2014-11-03T00:14:51Z edgar-rft joined #sbcl
2014-11-03T00:31:18Z leo2007 joined #sbcl
2014-11-03T00:40:07Z oleo quit (Read error: Connection reset by peer)
2014-11-03T00:41:27Z oleo joined #sbcl
2014-11-03T00:54:00Z leo2007 quit (Remote host closed the connection)
2014-11-03T02:11:03Z Hache_ quit (Remote host closed the connection)
2014-11-03T02:29:04Z cmack` quit (Ping timeout: 255 seconds)
2014-11-03T02:58:03Z Hydan quit (Quit: ZNC - http://znc.in)
2014-11-03T03:00:49Z Hydan joined #sbcl
2014-11-03T03:04:54Z ehaliewicz joined #sbcl
2014-11-03T03:39:05Z cmack` joined #sbcl
2014-11-03T03:39:35Z christoph_debian quit (Ping timeout: 265 seconds)
2014-11-03T03:45:42Z cmack`` joined #sbcl
2014-11-03T03:48:29Z cmack` quit (Ping timeout: 264 seconds)
2014-11-03T03:52:53Z christoph_debian joined #sbcl
2014-11-03T04:54:41Z LiamH quit (Quit: Leaving.)
2014-11-03T05:54:02Z pranavrc joined #sbcl
2014-11-03T06:10:53Z psy_ quit (Read error: No route to host)
2014-11-03T06:11:53Z psy_ joined #sbcl
2014-11-03T06:23:04Z oleo quit (Quit: Verlassend)
2014-11-03T06:39:38Z scymtym_ quit (Ping timeout: 256 seconds)
2014-11-03T08:40:01Z psy_ quit (Ping timeout: 244 seconds)
2014-11-03T08:59:13Z jdz quit (Ping timeout: 260 seconds)
2014-11-03T09:03:57Z jdz joined #sbcl
2014-11-03T09:11:56Z segv- joined #sbcl
2014-11-03T09:29:40Z rjmacready_ joined #sbcl
2014-11-03T09:42:30Z ehaliewicz quit (Ping timeout: 258 seconds)
2014-11-03T09:45:52Z ehaliewicz joined #sbcl
2014-11-03T10:00:41Z psy_ joined #sbcl
2014-11-03T10:06:19Z ehaliewicz quit (Ping timeout: 265 seconds)
2014-11-03T11:58:08Z DeadTrickster joined #sbcl
2014-11-03T11:58:57Z DeadTrickster: hey guys, question here: why sbcl doesn't call compiler macro function in apply and funcall forms?
2014-11-03T11:59:39Z DeadTrickster: at least in repl
2014-11-03T12:00:00Z DeadTrickster: yep just in repl
2014-11-03T12:04:44Z Krystof: it might have decided not to compile the form it's evaluating
2014-11-03T12:09:38Z DeadTrickster: but sb-ext:*evaluator-mode* is :comple
2014-11-03T12:11:13Z DeadTrickster: I'm looking at simple-eval-in-lexenv source now, it just get symbol function and applies it to args
2014-11-03T12:11:44Z DeadTrickster: doesn't look like compilation to me
2014-11-03T12:16:41Z Krystof: but :compile doesn't mean "compile every single thing, yes even that"
2014-11-03T12:16:54Z Krystof: it means "compile everything except basically trivial things"
2014-11-03T12:22:03Z DeadTrickster: can compiler macros be used by evaluator?
2014-11-03T12:22:37Z Krystof: generally, you shouldn't care
2014-11-03T12:22:40Z Krystof: why do you care?
2014-11-03T12:24:54Z DeadTrickster: i wrote string pattern support for local time, then I wrote compiler macro to compile string pattern to native lists. If format-timestring placed in body of another function everything forks but if It
2014-11-03T12:25:17Z DeadTrickster: 's repl top level compiler macro never invoked
2014-11-03T12:25:29Z Krystof: so?  Why do you care?
2014-11-03T12:25:44Z Krystof: a compiler macro is for doing work in the hope of making things faster in the long run
2014-11-03T12:25:54Z Krystof: at the repl, you are executing things exactly once
2014-11-03T12:26:48Z DeadTrickster: I understand this it's just strange too me that it works here but doesn't work there
2014-11-03T12:26:52Z jsnell: compiler macros should not be used for implementing different semantics than the normal function
2014-11-03T12:27:28Z DeadTrickster: it's the same semantic - construct string representation of timestamp according to format
2014-11-03T12:27:57Z Krystof: eval knows that it will evaluate its argument exactly once, so it doesn't try to do extra work in order to make that evaluation fast, because it knows it only gets to amortize that extra work over one evaluation
2014-11-03T12:28:29Z Krystof: compile and compile-file assume that they can do one bit of work at compile-time to make potentially many evaluations, later, faster, and that this will be beneficial
2014-11-03T12:29:02Z Krystof: so, maybe now it is less strange that EVAL doesn't go out of its way to use compiler macros when it doesn't have to
2014-11-03T12:29:07Z segv- quit (Ping timeout: 244 seconds)
2014-11-03T12:29:28Z DeadTrickster: OK I will add strings support directly to format-timestring
2014-11-03T12:29:45Z DeadTrickster: Thank you I understand reasoning now
2014-11-03T12:34:41Z Hache_ joined #sbcl
2014-11-03T13:00:52Z eudoxia joined #sbcl
2014-11-03T13:01:35Z psy_ quit (Remote host closed the connection)
2014-11-03T13:02:56Z attila_lendvai quit (Quit: Leaving.)
2014-11-03T13:14:01Z stassats joined #sbcl
2014-11-03T13:22:52Z Krystof: csr21@aleph-null:/tmp/tmp.kk7O7RBa8G$ diff -qr sbcl/obj/ ccl32/obj | wc
2014-11-03T13:22:52Z Krystof:       0       0       0
2014-11-03T13:28:35Z pranavrc quit
2014-11-03T13:50:08Z psy_ joined #sbcl
2014-11-03T13:54:22Z eudoxia quit (Quit: Lost terminal)
2014-11-03T13:56:54Z pkhuong: neat
2014-11-03T13:58:28Z Krystof: now I just need to find a way to not double the build time, and we're done
2014-11-03T13:59:30Z Krystof: still, it's nice to know that it's possible to write portable common lisp, even if it is slow
2014-11-03T14:16:01Z LiamH joined #sbcl
2014-11-03T14:24:38Z stassats quit (Ping timeout: 244 seconds)
2014-11-03T14:26:49Z psy_ quit (Read error: Connection reset by peer)
2014-11-03T14:31:17Z Hydan is now known as Hydan`
2014-11-03T14:31:18Z Hydan` is now known as Hydan
2014-11-03T14:43:21Z LiamH quit (Quit: Leaving.)
2014-11-03T15:06:20Z Hydan is now known as Hydan`
2014-11-03T15:07:12Z oleo joined #sbcl
2014-11-03T15:09:01Z stassats joined #sbcl
2014-11-03T15:22:18Z psy_ joined #sbcl
2014-11-03T15:33:02Z specbot quit (Remote host closed the connection)
2014-11-03T15:34:17Z specbot joined #sbcl
2014-11-03T15:36:15Z lacedaemon joined #sbcl
2014-11-03T15:36:21Z fe[nl]ix quit (Ping timeout: 260 seconds)
2014-11-03T15:36:42Z lacedaemon is now known as fe[nl]ix
2014-11-03T15:43:38Z irsol quit (Ping timeout: 256 seconds)
2014-11-03T15:45:48Z irsol joined #sbcl
2014-11-03T16:05:22Z specbot quit (Remote host closed the connection)
2014-11-03T16:05:28Z specbot joined #sbcl
2014-11-03T16:07:37Z snafuchs quit (Ping timeout: 260 seconds)
2014-11-03T16:08:09Z snafuchs joined #sbcl
2014-11-03T16:17:22Z krzysz00 quit (Ping timeout: 260 seconds)
2014-11-03T16:20:40Z cmack`` quit (Read error: Connection reset by peer)
2014-11-03T16:20:56Z cmack`` joined #sbcl
2014-11-03T16:39:18Z edgar-rft quit (Quit: activity corrupted because computer crashed)
2014-11-03T16:57:26Z krzysz00 joined #sbcl
2014-11-03T17:23:32Z slyrus joined #sbcl
2014-11-03T17:30:29Z rpg joined #sbcl
2014-11-03T17:39:14Z Krystof: csr21@aleph-null:/tmp/tmp.kk7O7RBa8G$ diff sbcl/output/sbcl.core ccl32/output/sbcl.core
2014-11-03T17:39:14Z Krystof: Binary files sbcl/output/sbcl.core and ccl32/output/sbcl.core differ
2014-11-03T17:39:16Z Krystof: waah
2014-11-03T17:46:41Z flip214: "Don't cry for me, Argentina...."
2014-11-03T17:51:57Z krzysz00 quit (Ping timeout: 245 seconds)
2014-11-03T17:54:51Z oleo: yo, i get a chill down my back when you cry.....
2014-11-03T17:54:55Z oleo: lol
2014-11-03T17:55:03Z Krystof: I suspect that my second most useless skill -- decompiling fasl files by hand -- is not going to help with the core file differences
2014-11-03T17:57:04Z flip214: what's the most useless one?
2014-11-03T17:59:18Z slyrus quit (Ping timeout: 272 seconds)
2014-11-03T18:03:38Z Krystof: five-dimensional general relativity
2014-11-03T18:05:15Z flip214: that's not useless.
2014-11-03T18:05:22Z flip214: that must be a hit with the girls!
2014-11-03T18:05:34Z flip214: and neither is reading fasl files....
2014-11-03T18:05:40Z Bike: gotta get as many components in that tensor as possible
2014-11-03T18:05:55Z flip214: if it was, then my knowledge of 6502 assembler might be called useless, too, and I can't have that.
2014-11-03T18:06:08Z stassats: a hack, what if instead of alloc_number for heap_eaxhausted, we allocate a SAP, dx allocate, that is
2014-11-03T18:06:29Z stassats: allocating saps is easier than allocating bignums, although that would mean we always have to, no fixnums here
2014-11-03T18:12:34Z krzysz00 joined #sbcl
2014-11-03T18:16:21Z jlarocco: is it a known issue that some of the new sb-bsd-socket tests fail when IPv6 is disabled?
2014-11-03T18:17:08Z stassats: sorta
2014-11-03T18:17:09Z scymtym: jlarocco: yes, stassats already fixed that in master
2014-11-03T18:17:15Z jlarocco: I think they just need an ipv6 feature check
2014-11-03T18:17:19Z jlarocco: okay, thanks
2014-11-03T18:17:20Z stassats: no, i didn't fix it, really
2014-11-03T18:17:32Z stassats: i disable the tests on windows, where ipv6 is not implemented
2014-11-03T18:17:52Z stassats: so, if you have it disabled on linux, you'll get your failures
2014-11-03T18:17:59Z scymtym: oh, are we talking about non-windows platforms without ipv6 addresses?
2014-11-03T18:18:09Z stassats: jlarocco: do you know how to check whether ipv6 is enabled?
2014-11-03T18:18:17Z jlarocco: yeah, Ubuntu with ipv6 disabled
2014-11-03T18:18:29Z Krystof: oh wait we embed a build-id hash value in sbcl.core
2014-11-03T18:18:30Z Krystof: d'oh
2014-11-03T18:19:00Z stassats: right, that's how src/runtime/sbcl knows it can mate with sbcl.core
2014-11-03T18:19:42Z Krystof: yeah.  So it's not (shouldn't be) surprising that my cores are different
2014-11-03T18:20:16Z jlarocco: I do, cat /proc/net/if_inet6 returns no results, and same with lsmod | grep ipv6
2014-11-03T18:21:01Z Hache_ quit (Remote host closed the connection)
2014-11-03T18:21:15Z scymtym: i think the ipv6 bind/connect test fails when the loopback interface does not have an ipv6 address. but i don't know how to inspect that property of the interface without adding more functionality to sb-bsd-sockets.
2014-11-03T18:22:47Z Krystof: is it not the case that the ipv4 tests fail when we don't have ipv4 support?
2014-11-03T18:23:28Z stassats: Krystof: you can cheat by creating the same output/build-id.tmp
2014-11-03T18:23:47Z slyrus joined #sbcl
2014-11-03T18:24:11Z Krystof: I might be inclined to document that the build needs to be run on a host supporting both ipv4 and ipv6
2014-11-03T18:24:47Z scymtym: the ipv4 binds/connects to 127.0.0.1, so i would expect it to fail similarly
2014-11-03T18:24:59Z scymtym: s/the ipv4/the ipv4 test/
2014-11-03T18:25:25Z stassats: cat /proc/net/if_inet6 is not that good a way
2014-11-03T18:25:41Z Krystof: or maybe we need to distinguish between test failures that suggest that a contrib is totally broken and test failures that suggest that the build environment is wrong (would also help with sb-posix, probably)
2014-11-03T18:26:04Z krzysz00 quit (Ping timeout: 256 seconds)
2014-11-03T18:26:56Z scymtym: sb-bsd-sockets could support enumerating devices and their addresses, but doing that portably is probably hard
2014-11-03T18:27:05Z scymtym: s/devices/interfaces/
2014-11-03T18:27:58Z krzysz00 joined #sbcl
2014-11-03T18:30:20Z kuimacro joined #sbcl
2014-11-03T18:39:34Z krzysz00 quit (Ping timeout: 244 seconds)
2014-11-03T18:41:24Z krzysz00 joined #sbcl
2014-11-03T18:51:17Z krzysz00 quit (Ping timeout: 264 seconds)
2014-11-03T18:54:22Z rjmacready_ quit (Ping timeout: 246 seconds)
2014-11-03T19:00:40Z segv- joined #sbcl
2014-11-03T19:45:02Z kuimacro quit (Quit: Nettalk6 - www.ntalk.de)
2014-11-03T19:46:05Z fiveop joined #sbcl
2014-11-03T19:51:37Z stassats quit (Ping timeout: 258 seconds)
2014-11-03T19:55:33Z Hache_ joined #sbcl
2014-11-03T20:01:10Z fiveop quit (Remote host closed the connection)
2014-11-03T20:02:25Z fiveop joined #sbcl
2014-11-03T20:03:24Z developernotes joined #sbcl
2014-11-03T20:07:07Z fiveop quit (Remote host closed the connection)
2014-11-03T20:10:13Z krzysz00 joined #sbcl
2014-11-03T20:12:03Z stassats joined #sbcl
2014-11-03T20:33:03Z attila_lendvai joined #sbcl
2014-11-03T20:35:12Z Bicyclidine joined #sbcl
2014-11-03T20:53:55Z stassats is excited about 1.2.6
2014-11-03T20:54:14Z stassats: finally fixes for longstanding issues
2014-11-03T20:56:31Z stassats: who cares about fast or featurful
2014-11-03T20:57:29Z krzysz00 quit (Ping timeout: 255 seconds)
2014-11-03T21:01:39Z slyrus quit (Ping timeout: 265 seconds)
2014-11-03T21:22:32Z krzysz00 joined #sbcl
2014-11-03T21:22:33Z slyrus joined #sbcl
2014-11-03T21:51:07Z cmack`` is now known as cmack
2014-11-03T21:59:52Z Bicyclidine quit (Ping timeout: 240 seconds)
2014-11-03T22:08:50Z segv- quit (Ping timeout: 265 seconds)
2014-11-03T22:08:54Z Bicyclidine joined #sbcl
2014-11-03T22:13:41Z stassats quit (Ping timeout: 264 seconds)
2014-11-03T22:20:08Z scymtym_ joined #sbcl
2014-11-03T23:02:05Z Bicyclidine quit (Ping timeout: 264 seconds)
2014-11-03T23:05:35Z developernotes quit (Quit: Textual IRC Client: www.textualapp.com)
2014-11-03T23:11:11Z Bicyclidine joined #sbcl
2014-11-03T23:37:53Z attila_lendvai quit (Quit: Leaving.)