00:20:36 fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 00:21:07 What magic do I need to do to get ccl running on leopard? 00:21:18 snow leopard 00:21:31 it should work; what's the problem? 00:21:40 I'm port installing it with macports 00:21:47 And getting alot of compile errors 00:22:47 fUD: That seems like a weird thing to do. Just install the dmg. 00:23:22 Or, download a Mac distribution of it. 00:23:24 ports are weird? And why? 00:23:33 Probably you need to install the optional 10.4 support when installing Xcode. 00:24:16 Shit... just found that... 00:24:26 > *** Install Xcode 10.4 support 00:26:23 I sure would like to drop Tiger support... 00:26:35 But I sound like a broken record. 00:27:09 fUD: I just think of macports as a way for managing non-Mac-oriented software on a Mac, and CCL is very Mac-oriented. 00:27:27 rme: Agreed. 00:27:41 (about dropping it, not about you sounding like a broken record) 00:27:45 heh 00:50:19 fmeyer [n=fmeyer@189.4.254.233] has joined #ccl 01:15:02 palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has joined #ccl 01:43:25 -!- fmeyer [n=fmeyer@189.4.254.233] has quit ["Leaving..."] 01:56:52 Adlai` [n=adlai@unaffiliated/adlai] has joined #ccl 02:02:14 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 60 (Operation timed out)] 02:02:32 billstclair [n=billstcl@gw3.tacwap.org] has joined #ccl 02:12:34 -!- Adlai [n=Adlai@unaffiliated/adlai] has quit [Read error: 110 (Connection timed out)] 02:38:18 fmeyer [n=fmeyer@189.4.254.233] has joined #ccl 03:09:00 I spoke to Shannon today and brought up the Tiger issue; I suggested that we could fix critical bugs in 1.4 for a while but not otherwise support Tiger, and he was OK with that. 03:09:23 *gbyers* is in the wrong channel ... 03:14:53 -!- palter [palter@clozure-BB1CFD88.hsd1.ma.comcast.net] has quit [Connection reset by peer] 03:14:54 palter_ [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has joined #ccl 03:14:54 -!- palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 03:14:56 -!- palter_ is now known as palter 03:14:57 -!- palter_ is now known as palter 03:24:48 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Read error: 60 (Operation timed out)] 03:40:01 -!- fmeyer [n=fmeyer@189.4.254.233] has quit ["Leaving..."] 03:40:41 billstclai [n=billstcl@dsl-65-219-213-14.taconic.net] has joined #ccl 03:41:53 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Nick collision from services.] 03:41:57 -!- billstclai is now known as billstclair 03:42:34 -!- billstclair is now known as wws 04:02:31 bfulgham_ [n=brent@adsl-69-234-109-204.dsl.irvnca.pacbell.net] has joined #ccl 04:43:42 lpolzer__ [n=lpolzer@dslb-088-073-197-072.pools.arcor-ip.net] has joined #ccl 04:48:52 fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 04:59:50 -!- lpolzer_ [n=lpolzer@dslb-088-073-220-022.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 05:04:49 Why couldn't Tiger support be an optional compile time flag? 05:19:26 There are Leopard-only features that we (or at least I myself) would like to use. Writing workarounds for Tiger systems that don't have those features is unappealing. And I'd hate to see questions like "are you running ccl compiled for Tiger or for Leopard?" 05:22:54 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Read error: 60 (Operation timed out)] 06:07:17 fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 06:07:46 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Client Quit] 06:08:05 billsb [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 06:29:44 -!- rme [rme@clozure-8353E8B.chi.dsl-w.verizon.net] has quit [Quit: rme] 06:29:45 -!- rme [n=rme@pool-70-104-125-173.chi.dsl-w.verizon.net] has quit [] 06:55:58 -!- billsb is now known as fUD 07:14:40 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Read error: 60 (Operation timed out)] 07:18:32 -!- Modius [n=Modius@cpe-70-123-130-159.austin.res.rr.com] has quit [Read error: 110 (Connection timed out)] 07:43:54 -!- gruseom [n=daniel@S0106001217057777.cg.shawcable.net] has quit [Read error: 60 (Operation timed out)] 08:10:12 -!- bfulgham_ [n=brent@adsl-69-234-109-204.dsl.irvnca.pacbell.net] has quit [] 08:18:24 -!- Adlai` [n=adlai@unaffiliated/adlai] has quit [verne.freenode.net irc.freenode.net] 08:18:26 -!- gbyers [n=gb@c-68-35-15-143.hsd1.nm.comcast.net] has quit [verne.freenode.net irc.freenode.net] 08:18:27 -!- pemryan [n=pem@159.226.35.246] has quit [verne.freenode.net irc.freenode.net] 08:18:27 -!- lisppaste5 [n=lisppast@common-lisp.net] has quit [verne.freenode.net irc.freenode.net] 08:18:27 -!- sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has quit [verne.freenode.net irc.freenode.net] 08:18:27 -!- mdc [n=mdc@flash.etherboot.org] has quit [verne.freenode.net irc.freenode.net] 08:20:43 Adlai` [n=adlai@unaffiliated/adlai] has joined #ccl 08:20:43 gbyers [n=gb@c-68-35-15-143.hsd1.nm.comcast.net] has joined #ccl 08:20:43 pemryan [n=pem@159.226.35.246] has joined #ccl 08:20:43 lisppaste5 [n=lisppast@common-lisp.net] has joined #ccl 08:20:43 sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has joined #ccl 08:20:43 mdc [n=mdc@flash.etherboot.org] has joined #ccl 09:00:56 billstclai [n=billstcl@gw3.tacwap.org] has joined #ccl 09:02:23 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Nick collision from services.] 09:02:28 -!- billstclai is now known as billstclair 09:03:07 -!- billstclair is now known as wws 09:40:58 -!- Adlai` is now known as Adlai 09:50:27 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 54 (Connection reset by peer)] 10:46:08 billstclair [n=billstcl@dsl-65-219-213-14.taconic.net] has joined #ccl 11:23:28 fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 11:29:07 fmeyer [n=fmeyer@189.4.254.233] has joined #ccl 11:47:38 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Read error: 110 (Connection timed out)] 14:29:36 Adlai` [n=adlai@unaffiliated/adlai] has joined #ccl 14:30:02 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Nick collision from services.] 14:30:08 -!- Adlai` is now known as Adlai 16:01:00 milanj [n=milan@77.46.249.106] has joined #ccl 16:48:00 anRch [n=markmill@64.134.65.181] has joined #ccl 17:28:21 gruseom [n=daniel@S0106001217057777.cg.shawcable.net] has joined #ccl 17:46:27 -!- fmeyer [n=fmeyer@189.4.254.233] has quit ["Leaving..."] 17:58:37 -!- anRch [n=markmill@64.134.65.181] has quit [] 18:46:22 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 18:46:48 Adlai [n=Adlai@unaffiliated/adlai] has joined #ccl 18:46:57 -!- gz [Clozure@clozure-93943513.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [Quit: gz] 19:04:31 -!- alms [n=alms@146-115-42-237.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has quit [] 19:10:40 rme [n=rme@pool-70-104-125-173.chi.dsl-w.verizon.net] has joined #ccl 19:19:40 I've looked a bit at the CCL kernel and the SBCL runtime, and one difference I've noticed is that SBCL uses SB-ALIEN to FFI into the runtime, while CCL uses interrupts. What's the reasoning behind that choice? 19:20:15 (I know very little about the details of how interrupts or FFI works) 19:20:47 fe[nl]ix [n=algidus@ABordeaux-158-1-16-102.w90-50.abo.wanadoo.fr] has joined #ccl 19:20:57 hello 19:21:00 hey fe[nl]ix 19:21:18 what brings you to #ccl ? 19:21:57 anyone here familiar with the innards of ffigen ? 19:22:27 gbyers knows it pretty well 19:22:43 I'd like to modify it to export GCC extensions too 19:22:49 and also function bodies 19:26:48 Adlai: in ccl, a foreign call saves the thread's register state and switches stacks. With an interrupt, the state of the thread is left intact. 19:28:00 (there may be other reasons, but that's one that comes to mind) 19:29:26 rme, that makes some sense, although don't the registers still get changed? 19:30:45 It's more likely left over from the days when CCL fit on a 1 meg floppy. 19:30:50 Yes, but thread state is saved by the operating system in a ucontext structure. 19:30:51 Probably smaller to do it that way 19:31:29 Or Gary and Gail just liked it that way 19:31:46 But hopefully, they'll chime in with the real reason 19:32:22 the ucontext is what gets passed around in {x86,ppc}-exceptions.c to access register values, right? 19:33:31 We call the interrupts UUOs, which appears to be terminology from the pdp-10; maybe the idea is from MacLisp? I don't know for sure. 19:34:33 Ununoctium ? 19:34:47 undefined user operation 19:35:40 On the 68K, IIRC, the UUOs were a sequence of unused instructions. It used some of the bits to encode the UUO number. 19:36:00 yes, the kernel calls the ucontext "ExceptionInformation". Macros like xpGPR(xp, Iarg_z) access the saved registers. 19:36:18 right 19:37:22 oh, "unimplemented user operation" -- http://trac.clozure.com/ccl/wiki/Internals/ExceptionHandling 19:39:44 rme, billstclair, thank you for clearing this up 19:40:32 Don't take what you heard as gospel. I didn't work much on that part of the lisp back in the MCL days. garyb and gz likely know it better 19:40:41 One difference between calling into the kernel and trapping into it is that exceptions are serialized (only one thread can actually handle an exception at any time, and there's a locking protocol to ensure that.) Some things that used to call into the kernel were changed when it was found that they needed to be serialized, and the easiest way to ensure that they were. 19:42:26 "that they were ... was to force an exception." 19:43:11 gbyers, does this mean that all threads have to run in the same POSIX thread group? 19:43:34 Since I don't remember what a POSIX thread group is ... no. 19:44:32 I'm wondering how the (os) kernel knows which threads need to be suspended for interrupt handling; I guess the way it tells, on POSIX platforms at least, is by thread groups. 19:45:42 Any thread that can run lisp code is known to the kernel (has a "Thread Context Record" - TCR - in a global list.) 19:48:34 known to the lisp kernel. Other threads run (in general) when a lisp thread gets an exception, and at the OS level they run an exception handler. The first things that that low-level exception handler does is to put its exception context where other threads can see it and wait for a lock. 19:49:56 ah, I see, this locking is done in signal_handler() right before calling handle_exception() 19:50:46 There's a function called (something like) 'prepare_to_wait_for_exception_lock' in there, too. 19:50:50 wait_for_exception_lock_in_handler 19:50:57 right, there's that too. 19:51:54 gbyers: do you have any idea if ffigen can be (easily) made to dump GCC extensions and function bodies too ? 19:51:56 thank you for explaining this, gbyers ! 19:52:17 NP. 19:52:47 fmeyer [n=fmeyer@189.4.254.233] has joined #ccl 19:53:57 fe[n]: it basically just hooks into a few things that process various types of declarations, and outputs lisp-like S-expressions based on GCC's parse-tree representations of those declarations. 20:03:15 -!- gbyers [n=gb@c-68-35-15-143.hsd1.nm.comcast.net] has quit ["Leaving"] 20:03:31 gbyers [n=gb@c-68-35-15-143.hsd1.nm.comcast.net] has joined #ccl 20:05:18 gbyers: I'm trying to deal with __attribute__((__aligned__(x))) in structs and __asm__("foo") and external inlines 20:06:29 currently I have no way to automatically handle the fact that a call to stat() gets inlined as __fxstat64(STAT_VER_LINUX, ...) 20:09:41 -!- fmeyer [n=fmeyer@189.4.254.233] has quit ["Leaving..."] 20:11:36 good old stat 20:14:11 There may be a way to pick up the alignment attribute as well as its effects; some old MacOS PPC code depended on some structs being aligned according to 68K rules, and I think that we used to represent that in the .ffi file explicitly. 20:16:11 As far as #define SOME_C_FRAGMENT SOME_OTHER_C_FRAGMENT goes (or equivalent effects of inlined C functions), that's a much harder problem: parse the C code, understand its effects, automatically generate something lisp-like ... 20:17:21 (er, #DEFINE SOME_IDENTIFIER ARBITRARY_C_FRAGMENT) 20:17:40 er, #define ... need caffiene, however it's spelled. 20:20:23 anRch [n=markmill@64.134.240.111] has joined #ccl 20:37:06 I'm starting to wonder if nameserver errors ought to have their own condition object. 20:37:38 fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has joined #ccl 20:48:48 -!- fUD [n=billsb@124-171-200-55.dyn.iinet.net.au] has quit [Read error: 60 (Operation timed out)] 20:59:41 -!- Adlai [n=Adlai@unaffiliated/adlai] has quit [Remote closed the connection] 21:00:05 Adlai [n=adlai@unaffiliated/adlai] has joined #ccl 21:05:58 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 21:06:54 Adlai [n=adlai@unaffiliated/adlai] has joined #ccl 21:13:47 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 21:14:13 Adlai [n=adlai@unaffiliated/adlai] has joined #ccl 21:19:35 -!- anRch [n=markmill@64.134.240.111] has quit [] 21:28:35 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 21:29:05 Adlai [n=adlai@unaffiliated/adlai] has joined #ccl 21:51:53 -!- palter [palter@clozure-BB1CFD88.hsd1.ma.comcast.net] has quit [Quit: palter] 21:51:53 -!- palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has quit [] 22:27:05 palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has joined #ccl 22:28:03 bfulgham_ [n=brent@adsl-69-234-109-204.dsl.irvnca.pacbell.net] has joined #ccl 22:47:27 alms [n=alms@146-115-42-237.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 23:06:00 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [] 23:06:06 billstclair [n=billstcl@dsl-65-219-213-14.taconic.net] has joined #ccl 23:08:30 -!- _3b` [i=foobar@cpe-70-112-214-100.austin.res.rr.com] has quit [verne.freenode.net irc.freenode.net] 23:08:34 -!- palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has quit [verne.freenode.net irc.freenode.net] 23:09:56 palter [n=palter@c-65-96-32-204.hsd1.ma.comcast.net] has joined #ccl 23:10:38 _3b` [i=foobar@cpe-70-112-214-100.austin.res.rr.com] has joined #ccl 23:37:34 fmeyer [n=fmeyer@189.4.254.233] has joined #ccl 23:41:02 -!- fmeyer [n=fmeyer@189.4.254.233] has quit [Client Quit] 23:48:01 -!- milanj [n=milan@77.46.249.106] has quit ["Leaving"]