T2 IRC Log: 2005-01-29
This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.
--- Log opened Sat Jan 29 00:00:46 2005
01:12 -!- nzg [~tschmidt@xchangecenter.de] has quit [Read error: 54 (Connection reset by peer)]
01:16 -!- nzg [~tschmidt@xchangecenter.de] has joined #t2
02:06 -!- ooo_ [~o@dsl-olugw3p33.dial.inet.fi] has joined #t2
02:06 -!- ooo [~o@dsl-olugw3p33.dial.inet.fi] has quit [Read error: 54 (Connection reset by peer)]
06:15 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has joined #t2
06:18 < rxr> moin
06:18 < _Ragnar_> moin moin
06:18 -!- ideal [~idealm@211.100.227.107] has joined #t2
06:18 < rxr> hi ideal
06:18 < ideal> Hi, rxr:)
06:19 < ideal> I'm dealing with WindowMaker:)
06:22 < CIA-10> rene * r5829 /tags/2.1.0-beta3/package/network/howl/howl.desc: * updated howl (0.9.8 -> 0.9.10)
06:33 < rxr> ehrm - I have a mail in my inbox from a person @ novell providing a patch for t2 ?!!
06:33 < _Ragnar_> nice;)
06:35 < Postal|Away> Hey
06:35 < Postal|Away> rxr you around?
06:38 < rxr> yes
06:38 < rxr> _Ragnar_: not that we really need that patch - but I'll apply it anyway ;-)
06:38 < _Ragnar_> *lol*
06:39 < CIA-10> rene * r5830 /tags/2.1.0-beta3/package/network/howl/howl-0.9.8-darwin-locking-api.patch:
06:39 < CIA-10> Hans Petter Jansson:
06:39 < CIA-10> * added a howl API conformance patch for MacOSX
06:39 < rxr> _Ragnar_: novell names in the commit stats ;-)
06:40 < _Ragnar_> =)
06:42 < Postal|Away> rxr, How do you feel about me creating a 32 bit tool chain for the compilation of grub in 64 bit?
06:42 < CIA-10> rene * r5831 /trunk/package/network/howl/ (howl-0.9.8-darwin-locking-api.patch howl.desc):
06:42 < CIA-10> * merged r5828:5829 from 2.1.0-beta3 into trunk: howl update and API
06:42 < CIA-10> patch
06:43 < rxr> Postal|Away: you only need to add one more package glibc32 if we would like to have that ...
06:44 < rxr> Postal|Away: are you in your t2 right now?
06:44 < Postal|Away> rxr, hmmm... but the 32 bit toolchain does not exist... so dont you also need the 32 bit binutils, and gcc..?
06:44 < Postal|Away> rxr yes
06:44 < rxr> no - the gcc and binutils are biarch - with -m32 they yield 32bit binaries ...
06:44 < Postal|Away> and im having a problem with ./scripts/Emerge-Pkg..
06:44 < rxr> the only thing is that grub' configure and Makefiles are not 64bit safe and brain dead
06:45 < rxr> this is why they need the 32bit C library
06:45 < Postal|Away> rxr, i noticed that
06:45 < rxr> one could also fix the grub buid system
06:45 < rxr> but I also do not feel like that ...
06:45 < _Ragnar_> *g*
06:45 < rxr> Postal|Away: I could add glibc32 ...
06:46 < _Ragnar_> why not use dietlibc for that?
06:46 < rxr> because it would be equally hacky to get grub use that or always force grub to use it ...
06:47 < rxr> a 32bit libc is hany anyway
06:47 < rxr> and I need it on sparc64 for the silo bootload (wonder wonder) anyway
06:47 < rxr> bootload_er_ ;-)
06:47 < _Ragnar_> *g*
06:47 < _Ragnar_> as you wish :)
06:48 < Postal|Away> rxr, Why would emerge system come back with the messege "Packages Scheduled to Build:"... but then exit...
06:48 < rxr> because they are already up-to-date
06:48 < rxr> try adding -debug
06:48 < rxr> it will list what it does then ...
06:49 < Postal|Away> ok, its working now...
06:49 < Postal|Away> never mind then
06:50 < Postal|Away> so about this glibc32
06:51 < Postal|Away> Oh and by the way, when i ran the Update system script, it failed on gcc, because the C compiler could not compile C programs...
06:51 < rxr> how long do you intend to stay up tonight to test a possible glibc32 injection from me ?
06:52 < Postal|Away> yes, i will stay up as long as you need me ;-)
06:52 < rxr> .oO
06:52 < rxr> what is it over there 0:52 ?
06:52 < Postal|Away> rxr si
06:52 < rxr> it is 6:52 here and I just got up - may I first get breakfast ?
06:52 < _Ragnar_> no!:)
06:53 < Postal|Away> rxr, go ahead... btw good to see you... havent seen you in a long time... I have some things for you...
06:53 < Postal|Away> I have been testing... and I also added some areas to the Wiki...
06:53 < rxr> some things means work, doesn't it ?
06:53 < rxr> yes - I noticed it in the wiki ;-)
06:54 < Postal|Away> rxr, i hope its ok that I added those things... i wasnt sure, but they seemed useful..
06:54 < Postal|Away> and yes things means work...
06:55 < rxr> sure - the wiki is for public edit
06:55 < rxr> we can revert single stuff when e.g. illegal links or so are injected, though ...
06:55 < rxr> ok - feel free to post TODO items here
06:55 < rxr> I go to the bakery now and get some food - and after that I craete a glibc32 package
06:56 < Postal|Away> There is a syntax error on line 4 in gcc "unexpected token near line 4 GCC_WRAPPER_INSERT
06:57 < Postal|Away> and it fails because the "C preprosessor fails the sanity check"
06:57 < rxr> where to you get the "near line 4" ?
06:57 < rxr> e.g. for which package build or so?
06:57 < Postal|Away> I would copy the tail of the log here, but i cannot paste.... its for 9-gcc
06:59 < Postal|Away> rxr go to the bakery, and eat, well talk when youve got some food in you ;-)
06:59 < rxr> hm
06:59 < rxr> interesting
06:59 < rxr> can you mail it later ?
07:01 < Postal|Away> rxr, no cause that I know of i have no mail client...
07:01 < Postal|Away> rxr, I have no X either... so im in text only
07:02 < rxr> ouhm
07:02 < Postal|Away> rxr, I am getting very accustomed to text based computing
07:02 < rxr> why no X ?
07:02 < Postal|Away> rxr cause it wont start
07:02 < Postal|Away> It wont detect my screens...
07:02 < Postal|Away> its says i have none... and i configured it about 3 times
07:03 < rxr> ok - we work on your X after my breakfast
07:09 -!- Postal [~root@ool-43562131.dyn.optonline.net] has joined #t2
07:09 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]
07:10 < Postal> Hey
07:10 < Postal> rxr, you still here or are you off at the bakery?
07:12 -!- Postal [~root@ool-43562131.dyn.optonline.net] has left #t2 []
07:12 -!- Postal [~root@ool-43562131.dyn.optonline.net] has joined #t2
07:12 < Postal> Ok hopefully this will work now, i hate this irc client
07:13 < CIA-10> rene * r5832 /trunk/package/graphic/pspresent/pspresent.desc: * updated pspresent (1.1 -> 1.2)
07:13 < CIA-10> rene * r5833 /trunk/package/graphic/magicpoint/magicpoint.desc: * updated magicpoint (1.11a -> 1.11b)
07:14 < rxr> so -now I have the packages for the university µC presentation I annoyingly have to do later this weekend - and can go to the bakery ,-)
07:15 < Postal> rxr, ok gimmie a jingle when you get back i am going to try and post these .err logs on the wiki...
07:20 < Postal> ok, the relevant portion is in the Wiki, in the 64bit section under error logs...
07:26 < rxr> I'll do basic glibc32 hacking onw ...
07:26 < rxr> so I can gr
07:26 < rxr> grab eggs on my way out, too ...
07:26 < rxr> -3.3°C here ...
07:26 < rxr> bibber
07:27 < Postal> rxr, why would gcc configure be reporting that the cpp fails the sanity check?
07:27 < CIA-10> rene * r5834 /trunk/package/base/glibc32/: * creating a glibc32 package part I (glibc copy)
07:29 < rxr> Postal: I would need to look into the configure, too to see why configure does there ...
07:30 < rxr> Postal: you are at svn:HEAD ?
07:31 < Postal> rxr, what is svn:HEAD?
07:32 < rxr> the top - newest revision of the svn tree ...
07:32 < Postal> rxr, oh ic what you mean, im pretty recent, i co, a new tree about 30 min ago
07:32 < rxr> ok
07:32 < Postal> rxr, can i make it tell me what revison im at?
07:33 < rxr> svn fino
07:33 < rxr> info even
07:33 < rxr> but no need
07:33 < rxr> I just wanted to know if you are in trunk or switched to tags/2.1.0-beta3 or so due to my announced trunk architecture reorganization ...
07:34 < Postal> its 5828, just in case your intersted
07:34 < Postal> rxr, i never saw that announcment, but i svn up once or twice a day so..
07:34 < rxr> sent it to the mailing list ...
07:34 < rxr> . . .
07:34 < Postal> rxr im not on the mailing list
07:35 < rxr> Postal: you _should_ be ;-)
07:35 < rxr> yeah - I seem to have a workign 64bit svn now ;-)
07:35 < Postal> rxr, if i had email and X i might be
07:35 < Postal> svn never had problems in my build...
07:36 < Postal> what you do have is a 64bit tester with no life, who also happens to be a perfectionist....
07:36 < _Ragnar_> =)
07:37 < Postal> But i suppose in this world those are good qualities...
07:42 < Postal> My GeForce6800 is too new, even nvidia, has only released drivers up to the 6600, but the was in November, I am hoping for a new release soon...
07:44 < Postal> _Ragnar_ how do i go about installing the kernel source tree in T2?
07:45 < _Ragnar_> huh?
07:46 < _Ragnar_> dl kernel, unpack ...
07:46 < rxr> Postal: it is not installed by the build system
07:46 < rxr> but the few patches we apply life in /usr/src/patches-$ver or so ..
07:47 < Postal> rxr, ok, but I need the full source, for the nvidia drivers...
07:47 < rxr> Postal: the nvidia driver is automatically compiled when you build t2 ...
07:47 < rxr> and compiling it manually is annoying anyway ...
07:48 < Postal> rxr, i dont think that it is in the minimal build
07:48 < rxr> what target have you currently selected? desktop ?
07:48 < rxr> no - not in the minimal
07:48 < rxr> but you could use t2 to rebuild a kernel with nvidia driver
07:48 < Postal> I am
07:48 < Postal> ;-)
07:49 < rxr> select destop
07:49 < rxr> desktop even
07:49 < rxr> o package preselection template
07:49 < rxr> download all the requirements
07:49 < rxr> and do BUild-Pkg linux26
07:49 < rxr> it will build the nvidia module ...
07:49 < Postal> I just use Emerge-Pkg... it dl and installes it automiticly...
07:49 < rxr> and then Build-Pkg nvidia
07:49 < rxr> that installs the X module ...
07:50 < Postal> rxr oh... ok, well it failed...
07:50 < rxr> well - Emerge is ok - but currently it has the drawback that it will not download the linux add on packages ...
07:50 < rxr> I need to think about how to add this "download extending packages"
07:50 < rxr> what did fail ?
07:50 < Postal> rxr, the nvidia build
07:51 < rxr> what is the error ?
07:51 < Postal> I cannot cut and paste... but it is...
07:52 < Postal> hook_eval62... and so on and so forth... but if i could cut and paste, that would be nice, prehaps we should concentrate on X
07:53 < rxr> are you really sure the open source nvidia driver can not handle your card ?
07:54 < Postal> rxr, yes it is the 6800... 256DDR3...
07:54 < Postal> ;-)
07:55 < _Ragnar_> you need the very latest version of the nvidia driver for amd64
07:56 < Postal> These are the specs on my box: AMD64 3400, 3GB ram, Chaintek motherboard.., Nvidia GeForce 6800, Soundblaster Audigy 2 sound card, Linksys Gigabyte ehternet card, CD-RW 52X, and DVD-RW +\- 8x
07:57 < rxr> the nvidia package is x86 only right now
07:57 < Postal> rxr, no its not
07:57 < rxr> jsaw: wanted to add the x86-64 binary but has not done so far ...
07:57 < Postal> www.nvidia.com... they have an IA64 and an AMD64 version
07:57 < rxr> I know
07:57 < rxr> I was talking about the t2 package ...
07:58 < Postal> rxr, well, any ideas on why X reports i have no screens?
07:58 < Postal> X works fine in any distro...
07:58 < Postal> so.. prehaps the config file is wrong, ill check it now..
08:00 < rxr> what do you mean with X works fine in any distro ?
08:00 < Postal> I mean in other distros..
08:01 < Postal> rxr, but the config file lists my screen, prehaps X is reading the wrong config...
08:01 < rxr> take a look into /var/log/Xorg.0.log
08:01 < rxr> when you say the nv driver does not suppor the card and you have not yet the nvidia one, ...
08:02 < rxr> it is most probably that youre card is not detected
08:02 < rxr> I add the x86-64 nvidia binary junk right now ?
08:02 < rxr> s/?/!/
08:02 < Postal> you could, it would be nice, and ill look into the log file above
08:05 < Postal> Ok, its not detecting the Card...
08:05 < rxr> what driver have you in your config?
08:05 < rxr> nv ?
08:05 < Postal> yes nv
08:06 < Postal> btw, does it matter that my monitor is Digital as opposed to analog?
08:07 < rxr> should not
08:07 < rxr> the driver /should/ handle that ..
08:07 < rxr> especially the binary only nvidia one ...
08:07 < rxr> so - in some secs you have nvidia x86-64 part I
08:09 < Postal> rxr, ok, well the driver from nvida, supports the card one below mine... the 6600, so i am thinking, it will not be perfect, with the 6800, but it should at least make it work..
08:10 < CIA-10> rene * r5835 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc): * added the x86-64 binary to the nvidia package - untested
08:10 < rxr> Postal: can you try Emerge-Pkg nvidia after an svn up ?
08:10 < Postal> yeah ok
08:11 < Postal> Ok its building now
08:12 < Postal> It wont build..
08:12 < rxr> error ?
08:12 < Postal> give me a sec, I have to type it by hand
08:13 < Postal> hook_eval:62 (last $?=0) eval 'unset hookdrity_premature...
08:14 < Postal> is the first line of the error..
08:14 < rxr> does not look like the error
08:15 < Postal> well... if i could copy and paste then i could give you the error, is there a good text email client, ill email it to you..?
08:15 < rxr> there is gpm
08:15 < rxr> which allows cut'n paste on the console ...
08:15 < rxr> and there are mutt, pine and so on as text mail clients ...
08:15 < rxr> but wait
08:15 < Postal> ok
08:15 < rxr> is there this in the log :
08:15 < rxr> No binary for this architecture detected!
08:16 < Postal> no, but it does mention the fact that there was not target or makefile specified..
08:17 < Postal> That sounds pretty important...
08:17 < CIA-10> rene * r5836 /trunk/package/x11/nvidia/ (nvidia.conf postlinux.conf):
08:17 < CIA-10> * fixed the nvidia match_source_file pattern and added the file
08:17 < CIA-10> selection to the postlinux.conf
08:17 < rxr> could you retry ?
08:18 < Postal> of course
08:18 < Postal> same error, no targets specified, and no makefile found
08:19 < Postal> that is the only "make: ***" type error it gives...
08:19 < rxr> .oO
08:19 < rxr> momemnt
08:19 < Postal> why dont we just install the kernel source tree, and ill build it by hand...?
08:20 < rxr> ok - now
08:20 < CIA-10> rene * r5837 /trunk/package/x11/nvidia/nvidia.conf:
08:20 < CIA-10> * readded the accidently removed custmain="main" to the nvidia
08:20 < CIA-10> package ;-)
08:20 < rxr> because we want it in t2, right ?
08:21 < rxr> this is also not yet the kernel driver - but the X one
08:21 < Postal> rxr, ok ill try again now
08:24 < Postal> rxr, /t2-trunk/package/x11/nvidia/nvidia.conf: line 11: cd: NVIDIA-Linux-*-pkg1: no such file or directory
08:24 < rxr> no error yet ? *g*
08:24 < rxr> ouhm ...
08:24 < rxr> ls src.nvidia..../
08:25 < rxr> ah - ic
08:25 < Postal> it says no such file or directory when i ls src.nvidia.110696
08:25 < rxr> why the h*ll is the x86-64 driver named NFORCE when the others are NVIDIA ... :-(
08:26 < Postal> rxr, because you downloaded the nforce driver for X, as opposed to the "nvidia" which is the kernel driver....
08:27 < rxr> so far they always included both of the m...
08:27 < rxr> I guess on x86-64 they only support recent cars so people buy the recent ones - and not place some older one into new hardware ...
08:28 < Postal> well, becuase if you have a 64bit box, you already spent a good deal of money, so you will probably buy a newer card...
08:28 * rxr would not ....
08:28 < rxr> especially never a nvidia card ....
08:29 < CIA-10> rene * r5838 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc postlinux.conf):
08:29 < CIA-10> * added the ia64 binary driver to the nvidia package and fixed the
08:29 < CIA-10> globbing pattern to work with the annoying x86-64 binary ...
08:29 < rxr> ok - you could give that a try
08:29 < Postal> I wanted the ATI but they were out of stock for about 3 months, so i would have had a box with no card, and im trying now..
08:29 < Postal> you could have let me know i had to dl it...
08:30 < rxr> ah - sorry
08:30 < rxr> moment
08:30 < Postal> did you forget a symlink?
08:30 < rxr> no - grabbed the wrong file from the nvidia homepage
08:31 < Postal> what is with the symlinks you just added... they were not there before..
08:31 < rxr> what symlinks ?
08:32 < Postal> never mind, its not important, but in the begining of the build it creates some symbolic links to somthing, i cannot tell what it is cause it is cut off with "..." cause my resolution is like 800x600 right now... and it is usually 1280x1024
08:33 < CIA-10> rene * r5839 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc postlinux.conf):
08:33 < CIA-10> * fixed to download the right file for the x86-64 nvidia package and
08:33 < CIA-10> fixed the globbing pattern again - sigh
08:33 < Postal> ok here we go--- again
08:33 < rxr> so - this /should/ work now
08:34 < rxr> (if the nvidia people did not do more stupid stuff in the package ...)
08:34 < rxr> so - bakery ...
08:35 < Postal> I have to dl the package, and i dont want to use emerge, cause it will compile X also you said so how do i dl it?
08:35 < rxr> Download nvidia
08:35 < rxr> ./scripts/Download nvidia
08:35 < rxr> that is
08:35 < rxr> Emerge will not build X
08:35 < rxr> just Emerge nvidia will work, too
08:35 < Postal> oh and btw, i would like to have my kernel source also...
08:35 < rxr> cu in some minutes
08:36 < Postal> ok cu in a few
08:36 < rxr> well - the kernel is in
08:36 < rxr> download/mirror/l/linux-2.6.10.tar.bz2 or so
08:36 < rxr> you can untar it to /usr/src
08:36 < rxr> the patched applied to the kernel by t2 are already there
08:36 < Postal> ok
08:36 < Postal> i will build the kernel nvidia also..?
08:36 < rxr> /usr/src/linux26-patches/
08:36 < rxr> yes
08:37 < rxr> if the X part installed while building the nvidia pacakge
08:37 < Postal> it built, now what do ?
08:37 < rxr> make sure desktop without package selection template is selected
08:37 < rxr> and Dowload -required
08:37 < rxr> so all the stuff needed to build is there
08:37 < rxr> and then
08:37 < rxr> Build-Pkg linux26
08:38 < rxr> what should include buliding the nvidia kernel module
08:38 < rxr> (by the t2 automatics)
08:38 < rxr> so - cu ;-)
08:39 < _Ragnar_> cu
08:40 < Postal> _Ragnar_ you can scroll up, can you give me the command to untar the source tree, because i cannot scroll up in centericq...
08:41 < Postal> oh wait never mind yes i can
08:41 < _Ragnar_> ;)
09:01 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has joined #t2
09:01 -!- Postal [~root@ool-43562131.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]
09:02 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has left #t2 []
09:02 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has joined #t2
09:03 < rxr> re
09:03 < rxr> so - breakfast
09:04 < Postal1> rxr, i got X to start, using the standard vga driver, but it froze... i had to reboot, i could not figure out how to make X use the new nvidia driver...
09:09 < rxr> your first need to build the kernel with the nvidia module
09:09 < rxr> have you done that ?
09:12 < Postal1> no, cause i untard the kernel source, and when nvidia found it, it complianed, hold on and ill get you the error
09:12 < rxr> that it does not match your running kernel I guess ...
09:12 < rxr> Postal1: just use Build-Pkg linux26
09:13 < rxr> it will build the nvidia module with the kernel
09:13 < rxr> and then we also know that the t2 package does now work on x86-64 ...
09:18 -!- rxr_ [~rene@p213.54.197.164.tisdip.tiscali.de] has joined #t2
09:18 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
09:18 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
09:18 [Users #t2]
09:18 [ _Ragnar_] [ ideal ] [ nzg ] [ praenti] [ sparc-kly]
09:18 [ CIA-10 ] [ jsaw ] [ ooo_ ] [ rxr ] [ valentin ]
09:18 [ daja77 ] [ mnemoc] [ Postal|Away] [ rxr_ ]
09:18 -!- Irssi: #t2: Total of 14 nicks [0 ops, 0 halfops, 0 voices, 14 normal]
09:18 -!- [freenode-info] please register your nickname...don't forget to auto-identify! http://freenode.net/faq.shtml#nicksetup
09:18 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
09:18 -!- Irssi: Join to #t2 was synced in 11 secs
09:18 < rxr_> re
09:22 -!- rxr [~rene@213.54.206.111] has quit [Read error: 60 (Operation timed out)]
09:28 -!- You're now known as rxr
09:44 < rxr> strange - sometimes sparc64 svn does illegal instruction here:
09:44 < rxr> #0 0xfffff80000c101a0 in ne_uri_parse
09:44 < rxr> and sometimes not ...
10:10 -!- mnemoc [~amery@200.75.27.25] has quit [Read error: 110 (Connection timed out)]
10:16 < CIA-10> rene * r5840 /trunk/package/base/glibc/glibc.conf: * cleaned glibc.conf
10:36 < rxr> I just found another rock inherited hack that made our 64bit life harder
10:37 < rxr> unbelievable what was injected into ROCK ages ago ...
10:38 < rxr> ouhm -4.4°C already
10:58 < CIA-10> rene * r5841 /trunk/package/base/glibc32/ (33 files): * first glibc32 draft - untested
11:06 < _Ragnar_> goodnight folks
11:07 < CIA-10> rene * r5842 /trunk/package/network/libidn/libidn.desc: * updated libidn (0.5.12 -> 0.5.13)
11:14 -!- ideal [~idealm@211.100.227.107] has quit ["Leaving"]
13:12 < rxr> hm
13:13 < rxr> on my sparc64 the illegal instruction inside svn is this:
13:13 < rxr> call 0x154
13:13 < rxr> during executing this somehow get's corrupted ...
13:14 < rxr> yep - it also is in the installed /usr/lib/libneon.a
13:14 < rxr> maybe the linker is unable link it properly ...
13:16 < rxr> hm - moment
13:16 < rxr> maybe this is strange:
13:16 < rxr> 0x0000000000000154: call 0x154
13:16 < rxr> does it call itself ?
13:18 < rxr> hm - ack - it is no runtime corruption
13:18 < rxr> it is also an illegal instruction after loading the svn binary into memory before main ...
13:18 < rxr> 0xfffff80000c181a0: unknown
13:18 < rxr> gdb does work mostly on sparc64 - which is good to know ....!
13:26 < rxr> hm - the gcc assembly generats is this:
13:26 < rxr> call strstr, 0
14:18 < rxr> I think -fPIC helped it ..
14:18 < rxr> yeah
14:19 < rxr> after one dive thru gcc -> assembly -> relocaton and gdb svn works on sparc64 !!!!!!!!!
14:22 < CIA-10> rene * r5843 /trunk/package/network/neon/neon.conf:
14:22 < CIA-10> * fixed neon to build a shared library by default - if static inject
14:22 < CIA-10> -fPIC - both ways help sparc64 to not illegal instructon due to
14:22 < CIA-10> not relocated call ;-) !!!
14:22 < rxr> ^- committed from sparc64 !!!
14:33 < CIA-10> lars * r5844 /trunk/package/powerpc/pbbuttonsd/ (pbbuttonsd.conf pbbuttonsd.desc):
14:33 < CIA-10> * updated pbbuttonsd (0.6.6 -> 0.6.7) (tested)
14:33 < CIA-10> * fixed /etc/pbbuttonds.conf to use ALSA instead of OSS
15:09 < CIA-10> rene * r5845 /trunk/scripts/Emerge-Pkg:
15:09 < CIA-10> * improved Emerge-Pkg to always check if a build is needed and
15:09 < CIA-10> added a -force option to Emerge-Pkg to force builds
15:13 < CIA-10> rene * r5846 /trunk/architecture/ (sparc/gcc-options sparc64/gcc-options): * removed -mno-app-regs from the sparc*/gcc-options files
15:23 < CIA-10> rene * r5847 /trunk/package/base/glibc/glibc.conf:
15:23 < CIA-10> * fixed glibc to use -fcall-used-g6 for sparc64 globally (not yet sure
15:23 < CIA-10> if it makes such a difference - but it is that Fedora does ...)
16:00 < CIA-10> rene * r5848 /trunk/package/base/procps/64bit-size_t.patch:
16:00 < CIA-10> * fixed procps abort due to SIGBUS on at least sparc64 (maybe other
16:00 < CIA-10> 64bit architectures, too)
17:00 < rxr> crazzy
17:00 < rxr> gcc on sparc64 optimizes a timeval memcpy into:
17:00 < rxr> add %l1, 8, %l0
17:00 < rxr> ldx [%l0+8], %g2
17:00 < rxr> ldx [%l1+8], %g1
17:00 < rxr> sethi %hi(.LLC50), %o0
17:00 < rxr> stx %g1, [%fp+1679]
17:00 < rxr> stx %g2, [%fp+1687]
17:01 < rxr> not bad
17:03 < valentin> how does that perform in MB/s in contrast to a non optimized single byte copy ?
17:03 < rxr> too bad the first load crashes ...
17:17 < rxr> well whole word load and stores are always way faster than sub word access on all archzes ...
17:21 < rxr> strange %l0 looks very much 8byte aligned ...
17:21 < rxr> 0x224c00
17:23 < rxr> well - sorry
17:23 < rxr> 0x204c04
17:29 < rxr> I might hit this bug:
17:29 < rxr> http://gcc.gnu.org/ml/gcc-patches/2003-11/msg02233.html
17:55 < rxr> 0x00000000001027ac: ld [ %l1 + 8 ], %g1
17:55 < rxr> 0x00000000001027b0: ld [ %l1 + 0xc ], %g2
17:55 < rxr> 0x00000000001027b4: ld [ %l1 + 0x10 ], %g3
17:55 < rxr> 0x00000000001027b8: ld [ %l1 + 0x14 ], %g4
17:55 < rxr> 0x00000000001027bc: st %g1, [ %fp + 0x68f ]
17:55 < rxr> 0x00000000001027c0: st %g2, [ %fp + 0x693 ]
17:55 < rxr> 0x00000000001027c4: st %g3, [ %fp + 0x697 ]
17:55 < rxr> 0x00000000001027c8: st %g4, [ %fp + 0x69b ]
18:01 < rxr> fixed without 64bit loads ;-)
18:01 < rxr> the next big task of fixing X is not for today ...
18:06 < rxr> root@sunrise:/t2-trunk# ping www.heise.de
18:06 < rxr> PING www.heise.de (193.99.144.71): 56 octets data
18:06 < rxr> 64 octets from 193.99.144.71: icmp_seq=0 ttl=247 time=104.0 ms
18:09 < CIA-10> rene * r5849 /trunk/package/network/netkit-base/64bit-alignment.patch: * fixed timeval cast in ping to not crash by SIGBUS on at least sparc64
18:14 < nzg> quite interesting... why use almost all german people www.heise.de as host for test-pings?
18:14 < nzg> moin btw
18:16 < rxr> hehe - ;-) used to "just works" ;-)
21:07 < rxr> n8 all
21:11 < _Ragnar_> gn8
21:14 -!- sparc-kly [~mubex@65-23-206-237.prtc.net] has left #t2 ["Leaving"]
21:58 -!- CIA-10 [~CIA@flapjack.navi.cx] has quit []
22:00 -!- CIA-temp267 [~CIA@flapjack.navi.cx] has joined #t2
22:00 -!- CIA-temp267 is now known as CIA-8
22:03 -!- Netsplit saberhagen.freenode.net <-> irc.freenode.net quits: daja77, jsaw, valentin, praenti, Postal|Away, _Ragnar_, nzg, ooo_
22:14 -!- rxr_ [~rene@p213.54.197.164.tisdip.tiscali.de] has joined #t2
22:14 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
22:14 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
22:14 [Users #t2]
22:14 [ _Ragnar_] [ daja77] [ nzg ] [ Postal|Away] [ rxr_ ]
22:14 [ CIA-8 ] [ jsaw ] [ ooo_] [ praenti ] [ valentin]
22:14 -!- Irssi: #t2: Total of 10 nicks [0 ops, 0 halfops, 0 voices, 10 normal]
22:14 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
22:14 -!- Irssi: Join to #t2 was synced in 11 secs
--- Log closed Sun Jan 30 00:00:46 2005
01:12 -!- nzg [~tschmidt@xchangecenter.de] has quit [Read error: 54 (Connection reset by peer)]
01:16 -!- nzg [~tschmidt@xchangecenter.de] has joined #t2
02:06 -!- ooo_ [~o@dsl-olugw3p33.dial.inet.fi] has joined #t2
02:06 -!- ooo [~o@dsl-olugw3p33.dial.inet.fi] has quit [Read error: 54 (Connection reset by peer)]
06:15 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has joined #t2
06:18 < rxr> moin
06:18 < _Ragnar_> moin moin
06:18 -!- ideal [~idealm@211.100.227.107] has joined #t2
06:18 < rxr> hi ideal
06:18 < ideal> Hi, rxr:)
06:19 < ideal> I'm dealing with WindowMaker:)
06:22 < CIA-10> rene * r5829 /tags/2.1.0-beta3/package/network/howl/howl.desc: * updated howl (0.9.8 -> 0.9.10)
06:33 < rxr> ehrm - I have a mail in my inbox from a person @ novell providing a patch for t2 ?!!
06:33 < _Ragnar_> nice;)
06:35 < Postal|Away> Hey
06:35 < Postal|Away> rxr you around?
06:38 < rxr> yes
06:38 < rxr> _Ragnar_: not that we really need that patch - but I'll apply it anyway ;-)
06:38 < _Ragnar_> *lol*
06:39 < CIA-10> rene * r5830 /tags/2.1.0-beta3/package/network/howl/howl-0.9.8-darwin-locking-api.patch:
06:39 < CIA-10> Hans Petter Jansson
06:39 < CIA-10> * added a howl API conformance patch for MacOSX
06:39 < rxr> _Ragnar_: novell names in the commit stats ;-)
06:40 < _Ragnar_> =)
06:42 < Postal|Away> rxr, How do you feel about me creating a 32 bit tool chain for the compilation of grub in 64 bit?
06:42 < CIA-10> rene * r5831 /trunk/package/network/howl/ (howl-0.9.8-darwin-locking-api.patch howl.desc):
06:42 < CIA-10> * merged r5828:5829 from 2.1.0-beta3 into trunk: howl update and API
06:42 < CIA-10> patch
06:43 < rxr> Postal|Away: you only need to add one more package glibc32 if we would like to have that ...
06:44 < rxr> Postal|Away: are you in your t2 right now?
06:44 < Postal|Away> rxr, hmmm... but the 32 bit toolchain does not exist... so dont you also need the 32 bit binutils, and gcc..?
06:44 < Postal|Away> rxr yes
06:44 < rxr> no - the gcc and binutils are biarch - with -m32 they yield 32bit binaries ...
06:44 < Postal|Away> and im having a problem with ./scripts/Emerge-Pkg..
06:44 < rxr> the only thing is that grub' configure and Makefiles are not 64bit safe and brain dead
06:45 < rxr> this is why they need the 32bit C library
06:45 < Postal|Away> rxr, i noticed that
06:45 < rxr> one could also fix the grub buid system
06:45 < rxr> but I also do not feel like that ...
06:45 < _Ragnar_> *g*
06:45 < rxr> Postal|Away: I could add glibc32 ...
06:46 < _Ragnar_> why not use dietlibc for that?
06:46 < rxr> because it would be equally hacky to get grub use that or always force grub to use it ...
06:47 < rxr> a 32bit libc is hany anyway
06:47 < rxr> and I need it on sparc64 for the silo bootload (wonder wonder) anyway
06:47 < rxr> bootload_er_ ;-)
06:47 < _Ragnar_> *g*
06:47 < _Ragnar_> as you wish :)
06:48 < Postal|Away> rxr, Why would emerge system come back with the messege "Packages Scheduled to Build:"... but then exit...
06:48 < rxr> because they are already up-to-date
06:48 < rxr> try adding -debug
06:48 < rxr> it will list what it does then ...
06:49 < Postal|Away> ok, its working now...
06:49 < Postal|Away> never mind then
06:50 < Postal|Away> so about this glibc32
06:51 < Postal|Away> Oh and by the way, when i ran the Update system script, it failed on gcc, because the C compiler could not compile C programs...
06:51 < rxr> how long do you intend to stay up tonight to test a possible glibc32 injection from me ?
06:52 < Postal|Away> yes, i will stay up as long as you need me ;-)
06:52 < rxr> .oO
06:52 < rxr> what is it over there 0:52 ?
06:52 < Postal|Away> rxr si
06:52 < rxr> it is 6:52 here and I just got up - may I first get breakfast ?
06:52 < _Ragnar_> no!:)
06:53 < Postal|Away> rxr, go ahead... btw good to see you... havent seen you in a long time... I have some things for you...
06:53 < Postal|Away> I have been testing... and I also added some areas to the Wiki...
06:53 < rxr> some things means work, doesn't it ?
06:53 < rxr> yes - I noticed it in the wiki ;-)
06:54 < Postal|Away> rxr, i hope its ok that I added those things... i wasnt sure, but they seemed useful..
06:54 < Postal|Away> and yes things means work...
06:55 < rxr> sure - the wiki is for public edit
06:55 < rxr> we can revert single stuff when e.g. illegal links or so are injected, though ...
06:55 < rxr> ok - feel free to post TODO items here
06:55 < rxr> I go to the bakery now and get some food - and after that I craete a glibc32 package
06:56 < Postal|Away> There is a syntax error on line 4 in gcc "unexpected token near line 4 GCC_WRAPPER_INSERT
06:57 < Postal|Away> and it fails because the "C preprosessor fails the sanity check"
06:57 < rxr> where to you get the "near line 4" ?
06:57 < rxr> e.g. for which package build or so?
06:57 < Postal|Away> I would copy the tail of the log here, but i cannot paste.... its for 9-gcc
06:59 < Postal|Away> rxr go to the bakery, and eat, well talk when youve got some food in you ;-)
06:59 < rxr> hm
06:59 < rxr> interesting
06:59 < rxr> can you mail it later ?
07:01 < Postal|Away> rxr, no cause that I know of i have no mail client...
07:01 < Postal|Away> rxr, I have no X either... so im in text only
07:02 < rxr> ouhm
07:02 < Postal|Away> rxr, I am getting very accustomed to text based computing
07:02 < rxr> why no X ?
07:02 < Postal|Away> rxr cause it wont start
07:02 < Postal|Away> It wont detect my screens...
07:02 < Postal|Away> its says i have none... and i configured it about 3 times
07:03 < rxr> ok - we work on your X after my breakfast
07:09 -!- Postal [~root@ool-43562131.dyn.optonline.net] has joined #t2
07:09 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]
07:10 < Postal> Hey
07:10 < Postal> rxr, you still here or are you off at the bakery?
07:12 -!- Postal [~root@ool-43562131.dyn.optonline.net] has left #t2 []
07:12 -!- Postal [~root@ool-43562131.dyn.optonline.net] has joined #t2
07:12 < Postal> Ok hopefully this will work now, i hate this irc client
07:13 < CIA-10> rene * r5832 /trunk/package/graphic/pspresent/pspresent.desc: * updated pspresent (1.1 -> 1.2)
07:13 < CIA-10> rene * r5833 /trunk/package/graphic/magicpoint/magicpoint.desc: * updated magicpoint (1.11a -> 1.11b)
07:14 < rxr> so -now I have the packages for the university µC presentation I annoyingly have to do later this weekend - and can go to the bakery ,-)
07:15 < Postal> rxr, ok gimmie a jingle when you get back i am going to try and post these .err logs on the wiki...
07:20 < Postal> ok, the relevant portion is in the Wiki, in the 64bit section under error logs...
07:26 < rxr> I'll do basic glibc32 hacking onw ...
07:26 < rxr> so I can gr
07:26 < rxr> grab eggs on my way out, too ...
07:26 < rxr> -3.3°C here ...
07:26 < rxr> bibber
07:27 < Postal> rxr, why would gcc configure be reporting that the cpp fails the sanity check?
07:27 < CIA-10> rene * r5834 /trunk/package/base/glibc32/: * creating a glibc32 package part I (glibc copy)
07:29 < rxr> Postal: I would need to look into the configure, too to see why configure does there ...
07:30 < rxr> Postal: you are at svn:HEAD ?
07:31 < Postal> rxr, what is svn:HEAD?
07:32 < rxr> the top - newest revision of the svn tree ...
07:32 < Postal> rxr, oh ic what you mean, im pretty recent, i co, a new tree about 30 min ago
07:32 < rxr> ok
07:32 < Postal> rxr, can i make it tell me what revison im at?
07:33 < rxr> svn fino
07:33 < rxr> info even
07:33 < rxr> but no need
07:33 < rxr> I just wanted to know if you are in trunk or switched to tags/2.1.0-beta3 or so due to my announced trunk architecture reorganization ...
07:34 < Postal> its 5828, just in case your intersted
07:34 < Postal> rxr, i never saw that announcment, but i svn up once or twice a day so..
07:34 < rxr> sent it to the mailing list ...
07:34 < rxr> . . .
07:34 < Postal> rxr im not on the mailing list
07:35 < rxr> Postal: you _should_ be ;-)
07:35 < rxr> yeah - I seem to have a workign 64bit svn now ;-)
07:35 < Postal> rxr, if i had email and X i might be
07:35 < Postal> svn never had problems in my build...
07:36 < Postal> what you do have is a 64bit tester with no life, who also happens to be a perfectionist....
07:36 < _Ragnar_> =)
07:37 < Postal> But i suppose in this world those are good qualities...
07:42 < Postal> My GeForce6800 is too new, even nvidia, has only released drivers up to the 6600, but the was in November, I am hoping for a new release soon...
07:44 < Postal> _Ragnar_ how do i go about installing the kernel source tree in T2?
07:45 < _Ragnar_> huh?
07:46 < _Ragnar_> dl kernel, unpack ...
07:46 < rxr> Postal: it is not installed by the build system
07:46 < rxr> but the few patches we apply life in /usr/src/patches-$ver or so ..
07:47 < Postal> rxr, ok, but I need the full source, for the nvidia drivers...
07:47 < rxr> Postal: the nvidia driver is automatically compiled when you build t2 ...
07:47 < rxr> and compiling it manually is annoying anyway ...
07:48 < Postal> rxr, i dont think that it is in the minimal build
07:48 < rxr> what target have you currently selected? desktop ?
07:48 < rxr> no - not in the minimal
07:48 < rxr> but you could use t2 to rebuild a kernel with nvidia driver
07:48 < Postal> I am
07:48 < Postal> ;-)
07:49 < rxr> select destop
07:49 < rxr> desktop even
07:49 < rxr> o package preselection template
07:49 < rxr> download all the requirements
07:49 < rxr> and do BUild-Pkg linux26
07:49 < rxr> it will build the nvidia module ...
07:49 < Postal> I just use Emerge-Pkg... it dl and installes it automiticly...
07:49 < rxr> and then Build-Pkg nvidia
07:49 < rxr> that installs the X module ...
07:50 < Postal> rxr oh... ok, well it failed...
07:50 < rxr> well - Emerge is ok - but currently it has the drawback that it will not download the linux add on packages ...
07:50 < rxr> I need to think about how to add this "download extending packages"
07:50 < rxr> what did fail ?
07:50 < Postal> rxr, the nvidia build
07:51 < rxr> what is the error ?
07:51 < Postal> I cannot cut and paste... but it is...
07:52 < Postal> hook_eval62... and so on and so forth... but if i could cut and paste, that would be nice, prehaps we should concentrate on X
07:53 < rxr> are you really sure the open source nvidia driver can not handle your card ?
07:54 < Postal> rxr, yes it is the 6800... 256DDR3...
07:54 < Postal> ;-)
07:55 < _Ragnar_> you need the very latest version of the nvidia driver for amd64
07:56 < Postal> These are the specs on my box: AMD64 3400, 3GB ram, Chaintek motherboard.., Nvidia GeForce 6800, Soundblaster Audigy 2 sound card, Linksys Gigabyte ehternet card, CD-RW 52X, and DVD-RW +\- 8x
07:57 < rxr> the nvidia package is x86 only right now
07:57 < Postal> rxr, no its not
07:57 < rxr> jsaw: wanted to add the x86-64 binary but has not done so far ...
07:57 < Postal> www.nvidia.com... they have an IA64 and an AMD64 version
07:57 < rxr> I know
07:57 < rxr> I was talking about the t2 package ...
07:58 < Postal> rxr, well, any ideas on why X reports i have no screens?
07:58 < Postal> X works fine in any distro...
07:58 < Postal> so.. prehaps the config file is wrong, ill check it now..
08:00 < rxr> what do you mean with X works fine in any distro ?
08:00 < Postal> I mean in other distros..
08:01 < Postal> rxr, but the config file lists my screen, prehaps X is reading the wrong config...
08:01 < rxr> take a look into /var/log/Xorg.0.log
08:01 < rxr> when you say the nv driver does not suppor the card and you have not yet the nvidia one, ...
08:02 < rxr> it is most probably that youre card is not detected
08:02 < rxr> I add the x86-64 nvidia binary junk right now ?
08:02 < rxr> s/?/!/
08:02 < Postal> you could, it would be nice, and ill look into the log file above
08:05 < Postal> Ok, its not detecting the Card...
08:05 < rxr> what driver have you in your config?
08:05 < rxr> nv ?
08:05 < Postal> yes nv
08:06 < Postal> btw, does it matter that my monitor is Digital as opposed to analog?
08:07 < rxr> should not
08:07 < rxr> the driver /should/ handle that ..
08:07 < rxr> especially the binary only nvidia one ...
08:07 < rxr> so - in some secs you have nvidia x86-64 part I
08:09 < Postal> rxr, ok, well the driver from nvida, supports the card one below mine... the 6600, so i am thinking, it will not be perfect, with the 6800, but it should at least make it work..
08:10 < CIA-10> rene * r5835 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc): * added the x86-64 binary to the nvidia package - untested
08:10 < rxr> Postal: can you try Emerge-Pkg nvidia after an svn up ?
08:10 < Postal> yeah ok
08:11 < Postal> Ok its building now
08:12 < Postal> It wont build..
08:12 < rxr> error ?
08:12 < Postal> give me a sec, I have to type it by hand
08:13 < Postal> hook_eval:62 (last $?=0) eval 'unset hookdrity_premature...
08:14 < Postal> is the first line of the error..
08:14 < rxr> does not look like the error
08:15 < Postal> well... if i could copy and paste then i could give you the error, is there a good text email client, ill email it to you..?
08:15 < rxr> there is gpm
08:15 < rxr> which allows cut'n paste on the console ...
08:15 < rxr> and there are mutt, pine and so on as text mail clients ...
08:15 < rxr> but wait
08:15 < Postal> ok
08:15 < rxr> is there this in the log :
08:15 < rxr> No binary for this architecture detected!
08:16 < Postal> no, but it does mention the fact that there was not target or makefile specified..
08:17 < Postal> That sounds pretty important...
08:17 < CIA-10> rene * r5836 /trunk/package/x11/nvidia/ (nvidia.conf postlinux.conf):
08:17 < CIA-10> * fixed the nvidia match_source_file pattern and added the file
08:17 < CIA-10> selection to the postlinux.conf
08:17 < rxr> could you retry ?
08:18 < Postal> of course
08:18 < Postal> same error, no targets specified, and no makefile found
08:19 < Postal> that is the only "make: ***" type error it gives...
08:19 < rxr> .oO
08:19 < rxr> momemnt
08:19 < Postal> why dont we just install the kernel source tree, and ill build it by hand...?
08:20 < rxr> ok - now
08:20 < CIA-10> rene * r5837 /trunk/package/x11/nvidia/nvidia.conf:
08:20 < CIA-10> * readded the accidently removed custmain="main" to the nvidia
08:20 < CIA-10> package ;-)
08:20 < rxr> because we want it in t2, right ?
08:21 < rxr> this is also not yet the kernel driver - but the X one
08:21 < Postal> rxr, ok ill try again now
08:24 < Postal> rxr, /t2-trunk/package/x11/nvidia/nvidia.conf: line 11: cd: NVIDIA-Linux-*-pkg1: no such file or directory
08:24 < rxr> no error yet ? *g*
08:24 < rxr> ouhm ...
08:24 < rxr> ls src.nvidia..../
08:25 < rxr> ah - ic
08:25 < Postal> it says no such file or directory when i ls src.nvidia.110696
08:25 < rxr> why the h*ll is the x86-64 driver named NFORCE when the others are NVIDIA ... :-(
08:26 < Postal> rxr, because you downloaded the nforce driver for X, as opposed to the "nvidia" which is the kernel driver....
08:27 < rxr> so far they always included both of the m...
08:27 < rxr> I guess on x86-64 they only support recent cars so people buy the recent ones - and not place some older one into new hardware ...
08:28 < Postal> well, becuase if you have a 64bit box, you already spent a good deal of money, so you will probably buy a newer card...
08:28 * rxr would not ....
08:28 < rxr> especially never a nvidia card ....
08:29 < CIA-10> rene * r5838 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc postlinux.conf):
08:29 < CIA-10> * added the ia64 binary driver to the nvidia package and fixed the
08:29 < CIA-10> globbing pattern to work with the annoying x86-64 binary ...
08:29 < rxr> ok - you could give that a try
08:29 < Postal> I wanted the ATI but they were out of stock for about 3 months, so i would have had a box with no card, and im trying now..
08:29 < Postal> you could have let me know i had to dl it...
08:30 < rxr> ah - sorry
08:30 < rxr> moment
08:30 < Postal> did you forget a symlink?
08:30 < rxr> no - grabbed the wrong file from the nvidia homepage
08:31 < Postal> what is with the symlinks you just added... they were not there before..
08:31 < rxr> what symlinks ?
08:32 < Postal> never mind, its not important, but in the begining of the build it creates some symbolic links to somthing, i cannot tell what it is cause it is cut off with "..." cause my resolution is like 800x600 right now... and it is usually 1280x1024
08:33 < CIA-10> rene * r5839 /trunk/package/x11/nvidia/ (nvidia.conf nvidia.desc postlinux.conf):
08:33 < CIA-10> * fixed to download the right file for the x86-64 nvidia package and
08:33 < CIA-10> fixed the globbing pattern again - sigh
08:33 < Postal> ok here we go--- again
08:33 < rxr> so - this /should/ work now
08:34 < rxr> (if the nvidia people did not do more stupid stuff in the package ...)
08:34 < rxr> so - bakery ...
08:35 < Postal> I have to dl the package, and i dont want to use emerge, cause it will compile X also you said so how do i dl it?
08:35 < rxr> Download nvidia
08:35 < rxr> ./scripts/Download nvidia
08:35 < rxr> that is
08:35 < rxr> Emerge will not build X
08:35 < rxr> just Emerge nvidia will work, too
08:35 < Postal> oh and btw, i would like to have my kernel source also...
08:35 < rxr> cu in some minutes
08:36 < Postal> ok cu in a few
08:36 < rxr> well - the kernel is in
08:36 < rxr> download/mirror/l/linux-2.6.10.tar.bz2 or so
08:36 < rxr> you can untar it to /usr/src
08:36 < rxr> the patched applied to the kernel by t2 are already there
08:36 < Postal> ok
08:36 < Postal> i will build the kernel nvidia also..?
08:36 < rxr> /usr/src/linux26-patches/
08:36 < rxr> yes
08:37 < rxr> if the X part installed while building the nvidia pacakge
08:37 < Postal> it built, now what do ?
08:37 < rxr> make sure desktop without package selection template is selected
08:37 < rxr> and Dowload -required
08:37 < rxr> so all the stuff needed to build is there
08:37 < rxr> and then
08:37 < rxr> Build-Pkg linux26
08:38 < rxr> what should include buliding the nvidia kernel module
08:38 < rxr> (by the t2 automatics)
08:38 < rxr> so - cu ;-)
08:39 < _Ragnar_> cu
08:40 < Postal> _Ragnar_ you can scroll up, can you give me the command to untar the source tree, because i cannot scroll up in centericq...
08:41 < Postal> oh wait never mind yes i can
08:41 < _Ragnar_> ;)
09:01 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has joined #t2
09:01 -!- Postal [~root@ool-43562131.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]
09:02 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has left #t2 []
09:02 -!- Postal1 [~root@ool-43562131.dyn.optonline.net] has joined #t2
09:03 < rxr> re
09:03 < rxr> so - breakfast
09:04 < Postal1> rxr, i got X to start, using the standard vga driver, but it froze... i had to reboot, i could not figure out how to make X use the new nvidia driver...
09:09 < rxr> your first need to build the kernel with the nvidia module
09:09 < rxr> have you done that ?
09:12 < Postal1> no, cause i untard the kernel source, and when nvidia found it, it complianed, hold on and ill get you the error
09:12 < rxr> that it does not match your running kernel I guess ...
09:12 < rxr> Postal1: just use Build-Pkg linux26
09:13 < rxr> it will build the nvidia module with the kernel
09:13 < rxr> and then we also know that the t2 package does now work on x86-64 ...
09:18 -!- rxr_ [~rene@p213.54.197.164.tisdip.tiscali.de] has joined #t2
09:18 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
09:18 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
09:18 [Users #t2]
09:18 [ _Ragnar_] [ ideal ] [ nzg ] [ praenti] [ sparc-kly]
09:18 [ CIA-10 ] [ jsaw ] [ ooo_ ] [ rxr ] [ valentin ]
09:18 [ daja77 ] [ mnemoc] [ Postal|Away] [ rxr_ ]
09:18 -!- Irssi: #t2: Total of 14 nicks [0 ops, 0 halfops, 0 voices, 14 normal]
09:18 -!- [freenode-info] please register your nickname...don't forget to auto-identify! http://freenode.net/faq.shtml#nicksetup
09:18 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
09:18 -!- Irssi: Join to #t2 was synced in 11 secs
09:18 < rxr_> re
09:22 -!- rxr [~rene@213.54.206.111] has quit [Read error: 60 (Operation timed out)]
09:28 -!- You're now known as rxr
09:44 < rxr> strange - sometimes sparc64 svn does illegal instruction here:
09:44 < rxr> #0 0xfffff80000c101a0 in ne_uri_parse
09:44 < rxr> and sometimes not ...
10:10 -!- mnemoc [~amery@200.75.27.25] has quit [Read error: 110 (Connection timed out)]
10:16 < CIA-10> rene * r5840 /trunk/package/base/glibc/glibc.conf: * cleaned glibc.conf
10:36 < rxr> I just found another rock inherited hack that made our 64bit life harder
10:37 < rxr> unbelievable what was injected into ROCK ages ago ...
10:38 < rxr> ouhm -4.4°C already
10:58 < CIA-10> rene * r5841 /trunk/package/base/glibc32/ (33 files): * first glibc32 draft - untested
11:06 < _Ragnar_> goodnight folks
11:07 < CIA-10> rene * r5842 /trunk/package/network/libidn/libidn.desc: * updated libidn (0.5.12 -> 0.5.13)
11:14 -!- ideal [~idealm@211.100.227.107] has quit ["Leaving"]
13:12 < rxr> hm
13:13 < rxr> on my sparc64 the illegal instruction inside svn is this:
13:13 < rxr> call 0x154
13:13 < rxr> during executing this somehow get's corrupted ...
13:14 < rxr> yep - it also is in the installed /usr/lib/libneon.a
13:14 < rxr> maybe the linker is unable link it properly ...
13:16 < rxr> hm - moment
13:16 < rxr> maybe this is strange:
13:16 < rxr> 0x0000000000000154
13:16 < rxr> does it call itself ?
13:18 < rxr> hm - ack - it is no runtime corruption
13:18 < rxr> it is also an illegal instruction after loading the svn binary into memory before main ...
13:18 < rxr> 0xfffff80000c181a0
13:18 < rxr> gdb does work mostly on sparc64 - which is good to know ....!
13:26 < rxr> hm - the gcc assembly generats is this:
13:26 < rxr> call strstr, 0
14:18 < rxr> I think -fPIC helped it ..
14:18 < rxr> yeah
14:19 < rxr> after one dive thru gcc -> assembly -> relocaton and gdb svn works on sparc64 !!!!!!!!!
14:22 < CIA-10> rene * r5843 /trunk/package/network/neon/neon.conf:
14:22 < CIA-10> * fixed neon to build a shared library by default - if static inject
14:22 < CIA-10> -fPIC - both ways help sparc64 to not illegal instructon due to
14:22 < CIA-10> not relocated call ;-) !!!
14:22 < rxr> ^- committed from sparc64 !!!
14:33 < CIA-10> lars * r5844 /trunk/package/powerpc/pbbuttonsd/ (pbbuttonsd.conf pbbuttonsd.desc):
14:33 < CIA-10> * updated pbbuttonsd (0.6.6 -> 0.6.7) (tested)
14:33 < CIA-10> * fixed /etc/pbbuttonds.conf to use ALSA instead of OSS
15:09 < CIA-10> rene * r5845 /trunk/scripts/Emerge-Pkg:
15:09 < CIA-10> * improved Emerge-Pkg to always check if a build is needed and
15:09 < CIA-10> added a -force option to Emerge-Pkg to force builds
15:13 < CIA-10> rene * r5846 /trunk/architecture/ (sparc/gcc-options sparc64/gcc-options): * removed -mno-app-regs from the sparc*/gcc-options files
15:23 < CIA-10> rene * r5847 /trunk/package/base/glibc/glibc.conf:
15:23 < CIA-10> * fixed glibc to use -fcall-used-g6 for sparc64 globally (not yet sure
15:23 < CIA-10> if it makes such a difference - but it is that Fedora does ...)
16:00 < CIA-10> rene * r5848 /trunk/package/base/procps/64bit-size_t.patch:
16:00 < CIA-10> * fixed procps abort due to SIGBUS on at least sparc64 (maybe other
16:00 < CIA-10> 64bit architectures, too)
17:00 < rxr> crazzy
17:00 < rxr> gcc on sparc64 optimizes a timeval memcpy into:
17:00 < rxr> add %l1, 8, %l0
17:00 < rxr> ldx [%l0+8], %g2
17:00 < rxr> ldx [%l1+8], %g1
17:00 < rxr> sethi %hi(.LLC50), %o0
17:00 < rxr> stx %g1, [%fp+1679]
17:00 < rxr> stx %g2, [%fp+1687]
17:01 < rxr> not bad
17:03 < valentin> how does that perform in MB/s in contrast to a non optimized single byte copy ?
17:03 < rxr> too bad the first load crashes ...
17:17 < rxr> well whole word load and stores are always way faster than sub word access on all archzes ...
17:21 < rxr> strange %l0 looks very much 8byte aligned ...
17:21 < rxr> 0x224c00
17:23 < rxr> well - sorry
17:23 < rxr> 0x204c04
17:29 < rxr> I might hit this bug:
17:29 < rxr> http://gcc.gnu.org/ml/gcc-patches/2003-11/msg02233.html
17:55 < rxr> 0x00000000001027ac
17:55 < rxr> 0x00000000001027b0
17:55 < rxr> 0x00000000001027b4
17:55 < rxr> 0x00000000001027b8
17:55 < rxr> 0x00000000001027bc
17:55 < rxr> 0x00000000001027c0
17:55 < rxr> 0x00000000001027c4
17:55 < rxr> 0x00000000001027c8
18:01 < rxr> fixed without 64bit loads ;-)
18:01 < rxr> the next big task of fixing X is not for today ...
18:06 < rxr> root@sunrise:/t2-trunk# ping www.heise.de
18:06 < rxr> PING www.heise.de (193.99.144.71): 56 octets data
18:06 < rxr> 64 octets from 193.99.144.71: icmp_seq=0 ttl=247 time=104.0 ms
18:09 < CIA-10> rene * r5849 /trunk/package/network/netkit-base/64bit-alignment.patch: * fixed timeval cast in ping to not crash by SIGBUS on at least sparc64
18:14 < nzg> quite interesting... why use almost all german people www.heise.de as host for test-pings?
18:14 < nzg> moin btw
18:16 < rxr> hehe - ;-) used to "just works" ;-)
21:07 < rxr> n8 all
21:11 < _Ragnar_> gn8
21:14 -!- sparc-kly [~mubex@65-23-206-237.prtc.net] has left #t2 ["Leaving"]
21:58 -!- CIA-10 [~CIA@flapjack.navi.cx] has quit []
22:00 -!- CIA-temp267 [~CIA@flapjack.navi.cx] has joined #t2
22:00 -!- CIA-temp267 is now known as CIA-8
22:03 -!- Netsplit saberhagen.freenode.net <-> irc.freenode.net quits: daja77, jsaw, valentin, praenti, Postal|Away, _Ragnar_, nzg, ooo_
22:14 -!- rxr_ [~rene@p213.54.197.164.tisdip.tiscali.de] has joined #t2
22:14 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
22:14 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
22:14 [Users #t2]
22:14 [ _Ragnar_] [ daja77] [ nzg ] [ Postal|Away] [ rxr_ ]
22:14 [ CIA-8 ] [ jsaw ] [ ooo_] [ praenti ] [ valentin]
22:14 -!- Irssi: #t2: Total of 10 nicks [0 ops, 0 halfops, 0 voices, 10 normal]
22:14 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
22:14 -!- Irssi: Join to #t2 was synced in 11 secs
--- Log closed Sun Jan 30 00:00:46 2005