00:12:21 -!- prxq [~mommer@mnhm-5f75c96b.pool.mediaWays.net] has quit [Remote host closed the connection] 01:41:43 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving] 01:42:43 echo-area [~user@182.92.247.2] has joined #sbcl 03:12:44 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 05:05:14 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 05:09:23 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 05:20:18 attila_lendvai [~attila_le@37.99.82.225] has joined #sbcl 05:20:19 -!- attila_lendvai [~attila_le@37.99.82.225] has quit [Changing host] 05:20:19 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 05:40:56 dtw [~dtw@pdpc/supporter/active/dtw] has joined #sbcl 05:57:28 sdemarre [~serge@91.176.38.72] has joined #sbcl 06:11:44 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 06:41:25 -!- edgar-rft [~user@HSI-KBW-078-043-123-191.hsi4.kabel-badenwuerttemberg.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 08:00:43 nikodemus [~nikodemus@cs181241043.pp.htv.fi] has joined #sbcl 08:00:43 -!- ChanServ has set mode +o nikodemus 08:02:08 morning 08:09:16 Morning. 08:21:38 prxq [~mommer@mnhm-5f75f76a.pool.mediaWays.net] has joined #sbcl 08:35:30 morning 08:37:19 morning 08:57:48 -!- dtw [~dtw@pdpc/supporter/active/dtw] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 11:05:42 -!- ASau [~user@95-27-143-55.broadband.corbina.ru] has quit [Remote host closed the connection] 11:06:36 ASau [~user@95-27-143-55.broadband.corbina.ru] has joined #sbcl 11:38:29 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 248 seconds] 11:39:34 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 12:30:28 pchrist [~spirit@gentoo/developer/pchrist] has joined #sbcl 12:54:49 dtw [~dtw@pdpc/supporter/active/dtw] has joined #sbcl 13:13:34 -!- flip215 [~marek@unaffiliated/flip214] has quit [Ping timeout: 276 seconds] 13:13:53 flip215 [~marek@86.59.100.100] has joined #sbcl 13:13:53 -!- flip215 [~marek@86.59.100.100] has quit [Changing host] 13:13:53 flip215 [~marek@unaffiliated/flip214] has joined #sbcl 13:42:53 -!- dtw [~dtw@pdpc/supporter/active/dtw] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 14:20:10 Kryztof: aroundp 14:25:24 LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has joined #sbcl 14:27:30 hi 14:27:35 nikodemus: what's up? 14:28:29 looking at one of my old sb-alien optimization projects 14:29:13 i think i'm going to need either def!global, or i need to make plain defglobal more xc-host/xc/target aware 14:30:37 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 14:30:43 I think the latter 14:31:02 def!foo is really for "do deffoo on the host as well as on the xc" 14:31:42 or "do deffoo on the host, and -- as much as it's possible -- make sure that the effects of deffoo are known to have happened to the cross-compiler/target as soon as possible" 14:32:03 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 244 seconds] 14:32:43 basically, i'm trying to intern some alien types 14:33:41 if you really want a host effect too (e.g. making a special binding) then def!global can be `(progn #+sb-xc-host (defvar foo ...) (sb-ext:defglobal foo ...)) and sb-ext:defglobal can be `(progn #+sb-xc-host (inform the xc about this) (normal-expansion of defglobal)) 14:34:10 hm 14:34:22 so you want that when you load a host fasl, some of the cross-compiler's alien datastructures should already be created? 14:34:29 i think i need to think harder 14:34:35 i'm not quite 100% sure yet what effects i need 14:35:19 i *think* so 14:36:04 leuler [~user@p5490545C.dip.t-dialin.net] has joined #sbcl 14:36:05 this sounds like the difference between cl:defstruct and sb-xc:defstruct 14:36:24 if you compile a file with cl:defstruct in it, the host compiler will get that structure and the cross-compiler won't know about it at all 14:36:27 putting (defglobal **alien-system-pointer-type** nil) in host-alieneval.lisp and referring to the variable in the same file currently breaks building the xc 14:37:05 if you compile a file with sb-xc:defstruct in it, the host compiler won't get that structure, but the cross-compiler will know how to compile its constructor and accessors 14:37:20 for bits of the compiler implementation itself, you want both, hence def!struct 14:37:58 similarly, if you compile a file with cl:defconstant, the host gets that constant; sb-xc:defconstant tells the cross-compiler about its value; sometimes you need both, so def!constant 14:38:14 let me look at what defglobal actually does. But yes, sounds like there you might need def!global 14:39:12 "tell the cross-compiler about this global variable, so that it can compile references to it appropriately. Also, we're going to need this global variable on the host, so please define it" 14:39:37 OTOH, maybe LOAD-TIME-VALUE would be a better match for this particular case :) 14:39:47 but i'm pretty sure i will want a couple of variables sooner or later as well 14:49:38 hey, nifty. LOAD-TIME-VALUE breaks in exact same manner as my attempt at making DEFGLOBAL smarter about this, so maybe it's interning the system-area-pointer type that's the problem... 14:58:36 how does it break 15:05:28 http://paste.lisp.org/display/128683 15:18:26 -!- nikodemus [~nikodemus@cs181241043.pp.htv.fi] has quit [Ping timeout: 252 seconds] 15:27:46 nikodemus [~nikodemus@188-67-118-40.bb.dnainternet.fi] has joined #sbcl 15:27:55 -!- ChanServ has set mode +o nikodemus 15:28:11 uh, that's not the way I would have expected it to break 15:28:20 any chance of a backtrace from there? 15:29:50 huh, turns out it wasn't broken, but spending absolute ages in gc_and_save 15:30:04 *now* it broke on target :) 15:30:42 ...but it broke on my debug printing, so... time to build again 15:45:04 hm. and taking ages doesn't seem to have anything to do with interning the type. maybe i just had another process go nuts at the same time 15:51:22 or maybe you've radically increased the size of the after-xc-core or the cold-sbcl.core? 16:00:42 possible. it's also possible i wasn't building the tree i thought i was building... 16:03:50 bots, Kryztof and nikodemus all present. just like the good ol' days. 16:08:43 did wnewman and danb use to come on IRC ? 16:09:54 danb certainly 16:10:09 -!- nikodemus [~nikodemus@188-67-118-40.bb.dnainternet.fi] has quit [Ping timeout: 265 seconds] 16:13:20 wnewman too 16:15:26 one of the pinnacles of lisp irc humour was between dan`b and wnewman 16:24:31 nikodemus [~nikodemus@188-67-118-40.bb.dnainternet.fi] has joined #sbcl 16:24:31 -!- ChanServ has set mode +o nikodemus 16:27:32 -!- sdemarre [~serge@91.176.38.72] has quit [Ping timeout: 246 seconds] 16:28:10 -!- nikodemus [~nikodemus@188-67-118-40.bb.dnainternet.fi] has quit [Client Quit] 16:55:23 dtw [~dtw@pdpc/supporter/active/dtw] has joined #sbcl 17:02:50 codeforkjeff [~codeforkj@c-69-249-201-156.hsd1.pa.comcast.net] has joined #sbcl 17:08:54 -!- cmm [~cmm@bzq-79-176-208-156.red.bezeqint.net] has quit [Ping timeout: 260 seconds] 17:10:31 -!- codeforkjeff [~codeforkj@c-69-249-201-156.hsd1.pa.comcast.net] has quit [Quit: Leaving] 17:10:36 cmm [~cmm@bzq-79-180-196-52.red.bezeqint.net] has joined #sbcl 17:19:56 -!- mgodshall [~quassel@pool-108-36-207-226.phlapa.fios.verizon.net] has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.] 17:23:11 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 246 seconds] 17:24:39 mgodshall [~quassel@pool-108-36-207-226.phlapa.fios.verizon.net] has joined #sbcl 17:26:41 -!- LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has quit [Ping timeout: 246 seconds] 17:44:38 -!- Phoodus [~foo@wsip-68-107-217-139.ph.ph.cox.net] has quit [Ping timeout: 245 seconds] 17:49:45 gabnet [~gabnet@ACaen-257-1-92-165.w86-220.abo.wanadoo.fr] has joined #sbcl 17:50:07 -!- leuler [~user@p5490545C.dip.t-dialin.net] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 18:06:08 -!- flip215 [~marek@unaffiliated/flip214] has quit [Ping timeout: 260 seconds] 18:38:00 LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has joined #sbcl 18:42:17 -!- LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has quit [Ping timeout: 246 seconds] 18:44:52 -!- dtw [~dtw@pdpc/supporter/active/dtw] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 18:46:36 -!- gabnet [~gabnet@ACaen-257-1-92-165.w86-220.abo.wanadoo.fr] has quit [Quit: Quitte] 19:39:46 LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has joined #sbcl 19:46:10 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 260 seconds] 20:21:28 edgar-rft [~user@HSI-KBW-078-043-123-191.hsi4.kabel-badenwuerttemberg.de] has joined #sbcl 20:51:47 -!- LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has quit [Ping timeout: 246 seconds] 20:54:24 sdemarre [~serge@91.176.38.72] has joined #sbcl 22:14:19 -!- prxq [~mommer@mnhm-5f75f76a.pool.mediaWays.net] has quit [Quit: Leaving] 22:25:35 -!- sdemarre [~serge@91.176.38.72] has quit [Ping timeout: 246 seconds] 22:54:42 daimrod_ [~daimrod@sbrk.org] has joined #sbcl 22:58:05 -!- daimrod [~daimrod@sbrk.org] has quit [Quit: leaving] 22:58:06 -!- daimrod_ is now known as daimrod 23:15:01 LiamH [~healy@pool-74-96-16-203.washdc.east.verizon.net] has joined #sbcl