z rgrau quit (Ping timeout: 260 seconds)
2017-09-21T18:38:54Z Jesin joined #sbcl
2017-09-21T19:25:12Z dougk quit (Ping timeout: 255 seconds)
2017-09-21T19:40:54Z dougk joined #sbcl
2017-09-21T19:51:08Z copec joined #sbcl
2017-09-21T19:51:51Z copec: I wonder if someone could fix this link: http://sbcl-internals.cliki.net/  under http://www.sbcl.org/manual/#Internals-Documentation
2017-09-21T19:53:59Z copec: well, google -> http://www.sbcl.org/sbcl-internals/
2017-09-21T19:54:11Z copec: So I guess the manual needs updated.
2017-09-21T20:08:54Z sjl quit (Ping timeout: 240 seconds)
2017-09-21T20:36:27Z edgar-rft joined #sbcl
2017-09-21T20:52:59Z shka_ quit (Ping timeout: 248 seconds)
2017-09-21T21:04:17Z Bicyclidine joined #sbcl
2017-09-21T21:07:20Z Bike quit (Ping timeout: 255 seconds)
2017-09-21T21:08:38Z Bicyclidine quit (Ping timeout: 260 seconds)
2017-09-21T21:20:23Z dougk quit (Ping timeout: 255 seconds)
2017-09-21T21:23:52Z dougk joined #sbcl
2017-09-21T21:41:27Z Bike joined #sbcl
2017-09-21T21:43:54Z abbe_ joined #sbcl
2017-09-21T21:44:33Z jyc quit (Ping timeout: 252 seconds)
2017-09-21T21:44:49Z abbe quit (Disconnected by services)
2017-09-21T21:45:01Z abbe_ is now known as abbe
2017-09-21T21:45:05Z easye` joined #sbcl
2017-09-21T21:45:06Z p_l quit (Ping timeout: 252 seconds)
2017-09-21T21:46:45Z swflint quit (Ping timeout: 252 seconds)
2017-09-21T21:46:45Z easye quit (Ping timeout: 252 seconds)
2017-09-21T21:46:45Z Blkt quit (Ping timeout: 252 seconds)
2017-09-21T21:47:07Z Blkt joined #sbcl
2017-09-21T21:47:08Z lacedaemon joined #sbcl
2017-09-21T21:47:18Z fe[nl]ix quit (Ping timeout: 252 seconds)
2017-09-21T21:48:44Z Bicyclidine joined #sbcl
2017-09-21T21:48:52Z Bicyclidine quit (Remote host closed the connection)
2017-09-21T21:50:59Z Bike quit (Ping timeout: 255 seconds)
2017-09-21T21:52:18Z swflint joined #sbcl
2017-09-21T21:59:59Z jyc joined #sbcl
2017-09-21T22:00:47Z p_l joined #sbcl
2017-09-21T22:02:10Z jamtho joined #sbcl
2017-09-21T22:03:12Z Bike joined #sbcl
2017-09-21T22:14:33Z dougk quit (Ping timeout: 248 seconds)
2017-09-21T22:16:29Z Bicyclidine joined #sbcl
2017-09-21T22:19:12Z Bike quit (Ping timeout: 246 seconds)
2017-09-21T22:19:20Z dougk joined #sbcl
2017-09-21T22:20:30Z Bicyclidine is now known as Bike
2017-09-21T22:24:02Z sjl joined #sbcl
2017-09-21T22:58:58Z angavrilov quit (Remote host closed the connection)
2017-09-21T23:07:21Z jamtho quit (Ping timeout: 248 seconds)
2017-09-21T23:14:09Z cromachina joined #sbcl
2017-09-21T23:14:12Z irsol quit (Remote host closed the connection)
2017-09-21T23:15:19Z irsol joined #sbcl
2017-09-21T23:20:48Z Bike quit (Ping timeout: 252 seconds)
2017-09-21T23:24:39Z Bike joined #sbcl
2017-09-21T23:29:54Z dougk quit (Ping timeout: 240 seconds)
2017-09-21T23:39:58Z karswell_ quit (Read error: Connection reset by peer)
2017-09-22T00:19:53Z attila_lendvai joined #sbcl
2017-09-22T00:40:19Z dougk joined #sbcl
2017-09-22T00:56:48Z muyinliu joined #sbcl
2017-09-22T00:56:57Z karswell joined #sbcl
2017-09-22T01:03:43Z stassats quit (Ping timeout: 260 seconds)
2017-09-22T01:14:43Z muyinliu quit (Quit: Textual IRC Client: www.textualapp.com)
2017-09-22T01:17:23Z dougk quit (Ping timeout: 252 seconds)
2017-09-22T01:20:40Z Jesin quit (Quit: Leaving)
2017-09-22T01:23:02Z attila_lendvai quit (Quit: Leaving.)
2017-09-22T01:29:44Z edgar-rft quit (Read error: Connection reset by peer)
2017-09-22T01:30:52Z dougk joined #sbcl
2017-09-22T01:31:01Z edgar-rft joined #sbcl
2017-09-22T01:45:06Z milanj quit (Quit: This computer has gone to sleep)
2017-09-22T02:22:23Z Bicyclidine joined #sbcl
2017-09-22T02:23:34Z Bike quit (Ping timeout: 240 seconds)
2017-09-22T02:41:41Z slyrus_ joined #sbcl
2017-09-22T02:41:54Z milanj joined #sbcl
2017-09-22T02:56:32Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-22T03:13:27Z shka_ joined #sbcl
2017-09-22T04:17:22Z cromachina_ joined #sbcl
2017-09-22T04:17:23Z cromachina quit (Ping timeout: 260 seconds)
2017-09-22T04:44:53Z milanj quit (Quit: This computer has gone to sleep)
2017-09-22T05:13:30Z knobo joined #sbcl
2017-09-22T05:25:55Z oleo quit (Quit: Leaving)
2017-09-22T05:32:11Z dougk quit (Ping timeout: 240 seconds)
2017-09-22T05:58:59Z scymtym quit (Ping timeout: 252 seconds)
2017-09-22T06:28:54Z dougk joined #sbcl
2017-09-22T06:32:40Z sigjuice quit (Quit: ZNC - http://znc.in)
2017-09-22T06:33:39Z dougk quit (Ping timeout: 264 seconds)
2017-09-22T06:37:51Z sigjuice joined #sbcl
2017-09-22T06:38:10Z easye` quit (Quit: ERC (IRC client for Emacs 25.2.1))
2017-09-22T06:38:26Z easye joined #sbcl
2017-09-22T06:45:22Z milanj joined #sbcl
2017-09-22T07:05:06Z flip214: I get a "no-applicable-method" error that I can't explain...
2017-09-22T07:05:13Z flip214: can somebody take a look and help me?
2017-09-22T07:05:26Z flip214: https://github.com/phmarek/sbcl/tree/no-applicable-method, the no-applicable-method/ directory.
2017-09-22T07:05:43Z anticrisis joined #sbcl
2017-09-22T07:05:44Z flip214: 3 commits - the last one includes the logfile.
2017-09-22T07:06:05Z flip214: basically, I'm changing a DEFSTRUCT; ASDF says the method is newly compiled (and loaded?!)
2017-09-22T07:06:15Z flip214: but then it's not found when called via the REPL
2017-09-22T07:16:09Z |3b|: do you get same results without the structure redefinition?
2017-09-22T07:17:13Z macdavid313 joined #sbcl
2017-09-22T07:17:25Z flip214: |3b|: well, when I'm in that state, it's broken.
2017-09-22T07:17:34Z flip214: If I don't change the structure, everything works fine.
2017-09-22T07:17:58Z |3b|: ok, so specifically talking about stuff not working after redefining a structure?
2017-09-22T07:18:19Z flip214: yes. sorry for not being clear enough.
2017-09-22T07:18:41Z flip214: if you read the logfile, you can see that the GF call in file3.lisp _does_ work as expected.
2017-09-22T07:18:48Z flip214: just the same line in the REPL doesn't.
2017-09-22T07:19:06Z shka_ quit (Ping timeout: 248 seconds)
2017-09-22T07:19:15Z flip214: perhaps it's a compiler/interpreter mismatch?!
2017-09-22T07:19:19Z |3b|: well, pretty sure redefining structures is undefined per spec, so technically conformant behavior :)
2017-09-22T07:20:21Z |3b|: what happened at the "CL-USER> ; Quit to level 1" line?
2017-09-22T07:22:32Z flip214: that's SBCL throwing a compile warning, and I said "retry compiling"
2017-09-22T07:22:55Z flip214: the second time through the file the structure is no longer different, and compilation continues
2017-09-22T07:23:10Z |3b|: right, figured it was something like that, just not sure from the log
2017-09-22T07:23:26Z flip214: I had a paste with quite a few comments....
2017-09-22T07:23:43Z flip214: but that would've required cutting it into the 4 pieces, and I thought the GIT way is easier to follow.
2017-09-22T07:24:01Z flip214: seems I should've left the comments in, and not rely only on my few text lines here.
2017-09-22T07:29:15Z scymtym joined #sbcl
2017-09-22T07:33:20Z flip214: and in case you wonder about the "dirty" sbcl, I set gencgc-card-bytes higher.
2017-09-22T07:34:54Z anticrisis quit (Quit: ERC (IRC client for Emacs 25.3.1))
2017-09-22T07:38:24Z |3b|: flip214: if you remove the call to do-that from the file, does it work on repl? or if you add another call to the file, do both work?
2017-09-22T07:39:33Z |3b| tried something similarly just interactively, and calling gf after redefining struct and method worked once but not 2nd time with no changes in between
2017-09-22T07:40:44Z flip214: |3b|: that's interesting.
2017-09-22T07:41:03Z flip214: repeating the GF line *in the same file* makes the first one go okay
2017-09-22T07:41:06Z flip214: but the next call fails
2017-09-22T07:41:23Z flip214: I only get one output of "Yeah, got #S(FOO :SLOT1 42 :SLOT6 0)!
2017-09-22T07:41:31Z |3b|: yeah
2017-09-22T07:41:34Z flip214: grrr
2017-09-22T07:41:41Z flip214: is that some GF corruption?
2017-09-22T07:42:10Z flip214: the interesting thing is that inspecting the GF shows that it still has the method for a FOO class
2017-09-22T07:43:02Z flip214: but thanks a lot for acknowledging that it seems to be an SBCL bug, and not (directly) my fault!
2017-09-22T07:43:24Z |3b|: well, weakness rather than bug, since spec says it can do anything at that point :)
2017-09-22T07:45:23Z flip214: perhaps *at that point*, but not later on, after a successful call to the GF ;)
2017-09-22T07:45:49Z |3b|: once you did undefined behavior, rest of that session can do anything
2017-09-22T07:46:06Z flip214: as structures are just classes in sbcl anyway, and redefining classes is supported, this looks strange, though
2017-09-22T07:46:44Z flip214: |3b|: well, the compiler itself says that the method is newly compiled - so it should get the new structure type!
2017-09-22T07:46:58Z |3b| wonders if undefined behavior is allowed to be contagious through .fasls... it is allowed to save any random garbage it wants in first session, but presumably the new session should refuse to load bad fasl files
2017-09-22T07:47:22Z knobo: What is immobile space in sbcl?
2017-09-22T07:47:43Z knobo reading commit log
2017-09-22T07:48:21Z |3b|: i don't thing they are /just/ classes, but probably relatively close for GF dispatch
2017-09-22T07:48:57Z |3b| 's uninformed guess would be pcl caches somewhere
2017-09-22T07:49:35Z flip214: knobo: memory ranges that may not be relocated. eg. symbols are stored there.
2017-09-22T07:50:11Z flip214: |3b|: well, the inspector says the parent classes are Class: {[1] #<STANDARD-CLASS COMMON-LISP:STRUCTURE-CLASS> []}
2017-09-22T07:51:40Z |3b|: right, but i think structure-class and standard-class have different rules (no idea if any of them apply here)
2017-09-22T07:53:05Z flip214: If I change DEFSTRUCT to DEFCLASS and add the MAKE-FOO function this works as expected.
2017-09-22T07:53:41Z |3b|: right, defclass has defined behavior in case of redefinition
2017-09-22T07:57:16Z |3b|: yeah, looks like adding another method after the method makes it recalculate whatever is broken, and calling the method on the struct works
2017-09-22T07:58:25Z knobo: flip214: is that because memory address is used as value for eq? or is it any other reason?
2017-09-22T08:00:50Z carenz joined #sbcl
2017-09-22T08:01:21Z angavrilov joined #sbcl
2017-09-22T08:02:40Z slyrus quit (Ping timeout: 240 seconds)
2017-09-22T08:03:50Z kini quit (Remote host closed the connection)
2017-09-22T08:04:07Z christoph_debian quit (Ping timeout: 240 seconds)
2017-09-22T08:05:55Z sebboh` joined #sbcl
2017-09-22T08:06:04Z kini joined #sbcl
2017-09-22T08:07:10Z sebboh quit (Ping timeout: 240 seconds)
2017-09-22T08:07:40Z flip214: knobo: I guess it's much easier and faster if eg. code keeps its address - then functions don't have to do indirect calls after a GC.
2017-09-22T08:16:33Z malm quit (Quit: Bye bye)
2017-09-22T08:18:22Z christoph_debian joined #sbcl
2017-09-22T08:29:57Z dougk joined #sbcl
2017-09-22T08:35:27Z dougk quit (Ping timeout: 264 seconds)
2017-09-22T08:43:45Z macdavid313 quit (Quit: Leaving.)
2017-09-22T09:32:05Z dougk joined #sbcl
2017-09-22T09:36:35Z dougk quit (Ping timeout: 255 seconds)
2017-09-22T09:44:53Z jamtho joined #sbcl
2017-09-22T09:52:05Z dustinm` quit (Quit: Leaving)
2017-09-22T09:56:15Z dustinm` joined #sbcl
2017-09-22T09:59:53Z jamtho quit (Ping timeout: 240 seconds)
2017-09-22T10:43:31Z milanj quit (Quit: This computer has gone to sleep)
2017-09-22T10:57:19Z Bike joined #sbcl
2017-09-22T11:00:46Z slyrus_ quit (Ping timeout: 240 seconds)
2017-09-22T11:08:14Z stassats joined #sbcl
2017-09-22T11:11:50Z nimbus7 joined #sbcl
2017-09-22T11:21:12Z flip214: stassats: I've got a case where after redefining a struct a defmethod call works for the first time afterwards, but not for the second or later calls.
2017-09-22T11:21:40Z flip214: would you like to take a look, or is your answer simply "redefining a structure is unsupported" as per the standard?
2017-09-22T11:21:49Z flip214: I'd fully understand that, BTW.
2017-09-22T11:23:04Z flip214: https://github.com/phmarek/sbcl/tree/no-applicable-method, the no-applicable-method/ directory - logfile included. just in case you're interested, and I'm already gone; thank you very much for any idea!
2017-09-22T11:25:04Z rgrau joined #sbcl
2017-09-22T11:27:34Z stassats: flip214: my answer would be "just use classes if you're using methods anyway"
2017-09-22T11:32:45Z flip214: thanks. we just found it interesting that it works for one call, but not any more.
2017-09-22T11:33:20Z flip214: although the method was recompiled according to ASDF...
2017-09-22T11:33:34Z |3b| suspects the problem is in the GF not the method
2017-09-22T11:34:07Z |3b|: some cache or dfun or whatever
2017-09-22T11:35:01Z milanj joined #sbcl
2017-09-22T11:35:45Z flip214: yeah, of course. but the DEFMETHOD might (or should?) have rebuilt the GF, I believe.
2017-09-22T11:36:09Z flip214: anyway, why does it work for one call?
2017-09-22T11:38:18Z |3b|: i think it doesn't need to rebuild GF stuff because it is still the same class
2017-09-22T11:38:52Z |3b|: no idea why it works once though
2017-09-22T11:39:40Z |3b|: possibly it changes what's in a cache and once it breaks it doesn't change anymore or something
2017-09-22T11:42:35Z flip214: sadly it's undefined behaviour... that immediately excludes lots of mindshare ;(
2017-09-22T11:45:25Z |3b|: yeah, being deep in PCL probably doesn't help :p
2017-09-22T12:15:20Z ryan_vw_ quit (Quit: leaving)
2017-09-22T12:34:18Z dougk joined #sbcl
2017-09-22T12:35:39Z Bike quit (Ping timeout: 246 seconds)
2017-09-22T12:38:33Z dougk quit (Ping timeout: 248 seconds)
2017-09-22T12:53:49Z macdavid313 joined #sbcl
2017-09-22T12:54:32Z Bike joined #sbcl
2017-09-22T13:03:22Z sjl quit (Ping timeout: 240 seconds)
2017-09-22T13:08:59Z oleo joined #sbcl
2017-09-22T13:47:41Z cromachina_ quit (Read error: Connection reset by peer)
2017-09-22T14:16:48Z nimbus7 quit (Quit: Leaving)
2017-09-22T14:17:03Z macdavid313 left #sbcl
2017-09-22T14:47:02Z dougk joined #sbcl
2017-09-22T14:56:02Z dougk_: flip214: there is no SBCL bug involving structure redefinition. Expect nothing to work after adding a slot. And don't try to reason about what methods are attached to a GF that had the old definition until you understand all of the PCL implementation and everything in src/code/*class*.lisp
2017-09-22T15:02:38Z stassats: it uses the class name to get cached functions
2017-09-22T15:02:45Z stassats: and the function refers to an invalid layout
2017-09-22T15:17:54Z stassats: i have several options: get rid of fngen and just always compile functions
2017-09-22T15:18:08Z stassats: slow down the test so that it works on redefinition
2017-09-22T15:18:29Z stassats: change the cache test to include the layout
2017-09-22T15:18:43Z stassats: or invalidate the function with invalid layouts in constants
2017-09-22T15:19:07Z stassats: i love removing code, so the first option is closer to me, but it might slow down compilation
2017-09-22T15:19:40Z stassats: or rather, first unoptimized hits
2017-09-22T15:20:07Z stassats: but maybe it was only a concern in 1985
2017-09-22T15:23:36Z stassats: for me, i only see two options a) do nothing b) remove code
2017-09-22T15:24:26Z dougk_: i would do nothing. why does anyone want to claim that "correctly" dealing with structure definition is an issue? It's disallowed and we make only a best effort to support it at all
2017-09-22T15:24:28Z stassats: also gets rid of the garbage retained by SB-PCL::*FGENS*
2017-09-22T15:24:53Z dougk_: s/definition/redefinition/
2017-09-22T15:30:01Z stassats: really, how important is fngen?
2017-09-22T15:32:32Z dougk_: refraining from at-runtime compilation is very important until we remove the world lock from compilation
2017-09-22T15:32:47Z dougk_: otherwise it would essentially single-thread anything that needs to compute a discriminating fun
2017-09-22T15:33:16Z stassats: that would have poor performance anyway
2017-09-22T15:37:24Z blackwolf joined #sbcl
2017-09-22T15:39:08Z stassats: anyway, i'm onto the pfdietz thing
2017-09-22T15:39:19Z stassats: something's wrong with constraint propagation
2017-09-22T15:39:49Z stassats: (i hate these parts)
2017-09-22T15:46:15Z stassats: why does a set confuse it?
2017-09-22T15:46:48Z DeadTrickster joined #sbcl
2017-09-22T16:10:37Z joshe quit (Ping timeout: 246 seconds)
2017-09-22T16:10:53Z joshe joined #sbcl
2017-09-22T16:23:10Z sebboh` quit (Changing host)
2017-09-22T16:23:10Z sebboh` joined #sbcl
2017-09-22T16:23:20Z sebboh` is now known as sebboh
2017-09-22T16:25:32Z Jesin joined #sbcl
2017-09-22T16:35:20Z dougk quit (Ping timeout: 252 seconds)
2017-09-22T16:57:43Z dougk joined #sbcl
2017-09-22T16:59:54Z carenz quit (Ping timeout: 248 seconds)
2017-09-22T17:06:38Z BitPuffin|osx joined #sbcl
2017-09-22T17:08:12Z oleo quit (Remote host closed the connection)
2017-09-22T17:09:14Z oleo joined #sbcl
2017-09-22T17:17:21Z oleo quit (Remote host closed the connection)
2017-09-22T17:33:52Z oleo joined #sbcl
2017-09-22T17:50:00Z sjl joined #sbcl
2017-09-22T17:58:52Z sjl quit (Ping timeout: 240 seconds)
2017-09-22T17:59:21Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-22T18:45:39Z stassats: constraint-propagate actually seems to be ok
2017-09-22T19:04:28Z stassats: so that makes a fix easy
2017-09-22T19:26:11Z dougk quit (Ping timeout: 246 seconds)
2017-09-22T19:29:13Z scymtym quit (Ping timeout: 255 seconds)
2017-09-22T19:53:55Z oleo quit (Remote host closed the connection)
2017-09-22T19:56:33Z dougk joined #sbcl
2017-09-22T20:02:08Z oleo joined #sbcl
2017-09-22T20:05:13Z dougk quit (Ping timeout: 255 seconds)
2017-09-22T20:22:09Z scymtym joined #sbcl
2017-09-22T20:25:54Z dougk joined #sbcl
2017-09-22T21:05:46Z Bike quit (Ping timeout: 240 seconds)
2017-09-22T21:18:37Z sjl joined #sbcl
2017-09-22T21:25:54Z dougk__ joined #sbcl
2017-09-22T21:27:02Z blackwolf quit (Quit: ERC (IRC client for Emacs 25.2.1))
2017-09-22T21:28:06Z dougk quit (Ping timeout: 255 seconds)
2017-09-22T21:29:10Z rgrau quit (Ping timeout: 240 seconds)
2017-09-22T21:40:40Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-22T21:41:04Z Bike joined #sbcl
2017-09-22T21:41:34Z dougk joined #sbcl
2017-09-22T21:46:01Z dougk quit (Ping timeout: 255 seconds)
2017-09-22T21:54:54Z Jesin quit (Quit: Leaving)
2017-09-22T21:55:08Z Shinmera joined #sbcl
2017-09-22T22:14:54Z angavrilov quit (Remote host closed the connection)
2017-09-22T22:22:28Z Bike quit (Ping timeout: 255 seconds)
2017-09-22T22:23:10Z jamtho joined #sbcl
2017-09-22T22:25:03Z Bike joined #sbcl
2017-09-22T22:36:33Z nimbus7 joined #sbcl
2017-09-22T22:41:26Z nimbus7 quit (Ping timeout: 240 seconds)
2017-09-22T22:49:46Z jamtho quit (Ping timeout: 240 seconds)
2017-09-22T22:55:18Z jamtho joined #sbcl
2017-09-22T22:57:21Z nimbus7 joined #sbcl
2017-09-22T23:26:06Z dougk joined #sbcl
2017-09-22T23:35:38Z nimbus7 quit (Ping timeout: 248 seconds)
2017-09-22T23:36:09Z nimbus7 joined #sbcl
2017-09-22T23:43:38Z jamtho quit (Ping timeout: 248 seconds)
2017-09-22T23:51:10Z stassats quit (Ping timeout: 240 seconds)
2017-09-23T00:18:27Z dougk quit (Ping timeout: 260 seconds)
2017-09-23T00:21:11Z stassats joined #sbcl
2017-09-23T00:22:57Z dougk joined #sbcl
2017-09-23T00:31:05Z dougk quit (Ping timeout: 240 seconds)
2017-09-23T00:50:02Z dougk joined #sbcl
2017-09-23T00:54:09Z dougk quit (Ping timeout: 240 seconds)
2017-09-23T00:56:51Z stassats quit (Ping timeout: 240 seconds)
2017-09-23T01:01:52Z dougk joined #sbcl
2017-09-23T01:17:04Z attila_lendvai joined #sbcl
2017-09-23T01:17:04Z attila_lendvai quit (Changing host)
2017-09-23T01:17:04Z attila_lendvai joined #sbcl
2017-09-23T01:26:16Z Bicyclidine joined #sbcl
2017-09-23T01:28:14Z Bike quit (Ping timeout: 264 seconds)
2017-09-23T01:29:02Z dougk quit (Ping timeout: 260 seconds)
2017-09-23T01:51:51Z sjl_ joined #sbcl
2017-09-23T01:52:57Z sjl quit (Ping timeout: 260 seconds)
2017-09-23T01:59:27Z attila_lendvai quit (Quit: Leaving.)
2017-09-23T02:10:17Z Gorgias joined #sbcl
2017-09-23T02:10:19Z Gorgias quit (Remote host closed the connection)
2017-09-23T02:12:09Z Gorgias joined #sbcl
2017-09-23T02:14:13Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-23T02:18:16Z nimbus7 quit (Quit: Leaving)
2017-09-23T02:21:47Z Gorgias is now known as Bike
2017-09-23T02:23:05Z milanj quit (Quit: This computer has gone to sleep)
2017-09-23T02:30:32Z sjl joined #sbcl
2017-09-23T02:31:07Z sjl_ quit (Ping timeout: 241 seconds)
2017-09-23T03:43:34Z Bike quit (Ping timeout: 252 seconds)
2017-09-23T05:03:37Z karswell quit (Remote host closed the connection)
2017-09-23T05:04:23Z karswell joined #sbcl
2017-09-23T05:06:39Z karswell quit (Remote host closed the connection)
2017-09-23T05:07:23Z karswell joined #sbcl
2017-09-23T05:10:41Z Intensity quit (Ping timeout: 246 seconds)
2017-09-23T05:28:37Z karswell_ joined #sbcl
2017-09-23T05:29:08Z karswell quit (Remote host closed the connection)
2017-09-23T05:43:13Z angavrilov joined #sbcl
2017-09-23T06:04:59Z eschatologist quit (Quit: ZNC 1.6.4+deb1 - http://znc.in)
2017-09-23T06:05:58Z eschatologist joined #sbcl
2017-09-23T06:52:22Z Intensity joined #sbcl
2017-09-23T07:21:35Z Intensity quit (Ping timeout: 246 seconds)
2017-09-23T07:28:30Z carenz joined #sbcl
2017-09-23T07:30:10Z Intensity joined #sbcl
2017-09-23T07:33:24Z dougk joined #sbcl
2017-09-23T07:48:51Z jamtho joined #sbcl
2017-09-23T07:49:14Z jamtho quit (Read error: Connection reset by peer)
2017-09-23T08:11:16Z DeadTrickster quit (Remote host closed the connection)
2017-09-23T08:19:50Z carenz quit (Ping timeout: 240 seconds)
2017-09-23T08:28:51Z milanj joined #sbcl
2017-09-23T08:30:55Z DeadTrickster joined #sbcl
2017-09-23T08:41:38Z leah2 quit (Quit: trotz alledem!)
2017-09-23T09:58:32Z Posterdati: hi
2017-09-23T09:59:05Z Posterdati: please is there anyone that checked get-decoded-time behaviour on sbcl?
2017-09-23T10:00:10Z |3b|: checked?
2017-09-23T10:00:26Z Posterdati: it seems to be wrong
2017-09-23T10:00:44Z |3b|: looks reasonable here
2017-09-23T10:01:13Z Posterdati: try ecl
2017-09-23T10:01:21Z Posterdati: it returns -2
2017-09-23T10:01:25Z Posterdati: sbcl -1
2017-09-23T10:01:34Z Posterdati: as timezone
2017-09-23T10:01:51Z Posterdati: for cet I men
2017-09-23T10:03:32Z Posterdati: should be -2 for CET
2017-09-23T10:03:36Z Shinmera: cet is -1. cest is -2.
2017-09-23T10:04:15Z Posterdati: anyway the CEST (with saving time)
2017-09-23T10:04:45Z |3b|: hmm, that does look wrong if i'm reading spec correctly
2017-09-23T10:05:32Z Posterdati: anyway get-decoded-time returns different on ecl and sbcl
2017-09-23T10:08:36Z Posterdati: Local time is now:      Sat Sep 23 12:09:13 CEST 2017.
2017-09-23T10:08:43Z Posterdati: Universal Time is now:  Sat Sep 23 10:09:13 UTC 2017.
2017-09-23T10:08:49Z Posterdati: so -2
2017-09-23T10:08:58Z Posterdati: sbcl is wrong then
2017-09-23T10:10:38Z |3b|: actually, not sure which is wrong, spec is confusing :/
2017-09-23T10:11:27Z Posterdati: now timezone depends on time savings
2017-09-23T10:12:03Z |3b|: "(When ``daylight savings time'' is separately represented as an argument or return value, the time zone that accompanies it does not depend on whether daylight savings time is in effect.) "
2017-09-23T10:17:45Z scymtym quit (Ping timeout: 248 seconds)
2017-09-23T10:19:24Z Posterdati: strange
2017-09-23T10:20:33Z Shinmera: Well, SBCL does emit a T if DST is in effect, so I guess according to that having the TZ be -1 is fine.
2017-09-23T10:37:05Z dougk quit (Ping timeout: 240 seconds)
2017-09-23T10:40:41Z attila_lendvai joined #sbcl
2017-09-23T10:40:41Z attila_lendvai quit (Changing host)
2017-09-23T10:40:41Z attila_lendvai joined #sbcl
2017-09-23T10:53:26Z stylewarning: can SAPs be stack allocated?
2017-09-23T10:54:21Z scymtym joined #sbcl
2017-09-23T11:06:11Z Bike joined #sbcl
2017-09-23T11:10:27Z oleo quit (Quit: Leaving)
2017-09-23T11:10:46Z Bicyclidine joined #sbcl
2017-09-23T11:12:39Z Bike quit (Ping timeout: 246 seconds)
2017-09-23T11:31:44Z leah2 joined #sbcl
2017-09-23T11:33:55Z dougk joined #sbcl
2017-09-23T11:38:13Z dougk quit (Ping timeout: 252 seconds)
2017-09-23T12:34:55Z dougk joined #sbcl
2017-09-23T12:36:18Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-23T12:39:16Z dougk quit (Ping timeout: 252 seconds)
2017-09-23T12:51:38Z Bike joined #sbcl
2017-09-23T12:51:49Z Posterdati: Shinmera: the same does ecl, but tz = -2
2017-09-23T12:56:17Z milanj quit (Quit: This computer has gone to sleep)
2017-09-23T13:19:01Z oleo joined #sbcl
2017-09-23T13:35:55Z dougk joined #sbcl
2017-09-23T13:40:23Z dougk quit (Ping timeout: 248 seconds)
2017-09-23T13:58:36Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-23T13:59:16Z DeadTrickster joined #sbcl
2017-09-23T14:37:01Z dougk joined #sbcl
2017-09-23T14:41:11Z dougk quit (Ping timeout: 248 seconds)
2017-09-23T15:36:10Z Bicyclidine joined #sbcl
2017-09-23T15:37:54Z Bike quit (Ping timeout: 255 seconds)
2017-09-23T15:38:25Z dougk joined #sbcl
2017-09-23T15:43:01Z dougk quit (Ping timeout: 260 seconds)
2017-09-23T15:56:52Z attila_lendvai quit (Quit: Leaving.)
2017-09-23T15:59:14Z Bike joined #sbcl
2017-09-23T16:01:45Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-23T16:49:37Z scymtym quit (Ping timeout: 246 seconds)
2017-09-23T16:53:16Z nicdev quit (Remote host closed the connection)
2017-09-23T16:54:04Z nicdev joined #sbcl
2017-09-23T17:14:01Z scymtym joined #sbcl
2017-09-23T17:21:18Z oleo quit (Quit: Leaving)
2017-09-23T17:34:57Z milanj joined #sbcl
2017-09-23T17:36:29Z dougk joined #sbcl
2017-09-23T17:48:58Z oleo joined #sbcl
2017-09-23T17:52:00Z milanj quit (Quit: This computer has gone to sleep)
2017-09-23T17:55:36Z dougk quit (Ping timeout: 255 seconds)
2017-09-23T18:01:02Z dougk joined #sbcl
2017-09-23T18:14:43Z flip214: dougk_: well, the strange thing is that the first call still works. only the second and later ones break.
2017-09-23T18:27:11Z dougk quit (Read error: Connection reset by peer)
2017-09-23T18:27:28Z dougk joined #sbcl
2017-09-23T18:41:07Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-23T18:41:33Z DeadTrickster joined #sbcl
2017-09-23T18:58:09Z dougk quit (Ping timeout: 240 seconds)
2017-09-23T19:07:24Z attila_lendvai joined #sbcl
2017-09-23T19:13:05Z attila_lendvai quit (Quit: Leaving.)
2017-09-23T19:17:25Z Bicyclidine joined #sbcl
2017-09-23T19:21:06Z Bike quit (Ping timeout: 255 seconds)
2017-09-23T19:40:20Z oleo quit (Quit: Leaving)
2017-09-23T19:42:07Z DeadTrickster quit (Read error: Connection reset by peer)
2017-09-23T19:42:31Z DeadTrickster joined #sbcl
2017-09-23T19:47:43Z Jesin joined #sbcl
2017-09-23T20:00:44Z Jesin quit (Quit: Leaving)
2017-09-23T20:04:47Z akovalenko joined #sbcl
2017-09-23T20:12:10Z stassats joined #sbcl
2017-09-23T20:12:19Z stassats: flip214: there's nothing strange about that
2017-09-23T20:20:21Z Jesin joined #sbcl
2017-09-23T20:26:08Z stassats quit (Ping timeout: 240 seconds)
2017-09-23T20:28:03Z Bike joined #sbcl
2017-09-23T20:31:31Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-23T20:35:25Z Bike quit (Ping timeout: 264 seconds)
2017-09-23T20:41:39Z Bike joined #sbcl
2017-09-23T20:42:30Z Bicyclidine joined #sbcl
2017-09-23T20:46:00Z Bike quit (Ping timeout: 252 seconds)
2017-09-23T21:34:46Z stassats joined #sbcl
2017-09-23T21:47:25Z Bicyclidine quit (Ping timeout: 264 seconds)
2017-09-23T21:58:06Z stassats` joined #sbcl
2017-09-23T22:05:26Z Bike joined #sbcl
2017-09-23T22:09:59Z angavrilov quit (Remote host closed the connection)
2017-09-23T22:40:40Z stassats` quit (Ping timeout: 260 seconds)
2017-09-23T22:44:49Z dougk joined #sbcl
2017-09-23T22:49:26Z milanj joined #sbcl
2017-09-23T22:53:51Z Jesin quit (Quit: Leaving)
2017-09-23T22:54:18Z dougk quit (Read error: Connection reset by peer)
2017-09-23T22:54:32Z dougk joined #sbcl
2017-09-23T23:19:11Z Bike quit (Ping timeout: 240 seconds)
2017-09-23T23:32:02Z Bike joined #sbcl
2017-09-23T23:43:41Z Jesin joined #sbcl
2017-09-24T00:00:22Z attila_lendvai joined #sbcl
2017-09-24T00:42:48Z stassats quit (Ping timeout: 248 seconds)
2017-09-24T01:29:48Z attila_lendvai quit (Read error: Connection reset by peer)
2017-09-24T01:30:44Z attila_lendvai joined #sbcl
2017-09-24T01:30:44Z attila_lendvai quit (Changing host)
2017-09-24T01:30:44Z attila_lendvai joined #sbcl
2017-09-24T02:20:57Z attila_lendvai quit (Ping timeout: 252 seconds)
2017-09-24T02:23:43Z flip214 quit (Ping timeout: 255 seconds)
2017-09-24T02:24:54Z flip214 joined #sbcl
2017-09-24T02:24:54Z flip214 quit (Changing host)
2017-09-24T02:24:54Z flip214 joined #sbcl
2017-09-24T02:35:47Z milanj quit (Quit: This computer has gone to sleep)
2017-09-24T04:06:11Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T04:28:46Z dougk joined #sbcl
2017-09-24T06:15:05Z dougk quit (Ping timeout: 260 seconds)
2017-09-24T06:52:01Z angavrilov joined #sbcl
2017-09-24T08:18:42Z milanj joined #sbcl
2017-09-24T08:49:00Z oleo joined #sbcl
2017-09-24T08:55:50Z Bike quit (Ping timeout: 252 seconds)
2017-09-24T09:44:15Z scymtym quit (Remote host closed the connection)
2017-09-24T09:51:48Z Bike joined #sbcl
2017-09-24T09:58:20Z Bike quit (Ping timeout: 246 seconds)
2017-09-24T10:18:05Z scymtym joined #sbcl
2017-09-24T10:32:12Z Xof joined #sbcl
2017-09-24T10:42:20Z Bike joined #sbcl
2017-09-24T11:00:44Z shka_ joined #sbcl
2017-09-24T11:37:49Z milanj quit (Quit: This computer has gone to sleep)
2017-09-24T11:42:05Z shka_ quit (Quit: Konversation terminated!)
2017-09-24T11:58:41Z shka joined #sbcl
2017-09-24T12:00:10Z milanj joined #sbcl
2017-09-24T12:35:06Z attila_lendvai joined #sbcl
2017-09-24T12:35:06Z attila_lendvai quit (Changing host)
2017-09-24T12:35:06Z attila_lendvai joined #sbcl
2017-09-24T12:42:35Z milanj quit (Quit: This computer has gone to sleep)
2017-09-24T12:58:41Z dougk joined #sbcl
2017-09-24T13:13:12Z shka_ joined #sbcl
2017-09-24T13:13:12Z shka quit (Ping timeout: 264 seconds)
2017-09-24T13:15:23Z shka_ quit (Excess Flood)
2017-09-24T13:15:45Z attila_lendvai quit (Quit: Leaving.)
2017-09-24T13:17:16Z shka_ joined #sbcl
2017-09-24T14:48:35Z shka joined #sbcl
2017-09-24T14:48:39Z shka_ quit (Ping timeout: 248 seconds)
2017-09-24T14:50:40Z shka quit (Excess Flood)
2017-09-24T14:51:32Z shka joined #sbcl
2017-09-24T14:59:44Z dougk__ joined #sbcl
2017-09-24T15:02:17Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T15:31:29Z specbot quit (Remote host closed the connection)
2017-09-24T15:31:29Z minion quit (Remote host closed the connection)
2017-09-24T15:31:34Z easye quit (Read error: Connection reset by peer)
2017-09-24T15:32:51Z specbot joined #sbcl
2017-09-24T15:33:30Z minion joined #sbcl
2017-09-24T15:38:37Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-24T15:51:41Z Jesin quit (Quit: Leaving)
2017-09-24T15:58:13Z scymtym quit (Ping timeout: 255 seconds)
2017-09-24T16:09:28Z dougk joined #sbcl
2017-09-24T16:13:25Z dougk quit (Read error: Connection reset by peer)
2017-09-24T16:13:42Z dougk joined #sbcl
2017-09-24T16:22:25Z stassats joined #sbcl
2017-09-24T16:25:07Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T17:17:32Z scymtym joined #sbcl
2017-09-24T17:22:02Z dougk joined #sbcl
2017-09-24T17:46:47Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T18:05:37Z karswell_ quit (Ping timeout: 240 seconds)
2017-09-24T18:23:20Z dougk joined #sbcl
2017-09-24T18:34:09Z dougk quit (Ping timeout: 260 seconds)
2017-09-24T18:36:38Z Bike quit (Ping timeout: 252 seconds)
2017-09-24T18:40:15Z Bike joined #sbcl
2017-09-24T18:41:18Z dougk joined #sbcl
2017-09-24T18:43:07Z cromachina joined #sbcl
2017-09-24T19:20:07Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T19:21:22Z shka quit (Remote host closed the connection)
2017-09-24T19:22:02Z shka joined #sbcl
2017-09-24T19:34:58Z malm joined #sbcl
2017-09-24T20:07:05Z Jesin joined #sbcl
2017-09-24T20:20:48Z Bicyclidine joined #sbcl
2017-09-24T20:24:23Z Bike quit (Ping timeout: 246 seconds)
2017-09-24T20:25:35Z Jesin quit (Quit: Leaving)
2017-09-24T20:32:32Z Bike joined #sbcl
2017-09-24T20:34:53Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-24T20:53:27Z stassats quit (Ping timeout: 248 seconds)
2017-09-24T21:15:35Z sjl quit (Ping timeout: 252 seconds)
2017-09-24T21:19:18Z sjl joined #sbcl
2017-09-24T21:20:04Z knobo: When is next sbcl comming?
2017-09-24T21:23:02Z scymtym: knobo: around the end of the month
2017-09-24T21:23:28Z knobo: great :)
2017-09-24T21:26:04Z milanj joined #sbcl
2017-09-24T21:42:07Z Jesin joined #sbcl
2017-09-24T22:10:37Z shka quit (Ping timeout: 240 seconds)
2017-09-24T22:23:07Z stassats joined #sbcl
2017-09-24T22:59:14Z dougk joined #sbcl
2017-09-24T23:01:12Z christoph_debian quit (Ping timeout: 264 seconds)
2017-09-24T23:13:05Z christoph_debian joined #sbcl
2017-09-24T23:20:23Z sjl quit (Quit: WeeChat 1.3)
2017-09-24T23:27:06Z sjl joined #sbcl
2017-09-24T23:27:37Z dougk quit (Ping timeout: 240 seconds)
2017-09-24T23:41:09Z dougk joined #sbcl
2017-09-24T23:59:16Z Bicyclidine joined #sbcl
2017-09-25T00:01:23Z Bike quit (Ping timeout: 240 seconds)
2017-09-25T00:26:19Z Bike joined #sbcl
2017-09-25T00:28:16Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-25T00:45:03Z stassats quit (Ping timeout: 240 seconds)
2017-09-25T00:48:45Z Bicyclidine joined #sbcl
2017-09-25T00:52:30Z Bike quit (Ping timeout: 240 seconds)
2017-09-25T03:19:16Z dougk quit (Ping timeout: 240 seconds)
2017-09-25T04:01:37Z jibanes quit (Ping timeout: 240 seconds)
2017-09-25T04:03:49Z jibanes joined #sbcl
2017-09-25T04:16:03Z dougk joined #sbcl
2017-09-25T04:20:23Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T04:26:44Z milanj quit (Quit: This computer has gone to sleep)
2017-09-25T04:47:54Z shka joined #sbcl
2017-09-25T04:48:07Z sjl quit (Ping timeout: 240 seconds)
2017-09-25T04:52:56Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-25T05:17:02Z dougk joined #sbcl
2017-09-25T05:21:17Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T05:55:46Z knobo quit (Ping timeout: 240 seconds)
2017-09-25T05:56:17Z oleo quit (Quit: Leaving)
2017-09-25T06:17:33Z dougk joined #sbcl
2017-09-25T06:21:50Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T06:25:12Z knobo joined #sbcl
2017-09-25T07:54:09Z dougk_ quit (Ping timeout: 240 seconds)
2017-09-25T07:58:12Z carenz joined #sbcl
2017-09-25T08:06:54Z dougk_ joined #sbcl
2017-09-25T08:19:48Z dougk joined #sbcl
2017-09-25T08:24:02Z dougk quit (Ping timeout: 240 seconds)
2017-09-25T08:29:32Z milanj joined #sbcl
2017-09-25T10:21:02Z dougk joined #sbcl
2017-09-25T10:25:06Z dougk quit (Ping timeout: 240 seconds)
2017-09-25T10:37:12Z BitPuffin|osx joined #sbcl
2017-09-25T10:51:50Z rgrau joined #sbcl
2017-09-25T11:22:11Z Bike joined #sbcl
2017-09-25T12:11:03Z stassats joined #sbcl
2017-09-25T12:15:33Z BitPuffin|osx quit (Ping timeout: 248 seconds)
2017-09-25T12:52:47Z jsjolen joined #sbcl
2017-09-25T12:57:49Z jsjolen: Hi, I'm currently going thru my OS course at uni. We get caught that there're 3 segments, heap, stack and text (which is non-extendable[?]) and of these 3 only text is executable. How does SBCL execute code if you can't extend the text segment?
2017-09-25T12:59:13Z foom2 joined #sbcl
2017-09-25T13:01:57Z foom quit (Ping timeout: 246 seconds)
2017-09-25T13:04:01Z dougk joined #sbcl
2017-09-25T13:12:55Z Bike: those are segments in a file, not a requirement of memory
2017-09-25T13:16:23Z Shinmera: Well, at least on Linux, you do have text, data, bss, heap, mmap, stack, kernel sections in memory.
2017-09-25T13:17:43Z Shinmera: Also on Linux, read up on mmap and mprotect. Basically you can give memory pages certain attributes, such as being executable.
2017-09-25T13:33:18Z dougk quit (Ping timeout: 255 seconds)
2017-09-25T13:33:39Z jsjolen: Shinmera: Aha
2017-09-25T13:34:53Z jsjolen: So SBCL just tells the kernel to make it executable
2017-09-25T13:38:09Z lacedaemon is now known as fe[nl]ix
2017-09-25T13:40:23Z rgrau quit (Ping timeout: 260 seconds)
2017-09-25T13:52:09Z oleo joined #sbcl
2017-09-25T14:12:13Z stassats` joined #sbcl
2017-09-25T14:14:07Z stassats`: jsjolen: that's from the old times
2017-09-25T14:14:17Z stassats`: memory segmentation is no longer used
2017-09-25T14:14:30Z stassats`: it's all just a big chunk of memory
2017-09-25T14:16:06Z blackwolf joined #sbcl
2017-09-25T14:16:31Z oleo quit (Quit: Leaving)
2017-09-25T14:17:24Z jsjolen: stassats`: But I guess there's some sort of internal segmentation in that big chunk of memory since Linux reports it as such? It's just all contigious now
2017-09-25T14:18:24Z oleo joined #sbcl
2017-09-25T14:18:56Z pkhuong_: you just ask the OS for ranges of (virtual) memory with mmap.
2017-09-25T14:20:16Z jsjolen: pkhuong_: and set up your stack and heap yourself?
2017-09-25T14:20:17Z stassats`: jsjolen: the MMU maps virtual address to real addresses in the big chunk of memory
2017-09-25T14:20:49Z stassats`: so both firefox and sbcl have 0xabc which points to different RAM locations
2017-09-25T14:21:07Z stassats`: jsjolen: there's no distinction between stack and heap
2017-09-25T14:21:38Z stassats`: stack is where [ER]SP points, that's it
2017-09-25T14:23:11Z jsjolen: stassats`: Yeah, that I know :-). The literature I read has shown how to calc. the mapping but described the heap and stack as being discontigious
2017-09-25T14:24:09Z stassats`: the stack grows downward, so it's useful to put it high up
2017-09-25T14:24:24Z jsjolen: Have them grow in opposite directions
2017-09-25T14:24:33Z jsjolen: them being heap & stack
2017-09-25T14:24:42Z pkhuong_: each process/thread has an initial stack, but it's just a suggestion on linux. You can point SP anywhere you want. The heap used to refer to the region at the bottom of the address space that grew with (s)brk, but we have mmap now.
2017-09-25T14:24:54Z stassats`: it's also safer to separate from the heap to somewhat protect from blown stacks
2017-09-25T14:25:48Z stassats`: usually, a single protected page does the job
2017-09-25T14:28:14Z jsjolen: Aha nice, pkhuong_. What you're saying about (s)brk reflects what I've read
2017-09-25T14:31:06Z jsjolen: Gotta restart -- won't log in again. Thank you for informing me
2017-09-25T14:31:12Z jsjolen quit (Remote host closed the connection)
2017-09-25T14:40:52Z cromachina quit (Read error: Connection reset by peer)
2017-09-25T14:51:58Z oleo quit (Remote host closed the connection)
2017-09-25T14:52:35Z oleo joined #sbcl
2017-09-25T15:00:54Z dougk joined #sbcl
2017-09-25T15:05:20Z oleo quit (Quit: Leaving)
2017-09-25T15:13:08Z stassats` quit (Ping timeout: 260 seconds)
2017-09-25T15:19:29Z rgrau joined #sbcl
2017-09-25T15:24:08Z dougk__ joined #sbcl
2017-09-25T15:24:48Z dougk quit (Ping timeout: 255 seconds)
2017-09-25T15:31:48Z knobo quit (Ping timeout: 260 seconds)
2017-09-25T15:32:37Z dougk joined #sbcl
2017-09-25T15:33:09Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-25T15:49:36Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T15:52:51Z dougk joined #sbcl
2017-09-25T16:00:42Z stassats` joined #sbcl
2017-09-25T16:01:32Z dougk__ joined #sbcl
2017-09-25T16:02:07Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T16:05:49Z stassats` quit (Ping timeout: 252 seconds)
2017-09-25T16:08:23Z stassats` joined #sbcl
2017-09-25T16:15:04Z oleo joined #sbcl
2017-09-25T16:32:45Z rgrau quit (Ping timeout: 240 seconds)
2017-09-25T16:33:37Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-25T16:42:22Z dougk joined #sbcl
2017-09-25T16:57:06Z stassats` quit (Remote host closed the connection)
2017-09-25T16:57:23Z stassats` joined #sbcl
2017-09-25T16:58:55Z Jesin quit (Quit: Leaving)
2017-09-25T16:59:57Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T17:02:06Z Jesin joined #sbcl
2017-09-25T17:07:47Z dougk joined #sbcl
2017-09-25T17:15:40Z carenz quit (Ping timeout: 252 seconds)
2017-09-25T17:22:15Z dougk quit (Ping timeout: 255 seconds)
2017-09-25T17:55:05Z Jesin quit (Quit: Leaving)
2017-09-25T17:58:34Z Jesin joined #sbcl
2017-09-25T18:04:53Z Jesin quit (Quit: brb rebooting)
2017-09-25T18:06:00Z dougk joined #sbcl
2017-09-25T18:20:45Z dougk quit (Ping timeout: 255 seconds)
2017-09-25T18:24:35Z Jesin joined #sbcl
2017-09-25T18:31:00Z dougk joined #sbcl
2017-09-25T18:31:42Z karswell_ joined #sbcl
2017-09-25T18:45:19Z dougk quit (Ping timeout: 252 seconds)
2017-09-25T18:51:16Z stassats` quit (Ping timeout: 248 seconds)
2017-09-25T18:59:00Z stassats` joined #sbcl
2017-09-25T19:20:23Z stassats quit (Ping timeout: 264 seconds)
2017-09-25T19:32:57Z stylewarning: does SBCL have superlinear time complexity for compilation unit size?
2017-09-25T19:34:05Z stassats`: what?
2017-09-25T19:42:25Z dougk joined #sbcl
2017-09-25T19:46:28Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T19:48:27Z pkhuong_: I don't think SBCL is carries enough state across toplevel functions to be capable of superlinear time.
2017-09-25T19:52:43Z Shinmera: Doesn't any kind of data flow analysis incur non-linear complexity?
2017-09-25T19:52:48Z dougk joined #sbcl
2017-09-25T19:52:49Z stassats` quit (Remote host closed the connection)
2017-09-25T19:53:03Z stassats` joined #sbcl
2017-09-25T19:53:14Z foom2 is now known as foom
2017-09-25T19:53:20Z foom: Within a function, sure.
2017-09-25T19:56:22Z Xof quit (Ping timeout: 248 seconds)
2017-09-25T19:58:13Z Shinmera: Ah, right.
2017-09-25T20:06:56Z stassats joined #sbcl
2017-09-25T20:09:14Z shka quit (Ping timeout: 240 seconds)
2017-09-25T20:13:35Z angavrilov quit (Remote host closed the connection)
2017-09-25T20:43:10Z dougk quit (Ping timeout: 246 seconds)
2017-09-25T20:51:11Z dougk joined #sbcl
2017-09-25T21:00:42Z rgrau joined #sbcl
2017-09-25T21:05:23Z sjl joined #sbcl
2017-09-25T21:23:00Z dougk quit (Ping timeout: 255 seconds)
2017-09-25T21:28:19Z stassats` quit (Remote host closed the connection)
2017-09-25T21:35:12Z stylewarning: Surprisingly SBCL is way faster than CCL at compiling this huge file containing 100s of generated CFFI bindings
2017-09-25T21:35:41Z stylewarning: I do have lots of pushnew’s at the top level though. Maybe that’s why.
2017-09-25T21:36:14Z stylewarning: But it’s something like 5x faster, SBCL
2017-09-25T22:18:46Z blackwolf quit (Quit: ERC (IRC client for Emacs 25.2.1))
2017-09-25T22:36:34Z milanj_ joined #sbcl
2017-09-25T22:38:14Z milanj quit (Ping timeout: 255 seconds)
2017-09-25T22:54:11Z dougk joined #sbcl
2017-09-25T23:07:44Z dougk quit (Ping timeout: 240 seconds)
2017-09-25T23:21:00Z dougk joined #sbcl
2017-09-25T23:31:33Z dougk quit (Read error: Connection reset by peer)
2017-09-25T23:32:18Z dougk joined #sbcl
2017-09-25T23:46:15Z cromachina joined #sbcl
2017-09-26T00:16:15Z Jesin quit (Ping timeout: 255 seconds)
2017-09-26T00:22:03Z rgrau quit (Ping timeout: 260 seconds)
2017-09-26T00:22:05Z Jesin joined #sbcl
2017-09-26T01:01:26Z Bicyclidine joined #sbcl
2017-09-26T01:03:30Z Bike quit (Ping timeout: 255 seconds)
2017-09-26T01:03:32Z stassats quit (Ping timeout: 248 seconds)
2017-09-26T01:42:07Z cromachina: can someone explain the purpose of this change in 1.3.18? "minor incompatible change: the result of WRITE-TO-STRING may be a BASE-STRING if all characters written are of type BASE-CHAR."
2017-09-26T01:44:26Z stylewarning: More efficient
2017-09-26T01:44:35Z cromachina: wouldnt this break anything that wants to (declare (type (simple-array character) ...))?
2017-09-26T01:44:50Z stylewarning: Yes
2017-09-26T01:48:01Z cromachina: do you have an easy way to relax this constraint or force WRITE-TO-STRING to always provide the old string type?
2017-09-26T01:51:07Z stylewarning: cromachina: WOTS can take a string to write to
2017-09-26T01:51:35Z stylewarning: Make a string of your preference and make it adjustable
2017-09-26T01:51:50Z dougk: and CLHS says that write-to-string returns a 'string'. 'string' is a union type, so the decl that is more constraining is likely wrong
2017-09-26T01:52:11Z stylewarning: Or, provide an element type to WOTS
2017-09-26T01:52:19Z cromachina: looks like some libraries need patching then
2017-09-26T01:52:40Z stylewarning: cromachina: WOTS takes a &key element-type
2017-09-26T01:52:58Z Bicyclidine: which libraries?
2017-09-26T01:53:47Z cromachina: cxml and closure-html
2017-09-26T01:54:09Z cromachina: library xpath explodes when trying to (xpath:compile-xpath ...)
2017-09-26T01:54:27Z Bike joined #sbcl
2017-09-26T01:54:27Z cromachina: maybe i should just switch to something less ancient
2017-09-26T01:56:42Z dougk: I told the cxml maintainer in January - https://sourceforge.net/p/sbcl/mailman/message/35636858/
2017-09-26T01:57:11Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-26T01:57:52Z _death: some test in plexippus-xpath tends to throw sbcl (w/ default settings) to ldb, so I comment it.. not clear to me why test code is even loaded if you just want to use it
2017-09-26T01:58:33Z cromachina: i have a feeling it's never going to get fixed in that branch
2017-09-26T02:08:56Z cromachina: took the nuclear route: modified the quicklisp source i had for closure-common characters.lisp and changed all the deftypes to t
2017-09-26T02:13:23Z cromachina: didnt work as well as i thought.. lol
2017-09-26T02:42:29Z dougk quit (Read error: Connection reset by peer)
2017-09-26T02:42:48Z dougk joined #sbcl
2017-09-26T02:52:55Z shka joined #sbcl
2017-09-26T02:57:31Z karswell_ quit (Remote host closed the connection)
2017-09-26T02:59:04Z stylewarning: Does SBCL provide a simple way to construct a specialized simple-array from a pointer to foreign data
2017-09-26T03:07:16Z stylewarning: By the looks of stelian's static-vectors, the answer seems to be "no"
2017-09-26T03:15:50Z cromachina: dougk, _death, i've just fixed my issue by switching to plump and lquery 🤷
2017-09-26T03:25:24Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-26T03:38:23Z jack_rabbit joined #sbcl
2017-09-26T03:59:01Z Bicyclidine joined #sbcl
2017-09-26T04:00:44Z Bike quit (Ping timeout: 246 seconds)
2017-09-26T04:04:54Z knobo joined #sbcl
2017-09-26T04:05:14Z milanj_ quit (Quit: This computer has gone to sleep)
2017-09-26T04:10:12Z knobo quit (Ping timeout: 248 seconds)
2017-09-26T04:38:06Z oleo quit (Quit: Leaving)
2017-09-26T05:04:34Z Bike joined #sbcl
2017-09-26T05:08:22Z Bicyclidine quit (Ping timeout: 264 seconds)
2017-09-26T06:08:16Z angavrilov joined #sbcl
2017-09-26T06:11:05Z Bike quit (Ping timeout: 240 seconds)
2017-09-26T06:12:48Z scymtym quit (Ping timeout: 252 seconds)
2017-09-26T06:22:56Z slyrus joined #sbcl
2017-09-26T06:42:18Z knobo joined #sbcl
2017-09-26T07:07:48Z dougk quit (Ping timeout: 248 seconds)
2017-09-26T07:08:48Z dougk joined #sbcl
2017-09-26T07:15:01Z shka quit (Ping timeout: 240 seconds)
2017-09-26T07:33:24Z scymtym joined #sbcl
2017-09-26T07:45:09Z dougk quit (Ping timeout: 248 seconds)
2017-09-26T07:48:02Z carenz joined #sbcl
2017-09-26T07:48:04Z milanj_ joined #sbcl
2017-09-26T08:01:07Z cromachina quit (Read error: Connection reset by peer)
2017-09-26T08:41:43Z dougk joined #sbcl
2017-09-26T08:46:10Z dougk quit (Ping timeout: 264 seconds)
2017-09-26T08:53:56Z adlai quit (Ping timeout: 248 seconds)
2017-09-26T09:07:40Z adlai joined #sbcl
2017-09-26T09:42:43Z dougk joined #sbcl
2017-09-26T09:46:43Z dougk quit (Ping timeout: 240 seconds)
2017-09-26T09:53:18Z stassats joined #sbcl
2017-09-26T10:07:28Z m00natic joined #sbcl
2017-09-26T10:22:12Z easye joined #sbcl
2017-09-26T10:43:48Z dougk joined #sbcl
2017-09-26T10:48:13Z dougk quit (Ping timeout: 240 seconds)
2017-09-26T11:02:48Z akovalenko quit (Ping timeout: 240 seconds)
2017-09-26T11:16:54Z Bike joined #sbcl
2017-09-26T11:21:52Z rgrau joined #sbcl
2017-09-26T13:00:57Z sjl quit (Ping timeout: 260 seconds)
2017-09-26T13:25:42Z cromachina joined #sbcl
2017-09-26T13:34:02Z scymtym quit (Ping timeout: 246 seconds)
2017-09-26T13:42:59Z oleo joined #sbcl
2017-09-26T13:55:50Z Bicyclidine joined #sbcl
2017-09-26T13:58:27Z Bike quit (Ping timeout: 246 seconds)
2017-09-26T14:01:52Z Jesin quit (Quit: Leaving)
2017-09-26T14:17:21Z cromachina quit (Read error: Connection reset by peer)
2017-09-26T14:21:19Z Jesin joined #sbcl
2017-09-26T14:27:38Z stassats quit (Ping timeout: 255 seconds)
2017-09-26T14:35:31Z scymtym joined #sbcl
2017-09-26T14:53:20Z alandipert joined #sbcl
2017-09-26T15:23:05Z knobo quit (Ping timeout: 240 seconds)
2017-09-26T15:35:33Z carenz quit (Ping timeout: 248 seconds)
2017-09-26T15:42:07Z dougk joined #sbcl
2017-09-26T15:49:24Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T16:10:54Z alandipert: hello, sbcl.org looks down
2017-09-26T16:15:20Z dougk joined #sbcl
2017-09-26T16:23:17Z rgrau quit (Ping timeout: 252 seconds)
2017-09-26T16:50:27Z shka joined #sbcl
2017-09-26T17:03:06Z m00natic quit (Read error: Connection reset by peer)
2017-09-26T17:13:32Z copec: (setf *crickets* t)
2017-09-26T17:19:06Z pkhuong_: sourceforge is down
2017-09-26T17:40:47Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T17:47:34Z dougk joined #sbcl
2017-09-26T17:54:47Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T18:07:49Z dougk joined #sbcl
2017-09-26T18:24:07Z dougk quit (Ping timeout: 240 seconds)
2017-09-26T18:36:20Z dougk joined #sbcl
2017-09-26T18:50:26Z milanj joined #sbcl
2017-09-26T18:50:43Z milanj_ quit (Ping timeout: 248 seconds)
2017-09-26T19:01:12Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T19:16:17Z dougk joined #sbcl
2017-09-26T19:21:32Z Bike joined #sbcl
2017-09-26T19:21:36Z Bike quit (Remote host closed the connection)
2017-09-26T19:25:45Z Bicyclidine quit (Ping timeout: 264 seconds)
2017-09-26T19:26:21Z Bike joined #sbcl
2017-09-26T19:33:45Z scymtym quit (Ping timeout: 246 seconds)
2017-09-26T19:35:35Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T19:37:53Z dougk joined #sbcl
2017-09-26T19:46:21Z dougk quit (Ping timeout: 246 seconds)
2017-09-26T19:57:53Z dougk joined #sbcl
2017-09-26T20:12:10Z dougk__ joined #sbcl
2017-09-26T20:12:20Z dougk quit (Ping timeout: 255 seconds)
2017-09-26T20:29:57Z dougk__ quit (Ping timeout: 264 seconds)
2017-09-26T20:30:35Z scymtym joined #sbcl
2017-09-26T20:53:14Z dougk joined #sbcl
2017-09-26T21:03:38Z dougk quit (Ping timeout: 255 seconds)
2017-09-26T21:07:10Z dougk joined #sbcl
2017-09-26T21:20:22Z sjl joined #sbcl
2017-09-26T21:31:19Z angavrilov quit (Remote host closed the connection)
2017-09-26T22:18:20Z dougk quit (Ping timeout: 255 seconds)
2017-09-26T23:26:20Z shka quit (Ping timeout: 240 seconds)
2017-09-26T23:37:29Z stassats joined #sbcl
2017-09-26T23:44:01Z cromachina joined #sbcl
2017-09-26T23:46:22Z myrkraverk joined #sbcl
2017-09-26T23:56:12Z karswell joined #sbcl
2017-09-27T00:22:10Z jdz quit (Ping timeout: 255 seconds)
2017-09-27T00:27:22Z jdz joined #sbcl
2017-09-27T00:40:00Z stassats quit (Ping timeout: 240 seconds)
2017-09-27T02:57:06Z milanj quit (Quit: This computer has gone to sleep)
2017-09-27T03:27:55Z Bike quit (Ping timeout: 255 seconds)
2017-09-27T03:53:02Z shka joined #sbcl
2017-09-27T04:22:14Z pkhuong joined #sbcl
2017-09-27T04:22:37Z pkhuong is now known as Guest10392
2017-09-27T04:30:19Z pkhuong_ quit (*.net *.split)
2017-09-27T04:32:31Z malm quit (Ping timeout: 260 seconds)
2017-09-27T04:33:47Z malm joined #sbcl
2017-09-27T04:57:22Z slyrus: maybe now's a good time to move away from sourceforge? it had a nice run that ended a long time ago.
2017-09-27T05:00:19Z oleo quit (Quit: Leaving)
2017-09-27T06:05:00Z scymtym quit (Ping timeout: 240 seconds)
2017-09-27T06:18:35Z knobo joined #sbcl
2017-09-27T07:02:11Z angavrilov joined #sbcl
2017-09-27T07:03:41Z shka quit (Ping timeout: 240 seconds)
2017-09-27T07:30:26Z scymtym joined #sbcl
2017-09-27T07:46:56Z milanj joined #sbcl
2017-09-27T07:50:48Z dim quit (Ping timeout: 255 seconds)
2017-09-27T07:51:55Z dim joined #sbcl
2017-09-27T07:52:15Z carenz joined #sbcl
2017-09-27T08:59:34Z panji joined #sbcl
2017-09-27T09:22:20Z panji quit (Quit: Leaving)
2017-09-27T10:13:27Z knobo quit (Ping timeout: 255 seconds)
2017-09-27T10:21:50Z m00natic joined #sbcl
2017-09-27T10:22:38Z scymtym: i made an #ifdef consistency checker using my dpll implementation and it found two issues in our runtime code. given the complexity, i'm surprised there are so few. are we already using some form of machine assistance for this?
2017-09-27T10:36:43Z carenz_ joined #sbcl
2017-09-27T10:39:52Z knobo joined #sbcl
2017-09-27T10:40:04Z carenz quit (Ping timeout: 240 seconds)
2017-09-27T11:05:19Z Bike joined #sbcl
2017-09-27T11:55:16Z carenz_ quit (Ping timeout: 248 seconds)
2017-09-27T12:24:32Z Bike quit (Ping timeout: 246 seconds)
2017-09-27T12:29:42Z dougk joined #sbcl
2017-09-27T12:34:40Z scymtym quit (Ping timeout: 255 seconds)
2017-09-27T12:37:35Z scymtym joined #sbcl
2017-09-27T12:47:43Z scymtym quit (Ping timeout: 246 seconds)
2017-09-27T12:49:36Z scymtym joined #sbcl
2017-09-27T12:51:24Z knobo quit (Ping timeout: 255 seconds)
2017-09-27T12:54:37Z Bike joined #sbcl
2017-09-27T12:59:37Z Bike quit (Remote host closed the connection)
2017-09-27T13:14:44Z knobo joined #sbcl
2017-09-27T13:25:40Z knobo quit (Ping timeout: 240 seconds)
2017-09-27T13:31:34Z sjl quit (Ping timeout: 252 seconds)
2017-09-27T13:31:40Z Bike joined #sbcl
2017-09-27T13:50:39Z knobo joined #sbcl
2017-09-27T14:04:26Z knobo quit (Ping timeout: 240 seconds)
2017-09-27T14:30:13Z stassats joined #sbcl
2017-09-27T14:30:17Z cromachina quit (Read error: Connection reset by peer)
2017-09-27T14:33:35Z dougk quit (Read error: Connection reset by peer)
2017-09-27T14:33:47Z edgar-rft quit (Quit: edgar-rft)
2017-09-27T14:33:54Z dougk joined #sbcl
2017-09-27T14:36:10Z oleo joined #sbcl
2017-09-27T14:37:11Z scymtym quit (Ping timeout: 246 seconds)
2017-09-27T15:06:03Z Jesin quit (Quit: Leaving)
2017-09-27T15:19:11Z Jesin joined #sbcl
2017-09-27T15:36:16Z Bike quit (Remote host closed the connection)
2017-09-27T15:36:32Z Bike joined #sbcl
2017-09-27T15:38:37Z Jesin quit (Quit: Leaving)
2017-09-27T15:51:09Z Jesin joined #sbcl
2017-09-27T15:52:32Z trinque joined #sbcl
2017-09-27T16:36:00Z scymtym joined #sbcl
2017-09-27T16:37:52Z ym joined #sbcl
2017-09-27T17:18:55Z shka_ joined #sbcl
2017-09-27T17:37:39Z m00natic quit (Remote host closed the connection)
2017-09-27T18:06:36Z edgar-rft joined #sbcl
2017-09-27T18:11:55Z flip214: I believe that clang and/or GCC have warnings for mismatched #if/#else/#endif... or what kind of inconsistency do you mean?
2017-09-27T18:20:16Z dougk quit (Ping timeout: 255 seconds)
2017-09-27T18:47:07Z dougk joined #sbcl
2017-09-27T18:58:44Z stassats quit (Ping timeout: 264 seconds)
2017-09-27T19:01:04Z dougk quit (Read error: Connection reset by peer)
2017-09-27T19:01:36Z dougk joined #sbcl
2017-09-27T19:33:39Z eschatologist quit (Ping timeout: 240 seconds)
2017-09-27T19:39:09Z eschatologist joined #sbcl
2017-09-27T19:52:09Z shka_ quit (Ping timeout: 240 seconds)
2017-09-27T20:04:47Z scymtym: flip214: http://paste.lisp.org/display/357032
2017-09-27T20:09:43Z flip214: scymtym: thanks, interesting.... yeah, such semantic things are not found by them (yet).
2017-09-27T20:16:07Z scymtym: the hardest part is the cpp parser
2017-09-27T20:16:23Z luis: Does not anyone have a link to recent SBCL binary tarballs for x86 and x86-64?
2017-09-27T20:16:30Z luis: *Does anyone
2017-09-27T20:49:48Z angavrilov quit (Remote host closed the connection)
2017-09-27T20:53:03Z Bike quit (Ping timeout: 240 seconds)
2017-09-27T20:58:27Z vydd joined #sbcl
2017-09-27T21:39:00Z Bike joined #sbcl
2017-09-27T21:45:11Z Bicyclidine joined #sbcl
2017-09-27T21:48:06Z Bike quit (Ping timeout: 246 seconds)
2017-09-27T21:52:36Z dougk quit (Read error: Connection reset by peer)
2017-09-27T21:53:10Z dougk joined #sbcl
2017-09-27T21:53:24Z rgrau joined #sbcl
2017-09-27T22:04:52Z HoloIRCUser joined #sbcl
2017-09-27T22:06:14Z HoloIRCUser is now known as Autcrock
2017-09-27T22:07:22Z Bicyclidine quit (Remote host closed the connection)
2017-09-27T22:07:54Z Bike joined #sbcl
2017-09-27T22:23:25Z dougk: flip214: scymtym: what's the context of that #ifdef checker?  do we have a current problem or was that an "fyi" ?
2017-09-27T22:24:02Z dougk: fe[nl]ix wants to take our insane pile of poo otherwise known as pseudo-atomic.h and require that each arch-os have an explicit file to include
2017-09-27T22:31:00Z scymtym: dougk: i was just toying around. the instances it finds are basically like #if defined (LISP_FEATURE_X86_64) ... #if defined (LISP_FEATURE_X86_64) #endif #endif. so harmless, but redundant. i have to verify the results some more before acting on anything
2017-09-27T22:34:13Z Bike quit (Ping timeout: 246 seconds)
2017-09-27T22:34:58Z Bike joined #sbcl
2017-09-27T22:39:12Z scymtym: with the current implementation, i only get a single instance in pseudo-atomic.h: the #elif defined(LISP_FEATURE_X86_64) in line 32 is redundant since the "else" branch of #if defined(LISP_FEATURE_X86) already implies this in the context of #if defined(LISP_FEATURE_X86) || defined(LISP_FEATURE_X86_64)
2017-09-27T22:41:06Z milanj quit (Ping timeout: 240 seconds)
2017-09-27T22:41:41Z milanj joined #sbcl
2017-09-27T22:48:44Z Jesin quit (Quit: Leaving)
2017-09-27T23:05:29Z sjl joined #sbcl
2017-09-27T23:18:31Z dougk__ joined #sbcl
2017-09-27T23:21:10Z dougk quit (Ping timeout: 252 seconds)
2017-09-27T23:24:10Z Jesin joined #sbcl
2017-09-27T23:36:31Z cromachina joined #sbcl
2017-09-27T23:47:47Z vydd quit (Ping timeout: 248 seconds)
2017-09-28T00:11:33Z dougk__ quit (Ping timeout: 240 seconds)
2017-09-28T00:14:27Z scymtym quit (Ping timeout: 248 seconds)
2017-09-28T00:19:08Z dougk joined #sbcl
2017-09-28T00:26:16Z Jesin quit (Quit: Leaving)
2017-09-28T00:41:39Z Bicyclidine joined #sbcl
2017-09-28T00:44:09Z Bike quit (Ping timeout: 246 seconds)
2017-09-28T00:49:56Z Jesin joined #sbcl
2017-09-28T01:14:19Z Bike joined #sbcl
2017-09-28T01:17:46Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-09-28T01:51:09Z Bicyclidine joined #sbcl
2017-09-28T01:51:11Z Bicyclidine quit (Remote host closed the connection)
2017-09-28T01:51:32Z Bicyclidine joined #sbcl
2017-09-28T01:53:37Z Bike quit (Ping timeout: 248 seconds)
2017-09-28T02:02:48Z karswell quit (Read error: Connection reset by peer)
2017-09-28T02:03:43Z karswell joined #sbcl
2017-09-28T02:05:50Z karswell quit (Remote host closed the connection)
2017-09-28T02:06:41Z karswell joined #sbcl
2017-09-28T02:08:56Z karswell quit (Remote host closed the connection)
2017-09-28T02:09:44Z karswell joined #sbcl
2017-09-28T02:10:14Z karswell quit (Read error: Connection reset by peer)
2017-09-28T02:12:21Z DeadTrickster quit (Ping timeout: 246 seconds)
2017-09-28T02:16:57Z Autcrock quit (Remote host closed the connection)
2017-09-28T02:37:19Z dougk quit (Ping timeout: 240 seconds)
2017-09-28T02:37:39Z dougk joined #sbcl
2017-09-28T02:40:02Z milanj quit (Quit: This computer has gone to sleep)
2017-09-28T02:41:39Z rgrau quit (Ping timeout: 248 seconds)
2017-09-28T02:42:48Z DeadTrickster joined #sbcl
2017-09-28T03:14:34Z Intensity quit (Ping timeout: 246 seconds)
2017-09-28T03:22:57Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-28T03:28:47Z Jesin quit (Read error: Connection reset by peer)
2017-09-28T03:32:09Z abbe quit (Ping timeout: 246 seconds)
2017-09-28T03:37:19Z abbe joined #sbcl
2017-09-28T03:41:12Z Intensity joined #sbcl
2017-09-28T03:52:01Z Colleen quit (Ping timeout: 248 seconds)
2017-09-28T03:53:03Z Shinmera quit (Ping timeout: 240 seconds)
2017-09-28T03:56:43Z shka_ joined #sbcl
2017-09-28T03:59:45Z Shinmera joined #sbcl
2017-09-28T04:00:10Z milanj joined #sbcl
2017-09-28T04:00:59Z jsnell quit (Ping timeout: 240 seconds)
2017-09-28T04:08:59Z jsnell joined #sbcl
2017-09-28T04:12:39Z dougk quit (Ping timeout: 240 seconds)
2017-09-28T04:12:39Z minion quit (Ping timeout: 252 seconds)
2017-09-28T04:13:07Z specbot quit (Ping timeout: 264 seconds)
2017-09-28T04:13:26Z fe[nl]ix quit (Ping timeout: 240 seconds)
2017-09-28T04:13:45Z easye quit (Ping timeout: 252 seconds)
2017-09-28T04:13:46Z Blkt quit (Ping timeout: 240 seconds)
2017-09-28T04:14:36Z milanj quit (Quit: This computer has gone to sleep)
2017-09-28T04:18:36Z Blkt joined #sbcl
2017-09-28T04:18:39Z minion joined #sbcl
2017-09-28T04:18:39Z specbot joined #sbcl
2017-09-28T04:18:45Z fe[nl]ix joined #sbcl
2017-09-28T04:40:10Z slyrus quit (Remote host closed the connection)
2017-09-28T04:50:40Z knobo joined #sbcl
2017-09-28T04:58:51Z knobo quit (Ping timeout: 252 seconds)
2017-09-28T05:09:00Z dougk joined #sbcl
2017-09-28T05:13:33Z stylewarning: is there a good way to profile compilation? Something is slowing down compilation of a large file and I can't figure out what it is
2017-09-28T05:23:15Z oleo quit (Quit: Leaving)
2017-09-28T05:32:18Z angavrilov joined #sbcl
2017-09-28T05:56:21Z Colleen joined #sbcl
2017-09-28T05:56:23Z Colleen quit (Remote host closed the connection)
2017-09-28T05:56:36Z Colleen joined #sbcl
2017-09-28T06:21:31Z knobo joined #sbcl
2017-09-28T06:29:45Z dougk quit (Ping timeout: 240 seconds)
2017-09-28T06:57:33Z carenz_ joined #sbcl
2017-09-28T07:25:15Z vydd joined #sbcl
2017-09-28T07:25:16Z vydd quit (Changing host)
2017-09-28T07:25:16Z vydd joined #sbcl
2017-09-28T07:27:13Z scymtym joined #sbcl
2017-09-28T07:33:53Z vydd quit (Quit: Leaving)
2017-09-28T07:33:54Z shka_ quit (Ping timeout: 248 seconds)
2017-09-28T08:07:25Z adlai quit (Ping timeout: 260 seconds)
2017-09-28T08:16:45Z adlai joined #sbcl
2017-09-28T08:28:15Z a_egorov joined #sbcl
2017-09-28T08:35:20Z Cain joined #sbcl
2017-09-28T09:11:38Z easye joined #sbcl
2017-09-28T09:24:51Z Bike joined #sbcl
2017-09-28T10:02:41Z m00natic joined #sbcl
2017-09-28T10:09:08Z adlai quit (Ping timeout: 240 seconds)
2017-09-28T10:20:12Z adlai joined #sbcl
2017-09-28T10:50:06Z dougk joined #sbcl
2017-09-28T11:01:34Z BitPuffin|osx joined #sbcl
2017-09-28T11:01:53Z stassats joined #sbcl
2017-09-28T11:02:28Z stassats: the latest pfdietz thing is strange
2017-09-28T11:03:10Z stassats: somehow an IF test gets component-tail as a successor
2017-09-28T11:05:15Z stassats: and somehow branches with different numbers of values are involved
2017-09-28T11:07:44Z stassats: usually, only the nodes with returns go to component-tail
2017-09-28T11:14:39Z stassats quit (Ping timeout: 260 seconds)
2017-09-28T11:19:06Z dougk quit (Read error: Connection reset by peer)
2017-09-28T11:19:23Z dougk joined #sbcl
2017-09-28T11:41:20Z stassats joined #sbcl
2017-09-28T11:44:26Z panji joined #sbcl
2017-09-28T11:50:55Z stassats: after the last ir1-optimize-if the IR looks in good shape, then something happens
2017-09-28T12:03:51Z nimbus7 joined #sbcl
2017-09-28T12:06:25Z stassats: the return gets deleted, huh
2017-09-28T12:09:02Z stassats: for some reason a [#<SB-KERNEL:NAMED-TYPE *> -> #<SB-KERNEL:NAMED-TYPE *>] cast gets directed to component-tail
2017-09-28T12:14:22Z stassats: another may-delete-vestigial-exit issue?
2017-09-28T12:14:26Z stassats: paging nyef
2017-09-28T12:25:17Z stassats: looks like may-delete-vestigial-exit stop the * -> * cast from being deleted, then the m-v receiver has type (or (eql 0) null) and the lvar-value has (values 0 0), it conflicts and gets terminated
2017-09-28T12:37:29Z stassats: (values-type-intersection (values-specifier-type '(values (or (integer 0 0) null) &optional (integer 0 0))) (values-specifier-type '(values (or bit null) &optional (or (integer 2 2) (integer 0 0)) (integer 3 3)))) => #<VALUES-TYPE (VALUES (OR (INTEGER 0 0) NULL) &OPTIONAL NIL)>
2017-09-28T12:37:40Z stassats: that's where the different number of values comes in
2017-09-28T12:39:39Z stassats: so it derives a tighter type, but derive-node-type intersects them, :from-scratch t will help there
2017-09-28T12:40:17Z stassats: or type intersection wrt the number of value types is just incorrect
2017-09-28T12:40:36Z stassats: it's for sure incorrect here, a tighter type suddenly conflicts
2017-09-28T12:45:01Z stassats: it's wrong that (intersection (union x y) x) doesn't yield X
2017-09-28T12:45:23Z edgar-rft quit (Quit: edgar-rft)
2017-09-28T12:49:20Z panji quit (Read error: Connection reset by peer)
2017-09-28T12:56:48Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-28T12:59:00Z stassats: a missing 0 start in (subseq optional i)? huh
2017-09-28T12:59:44Z stassats` joined #sbcl
2017-09-28T13:00:22Z stassats: what's 14 years for a one letter bug?
2017-09-28T13:03:11Z jdz: Cute.
2017-09-28T13:05:50Z panji joined #sbcl
2017-09-28T13:07:30Z sjl quit (Ping timeout: 255 seconds)
2017-09-28T13:14:46Z stassats: most of the time i spent finding the moment IR goes wrong
2017-09-28T13:14:54Z stassats: i fish i could place breakpoints on IR modification
2017-09-28T13:15:07Z stassats: or at least some kind of history
2017-09-28T13:26:59Z milanj joined #sbcl
2017-09-28T13:27:21Z stassats` quit (Ping timeout: 258 seconds)
2017-09-28T13:27:30Z stassats quit (Ping timeout: 248 seconds)
2017-09-28T13:35:47Z BitPuffin|osx joined #sbcl
2017-09-28T13:49:28Z Cain_ joined #sbcl
2017-09-28T13:49:43Z dougk quit (Ping timeout: 264 seconds)
2017-09-28T13:52:34Z Cain quit (Ping timeout: 255 seconds)
2017-09-28T13:53:42Z oleo joined #sbcl
2017-09-28T14:04:01Z slyrus joined #sbcl
2017-09-28T14:06:53Z Jesin joined #sbcl
2017-09-28T14:10:30Z carenz_ quit (Ping timeout: 252 seconds)
2017-09-28T14:11:26Z Bike quit (Ping timeout: 258 seconds)
2017-09-28T14:15:13Z rgrau joined #sbcl
2017-09-28T14:21:11Z cromachina quit (Read error: Connection reset by peer)
2017-09-28T14:28:00Z Bike joined #sbcl
2017-09-28T15:12:34Z knobo quit (Ping timeout: 248 seconds)
2017-09-28T15:14:41Z a_egorov quit (Ping timeout: 248 seconds)
2017-09-28T15:23:33Z dougk joined #sbcl
2017-09-28T15:28:39Z dougk__ joined #sbcl
2017-09-28T15:30:57Z dougk quit (Ping timeout: 246 seconds)
2017-09-28T15:37:06Z rgrau quit (Ping timeout: 248 seconds)
2017-09-28T15:53:07Z Cain joined #sbcl
2017-09-28T15:55:58Z Cain_ quit (Ping timeout: 240 seconds)
2017-09-28T15:59:18Z dougk__ quit (Ping timeout: 246 seconds)
2017-09-28T16:03:38Z dougk joined #sbcl
2017-09-28T16:14:35Z panji quit (Quit: Leaving)
2017-09-28T16:17:32Z tonton quit (Ping timeout: 240 seconds)
2017-09-28T16:19:29Z tonton joined #sbcl
2017-09-28T16:23:47Z attila_lendvai joined #sbcl
2017-09-28T16:23:56Z attila_lendvai quit (Remote host closed the connection)
2017-09-28T16:25:38Z dougk quit (Ping timeout: 246 seconds)
2017-09-28T16:34:02Z Cain quit (Ping timeout: 258 seconds)
2017-09-28T16:35:15Z edgar-rft joined #sbcl
2017-09-28T16:54:20Z dougk joined #sbcl
2017-09-28T17:08:39Z m00natic quit (Remote host closed the connection)
2017-09-28T17:10:05Z shka_ joined #sbcl
2017-09-28T17:21:01Z knobo joined #sbcl
2017-09-28T17:31:09Z knobo quit (Ping timeout: 258 seconds)
2017-09-28T17:43:53Z carenz_ joined #sbcl
2017-09-28T18:12:56Z dougk quit (Ping timeout: 258 seconds)
2017-09-28T18:14:30Z attila_lendvai joined #sbcl
2017-09-28T18:25:02Z myrkraverk quit (Ping timeout: 240 seconds)
2017-09-28T18:45:50Z whoman joined #sbcl
2017-09-28T18:47:39Z myrkraverk joined #sbcl
2017-09-28T18:50:00Z dougk joined #sbcl
2017-09-28T19:20:39Z Jesin quit (Quit: Leaving)
2017-09-28T19:24:50Z carenz_ quit (Ping timeout: 248 seconds)
2017-09-28T19:25:27Z scymtym quit (Ping timeout: 246 seconds)
2017-09-28T19:39:45Z Jesin joined #sbcl
2017-09-28T20:01:02Z rpg joined #sbcl
2017-09-28T20:01:25Z dougk quit (Ping timeout: 258 seconds)
2017-09-28T20:04:02Z sjl joined #sbcl
2017-09-28T20:07:46Z knobo joined #sbcl
2017-09-28T20:12:32Z knobo quit (Ping timeout: 258 seconds)
2017-09-28T20:16:30Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-28T20:26:45Z rpg joined #sbcl
2017-09-28T20:28:09Z scymtym joined #sbcl
2017-09-28T20:45:09Z nimbus7 quit (Quit: Leaving)
2017-09-28T21:11:05Z rpg quit (Ping timeout: 240 seconds)
2017-09-28T21:14:51Z prxq joined #sbcl
2017-09-28T21:19:37Z shka_ quit (Ping timeout: 240 seconds)
2017-09-28T21:23:49Z angavrilov quit (Remote host closed the connection)
2017-09-28T21:24:31Z whoman quit (Read error: Connection reset by peer)
2017-09-28T21:28:50Z whoman joined #sbcl
2017-09-28T21:33:16Z Bicyclidine joined #sbcl
2017-09-28T21:34:34Z Bike quit (Ping timeout: 258 seconds)
2017-09-28T21:44:37Z rpg joined #sbcl
2017-09-28T21:51:57Z prxq quit (Remote host closed the connection)
2017-09-28T21:59:05Z jdz quit (Ping timeout: 240 seconds)
2017-09-28T21:59:34Z jdz joined #sbcl
2017-09-28T22:22:39Z dougk joined #sbcl
2017-09-28T22:36:07Z dougk quit (Ping timeout: 240 seconds)
2017-09-28T23:09:40Z dougk joined #sbcl
2017-09-28T23:13:30Z Bike joined #sbcl
2017-09-28T23:14:26Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-28T23:49:22Z cromachina joined #sbcl
2017-09-28T23:51:02Z trinque quit (Ping timeout: 240 seconds)
2017-09-28T23:53:43Z trinque joined #sbcl
2017-09-29T00:01:18Z stassats joined #sbcl
2017-09-29T00:02:33Z attila_lendvai quit (Quit: Leaving.)
2017-09-29T00:02:53Z attila_lendvai joined #sbcl
2017-09-29T00:03:00Z attila_lendvai quit (Client Quit)
2017-09-29T00:16:52Z eschatologist quit (Quit: ZNC 1.6.4+deb1 - http://znc.in)
2017-09-29T00:17:44Z eschatologist joined #sbcl
2017-09-29T00:28:37Z whoman quit (Ping timeout: 240 seconds)
2017-09-29T00:32:06Z Bicyclidine joined #sbcl
2017-09-29T00:35:07Z Bike quit (Ping timeout: 258 seconds)
2017-09-29T00:40:29Z stassats quit (Ping timeout: 258 seconds)
2017-09-29T01:05:25Z Bike joined #sbcl
2017-09-29T01:06:10Z dougk quit (Ping timeout: 258 seconds)
2017-09-29T01:08:51Z Bicyclidine quit (Ping timeout: 258 seconds)
2017-09-29T01:38:27Z Bicyclidine joined #sbcl
2017-09-29T01:39:31Z Bike quit (Ping timeout: 258 seconds)
2017-09-29T02:03:05Z dougk joined #sbcl
2017-09-29T02:03:46Z Gorgias joined #sbcl
2017-09-29T02:07:01Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T02:07:47Z Bicyclidine quit (Ping timeout: 255 seconds)
2017-09-29T02:24:29Z Bike joined #sbcl
2017-09-29T02:25:37Z Gorgias quit (Ping timeout: 248 seconds)
2017-09-29T02:52:49Z shka_ joined #sbcl
2017-09-29T03:04:05Z dougk joined #sbcl
2017-09-29T03:08:31Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T03:11:33Z milanj quit (Quit: This computer has gone to sleep)
2017-09-29T03:12:39Z adlai quit (Ping timeout: 258 seconds)
2017-09-29T03:28:50Z adlai joined #sbcl
2017-09-29T03:50:30Z Bike quit (Ping timeout: 246 seconds)
2017-09-29T04:05:19Z dougk joined #sbcl
2017-09-29T04:09:46Z dougk quit (Ping timeout: 258 seconds)
2017-09-29T04:31:53Z slyrus quit (Quit: Client Quit)
2017-09-29T04:36:41Z slyrus joined #sbcl
2017-09-29T05:08:55Z rpg quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2017-09-29T05:40:12Z oleo quit (Quit: Leaving)
2017-09-29T06:13:55Z angavrilov joined #sbcl
2017-09-29T06:19:13Z scymtym quit (Ping timeout: 248 seconds)
2017-09-29T06:38:11Z carenz_ joined #sbcl
2017-09-29T06:48:43Z ym_ joined #sbcl
2017-09-29T07:01:16Z knobo joined #sbcl
2017-09-29T07:18:17Z shka_ quit (Ping timeout: 240 seconds)
2017-09-29T07:24:14Z scymtym joined #sbcl
2017-09-29T07:36:59Z dougk joined #sbcl
2017-09-29T07:41:22Z dougk quit (Ping timeout: 258 seconds)
2017-09-29T07:43:04Z rgrau joined #sbcl
2017-09-29T07:51:55Z ym_ quit (Quit: Leaving)
2017-09-29T08:01:28Z attila_lendvai joined #sbcl
2017-09-29T08:38:24Z dougk joined #sbcl
2017-09-29T08:42:42Z dougk quit (Ping timeout: 258 seconds)
2017-09-29T08:48:59Z dougk joined #sbcl
2017-09-29T09:09:38Z dougk quit (Read error: Connection reset by peer)
2017-09-29T09:09:56Z dougk joined #sbcl
2017-09-29T09:20:06Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T09:28:04Z panji joined #sbcl
2017-09-29T09:42:24Z kini quit (Ping timeout: 248 seconds)
2017-09-29T09:45:05Z kini joined #sbcl
2017-09-29T09:59:23Z dougk joined #sbcl
2017-09-29T10:03:35Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T10:10:15Z milanj joined #sbcl
2017-09-29T10:28:59Z stassats joined #sbcl
2017-09-29T10:29:51Z stassats: (progn (defparameter *abc* (make-string 1 :element-type 'base-char)) (setf (schar *abc* 0) #\cyrillic_small_letter_yu) *abc*) => "N"
2017-09-29T10:30:01Z stassats: the interpreter stub doesn't seem to be checking the type
2017-09-29T10:31:29Z stassats: weird, cause when i recompile %scharset it starts to behave
2017-09-29T10:44:26Z stassats quit (Quit: ERC (IRC client for Emacs 25.1.2))
2017-09-29T10:46:43Z stassats joined #sbcl
2017-09-29T10:54:53Z stassats: (speed 2) breaks it
2017-09-29T10:55:11Z stassats left #sbcl
2017-09-29T10:55:43Z stassats joined #sbcl
2017-09-29T10:59:42Z dougk joined #sbcl
2017-09-29T11:04:05Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T11:19:39Z stassats: MAYBE-WEAKEN-CHECK weakens base-char to character?
2017-09-29T11:20:35Z stassats: (weaken-type (specifier-type 'base-char)) => #<CHARACTER-SET-TYPE CHARACTER>
2017-09-29T11:23:11Z stassats: checking for CHARACTER is absolutely not faster
2017-09-29T11:39:28Z stassats: special-casing base-char seems like a solution
2017-09-29T11:51:46Z dougk joined #sbcl
2017-09-29T11:59:13Z rpg joined #sbcl
2017-09-29T12:01:05Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T12:06:04Z Bike joined #sbcl
2017-09-29T12:30:53Z whoman joined #sbcl
2017-09-29T12:33:38Z panji quit (Ping timeout: 260 seconds)
2017-09-29T12:35:45Z attila_lendvai quit (Quit: Leaving.)
2017-09-29T12:46:11Z panji joined #sbcl
2017-09-29T12:52:05Z dougk joined #sbcl
2017-09-29T13:02:13Z sjl quit (Ping timeout: 258 seconds)
2017-09-29T13:32:57Z dougk quit (Ping timeout: 240 seconds)
2017-09-29T13:41:44Z cromachina quit (Read error: Connection reset by peer)
2017-09-29T13:57:21Z oleo joined #sbcl
2017-09-29T14:11:27Z scymtym quit (Remote host closed the connection)
2017-09-29T14:19:00Z slyrus quit (Quit: Client Quit)
2017-09-29T14:29:09Z slyrus joined #sbcl
2017-09-29T14:31:27Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-29T14:46:44Z jack_rabbit joined #sbcl
2017-09-29T14:52:15Z jack_rabbit quit (Ping timeout: 248 seconds)
2017-09-29T15:04:21Z BitPuffin|osx quit (Remote host closed the connection)
2017-09-29T15:05:00Z jack_rabbit joined #sbcl
2017-09-29T15:05:39Z dougk joined #sbcl
2017-09-29T15:08:16Z slyrus quit (Remote host closed the connection)
2017-09-29T15:12:31Z stassats quit (Ping timeout: 248 seconds)
2017-09-29T15:34:38Z m00natic joined #sbcl
2017-09-29T16:00:16Z rgrau quit (Ping timeout: 255 seconds)
2017-09-29T16:03:34Z galdor quit (Read error: Connection reset by peer)
2017-09-29T16:03:53Z galdor joined #sbcl
2017-09-29T16:33:05Z carenz_ quit (Ping timeout: 248 seconds)
2017-09-29T16:57:23Z stassats joined #sbcl
2017-09-29T16:59:57Z shka_ joined #sbcl
2017-09-29T17:00:36Z trinque quit (Quit: trinque)
2017-09-29T17:06:41Z sjl joined #sbcl
2017-09-29T17:22:41Z panji quit (Ping timeout: 248 seconds)
2017-09-29T17:26:31Z m00natic quit (Remote host closed the connection)
2017-09-29T17:32:18Z sjl quit (Ping timeout: 260 seconds)
2017-09-29T17:32:49Z dougk quit (Ping timeout: 248 seconds)
2017-09-29T17:35:09Z panji joined #sbcl
2017-09-29T17:39:09Z rick-monster joined #sbcl
2017-09-29T17:40:16Z whoman quit (Ping timeout: 240 seconds)
2017-09-29T17:40:32Z whoman joined #sbcl
2017-09-29T17:49:49Z dougk joined #sbcl
2017-09-29T17:49:57Z rick-monster: hi I'm trying to add a directive to format by compiling & loading lisp files into a 'stock' sbcl.  Got it working for expressions eval-ed at the repl. But compiler still complains it doesn't know about my directive.
2017-09-29T17:51:20Z rick-monster: Is there any way to add a directive without modifying sbcl source & making a custom sbcl binary?
2017-09-29T17:57:57Z Guest10392 is now known as pkhuong
2017-09-29T17:58:15Z pkhuong: rick-monster: is it that important to avoid ~// ?
2017-09-29T18:10:13Z stassats: rick-monster: interpreted and compiled format directives are different
2017-09-29T18:10:18Z stassats: but it's a bad idea modifying either
2017-09-29T18:18:51Z rick-monster: pkhuong: thanks for the tip, didn't know about ~//
2017-09-29T18:22:44Z rick-monster: stassats: broadly agree it's a bad idea - it's someone else's large, old & valuable codebase targetting allegro. already makes extensive use of a custom directive.
2017-09-29T18:26:05Z dougk_: now that we know you agree, see if a quick win is to globally declaim (notinline format), to get the compiler to back off from its complaint
2017-09-29T18:26:53Z dougk_: you're likely just missing the compile-time handler for your directive.
2017-09-29T18:29:34Z dougk_: stassats: aren't we "overaligning" the stack in x86-64-assem.S alloc-tramp with "and $-32,%rsp" ? It should be a 16-byte alignment, which is $-16
2017-09-29T18:29:52Z rick-monster: dougk_: that seems to do the trick, thanks very much all.
2017-09-29T18:52:00Z sjl joined #sbcl
2017-09-29T18:57:08Z karswell joined #sbcl
2017-09-29T19:00:38Z dougk quit (Ping timeout: 258 seconds)
2017-09-29T19:14:00Z sjl quit (Ping timeout: 240 seconds)
2017-09-29T19:14:40Z stassats: dougk_: yeah
2017-09-29T19:18:05Z stassats: dougk_: but really inconsequential
2017-09-29T19:27:58Z Jesin quit (Quit: Leaving)
2017-09-29T19:31:41Z Jesin joined #sbcl
2017-09-29T19:55:05Z panji quit (Quit: Leaving)
2017-09-29T20:18:14Z milanj quit (Quit: Leaving)
2017-09-29T20:43:04Z stassats quit (Ping timeout: 240 seconds)
2017-09-29T20:54:09Z jack_rabbit quit (Remote host closed the connection)
2017-09-29T20:58:04Z sjl joined #sbcl
2017-09-29T21:00:13Z dougk joined #sbcl
2017-09-29T21:02:49Z dougk quit (Read error: Connection reset by peer)
2017-09-29T21:03:18Z dougk joined #sbcl
2017-09-29T21:06:27Z attila_lendvai joined #sbcl
2017-09-29T21:38:27Z sjl quit (Ping timeout: 260 seconds)
2017-09-29T21:55:25Z Bike quit (Ping timeout: 258 seconds)
2017-09-29T22:00:14Z Bike joined #sbcl
2017-09-29T22:13:00Z Merv left #sbcl
2017-09-29T22:20:20Z sjl joined #sbcl
2017-09-29T22:30:31Z rick-monster quit (Remote host closed the connection)
2017-09-29T22:34:04Z angavrilov quit (Remote host closed the connection)
2017-09-29T22:42:18Z attila_lendvai quit (Quit: Leaving.)
2017-09-29T23:04:34Z cromachina joined #sbcl
2017-09-29T23:19:53Z zulu_inuoe joined #sbcl
2017-09-30T00:26:04Z dougk quit (Ping timeout: 240 seconds)
2017-09-30T00:37:48Z dougk joined #sbcl
2017-09-30T01:06:39Z whoman quit (Ping timeout: 248 seconds)
2017-09-30T01:41:28Z jack_rabbit joined #sbcl
2017-09-30T01:43:05Z dougk__ joined #sbcl
2017-09-30T01:46:36Z dougk quit (Ping timeout: 240 seconds)
2017-09-30T03:07:00Z zulu_inuoe quit (Read error: Connection reset by peer)
2017-09-30T03:39:39Z Bike quit (Ping timeout: 258 seconds)
2017-09-30T04:47:31Z PuercoPop left #sbcl
2017-09-30T05:11:28Z dougk__ quit (Ping timeout: 248 seconds)
2017-09-30T06:33:23Z stylewarning: Hey folks, I have some small-to-medium sized SBCL projects open for contracting agreement. If anyone would like to inquire, shoot me a PM.
2017-09-30T08:07:43Z angavrilov joined #sbcl
2017-09-30T08:38:20Z dim: Hi!
2017-09-30T08:38:37Z dim: https://github.com/dimitri/pgloader/issues/639 -- fatal error encountered in SBCL ... zlib inflate error: -3
2017-09-30T08:38:50Z dim: it looks like something specific to SBCL rather than to pgloader
2017-09-30T08:39:10Z dim: anything we could ask to the OP of the issue for debugging?
2017-09-30T08:39:24Z Shinmera: Ask him to redownload and verify hash, if that's possible.
2017-09-30T08:39:36Z dim: ok
2017-09-30T08:40:19Z Shinmera: I've had issues where people's binaries were screwed up from a bad download. Haven't encountered that specific one, but it seems probable.
2017-09-30T08:41:06Z dim: I realise I don't provide checksums for pgloader releases on github, that said
2017-09-30T08:44:04Z |3b|: yeah, -3 seems to be "data error" = bad/corrupted input to inflate
2017-09-30T08:44:47Z Shinmera: My issue was not with SBCL itself, but with a self-extracting xz of another kind: https://github.com/portacle/portacle/issues/42
2017-09-30T09:24:15Z josemanuel joined #sbcl
2017-09-30T09:54:10Z Xach joined #sbcl
2017-09-30T09:54:53Z Xach: Hi! quicklisp builds are a lot faster with the latest from git.
2017-09-30T09:54:59Z Xach: latest sbcl, that is.
2017-09-30T09:55:17Z Xach: is there a specific thing someone did lately or is it just a lot of little things adding up?
2017-09-30T10:00:06Z Xach: (or maybe my computer got faster?)
2017-09-30T10:01:33Z Xach: anyway, what used to take about 2 hours now takes 1 hour 40 minutes
2017-09-30T10:13:31Z josemanuel quit (Quit: leaving)
2017-09-30T10:58:24Z scymtym joined #sbcl
2017-09-30T11:42:33Z stassats joined #sbcl
2017-09-30T11:42:52Z stassats: Xach: the only thing i can think of is GC improvements
2017-09-30T11:44:53Z Bike joined #sbcl
2017-09-30T12:01:29Z Bicyclidine joined #sbcl
2017-09-30T12:03:21Z Bike quit (Ping timeout: 258 seconds)
2017-09-30T12:34:00Z leah2 quit (Ping timeout: 252 seconds)
2017-09-30T12:50:00Z leah2 joined #sbcl
2017-09-30T13:13:22Z scymtym: speaking of compiler optimizations: does this look ok? http://paste.lisp.org/display/357293
2017-09-30T13:18:45Z stassats: scymtym: is the first call to (ensure-blocks-for-env component env block-cache) necessary?
2017-09-30T13:21:38Z stassats: and is the order relevant? i.e. is the nreverse necessary?
2017-09-30T13:25:42Z scymtym: stassats: first call: it should not be necessary. should i try to allocate the hash-table lazily as well?
2017-09-30T13:26:49Z scymtym: order: i don't know. i will try to find out
2017-09-30T13:26:50Z stassats: is it often not required?
2017-09-30T13:28:14Z scymtym: i guess the only way to tell is instrumenting and measuring
2017-09-30T13:33:34Z scymtym: the hash-table may not be necessary at all
2017-09-30T13:34:01Z stassats: storing in the node?
2017-09-30T13:34:05Z stassats: in the environment?
2017-09-30T13:34:38Z scymtym: just local scope may be enough
2017-09-30T13:34:59Z scymtym: can different lambdas have the same physenv?
2017-09-30T13:35:50Z stassats: as in EQ?
2017-09-30T13:36:10Z scymtym: yes
2017-09-30T13:36:39Z stassats: for lets, yeah
2017-09-30T13:36:46Z stassats: that's what it says
2017-09-30T13:37:44Z scymtym: then the question is whether to use cache for the component so that component lambdas can shared block lists or just compute the block list for each component lambda and not use the hash-table at all
2017-09-30T13:39:45Z scymtym measures how often that kind of sharing occurs in practice
2017-09-30T13:41:57Z scymtym: answer: never (and 90 % of the components compiled had either 1 or 2 lambdas)
2017-09-30T13:44:23Z scymtym: so i guess i can get rid of the hash-table and precompute the block list into a local variable
2017-09-30T13:47:29Z AndreasO joined #sbcl
2017-09-30T13:57:28Z scymtym: regarding order: i think i can change DO-BLOCKS-BACKWARDS to DO-BLOCKS and eliminate the NREVERSE
2017-09-30T14:12:51Z stassats: Value of (BAR) in (THE NIL (BAR)) is NIL, not a NIL.
2017-09-30T14:12:56Z stassats: what a great wording
2017-09-30T14:13:49Z scymtym: maybe "… is NIL which is not of type NIL" instead?
2017-09-30T14:14:07Z stassats: thinking special casing NIL into that
2017-09-30T14:14:46Z stassats: but i bet it's the same confusion if you're new to this
2017-09-30T14:15:00Z stassats: maybe "of type NIL (empty type)"
2017-09-30T14:15:08Z stassats: clhs nil/t
2017-09-30T14:15:08Z specbot: http://www.lispworks.com/reference/HyperSpec/Body/t_nil.htm
2017-09-30T14:15:26Z stassats: and "empty type" is an official term
2017-09-30T14:15:35Z AndreasO quit (Quit: Found more important stuff than irc!)
2017-09-30T14:16:10Z scymtym: colon and at modifiers in PRINT-TYPE[-SPECIFIER] could be used to enable that
2017-09-30T14:16:14Z stassats: but (the (and fixnum list) (bar)) folds down to that
2017-09-30T14:16:18Z dougk joined #sbcl
2017-09-30T14:16:18Z stassats: how to make it less confusing?
2017-09-30T14:17:05Z scymtym: is (THE NIL (BAR)) reconstructed or is it the source form?
2017-09-30T14:17:27Z stassats: both
2017-09-30T14:17:35Z stassats: but i doubt anyone will do (the nil (bar))
2017-09-30T14:18:18Z stassats: (the (and fixnum list) (bar)) does appear in the error message
2017-09-30T14:18:29Z stassats: Value of (BAR) in (THE (AND FIXNUM LIST) (BAR)) is NIL, not a NIL.
2017-09-30T14:20:26Z stassats: maybe nobody actually sees this
2017-09-30T14:21:16Z stassats: what about warning when (the x) resolves to an empty type?
2017-09-30T14:22:15Z scymtym: i like it
2017-09-30T14:22:55Z stassats: and i forgot what i was doing
2017-09-30T14:23:16Z stassats: ah, testing nil-fun-returned-error
2017-09-30T14:24:43Z scymtym: stassats: btw, did you see the experiments i did based on beach's fastgf stuff?
2017-09-30T14:24:48Z stassats: nope
2017-09-30T14:25:37Z scymtym: are you interested? (i don't want to keep you from nil-fun-returned-error)
2017-09-30T14:25:45Z |3b| wonders how (the nil (values)) should be interpreted
2017-09-30T14:26:07Z stassats: an error
2017-09-30T14:26:24Z stassats: (values &optional) is the type of (values)
2017-09-30T14:27:13Z stassats: scymtym: go ahead
2017-09-30T14:27:58Z |3b|: so (the NIL (values)) is saying it will return 1 value, of type NIL?
2017-09-30T14:28:09Z stassats: rather that it will never return
2017-09-30T14:28:44Z |3b|: ah, THE considers missing values to be NIL, so same as any other value
2017-09-30T14:30:59Z dougk quit (Ping timeout: 240 seconds)
2017-09-30T14:31:42Z scymtym: stassats: beach's is to use a decision tree to discriminate between "class versions" (beach has "stamps" in "general instances", we have layouts). i extended this to use a cascade of decision trees for lowtag, lowtag then widetag, lowtag then layout and lowtag then eq. example: http://paste.lisp.org/display/356932 . bottom line so far: faster than pcl but requires more compilation in case of dispatch misses
2017-09-30T14:31:58Z scymtym: * beach's approach
2017-09-30T14:32:33Z stassats: yeah, i'm familiar with the thing
2017-09-30T14:33:37Z stassats: is it still limited to one argument functions?
2017-09-30T14:34:16Z scymtym: yes. going to multiple arguments is probably the critical experiment
2017-09-30T14:35:09Z scymtym: could go both ways since pcl doesn't need much extra work for additional arguments, but decision trees for additional arguments should also be shallow
2017-09-30T14:35:59Z stassats: except that changing anything in PCL is a pain
2017-09-30T14:36:05Z scymtym: benchmark for a generic function that has methods for (and has been called with) lots of different specializers https://techfak.de/~jmoringe/benchmark-1.png
2017-09-30T14:36:25Z scymtym: sure. it would be a vast simplification, though
2017-09-30T14:37:24Z |3b|: is that 1 second per call?
2017-09-30T14:37:27Z scymtym: also, PCL seems to fall back to the TYPEP-based "discrimination net" strategy if you add the wrong specializer. and then it's realy slow
2017-09-30T14:37:28Z stassats: well, if you can make it work on accessors first, that'd be already an improvement
2017-09-30T14:37:37Z |3b|: or for 200M?
2017-09-30T14:37:43Z scymtym: |3b|: the latter
2017-09-30T14:38:23Z stassats: and invalid layouts?
2017-09-30T14:39:17Z scymtym: invalid layouts will be dispatch misses but invalidating layouts will be more expensive
2017-09-30T14:39:35Z stassats: invalidating layouts is even broken
2017-09-30T14:39:40Z stassats: (for multiple threads)
2017-09-30T14:39:57Z scymtym: in accessors?
2017-09-30T14:40:07Z scymtym: or during dispatch?
2017-09-30T14:40:15Z stassats: anywhere
2017-09-30T14:40:49Z stassats: it's done in two steps, not in one, as you can imagine, that's a problem for multiple threads
2017-09-30T14:41:24Z stassats: it's been on my "really nice to sort out" list for too long
2017-09-30T14:41:49Z scymtym: setting the hash to zero and removing the layout from dispatch caches?
2017-09-30T14:43:35Z leah2 quit (Ping timeout: 258 seconds)
2017-09-30T14:43:40Z stassats: it's the classoid-typep problem
2017-09-30T14:44:23Z scymtym: that would be the case of accessors of classes which are not specializers in a method?
2017-09-30T14:45:11Z stassats: basically the layout has two invalid states, :invalid and something else, T or something
2017-09-30T14:45:31Z stassats: so it's like a state machine, except bad things happen if two threads compete between the steps
2017-09-30T14:45:36Z scymtym: i thought there were more, but i'm not familiar with the details
2017-09-30T14:47:09Z stassats: i don't remember the details
2017-09-30T14:48:18Z scymtym: we could compile every TYPEP (or better TYPECASE) into something like a dispatch function that updates itself. but that would only work efficiently if we could dynamically inline the new code
2017-09-30T14:49:01Z stassats: i think just making invalidation be done in one step should solve this
2017-09-30T14:49:38Z scymtym: if that is possible. i remember that beach takes an extra indirection to get at the slot vector "just" to make invalidation atomic
2017-09-30T14:50:06Z stassats: well, each step itself is locked
2017-09-30T14:50:11Z stassats: so if it's one step, there's no problem
2017-09-30T14:51:43Z scymtym: but taking a lock to /check/ validity would be expensive
2017-09-30T14:51:58Z stassats: validity checks are not locked
2017-09-30T14:52:22Z stassats: only when it's invalid
2017-09-30T14:52:48Z scymtym: ok, i looked at that code once, but forgot since
2017-09-30T14:54:11Z stassats: https://bugs.launchpad.net/sbcl/+bug/1153309/comments/7 still fails
2017-09-30T15:00:55Z leah2 joined #sbcl
2017-09-30T15:06:20Z stassats quit (Quit: ERC (IRC client for Emacs 25.1.2))
2017-09-30T15:07:01Z stassats joined #sbcl
2017-09-30T15:47:35Z leah2 quit (Ping timeout: 258 seconds)
2017-09-30T15:59:36Z scymtym: i'm not 100 % confident i measured this correctly, but if so, the optimization discussed earlier takes compiling ironclad from 78 to 65 seconds on my machine
2017-09-30T16:01:30Z leah2 joined #sbcl
2017-09-30T16:07:27Z scymtym quit (Ping timeout: 255 seconds)
2017-09-30T16:22:27Z Bike joined #sbcl
2017-09-30T16:26:18Z Bicyclidine quit (Ping timeout: 258 seconds)
2017-09-30T16:41:04Z panji joined #sbcl
2017-09-30T16:51:44Z flip214: isn't it good enough to store the last (or most often seen) type, and have the right dispatch cached for that one?
2017-09-30T16:51:53Z flip214: that should be nearly as fast as changing the inline code, I guess
2017-09-30T16:56:51Z whoman joined #sbcl
2017-09-30T17:05:32Z panji quit (Ping timeout: 240 seconds)
2017-09-30T18:44:37Z dougk joined #sbcl
2017-09-30T18:58:52Z Xof joined #sbcl
2017-09-30T19:30:38Z knobo: what's the mail reason for the big verion number jump?
2017-09-30T19:34:07Z stassats: is it big?
2017-09-30T19:34:55Z knobo: ah.. not mail reason, but main reason...
2017-09-30T19:35:05Z knobo: stassats: depends on how you count.
2017-09-30T19:37:01Z stassats: base 22
2017-09-30T19:38:21Z knobo: then not so big.
2017-09-30T19:38:42Z Xof: knobo: I mailed both the sbcl developers and the sbcl announce list to explain the rationale for the version number "jump"
2017-09-30T19:39:03Z knobo: but in base 58 it's big.
2017-09-30T19:39:26Z knobo: Aha. I'll check it out.
2017-09-30T19:41:17Z Xof: (the sbcl developers at freeze time; the announce list a month ago)
2017-09-30T19:41:44Z Xof: spoiler: there is no particular reason other than I find it hard to remember whether we're on 21 or 22, but 0 vs 1 is easy
2017-09-30T19:42:23Z knobo: Aha.. because of why now :)
2017-09-30T19:42:50Z knobo: I like it.
2017-09-30T19:43:25Z knobo: Event though my first thought is this a less stable release then 21 to 22 would be.
2017-09-30T19:44:05Z knobo: first thought was: is ....
2017-09-30T19:51:24Z prxq joined #sbcl
2017-09-30T20:10:46Z scymtym joined #sbcl
2017-09-30T20:18:24Z knobo: Thank you guys for the great work you are doing :)
2017-09-30T20:20:30Z sjl_ joined #sbcl
2017-09-30T20:22:26Z sjl quit (Ping timeout: 258 seconds)
2017-09-30T20:24:21Z stassats quit (Ping timeout: 258 seconds)
2017-09-30T21:25:09Z Bicyclidine joined #sbcl
2017-09-30T21:26:50Z Bike quit (Ping timeout: 258 seconds)
2017-09-30T21:37:29Z shka_ quit (Ping timeout: 240 seconds)
2017-09-30T21:58:16Z dougk quit (Ping timeout: 258 seconds)
2017-09-30T21:58:36Z dougk joined #sbcl
2017-09-30T22:00:33Z angavrilov quit (Remote host closed the connection)
2017-09-30T22:20:09Z prxq quit (Remote host closed the connection)
2017-09-30T22:21:19Z attila_lendvai joined #sbcl
2017-09-30T22:23:27Z Bike joined #sbcl
2017-09-30T22:25:35Z eschatologist quit (Read error: Connection reset by peer)
2017-09-30T22:26:05Z eschatologist joined #sbcl
2017-09-30T22:26:09Z Bicyclidine quit (Ping timeout: 246 seconds)
2017-09-30T22:27:44Z eschatologist quit (Remote host closed the connection)
2017-09-30T22:30:01Z eschatologist joined #sbcl
2017-09-30T22:45:03Z White_Flame quit (Ping timeout: 252 seconds)
2017-09-30T23:04:07Z White_Flame joined #sbcl
2017-09-30T23:12:29Z dougk quit (Ping timeout: 248 seconds)
2017-09-30T23:23:20Z stassats joined #sbcl
2017-09-30T23:26:47Z dougk joined #sbcl