October 29, 2007 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
October 29, 2007 [00:00:04] <oco> mmu_man : your deadline is in one week ;-)[00:00:05] <geist> i had looked at the 68k to port newos to early on, but was pretty turned off by it[00:00:08] <mmu_man> 020 needed a 68551 as coprocessor[00:00:15] <mmu_man> 030 had a subset of it[00:00:23] <geist> however, I have a next box might be fun to get running once you get your stuff working[00:00:24] <mmu_man> 040 too but it's very limited[00:00:27] <geist> has an 040, 32MB ram[00:00:32] <mmu_man> and even the opcodes aren't the same[00:00:43] <mmu_man> pflush, ...[00:00:50] <mmu_man> same mnemonic, different opcode :^)[00:00:59] <geist> grody[00:01:00] <mmu_man> yeah that'd rock[00:01:09] <mmu_man> got the specs handy ?[00:01:27] <geist> not really, but it didn't look to hard, based on the netbsd port[00:01:30] *** DaaT has joined #haiku[00:01:30] <geist> I had it booting, etc[00:01:38] <mmu_man> at least I checked the 040 book I had before starting the vm tmap code :)[00:01:43] <geist> the only thing that may cause haiku some problems is the memory is not physically contiguous[00:01:46] <DaaT> mmu_man![00:01:49] <mmu_man> I'm sure 060 has even more limits[00:01:57] <geist> ah, so the thing you're doing is 030 based?[00:02:02] <geist> or 060 based?[00:02:22] <mmu_man> also it's not even supporting CAS2 (double 32bit compare&swap)[00:02:41] <mmu_man> I'll target 030 for now as I have only a falcon emu[00:02:44] <mmu_man> (ARAnyM)[00:02:46] <mmu_man> plop DaaT ![00:02:54] <geist> ah, not real hardware? hrm[00:03:05] <mmu_man> I have a Mac LC III (030 too)[00:03:08] <mmu_man> but no emu for it[00:03:39] <mmu_man> the only emu I know for 68k that emulates teh mmu is ARAnyM[00:03:42] <mmu_man> UAE doesn't[00:03:49] <mmu_man> AmigaOS doesn't use it, so...[00:04:04] <DaaT> mmu_man: you HAVE to watch the film: "a dog's breakfast"[00:04:07] <DaaT> go download it :P[00:04:16] <mmu_man> DaaT: B_WILL_DO :)[00:04:21] <DaaT> :)[00:04:32] *** korli has quit IRC[00:04:51] <mmu_man> I wanted to use 3 level page tree, with a va split like 1 bit, 9 bit, 10 bits (and 4K pages)[00:05:34] <mmu_man> to look almost like x86, but have a single entry forthe kernel space (2 entries pgdir on 1st level)[00:05:40] <geist> yeah[00:05:54] <mmu_man> that would have made things too easy :)[00:06:21] *** NeoOokami has quit IRC[00:07:07] <mmu_man> 040 hardcodes a 3 level tree like 7bits, 7bits, 6 bits[00:07:11] <umccullough> mmu_man, wasn't there basilisk?[00:07:28] <geist> ah it was previously configurable?[00:07:29] <mmu_man> umccullough: that's like 68000 isn't it ?[00:07:48] <umccullough> seems like it emulated 030/040 also[00:07:49] <geist> the other thing I remember was the big pile of crap the cpu spits on interrupt changes between revisions, right?[00:08:09] <mmu_man> yes 030 had a big translation controlm reg where each was settable, up to 6 levels[00:08:12] <umccullough> http://en.wikipedia.org/wiki/Basilisk_II[00:08:17] <mmu_man> and 256 to 32K page size...[00:08:22] <umccullough> ran OS 8.1[00:08:37] <mmu_man> yes there are many int frame types[00:08:48] <mmu_man> umccullough: MacOS doesn't use teh mmu by default AFAIK[00:08:58] <geist> yeah, macos isn't a great stresser of the mmu[00:09:08] <umccullough> ah, so you don't think it supported emulation of that?[00:09:11] *** brycegroff has joined #haiku[00:09:15] <geist> though I think it may have sort of used it around 8 or 9[00:09:23] <geist> for simple protection or something[00:09:43] <mmu_man> Uses UAE 68k emulation or (under AmigaOS and NetBSD/m68k) real 68k processor[00:09:56] <umccullough> says it emulates Mac II series - I'm pretty sure I've run it with a performa ROM[00:10:09] <mmu_man> yes it can use it for virtual mem[00:10:21] <umccullough> might have been a quadra rom even[00:10:46] <umccullough> ah yes, quadra 650[00:10:51] <umccullough> still have the rom here if you need it ;)[00:12:57] <umccullough> i should set that up again - i've got some old mac proggies i'd like to mess with again sometime[00:13:04] <umccullough> and executor sucks[00:13:42] <mmu_man> it had an LC040 (no fpu)[00:13:45] <mmu_man> maybe even no mmu[00:14:08] <mmu_man> maybe aranym can be patched to emulate a mac[00:14:15] <mmu_man> or you can finish 68k in qemu :)[00:14:21] *** flyinghippo has joined #haiku[00:14:22] <mmu_man> only supports coldfire now[00:14:33] <umccullough> yeah, i was wondering about QEMU... but sounds like it's not finished?[00:14:44] *** Darknesss has joined #haiku[00:15:13] <umccullough> coldfire was 060?[00:15:16] <mmu_man> no[00:15:29] <mmu_man> it's a very small subset of 68k for embedded[00:15:32] <umccullough> ok[00:15:40] <mmu_man> with some incompatibilities aven[00:15:51] <umccullough> nice... :P[00:16:04] *** TheNerd_ has joined #haiku[00:16:24] <umccullough> i need to figure out why my LCIII won't post any more :/[00:16:34] <umccullough> i replaced the cmos battery and that didn't help[00:16:38] <mmu_man> anyway, I suppose I should write a vm_translation_map_030.cpp and use function ptrs[00:16:58] <mmu_man> check http://68kmla.net :)[00:20:31] *** TheNerd_ is now known as TheNerd_LT[00:22:02] <umccullough> mmu_man, this you? http://68kmla.net/forums/viewtopic.php?t=1604[00:22:05] <umccullough> :)[00:22:14] <umccullough> must be[00:22:18] <umccullough> "porting haiku"[00:22:35] <umccullough> I think i upgraded my LCIII to 24mb RAM and a 2gb disk[00:23:02] <mmu_man> yep[00:23:50] <mmu_man> I just checked the old SCSI card I have in a shoes box (with 2 big and noisy 10GB hdds) and it seems to have an scsi2 connector, so I can probably plug the LC disk in[00:24:00] <mmu_man> to image it and stuff the port in[00:24:17] <mmu_man> the LC has an ethernet card but I lack the AAUI to RJ adapter[00:24:36] <umccullough> yeah, same here - i bought an adapter to RJ45 for mine[00:24:54] <umccullough> only cost me like $15 or something...let me find out what brand i got[00:25:35] *** Lelldorin1 has joined #haiku[00:26:23] <umccullough> hm... appears to be MIA - i probably threw it in a box of stuff when i was cleaning up[00:30:40] <mmu_man> there are some on ebay at 2 or 3 $ usually (+20 shipping :)[00:32:50] <AlienSoldier> seem transmision 0.91 will be out tonight, too bad beos miss the boat :([00:32:54] * pyCube_ curses regular expressions[00:33:13] <pyCube_> f'ing hate the damn things[00:34:28] <pyCube_> expressing something like "starts with BLAH, ends with WOO, and has any number of chars between as long as DUMB isnt in there" isfrustrating the hell out of me[00:41:04] <DeadYak> ^BLAH.*[^dD][^uU][^mM][^bB].*WOO$ I think[00:42:49] *** nils_t has joined #haiku[00:43:20] *** Lelldorin1 has quit IRC[00:45:49] <pyCube_> hmm[00:46:01] *** MangoFusion has quit IRC[00:46:05] <DeadYak> might be a more elegant way to express that but I *think* it works[00:46:30] <pyCube_> it doesnt seem to be skipping on DUMB[00:46:42] <DeadYak> :/[00:46:53] <pyCube_> it reads like it should.. hehe[00:46:59] <pyCube_> hence my hatred of regex[00:47:24] <DeadYak> might need to use backreferences here[00:47:30] <DeadYak> which I don't know so well[00:48:41] *** Serpentor has quit IRC[00:54:46] * DaaT pets DeadYak[00:56:30] *** Sikosis has joined #haiku[00:56:30] *** ChanServ sets mode: +o Sikosis[00:56:37] * DeadYak pets DaaT[00:57:35] <stargater> n8[00:57:41] *** stargater has quit IRC[00:57:53] <geist> mmu_man: are you ever going to try to run it on a real 68k?[01:00:04] <DeadYak> geist: I'd guess his LC3[01:01:12] *** jan__64 has quit IRC[01:05:54] <umccullough> mmu_man, found my mini-AAUI->RJ45 but it's generic :/[01:05:57] *** AlienSoldier has quit IRC[01:06:22] <umccullough> also conveniently, the previous owner of this LCIII put a VGA convertor on the video port :)[01:08:16] *** AlienSoldier has joined #haiku[01:09:35] <umccullough> hmm... one of the caps on this LC3 board appears to be bulging slightly :([01:09:35] <umccullough> C13[01:09:40] <umccullough> perhaps that's why it doesn't post[01:09:45] <geist> pop[01:10:51] <umccullough> oh yes, i see it leaked all over the board now[01:10:53] <umccullough> lame[01:10:58] <umccullough> okydoke - well that explains it[01:11:04] <geist> putergoboom[01:11:25] <umccullough> looks like a couple of them leaked[01:11:41] <geist> puterhaveaccident[01:12:01] <cps1966> its need pampers[01:14:02] <geist> hah, cute: http://www.1up.com/do/newsStory?cId=3119381&did=1[01:16:00] <mmu_man> umccullough: it's easy to make a Mac to VGA[01:16:08] <mmu_man> check hardwarebook.net[01:16:09] *** oco has quit IRC[01:16:12] <mmu_man> I have one for my ppc mac[01:16:24] <umccullough> mmu_man, yeah i can see - it's just little block that basically just changes the connector[01:16:41] <mmu_man> though it will need one more wire for the LC to make it know it's a color monitor[01:16:54] *** TheNerd_LT has quit IRC[01:17:29] <umccullough> crap, these are surface mount caps[01:17:37] <umccullough> that's gonna suck[01:18:45] *** insomnia has joined #Haiku[01:21:00] *** flyinghippo|AWAY has joined #haiku[01:21:50] *** flyinghippo has quit IRC[01:22:09] <cps1966> 84 bucks to get it fixed[01:24:07] *** NeoOokami has joined #haiku[01:27:36] *** insomnia has quit IRC[01:27:53] *** petterhj- has joined #haiku[01:28:06] *** insomnia has joined #Haiku[01:30:07] <umccullough> i think I paid $35 for the whole deal (including 14" monitor)[01:30:13] <umccullough> so... no thanks :)[01:30:37] *** petterhj- has quit IRC[01:45:37] *** petterhj has quit IRC[01:50:29] *** DaaT has quit IRC[02:02:59] <jjkola> night[02:03:21] *** jjkola has quit IRC[02:15:12] *** cps1966 has quit IRC[02:16:51] *** NeoOokami has left #haiku[02:16:52] *** cps1966 has joined #haiku[02:17:42] *** Darknesss has quit IRC[02:25:52] *** mmu_man has quit IRC[02:38:18] *** brycegroff has quit IRC[02:46:17] <umccullough> ah, buying the caps to fix this LCIII now :)[02:48:21] *** brycegroff has joined #haiku[02:51:21] *** brycegroff has quit IRC[03:32:32] *** nils_t has quit IRC[03:34:33] *** acidx has quit IRC[03:44:06] *** M_199 has joined #haiku[03:46:32] *** flyinghippo|AWAY has quit IRC[03:51:35] *** flyinghippo has joined #haiku[03:59:00] *** M199 has quit IRC[04:18:06] *** DocPheniX has joined #haiku[04:22:12] *** flyinghippo has quit IRC[05:03:29] *** frodobeg has quit IRC[05:05:20] *** kad77 has joined #haiku[05:05:45] <kad77> hey folks[05:26:48] *** frodobeg has joined #haiku[05:44:43] *** Sikosis has quit IRC[06:00:33] *** pyCube_ has quit IRC[06:10:26] *** pyCube has joined #haiku[06:23:01] *** Sikosis has joined #haiku[06:23:01] *** ChanServ sets mode: +o Sikosis[06:52:52] *** SiCuTDeUx has joined #haiku[06:56:34] *** Sikosis has quit IRC[07:14:15] *** DeadYak has quit IRC[07:16:26] *** AndrevS has quit IRC[07:50:01] *** CIA-39 has quit IRC[07:56:56] *** AlexForster has quit IRC[08:02:48] *** AlexForster has joined #haiku[08:26:46] *** kr1stof has joined #haiku[08:43:05] *** Ingenu has joined #Haiku[08:48:58] *** AlexForster has quit IRC[08:50:15] *** CIA-38 has joined #haiku[08:59:02] *** Ingenu_ has joined #Haiku[09:06:00] *** Ingenu has quit IRC[09:25:05] *** kr1stof has quit IRC[09:27:08] *** stargater has joined #haiku[09:27:11] <stargater> moin[09:48:35] <ddew|bofh> gmorning[09:50:34] <stargater> ddew|bofh: thx[09:52:31] <stargater> http://www.youtube.com/watch?v=jMjIW5qyq3s[09:53:20] *** jjkola1 has joined #haiku[09:54:27] *** jjkola1 is now known as jjkola[09:55:17] <jjkola> hi[09:57:34] <ddew|bofh> ho[10:01:55] <stargater> http://www.youtube.com/watch?v=P90NfUw4JFw[10:02:12] <ddew|bofh> another eeepc video?[10:02:57] <geist> oh wow, more laptops![10:04:07] <ddew|bofh> the eeepc is pretty neat, tiny ftw :)[10:10:16] <pyCube> heh[10:11:03] <pyCube> hm[10:12:05] <stargater> dell xps m1330 looks good[10:17:42] <pyCube> ya know.. if you spend a couple evenings f'ing arounf with all the settings and whatnot in kde, you can really get it to be quite nice[10:26:31] *** Ingenu has joined #Haiku[10:42:39] *** Ingenu_ has quit IRC[10:48:57] *** rdmr has joined #haiku[10:51:31] *** Max__ has joined #haiku[10:51:40] <Max__> hi[10:51:50] *** Max__ is now known as Hugen[10:52:21] <stargater> hi Hugen[10:52:32] <Hugen> hi stargater[11:03:14] *** insomnia has quit IRC[11:03:45] *** insomnia has joined #Haiku[11:14:33] *** AndrevS has joined #haiku[11:22:17] *** MrSunshine_ has joined #haiku[11:28:31] *** MrSunshine has quit IRC[11:28:36] *** Darknesss has joined #haiku[11:36:15] *** Lelldorin1 has joined #haiku[11:36:59] *** stargater has quit IRC[11:39:32] <Lelldorin1> moin[11:41:30] *** flyinghippo has joined #haiku[11:46:18] <Hugen> bye[11:46:21] *** Hugen has quit IRC[12:16:51] *** NeonLicht has quit IRC[12:17:25] *** NeonLicht has joined #haiku[12:28:49] *** rdmr has quit IRC[12:40:48] *** SiCuTDeUx_ has joined #haiku[12:42:47] *** SiCuTDeUx has quit IRC[12:52:07] *** mmu_man has joined #haiku[12:56:02] *** brycegroff has joined #haiku[13:03:30] *** Lelldorin1 has quit IRC[13:03:41] *** brycegroff has quit IRC[13:30:59] *** DeadYak has joined #haiku[13:33:57] *** nils_t has joined #haiku[13:38:40] * JonathanThompson lobs an egg roll at DeadYak[13:56:17] <Ed___> AndrevS: congratulations with DSP[13:56:28] <Ed___> you're one of the lucky guys that passed :D[13:56:31] * Ed___ got a 4.4[14:10:06] <DeadYak> JonathanThompson: an egg roll?[14:10:21] <JonathanThompson> It was on my mind for no logical reason.[14:10:30] <JonathanThompson> It's out of my hair now, though, and in yours :)[14:12:25] <AndrevS> Thxies[14:48:43] *** SiCuTDeUx has joined #haiku[14:50:51] *** SiCuTDeUx_ has quit IRC[15:12:08] *** SprMa has joined #haiku[15:15:14] *** WindowsUninstall has joined #haiku[15:27:04] * TuneTracker plays with Guido to see if it still works with BeOS 5[15:27:19] * TuneTracker would get system lockups from it under ZETA[15:27:47] <TuneTracker> woo hoo! It works![15:33:37] *** flyinghippo has quit IRC[15:34:26] <mmu_man> hmm MiNT has an ext2 driver with write support it seems :)[15:34:41] <DeadYak> MiNT?[15:36:03] <mmu_man> MiNT is Not TOS[15:36:17] <DeadYak> which is? :)[15:36:32] <mmu_man> an OS for Atari Falcon[15:37:00] <DeadYak> ahh[15:37:02] <DeadYak> nice[15:37:02] <cps1966> i thought your were refering to mintlinux[15:37:25] *** myrkraver1 has joined #haiku[15:45:59] <MrSunshine_> gaaah why wont cupsd start for me :P[15:46:01] <MrSunshine_> stupid gentoo[15:46:07] <MrSunshine_> something has seriously fucked up[15:46:28] *** myrkraverk has quit IRC[15:46:43] *** SiCuTDeUx_ has joined #haiku[15:48:59] *** SiCuTDeUx has quit IRC[15:54:18] *** SprMa has quit IRC[16:08:31] *** cps1966 has quit IRC[16:09:25] *** AndrevS has quit IRC[16:11:24] *** flyinghippo has joined #haiku[16:12:38] *** myrkraverk has joined #haiku[16:17:30] *** myrkraver1 has quit IRC[16:25:59] *** petterhj has joined #haiku[16:44:05] <TuneTracker> Is there a way to KILL all as a shutdown is happening?[16:46:16] <DeadYak> how do you mean?[16:48:31] <TuneTracker> DeadYak Ok, here's the scenario...[16:49:23] <TuneTracker> I need a shutdown script that tries to gracefully shut down, but if there are zombies that won't close through the normal shutdown, it kills them.[16:50:12] <TuneTracker> DeadYak I have found that sometimes an "alert" window will zombie and cause a shutdown/restart from happening properly[16:50:17] <DeadYak> ok, first problem...how do you determine which ones are zombies? some apps take a bit to shut down[16:50:32] <TuneTracker> alert has been the big problem[16:50:53] <TuneTracker> If it zombies, only CTRL-ALT-SHIFT-Click on it in the deskbar will kill it...and that has to be done manually[16:51:21] <DeadYak> what app's the alert from?[16:51:29] <TuneTracker> DeadYak BeOS[16:51:34] <DeadYak> uhh....?[16:51:35] <TuneTracker> type alert "hi" in a terminal[16:51:38] <DeadYak> oh that[16:51:45] <DeadYak> I mean, what's spawning it?[16:51:52] <DeadYak> some script of yours or what?[16:51:58] <TuneTracker> some script of mine, yes[16:52:07] <TuneTracker> such as a backup script that says "backups completed"[16:52:21] <DeadYak> well, you could just have a kill -9 alert in there[16:52:28] <TuneTracker> doesn't do it[16:52:50] <TuneTracker> kill -9 alert will kill an active alert window but not a zombie[16:53:01] <DeadYak> if -9 doesn't kill it then you have bigger issues[16:53:21] <TuneTracker> It's been a universal problem in BeOS and ZETA[16:53:25] <TuneTracker> Not just a problem here[16:54:58] <DeadYak> not much else I can think of offhand then, that kind of thing used to work for me[16:55:15] <DeadYak> but it's been a while since I've run BeOS in any context other than messing with Haiku builds[16:55:25] <TuneTracker> What I was hoping for was a KILL -KILL ALL or something[16:55:32] <DeadYak> kill -9 == -kill[16:55:45] <DeadYak> there is killall but I'm not sure BeOS has that[16:55:56] <DeadYak> if it does, it'd be killall -9 alert[16:56:13] <TuneTracker> There MUST be something different between kill -9 and CTRL-ALT-SHIFT-CLICK, because one works and one doesn't in this case. Mysterious![16:58:01] <DeadYak> the difference might be kill not being able to find the team corresponding to "alert"[16:58:05] <DeadYak> depending on how it was launched[16:58:16] <DeadYak> with deskbar you're being very explicit about which team you're asking it to kill[16:58:31] <DeadYak> with kill you're asking it "kill whatever the first thing is you can find with this exact name"[16:59:07] <DeadYak> which won't necessarily work if for instance several alerts are currently active[16:59:24] <DeadYak> or if it's launched in some strange way that results in its team name not being alert[16:59:41] <DeadYak> this might be a case where you'd want something smarter than a shell script[17:00:13] <DeadYak> though if you're doing alerts for things like backups being complete, I'd suggest possibly sending out an email as a non-interactive notification instead[17:02:05] *** geist_ has joined #haiku[17:05:19] *** Ingenu has quit IRC[17:10:32] <TuneTracker> DeadYak TYhat's why I was hoping for a KILL All option[17:10:36] <TuneTracker> That's[17:11:57] * TuneTracker envisions a script that runs Kill on every possible program ID, or at least the most likely ones :-)[17:12:22] <DeadYak> if you do that you'll very likely screw up the shutdown process[17:12:42] <TuneTracker> Well that's what I was wondering too...if kill would kill shutdown[17:12:49] <DeadYak> indeed it can[17:12:52] *** JBurton has joined #haiku[17:12:53] *** ChanServ sets mode: +o JBurton[17:12:56] <JBurton> hi all[17:13:00] <TuneTracker> Hi JBurton[17:13:13] *** geist has quit IRC[17:13:24] <JBurton> hmmm let's see if someone knows this...[17:13:55] <JBurton> in kernel land... let's say I have a thread trying to acquire a semaphore with B_INFINITE_TIMEOUT... would the semaphore acquisition fail if the thread receives a SIGINT ?[17:14:08] <DeadYak> JBurton: yeah, that'd drop out with a B_WOULD_BLOCK[17:14:24] <JBurton> hmmm then something isn't working in haiku, or I just wrote wrong code[17:14:27] <JBurton> I have something like this[17:14:28] *** AlexForster has joined #haiku[17:14:28] <DeadYak> or some equivalent error code[17:14:35] <JBurton> a semaphore created with count == 0[17:14:38] <DeadYak> ok[17:14:43] <DeadYak> actually wait[17:14:52] <DeadYak> you might have to pass in a flag to acquire_sem_etc to allow it to be interruptible[17:15:01] <JBurton> one thread tries to acquire with acquire_sem_etc(fSem, 0, B_CAN_INTERRUPT, B_INFINITE_TIMEOUT)[17:15:18] <JBurton> the other thread does a send_signal(thread, SIGINT)[17:15:23] <DeadYak> ok, that's the right flag...[17:15:26] <JBurton> but the thread never unblocks[17:15:28] <DeadYak> that should definitely interrupt acquire_sem then[17:15:45] <JBurton> I see[17:15:55] <DeadYak> actually....[17:16:04] <DeadYak> B_CAN_INTERRUPT is according to the Be Book only applicable from kernel land[17:16:08] <JBurton> ok[17:16:13] <JBurton> I AM in kernel land :P[17:16:16] <DeadYak> ah[17:16:23] <DeadYak> then yeah, something might be wrong[17:16:30] <JBurton> that's because user land acquisitions are always interruptible[17:16:33] <DeadYak> though why're you sending a signal to a kernel thread?[17:16:58] <DeadYak> also I'm not sure a userspace app can send a signal to a kern thread anyways[17:17:09] <JBurton> I send a signal from kernelland to kernelland[17:17:10] <DeadYak> assuming your send_signal is running from userspace[17:17:11] <DeadYak> ah.[17:17:14] <DeadYak> never mind that idea then[17:17:35] <DeadYak> may be worth filing a bug then, that sounds incorrect[17:17:35] <JBurton> and AFAIK, when you send a signal from userladn to kernelland, the kernel thread just wakes up[17:17:43] <JBurton> yeah, although it might well be my code[17:17:53] *** petterhj- has joined #haiku[17:18:00] <JBurton> oh well[17:18:03] <JBurton> worth opening a bug anyway[17:18:08] <JBurton> it'll be closed if invalid[17:19:55] <JBurton> I wonder if there's an example of that kind of design in the kernel[17:20:11] <JBurton> I mean, one thread blocking on a sem, and some other thread waking it up by sending a signal to it[17:20:14] <DeadYak> if you don't mind me asking, what's your code aiming to do?[17:20:25] <DeadYak> usually I'd assume the other thread would wake up the first one by releasing the sem :)[17:20:38] <DeadYak> unless there's a special reason to try and wake it up with an error[17:20:57] <JBurton> well let's say that many threads try to acquire that sem, but only one has to be woken up[17:21:11] <DeadYak> that would be a circumstance where you'd want a condition variable[17:21:18] <JBurton> hmm yeah I thought about that too[17:21:29] <DeadYak> they're designed to do exactly what you're describing[17:21:34] <JBurton> ok thought so[17:21:43] <JBurton> I'll try that approach them[17:21:44] <JBurton> thenn[17:21:45] <JBurton> then[17:21:46] <JBurton> grr[17:21:55] <JBurton> can't tpye corrctly[17:21:57] <DeadYak> bunch of threads = cond->wait(), other thread = cond->signal(), one thread wakes up[17:22:23] <JBurton> oh wait... but can I decide which thread will wake up?[17:22:26] <DeadYak> not sure exactly how that maps to Ingo's Condition variable stuff but that's the general semantics in POSIX anyways[17:22:29] <DeadYak> nope[17:22:36] <JBurton> then it doesnt' apply[17:22:41] <JBurton> I need to wake up that specific thread[17:22:46] <JBurton> I have the thread_id[17:22:59] <DeadYak> then it sounds like you want a different sem for each thread[17:23:16] <JBurton> hmm why wouldn't the design I described work ?[17:23:23] <JBurton> I mean, one sem where everyone blocks[17:23:30] <DeadYak> because if everyone's blocking on it[17:23:31] <JBurton> and sending a signal to the correct thread[17:23:32] <JBurton> ?[17:23:39] <DeadYak> then if you release it, the first thread to try acquiring it will wake up[17:23:48] <DeadYak> in FIFO order[17:23:48] <JBurton> yeah that's why I don't release it[17:23:54] <JBurton> but just send a signal to the thread[17:23:58] <DeadYak> so you basically want some way to put the thread to sleep?[17:24:01] <JBurton> to interrupt the acquisition[17:24:02] <JBurton> yeah[17:24:11] <DeadYak> um....hmm[17:24:14] <JBurton> I guess suspend_thread() would work[17:24:16] <DeadYak> why is it thread-speicifc?[17:24:18] <DeadYak> just curious?[17:24:19] <JBurton> but would be overkill, no ?[17:24:25] <DeadYak> why?[17:24:33] <DeadYak> acquire_sem basically results in a suspend_thread[17:24:36] <DeadYak> if it would block[17:24:38] <JBurton> oh[17:24:43] <JBurton> oh well worth trying then[17:24:47] <DeadYak> either way, net effect: move thread out of run queue[17:24:55] <DeadYak> so yeah, suspend/resume sounds like a better bet here[17:24:57] <JBurton> yeah[17:25:02] <JBurton> thanks[17:25:04] <DeadYak> np[17:28:04] <JBurton> anyway I'm off[17:28:05] <JBurton> bye all[17:28:09] <JBurton> thank you again DeadYak[17:28:20] <DeadYak> np :)[17:28:23] *** JBurton has quit IRC[17:35:30] *** petterhj has quit IRC[17:56:07] *** Ingenu has joined #Haiku[18:00:40] <TuneTracker> mmu_man[18:01:15] *** mats has joined #haiku[18:01:38] *** SiCuTDeUx_ has quit IRC[18:01:45] <TuneTracker> mmu_man http://www.bebits.com/app/4070 <== Does this actually provide an encoder that could be used for streaming? It seems to, but SoundPlay doesn't see it as an option in LiveEncoder window.[18:02:17] *** noph has joined #haiku[18:06:15] *** wildur has joined #Haiku[18:08:56] <mmu_man> TuneTracker: is the SP plugin encoder too or decoder only ?[18:09:25] <TuneTracker> sec[18:10:04] <TuneTracker> mmu_man[18:10:05] <TuneTracker> You can also use SoundPlay's LiveEncoder to create your own AAC streaming[18:10:05] <TuneTracker> station. The easiest way to do that is to copy the file called 'faac AAC[18:10:06] <TuneTracker> encoder' to the 'Plugins/(LiveEncoder)' folder (again, the INSTALL script does[18:10:07] <TuneTracker> this), and then just select the AAC encoder and desired preset from[18:10:07] <TuneTracker> LiveEncoder's presets.[18:10:18] <TuneTracker> oops, just saw it...[18:10:29] <mmu_man> :)[18:10:58] * TuneTracker is gonna test it now... will let you see if you can listen![18:12:04] *** jjkola1 has joined #haiku[18:12:20] *** SiCuTDeUx has joined #haiku[18:15:03] *** jjkola has quit IRC[18:15:21] *** jjkola1 is now known as jjkola[18:17:00] *** tombhadAC has quit IRC[18:24:59] *** dr_evil has joined #haiku[18:26:18] *** TuneTracker has quit IRC[18:32:43] *** ChanServ sets mode: +o dr_evil[18:33:26] <dr_evil> "-NickServ- You cannot GHOST yourself." *wimp*[18:43:40] <SiCuTDeUx> lol[18:44:19] *** tombhadAC has joined #haiku[18:48:11] *** AndrevS has joined #haiku[19:02:08] *** Styrbjorn has quit IRC[19:02:10] *** fanskap has joined #haiku[19:03:06] <mmu_man> shame :)[19:03:17] *** mmu_man has quit IRC[19:04:14] *** jjkola1 has joined #haiku[19:05:05] *** jjkola has quit IRC[19:05:30] *** jjkola1 is now known as jjkola[19:07:02] *** mats has quit IRC[19:11:15] *** Yaroze has joined #haiku[19:17:51] <noph> still no one who got the nForceNIC_Zeta_0_2.zip file?[19:19:56] *** dr_evil has quit IRC[19:20:20] <DeadYak> I have a zeta CD at home somewhere with the nForce driver on it, I don't have that exact zip file though[19:21:35] *** flyinghippo has quit IRC[19:21:48] *** flyinghippo has joined #haiku[19:25:02] *** urnenfeld has joined #haiku[19:26:37] *** alpha-li1n has joined #haiku[19:30:30] <noph> alright[19:30:50] <DeadYak> however that's at home[19:30:54] <noph> i suppose there are no source-code on that cd as well?[19:31:07] <DeadYak> for that driver? no idea offhand[19:31:11] <noph> alright[19:31:20] <DeadYak> I was under the impression it was more or less the same as the forcedeth driver on bebits[19:32:01] <noph> okay, well as it is a GPL driver, the source code might be there somewhere as well[19:32:13] <noph> i would be glad if you could post the driver when you have time[19:32:19] <DeadYak> it used to be on bebits, I'm not sure the download link is still good though[19:32:26] <noph> no, its broken[19:32:35] <DeadYak> well, I can look later[19:33:25] <DeadYak> not sure how easily I'll be able to post it somewhere since the only thing I have installed right now that can read BFS is Haiku[19:35:37] <noph> :)[19:36:20] *** oco has joined #haiku[19:41:46] *** xendor has joined #haiku[19:45:00] *** alpha-lion has quit IRC[19:50:11] *** geist_ is now known as geist[19:50:16] *** ChanServ sets mode: +o geist[19:51:19] *** alpha-li1n is now known as alpha-lion[20:04:23] *** genkie has joined #haiku[20:10:27] *** TuneTracker has joined #Haiku[20:18:30] *** rdmr has joined #haiku[20:19:48] *** rdmr has quit IRC[20:21:40] *** xendor has quit IRC[20:22:38] *** lorglas has joined #haiku[20:23:31] *** lorglas has quit IRC[20:29:37] *** genkie has quit IRC[20:36:37] <pyCube> yep[20:37:51] *** lorglas has joined #haiku[20:38:00] *** lorglas has quit IRC[20:38:05] <MrSunshine_> http://www.youtube.com/watch?v=IAmwy-bDKxs[20:45:14] <noph> :D[20:59:16] <pyCube> hehe[21:15:00] *** urnenfeld has left #haiku[21:20:40] *** PulkoMandy has joined #haiku[21:31:23] *** MangoFusion has joined #haiku[21:34:53] *** DeadYak has quit IRC[21:51:03] *** rdmr has joined #haiku[21:57:36] *** stargater has joined #haiku[21:57:40] <stargater> hi[21:58:21] *** DeadYak has joined #haiku[22:10:49] *** MrSunshine_ has quit IRC[22:11:31] *** MrSunshine has joined #haiku[22:12:24] *** SiCuTDeUx has quit IRC[22:15:17] *** dr_evil has joined #haiku[22:16:39] *** rgb has joined #haiku[22:17:37] *** SiCuTDeUx has joined #haiku[22:18:08] *** Kokito_ has joined #haiku[22:18:34] <Kokito_> howdy[22:21:03] *** myrkraver1 has joined #haiku[22:26:14] *** Ingenu has quit IRC[22:28:55] *** myrkrave2 has joined #haiku[22:32:08] *** Kokito_ has quit IRC[22:35:55] *** myrkraverk has quit IRC[22:36:00] *** flyinghippo has quit IRC[22:36:12] *** flyinghippo has joined #haiku[22:36:21] *** flyinghippo has quit IRC[22:43:19] *** noph has quit IRC[22:44:47] *** myrkraver1 has quit IRC[22:51:58] *** cc^shinobi has joined #haiku[22:54:03] * pyCube sits[22:54:43] <MangoFusion> you forgot to[22:54:46] <MangoFusion> import chair[22:55:21] <MangoFusion> :)[22:55:28] <pyCube> isinstance(pyCube, chair) -> True[22:58:09] *** Yaroze has quit IRC[23:01:16] *** Yaroze has joined #haiku[23:04:29] *** Kokito has joined #haiku[23:10:01] *** oco has quit IRC[23:10:32] *** oco has joined #haiku[23:11:31] <MangoFusion> ah i see[23:17:17] *** mmu_man has joined #haiku[23:17:17] *** ChanServ sets mode: +o mmu_man[23:21:33] *** jjkola has quit IRC[23:26:20] <DeadYak> pyCube: you are a chair?[23:28:39] <MangoFusion> no[23:28:40] <MangoFusion> he's[23:28:43] <MangoFusion> the Chairman[23:28:59] <pyCube> my cats seem to think so[23:29:12] <MangoFusion> *badumdum*[23:38:02] *** cc^shinobi has quit IRC[23:42:07] *** Sikosis has joined #haiku[23:42:07] *** ChanServ sets mode: +o Sikosis[23:45:04] *** axeld has joined #Haiku[23:45:04] *** ChanServ sets mode: +o axeld[23:45:46] <dr_evil> hi axeld[23:45:57] <axeld> Hi Marcus[23:46:20] <mmu_man> plop![23:46:45][23:47:12] <mmu_man> you know what ? 68k is fun, you think it'll be simpler than braindead ppc or x86... btu then you start digging the differences and you find that the mmu from 030, 040 and 060 aren't compatible ;)[23:47:33] <mmu_man> you find a spare register for thread ptr, but then it disapears :D[23:48:14] <axeld> mmu_man: looking at how other systems do it usually helps a lot, though :-)[23:48:43] <mmu_man> yes I cvs coed netbsd and freemint :)[23:49:07] <axeld> mmu_man: and as long there can only be a single CPU, there isn't a big problem in saving the thread pointer in a fixed memory location, either[23:49:14] <mmu_man> I'll just make a cpu_ops table and write the 030 one first[23:49:22] <mmu_man> right[23:49:43] <mmu_man> there actually are SMP 68k systems, but I've yet to see one[23:49:51] <mmu_man> netbsd doesn't upport that even anyway[23:52:17] <mmu_man> oh, they start back at the pilot for Futurama on NRJ12 at midnight[23:53:17] <mmu_man> ... I shouldn't have tried svn cleanup in zeta[23:53:24] <mmu_man> it will take half an hour again :)[23:54:30] <mmu_man> I'll reboot to nux as soon as it's done, way faster :-([23:55:05] <mmu_man> btw, anyone wanting to help with userland, still missing some math stuff in libroot[23:55:51] *** WindowsUninstall has quit IRC[23:56:09] *** rgb has quit IRC[23:58:18] *** ChanServ sets mode: +o dr_evil