[00:00:17] *** mmu_man has quit IRC
[00:00:28] *** mmu_man has joined #haiku
[00:00:28] *** ChanServ sets mode: +o mmu_man
[00:02:00] *** rdmr has quit IRC
[00:02:49] <Ingenu> night
[00:03:14] *** Lt_Henry has joined #haiku
[00:03:38] <Lt_Henry> hi :)
[00:03:40] *** koki_haiku has joined #haiku
[00:04:54] *** luroh has quit IRC
[00:05:01] *** Ingenu has quit IRC
[00:06:03] <mmu_man> re
[00:06:05] <mmu_man> got a freeze
[00:08:38] *** daste has joined #haiku
[00:10:15] *** AlienSoldier has joined #haiku
[00:13:58] *** stargater has quit IRC
[00:16:03] *** Stargater has joined #haiku
[00:16:04] <Stargater> re
[00:16:22] <umccullough> dr_evil, I asked yesterday, but wasn't sure if you saw it: do you still happen to have an R5 net_server version of the ipro1000 driver around somewhere?
[00:16:54] *** urnenfeld has left #haiku
[00:17:49] <dr_evil> yes, should still have a backup somewhere
[00:20:13] <dr_evil> I can have a look tomorrow
[00:20:16] <dr_evil> good night
[00:20:19] *** dr_evil has quit IRC
[00:20:20] <umccullough> thanks - night :)
[00:20:22] <umccullough> doh
[00:20:53] *** daste has quit IRC
[00:21:30] *** burfi has left #haiku
[00:23:38] *** stippi has quit IRC
[00:24:43] *** WindowsUninstall has joined #haiku
[00:26:17] *** burfi has joined #haiku
[00:28:13] *** capycube has joined #haiku
[00:30:09] <burfi> I had to delete and recreate my Haiku partition to force the resize from 100M to 512M. It doesn't happen from simply changing the value in UserBuildConfig
[00:31:32] <geist> you're building directly into it?
[00:31:40] <burfi> yes
[00:31:48] <geist> not surprised
[00:32:08] <geist> guess it doesn't mkbfs it if the target is a partition
[00:34:12] <burfi> when it is already bfs formatted
[00:35:45] <CIA-38> bonefish * r24701 /haiku/trunk/src/system/kernel/fs/pipefs.cpp: (log message trimmed)
[00:35:45] <CIA-38> * Since writers also might have to wait when there is still some room
[00:35:45] <CIA-38>  in the buffer, they need to be notified after a read when their
[00:35:45] <CIA-38>  minimal write size requirement has just become satisfied. We were
[00:35:45] <CIA-38>  notifying only when there was no space in the buffer before, which
[00:35:46] <CIA-38>  caused bug #1755.
[00:35:50] <CIA-38> * Removed Inode::NotifyWriteDone(). It's not needed anymore, since we
[00:38:23] <umccullough> actually, i haven't had any problems building directly to  a partition
[00:38:42] <umccullough> but maybe if it's been dd'd to already that causes an issue
[00:39:09] <umccullough> in fact, i don't even have to specify a size building to a partition - it seems to detect the size automatically and mkbfs the entire thing for me
[00:39:42] <NeonLicht> umccullough: from Linux?
[00:39:47] <umccullough> yeah
[00:40:06] <burfi> not here
[00:40:10] <umccullough> i can set the size to whatever in UserBuildConfig - but it seems to always initialze the entire size
[00:40:16] <umccullough> probably because mine start off as empty
[00:40:43] <burfi> shrug
[00:40:51] <NeonLicht> cool... I think I'll try to build Haiku from Linux myslef instead of trying to install it from R5MaxED (which has driven me nowhere)
[00:41:14] <umccullough> NeonLicht, yeah, building haiku on beos has become more difficult now ;)
[00:41:46] <NeonLicht> umccullough: I haven't tried to build it, just copy the prebuilt image to a partition
[00:42:05] <NeonLicht> umccullough: but it has never booted
[00:42:23] <umccullough> even after running makebootable on it?
[00:42:46] <NeonLicht> yes, even then
[00:45:52] <EuanK> what partition you trying to copy to?  are your partitions out of order?
[00:45:53] *** Kancept has quit IRC
[00:46:06] *** Kancept has joined #haiku
[00:46:19] <umccullough> i think there's possibly a problem with "out of order" partitions :/
[00:46:31] <umccullough> seems like i have problems whenever i try to do that
[00:47:02] *** mmu_man has quit IRC
[00:47:13] <NeonLicht> EuanK: me?  I have R5MaxED on partition 2, data (Haiku images and others) on partition 3, and I've tried to copy Haiku images into partitions 1 and 4; all 4 partitions are 25% of a 4GB HDD
[00:47:37] <NeonLicht> umccullough, EuanK: what do you mean by "out of order" partitions?
[00:47:54] <EuanK> sda1 sda4 sda 2 sda 3
[00:48:03] <EuanK> gparted seems to do that.
[00:48:08] <umccullough> NeonLicht, if i used gparted to resize and "insert" a partition - the new partition in the middle has a higher number than the one following it
[00:48:09] <NeonLicht> EuanK: mine are hda1, 2, 3, 4
[00:48:10] <EuanK> I can't get my laptop to boot
[00:48:26] <EuanK> I'm trying to fix it to see if that's the issue.
[00:48:27] <umccullough> and grub also seems to honor those "out of order" numbers
[00:48:44] <umccullough> i don't understand if it's something funky in the partition table, or what
[00:49:16] <NeonLicht> I started out with an empty HDD and let R5MaxED Live to partition the HDD (4 25% partitions), and then let install R5MaxED on partition 2
[00:49:27] <umccullough> sounds reasonable then :)
[00:49:32] <umccullough> NeonLicht, how big is the disk?
[00:49:52] <NeonLicht> umccullough: 4GB, as I've just said
[00:50:03] <umccullough> oh jeez, that's tiny :D
[00:50:31] <EuanK> compact flash?
[00:50:45] <NeonLicht> well, it shpuld be enough to try a couple of different Haiku images on two of them, shouldn't it?
[00:51:18] <EuanK> do you get the beos boot menu? what symptoms?
[00:51:41] <NeonLicht> anyways, I think I'll put in a lager (8 or 10GB) HDD next time to try to build from Linux
[00:52:03] *** WindowsUninstall has quit IRC
[00:52:29] <NeonLicht> yes, EuanK, I get the R5MaxED boot menu, which lets me either boot R5MaxED or Haiku... the first worsk fine
[00:52:52] <EuanK> hmmm does the R5 bootloader work with haiku???
[00:53:26] <umccullough> NeonLicht, 1gb is plenty for testing haiku - yes
[00:53:29] <NeonLicht> when I choose Haiku, I see the splash screen, and I get the Haiku boot menu, but I can't do anything else than rebooting, since the menu tells that it hasen't found any bootable thing
[00:53:33] <umccullough> EuanK, yes, bootman works fine with haiku
[00:54:17] <umccullough> NeonLicht, i assume a serial log and info has been provided in a trac ticket already?
[00:54:21] *** slaad has joined #haiku
[00:54:36] <NeonLicht> (regardless of whether or not I make partitions 1 and 4 bootable with makebootable=
[00:55:05] <umccullough> makebootable really doesn't ... make it bootable
[00:55:10] <NeonLicht> umccullough: no idea, do you mean if I provided that?
[00:55:37] <umccullough> what you're running makebootable for is to change the offset in the first block or two of the partition that tells haiku's kernel where the partition physically starts
[00:55:52] <NeonLicht> if so, I don't know what a trac ticket means :-)
[00:56:03] <umccullough> NeonLicht, http://dev.haiku-os.org !
[00:56:25] <EuanK> is the build broken?
[00:56:31] <umccullough> is it?
[00:56:41] <EuanK> i get a error on bluetooth shared library
[00:56:48] <EuanK> svn st shows no changes
[00:56:54] <NeonLicht> ok, umccullough, I'll have a look and try to see if I can get such serial log
[00:57:35] <burfi> bluetooth related commits happened very recently
[00:57:37] <EuanK> ah  r24697
[00:57:40] <EuanK> yes 22.42
[00:57:45] <EuanK> bah!
[00:58:11] <umccullough> looks like oliver left about  40 mins ago :(
[00:58:39] <EuanK> i'll just remove bluetooth from the build...
[01:00:13] <EuanK> yeah the sharedlibrary line is duplicated
[01:07:33] *** EuanK has left #haiku
[01:15:26] *** capycube has quit IRC
[01:20:50] *** wildur_ has quit IRC
[01:22:42] *** oco has quit IRC
[01:27:50] *** Kokito has quit IRC
[01:32:55] *** burfi has left #haiku
[01:32:59] *** hakimrie has joined #haiku
[01:35:38] *** Kancept has quit IRC
[01:45:00] *** Kancept has joined #haiku
[01:46:12] *** capycube has joined #haiku
[01:49:11] *** Animal-X has quit IRC
[01:55:01] *** capycube has quit IRC
[02:01:05] *** Zenja has joined #haiku
[02:01:42] <Zenja> Morning/evening.  geist, are you on-line by any chance?
[02:09:47] *** clsk has joined #haiku
[02:10:49] <ddew|bofh> o/
[02:11:23] <ddew|bofh> lots of coffee and cat stevens on the stereo. an awesome hack.night :)
[02:11:28] <ddew|bofh> *hack-night
[02:12:44] <stpere1> cool
[02:13:08] <ddew|bofh> scary, running bootman in VirtualBox picked up one of my hard-drives :O
[02:14:04] *** EuanK has joined #haiku
[02:14:09] <ddew|bofh> i'll stay away from that app :)
[02:14:24] <EuanK> lol the pres is getting booed on tele opening a new ball park.
[02:14:31] <ddew|bofh> he
[02:14:32] <EuanK> (US pres that is)
[02:14:41] <EuanK> suppose it is 50/50 over there anyway.
[02:14:50] <EuanK> expected I guess hehe
[02:15:09] <mmadia> mmm... coffee sounds good.
[02:15:32] <EuanK> at 1:15am that wouldn't be a good idea for me.
[02:15:48] <ddew|bofh> heh, it's 2:15 here :)
[02:15:53] *** hakimrie has quit IRC
[02:17:39] <mmadia> 8:25 but no coffee in the pot : (
[02:17:46] <mmadia> *pm
[02:20:05] *** Tonik has quit IRC
[02:20:46] <stpere1> mmadia: same timezone here :-)
[02:20:50] <EuanK> i'm looking forward to my coffee in 6 hours time. :)
[02:31:28] *** Lt_Henry has quit IRC
[02:34:52] *** StarWarsLegacy has joined #haiku
[02:36:35] *** hakimrie has joined #haiku
[02:36:54] *** MindChld2 is now known as MindCHild
[02:36:59] *** MindCHild is now known as MindChild
[02:37:10] *** Kancept has quit IRC
[02:38:48] *** Zenja has quit IRC
[02:45:06] *** Kancept has joined #haiku
[02:51:45] *** pyCube has joined #haiku
[03:01:18] <CIA-38> anevilyak * r24702 /haiku/trunk/src/kits/bluetooth/Jamfile: Clean up build break - library declaration line was doubled.
[03:02:24] *** mmadia has quit IRC
[03:21:08] *** EuanK has left #haiku
[03:25:55] *** MikeW has quit IRC
[03:28:18] <stpere1> ok, I flush the vCard part of my project :)
[03:28:30] <stpere1> there is already People and DeeperPeople to do that part
[03:28:41] <stpere1> I will focus on calendar
[03:28:46] <stpere1> one thing, but one thing right
[03:31:53] <Schmedly3D> cool beans, BeIDE and gcc actually work well enough in Haiku to compile my game
[03:34:56] <DeadYak> nice :)
[03:35:17] *** Penix has quit IRC
[03:35:44] <Schmedly3D> when did the 32 vs 16 bit issue for nvidia cards crop up any clues?
[03:35:51] <DeadYak> 32 vs 16 bit issue?
[03:36:13] <stpere1> 32 suffered a performance hit?
[03:36:26] <Schmedly3D> in 32bit, large sections of the display are black
[03:36:35] <Schmedly3D> but dropping back to 16 it's normal
[03:36:38] <stpere1> ah in that way
[03:36:48] <DeadYak> wouldn't know, I don't have an nv chipset here
[03:37:22] <Schmedly3D> I'll see if there's already a ticket
[03:37:50] <Schmedly3D> otherwise.. that's just cool to have a working environment, even if it might not be super stable
[03:38:38] <DeadYak> Schmedly3D: 1950 and 1959 sound right?
[03:38:44] <stpere1> :)
[03:39:23] <DeadYak> ticket #s that is
[03:39:31] <DeadYak> 1950 in particular though
[03:40:37] <Schmedly3D> yes 1950 is it
[03:40:43] <DeadYak> looks like it's MTRR-related then
[03:41:18] <Schmedly3D> completely workable though
[03:42:48] <Schmedly3D> even my mouse and keyboard (usb) work through the kvm
[03:43:01] <Schmedly3D> although the \ key is dead for whatever reason
[03:43:05] <DeadYak> keymap
[03:43:08] <DeadYak> most likely
[03:43:10] <umccullough> yeah
[03:43:17] <umccullough> i was just gonna say that ;)
[03:43:18] <DeadYak> it defaults to US-International, which renders ~ non-functional for me too
[03:43:27] <umccullough> DeadYak, you have to tap it twice :(
[03:43:37] <DeadYak> umccullough: switched to the American keymap and it works fine
[03:43:42] <umccullough> yeah, i know
[03:43:47] <umccullough> i'm usually too lazy
[03:43:52] <DeadYak> I have that set up in my settings.zip that I unzip to the image
[03:43:55] <umccullough> guess i could change my build
[03:43:56] <DeadYak> so never have to worry about it any more
[03:44:05] <umccullough> i think it's even easy to add something to the UserBuildConfig to do it
[03:44:10] <DeadYak> probably
[03:44:21] <DeadYak> you could also edit HaikuImage to alter the default it sets :)
[03:44:26] <umccullough> nah
[03:44:40] <umccullough> i used to do that, but it gets tedious every time someone changes it
[03:45:05] <umccullough> that's one thing about the UserBuildCOnfig - I don't think you can *remove* something that's already in the HaikuImage using a rule
[03:45:06] <DeadYak> want my keymap settings file?
[03:45:09] <umccullough> nah
[03:45:23] <DeadYak> well, in this case you can't because you can't overwrite a symlink that easily
[03:45:35] <Schmedly3D> doesn't seem to fix it by switching
[03:45:43] <umccullough> Schmedly3D, sounds crappy :)
[03:45:46] <DeadYak> laptop or desktop?
[03:45:51] <umccullough> usb keyboard
[03:45:57] <DeadYak> oh
[03:45:58] <DeadYak> never mind
[03:45:59] <umccullough> through a kvm no less
[03:46:02] <Schmedly3D> tried a few, but no biggie because I've got a ps2 keyboard on it too
[03:46:11] <umccullough> and that one wrosk?
[03:46:13] <umccullough> works?
[03:46:26] <Schmedly3D> still, for the keyboard to work at all.. through the kvm no less AND survive switches.. hell yeah!
[03:46:29] <umccullough> strange...must be a USB thing
[03:46:55] <umccullough> well, USB *is* a hot-swap architecture ;)
[03:47:06] <umccullough> my PS2 kvm plays tricks on Haiku
[03:47:13] <umccullough> with the mouse anyway
[03:47:25] <Schmedly3D> indeed, even though through a KVM, usb functionality is simply emulated ps2
[03:47:30] *** Androidz has quit IRC
[03:47:32] <stpere1> it has to be plugged in at boot time I suppose
[03:47:37] <Schmedly3D> unless it's hooked to the accessory port
[03:48:02] <DeadYak> I've had all kinds of strange behavior with KVMs :?
[03:48:03] <Schmedly3D> these aren't complaints at all though
[03:48:04] <DeadYak> er :/
[03:48:13] <DeadYak> especially depending on what keys they take over for switching shortcuts and stuff
[03:48:18] <umccullough> heh
[03:48:29] <DeadYak> for instance, on my linux box, hooking my mouse through a KVM breaks the middle scroll wheel'
[03:48:34] <umccullough> mine you gotta push ctrl-alt-shift, let go, press a number, and then enter
[03:49:00] <umccullough> that's why you always see me typing numbers into iRC ;)
[03:49:04] <DeadYak> ahh
[03:49:06] <Schmedly3D> yes weirdness abounds, when for whatever reason a reboot kills any and all keyboard input until you actually unplug the kvm
[03:49:37] <umccullough> for me - every other boot into Haiku causes the mouse to go wonky
[03:49:49] <umccullough> but only when it's connected through the KVM
[03:49:56] <umccullough> same mouse attached directly works fine every time
[03:50:14] <umccullough> and seriously, every*other*boot
[03:50:17] <umccullough> pisses me off
[03:50:28] *** AnEvilYak has joined #haiku
[03:51:31] <Schmedly3D> is there a 1gb+ kernel patch for R5?
[03:52:00] *** hakimrie has quit IRC
[03:52:11] <Schmedly3D> actually memory fails, does R5 just ignore the extra memory or does it cause problems
[03:52:23] <AnEvilYak> causes problems, there's a patch that limits it to recognizing 768MB
[03:52:26] <Schmedly3D> heh memory ... memory
[03:54:40] *** mmadia has joined #haiku
[03:54:45] <Schmedly3D> AnEvilYak: is it on bebits or elsewhere?
[03:55:10] <DeadYak> http://www.bebits.com/app/3851 I believe
[03:55:41] <Schmedly3D> ah yes Euan's name I remember
[03:55:59] <mmadia> additional bootloaders are at http://bootdisks.beuser.de
[03:56:32] <DeadYak> mmadia: thanks! :)
[03:56:48] <Schmedly3D> that's awesome
[03:58:05] <mmadia> it wasn't much.  euan + zadig did the initial work,    scanty helped my modify the zbeos's to do 256, 768, and 1024(for net_server)
[03:58:37] <mmadia> s/my/me
[04:03:14] <umccullough> btw, beos max v4b1 is pre-patched
[04:05:25] <umccullough> could be wrong about that :/
[04:05:52] <stpere1> night!
[04:06:06] <umccullough> maybe it just comes with boot disk images that are pre-patched
[04:06:29] *** stpere1 has left #haiku
[04:18:10] *** Al2O3 has quit IRC
[04:22:21] <umccullough> i'm gonna pull out one of my atari 2600's and hook it up to the 70" TV :)
[04:22:43] <DeadYak> ow.
[04:22:55] <DeadYak> my SNES hooked to my 50" TV is bad enough :/
[04:23:45] <umccullough> i hooked the NES up the other day
[04:23:57] <umccullough> used the AV cable to ensure "high quality" ;)
[04:24:43] <Schmedly3D> nice
[04:24:46] <umccullough> guess i'll wait till the TV's not in use :P kids are watching a movie
[04:25:01] <pyCube> hmmm.. now i feel like playing excite bike
[04:25:13] <umccullough> there's a game i haven't played in a LONG time...
[04:25:20] <umccullough> i don't even own it now :(
[04:25:25] <Schmedly3D> I got a boxed motorodeo for christmas
[04:25:44] <Schmedly3D> well, I bought it for myself :P
[04:27:18] <umccullough> does a NES RF switch work with a 2600? i can't remember if i've ever tried
[04:28:24] <umccullough> ah, need an RCA-RCA barrel connector :P
[04:29:25] <DeadYak> don't know
[04:29:29] <DeadYak> the SNES I have hooked up via A/V
[04:29:30] <Schmedly3D> I'd eventually like to end up with everything hooked up permanently but I don't have enough shelf space for everything
[04:30:15] <DeadYak> I have way too many consoles to hook up everything :/
[04:30:20] <DeadYak> both due to inputs and space
[04:30:45] <DeadYak> though speaking of the 2600, I kept randomly having the urge to play Choplifter the other day for some reaosn
[04:30:46] <DeadYak> reason*
[04:30:52] *** koki_haiku has quit IRC
[04:31:12] *** koki_haiku has joined #haiku
[04:31:47] <Schmedly3D> I've got a Gyruss cabinet in the garage, but I don't play it much
[04:31:57] <DeadYak> wb koki :)
[04:45:22] <pyCube> i have been playing choplifter a lot lately
[04:45:32] <pyCube> for some weird reason
[04:45:41] <pyCube> c64 version, via vice
[04:46:50] <DeadYak> pyCube: that's actually what I had it on too
[04:46:57] <DeadYak> but it was a 2600 title originally I thought
[04:47:40] <StarWarsLegacy> Choplifter was ported to the 2600?
[04:48:20] <DeadYak> I might be remembering wrong :)
[04:48:25] <DeadYak> I thought it was 2600 originally
[04:49:24] <DeadYak> I only played it on my c128 myself like I said :)
[04:49:51] *** koki_haiku has quit IRC
[04:50:02] <pyCube> great game
[04:50:34] <DeadYak> indeed
[04:50:37] <DeadYak> so simple, yet so addictive
[04:50:39] *** koki_haiku has joined #haiku
[04:51:46] <StarWarsLegacy> The first I ever saw of it was in a Sega Arcade Cabinet, then I played it on my Sega Master system, but beyond that I'm not sure what other ports there were, though I do believe it had at least 2 sequels
[04:52:22] <DeadYak> that game was Sega? I thought it was..Activision
[04:52:26] <DeadYak> I might be smoking crack though
[04:52:29] * DeadYak googles
[04:52:32] <StarWarsLegacy> :)
[04:52:41] <Schmedly3D> Sega did their port after the original
[04:52:44] <DeadYak> ah
[04:52:47] <DeadYak> Sega released the arcade version
[04:52:52] <DeadYak> looks like Broderbund originally
[04:52:54] <pyCube> cyeah
[04:52:55] <Schmedly3D> it started out life on the Apple II
[04:52:55] <DeadYak> so I was wrong either way
[04:53:10] <StarWarsLegacy> Ya, just saw that, who would have known! :)
[04:53:19] *** absabs has joined #haiku
[04:53:30] <Schmedly3D> great game no matter where it was played though
[04:53:36] <DeadYak> :)
[04:53:41] <StarWarsLegacy> For sure! Frustrating at times too! :)
[04:53:42] <DeadYak> absabs: finally applied I see :)
[04:53:52] <DeadYak> for frustrating nothing beats Super Ghouls and Ghosts imo :)
[04:53:53] <absabs> Hi DeakYak
[04:53:57] *** Stargater has quit IRC
[04:54:14] <absabs> I will update it tonight
[04:54:19] <StarWarsLegacy> That's another series that I Love, but also hate! :)
[04:54:22] <pyCube> or anything psygnosis ever made
[04:54:23] <absabs> some fix
[04:54:33] <DeadYak> absabs: ok :)
[04:54:38] <absabs> :P
[04:54:41] <StarWarsLegacy> I think I beat the one on the NES, Arcade, Genesis, and SNES, but have not gotten my hands on the new PSP one yet
[04:54:56] <DeadYak> ah
[04:55:03] <DeadYak> I beat it on the SNES...that was rough though
[04:55:50] <StarWarsLegacy>  Oh ya, it was also on the Master System, but I never beat that port.
[04:56:27] <DeadYak> was that Ghosts and Goblins or Ghouls and Ghosts?
[04:57:02] <DeadYak> Maximo and Maximo vs Army of Zin on the PS2 were fun as well
[04:57:07] <DeadYak> not quite as hard as the originals though :)
[04:57:16] *** DHowett has joined #haiku
[04:57:42] <StarWarsLegacy> Ghost and Goblins was only on the NES, Gameboy, and arcade as far as I know, all other ports were variations of Ghouls & Ghosts
[04:57:46] <DeadYak> oh
[04:57:56] <DeadYak> I didn't think Ghouls and Ghosts was on anything other than Genesis and SNES
[04:57:57] *** mmadia has quit IRC
[04:58:10] <StarWarsLegacy> well wait, there was the C64 and other old compturs that had ports of them too
[04:58:22] <DeadYak> yeah, I had Ghosts and Goblins on my c4
[04:58:23] <DeadYak> c64*
[04:58:54] <pyCube> troll and tribulations
[04:58:56] <pyCube> trolls
[04:59:35] <umccullough> disappointment - the antenna in on my TV doesn't appear to work
[04:59:38] <umccullough> i'll have to look into that
[05:02:56] <MindChild> Ghouls and Ghosts was an NES game
[05:02:57] <MindChild> not SNES
[05:03:05] <DeadYak> uhh
[05:03:12] <MindChild> SNES had super Ghouls and Ghosts
[05:03:13] <DeadYak> my Super Ghouls and Ghosts SNES cartridge says otherwise.
[05:03:16] <MindChild> completely different game
[05:03:31] <MindChild> I am SORRY that you cannot read
[05:03:52] <DeadYak> do you really have to be a complete jackass about it?
[05:04:25] <MindChild> says the person who tried to infer I was a dumbass because he didn't understand the situation
[05:04:43] <DeadYak> .....
[05:04:44] <DeadYak> whatever.
[05:04:49] <MindChild> booo hoo
[05:04:54] <DeadYak> I'm sorry I finished typing that sentence right as yours came across the network
[05:05:02] <DeadYak> whatever.
[05:05:17] <pyCube> how do you try to infer?
[05:05:18] <DeadYak> doesn't negate the fact that you're being a complete ass about it
[05:05:25] <MindChild> have your pubes started growing in yet?
[05:06:07] <DeadYak> .....
[05:06:34] <umccullough> jeez...
[05:07:57] <geist> wow
[05:08:18] <slaad> See what you people did? You went and attracted cock monkeys to the channel. With your development of a new OS.
[05:08:27] <slaad> None of this would have happened if you all stuck to using Linux.
[05:08:35] * DeadYak pets /silence
[05:08:36] <geist> clearly
[05:08:40] <DeadYak> problem solved either way
[05:08:49] * slaad wedgies Yak.
[05:09:20] <DeadYak> slaad: what's new in the land where toilets flush counterclockwise? :P
[05:09:27] <geist> badger badger badger badger
[05:09:32] <DeadYak> mushroom mushroom
[05:09:43] <slaad> Lasers. Freakin' lasers.
[05:09:53] <slaad> Not much, taking today off because I'd stab someone if I went to work.
[05:09:58] <DeadYak> damn you for getting weebls in my head :)
[05:10:38] * [Katisu] wonders if slaad works as a hitman
[05:10:48] <slaad> SE, but close enough :)
[05:11:32] <MindChild> You do know that the whole Toilet-flushes-the-other-way in the opposite hemisphere is bullshit, right?
[05:11:37] <MindChild> Apparently not
[05:11:41] <MindChild> two for two
[05:11:44] <slaad> Quiet little man.
[05:11:47] <DeadYak> someone evidently hasn't seen the simpsons
[05:12:00] * geist passes on the headworm
[05:12:08] <MindChild> YEAH THE SIMPSONS IS THE BEST RESOURCE EVAR!
[05:12:13] <DeadYak> it was a joke dipshit.
[05:12:18] <MindChild> my bad
[05:12:21] <MindChild> seriously
[05:12:28] <DeadYak> grow up dude
[05:12:29] <DeadYak> or just leave
[05:12:31] <MindChild> I mistook the atmosphere
[05:12:36] <[Katisu]> slaad, SE?
[05:12:41] <slaad> Software Engineer.
[05:12:42] <DeadYak> = Software Engineer
[05:13:03] <geist> well, the primary reasons toilets flush in either direction is the aim of the nozzles
[05:13:16] *** DHowett has left #haiku
[05:13:21] <geist> and i would assume to keep it simple most toilets in the world point in one direction
[05:13:32] <MindChild> which they do
[05:13:43] <slaad> Toilets don't even flush in a direction here.
[05:13:48] * umccullough sometimes tries to make it go the opposite direction with his own nozzle
[05:13:54] <slaad> It's more *woosh, woosh, woosh*. And then no more poop.
[05:14:02] <geist> in the southerm hemisphere stuff comes out of the toilet?
[05:14:09] <umccullough> my toilet usually flushes down...i'm not sure i'd like if it flushed the other way :(
[05:14:26] <DeadYak> haha
[05:14:32] <geist> umccullough: yeah, trying to fight it with your own nozzle is pretty tough
[05:14:43] <umccullough> geist, but i still try!
[05:14:56] <geist> nothing beats peeing on chickens
[05:15:03] <geist> that was may favorite pastime as a kid
[05:15:06] <umccullough> electric fences?
[05:15:08] <geist> when we had chickens everywhere
[05:15:09] <slaad> W... t... f?
[05:15:16] <geist> they love it. they'll get in the stream and play around
[05:15:22] <umccullough> lol
[05:15:33] <umccullough> just imagine them clucking "golden showers!"
[05:15:38] <geist> yeah
[05:15:43] <slaad> My girlfriend apparently knows this too, geist, and thinks it's hilarious.
[05:15:49] <umccullough> i'll have to try that sometime geist
[05:15:50] <slaad> I am totally without knowledge of this.
[05:15:54] <slaad> How have I lived 25 years without knowing this?
[05:16:00] <slaad> I'd have spent far more time peeing on chickens!
[05:16:13] <umccullough> maybe i'll have my wife and daughters watch too for the entertainment factor
[05:16:28] <DeadYak> ....
[05:16:31] <DeadYak> I'm scared now
[05:16:48] <umccullough> funny - i can actually see chickens in my backyard from here - wonder if I could open the window and...
[05:16:56] <pyCube> i always kinda liked peeing out of really tall trees
[05:17:00] <umccullough> just barely enough light :)
[05:17:04] <DeadYak> umccullough: you live on/near a farm?
[05:17:11] <umccullough> DeadYak, might as well :/
[05:17:17] <geist> yes, men are blessed with the ability to aim. USE IT
[05:17:47] <MindChild> just because you aim forward doesnt mean it goes forward
[05:17:48] <umccullough> standers unite!
[05:17:50] <slaad> I think most wimmenfolk would disagree, geist ;)
[05:18:06] <geist> they're just jealous
[05:18:14] <umccullough> lowly squatters
[05:18:20] <slaad> Travelmate! For all your jealousy needs.
[05:21:41] *** Kancept has quit IRC
[05:42:51] *** koki_haiku has quit IRC
[05:43:15] *** AlexForster has quit IRC
[05:44:19] *** bhy has joined #haiku
[05:45:31] *** rcjsuen has quit IRC
[05:48:08] *** Stargater has joined #haiku
[05:48:11] <Stargater> moin
[05:51:57] *** absabs has left #haiku
[05:52:08] *** absabs has joined #haiku
[05:52:16] *** absabs has left #haiku
[05:53:54] *** MindChild has quit IRC
[05:53:57] *** MindChild has joined #haiku
[05:53:59] *** pyCube_ has joined #haiku
[05:57:12] <Stargater> http://www.haikuware.com/poll/
[05:57:51] <geist> select
[06:01:57] *** dchis has quit IRC
[06:02:54] <umccullough> jeez, if the bezilla team wins they'll get $50 to split what - 4 ways?
[06:03:14] <DeadYak> wins what?
[06:03:22] <umccullough> the haikuware thank you award
[06:03:25] <DeadYak> oh
[06:03:31] <DeadYak> the April 1st thing made wonder if the poll was a joke
[06:03:38] <umccullough> nah
[06:05:42] *** dchis_ has joined #haiku
[06:05:53] *** dchis_ is now known as dchis
[06:08:27] *** cherry` has joined #haiku
[06:09:39] *** pyCube has quit IRC
[06:16:02] *** Kokito has joined #haiku
[06:16:49] <Stargater> haiku-os.org needs a april joke
[06:17:02] <umccullough> how bout..no
[06:17:14] <Stargater> :-)
[06:17:15] *** cherrypie has quit IRC
[06:17:35] <Kokito> Stargater: something like "Haiku 2.0 is released" ?
[06:17:57] <AnEvilYak> Haiku switches to X11
[06:18:16] <Kokito> or to the Linux kernel!
[06:18:30] <AnEvilYak> or better yet, HURD
[06:19:34] <Kokito> HURD seems interesting
[06:19:43] <Stargater> Mac OS 11 is based on Haiku
[06:19:51] <Kokito> LOL
[06:20:22] <Kokito> MacOS Cheetah to be based on Haiku. How about that? :P
[06:20:40] <ozy`> they already had a cheetah
[06:20:41] <ozy`> I think
[06:20:45] <AnEvilYak> wasn't Cheetah one of the previous code names?
[06:20:48] <Kokito> did they?
[06:20:49] <ozy`> wasn't 10.1 cheetah?
[06:20:50] <ozy`> yeah
[06:20:53] <ozy`> or 10.0
[06:21:04] <Kokito> oh, then make the another animal
[06:21:05] <AnEvilYak> I couldn't remember 10.1's... I only remember Jaguar, Panther, Tiger, Leopard
[06:21:16] <ozy`> 10.0 was cheetah, 10.1 was puma
[06:21:17] <AnEvilYak> ah
[06:21:29] <Kokito> Cat?
[06:21:30] <AnEvilYak> wonder what they'll choose next...Lion?
[06:21:31] <ozy`> it was only starting with 10.2 that they made a big deal of the code names though
[06:21:33] <umccullough> running out of cat names...
[06:21:36] <ozy`> yeah
[06:21:43] <AnEvilYak> lol
[06:21:43] <ozy`> they could move onto bears
[06:21:44] <AnEvilYak> I can just see it now
[06:21:47] <AnEvilYak> 10.6 Tabby
[06:21:48] <ozy`> that would be pretty epic
[06:21:49] <ozy`> haha
[06:21:53] <umccullough> Siamese
[06:22:01] <umccullough> if you please
[06:22:07] <AnEvilYak> though Grizzly would have an interesting ring to it
[06:22:09] <Stargater> STeve Jobs says " the future of Operation Systems is Haiku (BeOS), now 13J after beos release for Mac, now i understand the power and yes i dont like Mac OS X , it is not powerfull enoght. .."
[06:22:12] <Kokito> umccullough: what is Luposian complaining about on the forums?
[06:22:23] <AnEvilYak> Kokito: not being able to build Haiku from R5 since the compiler switch
[06:22:39] <AnEvilYak> he sent me around 10 private messages on OSN asking what to do as well, despite me telling him repeatedly that I had no idea since I don't build on R5
[06:23:17] <umccullough> yeah
[06:23:22] <umccullough> he emailed me privately also :P
[06:23:27] <Kokito> AnEvilYak: I suppose there will be an R5 cross compiler at some point in time?
[06:23:37] <umccullough> Kokito, you have to build it yourself currently
[06:23:53] <umccullough> just like on linux
[06:24:05] <umccullough> although, mlotz built one
[06:24:08] <Kokito> aha, ok,
[06:24:20] <umccullough> i was gonna test it, but haven't
[06:24:29] <AnEvilYak> hey umccullough
[06:24:30] <Kokito> but it would be possible to post a binary for people with R5 to use
[06:24:33] <AnEvilYak> you've tried building Haiku from Haiku yes?
[06:24:44] <umccullough> AnEvilYak, not in a couple days
[06:24:47] <Stargater> Kokito http://www.freelists.org/archives/openbeos/03-2008/msg00144.html
[06:24:47] <AnEvilYak> Kokito: that's what mlotz tried to do, but he only had a BONE sys to build/test on so he has no way of being sure it works on R5
[06:24:58] *** AndrevS has quit IRC
[06:25:07] <Kokito> got it AnEvilYak
[06:25:30] <Kokito> maybe we can post the BONE version as a start?
[06:25:39] <umccullough> he did somewhere...
[06:25:43] <AnEvilYak> he did
[06:26:41] <umccullough> somewhere in teh svn mailing list
[06:26:43] <Kokito> how about we link to that from the Downloads page?
[06:26:55] <umccullough> looking...
[06:27:10] <AnEvilYak> Kokito: don't know if anyone yet confirmed that it works ok...
[06:27:21] <umccullough> the problem with Luposian is that he's going to immediately ask for "concise instructions on how to use it"...
[06:27:23] <AnEvilYak> until then I probably wouldn't post a link to it
[06:27:32] <umccullough> at least, that's the verbiage he used in an email to me
[06:27:41] <AnEvilYak> indeed, that's what he said repeatedly to me as well
[06:27:56] <Kokito> AnEvilYak: I will ask on the mailing list by replying to mlotz post :)
[06:28:02] <AnEvilYak> okey :)
[06:28:20] <Kokito> hmmm... found the message, but the link there is dead
[06:29:11] <Stargater>  http://haiku.mlotz.ch/haiku_cross_compiler_bone_2.95.3-haiku-080323.zip
[06:29:39] <umccullough> this one? https://lists.berlios.de/pipermail/haiku-commits/2008-March/012278.html
[06:30:09] <Stargater> umccullough this is going
[06:30:22] <umccullough> Stargater, yeah, looks like the same link in that message
[06:30:33] <Stargater> :-)
[06:32:20] <Kokito> got it
[06:33:57] *** helf_ has joined #haiku
[06:34:03] *** DeadYak has quit IRC
[06:34:33] *** helf has quit IRC
[06:35:39] <Kokito> should we not remove the "GCC compiler 2.95.3 (for BeOS 5.0 or later)" linked from the Development page?
[06:35:56] <AnEvilYak> probably since that one's no longer usable
[06:36:23] <Kokito> OK
[06:36:28] <Stargater> the bebits gcc will not more work to compile haiku from beos /zeta
[06:36:53] <umccullough> Kokito, we probably should remove it
[06:37:19] <umccullough> and possibly we need an updated jam there also
[06:37:48] <Stargater> optinal packages list is nice too
[06:38:38] <Kokito> umccullough: where to get the updated jam from?
[06:38:58] <umccullough> Kokito, someone usually compiles a new one and posts it - often it's sikosis
[06:38:59] <AnEvilYak> up-to-date jam's already on the dev page
[06:39:05] <umccullough> oh is it?
[06:39:12] <umccullough> from the last couple days?
[06:39:26] <AnEvilYak> jam changed how?
[06:39:42] <umccullough> it changed, not sure how
[06:39:51] <umccullough> perhaps it's a minor thing
[06:40:04] <AnEvilYak> don't recall seeing a change to it apart from maybe some haiku-specific ifdef
[06:40:08] <umccullough> i know it reports a newer date int eh version string
[06:40:36] <AnEvilYak> yeah, that was just for a haiku-native jam change
[06:40:37] <umccullough> mine is 20080327
[06:40:40] <umccullough> ok
[06:40:46] <AnEvilYak> shouldn't affect jam for BeOS
[06:49:01] <geist> what you talkin bout
[06:49:33] * pyCube_ thinks willis
[06:58:26] <skoe> OpenOffice is too bloated x(
[06:58:48] <umccullough> such is life :)
[07:00:18] *** StarWarsLegacy has quit IRC
[07:01:07] <skoe> There needs to be a well designed, cleanly written and (ONLY) C++ office suite written for Haiku :-).
[07:01:27] <skoe> Maybe even each App should be separately developed...
[07:08:16] *** Kancept has joined #haiku
[07:13:01] <Kokito> is the haiku-commits list down?
[07:13:36] <Kokito> I am getting a  failed connection when trying to open https://lists.berlios.de/mailman/listinfo/haiku-commits
[07:15:39] <Kokito> ahhh... never mind
[07:16:26] *** Kernel86 has quit IRC
[07:16:36] *** Kernel86 has joined #Haiku
[07:26:34] *** ynop_ has joined #haiku
[07:30:10] <Thom_Holwerda> boo
[07:30:35] <Kokito> ola Thom_Holwerda
[07:30:49] <Thom_Holwerda> morning
[07:32:55] <Thom_Holwerda> http://blogs.sun.com/richb/entry/you_know_you_re_tired
[07:33:00] <Thom_Holwerda> heh :)
[07:34:41] * Kokito recalls doing that more than once :)
[07:34:58] <Thom_Holwerda> same here
[07:35:19] <pyCube_> i really dont get kde4
[07:39:29] *** MindChild has quit IRC
[07:39:36] *** MindChild has joined #haiku
[07:47:41] *** darbelo has joined #haiku
[07:57:03] *** skoe has quit IRC
[07:59:46] *** Kancept has quit IRC
[08:02:49] *** Kancept has joined #haiku
[08:10:08] *** Kernel86 has quit IRC
[08:10:08] *** clsk has quit IRC
[08:10:08] *** slaad has quit IRC
[08:10:08] *** Vampyre has quit IRC
[08:10:10] *** crash|_ has quit IRC
[08:10:10] *** serpentor has quit IRC
[08:10:12] *** adamblokus has quit IRC
[08:10:12] *** Zaranthos has quit IRC
[08:10:37] *** Stargater has quit IRC
[08:13:39] *** Kernel86 has joined #haiku
[08:13:39] *** clsk has joined #haiku
[08:13:39] *** slaad has joined #haiku
[08:13:39] *** Vampyre has joined #haiku
[08:13:39] *** crash|_ has joined #haiku
[08:13:39] *** serpentor has joined #haiku
[08:13:39] *** adamblokus has joined #haiku
[08:13:39] *** Zaranthos has joined #haiku
[08:18:15] *** Wiss_ has joined #haiku
[08:18:18] <Wiss_> hi all !
[08:18:24] <ddew|bofh> hiya
[08:18:56] <Wiss_> I'm student who search a GSoC Application... I know, it's late, but not too late and I really hope find a good adventure with this high opportunity
[08:19:13] <ddew|bofh> best of luck :)
[08:19:21] <ddew|bofh> what's your task?
[08:19:28] <ddew|bofh> or suggestion
[08:19:44] *** absabs has joined #haiku
[08:20:21] <Wiss_> I read quickly about haiku projects, and I have the skills for "Implement XSI semaphore and IPC API" or "Expand our automated unit tests" ... Do you know if there are student which have submit on these projects ? or other free projects ?
[08:20:47] <Wiss_> thanks for your welcome, ddew|bofh :)
[08:21:18] <ddew|bofh> i'm not a Haiku dev, just a devote supporter who likes to try out things in Haiku :)
[08:21:48] <Wiss_> oh nice :D
[08:22:01] <ddew|bofh> but for gsoc i'd say that helping out with stuff to move towards an alpha release is welcome
[08:23:28] <ddew|bofh> ipc support is a good one imo
[08:24:16] <Wiss_> ok
[08:24:48] <ddew|bofh> have you played around anything with Haiku so far?
[08:25:10] <Wiss_> Do you know what are the developpers who know these informations that I can ping  ?
[08:25:43] <Wiss_> no I have to recognize that I never heard about Haiku before.. thanks Google to permit me this discover :)
[08:25:44] <umccullough> Wiss_, you should send email to the development mailing list
[08:25:46] <ddew|bofh> dropping a note to the development mailing list is the best way
[08:26:02] <absabs> I think so
[08:26:14] <umccullough> list of mailing lists are here: http://www.haiku-os.org/community/ml
[08:26:28] <umccullough> second one down is the dev list
[08:26:33] <Wiss_> umccullough : Thanks for advise, but I think it's too late, I can't wait a response which probably come too late, the GSoC deadline is today :S
[08:26:50] <umccullough> Wiss_, today in which timezone? ;)
[08:27:06] <ddew|bofh> heh
[08:27:11] <Wiss_> 13h UTC I think
[08:27:20] <umccullough> there are no developers in this channel - you are likely to get several on the dev list immediately since many are european
[08:27:23] <Wiss_> so 9h in montreal
[08:27:32] <umccullough> well, there's no devs awake in this channel anyway ;)
[08:27:39] <Wiss_> ok lol
[08:27:42] <ddew|bofh> umccullough: occasionaly devs drop by though ;)
[08:27:47] <umccullough> yes
[08:27:52] <umccullough> just not at the moment :P
[08:28:05] <ddew|bofh> they're getting some well-deserved sleep
[08:28:10] <Wiss_> I have an exam in 5 hours lol... xD
[08:28:16] <Wiss_> poor me, that's dead
[08:28:21] <ddew|bofh> the past few datys have seen a torrent of neato commits
[08:28:26] <ddew|bofh> *days
[08:29:28] <absabs> Wiss_, there is about 17 hours left, so you can try
[08:29:44] <Wiss_> 17 hours ? :o
[08:29:57] <Wiss_> ok thanks, I write now :)
[08:31:45] <absabs> Wiss, #gsoc says
[08:31:47] <absabs> (14时31分16秒) socinfo: "extension" is There may be a deadline extension for student applications, to make sure that orgs have enough great applications to work with. But it is not certain at this time, *SO DON'T COUNT ON IT!*
[08:32:11] <absabs> hope it helps
[08:33:01] <Wiss_> ok I didn't heard about this news :)
[08:33:26] <absabs> but just as it says, DON'T COUNT ON IT!
[08:33:38] <Wiss_> I don't find the exact mail address... it's haiku-development@xxxxxxxxx  what is after @  ?
[08:34:03] <Wiss_> Yes ok, I don't count on it and wait about a haiku developer answer :)
[08:34:41] <absabs> Let me see
[08:35:32] <absabs> haiku-development at freelists dot org
[08:37:02] <Wiss_> thanks for all absabs :)
[08:37:19] <absabs> It's my pleasure
[08:37:29] <absabs> :)
[08:37:47] <umccullough> Wiss_, you can't send mail to that list until you register on freelists.org
[08:38:23] <umccullough> subscribe i mean
[08:38:48] <absabs> Wiss_, why not subscribe it
[08:39:01] <Wiss_> I just subscribe and had answer, don't worry :P
[08:39:13] <absabs> :)
[08:40:06] <ddew|bofh> uhhm, has the python devs decided that building python requries python? :O
[08:40:34] *** emitrax has joined #haiku
[08:40:42] <absabs> cpython?
[08:41:26] <pyCube_> rpython
[08:41:40] <ddew|bofh> nvm :)
[08:42:43] <ddew|bofh> one of these days i'll learn to read documentation :P
[08:44:07] *** axeld has joined #Haiku
[08:44:07] *** ChanServ sets mode: +o axeld
[08:44:42] <umccullough> hello axeld :)
[08:44:52] <ddew|bofh> gmorning axel
[08:45:19] <umccullough> almost morning here even
[08:45:20] <axeld> Hi umccullough!
[08:45:34] <axeld> It definitely is here :-)
[08:45:39] <umccullough> yep
[08:46:12] *** rdmr has joined #haiku
[08:46:34] <Wiss_> mail send : thanks again absabs and umccullough for your help
[08:46:41] <umccullough> Wiss_, you still want to talk to someone about GSoC?
[08:46:46] <Wiss_> yes
[08:47:00] <umccullough> maybe axeld has time :)
[08:47:11] <Wiss_> oh ok
[08:47:14] <Wiss_> hi axeld :)
[08:47:35] <umccullough> and i see your email found the mailing list as well :)
[08:47:46] <Wiss_> :)
[08:47:46] <axeld> I never have any time, I just hang around here because I was told to
[08:48:17] <Wiss_> ok :/  I'm just searching about a GSoC project
[08:49:50] <emitrax> mornign all
[08:50:10] <emitrax> Wiss_: did you apply for gsoc ?
[08:50:25] <emitrax> has the deadline been postpone by google ?
[08:50:26] <Wiss_> yes
[08:50:34] <emitrax> Wiss_: what projects for ?
[08:50:57] <axeld> Wiss_: I was just kidding
[08:51:04] <axeld> Hi emitrax
[08:51:08] <Wiss_> I applied for Adium, BlueZ, MacPorts, OpenMRS and Vim projects...
[08:51:08] <umccullough> i was wondering... :D
[08:51:25] <axeld> Wiss_: you're Adrien?
[08:51:44] <Wiss_> yes :)
[08:51:54] <Wiss_> Sorry I didn't precise my nickname on irc
[08:52:08] <emitrax> hi axeld! is that _you_ or an alias? :)
[08:52:39] <emitrax> Wiss_: did you realize this is #Haiku right? :)
[08:52:45] <axeld> emitrax: it's my nick if you mean that :-)
[08:52:57] <axeld> Wiss_: Just saw your mail
[08:53:02] <Wiss_> emitrax : yes. Why ?
[08:53:12] <emitrax> Wiss_: how about bluetooth for Haiku?
[08:53:40] *** darbelo has quit IRC
[08:54:35] <emitrax> axeld: did you by any chance start looking into the device manager and/or our I/O architecture (btw is it documented somewhere? ) ?
[08:54:53] <Wiss_> emitrax : It could be great :) I don't have good skills on bluetooth, but the first meaning in GSoC is to learn :)
[08:55:01] <axeld> emitrax: not really, I didn't find much time yet
[08:55:33] *** Kancept has quit IRC
[08:55:50] <emitrax> Wiss_: I agree with that, but it requires lots of time and efforts, I suggested bluetooth because you mentioned BlueZ and I thought it was bluetooth related and you had some skills already ;)
[08:55:57] <axeld> Wiss_: so what kind of experiences do you have already? Have you worked on anything in particular that would sound good in your application? If so, put it in.
[08:56:22] <Wiss_> emitrax : ok lol.. I applied for BlueZ to learn this technology ;)
[08:56:50] <Wiss_> axeld : I worked most on the web development
[08:58:54] <Wiss_> I did the Quality control interface for Clarisys (french informatic company which work for medical analisys laboratory ... I used Python, Perl, PHP, Ajax, Templates, SVG
[08:59:04] <emitrax> Wiss_: just out of curiosity, are you interested in kernel development or user space applications ?
[08:59:27] <Wiss_> emitrax : You don't imagine in how high interests I am for this
[08:59:43] <axeld> Wiss_: I've just answered your mail
[08:59:50] <Wiss_> but it's a new thing for me, I just search the opportunity to learn more about
[09:00:02] <Wiss_> axeld : thanks for your quick answer :)
[09:00:52] <geist> hey it's axeld
[09:01:01] <geist> don't see him around here that much
[09:01:03] <geist> guess he hax
[09:01:31] <Thom_Holwerda> who is this axeld of whom we speaketh?
[09:01:38] <geist> axeld: hey, what are the thoughts on finally enabling the swap path
[09:01:47] <geist> now that there are real block devices, it might be time to start really doing it
[09:02:22] <geist> will be interesting, the way pages are committed seems to have changed a bunch since newos, so i dont really recognize that low level code that much anymore
[09:02:31] <geist> hopefully the new changes haven't painted it into a corner
[09:02:51] <axeld> Morning geist :-)
[09:03:04] <geist> evening axeld
[09:03:17] <axeld> geist: yeah, it's probably about time to do so
[09:03:36] <emitrax> nice topics to discuss when just woke up :)
[09:03:47] <Wiss_> emitrax : Do you think I have chances to be accepted on a kernel development or user space applications without skill on it for the moment ? If yes, I'll submit now for these project..
[09:04:02] <axeld> There are still some bugs to sort out, though, that have a higher ranking in my TODO list :-)
[09:04:08] <geist> yah
[09:04:09] <Wiss_> "on a kernel development project", sorry
[09:04:30] <geist> axeld: will be interesting to see if the block device paths are swap-safe
[09:04:48] <geist> that always scared the crap outta me, and i think it's generally solved in gnarly and nasty ways
[09:05:11] <geist> you guys did hack a new pageout daemon, right?
[09:05:17] <emitrax> Wiss_: I suggest you to read axeld email and follow his suggestions. If you are very motivated, interested and have time then just apply :)
[09:05:19] <geist> ie, write mapped mmap()s actually write pages out
[09:05:31] <axeld> Wiss_: it always depends on the quality of other applications, too - it's always better to be already qualified, but we can all learn
[09:05:51] <Wiss_> yes ok its totally true
[09:06:08] <Wiss_> Thanks for all your answer !
[09:06:19] <axeld> geist: yes, we have a new page daemon. Our I/O path might be swap-safe, but we'll see about that ;-)
[09:06:33] <emitrax> Wiss_: you can apply for different Haiku related projects
[09:06:34] <axeld> geist: we still have some dead-lock problems under high memory pressure
[09:06:54] <geist> yeah
[09:07:03] <geist> that's where a lot of black magic comes in, it seems
[09:07:14] <geist> lots of reserved pools of memory, watermarks, etc
[09:07:31] <axeld> I don't like that part either
[09:07:48] <axeld> There doesn't seem to be a clean solution that is guaranteed to work
[09:07:55] <axeld> (and simple)
[09:09:15] <geist> nope
[09:10:10] <axeld> Our slab allocator has similar problems itself, too :-)
[09:11:01] <geist> yeah, that's why i was a little concverned with that
[09:11:07] <geist> see, a lot of the stuff that
[09:11:22] <geist> that's 'missing' on newos is stuff that i didn't have a good solution for so i didn't solve it
[09:11:38] <geist> or was slated to be written, but didn't want any half solutions
[09:11:50] <geist> i preferred to leave it missing than have a noncomplete solution
[09:11:58] <geist> i have the feeling the slab allocator is a halfsolution
[09:12:04] <axeld> it currently is, yes
[09:12:13] <geist> and i'm really not sure that's a good idea
[09:12:24] <geist> though you obviously serve a different master
[09:12:31] <axeld> Definitely not, that's why almost no one is using it yet
[09:12:32] <geist> people actually want to use it, etc
[09:12:47] <geist> ah good, I thought you had moved it all over
[09:12:58] <geist> the kernel allocator was something i really shouldn have gotten to earlier in newos
[09:13:12] <axeld> I haven't written the slab allocator mostly because I didn't have a good solution for that yet
[09:13:15] <geist> it's a nasty problem to solve, but really should be solved early on because it effects the design of things that come later
[09:13:23] <axeld> exactly
[09:13:41] <axeld> I tried to move everything to the slab, but I realized it's not ready for that
[09:13:56] <geist> i dont think there's a good solution other than exposing more hints to the caller about what it's for
[09:14:08] <geist> pageable, nonpageable, need to reserve this, this is for the swap path, etc
[09:14:13] <geist> thats what most systems seem to do
[09:14:22] <geist> which of course effects *everyone* that alloates memory from it
[09:14:32] <axeld> and that's definitely helpful, too.
[09:14:49] <axeld> The heap implementation is the only problematic one, as that is truly generic
[09:15:17] <geist> right, though it's nongeneric in that the memory is wired, so there's no change of it getting swapped out
[09:15:18] <axeld> But as long as swap/VM doesn't use that, it should work
[09:15:29] <geist> yeah, beos actually had two heaps. one was used by the vm
[09:15:40] <axeld> but you might need to swap out stuff in order to enlarge the heap :-)
[09:15:41] <geist> the main heap grew on demand
[09:15:48] <geist> the vm heap was smaller and fixed
[09:16:07] <axeld> I am not yet sure which path to take
[09:16:49] <geist> what i was thinking of doing in newos is making the heap expandable, and requiring the caller make some additional hints as to what it was for
[09:17:02] <geist> so it can dip into reserve pools to avoid expansion in cases of extreme duress
[09:17:23] <geist> ie, if the thread is part of the swap path it'd use a different arg
[09:17:41] <geist> of course the actual expansion of the heap I hadn't thought of
[09:17:51] <axeld> it should only use that kind of memory to free other memory, though
[09:17:52] <geist> but that's mostly moot nowadays, since everythigns going 64bit
[09:18:02] <geist> where you have essentially an infinite supply of kernel address space
[09:18:11] <axeld> Haiku isn't there yet ;-)
[09:18:31] <geist> which actually brings up another problem, that beos would have slammed headfirst into
[09:18:35] <geist> and haiku may be as well
[09:18:43] <geist> >4GB ram for device drivers with 32bit DMA
[09:19:12] <geist> the beos/haiku strategy of not really abstractin the bus away so that the drivers themselves have to deal with it is a Really Bad Idea
[09:19:56] <geist> it's unfortunately a real problem that everyone has to deal with now, even if you're not x86-64
[09:20:37] <geist> only system i've seen that dealt with it really elegantly is OSX. the iokit already completely abstracted away all notions of buffers and physical memory handles behind a pile of c++ classes
[09:20:48] <geist> no driver ever gets anywhere close to a real hard pointer
[09:21:46] <axeld> Yeah, I didn't look into how solving that best for Haiku yet
[09:21:47] <geist> the only temporary solution is to do what 32bit windows does and refuse to use any RAM > 4GB physical
[09:22:02] <geist> why my 4GB machine only sees 3.5GB on windows :)
[09:22:08] <axeld> Same for Haiku :-)
[09:22:27] <axeld> And I'm afraid we won't change that for R1 anymore
[09:22:43] <geist> yeah, trouble is changing it involves heavily modifying every driver
[09:22:51] <geist> so the farther down the road, the more painful it gets
[09:23:18] <axeld> True, but we have to deliver something sometime, and that is already really painful now
[09:23:21] <geist> MS never abstracted that away for the drivers in windows either, which is why they have so many problems
[09:23:53] <axeld> So I better have a look at Apple's I/O Kit before thinking more about that :)
[09:24:19] <geist> might be worth looking at, if anything just to see the extreme solution
[09:24:35] <geist> i'd look at BSD and linux too, though a lot of their solutions may be 'recompile it'
[09:25:00] <geist> i suspect that linux just requires that the drivers themselves do some bounce buffer management, and they just provide an api to help them doing it
[09:25:07] <axeld> BSD has specific functions to reserve DMA memory
[09:25:25] <geist> linux does too, but the trouble happens if you get a buffer directly from userspace that happens to have a page > 4GB
[09:25:29] <axeld> so in the end, the driver has to take care of that, too, I guess
[09:25:32] <geist> you gotta bounce through some sort of buffer
[09:26:01] <geist> but linux definitely has to have multiple queues of pages
[09:26:14] <axeld> Linux already needs something like that for the memory it hasn't mapped directly yet, anyway, or is that a thing of the past?
[09:26:17] <geist> grody grody stff. PC hardware sucks
[09:26:53] <AlienSoldier> ir suck beyond belief
[09:26:55] <AlienSoldier> *it
[09:27:00] <geist> on 32bit linux I think it has like 4 queues: < 1MB, 'low memory' (< 896MB, directly mapped), high memory, and stuff above 4GB
[09:27:16] <geist> on x86-64 the low memory goes away, but it still has < 4GB and >= 4GB to think about
[09:27:44] <axeld> For the PCI bus
[09:27:49] <geist> right
[09:27:59] <axeld> It's a pity
[09:28:11] <geist> yep, it's the ISA dma prblem, all over again, 20 years later
[09:28:18] <geist> thankfully most newer stuff is 64bit capabile
[09:28:29] <geist> it's not even a pci problem, it's a card implementation problem
[09:28:43] <geist> pci has supported 64bit addresses for a while, a lot of cards just ignore the top 64bits
[09:28:50] <geist> top 32bits of 64bits that is
[09:29:15] <geist> ie, all those dumbass rtl8139s out there. the TX buffer pointers are only 32bit wide
[09:29:45] <axeld> A lot of cards do that, though
[09:29:48] <geist> yep
[09:30:10] <geist> though nics are probably the simplest to deal with: make sure the net stack allocates all it's buffers out of the <4GB pool
[09:30:23] <geist> you never have a user space pointer from the wild to deal with
[09:30:37] <axeld> true
[09:30:56] <axeld> At least if you don't try zero-copying :-)
[09:31:02] <Thom_Holwerda> i never realsed the 32/64 bit issue was that deeply rooted into computing
[09:31:05] <geist> though of course doing that means the VM needs to have the concept of different typs of physical pages int he first case
[09:31:05] <Kokito> axeld! good to see you!
[09:31:12] <axeld> Hi Kokito
[09:31:24] <geist> Thom_Holwerda: yep. that's why MS has so many problems with 32/64 transition
[09:31:29] <Thom_Holwerda> i knew it got deep, but that even pci cards were affected in such a way
[09:31:31] <geist> one of them at least
[09:31:52] <umccullough> geist, the problem started with PAE, though, didn't it?
[09:31:59] <umccullough> or is that different?
[09:32:09] <geist> yep. 32bit windows XP wont 'see' memory > 4GB, even though the cpu can just fine
[09:32:24] <geist> and in fact the shipping version did, but MSFT quietly disabled it in SP1
[09:32:31] <umccullough> lol
[09:32:33] <geist> because they were having too many problems with it
[09:32:37] <umccullough> cuz it busted drivers and stuff?
[09:32:40] <geist> drivers out there that didn't deal with large pointers
[09:32:42] <geist> right
[09:33:04] <geist> that's why my machine only sees 3.5GB of the 4GB here
[09:33:23] <umccullough> yeah, same here - although i updated to xp64 not that long ago
[09:33:23] <geist> the top 512MB is mapped at 4GB to 4.5 to leave a hole for pci addresses
[09:33:37] <geist> yeah, with xp64 they changed the driver api to deal with it better
[09:33:50] <geist> or at least required that all new drivers be updated to pass their WHCL
[09:34:24] <geist> i'm sure it's part of the whcl tests now, it's just all the 10 year old 32bit drivers for 32bit versions of their os they have to worry about
[09:34:40] <geist> plus, it gives them a reason to push users to 64bit windows, because there's little reason otherwise
[09:41:15] *** Barrett666 has joined #haiku
[09:41:54] *** slaad has quit IRC
[09:45:42] <AlienSoldier> so, as it is, kernel is no longuer lacking development time, but stumbling block start to appear?
[09:54:37] *** pyCube has joined #haiku
[09:54:43] *** Stargater has joined #haiku
[09:54:44] <Kokito> good night folks
[09:54:48] <Stargater> re
[09:55:02] <Stargater> hello axeld :-)
[09:55:05] *** bhy has quit IRC
[09:55:13] *** Kokito has quit IRC
[09:56:28] *** jiuda_D`arkness has joined #haiku
[09:57:08] *** stef has quit IRC
[09:57:09] *** rennj has quit IRC
[10:00:05] *** stef has joined #haiku
[10:04:49] *** Wiss has joined #haiku
[10:05:47] *** Barrett666 has quit IRC
[10:09:20] *** rennj has joined #haiku
[10:11:11] *** pyCube_ has quit IRC
[10:13:19] *** W0rmDrink has joined #haiku
[10:13:22] <Wiss> axeld, umccullough : I have submitted two applications, but like you said axeld, it wont sound that convincing :(  I don't have skills so I can't give detailed plan, I just speak about motivation, true but.. not convincing :/
[10:13:56] *** WormDrink has quit IRC
[10:14:24] <Thom_Holwerda> Wiss: motivation can do wonders.
[10:15:04] <Thom_Holwerda> an unskilled person with a lot of motivation can be of much greater value an a skilled person lacking motivation
[10:15:17] <Thom_Holwerda> than*
[10:16:24] <axeld> Wiss: we'll see, and thanks for your applications!
[10:18:29] <AlienSoldier> the best is always someone with lot of motivation torturing a skillfull one
[10:18:38] *** rennj has quit IRC
[10:18:56] <Wiss> ok thanks you too :)
[10:19:04] *** Wiss_ has quit IRC
[10:19:14] *** WindowsUninstall has joined #haiku
[10:19:14] *** rennj has joined #haiku
[10:21:45] *** Kancept has joined #haiku
[10:22:13] <umccullough> hm? oh... ok - i'm half asleep and soon will be fully - 'night :)
[10:22:24] <ddew|bofh> gnight
[10:26:27] *** tic has quit IRC
[10:27:29] *** Wiss_ has joined #haiku
[10:27:39] *** tic has joined #haiku
[10:27:43] *** _WindowsUninstal has joined #haiku
[10:32:57] *** SprMa has joined #haiku
[10:35:04] *** Wiss has quit IRC
[10:37:02] *** petterhj has joined #haiku
[10:38:16] *** hUMUNGUs has joined #haiku
[10:39:00] *** WindowsUninstall has quit IRC
[10:41:46] *** adrien__ has joined #haiku
[10:47:50] <adrien__> axeld : I have submit an application for  "Samba Port" project... Could you look it quickly ? I gave some ideas, it's better than just motivation (I hope :P )
[10:48:55] *** Wiss_ has quit IRC
[10:49:35] *** Wiss has joined #haiku
[10:50:00] <Wiss> axeld : hi again, sorry my connexion is bugged
[10:50:55] <axeld> Wiss: BTW "kernel userland" stuff requires quite a bit of experience
[10:50:58] <Wiss> axeld : I submitted another application about "Port SAMBA" application... I have more skills about it, so could you read this quickly and give me advise if I have more details to write ? thanks :)
[10:51:43] <Wiss> axeld : Yes I supposed it :/   But we always have to begin one day :P
[10:53:13] <Stargater> reboot
[10:53:17] <axeld> Wiss: SAMBA is essentially the server part, not the client (the file system), though. What we would actually need is both parts, though. IIRC the file system is usually called CIFS. I don't know which part you would prefer to work on
[10:54:49] *** absabs has quit IRC
[10:54:49] *** Stargater has quit IRC
[10:54:55] *** Kancept has quit IRC
[10:55:12] <axeld> Wiss: and the application seems to be plain text, and not HTML enabled
[10:55:14] <Wiss> Oh ... I don't mind, I can work on the most important part
[10:55:15] *** Kancept has joined #haiku
[10:55:33] <Wiss> erf... It was proposed by Haiku
[10:55:40] <Wiss> ok, I'll correct that
[10:55:46] <axeld> Wiss: both parts are important, but I guess the file system would be the more important (but also more difficult) part
[10:55:58] *** Stargater has joined #haiku
[10:56:00] <Stargater> re
[10:56:08] <Wiss> Yes, but I'm not afraid working on it :)
[10:56:34] <axeld> Wiss: there is some userland FS layer that you could use to develop it, though
[10:57:12] <axeld> Wiss: I just don't know very little about that one - you should ask Ingo Weinhold (also a mentor) if you had any questions about that one
[10:57:36] <Wiss> Ok
[10:58:33] *** axeld has quit IRC
[10:59:21] *** axeld has joined #Haiku
[10:59:22] *** ChanServ sets mode: +o axeld
[10:59:35] <axeld> oops, crashed
[11:00:08] <Wiss> no problem ;)
[11:02:01] <axeld> Don't like it a lot, though
[11:02:13] <axeld> Did you reply anything?
[11:02:42] <Wiss> <axeld> Wiss: I just don't know very little about that one - you should ask Ingo Weinhold (also a mentor) if you had any questions about that one
[11:02:42] <Wiss>  <Wiss> Ok
[11:02:58] <Thom_Holwerda> ddew|bofh: can i extend my gratitude to you for your country's excellent contributions to music?
[11:03:17] <ddew|bofh> sure, why not? :)
[11:03:26] <ddew|bofh> what's todays reason then?
[11:03:45] <Thom_Holwerda> im a long-time roxette fan, started back in 1991 when i was 7
[11:03:53] <Thom_Holwerda> and the last few days i decided to travel back in time :)
[11:04:05] <ddew|bofh> they're both great musicians
[11:04:09] <Thom_Holwerda> yup
[11:04:11] <ddew|bofh> i prefer their solo works though
[11:04:14] *** adrien__ has quit IRC
[11:04:27] <Thom_Holwerda> they are much deeper, yes
[11:04:39] <Thom_Holwerda> i like the magic between per and marie
[11:04:46] <Thom_Holwerda> they are such a spot-on match, musicallty
[11:05:05] <ddew|bofh> well per is like the most shallow person ever and his lyrics tend to be a lot of "lalala" and summer things
[11:05:10] <Thom_Holwerda> their stuff from 1990 still sounds modern, it just doesnt age
[11:05:10] <Stargater> re
[11:05:11] <ddew|bofh> but the tunes are catchy :)
[11:05:23] <Thom_Holwerda> ddew|bofh: yes, but at least he doesnt pretend to be anything more
[11:05:28] <Thom_Holwerda> i appreciate that
[11:05:34] <ddew|bofh> per has done one song and repeated it a few hundred times with different lyrics :P
[11:05:58] *** petterhj has quit IRC
[11:05:58] <Thom_Holwerda> nah thats not true
[11:06:06] <ddew|bofh> ever listened to gyllene tider? :)
[11:06:22] <Thom_Holwerda> i dont have their swedish material
[11:06:27] <Thom_Holwerda> very hard to come by here
[11:06:49] <ddew|bofh> yeah, i just remembered that. gyllene tider however is per's "main" band
[11:07:08] <Thom_Holwerda> i liked The World According To Gessle
[11:07:10] <ddew|bofh> insanely catchy tunes, perfect summer music. but the lyrics are very banale
[11:07:17] <Thom_Holwerda> has some very interesting experiments on it
[11:07:28] *** mmu_man has joined #haiku
[11:07:28] *** ChanServ sets mode: +o mmu_man
[11:07:48] <ddew|bofh> yeah, that was an interesting record.
[11:07:52] <Thom_Holwerda> and let's face it, roxette invented the power ballad, perfected it, and made sure no one else could repeat it
[11:08:44] <ddew|bofh> both per and marie are great musicians, roxette really struck out of nowhere
[11:09:01] <ddew|bofh> marie was like a folk-singer before roxette :)
[11:09:02] <Thom_Holwerda> yup, 75 million albums sold
[11:09:25] <Thom_Holwerda> two of my favourite bands are from sweden - roxette and the cardigans
[11:09:30] <Thom_Holwerda> therefore, i wanted to thank you :P
[11:09:39] <ddew|bofh> hehe, you're welcome ;)
[11:10:21] <ddew|bofh> guess i'll have to return the gesture by thanking for Anouk
[11:10:49] <Stargater> cu later
[11:11:13] <Thom_Holwerda> ddew|bofh: anouk has her moments, yeah
[11:11:33] <Thom_Holwerda> dont like all of it, but she's a good musician, and as a person, she's a lot of fun
[11:11:35] <Thom_Holwerda> very down to earth
[11:11:58] *** Stargater has quit IRC
[11:12:23] <ddew|bofh> her lyrics can be pretty lame sometimes but she's got a great voice
[11:13:26] <Thom_Holwerda> i like "Girl"
[11:13:31] <Thom_Holwerda> that song is pounding
[11:13:36] <Thom_Holwerda> awesome stuff
[11:13:43] <Wiss> axeld : I have updated the "Port Samba" project, including more informations about server configuration.;. It's probably my best chance to have a Haiku project, so if you have another advise, they are welcome :)
[11:13:48] <Thom_Holwerda> and her ballads are generally good too
[11:14:04] <Wiss> Do you know how many students have submitted for this project ?
[11:14:39] <ddew|bofh> i love the song "Michel", great ballad
[11:16:12] *** Ingenu has joined #Haiku
[11:16:43] <ddew|bofh> oh, and "Pictures on your skin"
[11:17:09] <clsk> hey axeld, are you still around?
[11:23:56] *** Penix has joined #haiku
[11:24:03] *** hUMUNGUs has quit IRC
[11:25:31] <mmu_man> axeld oh, you're 10h late or 13h early:)
[11:27:11] <Penix> is socket.h supposed to be included in headers/os/NetKit.h? as in, do we have a socket.h somewhere?...
[11:27:32] <Penix> if I comment the include line I can compile the app I'm building without any problems
[11:28:31] <Wiss> hi mmu_man !
[11:28:48] <Wiss> Are you one of the haiku mentors for GSoC ?
[11:28:57] *** emitrax has quit IRC
[11:28:58] <mmu_man> Penix posix/sys/socket.h
[11:29:08] <Penix> ah. thanks
[11:29:13] <mmu_man> Wiss I've been volounteered it seems
[11:29:17] <mmu_man> :)
[11:29:19] <Penix> must be something up with my system
[11:29:27] <Wiss> ok :)
[11:31:38] <Wiss> mmu_man : could you give me your advises about my "Port SAMBA" application ?  They are all welcome, I really hope to join your community with GSoC :)
[11:32:26] <Penix> ooo. finally found a valid patch studio URL. hooray for haikuware
[11:33:04] <Penix> if only I didn't have OHCI USB. could try out my proper midi keyboard.
[11:33:59] <mmu_man> Wiss well I've worked on samba ports, so...
[11:34:11] <mmu_man> someone was supposed to be doing one, but he seems to be missing in action
[11:34:22] *** Kancept has quit IRC
[11:34:25] *** Kancept has joined #haiku
[11:34:47] <Wiss> mmu_man : how, you're my future mentor, so (if you choose me, of course :) )
[11:35:54] <Wiss> How many students have submit for this application ?
[11:36:58] *** emitrax has joined #haiku
[11:37:31] <mmu_man> no idea
[11:37:46] <Wiss> ok lol
[11:39:03] <Wiss> So... if you have any advise to do on my app, it's welcome. If not, that's means I'm in good position to be chosen, that's great :P
[11:40:25] <mmu_man> actually the title seems to be incorrect
[11:40:47] <mmu_man> porting samba itself and writing a CIFS client is 2 entirely different things...
[11:40:54] <mmu_man> that should be clarified, maybe split
[11:42:11] <axeld> mmu_man/Wiss: at least you both speak French natively :-)
[11:42:46] <Wiss> lol axeld, you're right :)
[11:43:40] <mmu_man> righto :)
[11:44:10] <Wiss> mmu_man : ok, I have to change my title subject, so..I don't have really good idea on how to port samba... But of course, I want to learn
[11:44:29] <mmu_man> lemaire... I recall a Lemaire at ESISAR at some point...
[11:44:38] *** axeld has quit IRC
[11:44:45] <mmu_man> but maybe I have bad blocks in /dev/brain
[11:44:49] <Wiss> there are lots of lemaire in France ;)
[11:44:51] <clsk> alexd: I have an idea for the problem we talked about on the list for the ICMP implementation.
[11:44:55] * clsk -> Alan Alvarez
[11:44:58] <mmu_man> right
[11:45:27] *** axeld has joined #Haiku
[11:45:27] *** ChanServ sets mode: +o axeld
[11:45:28] <mmu_man> Wiss: well I can dig my old diffs
[11:45:43] <clsk> erm
[11:45:46] <clsk> alexd: I have an idea for the problem we talked about on the list for the ICMP implementation.
[11:45:49] * clsk -> Alan Alvarez
[11:46:38] <axeld> Hi clsk
[11:46:42] <clsk> I think what we could do is instead of remove the header, is just make a function that move the pointers down to the upper layer header, call the upper layer handler, wait until it's finished handing the packet, then free the header.
[11:46:55] <clsk> moves* the pointer down
[11:47:18] <axeld> Not sure what you mean by moving the pointer down
[11:47:20] <mmu_man> Wiss that's what I came up with on samba 3.0: http://revolf.free.fr/beos/patches/samba-3.0.0-0005.patch.txt
[11:47:28] <mmu_man> was not finished but gives an idea
[11:47:35] <clsk> hello btw
[11:47:50] <mmu_man> of course you can also try writing a CIFS client but that's probably a bigger thing.
[11:48:03] <Wiss> ok yes
[11:48:12] <mmu_man> should be possible to take pieces from the linux fs code, but it's not really portable on its own :^)
[11:48:26] <axeld> Besides it's GPL
[11:48:28] <mmu_man> hmm maybe BSD has a BSD-licenced one that would work better
[11:48:30] <mmu_man> yes
[11:49:13] <Wiss> I have no experiences with BSD... only with linux (Mandriva, Ubuntu, Debian) :(
[11:50:20] <clsk> so move_header would have to know how big the ip header is, then move the pointer down (or up?) to where the tcp or udp header starts
[11:50:36] <mmu_man> Wiss it's just code anyway
[11:50:54] <Wiss> yes
[11:50:55] <clsk> I'm not sure how easy this would be to implement in the current buffer module
[11:51:53] <axeld> clsk: You mean to put this into the protocol code? Or into the net_buffer module?
[11:52:21] *** Barrett666 has joined #haiku
[11:53:29] <clsk> it would mean to add a functionality to the net_buffer module (to have some more flexibility on moving the pointer up and down).
[11:53:53] <clsk> the rest would have to be implemented in the corresponding net module
[11:54:09] *** jiuda_D`arkness has quit IRC
[11:55:20] <clsk> so from the IP net module, instead of doing remove_header, it would be more like move_header. The next protocol will not know of the IP header being there because the net_buffer will be pointing at the next layer's header.
[11:57:08] <clsk> maybe this is already possible? I haven't really looked at the net_buffer code deeply enough.
[11:57:13] <geist> clsk: it doesn't already do that?
[11:57:17] <geist> that's usually how it's done
[11:57:22] <geist> you trim from the front or add to the front
[11:57:46] <geist> and as an optimization as you're building the packet you prepad the buffer so as you add to the front it can just optimize by moving the pointer
[11:57:57] <geist> that's precisley how i did it in the newos net stack
[11:58:15] <clsk> but from what Axel told me it's not guaranteed that when you move the pointer down that the IP header will still be there. It's possible that something else might have claimed that chunk of memory.
[11:58:22] <geist> oh absolutely
[11:58:32] <axeld> clsk: That's pretty much what I suggested in http://www.freelists.org/archives/haiku-development/03-2008/msg00545.html
[11:58:39] <clsk> that's my issue
[11:58:48] <axeld> clsk: so yes, I think this would be a good way to do it :-)
[11:59:06] <geist> hmm, okay
[11:59:27] <geist> is this some case where the upper layer protocols need to know the lower layer header?
[11:59:35] <clsk> geist: does newos handles icmp packets?
[11:59:40] <axeld> geist: exactly
[11:59:48] <geist> clsk: yeah
[12:00:07] <clsk> I shall see how you do it then.
[12:00:21] <geist> doesn't do much with it
[12:00:34] <clsk> as you might know then, you need to send back the ip header back and part of the next layer protocol header
[12:00:37] *** SprMa has quit IRC
[12:00:38] <clsk> 64 bytes to be exact
[12:00:57] <geist> okay
[12:01:21] <geist> i think the general strategy I had in newos was to strip the header from the lower level before passing to the upper level, but passing alongside it the necessary info
[12:01:32] <geist> in the case of icmp, the ip layer also passes the sender's ip address
[12:01:39] <geist> so it can build a packet in case it needs to respond
[12:01:49] <geist> and in the case of tcp it passes the pseudo header alongside
[12:01:56] <clsk> so if for example, someone sends a packet for a port that is not open, the udp module would need to have the ip packet to send back an icmp response.
[12:02:11] <axeld> clsk: btw, it's always nice to answer to questions about your application; Oliver had some lengthy question there
[12:02:15] <geist> does it need the ip packet, or the ip address
[12:02:19] <geist> there's a difference
[12:02:33] <clsk> well, according to the protocol specification, you're supposed to send back the whole header.
[12:02:46] <geist> can you not reconstruct the header from the sender ip?
[12:03:08] <clsk> axeld: We've been exchanging emails about it actually.
[12:03:37] <clsk> the ip packet. You're supposed to send all fields of the ip packet intact
[12:03:56] <clsk> the same way you received them
[12:04:01] <axeld> clsk: it would be better to answer on the comment section, so that other mentors (who need to vote on your application) can see your answers, too
[12:04:04] <geist> okay, well. guess so
[12:04:17] <geist> of course the other option to do all of this is not to actually trim the net buffer
[12:04:32] <geist> you could just pass to the upper layer the net buffer and the offset
[12:04:50] <geist> but it'd still be problematic, because they'd need to know how far back to find the ip header
[12:04:52] <mmu_man> isn't there a size limit on the sent-back IP header ?
[12:04:55] <clsk> axeld: I'm planning on submitting a response on the webapp today.
[12:04:59] <mmu_man> maybe that's only for v4
[12:05:14] <axeld> clsk: okay then, just wanted to mention it :-)
[12:05:16] <geist> it's annoying how much tcp/udp/icmp/arp/ip need to know about each other
[12:05:20] <geist> layering my ass
[12:05:30] <clsk> heh
[12:05:45] <geist> but i guess they're really one big blob, designed only to interoperate with themselves, so what the hey
[12:06:05] <axeld> I think you can separate TCP and UDP pretty nicely, but ICMP and ARP are pretty much dependent on IPv4
[12:06:13] <geist> you could pretty easily define an out of band bit of info that travels with the net buffer up and down the stack that has all the info that would ever be needed
[12:06:24] <geist> i think that's what BSD did with their mbufs
[12:06:34] <clsk> well they way I plan to do it is have the udp module pass whatever was wrong. Kinda like telling the icmp module through the IP module, "this is wrong, let them know"
[12:06:51] <clsk> the UDP or TCP module won't actually build the ICMP packet
[12:06:53] <geist> they had some mode where a piece of the first mbuf is consumed with a structure that contains protocol specific stuff, and the 'start' of the buffer is farther into the block, I believe
[12:07:14] <axeld> that's another option, yes, but it would require allocating a buffer, and copy the info, so protecting the header in the buffer (and sending it alongside the net buffer) seems like a good idea to me
[12:07:53] <clsk> bsd's?
[12:08:10] *** SprMa has joined #haiku
[12:08:42] <geist> axeld: but if you protect the header by leaving it in the net bffer, you'd still need to 'find' it
[12:09:05] <geist> and since it can be variable length, having to figure out how far back to move the pointer to find the header may be pretty expensive
[12:09:16] <axeld> geist: yes, you would need to pass an extra pointer somewhere.
[12:09:20] <geist> sounds likme you also need to store an offset info the net buffer that says where it is
[12:09:20] <clsk> Do IP headers' length change?
[12:09:25] <geist> options, right?
[12:09:32] <geist> or is that tcp. it's been a while
[12:09:34] <clsk> yea, that's what I thought.
[12:09:43] <clsk> that's IP I believe
[12:09:56] <geist> plus i dont know what happens if this is a fragment
[12:10:08] <geist> if it's on top of a fragment that you're having to send back, you've lost the IP header a long time ago
[12:10:18] <geist> because it was consumed when reconstructing the fragment
[12:10:22] <axeld> Actually, both have options
[12:10:34] *** frankps has joined #haiku
[12:10:49] <axeld> geist: yes, but I guess the specs will tell us
[12:10:57] <geist> yeah you know I dont think the need to send back the exact ip header is even feasible
[12:11:03] <axeld> clsk: have you read anything about that?
[12:11:27] <geist> ip fragment would completely destroy it
[12:11:31] <clsk> about what?
[12:11:39] <geist> though it may also be defined that icmp messages cannot be fragmented
[12:11:46] <axeld> clsk: what header needs to be send in case of a fragment?
[12:11:57] <geist> though if you can send an icmp as part of udp or tcp then you're screwed
[12:12:00] <clsk> it doesn't say.
[12:12:05] <clsk> I guess I'll have to do more research
[12:12:18] <geist> yeah, a single udp packet is split across 3 ip fragments. they arrive out of order
[12:12:21] <geist> 1 then 3 then 2
[12:12:21] <clsk> you can't
[12:12:33] <axeld> clsk: okay, it's best to know the situation best before changing the stack according to what we need :-)
[12:12:44] <geist> then it gets passed to udp which needs to send a bad port icmp (isn't that right?)
[12:12:56] <axeld> geist: that's right
[12:12:57] <geist> can't use the original header, since there is no 'original' header for that udp packet
[12:13:00] <clsk> right
[12:13:39] <geist> you obviously need to know the sender so the udp layer can bounce it back, but it needs to know that anyway
[12:13:47] <clsk> hm wait a second.
[12:14:17] <clsk> if an ip packet is defragmented, don't you wait until you have all fragments to send it up to the next layer?
[12:14:22] <geist> exactly
[12:14:35] <geist> in my case it gets 1, 3, then 2
[12:14:42] <geist> 2 completed it so it sent it up
[12:15:21] <clsk> wel the IP header would be the same for the most part.
[12:15:37] <clsk> But I have to check which one has to be sent. The icmp rfc does
[12:15:41] <geist> but there is no original ip header
[12:15:42] <axeld> The IP header of the first packet will be part of the buffer in our implementation
[12:15:46] <clsk> the icmp rfc doesn't specify
[12:15:57] <geist> you'd have to assume one of them, yeah
[12:16:10] <geist> and they'll actually vary, since the flags will be different
[12:16:15] <axeld> I guess in the end it doesn't really matter
[12:16:22] <geist> my guess is it doesn't matter at all
[12:16:39] <geist> i'd look at what other stacks do. if it turns out they dont honor it, dont bother
[12:16:44] <geist> i doubt they do
[12:16:52] <axeld> The only important check would be the checksum, anything else could be made up
[12:17:24] <clsk> I guess it does
[12:17:25] <clsk> Also ICMP
[12:17:25] <clsk>    messages are only sent about errors in handling fragment zero of
[12:17:25] <clsk>    fragemented datagrams.
[12:17:32] <axeld> And I doubt anyone would check if the sum is the same as in the outgoing packet
[12:17:52] <geist> interesting. so technically you need to keep the ip header around for the fragment
[12:17:55] <geist> that's really icky
[12:18:04] <clsk> indeed
[12:18:27] <geist> in the newos implementation I just glue the net buffers together, so that by the time they come out of the fragment, it's a single net buf
[12:18:39] <geist> and who knows where the original pointers went
[12:19:17] <axeld> In Haiku, the first packet remains unchanged, the header of all fragments is thrown away even before they are put together in a single net_buffer
[12:19:56] <clsk> hopefully it won't be doing much fragmanting after path MTU discovery is implemented though
[12:20:03] <geist> I assume the net_buffer in haiku is pretty much the standard chain of buffers that looks like a single one, right?
[12:20:07] <geist> clsk: yeah, exactly
[12:20:10] <clsk> so I guess I'll have to modify that too?
[12:20:22] <axeld> geist: yes, pretty much
[12:21:50] <axeld> clsk: modify what?
[12:22:22] <clsk> the way fragments are handled, so that the header of the first fragmnet is kept.
[12:22:53] <geist> seems like extra credit to me
[12:24:19] <clsk> I'm all for extra credit :)
[12:24:46] <axeld> clsk: as I said, that's already the case in Haiku
[12:25:30] <geist> it just so happens the header is still there somewhere, but it's not 'officially' there, right?
[12:26:03] <axeld> It's still officially there before the packet is forwarded to the upper layers
[12:26:19] <axeld> Raw sockets do see them, for example
[12:26:29] <clsk> oh I'm sorry I thought you said it was the last header that was kept.
[12:26:50] <clsk> I need to work on my reading skills
[12:28:57] <axeld> looks like it wouldn't hurt ;-)
[12:29:32] <clsk> that's mean
[12:30:27] <clsk> either way, I need to start getting ready for work.
[12:30:36] <clsk> bye now
[12:32:30] <axeld> hehe
[12:32:43] <axeld> have a good time!
[12:34:06] *** cherrypie has joined #haiku
[12:38:53] *** mmu_man has quit IRC
[12:41:50] *** hakimrie has joined #haiku
[12:43:13] *** cherry` has quit IRC
[12:46:32] *** frankps has left #haiku
[12:48:34] *** absabs has joined #haiku
[12:51:21] *** waveshaper has joined #haiku
[12:51:31] *** Kancept has quit IRC
[12:52:13] *** hakimrie has quit IRC
[12:55:12] <axeld> Wiss: BTW always have an eye on the comment section to your applications, just in case :-)
[12:56:39] *** Kancept has joined #haiku
[12:59:48] <Wiss> axeld : excuse me, who is BTW ?
[13:00:52] 
[13:01:01] <Wiss> excuse me X_x
[13:01:22] <axeld> np
[13:01:31] <Wiss> I am noticed by email when I receive a comment on my application ;)
[13:01:50] <Wiss> thanks for the advise :)
[13:02:14] <axeld> okay great, I just mentioned it because a lot of applicants never answer to those comments
[13:02:25] <Wiss> ho ? ok strange
[13:02:33] <axeld> you tell me :-)
[13:06:28] <Wiss> I see your comment :)
[13:10:57] *** wildur has joined #Haiku
[13:17:47] *** burfi has joined #haiku
[13:21:06] *** _WindowsUninstal is now known as WindowsUninstall
[13:26:39] <Wiss> I see you later, bye !
[13:26:44] *** Wiss has quit IRC
[13:29:58] *** Tonik has joined #haiku
[13:33:07] <stpere> axeld: are you Axel Doerfler?
[13:34:31] <axeld> stpere: yes
[13:34:46] *** PulkoMandy has joined #haiku
[13:35:08] <stpere> ah cool
[13:35:34] <stpere> pleasure to meet you!
[13:35:37] <stpere> "meet" I should say
[13:41:07] <axeld> you're welcome
[13:47:31] *** Kancept has quit IRC
[13:48:49] *** slaad has joined #haiku
[13:54:13] *** PulkoMandy has quit IRC
[13:54:37] *** PulkoMandy has joined #haiku
[14:01:09] *** goosemo_ has joined #haiku
[14:02:35] *** SprMa has quit IRC
[14:10:25] <stpere> someone got a few minutes to spare, must be an easy problem to solve
[14:10:32] <stpere> I'm just confused right now..
[14:10:53] <stpere> I'm trying to get networking going on my Haiku vmware image
[14:11:00] <stpere> I tried bridged, NAT, ...
[14:11:10] <stpere> without conclusive results I would say
[14:13:00] *** goosemo_ has quit IRC
[14:13:11] <stpere> nevermind, found an educative webpage on the topic!
[14:13:12] *** goosemo_ has joined #haiku
[14:14:41] *** goosemo has quit IRC
[14:15:30] *** goosemo_ is now known as goosemo
[14:16:24] *** Kancept has joined #haiku
[14:17:06] *** DeadYak has joined #haiku
[14:19:18] <stpere> hi DeadYak
[14:19:44] <DeadYak> hiya
[14:20:42] <stpere> are you using Haiku in vmware by any chance?
[14:20:47] <DeadYak> me? nope
[14:20:52] <DeadYak> native hardware only
[14:20:54] <stpere> ok
[14:21:05] <stpere> I'm having trouble setting up the bridged networking
[14:21:33] <axeld> stpere: VMware networking works fine over here using bridged
[14:21:34] <DeadYak> I don't know anything about that sorry
[14:22:20] <stpere> I did setup the bridging on eth1 (my wifi)
[14:22:27] <stpere> it's currently used by my linux
[14:22:44] <stpere> I did setup Haiku to use DHCP
[14:22:50] <stpere> it's not getting the address
[14:22:59] <stpere> well, not a valid one in any case
[14:23:15] <stpere> I wonder where the error might be, in my vmware setup or my haiku setup
[14:23:20] <stpere> or linux..
[14:24:50] <axeld> I am no expert on this either, I'm afraid; I just installed VMware (player) on Ubuntu, pressed <enter> a few times during installation, and got a working network out of that
[14:24:59] <stpere> hmm
[14:25:28] <stpere> yes, I have the feeling it should work and I just have something obvious to do
[14:25:41] <stpere> is there a "dhclient" port in haiku?
[14:26:20] <stpere> hmm, looks like no
[14:26:31] <axeld> stpere: nope, the DHCP functionality is built into the net_server
[14:27:18] <axeld> stpere: currently, you have to quit that one, remove the /dev/net/ipro1000/0 (assuming you're using that one, I do) interface, and restart it to renew the DHCP lease
[14:27:42] <axeld> Kind of inconvenient, but that will definitely improve one day :-)
[14:33:29] <burfi> ES1969 Solo-1 is recognized with the OpenSound add-on. Mixer all there, but still no sound :-(
[14:39:06] *** rcjsuen has joined #haiku
[14:40:03] <clsk> stpere: I've read that bridged networking doesn't work on wireless cards
[14:40:10] <DeadYak> hi clsk :)
[14:40:19] <clsk> There's a patch to get it to work though. Just google for it
[14:40:26] <clsk> hello DeadYak
[14:40:32] <clsk> how's it going?
[14:40:38] <DeadYak> not bad, at work
[14:40:38] <stpere> thx clsk
[14:40:39] <DeadYak> you?
[14:40:50] <clsk> about to head to work :(
[14:45:03] *** CIA-38 has quit IRC
[14:47:39] *** jiuda_D`arkness has joined #haiku
[14:52:06] *** Barrett666 has quit IRC
[14:52:06] *** jiuda_D`arkness is now known as Barrett666
[14:56:31] *** Wiss has joined #haiku
[15:01:44] *** WormDrink has joined #haiku
[15:03:14] *** waveshaper has quit IRC
[15:05:09] *** hUMUNGUs has joined #haiku
[15:07:44] *** Penix has quit IRC
[15:08:15] *** Kancept has quit IRC
[15:09:12] *** CIA-48 has joined #haiku
[15:12:19] *** absabs has quit IRC
[15:19:24] *** W0rmDrink has quit IRC
[15:22:46] *** emitrax has quit IRC
[15:28:07] *** slaad has quit IRC
[15:35:28] *** Kancept has joined #haiku
[15:35:58] *** clsk has quit IRC
[15:36:08] *** leszek has joined #haiku
[15:37:05] <leszek> hi
[15:41:31] *** Ceptikan has joined #haiku
[15:54:57] *** Kancept has quit IRC
[15:56:28] *** luroh has joined #haiku
[15:58:20] <helf_> hi
[15:58:28] *** helf_ is now known as helf
[15:58:44] <helf> morning Ceptikan
[15:59:40] <DeadYak> helf!
[16:01:56] <helf> hey yak
[16:02:42] <helf> whatcha up to?
[16:02:48] *** Ceptikan has quit IRC
[16:03:05] <burfi> My sound chip i supported by OpenSound add-on. The chip is recognized, mixers all there, yet still no sound. Something I could try?
[16:03:11] <DeadYak> the usual, you?
[16:05:11] *** mata_svada has joined #haiku
[16:05:26] <helf> ok, i suppose :)
[16:05:35] <helf> enjoying the nice, dreary weather
[16:05:48] <DeadYak> I dunno, variety is nice
[16:09:07] <helf> heh
[16:10:00] <burfi> Do you know the muffin man?
[16:11:04] <helf> the muffin man?
[16:11:11] <helf> who lives in dreary lane?
[16:11:23] <burfi> :)
[16:11:32] <helf> that was my favorite part of shrek
[16:11:45] <burfi> Not the buttons =-O
[16:12:53] <helf> not the gum drop buttons!
[16:12:57] <DeadYak> :)
[16:13:30] <helf> i noticed something about the history channels phone number yesterday
[16:13:43] <helf> im either reading WAY too deeply into it or they picked it on purpose
[16:14:20] <helf> its 1-800-708-1776.. aka, July 8th, 1776. The day the Col nixon or whatever his name was first publically read the declaration of independance
[16:16:03] <helf> I wonder how many other people have noticedthat
[16:16:31] <helf> They showed the number for ordering DVDs of the show I was watching, and i was like, "Hang on! 1776? That can't be a coincidence!"
[16:17:45] *** W0rmDrink has joined #haiku
[16:25:23] <helf> c'mon, people. you are supposed to be going "ooo! your powers of observation serve you well, helf!"
[16:25:40] <helf> With proper capitalization
[16:27:38] <DeadYak> so, about that hard disk...
[16:27:43] <DeadYak> :P
[16:30:11] <helf> ha.. i need to do that.. I haven't had much of a chance.
[16:30:22] <helf> ill see if i cant get it done this afternoon. I get off work early :)
[16:30:37] <helf> we had our wells bladder tank go out :D
[16:30:47] <helf> that was fun to get semi working again until we get a new one
[16:31:40] *** tic has quit IRC
[16:31:44] <helf> what were the specs on that one, again? 25mhz 040 and ~20mb of ram?
[16:31:57] *** Kancept has joined #haiku
[16:32:03] <helf> morning, Kancept
[16:33:05] *** thorn has quit IRC
[16:33:10] *** thorn__ has joined #haiku
[16:33:32] *** WormDrink has quit IRC
[16:34:07] <helf> deadyak, does he want DHCP or a static setup? I can do which ever
[16:34:14] <helf> if static, ill need numbers
[16:38:00] *** mata_svada_ has joined #haiku
[16:47:22] *** dr_evil has joined #haiku
[16:53:11] *** mmu_man has joined #haiku
[16:53:11] *** ChanServ sets mode: +o mmu_man
[16:55:02] *** Kancept has quit IRC
[16:56:13] *** mata_svada has quit IRC
[16:56:34] *** mata_svada_ has quit IRC
[17:00:54] <DeadYak> helf: DHCP please
[17:01:02] <helf> k
[17:01:45] <helf> yak, do you remember if all the RAM slots were filled on that one or if it had any free slots? I might have some 30 pin ram that'll work in it
[17:02:09] <DeadYak> um...I think it was only using 4 out of 8 slots
[17:02:17] <DeadYak> I'm not entirely certain on that though
[17:02:51] *** ChanServ sets mode: +o dr_evil
[17:03:31] <helf> ok
[17:03:34] <CIA-48> stippi * r24703 /haiku/trunk/src/apps/launchbox/ (LaunchButton.cpp MainWindow.cpp): (log message trimmed)
[17:03:34] <CIA-48> * Better error output in LaunchButton
[17:03:34] <CIA-48> * Load both the app signature and the entry_ref from the settings
[17:03:34] <CIA-48>  in any case, but do not use the entry_ref if the entry does not exist
[17:03:34] <CIA-48>  anymore.
[17:03:37] <CIA-48> * Prefer the entry_ref when launch something. This make it much easier
[17:03:39] <helf> ill check my big box -o ram
[17:03:40] <CIA-48>  to launch a specific executable, especially if multiple copies exist on
[17:10:11] <CIA-48> axeld * r24704 /haiku/trunk/ (6 files in 2 dirs):
[17:10:11] <CIA-48> * Rearranged pthread thread support a bit: there is now a pthread_thread
[17:10:11] <CIA-48>  structure that is attached (via TLS) to each pthread.
[17:10:11] <CIA-48> * Implemented support for pthread_cleanup_{push|pop}().
[17:10:11] <CIA-48> * I haven't really been able to test these changes, yet, though.
[17:10:42] <MindChild> How many commits have I seen with the "I havent/cant/wont test this"
[17:10:54] <MindChild> seriously
[17:11:40] <dr_evil> how many commits did you do MindChild?
[17:11:45] <dr_evil> seriously
[17:12:59] <MindChild> Wow. Scarecrow defense
[17:13:18] <MindChild> "EVERYONE look away from the subject at hand and look at the shiney!"
[17:14:14] <stpere> the subject being..
[17:14:21] <stpere> the dev doesn't have a test case handy
[17:14:45] <stpere> hopefully, someone can make one and test
[17:14:58] <MindChild> Then how do you even know your code does anything different/useful
[17:15:04] <MindChild> clue me in if I am missing something
[17:15:07] <axeld> MindChild: you see that "yet" in there? I often test this stuff on another box, and therefore commit before, that's why I add this disclaimer
[17:15:24] <MindChild> axeld: I do see that in this perticular one, but I meant overall
[17:15:36] <axeld> I usually only see that in my commits ;-)
[17:15:37] <MindChild> it just reminded me, for lack of a better owrd
[17:15:46] <axeld> Others don't test either but don't add a comment ;-P
[17:15:55] <MindChild> heh
[17:18:53] *** rcjsuen has quit IRC
[17:20:54] *** tqh has joined #haiku
[17:25:53] *** Wiss has quit IRC
[17:32:57] <helf> hoyl shit
[17:32:58] <helf> what morons
[17:32:59] <helf> http://www.katu.com/news/local/17126571.html
[17:33:02] <helf> *holy
[17:34:32] *** emitrax has joined #haiku
[17:34:37] <MindChild> Not really. By my fathers house, there are hundreds of miles of unused tracks
[17:34:43] <MindChild> thats what it looks like there
[17:34:55] <MindChild> Literally, trees are growing up in the middles
[17:35:06] <helf> i can understand that, sort of. but these weren unused tracks
[17:35:33] <helf> we have miles of unused track here, too
[17:35:42] <helf> heh, there are several places where they just paved over the track instead of removing it :)
[17:44:49] *** jiuda_D`arkness has joined #haiku
[17:47:33] <DeadYak> FYI: 1 week extension for GSoC student applications is confirmed.
[17:48:32] <PulkoMandy> was it needed ? :)
[17:49:19] <axeld> PulkoMandy: the more, the better, now we have enough time to let you all solve some simple tests ;-)
[17:49:24] <DeadYak> many many projects asked for it
[17:49:38] <PulkoMandy> :)
[17:49:54] *** _Lucretia_ has quit IRC
[17:49:59] <helf> holy smokes
[17:50:10] <helf> some girl just left that looked JUST like the girl fro mDead Like Me
[17:50:17] <axeld> PulkoMandy: at least we were really thinking about something like that, but never got around to actually do something about it
[17:50:19] <dr_evil> helf check this: http://www.boingboing.net/2008/02/20/4ton-railway-bridge.html
[17:51:02] <DeadYak> uhhh
[17:51:07] <DeadYak> how do you steal a 4 ton bridge?
[17:51:58] <dr_evil> using the right tools, that should be easier than stealing railroads
[17:52:05] *** bSON has joined #haiku
[17:52:43] <DeadYak> yeah but moving 4 tons of stuff has got to take some non-trivial equipment regardless :)
[17:56:10] <helf> wow
[17:56:11] <helf> thats awesome :d
[17:56:14] <dr_evil> well, you can load 3 tons in a normal (5T) Sprinter van, or 6 tons in a small truck
[17:56:22] <helf> can you imagine what went into coming up with THAT idea?
[17:56:23] <DeadYak> really? didn't know they could carry that much
[17:56:43] <helf> yak, my van can haul 6tons with a hitch and a trailer attached
[17:56:44] <helf> er
[17:56:49] <helf> 3 tons
[17:56:51] <helf> heh
[17:56:57] <DeadYak> nice
[17:56:57] <helf> 6000lbs
[17:57:11] <helf> if i didnt care about MPG, i could probably haul more with it
[17:57:49] <helf> I'm sure something like that happened in the UK a few years ago: a disused railway bridge over a road was removed overnight and made the news. However it then transpired that the work had been carried out by National Rail who had just forgotten to tell anyone. I can't find the story now though, so maybe I just dreamt it.
[17:57:56] <helf> heh
[18:01:38] <helf> yak, i wont be able to mail that drive out until i get paid friday. but ill try to get it out then :)
[18:01:52] <mmu_man> axeld did you ever see the libpthread from zeta 1.5 ?
[18:02:09] <axeld> mmu_man: nope
[18:02:09] <mmu_man> it was supposedly quite complete and was sent to the maintainers and also the moz team
[18:02:20] <mmu_man> I don't know how complete ours is
[18:02:21] <DeadYak> helf: thanks :)
[18:02:30] <axeld> What maintainers?
[18:02:37] <axeld> mmu_man: not very
[18:02:49] *** Barrett666 has quit IRC
[18:03:17] <helf> oh, wow. axel is on
[18:03:19] <helf> hi axel :)
[18:03:39] <mmu_man> I'll get the source and see how finished it is
[18:04:09] <stpere> yay! the checkout just finished..
[18:04:16] <stpere> started friday night ;-)
[18:04:20] <DeadYak> o.0
[18:04:23] <DeadYak> on what kind of connection?
[18:04:25] <stpere> dialip
[18:04:28] <stpere> dialup*
[18:04:29] <DeadYak> ahh
[18:04:31] <helf> lol
[18:04:33] <helf> ouch
[18:04:35] <stpere> with some interruptions in between
[18:04:40] <DeadYak> was about to say, berlios is slow for me, but not that slow :)
[18:04:45] <stpere> haha
[18:04:53] *** bga has joined #haiku
[18:04:53] *** ChanServ sets mode: +o bga
[18:05:08] <bga> axeld: back to the basement, NOW!
[18:05:08] <stpere> now I'm at work tho, and on broadband
[18:05:09] <bga> :)
[18:05:13] <DeadYak> BAG!
[18:05:18] <axeld> Hey bga!
[18:05:39] <DeadYak> we need to rename one of the kernel threads to "basement lurker"
[18:06:25] <bga> axeld: FYI, I am seeing exactly the same problem as Rene mentioned to you.
[18:06:40] <bga> strings/endian.h is giving an operation not allowed error.
[18:06:45] <axeld> bga: he actually mentioned several ones :-)
[18:06:47] <bga> But I can cat the file just fine.
[18:07:04] <DeadYak> bga: try strace on gcc yet?
[18:07:13] <axeld> Any minimal procedure to reproduce it?
[18:07:20] <bga> When building haiku inside haiku, the build aborts with an "operation not allowed" error in strings/endian.h
[18:07:25] <bga> Doing that now.
[18:07:29] <bga> DeadYak:
[18:07:30] <DeadYak> axeld: I guess "try and build haiku in haiku" isn't considered "minimal"? :)
[18:07:39] <helf> heh
[18:07:43] <bga> axeld: Trying to reproduce it.
[18:07:48] <DeadYak> I hit that error within 5 minutes of the build starting on my a64 in any event
[18:08:25] <axeld> DeadYak: hmm :-)
[18:08:33] <DeadYak> I wonder...what's Haiku's default limit on fds per team?
[18:08:55] <axeld> DeadYak: it could be a preliminary minimal procedure :-)
[18:08:56] <bga> here is a minimail procedure:
[18:09:04] <bga> ./home/bga/development/clients/bga-healthcare/google3/releasetools/bounty/config/gtape.rc
[18:09:09] <bga> ops...
[18:09:10] <bga> :)
[18:09:12] <DeadYak> hahahaha
[18:09:20] <DeadYak> wrong paste buffer? :)
[18:09:23] <bga> Sorry, Copy paste does not work form inside VMWARE.
[18:09:40] <mmu_man> hey
[18:09:42] <mmu_man> plop bga
[18:09:54] <bga> Anyway, I can reproduce it but just manually running the GCC line that compiles xstrio.o
[18:09:57] <DeadYak> axeld: it stops there every time in any case, it's not one of those things that fixes itself on repeat runs :)
[18:09:58] <bga> Will run strace on it.
[18:10:02] <axeld> DeadYak: 256 FDs
[18:10:05] <DeadYak> axeld: ah
[18:10:08] <DeadYak> more than R5 at least :)
[18:10:23] <DeadYak> do we have an equivalent to _kset_fd_limit?
[18:10:35] <mmu_man> _kset_fd_limit :)
[18:10:39] <DeadYak> lol
[18:10:53] <mmu_man> DeadYak actually I mapped it to setrlimit(RLIM_NOFILE) or something
[18:10:59] <DeadYak> ahh.
[18:11:06] <DeadYak> was about to ask if we were going to support the posix way
[18:11:19] <mmu_man> yes it's actually that one the kernel implements
[18:11:26] <stpere> what is the best way to build mockups of an haiku apps..?
[18:11:35] <stpere> I wonder how to get all possible widgets
[18:11:51] <mmu_man> stpere you could try with MeTOS or some other thing
[18:11:51] <dr_evil> i'm bored and tired but need to clean up here :(
[18:12:39] <DeadYak> know the feeling
[18:12:40] <bga> This is weird.
[18:12:46] <DeadYak> bga: what is?
[18:12:52] <bga> axeld: strace does not report any call whatsoever on that file.
[18:13:11] <bga> So the only message about it I get in the strace output is the result from the gcc line.
[18:13:30] <bga> according to strace, gcc does not even try to open the file.
[18:13:40] <DeadYak> does strace catch stuff from child processes?
[18:13:45] <DeadYak> I thought gcc forks cc1plus, etc.
[18:13:54] <bga> Hmmmm... Not sure.
[18:13:57] <bga> axeld?
[18:14:00] *** _Lucretia_ has joined #haiku
[18:14:13] <bga> I did see a SIGCHLD on the output.
[18:14:48] <axeld> bga: you have to tell strace that the children are monitored as well
[18:15:27] <DeadYak> that?
[18:15:32] <bga> strace -s didn't help.
[18:16:43] <stpere> yay! network works in vmware
[18:16:45] <DeadYak> bga: try -f -F
[18:16:59] <stpere> I don't remember who suggested it, but yes, it works on ethernet but not wifi
[18:17:12] <DeadYak> that was clsk I think
[18:17:21] <bga> there is no -F argument in ours.
[18:17:24] <DeadYak> oh
[18:17:29] <DeadYak> sorry, I was looking at linux's arguments
[18:17:43] <DeadYak> I see -f for follow-forks, -F for follow vforks
[18:20:02] <bga> Hmmmmm... strace finishes with a: Failed to run thread 279 : Bad port ID.
[18:20:15] <bga> Maybe that why it is not managing to handle the children processes?
[18:21:05] *** PascalR has joined #haiku
[18:21:15] <DeadYak> I'm assuming strace works by attaching as the debugger?
[18:22:01] <bga> This is interesting...
[18:22:11] *** pyCube has quit IRC
[18:22:13] <bga> I am runing the gcc line manually and get the error...
[18:22:21] <bga> But nothing shows up on syslog. At all.
[18:22:38] <bga> So I think the syslog message I saw earlier is not related...
[18:22:55] <bga> Unless it could be related to the syslog buffer for the "last message repeated" thing.
[18:23:09] <bga> I have no idea how strace works on Haiku.
[18:24:07] <helf> haha, http://4lfa.com/page.php/20080306/083213/PollyMorfs_Love2Hate
[18:25:39] *** urnenfeld has joined #haiku
[18:27:06] <bga> axeld: The syslog buffer should also flush itself form time to time.
[18:27:14] <bga> I got confused by it.
[18:27:29] *** mats has joined #haiku
[18:27:39] *** absabs has joined #haiku
[18:27:48] <mmu_man> bga I heard SoC deadline for student app was pushed to 7th apr ?
[18:28:06] <mmu_man> (said on ffmpeg-devel)
[18:28:14] <helf> yep
[18:28:39] <CIA-48> laplace * r24705 /haiku/trunk/src/apps/bootman/ (6 files):
[18:28:42] <absabs> yes. deadline extended by one week
[18:29:23] *** PascalR has quit IRC
[18:29:33] *** ynop_ has quit IRC
[18:30:40] <stpere> absabs: maybe it's an april fool..
[18:30:52] <DeadYak> it's not april fool's yet
[18:30:54] <DeadYak> :)
[18:31:04] <DeadYak> but yeah, it's official
[18:31:15] <mmu_man> bbl
[18:31:16] *** mmu_man has quit IRC
[18:32:01] <stpere> I encountered a bug in vmplayer :-\
[18:32:15] <stpere> it's pretty nasty
[18:32:56] <absabs> :D
[18:33:07] <stpere> if I maximise the window, it gets fullscreen.. ok, no problem so far.. then I try to release the control (ALT+CTRL), the window return to its initial size then goes back to fullscreen
[18:33:14] <stpere> and I still have control within vmware
[18:33:22] <stpere> so, I can't really quit properly
[18:33:31] <absabs> Here It's April 1st
[18:33:55] <CIA-48> laplace * r24706 /haiku/trunk/src/apps/bootman/Jamfile: Fixed jam build.
[18:34:23] <burfi> down under?
[18:35:13] <burfi> vladivostok?
[18:35:24] <DeadYak> absabs: ah right, I forgot you're on the opposite side of the intl date line :)
[18:35:31] <DeadYak> burfi: he's in .cn I believe
[18:35:49] <absabs> :P
[18:36:20] *** jiuda_D`arkness is now known as Barrett666
[18:39:53] <burfi> <|:)
[18:40:32] <burfi> "<|:)"
[18:40:58] <burfi> fu..
[18:41:03] *** mmu_man has joined #haiku
[18:41:03] *** ChanServ sets mode: +o mmu_man
[18:42:22] <burfi> <|-)
[18:43:17] *** AlexForster has joined #haiku
[18:49:09] *** helf has quit IRC
[18:51:41] *** mmu_man has quit IRC
[18:53:53] *** Teknomancer has joined #haiku
[18:54:03] *** tic has joined #haiku
[18:55:16] *** tic has joined #haiku
[18:55:19] <stpere> how can I open a .pkg in haiku?
[18:55:42] *** rcjsuen has joined #haiku
[18:57:14] <Thom_Holwerda> stpere: dunno if haiku has softwarevalet yet
[18:57:15] *** AndrevS has joined #haiku
[18:57:28] <stpere> m'k
[19:02:43] *** pyCube has joined #haiku
[19:02:44] <AlienSoldier> stpere does it work renaming the mime to zip, i tought they were just zip
[19:03:36] <stpere> unzip doesn
[19:03:41] <stpere> 't recognize it
[19:03:44] <AlienSoldier> ok
[19:03:58] <Teknomancer> no softwarevalet pkgs!=zip .. internally they use zlib so yes, but they have some additional header infos as well
[19:04:19] <stpere> is it closed source?
[19:04:34] <Teknomancer> unfortunately yes
[19:04:41] <stpere> kk
[19:04:50] <stpere> guess we will need some reverse engineering here :)
[19:05:06] <stpere> we need that because many apps are packaged with it
[19:05:08] <AlienSoldier> i don't think it's closed in a sense, as you could figure it by pkging an empty file
[19:05:10] <Teknomancer> there was already someone trying it with some fair degree of success, but this channel is logged and i'm not talking more :P
[19:05:27] <stpere> :)
[19:05:50] <burfi> that is legal in the EU
[19:06:03] <burfi> AFAIK'
[19:06:17] <DeadYak> Thom_Holwerda: /boot/beos/apps/PackageInstaller
[19:06:22] <DeadYak> was one of last year's GSoC projects
[19:06:29] <AlienSoldier> that is one thing we could ask Acces about btw and have a good chance for them to reply yes
[19:06:42] <Thom_Holwerda> DeadYak: i actually remember that now that you mention it :/
[19:07:12] <stpere> great!
[19:07:17] <stpere> DeadYak: thanks
[19:07:18] <stpere> it works
[19:07:20] <Teknomancer> there was talks of mac style .apps for haiku as well as stuff like pacman ... not sure how much these have gone
[19:07:49] <Teknomancer> i guess Beezer's packager should work also then ;)
[19:08:05] <DeadYak> stpere: np :)
[19:08:08] <AlienSoldier> so it's already solved, great :)
[19:08:22] <DeadYak> for some reason it's not linked from the apps menu, probably should be
[19:08:36] <absabs> time to go to bed
[19:08:45] <absabs> good night
[19:08:53] *** absabs has quit IRC
[19:11:25] *** rcjsuen has quit IRC
[19:12:32] <stpere> I get some MessageReceived() debug infos..
[19:12:52] <stpere> MessageReceivedcouldn't
[19:13:00] <stpere> MessageReceived() couldn't understand message*
[19:13:17] <DeadYak> from what?
[19:13:20] <stpere> MeTOS
[19:13:24] <DeadYak> ah.
[19:13:37] <DeadYak> sounds like a debug printf in MeTOS
[19:13:43] <stpere> is it a backward compatibility bug that is probably fixable by me? :)
[19:13:45] <DeadYak> doesn't sound like one of ours anyways
[19:13:51] <DeadYak> not necessarily
[19:13:56] <DeadYak> it might spit that out on R5 too
[19:14:01] <DeadYak> depends what msgs it's talking about
[19:14:37] *** PascalR has joined #haiku
[19:15:52] <stpere> I will download vision and paste privately some of those messages if you don't mind
[19:16:10] <stpere> it's like 5 lines, but I can't copy/paste
[19:16:40] <PascalR> Hi everybody, where can i get gcc from 080323
[19:17:33] <DeadYak> stpere: have any way of checking if it says those messages on R5 too?
[19:17:43] <stpere> sadly no
[19:17:51] <stpere> because I have no R5 on hand
[19:17:54] <DeadYak> because I remember seeing lots of CLI debug from MeTOS back then as well
[19:17:55] <stpere> I should download PE
[19:18:01] <stpere> ah ok
[19:18:16] <DeadYak> it's probably just printing that out on any message it doesn't handle
[19:23:47] *** _Lucretia__ has joined #haiku
[19:26:11] *** nils_t has joined #haiku
[19:29:48] *** _Lucretia_ has quit IRC
[19:29:54] *** _Lucretia__ is now known as _Lucretia_
[19:32:16] *** bSON has quit IRC
[19:32:22] <Gareth> WindowsUninstall: 23
[19:32:24] <Gareth> erm
[19:32:32] <WindowsUninstall> uh?
[19:32:45] <Gareth> tab completion gone horribly, horribly wrong.
[19:33:11] *** bSON has joined #haiku
[19:33:23] <stpere> it's not a coincidence if you typed 23!
[19:33:26] <WindowsUninstall> np
[19:33:29] * stpere return to his book..
[19:33:34] <stpere> "the Secret"
[19:36:02] <AlienSoldier> the original emerald tablet were lot better :P
[19:41:02] *** Euver has joined #haiku
[19:42:01] *** emitrax has quit IRC
[19:44:10] *** Lelldorin1 has joined #haiku
[19:44:26] <Lelldorin1> hi all
[19:48:13] *** _Lucretia__ has joined #haiku
[19:48:27] *** _Lucretia_ has quit IRC
[19:48:44] *** _Lucretia__ is now known as _Lucretia_
[19:49:56] <stpere> hi Lelldorin1
[19:50:04] <stpere> hi _Lucretia_
[20:00:34] *** burfi has left #haiku
[20:00:39] <dr_evil> game rating by crate (the earlier you see one, the designer ran out of ideas) http://www.oldmanmurray.com/features/40.html
[20:02:58] *** phlpp has joined #haiku
[20:03:25] *** Teknomancer has left #haiku
[20:03:46] *** MindChild has quit IRC
[20:04:16] <phlpp> lhmm, is there sth. like "first steps in the haiku enviroment"? (i'm using qemu btw.) the wikibook didn't present very much information and it's quite difficult to find sth. other than coding guidelines/howto on the mainpage
[20:06:35] *** Hugen_ has joined #haiku
[20:06:44] <Hugen_> hi all
[20:10:34] <stpere> dr_evil: hehehe, StoC
[20:10:37] *** MindChild has joined #haiku
[20:11:01] <PulkoMandy> +++
[20:11:13] *** PulkoMandy has quit IRC
[20:14:45] <CIA-48> mauricek * r24707 /haiku/trunk/headers/private/media/DataExchange.h: put enum to appropiate place as pointed out by marcus
[20:15:03] *** MauriceK has joined #haiku
[20:15:15] <phlpp> hm
[20:15:26] <phlpp> so there's nothing like "first steps in haiku"? :>
[20:16:43] <bga> axeld: Around?
[20:17:22] *** stippi has joined #haiku
[20:17:22] *** ChanServ sets mode: +o stippi
[20:17:30] <stippi> axeld!
[20:17:57] <NeonLicht> phlpp: is that a document?
[20:18:11] <DeadYak> hi stippi :)
[20:18:17] <stippi> hi DeadYak
[20:18:58] <NeonLicht> opppsss... phlpp, you were asking :)
[20:19:35] <Hugen_> aaaa
[20:19:38] <phlpp> yep, i was asking ;)
[20:19:43] <Hugen_> hi axeld
[20:20:26] <phlpp> i'm just looking for some more or less detailed introduction to haiku, at sth. like a archive of papers which describe the system in detail
[20:21:00] <phlpp> there only 2 entries in the wikibook and they aren't satisfying ;)
[20:21:28] *** Lelldorin1 has quit IRC
[20:23:38] <stippi> Are there any people applying as students for GSoC lurking here?
[20:25:00] <dr_evil> MauriceK did the StartControlPanel work even with misplaced define?
[20:30:07] <dr_evil> BMediaNode::WaitForMessage does the dispatching in haiku for all known messages, but applications that properly implement HandleMessage() like in the bebook example should work even with unkown values
[20:33:12] <PascalR> Does anybody know where i can download gcc from 080323 as configure says?
[20:33:32] *** burfi has joined #haiku
[20:33:34] <PascalR>  $ ./configure --target=haiku
[20:33:34] <PascalR>  GCC version 2.95.3-haiku-080323 is required!
[20:33:35] <PascalR>  Please download it from www.haiku-os.org...
[20:33:50] *** rcjsuen has joined #haiku
[20:34:09] <stippi> what host OS, PascalR?
[20:34:36] <PascalR> Hi stippi. It's Zata 1.5
[20:35:11] <stippi> just a sec
[20:36:56] <stippi> see https://lists.berlios.de/pipermail/haiku-commits/2008-March/012278.html
[20:37:49] <stippi> download it from http://haiku.mlotz.ch/haiku_cross_compiler_bone_2.95.3-haiku-080323.zip
[20:37:52] <stippi> and configure like this:
[20:38:00] <stippi> wait
[20:38:07] <stippi> extract this archive to /boot
[20:38:14] <stippi> then configure like this:
[20:39:12] <stippi> configure --cross-tools-prefix /boot/apps/cross-tools/bin/i586pc-haiku-
[20:39:52] *** tombhad-AC has joined #haiku
[20:40:18] *** jiuda_D`arkness has joined #haiku
[20:40:21] <CIA-48> laplace * r24708 /haiku/trunk/src/apps/bootman/PartitionsPage.cpp:
[20:40:21] <CIA-48> Replace partition size pretty printing with a version by Ingo copied
[20:40:21] <CIA-48> from file Partitioner.cpp.
[20:41:15] <CIA-48> laplace * r24709 /haiku/trunk/src/apps/bootman/TestBootDrive.h: Fixed typo in header guard.
[20:43:19] *** Korli has joined #haiku
[20:43:19] *** ChanServ sets mode: +o Korli
[20:43:33] <Korli> evening
[20:43:36] <stippi> hi Korli
[20:43:38] *** Kernel86_ has joined #Haiku
[20:44:05] <stippi> we could hold an admin meeting... there is so many ops tonight
[20:44:10] *** Kernel86 has quit IRC
[20:44:44] <Korli> stippi yeah even bga is present :)
[20:45:15] *** Kancept has joined #haiku
[20:45:18] * bga is not. :P
[20:45:23] <stpere> haha
[20:46:18] *** rcjsuen_ has joined #haiku
[20:47:15] *** Barrett666 has quit IRC
[20:47:16] *** jiuda_D`arkness is now known as Barrett666
[20:49:22] <CIA-48> axeld * r24710 /haiku/trunk/ (5 files in 2 dirs):
[20:49:22] <CIA-48> * Added start of an implementation of pthread_cancel().
[20:49:22] <CIA-48> * Implemented pthread_setcanceltype(), pthread_setcancelstate(), and
[20:49:22] <CIA-48>  pthread_testcancel().
[20:49:22] <CIA-48> * In the previous commit, I also made pthread_private.h self-contained.
[20:49:46] *** bSON has quit IRC
[20:51:42] <axeld> Hey stippi!
[20:52:46] <stippi> hi axeld
[20:53:00] <stippi> why are you messing with pthreads? What are you up to?
[20:55:00] <dr_evil> something evil for sure ;)
[20:55:35] *** tombhadAC has quit IRC
[20:55:40] <tqh> ooh, evil as in evil evil? :P
[20:55:57] <stpere> must be to simplifying app porting..
[20:56:58] <tqh> phlpp: You can always read the BeOS r5 docs: http://www.tunetrackersystems.com/bedocs/documentation/User's%20Guide/index.html
[20:57:01] <axeld> stippi: I started looking into porting http://synergy2.sourceforge.net/
[20:58:41] <tqh> axel is that between different computers over network?
[20:59:02] <axeld> tqh: exactly
[20:59:06] <stippi> cool
[20:59:17] *** tombhad-AC has quit IRC
[20:59:28] <tqh> ah, go ahead then :)
[20:59:37] <stippi> but axeld: You should finish the HDA driver!
[20:59:56] *** rcjsuen_ has quit IRC
[21:00:10] <axeld> stippi: well, I just got all platform independent parts (using Unix as platform) building, I didn't do much more yet
[21:00:22] <axeld> stippi: but I can't work on that all day - that would drive me nuts
[21:00:56] *** tombhadAC has joined #haiku
[21:01:36] <stippi> I can sympathize
[21:01:57] <axeld> bbl
[21:02:05] <Thom_Holwerda> latest comic up, dont think anyone will get this one though
[21:02:06] <Thom_Holwerda> http://www.osnews.com/comics
[21:02:58] <burfi> Has the GSoC flyer been updated with extended date? I'll be stopping by at my Alma Mater tomorrow - opportunity to pin the flyer there.
[21:03:20] *** rcjsuen has quit IRC
[21:03:35] *** rcjsuen has joined #haiku
[21:04:32] <stpere> we should make Haiku run on the EEE Pc
[21:06:06] <dr_evil> stpere I know it doesn't boot, but I don't have an EEE PC
[21:06:28] <dr_evil> and I'm not going to buy one, becauseI don't like the small screen
[21:06:36] <burfi> New models out next week anyway
[21:07:00] *** TheNerd_LT has joined #haiku
[21:07:17] <dr_evil> I think those are still too small
[21:07:36] <stippi> burfi: More interesting, but I just got a new notebook.
[21:07:42] <Thom_Holwerda> i kinda like the new models, with the bigger screens
[21:07:49] <stippi> which coincidentally runs Haiku very well... :-P
[21:07:52] <burfi> Atom processor, new chipset and all, presumably
[21:08:10] <stippi> stpere: Can you code?
[21:08:44] *** petterhj has joined #haiku
[21:09:55] <dr_evil> I should really fix the ATAPI handing in the new ATA stack, found little time lately
[21:10:29] <stpere> stippi: I think I do
[21:10:38] <stpere> I've never coded kernels and stuff tho
[21:10:53] <stpere> business apps and small games
[21:12:00] * stpere seriously lacks confidence..
[21:12:36] <DeadYak> dr_evil: idle curiosity, how much work is it to finish DMA transfers in the new stack? or is that done already?
[21:14:22] *** Kokito has joined #haiku
[21:14:49] <Kokito> hello folks
[21:14:51] <stpere> hi!
[21:16:44] <stpere> stippi: was it a "do it yourself" kind of remark? :)
[21:17:22] <stippi> :-)
[21:17:38] <stippi> ... just asking
[21:18:33] <stippi> dr_evil: while you're here, does the SATA AHCI driver already support ATAPI devices?
[21:18:36] * DeadYak pets Kokito
[21:18:55] *** rcjsuen has quit IRC
[21:19:10] <stpere> meow?!
[21:19:38] <Thom_Holwerda> darn, arent all the big shots in here today or what
[21:19:56] <stippi> stpere: seriously, it will happen, if these things all happen at once:
[21:20:10] <stippi> 1) A skilled developer gets himself an EEEPC
[21:20:29] <Kokito> hey DeadYak
[21:20:31] <stippi> 2) He feels very motivated to port Haiku to it and/or fix any bugs which prevent it from working on this device
[21:20:53] <stippi> it certainly will *not* happen only because we all feel it is a good idea.
[21:21:05] <stippi> that's a bummer, but that's just how it works.
[21:21:25] *** rcjsuen has joined #haiku
[21:21:42] <Thom_Holwerda> i suspect the eeepc hardware requires some special tricks here and there
[21:21:58] <stippi> Thom_Holwerda: The way bigger problem is the screen size
[21:21:59] <Thom_Holwerda> luckily any potential dev can study linux' implementation
[21:22:13] <stippi> why do you think ASUS invented a special GUI to support it?
[21:22:22] <stippi> I know all my apps don't run at this screen size
[21:22:26] <Thom_Holwerda> you're right stippi
[21:23:11] <stippi> I don't think the hardware is even any special, it's just we are lacking some drivers to support it.
[21:23:13] <Korli> stippi virtual screen is ok
[21:23:31] <stippi> unfortunately, the current app_server does not support that.
[21:23:41] <stippi> no auto scrolling...
[21:24:04] <Thom_Holwerda> virtual screen space? that really turns off every usability related synaps :(
[21:24:12] <tqh> I'm actually a bit interested in the new model.
[21:24:48] <tqh> although I'm also interested in an ebook reader.
[21:27:34] <stippi> bbl
[21:28:16] <phlpp> tqh: okay, thanks
[21:30:00] *** nils_t has quit IRC
[21:30:14] <phlpp> tqh: The requested URL /bedocs/documentation/User's0Guide/index.html was not found on this server.
[21:31:25] *** nils_t has joined #haiku
[21:32:26] <phlpp> ah, okay, i see what's wrong in the url
[21:32:57] <tqh> phlpp: yes, clicking on it works here
[21:33:20] <tqh> 0 -> %20
[21:33:25] *** dchis has quit IRC
[21:33:25] <phlpp> yup
[21:34:22] *** leszek has quit IRC
[21:41:07] <MauriceK> dr_evil: yup it did work... sorry for late response, watching football and was in the shower while break
[21:43:35] *** Hugen_ has quit IRC
[21:45:18] <MauriceK> stippi: you mentioned encoding support lately on the mailing list... any concrete plans so far?
[21:47:02] <CIA-48> mmlr * r24711 /haiku/trunk/src/add-ons/kernel/file_systems/fat/file.c:
[21:47:02] <CIA-48> This was probably changed by accident in r22886. It caused all read/write
[21:47:02] <CIA-48> pages calls of the fat filesystem to return an error instead of the effective
[21:47:02] <CIA-48> result. Please shout if this was done on purpose. I tested a bit with a fat
[21:47:02] <CIA-48> volume and where it failed previously when opening files it now works as
[21:47:02] <CIA-48> expected.
[21:47:53] <stippi> hi MauriceK
[21:48:00] <stippi> no, not really
[21:48:05] <CIA-48> laplace * r24712 /haiku/trunk/src/apps/bootman/DefaultPartitionPage.cpp: Forgot to remove debug code.
[21:48:11] <stippi> I didn't even follow any unimplemented code paths
[21:48:20] <stippi> so see what exactly is lacking
[21:49:13] *** urnenfeld has left #haiku
[21:49:43] *** tombhadAC has quit IRC
[21:50:27] *** tombhadAC has joined #haiku
[21:50:33] <MauriceK> stippi: ok... I was always interested in it, but didn't have the time to do something concrete on it. Additionally still getting into this whole thing
[21:54:11] <stippi> MauriceK: You are doing a good job
[21:55:00] *** Yaroze has quit IRC
[21:55:01] <MauriceK> thx
[21:55:10] *** Barrett666 has quit IRC
[21:55:58] *** Yaroze has joined #haiku
[21:56:18] <phlpp> ftp://ftp.haikuhost.com/haiku/videos/NUMERICA07_conference_Haiku_operating_system_divx.avi <- does the link work for somebody?
[21:59:41] <burfi> not me. you can get it from archive.org though
[22:00:09] <stpere> ftp.haikuhost.com isn't working at all for me
[22:00:21] <stpere> it bails out before the login prompt
[22:02:00] <CIA-48> mmlr * r24713 /haiku/trunk/src/add-ons/kernel/drivers/disk/ (8 files in 3 dirs): (log message trimmed)
[22:02:00] <CIA-48> Introduce a simple usb_disk driver that supports USB mass storage devices of
[22:02:00] <CIA-48> the bulk-only class using transparent SCSI commands (i.e. most of the current
[22:02:00] <CIA-48> external harddrives and flash media). It emulates the few SCSI commands needed
[22:02:00] <CIA-48> to get this sort of devices working and does not interface with the SCSI
[22:02:03] <CIA-48> subsystem. This makes it far easier to get working and also far better fits
[22:02:05] <CIA-48> how the USB stack works (as drivers can be dynamically rescanned when device
[22:02:16] <phlpp> burfi: erm
[22:02:21] <phlpp> how to get it from archive.org?
[22:02:28] <phlpp> i have get no matching results :/
[22:02:33] <DeadYak> oh wow.
[22:03:07] <burfi> don't have the link handy. moment...
[22:03:18] <phlpp> kay
[22:03:43] <stpere> wow, nice commit :)
[22:04:26] *** PascalR has quit IRC
[22:04:46] <burfi> http://www.archive.org/details/NUMERICA_2007_conference_Haiku_operating_system_divx
[22:05:08] <phlpp> thanks
[22:08:18] <burfi> Might Haiku connect to my Lumix camera now?
[22:09:41] *** emitrax has joined #haiku
[22:09:42] *** siarzhuk has joined #haiku
[22:09:48] <stippi> I need to check this out
[22:09:51] <stippi> see you all later
[22:10:03] *** stippi has quit IRC
[22:12:29] <emitrax> siarzhuk: hi
[22:12:42] <siarzhuk> hi
[22:13:09] <emitrax> siarzhuk: did you see the latest mmlr commit ?
[22:13:26] <siarzhuk> not yet
[22:14:52] <stpere> siarzhuk: apparently it brings a initial usb mass storage driver
[22:15:32] <emitrax> maybe that's only a temporary solution until the device manager structure has been discussed
[22:16:07] <stpere> yeah
[22:16:18] *** Kokito has quit IRC
[22:16:23] * stpere isn't sure what was discussed tho :)
[22:17:03] *** burfi has quit IRC
[22:17:23] <siarzhuk> Russianas say: There are nothing more persistent than temporary one. :-)
[22:18:08] <emitrax> I hope not
[22:18:27] <siarzhuk> I'm too
[22:19:59] <emitrax> let's see if any comments come out :p
[22:20:07] *** burfi has joined #haiku
[22:20:51] <DeadYak> bbl, heading home
[22:20:56] *** DeadYak has quit IRC
[22:22:39] *** aljen has joined #haiku
[22:22:41] <aljen> hey :)
[22:23:19] *** rcjsuen has quit IRC
[22:23:49] *** daste has joined #haiku
[22:24:27] *** MichaelHenry has joined #haiku
[22:24:36] *** daste_ has joined #haiku
[22:25:00] *** Stargater has joined #haiku
[22:25:03] <Stargater> hi
[22:25:11] *** daste has quit IRC
[22:25:29] <stpere> is there some interest in a more evolved address book app than what People provide?
[22:25:37] <stpere> no offense to the author, mind you
[22:25:39] <stpere> it's great
[22:26:04] <emitrax> stpere: better ask this question in the mailing list
[22:26:11] <stpere> ah yes, good idea
[22:28:05] *** bga has quit IRC
[22:28:35] *** MindChild has quit IRC
[22:28:37] *** pfoetchen has joined #haiku
[22:34:37] *** MindChild has joined #haiku
[22:35:07] *** MindChild has quit IRC
[22:35:22] *** MindChild has joined #haiku
[22:38:52] *** plfiorini has joined #haiku
[22:39:23] <plfiorini> hi
[22:39:52] *** mmu_man has joined #haiku
[22:39:53] *** ChanServ sets mode: +o mmu_man
[22:39:59] *** ozy` has quit IRC
[22:41:12] *** daste_ is now known as daste
[22:45:42] <Stargater> hmm haikuware have a image vor virtuelbox and haiku-os.org not ?
[22:46:42] <ddew|bofh> yup
[22:46:54] *** phlpp_ has joined #haiku
[22:47:05] <ddew|bofh> there's some extra work involved in getting a vbox image and it's not guaranteed to work for everyone
[22:48:24] *** Kokito has joined #haiku
[22:54:05] *** Euver has quit IRC
[22:54:24] <plfiorini>  shouldn't vbox be the same as qemu, or am i missing something?
[22:55:21] *** DeadYak has joined #haiku
[22:55:34] *** TheNerd_LT has quit IRC
[22:56:12] *** ozy` has joined #haiku
[22:56:38] <tqh> plfiorini: I think it has some extra options enabled for accel or such. There is a bug for virtualbox and one for qemu with special options.
[22:57:21] *** AnEvilYak has quit IRC
[22:57:41] <plfiorini> tqh: thanks
[22:57:49] *** MichaelHenry has quit IRC
[22:58:45] *** burfi has quit IRC
[23:00:13] *** burfi has joined #haiku
[23:03:50] *** tqh has quit IRC
[23:04:32] *** daste has quit IRC
[23:04:57] *** dchis has joined #haiku
[23:05:05] *** urnenfeld has joined #haiku
[23:05:26] *** axeld has quit IRC
[23:05:49] *** va1 has joined #haiku
[23:06:54] *** va1 has left #haiku
[23:10:06] *** phlpp has quit IRC
[23:12:53] *** emitrax has quit IRC
[23:14:32] *** User328 has joined #haiku
[23:22:12] *** Korli has quit IRC
[23:23:40] <CIA-48> mauricek * r24714 /haiku/trunk/src/add-ons/media/media-add-ons/ (7 files in 2 dirs):
[23:23:40] <CIA-48> * get reader.media_addon compiling (both gcc2 and 4)(used for testing BFileInterface node kind)
[23:23:40] <CIA-48> * warning/style fixes
[23:23:40] <CIA-48> * removed some dubious comments :)
[23:23:40] <CIA-48> * no behaviour changes so far...
[23:23:48] *** siarzhuk has quit IRC
[23:24:35] <pyCube> psshh.. i like my oses with dubious comments
[23:25:28] <plfiorini> :)
[23:25:46] *** clsk has joined #haiku
[23:27:30] *** Ingenu has quit IRC
[23:27:31] <MauriceK> well, comments like "We don't crash, because you are so stupid" shouldn't be in i guess
[23:28:33] <plfiorini> grrr my reply to haiku-dev ml has not been received, i'm wondering what gmail's smtp is doing
[23:30:10] <MauriceK> gn8
[23:30:14] *** MauriceK has quit IRC
[23:33:06] *** Wiss has joined #haiku
[23:36:05] <mmu_man> best being panic("We never get here!");
[23:36:44] <mmu_man> or // this is useless but it only works with it
[23:41:09] *** User328 has quit IRC
[23:41:32] <Thom_Holwerda> omg this is fcuking unbelievable
[23:41:46] <Thom_Holwerda> we have this mandatory course at university where we need to build a large website
[23:41:55] <Thom_Holwerda> THEY'RE MAKING US DO IT IN FRONTPAGE
[23:42:28] <Thom_Holwerda> the guy better tell me tomorrow im allowed to use notepad or else im in for 8 weeks of pure pain
[23:45:02] <mmu_man> use XEmacs :
[23:45:03] <mmu_man> !
[23:45:24] <mmu_man> (*yes* there is a windows version and *yes* it even looks better in win than linux)
[23:49:05] *** stpere_ has joined #haiku
[23:51:08] <Stargater> ore vim
[23:51:47] <mmu_man> yeah
[23:51:55] <mmu_man> should be available in cygwin
[23:52:11] <Stargater> Kokito the bebook can we have to download ? (zip)
[23:52:14] <stpere_> I'm having issues compiling the buildtools
[23:52:21] <stpere_> am I the only one?
[23:52:24] <pyCube> umm.. why would anybody even suggest the use of frontpage.. for anything beyond being an object of ridicule
[23:52:56] <stpere_> ~/haiku$ ./configure --build-cross-tools ../buildtools/                     doesn't succeed
[23:53:13] <stpere_> it fails within bfd
[23:53:25] <stpere_> I've just updated my svn
[23:53:26] <plfiorini> stpere_: what error exactly?
[23:53:30] <stpere_> sec
[23:53:47] <Stargater>  ./configure --cross-tools-prefix /boot/apps/haiku/cross-tools/bin/i586-pc-haiku-
[23:53:51] <plfiorini> i have rev24713 and it is fine
[23:55:19] <stpere_> http://pastebin.com/m1332be36
[23:55:36] <stpere_> I have pasted what I think might be relevent, but I'm not quite sure
[23:56:01] *** geist_ has joined #haiku
[23:56:08] <stpere_> oh, I'm missing a build tool I think
[23:56:14] <stpere_> makeinfo
[23:56:51] <plfiorini> install texinfo

   March 31, 2008  
< | 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 | >