[00:04:23] *** Racer__X has quit IRC
[00:18:25] *** lizdeika has joined #haiku
[00:32:01] *** oco has quit IRC
[00:33:34] *** Methe has quit IRC
[00:37:24] *** znation has quit IRC
[00:37:49] *** znation has joined #haiku
[00:38:19] *** fyy_vision has quit IRC
[00:47:41] *** shackan_ has joined #haiku
[00:48:26] *** ShackaN has quit IRC
[00:50:29] *** ConneX has quit IRC
[00:52:56] *** fyysik has joined #haiku
[00:54:06] *** slaad has quit IRC
[01:14:19] *** shackan_ has quit IRC
[01:16:59] *** AnEvilYak has joined #haiku
[01:27:13] *** Sg_Henry has quit IRC
[01:37:41] *** dipp has quit IRC
[01:38:05] <guerrilla> anyone know when the build factory going to be back online?
[01:38:17] <guerrilla> s/going/is going
[01:49:26] *** badonaway has joined #haiku
[01:58:07] *** fyysik has quit IRC
[02:12:14] *** Fanskapet has quit IRC
[02:19:14] <CIA-6> bonefish * current/src/kernel/libroot/posix/sys/uname.c:
[02:19:14] <CIA-6> * utsname::sysname is the name of the OS while system_info::kernel_name is
[02:19:14] <CIA-6> the name of the kernel file. Now return "Haiku" as operating system.
[02:19:14] <CIA-6> * Even if we can't fill in utsname::nodename with something useful yet,
[02:19:14] <CIA-6> fill in at least something.
[02:20:42] <CIA-6> bonefish * current/ (headers/private/app/AppMisc.h src/kits/app/AppMisc.cpp): Added function is_running_on_haiku(), which returns at runtime whether we're running under Haiku.
[02:22:30] <CIA-6> bonefish * current/src/kits/storage/mime/database_support.cpp: Use the standard MIME DB path when running under Haiku, but a different one when running under BeOS (so BeOS' MIME DB is not affected when running tests).
[02:24:41] <CIA-6> bonefish * current/src/tools/ (resattr/Jamfile resattr/resattr.cpp Jamfile): Added resattr, a small tool that converts resources into attributes and vice versa.
[02:31:51] <CIA-6> bonefish * current/src/data/beos_mime/ (19 files in 7 dirs):
[02:31:52] <CIA-6> Added source files for our MIME database. The files are in rdef format,
[02:31:52] <CIA-6> specifying resources. They are converted via rc to resource files and
[02:31:52] <CIA-6> those in turn via resattr to empty files with attributes.
[02:31:52] <CIA-6> Ideally rc would understand attributes, too...
[02:33:43] <CIA-6> bonefish * current/makehdimage: Install the MIME DB in the image.
[02:33:56] *** Trevor1 has quit IRC
[02:35:03] *** Trevor1 has joined #haiku
[02:35:27] *** Fanskapet has joined #haiku
[02:46:20] *** lemon0 has joined #haiku
[02:46:23] <lemon0> hello
[02:46:50] <lemon0> is anonymous access to cvs working ?
[02:53:34] <mmadia> yes.   just tried.
[02:57:42] *** voidref has quit IRC
[03:03:32] *** slaad has joined #haiku
[03:04:40] <CIA-6> darkwyrm * current/headers/private/servers/app/ (FontFamily.h FontServer.h ServerFont.h): Retooled the server-side font API to make it work better
[03:04:46] *** brennanOS has joined #haiku
[03:04:48] <brennanOS> Hi all
[03:06:08] <CIA-6> darkwyrm * current/src/servers/app/server/ (7 files): Retooled the API and implemented some of the message handlers
[03:08:52] <brennanOS> what type of system call interface does BeOS use?
[03:09:36] <brennanOS> at the assembly language level?  i.e. is it like Unix where you push arguments on the stack, put a call number in EAX, and do a software interrupt?
[03:10:02] <@geist> "unix" doesn't do it any particular way
[03:10:12] <@geist> but that's the traditional way of doing it, and I imagine how beos does
[03:10:21] <@geist> seeing as it hasn't been developed in like 6 years
[03:10:40] <brennanOS> ok, FreeBSD does the way I described
[03:11:14] <@geist> that will always work on x86
[03:11:28] <@geist> depending on the architecture, you can use syscall or sysenter
[03:11:36] <@geist> but no matter what you have to provide some sort of fallback mechanism
[03:11:40] <brennanOS> well, Linux expects arguments in registers
[03:12:03] <@geist> well, some of the arguments
[03:12:10] <brennanOS> but yeah I would guess BeOS passes them on the stack
[03:12:11] <@geist> I forget what I defined for newos
[03:12:12] <brennanOS> yeah
[03:12:43] <@geist> iirc, I put the syscall number in eax, the number of args in another register, and then the pointer to the args in the third
[03:12:46] <@geist> on x86 at least
[03:13:17] <brennanOS> hmm
[03:13:38] <brennanOS> so theoretically, the args themselves could be anywhere in RAM the kernel has access to?
[03:13:43] <@geist> right
[03:13:49] <brennanOS> hm
[03:13:49] <@geist> but in reality it points to the stack
[03:13:57] <@geist> but that's for the legacy implementation
[03:13:58] <brennanOS> yeah, I figured
[03:14:14] <@geist> for newer cpus I can (haven't done it yet) use sysenter/syscall
[03:14:26] <@geist> but it requires a bunch of gruntwork to get that going and there are bigger fish to fry
[03:14:36] <brennanOS> is that a new(er) cpu instruction?
[03:14:38] <@geist> I doubt haiku will take that though since they need to be binary compat with beos
[03:14:40] <@geist> yes
[03:15:16] <brennanOS> how does it differ from the INT <num> method?
[03:15:23] <brennanOS> I guess I could just look it up myself
[03:15:29] <@geist> it goes in through a different mechanism
[03:15:52] <@geist> instead of going through the full interrupt mechanism, both methods (one is amd, one is intel) have a fast way of getting into ring0
[03:16:02] <brennanOS> I see
[03:16:03] <@geist> an 17 02:16:03 <@geist>	an int instruction takes quite a few cycles
[03:16:19] <@geist> I forget how many, but at least a few hundred
[03:16:38] <brennanOS> hmm.. sounds interesting
[03:16:52] <@geist> iirc, the amd version is a lot more sane than the intel version
[03:16:53] <agentmumu> 7win 2
[03:17:24] <@geist> and since amd pretty much defined the new x64 architecture, I betcha intel x64 cpus will switch to amds method
[03:19:07] <brennanOS> nice
[03:19:22] <AnEvilYak> geist: what's the cycle time on syscall/sysenter?
[03:19:28] <@geist> I dunno
[03:19:37] <@geist> I'm sure it's well documented
[03:19:44] <@geist> as you can see I haven't actually *done* anything with it
[03:19:50] * AnEvilYak nods
[03:20:38] <lemon0> sorry for interrupting, people, google said that comes new driver architecture where i read about that ?
[03:20:56] <lemon0> point me to it
[03:21:16] <CIA-6> darkwyrm * current/src/servers/app/server/FontFamily.cpp: Fixed style name -> face translation
[03:21:52] * geist points in the general direction of sourceforge servers
[03:21:58] <@geist> it's over there somewhere
[03:23:45] <brennanOS> syscall is AMD, and sysenter is Intel
[03:24:03] <@geist> yeah
[03:24:13] <@geist> syscall is much more sane
[03:25:14] <AnEvilYak> in what sense?
[03:25:48] <brennanOS> syscall takes its medication every day
[03:25:51] <brennanOS> brb
[03:26:13] <@geist> syscall works under the assumption you have a flat memory model
[03:26:18] <@geist> which everyone does nowadays
[03:26:34] <@geist> so I think you just in a few special register set the address to jump to and the stack to switch to
[03:26:42] <@geist> and it just calls right into ring0
[03:27:01] <@geist> the sysenter method is older and tries to still work with any sort of crazy segmentation model you can come up with
[03:27:09] <@geist> so it does some crazy shit that makes my head hurt
[03:27:40] <@geist> the reason they're faster is they avoid all the overhead with doing a cross segment call
[03:27:54] <brennanOS> I really need to read up on x86 again
[03:27:58] <@geist> there are lots of permission checks that go on in those cases that can be streamlined if there's only one way to do it
[03:28:09] <@geist> ist>	i wish I could forget this stuff actually
[03:29:36] <brennanOS> INT takes 30 cycles on 486
[03:29:36] <brennanOS> this is an old book, heh
[03:30:05] <@geist> I betcha it'll take longer if you go across priviledge levels
[03:30:10] <@geist> which a syscall int will do
[03:32:38] <brennanOS> yeah...
[03:33:07] <brennanOS> the 30 it looks like is just the saving of state and jmp to the new address
[03:33:48] <brennanOS> forgive another question... if you use a flat memory model, how do you separate kernel memory from user space memory?
[03:36:09] <@geist> paging
[03:36:20] <@geist> the paging system provides a second layer of memory protection
[03:36:23] <@geist> supervisor vs user
[03:36:33] <@geist> so you do it on a page by page basis
[03:36:40] <@geist> that's how virtually all other architectures do it
[03:37:04] <@geist> though sometimes they let you reference alternate address spaces at the instruction level, like on ultrasparc
[03:39:20] <brennanOS> hm...
[03:39:37] <brennanOS> Linux is segmented I think.. is that right?
[03:39:40] <brennanOS> no
[03:39:47] * brennanOS goes and gets another book
[03:40:39] <brennanOS> so each page is 'marked' somehow as belonging to the system vs. the user, and the memory management hardware is able to check if an access is valid
[03:44:46] <@geist> that's right
[03:45:47] <brennanOS> I just did one of the stupidest things Ive done in a long time
[03:45:53] <brennanOS> funny
[03:46:36] <brennanOS> I just downloaded, installed and assembled some code with NASM
[03:46:46] <brennanOS> but then I realized Im on a Mac
[03:47:28] <@geist> yeah it wont run
[03:47:51] <brennanOS> yeah, no shit ;)
[03:47:58] <brennanOS> thats a funny one
[03:48:18] <brennanOS> I guess I could download dosbox or something
[03:48:43] <brennanOS> or just get on a PC
[03:54:53] *** liquidboy has joined #haiku
[04:00:32] *** liquidboy has quit IRC
[04:01:51] *** lizdeika has quit IRC
[04:08:17] *** Fanskapet has quit IRC
[04:10:48] *** brennanOS has quit IRC
[04:17:37] *** nPHYN1T3 has quit IRC
[04:45:53] *** mmadia has quit IRC
[04:59:47] *** brennanOS has joined #haiku
[05:30:24] *** Potn has quit IRC
[05:31:04] *** Potn has joined #haiku
[05:49:01] *** lemon0 has quit IRC
[06:06:11] *** mmadia has joined #haiku
[06:09:46] *** Potn has quit IRC
[06:10:16] <guerrilla> hey brennanOS you should check out "Intel Architecture Software Developer's Manual. Vol 3: System Programming"
[06:10:25] <guerrilla> it goes over that in the 2nd chapter
[06:11:06] <guerrilla> (segmentation, privledge level, rings (what systems use to seperate kernel space from user space)
[06:11:10] <guerrilla> etc
[06:11:12] <guerrilla> )
[06:11:34] <guerrilla> http://www.cs.ucf.edu/~ekambara/cop6614/Pentium_Sherri.pdf
[06:11:45] <guerrilla> that's also a good little explanation with diagrams
[06:12:59] <guerrilla> and btw, linux doesn't use the stack for system call arguments it uses registers, with the exception of socket calls which one of the arguments in the registers is a pointer to yet more arguments in memory, specific to that socket call...
[06:13:39] <guerrilla> what version of "Unix" were you reffering to?
[06:18:49] <guerrilla> (i now realize that conversation was an hour or two ago, sorry :P)
[06:34:10] *** znation has quit IRC
[06:34:48] *** znation has joined #haiku
[06:38:11] <brennanOS> guerilla: thanks
[06:39:02] <brennanOS> guerilla: yeah I knew about Linux using registers for system calls, I was referring to FreeBSD
[06:44:13] <brennanOS> where are the standard C library functions implemented? in what library?
[06:44:19] <guerrilla> glibc
[06:44:24] <guerrilla> in linux
[06:44:31] <brennanOS> In BeOS I mean
[06:44:52] <guerrilla> yeah, sorry (i'm not the person to ask)
[06:45:10] <brennanOS> I assumed libroot or libbe
[06:45:50] <brennanOS> Im trying to link an ELF object file to create a BeOS executable... I wrote it in asm, and I need to resolve a couple of symbols
[06:46:02] <brennanOS> usind ld
[06:46:04] <brennanOS> using
[06:46:32] <brennanOS> which actually doesnt have anything really to do with my previous question, I was just curious
[06:46:41] <guerrilla> oh
[06:46:49] <guerrilla> illa>	i don't know, sounds like an interesting task though
[06:46:49] <guerrilla> :)
[06:47:50] <brennanOS> beos/develop/lib/x86/libroot.so: undefined reference to `memset_internal'
[06:48:05] <brennanOS> not sure where memset_internal is exported
[06:48:09] <brennanOS> if not in libroot
[06:48:23] <guerrilla> use nm
[06:48:31] <guerrilla> nm /lib/* | grep memset_internal
[06:48:39] <guerrilla> or readelf, if that's on there
[06:50:31] <brennanOS> $ nm /boot/develop/lib/x86/* | grep memset_internal
[06:50:31] <brennanOS> U memset_internal
[06:50:31] <brennanOS> U memset_internal
[06:50:42] <brennanOS> boot/develop/lib/x86/libstdc++.r4.so: :Unknown command
[06:50:47] <brennanOS> thats the output of nm
[06:51:12] <brennanOS> oops, last line is:
[06:51:13] <brennanOS> boot/develop/lib/x86/libstdc++.r4.so: no symbols
[06:51:44] <brennanOS> what does 'U memset_internal' mean?
[06:51:48] <slaad> libbe, brennanOS?
[06:52:00] <brennanOS> slaad: in response to what question?
[06:52:18] <slaad> Where memset_internal may be located.
[06:52:30] <slaad> But I doubt it's exported
[06:52:35] <brennanOS> hm
[06:52:50] <brennanOS> I wonder how I can manually link this thing, then
[06:54:01] <brennanOS> well, if libroot.so refers to memset_internel, and that name isnt exported, how does a program get successfully linked?
[06:58:01] <brennanOS> well, Ill give up the fight for tonight
[07:02:58] <brennanOS> later
[07:03:00] *** brennanOS has quit IRC
[07:04:01] *** RageMax has quit IRC
[07:06:10] <guerrilla> slaad: no
[07:06:15] <guerrilla> that's not where it is located
[07:06:18] <guerrilla> U means undefined
[07:06:22] <guerrilla> T would be where it really is
[07:06:54] <guerrilla> U means that program page faults when it executes that and asks the system for the real code
[07:07:03] <guerrilla> he left though..
[07:07:03] <guerrilla> :)
[07:07:08] <guerrilla> oh well
[07:08:26] <slaad> You'll note he never mentioned libbe. I wasn't referring to the output of nm
[07:08:44] *** guerrilla has quit IRC
[07:08:45] <slaad> So... neener, neener, neener!
[07:08:46] <slaad> Damn.
[07:13:11] *** m_eiman has joined #haiku
[07:20:21] *** mmu_man has joined #haiku
[07:20:21] *** ChanServ sets mode: +o mmu_man
[07:33:57] *** stippi has joined #haiku
[07:53:26] *** m_eiman has quit IRC
[08:15:41] <w-ber> http://apz.fi/apz/pwned.jpg <-- that's a broken down Matrox drive
[08:16:23] <w-ber> uh
[08:16:25] <w-ber> Maxtor
[08:16:28] <w-ber> seems like I'm still sleeping
[08:16:58] *** guerrilla has joined #haiku
[08:32:14] *** stippi has quit IRC
[08:37:00] *** voidref has joined #haiku
[08:37:02] *** ChanServ sets mode: +o voidref
[08:54:31] *** voidref has quit IRC
[08:57:14] *** mmadia has quit IRC
[09:00:24] *** NathanW has quit IRC
[09:01:53] *** thaflo has joined #haiku
[09:07:19] *** voidref has joined #haiku
[09:07:20] *** ChanServ sets mode: +o voidref
[09:08:46] *** nielx has joined #haiku
[09:08:58] <nielx> Could someone try to ssh to me?
[09:09:08] <nielx> I have a major problem with configuring my router
[09:09:30] <znation> nielx: sure
[09:11:20] *** JBurton has joined #haiku
[09:11:28] <JBurton> hi
[09:12:24] *** nielx has quit IRC
[09:14:06] *** nielx has joined #haiku
[09:14:34] <nielx> znation: sorry about that
[09:14:47] <nielx> my router has found a new way of entertaining itself:
[09:14:50] <nielx> random restarting
[09:16:04] <w-ber> refreshing
[09:16:36] <nielx> yeah, extremely, especially when you're ssh'ed into a gentoo system which was almost done building glibc
[09:16:57] <w-ber> always, always use screen :)
[09:17:15] *** pst has joined #haiku
[09:17:54] *** pst is now known as |pst|
[09:18:24] <JBurton> hi |pst|
[09:18:25] <nielx> well, I'm getting an airport basestation
[09:19:54] *** Karina``_ has joined #haiku
[09:19:55] *** Karina`` has quit IRC
[09:22:52] *** sl44d has joined #haiku
[09:23:50] <nielx> anyway, later
[09:23:51] *** nielx has quit IRC
[09:34:18] *** Fanskapet has joined #haiku
[09:36:16] <CIA-6> sikosis * current/src/prefs/fonts/ (16 files): Updated
[09:40:26] *** slaad has quit IRC
[09:40:26] *** sl44d is now known as slaad
[09:50:37] *** znation has quit IRC
[09:50:57] *** znation has joined #haiku
[10:08:32] *** emwe has joined #haiku
[10:29:23] *** slaad has quit IRC
[10:36:06] *** BGA has joined #haiku
[10:36:07] *** ChanServ sets mode: +o BGA
[11:00:16] *** ShackaN has joined #haiku
[11:00:18] *** AnEvilYak has quit IRC
[11:21:03] *** BGA has quit IRC
[11:25:40] *** emwe has quit IRC
[11:27:44] *** emwe has joined #haiku
[11:41:12] *** thies has quit IRC
[11:42:10] *** Undrew has joined #haiku
[11:42:10] *** thies has joined #haiku
[11:42:55] *** emwe has quit IRC
[11:44:33] *** emwe has joined #haiku
[11:50:54] *** mmu_man has quit IRC
[11:55:57] <CIA-6> rudolfc * current/src/add-ons/accelerants/nvidia/engine/nv_acc_dma.c: fixed FIFO channel pointer error that caused assert_fifo to always fail (DMA, >NV40). Should fix driver hanging after changing workspaces with different modesettings on NV40 and later.
[12:00:27] *** Methe has joined #haiku
[12:36:27] *** thies has quit IRC
[12:36:27] *** znation has quit IRC
[12:40:03] *** znation has joined #haiku
[12:42:32] *** thies has joined #haiku
[12:42:45] *** Methe has quit IRC
[12:42:59] *** Fanskap has joined #haiku
[12:45:57] *** mmu_man has joined #haiku
[12:45:58] *** ChanServ sets mode: +o mmu_man
[12:46:06] *** Fanskapet has quit IRC
[12:49:58] *** xeD has joined #haiku
[12:50:49] *** xeD has quit IRC
[12:51:24] *** xeD has joined #haiku
[12:54:14] *** GregCrain has joined #haiku
[12:55:11] *** mes_ has joined #haiku
[13:01:43] *** Undrew has quit IRC
[13:05:49] *** emwe has quit IRC
[13:14:06] *** stippi has joined #haiku
[13:16:52] <stippi> Anybody? strcasecmp() and strcmp()?
[13:17:01] <|pst|> ?
[13:17:06] <stippi> what is the difference between strcasecmp() and strmp()?
[13:17:23] <CIA-6> rudolfc * current/src/add-ons/accelerants/nvidia/ (4 files in 2 dirs): added 2D_surface setup (DMA), added fifo channel assignent check for aquire_engine, cleaned some stuff up.
[13:17:59] <@mmu_man> strcmp("abc","ABC") != 0
[13:18:08] <@mmu_man> strcasecmp("abc","ABC") == 0
[13:18:50] <stippi> mmu_man: thanks a bunch, I think the functions is badly misnamed, thus my asking. But it is exactly what I hoped for.
[13:19:03] <CIA-6> rudolfc * current/headers/private/graphics/nvidia/nv_acc.h: added more DMA defines (for FIFO channel assigment).
[13:19:45] <@mmu_man> should be strnocasecmp but it's 2 chars more to type :)
[13:19:54] *** emwe has joined #haiku
[13:20:03] <@mmu_man> ELAZY
[13:20:26] <stippi> or maybe stricmp(), which would have helped-
[13:21:48] *** Methe has joined #haiku
[13:44:21] *** ahwayakchih has joined #haiku
[13:44:24] <ahwayakchih> hih>	hi everyone
[13:44:33] <ahwayakchih> hih>	hi JBurton
[13:44:36] <ahwayakchih> hih>	hi stippi
[13:52:51] <JBurton> re
[13:52:52] <JBurton> hi ahwayakchih
[13:52:53] *** GregCrain has quit IRC
[13:53:28] <JBurton> stippi there is stricmp() actually
[13:53:35] <JBurton> and also strcmpi()
[13:53:42] <JBurton> it depends on the C implementor
[13:54:03] <JBurton> IIRC stricmp() is/was used on bsd systems
[13:54:07] <JBurton> strcmpi() on windows
[13:54:15] <JBurton> strcasecmp() should be posix
[13:54:25] <JBurton> but I agree it's misnamed
[13:54:53] <@mmu_man> hmm some of them also handle locale stuff
[13:55:15] 
[13:55:23] <@mmu_man> depending on the locale
[13:56:17] <JBurton> ahwayakchih around ?
[13:56:29] <ahwayakchih> yep
[13:57:10] <JBurton> ahwayakchih usually you greet me with "re" everytime I "re" :P
[13:57:27] <ahwayakchih> oops, OMG, i'll be damned!
[13:57:29] <ahwayakchih> sorry
[13:57:34] <ahwayakchih> FORGIVE ME!!!
[13:57:35] <JBurton> mmu_man do you know if beos stropt() has some weird bugs by chance ?
[13:57:37] <JBurton> ahwayakchih ahah :)
[13:57:39] <ahwayakchih> re JBurton :))
[13:57:44] <JBurton> thanks ahwayakchih :P
[13:57:47] <ahwayakchih> hehe
[13:58:30] <@mmu_man> do'nt you mean getopt ?
[13:58:42] <JBurton> ah damn yes :P
[13:58:47] <JBurton> sorry, Freudian typo :P
[13:58:48] <JBurton> mmu_man
[13:58:56] <@mmu_man> dunno
[13:59:20] <JBurton> I got weird results depending on the order of the parameters in the parameter string
[14:01:10] *** stippi has quit IRC
[14:10:22] <CIA-6> rudolfc * current/src/add-ons/accelerants/nvidia/engine/nv_acc_dma.c: added setting pattern colorspace (DMA)
[14:11:58] <CIA-6> rudolfc * current/headers/private/graphics/nvidia/nv_acc.h: added pattern cmd colorspace define (PIO/DMA)
[14:29:58] <ahwayakchih> JBurton i have an request for You :)
[14:47:11] <ahwayakchih> JBurton would it be possible to make menu's scrolling public? so one coudl manipulat menus too big for screen? or at least make menu/popup menu autoscroll to currently selected item (at least in radio mode) :))
[14:50:30] *** emwe has quit IRC
[14:51:53] *** emwe has joined #haiku
[14:53:39] *** nPHYN1T3 has joined #haiku
[14:54:24] *** mes_ has quit IRC
[14:57:25] <JBurton> ahwayakchih back
[14:57:31] <ahwayakchih> re
[14:57:39] <JBurton> ahwayakchih menu scrolling public...
[14:57:40] <ahwayakchih> JBurton
[14:57:46] <JBurton> hmmmm I have yet to implement it actually :P
[14:58:23] <ahwayakchih> heh
[14:59:51] <w-ber> http://www.qdb.us/41739
[15:00:38] <JBurton> I think I'll leave that as the last thing to do :P
[15:01:45] <w-ber> I have to admire C's simplicity
[15:01:53] <w-ber> then again...
[15:06:22] <ahwayakchih> JBurton hmm ok, though my font popup in WonderBrush is too big to fit in screen, and i would like it to be scrolled properly after opening :)
[15:07:27] <JBurton> actually it should scroll automatically if it goes beyond the screen
[15:12:00] <ahwayakchih> yep, but current popup doesn't scroll, so if i have z* font selected and want to select x* font i have to scroll all the fonts each time... ;]
[15:12:19] <JBurton> ah, yes, I see the problem
[15:21:25] *** xeD has quit IRC
[15:34:32] *** BeMonni has joined #haiku
[15:35:10] <ahwayakchih> cya everyone
[15:35:22] *** ahwayakchih has quit IRC
[15:36:11] <JBurton> bye ahwayakchih
[15:36:11] <JBurton> oh
[15:36:14] <JBurton> he already left
[15:49:34] <CIA-6> axeld * current/src/servers/syslog_daemon/syslog_output.cpp:
[15:49:34] <CIA-6> The daemon now runs on top of our kernel, so it should now write to
[15:49:34] <CIA-6> the real syslog.
[15:49:34] <CIA-6> Also, find_directory() now makes sure that the path to the syslog
[15:49:34] <CIA-6> actually exists.
[15:51:11] *** ConneX has joined #haiku
[15:59:21] *** znation has quit IRC
[15:59:36] *** znation has joined #haiku
[16:03:46] *** dipp has joined #haiku
[16:11:54] *** Fanskap has quit IRC
[16:13:15] <CIA-6> rudolfc * current/src/add-ons/accelerants/nvidia/engine/nv_acc_dma.c: added cmd defines (CTX registers) for pre-NV40 cards, corrected a CTX programming error for NV40 and higher (2d_surface).
[16:14:26] <sys2> hows that accel comming? ... anyone know if there is a differnece yet? :>
[16:40:01] <CIA-6> axeld * current/Jamrules: Removed now unnecessary comment.
[16:41:57] *** Methe has quit IRC
[16:45:19] *** mahlzeit has joined #haiku
[16:45:19] *** ChanServ sets mode: +o mahlzeit
[16:45:45] <JBurton> hi mahlzeit
[16:45:49] <@mahlzeit> hey JBurton
[16:47:16] *** mmadia has joined #haiku
[16:49:44] *** SamuelGZ has joined #haiku
[16:50:45] *** mes_ has joined #haiku
[16:52:02] <CIA-6> rudolfc * current/src/add-ons/accelerants/nvidia/engine/nv_acc_dma.c: completed NV10 architecture DMA programming to the same level as NV43. Hopefully this engine will show more signs of life than NV43 still does..
[16:52:11] *** brennanOS has joined #haiku
[16:54:13] *** Methe has joined #haiku
[16:54:17] <CIA-6> rudolfc * current/headers/private/graphics/nvidia/nv_macros.h: added NV4/NV10 architecture register define needed for DMA stuff
[16:54:34] *** mes_ has quit IRC
[16:55:03] <Methe> mahlzeit!
[16:55:06] <Methe> howdy ?
[16:55:09] <@mahlzeit> uh oh
[16:55:11] * mahlzeit runs away
[16:55:21] *** thaflo has quit IRC
[16:55:30] <JBurton> hey Methe
[16:55:40] <Methe> hola JBurton
[16:55:43] <Methe> come es tai ?
[16:56:06] <@mahlzeit> bah, just lost a game of rummy
[16:56:25] <Methe> I'm doing a playlist for a radio atm :DD I'm in a webradio. I have already done one and i'm mkaing another one :DDD
[16:56:48] <@mahlzeit> you have many listeners?
[16:57:12] <Methe> last week during my playlist got between 40 and 60
[16:57:21] <@mahlzeit> not too bad
[16:57:30] <Methe> indeed
[16:58:03] <Methe> its not a very well know radio
[16:58:06] <Methe> that's why
[16:58:09] <Methe> but anyway it's fun :
[16:58:11] <Methe> :)
[16:58:19] <@mahlzeit> nice
[16:58:52] <Methe> next one will be the the night in a strange gardent
[16:58:54] <Methe> :DD
[16:58:55] *** badonaway has quit IRC
[16:59:01] *** Greyhawk_Dragon has joined #haiku
[17:00:00] *** thaflo has joined #haiku
[17:00:08] <@mahlzeit> hey, maybe you can include this song (look away little children): http://home.tiscali.nl/mahlzeit/suckmycunt.mp3
[17:00:20] <@mahlzeit> my friend made it :-)
[17:00:36] <@mahlzeit> it's very funny
[17:00:48] * Methe lol @ title
[17:01:10] <JBurton> fine thanks Methe
[17:01:21] <Methe> JBurton: nice delay :]]
[17:01:52] <Methe> mahlzeit: I can't pu it into my playlist cause it's not in the theme i'm making
[17:01:56] <Methe> but anysong is welcome
[17:01:58] <Methe> for the radio
[17:02:07] <Methe> he>	he can just dend links to the radio
[17:02:10] <Methe> and hop they'll get in it
[17:02:18] *** badonaway has joined #haiku
[17:02:39] <Methe> kontakt at erreur404 dot org is the mail adress
[17:02:43] <@mahlzeit> did you listen to the song? :-)
[17:03:15] * Methe Dls it
[17:03:21] <Methe> AHAH
[17:03:24] <Methe> MIDI powa
[17:03:48] <Methe> that's indeed fun.
[17:04:03] <Methe> and it will be playlisted without a doubt :)
[17:04:27] <Methe> I dont know yet how to include songs in main playlist
[17:04:31] <Methe> ahah
[17:04:36] <Methe> who's thekid
[17:04:39] <Methe> who's the kid
[17:04:45] <@mahlzeit> kid?
[17:04:49] <Methe> girl
[17:04:50] <Methe> ?
[17:04:58] <@mahlzeit> it's a guy, pitched up :-)
[17:05:05] <Methe> huhu
[17:11:41] <JBurton> bye all
[17:11:45] <@mahlzeit> cya
[17:12:51] *** JBurton has quit IRC
[17:13:58] *** mmadia has quit IRC
[17:16:29] *** tqh has joined #haiku
[17:19:35] *** Greyhawk_Dragon has quit IRC
[17:29:09] *** MikeW has joined #haiku
[17:29:28] *** _gingerninja is now known as gingerninja
[17:30:14] *** ConneX has quit IRC
[17:44:32] *** gingerninja has quit IRC
[17:48:44] *** gingerninja has joined #haiku
[17:50:30] *** Chatarra has joined #haiku
[17:51:40] <CIA-6> laplace * current/headers/private/print/BeUtils.h: Added function ScaleRect().
[17:54:34] *** fyysik has joined #haiku
[17:55:00] <CIA-6> laplace * current/src/add-ons/print/drivers/pcl6/DeltaRowCompression.h: Removed misleading comment
[17:55:04] <fyysik> tqh - there are two existsd in last patch - what does it mean?
[17:55:06] *** Chatarra has quit IRC
[17:55:11] <fyysik> Exists()
[17:55:26] *** Chatarra has joined #haiku
[17:55:57] *** Chatarra has quit IRC
[17:56:12] *** Chatarra has joined #haiku
[17:56:51] *** mmu_man has quit IRC
[18:03:22] *** illissius has joined #haiku
[18:05:56] *** fyysik has quit IRC
[18:06:26] *** SamuelGZ has quit IRC
[18:18:15] *** thaflo has quit IRC
[18:44:29] *** mmadia has joined #haiku
[18:46:10] *** emwe has quit IRC
[18:47:31] *** emwe has joined #haiku
[19:09:08] *** TuneTracker has joined #haiku
[19:10:33] *** MikeW has quit IRC
[19:11:12] *** fyysik has joined #haiku
[19:12:40] *** m_eiman has joined #haiku
[19:12:54] <fyysik> for now - http://beos.spb.ru/program/123/Samba2.2.7a-BeOS-BONE-test2.pkg.zip
[19:16:01] *** RageMax has joined #haiku
[19:17:42] *** Potn has joined #haiku
[19:19:14] *** oco has joined #haiku
[19:25:51] <brennanOS> fyysik: how long does it take to build Mozilla?
[19:36:00] *** Chatarra has quit IRC
[19:38:43] *** |DaGo| has joined #haiku
[19:40:34] <fyysik> brennanOS - depends on CPU
[19:41:16] <fyysik> on PIII-550 with IDE replacement driver at indexless partitions - 4-5 hours
[19:41:36] <fyysik> much faster under Linux and Windows
[19:41:55] <fyysik> ~ 1 hour at same machine
[19:47:04] * tqh needs reboot
[19:47:07] *** tqh has quit IRC
[19:47:31] <tic> fyysik, eek. why do you think that is?
[19:47:34] <tic> fyysik, or perhaps you know
[19:47:39] *** Fanskapet has joined #haiku
[19:48:14] *** tqh has joined #haiku
[19:54:13] * fyysik tried
[19:54:46] *** Fanskapet has quit IRC
[19:54:49] <fyysik> someone said that's due ugly BeOS VM implementation + BFS overheads
[19:55:28] *** Begasus has joined #haiku
[19:55:43] *** DaaT has joined #haiku
[19:55:47] <brennanOS> ick
[19:55:47] *** |DaGo| has quit IRC
[19:55:52] <brennanOS> thats a huge time difference
[19:56:33] *** Fanskapet has joined #haiku
[19:56:50] <mmadia> is it possible to build mozilla on a fat32 partition?
[19:56:55] <fyysik> yup
[19:57:17] <fyysik> timeless from irc.mozill.org managed to build BeZilla on Fat32
[19:57:59] <mmadia> did he mention if it was significantly faster than a nonindexed bfs?
[19:58:11] <fyysik> ik>	i think he didn't compare
[19:58:12] <tic> it was probably ot. :)
[19:58:13] <tic> not.
[19:58:25] <tic> as the fat32 driver in beos is way slower than bfs.
[19:58:30] <tic> but on a windows _system_, it's faster.
[19:58:35] <tic> different VM and all.
[19:59:03] <fyysik> tic - "fast/slow" is relative uncertain word. There are very various fail operations
[19:59:15] <fyysik> one is copying big files at once
[19:59:21] <tic> m'kay.
[19:59:27] <fyysik> second is permanent open/close for little files
[19:59:34] <fyysik> for example
[19:59:35] <tic> I find fat32 volumes slower than bfs.
[20:03:39] *** mmadia42 has joined #haiku
[20:06:49] * tqh is waiting on bugzilla mail...
[20:08:12] *** NathanW has joined #haiku
[20:09:55] *** YNOP has joined #haiku
[20:11:39] <fyysik> for now - http://beos.spb.ru/program/123/Samba2.2.7a-BeOS-BONE-test2.pkg.zip
[20:11:58] *** Fanskap has joined #haiku
[20:12:02] *** Fanskap has quit IRC
[20:12:09] <fyysik> if someone needs file sharing in BONE urgently:)
[20:12:28] *** Fanskap has joined #haiku
[20:12:42] *** Fanskapet has quit IRC
[20:20:33] *** mmadia has quit IRC
[20:20:34] *** mmadia42 is now known as mmadia
[20:24:08] *** stippi has joined #haiku
[20:24:37] *** Begasus has quit IRC
[20:25:24] *** illissius- has joined #haiku
[20:27:28] *** Begasus has joined #haiku
[20:33:42] *** illissius has quit IRC
[20:42:39] <fyysik> tqh
[20:45:39] *** znation has quit IRC
[20:46:55] *** znation has joined #haiku
[20:51:02] *** thies_ has joined #haiku
[20:54:11] *** mahlzeit has quit IRC
[20:55:26] <tqh> fyysik yes?
[20:56:13] <fyysik> tqh - did you receiceived your promised Zera copy already? No info from z_eight fro long time
[20:58:00] <tqh> no, havn't heard about that. Although I'm not very concerned about it either.
[20:58:11] <CIA-6> darkwyrm * current/src/kits/interface/ (Font.cpp InterfaceDefs.cpp):
[20:58:11] <CIA-6> Moved _init_global_fonts to Font.cpp
[20:58:11] <CIA-6> Implemented _init_global_fonts
[20:58:11] <CIA-6> BFont::GetHeight returns proper values
[20:58:21] <tqh> If I was I would have bought it already.
[20:59:26] <fyysik> tqh - me to, but i wish biesi and timeless will get their copies at last
[20:59:54] <tqh> ah yes that would be good
[21:00:59] *** thies has quit IRC
[21:01:15] <tqh> fyysik Is IFRAMEs in their own BWindow?
[21:05:56] <fyysik> hmm, that's interesting question...wondered about it myself long ago
[21:05:57] <CIA-6> darkwyrm * current/headers/private/servers/app/ (FontFamily.h FontServer.h ServerFont.h): More API fixes and tweaks, mostly to get BFont::GetHeight working
[21:06:40] <fyysik> tqh - drop-downs are
[21:07:10] <fyysik> forms in buzilla main query are. Lot of them at least
[21:09:10] <CIA-6> darkwyrm * current/src/servers/app/server/ (5 files):
[21:09:10] <CIA-6> Changed keyboard nav color to match R5 even though we use Dano's attributes
[21:09:10] <CIA-6> BFont::GetHeight now works properly
[21:11:40] <fyysik> tqh, but i again suspecy from my current picture in ProcessController, that most objects at https://bugzilla.mozilla.org/query.cgi?format=advanced are InvisibleWindows now:)
[21:12:11] <fyysik> after that patch you know about, no more w->   in PC, just empty places for threads
[21:13:24] <tqh> ah
[21:13:35] *** Racer__X has joined #haiku
[21:13:48] *** thaflo has joined #haiku
[21:14:36] <fyysik> http://beos.spb.ru/mozilla/BugZilla-threads.png tqh
[21:15:25] <fyysik> one w> is main window with title, second is URL-bar drop-down
[21:16:25] <tqh> ah, and lots of untitled threads. hmm.
[21:16:55] <fyysik> this is nsprthreads+bthreads build
[21:17:10] <fyysik> interesting is to compare with different build
[21:17:13] <tqh> I use it here as well
[21:17:30] <fyysik> maybe some older version on your disk?
[21:19:44] <tqh> yes, I'll need to install process controller...
[21:20:43] <fyysik> what!!!??? You lived with PC till now???:)
[21:20:47] <fyysik> without
[21:23:14] <tqh> I had it before, but I did reinstall some time ago and didn't install it.
[21:24:19] <tic> bah. how can you avoid it? :)
[21:25:27] <tqh> Must be because I'm good :P
[21:25:44] <tic> heh.
[21:26:00] <tic> there's some ice-cream in the freezer waiting to be eaten by me.
[21:26:09] *** AtomoZero has joined #haiku
[21:28:52] *** lordcoxis has joined #haiku
[21:29:35] <tqh> fyysik: bthreads one on the left: http://europe.dnsalias.net/bthreads_on_the_left.png
[21:29:37] *** daguito has joined #haiku
[21:30:49] <daguito> v
[21:30:55] * tqh pets poorman
[21:31:31] <daguito> ?
[21:31:51] *** daguito has quit IRC
[21:32:29] *** mmadia has quit IRC
[21:33:02] <fyysik> tqh - those on the left looks like explicit BWindows, though, titleless
[21:33:18] <tqh> fyysik Yes, looks so.
[21:33:32] <CIA-6> laplace * current/src/add-ons/print/drivers/shared/libprint/ (GraphicsDriver.cpp JobData.cpp): Fixed bugs.
[21:34:49] * tqh goes to checks once more to be safe
[21:35:29] <tqh> I suspect one BWin for every dropdown.
[21:36:27] <tqh> Yep sure looks like it.
[21:36:52] <fyysik> eWindow_type == popup
[21:36:53] *** SamuelGZ has joined #haiku
[21:37:55] *** SamuelGZ has quit IRC
[21:38:10] *** SamuelGZ has joined #haiku
[21:38:40] *** daguito has joined #haiku
[21:39:05] <daguito> hi ppl
[21:39:49] *** BGA has joined #haiku
[21:39:49] *** ChanServ sets mode: +o BGA
[21:42:22] *** Konrad has joined #haiku
[21:47:31] <tqh> fyysik it's the popups:  http://europe.dnsalias.net/popups.png
[21:50:28] *** AtomoZero has quit IRC
[21:52:09] *** m_eiman has quit IRC
[21:54:10] <fyysik> tqh - which ones? bugzilla forms?
[21:54:20] *** AtomoZero has joined #haiku
[21:54:32] <fyysik> or all IFRAMEs ?
[21:55:08] *** DaaT has quit IRC
[21:55:51] *** DaaT has joined #haiku
[21:57:59] *** mmu_man has joined #haiku
[21:57:59] *** ChanServ sets mode: +o mmu_man
[21:58:02] <tqh> fyysik bugzilla forms.
[21:58:27] *** mmadia has joined #haiku
[22:00:35] <tqh> idg.se has three popups, but I can't dind one (url-bar is probably one)
[22:00:43] *** Konrad has quit IRC
[22:01:18] <tqh> ah found them, so iframes are probably not windows, if idg.se has iframe.
[22:02:58] <fyysik> ik>	i think not ALL forms are popups there though
[22:03:20] <fyysik> but maybe i'm wrong. looking at page source may help
[22:04:37] <tqh> fyysik idg.se has iframes but all windows are accounted for, so IFrames are probably not in their own windows.
[22:05:21] <tqh> which should be good. I suspect drawing order on children to cause that redraw problem.
[22:09:46] <fyysik> ik>	i suspected it too, 2 years ago, but then dropped those suspects
[22:10:26] <fyysik> vice versa, had thoughts about doing in MS Window way - replace BViews with BWindow+BView:)) but then dropped that idea too:)
[22:13:11] <@mmu_man> there is a finite number of BWindow you can cerate
[22:13:22] <@mmu_man> ue to the thread limit in app_server
[22:14:01] *** mmadia has quit IRC
[22:16:06] <fyysik> http://beos.spb.ru/mozilla/BugZilla-threads.png mmu_man
[22:16:28] <@mmu_man> hehe :)
[22:16:34] <fyysik> https://bugzilla.mozilla.org/query.cgi?format=advanced for that page
[22:19:41] * tqh should learn to only run one firefox build at a time.
[22:20:18] <@mmu_man> ouch =)
[22:20:54] <@mmu_man> or have a big swap
[22:21:20] <@mmu_man> [revol@patrick ~]$ cat /dev/misc/sysinfo
[22:21:21] <@mmu_man> max_swap_size   436207616
[22:21:21] <@mmu_man> swap_size       110944256
[22:21:22] <@mmu_man> used_swap_size  24576
[22:21:59] <fyysik> mmu_man - IIRC there were something like CIFS browsers since SAMBA 2.0 - did you investigate that problem? (I forgot names of those apps:(
[22:22:51] <@mmu_man> EALREADY
[22:22:51] *** thies_ is now known as thies
[22:23:46] <Methe> http://abfhm.free.fr/basket.swf OMFG !
[22:28:42] *** xeD has joined #haiku
[22:31:34] *** Methe has quit IRC
[22:33:43] *** ShackaN has quit IRC
[22:36:36] *** slaad has joined #haiku
[22:38:01] *** ShackaN has joined #haiku
[22:38:49] *** daguito has quit IRC
[22:47:06] *** stippi has quit IRC
[22:47:54] *** LupusMichaelis has joined #haiku
[23:00:27] *** brennanOS has quit IRC
[23:00:47] * tqh wonders why it's impossible to get stackcrawls with symbols in libwidget_beos.so
[23:01:51] <@mmu_man> is it built with -g ?
[23:02:38] <tqh> dunno, don't think it should be.
[23:02:45] <tqh> -g is strip?
[23:02:58] *** illissius- is now known as illissius[sleep]
[23:03:08] <w-ber> -g means include debugging symbols
[23:03:21] <@mmu_man> -g is include debug syms
[23:04:36] <tqh> well it is a debugbuild, and nothing interesting in makefile so I'll look at build log
[23:06:40] <tqh> it's built with -g
[23:07:35] <fyysik> Skype rules. Only reason to boot Windoze at home
[23:08:41] *** axeld has joined #haiku
[23:08:41] *** ChanServ sets mode: +o axeld
[23:09:13] <tqh> mmu_man can I use objdump to lookup an  #File libwidget_beos.so text + 0003adaa adress easily?
[23:09:36] <@mmu_man> use nm
[23:10:10] <@mmu_man> nm libwidget_beos.so | sort | more '+/03adaa'
[23:10:50] <@axeld> tqh: or better forget about the last "aa" because "nm" only shows the beginning of symbols
[23:10:53] <@mmu_man> http://clapcrest.free.fr/revol/beos/shot_BeOS_powah_004.png
[23:10:57] <@mmu_man> hey axeld
[23:11:02] <@axeld> Hi mmu_man
[23:11:23] <@mmu_man> you're in early :)
[23:11:36] <@axeld> mmu_man: true enough :)
[23:11:42] * tqh cuts and pastes another one of mmu_mans nuggets to his Styledit text of wisdoms :)
[23:11:52] <fyysik> hi axeld
[23:11:56] <@axeld> hi fyysik!
[23:12:09] <tqh> and axeld's of course
[23:12:17] *** emwe has quit IRC
[23:12:39] * fyysik is in uncertain mood. Wife is away for next 1.5 months. maybe it is time to do some coding?
[23:13:10] <@axeld> fyysik: sounds good to me :)
[23:13:45] * fyysik thinks how to send also his daily work away. For 1.5 months:)
[23:13:58] <fyysik> and rent basement
[23:14:29] <@mmu_man> fyysik how about cloning virtual girl in beos ? :)
[23:14:49] <@mmu_man> an 17 22:14:49 <@mmu_man>	a "grown up" verion of Moe so to speak
[23:14:52] <fyysik> ghm, idea worth consideration
[23:16:59] <fyysik> axeld - actually. i'm almost ready - http://beos.spb.ru/fyysik/FyysiksCorner-SweetHome.jpg
[23:17:29] <YNOP> nice real to real
[23:17:34] <@axeld> Hey, there are almost as many computers as are here :)
[23:17:50] <@mmu_man> fyysik hmm is that a windoze screen I see on teh left ???
[23:18:01] <@axeld> fyysik: although the laptop on the left appears to run Windows :)
[23:18:05] <fyysik> axeld - one hidden is left from that picture and one in another room:)
[23:18:27] <fyysik> Skype, axeld - as i said, only reason to boot windoze at home for me:)
[23:18:43] <tqh> grr, I fix the seg. violation only to be left with automatic shutdown instead.
[23:19:11] * tqh wonders if nsWindow::Destroy does to much for new structure.
[23:19:22] <fyysik> and one of those laptops is "debug terminal":)
[23:20:03] <CIA-6> axeld * current/src/kernel/core/device_manager/ (probe.cpp Jamfile probe.c): Renamed probe.c to probe.cpp to prepare if for all the changes it got.
[23:20:10] <@axeld> fyysik: sounds like a good environment for kernel debugging :)
[23:20:14] * mmu_man runs away
[23:20:15] <fyysik> tqh - how dod you manage to get such interesting results?
[23:20:19] <@mmu_man> one more C++ file :-(
[23:20:38] <oco> :-)
[23:20:38] <@mmu_man> I heard Travis wasn't really fond of C++ in the kernel either :D
[23:20:48] <tqh> fyysik probably because nsWindow destroy almost always want window to quit.
[23:21:10] <tqh> and I rewrote StandardWindowCreate
[23:21:13] <fyysik> axeld - sure. Until i reanimated that old laptop i used (tried to use) ancient serial dot-matrix printer for that purpose, hehe
[23:21:54] <@axeld> mmu_man: who cares who is. C++ is *ways* better than C
[23:21:59] <fyysik> tqh - ahh, that's nice you reached :Destroy() code now. Looks weird for me
[23:22:10] <@axeld> mmu_man: and you know, I only do this do annoy you :-))
[23:22:37] <oco> maybe i can contribute some pascal sources :-)
[23:22:37] <@axeld> mmu_man: Just because I like you so much :-)
[23:22:53] <@mmu_man> it's an honour  :))
[23:23:00] <tqh> fyysik I am not sure I do I'm just guessing it just quit with a 'Kill Thread' in terminal no info at all
[23:23:32] <tqh> and bdb won't work for me. Having obj dir seperate from src seems to be bad for that.
[23:23:36] <fyysik> tqh - when you did chkbfs last time?
[23:23:40] <CIA-6> axeld * current/src/kernel/core/device_manager/probe.cpp: Fixed compilation issues that came with changing it to be C++ (hi mmu_man :-)).
[23:24:04] <@axeld> mmu_man: and btw, the evil factor of C++ is also much greater than the one of C. Why don't you like it more? :-))
[23:24:25] <Procton> tqh: bone?
[23:24:27] <tqh> fyysik: Long time ago.
[23:24:32] <tqh> Procton: yes
[23:24:43] <Procton> tqh: afaik bdb is broken in bone systems.
[23:24:48] <@axeld> Hey Procton
[23:24:52] <fyysik> tqh...time to do it for both partititions, it seems:)
[23:24:54] <Procton> hey axeld.
[23:24:55] <tqh> nope I've used it a lot.
[23:25:00] <Procton> you have?
[23:25:02] <@axeld> The latest one on BeBits should work
[23:25:07] <Procton> I've never managed it to run.
[23:25:19] <Procton> atleast not with proper info.
[23:25:43] <tqh> yes. just make sure your library and addon path is set up then run bdb app
[23:26:00] <@mmu_man> find current -name '*.cpp' -exec mv {} '`echo '{}' | sed 's/cpp$/c/'`' \;
[23:26:05] <Procton> odd.
[23:26:09] <@mmu_man> axeld it just feeld unright
[23:26:23] <@mmu_man> when I'm in kernel space I want to be in control
[23:26:26] <@axeld> mmu_man: Oh yeah, that it is :-)
[23:26:32] <@mmu_man> I want to feel the flow of control
[23:26:35] <tqh> I've used it several times for firefox-bin so it must work quite well :)
[23:26:37] <@mmu_man> doesn't wotrk with OO
[23:27:03] <@mmu_man> like the blood in your veins :)
[23:27:12] <@axeld> mmu_man: You have total control. You decide where to hide the obvious - to make the code clearer and less error prone
[23:27:27] <@mmu_man> I guess I should try
[23:27:36] <Procton> axeld: you can't do that with C?
[23:27:38] <@mmu_man> but keep the aspirine tube around
[23:28:29] <@axeld> Procton: not as good as. For example, with C++ you can easily allocate temporary objects using a helper object on the stack - it will be automatically deleted for you once you leave the stack frame
[23:28:44] <@axeld> Procton: And simplifications like these help a lot
[23:28:47] *** dipp has quit IRC
[23:29:01] <Procton> axeld: true.. hmm...
[23:29:22] <tqh> Procton Debugging firefox http://europe.dnsalias.net/hanging-threads.png
[23:30:11] <Procton> tqh: interesting... I suppose I should set up bdb then. I was told is was broken a long time ago, so I never bothered.
[23:30:35] *** thaflo has quit IRC
[23:30:58] <tqh> Well, it doesn't work today for me. But then again seems much doesn't work with the new firefox building instructions.
[23:31:33] <tqh> ofc it's nicer to debug debug-builds though.
[23:35:55] <CIA-6> axeld * current/src/kernel/libroot/posix/arch/x86/setjmp.c: Fixed compilation in case COMPILE_FOR_R5 is defined.
[23:37:12] <NathanW> axeld: On my system, the build complains about not knowing what kernel.so is
[23:37:22] <NathanW> Any idea what's up with that?
[23:37:25] <@mmu_man> must link with _KERNEL_ :)
[23:37:27] <@axeld> NathanW: since when?
[23:37:40] <@axeld> NathanW: since always or recently?
[23:37:40] <@mmu_man> (and not __KERNEL__ either as some ppl tried elsewhere =)
[23:37:43] <NathanW> an 17 22:37:43 <NathanW>	a few weeks, at the least
[23:37:51] <NathanW> It worked at the beginning of December
[23:38:08] <@axeld> NathanW: good, then this should be fixed since today or yesterday - can you confirm?
[23:38:12] <NathanW> the next time I tried building the kernel was 2 weeks ago, when it gave me that error
[23:38:13] <NathanW> sure
[23:38:20] <NathanW> let me sync the tree
[23:38:24] <@axeld> NathanW: thanks!
[23:38:29] <NathanW> may take a while :)
[23:42:24] <@mmu_man> axeld btw does our vfs use linux-like dentry stuff or just inodes as in-mem structs ?
[23:42:59] <@mmu_man> I do'nt think dentries would mix with out-of-namespace thigns like indices, attr dirs anyway
[23:43:04] <@axeld> mmu_man: the VFS itself has no dentry stuff yet
[23:43:09] <@mmu_man> yet ?
[23:43:32] <@axeld> mmu_man: we might introduce *partial* dentries - like remembering the parent directory, etc.
[23:43:35] <@mmu_man> not sure what it would bring in
[23:43:42] <@mmu_man> ah yeah well
[23:43:54] <@mmu_man> in Zeta I added a parent_vnid field to the vnode
[23:43:59] * tqh might have fixed instability
[23:44:04] <@mmu_man> tried in the ofile
[23:44:04] <@axeld> But we won't copy the name for now
[23:44:16] <@mmu_man> but it was problematic to use
[23:44:20] <@mmu_man> and maintain
[23:44:22] <@axeld> mmu_man: hm, so you've broken all hard-link supporting file systems? :)
[23:44:30] <@mmu_man> I wanted to minimize the impact on notify_listener
[23:44:35] <@mmu_man> axeld no
[23:44:41] <@mmu_man> it's just a hint for lsof
[23:44:45] <@mmu_man> it's not used elsewhere
[23:44:49] <@axeld> ah, okay
[23:44:57] <@mmu_man> so just if you use lsof you might get the wrong link
[23:44:59] <@mmu_man> that's all
[23:45:08] <@mmu_man> well first I mist fidn a way to not crash :)
[23:45:20] <@mmu_man> I access stuff under ints off
[23:45:23] * tqh goes to sleep
[23:45:29] <@mmu_man> but other stuff need to be locked
[23:45:31] <@axeld> night tqh
[23:45:38] *** lordcoxis has quit IRC
[23:45:43] <@BGA> axeld!
[23:45:46] <tqh> good night
[23:45:47] <@axeld> Hey BGA!
[23:45:52] *** tqh has quit IRC
[23:47:49] *** BGA_ has joined #haiku
[23:47:54] *** BGA has quit IRC
[23:48:24] *** brennanOS has joined #haiku





top