: T2 SDE :

T2 IRC Log: 2007-08-07

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.

« prev | next »

--- Log opened Tue Aug 07 00:00:04 2007
00:18 -!- renderfil [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has joined #t2
00:18 -!- renderfil is now known as renderfull
00:25 -!- yokoy [n=yokoy@e178087143.adsl.alicedsl.de] has quit [Read error: 113 (No route to host)]
00:55 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has quit [Read error: 110 (Connection timed out)]
01:00 -!- yokoy [n=yokoy@e179042243.adsl.alicedsl.de] has joined #t2
01:59 -!- Enqlave [i=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
02:06 -!- Enqlave [i=stealth@unaffiliated/enqlave] has joined #t2
02:13 -!- Enqlave [i=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
02:46 -!- Enqlave [i=stealth@unaffiliated/enqlave] has joined #t2
03:30 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has joined #t2
04:16 -!- mqueiros_ [n=mqueiros@c-217-70-66-47.bragatel.pt] has quit ["Ex-Chat"]
04:22 -!- mqueiros [n=mqueiros@c-217-70-66-47.bragatel.pt] has joined #t2
04:54 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has quit []
04:55 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has joined #t2
05:32 -!- yokoy [n=yokoy@e179042243.adsl.alicedsl.de] has quit [Read error: 113 (No route to host)]
06:03 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has quit [Read error: 110 (Connection timed out)]
06:17 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has joined #t2
07:53 -!- DeeKey [n=kulandin@jabber.hst.ru] has joined #t2
07:58 -!- renderfull [n=codey@c-67-190-0-98.hsd1.co.comcast.net] has quit []
08:34 < rxr> re
08:34 [Users #t2]
08:34 [@ChanServ] [ cyda ] [ Fatal ] [ mtr ] [ rxr ] [ valentin]
08:34 [ cbrake ] [ DeeKey ] [ LMJ ] [ pdinoto ] [ sepp_]
08:34 [ CIA-8 ] [ Enqlave] [ mqueiros] [ Ragnarin] [ TobiX]
08:34 -!- Irssi: #t2: Total of 16 nicks [1 ops, 0 halfops, 0 voices, 15 normal]
08:34 < DeeKey> hi
09:16 < CIA-8> susan * r25397 /trunk/package/xorg/xclock/xclock.desc: * updated xclock (1.0.2 -> 1.0.3)
09:16 < CIA-8> susan * r25398 /trunk/package/xorg/xf86-input-mouse/xf86-input-mouse.desc: * updated xf86-input-mouse (1.1.2 -> 1.2.2)
09:17 < CIA-8> susan * r25399 /trunk/package/xorg/xf86-video-ati/xf86-video-ati.desc: * updated xf86-video-ati (6.6.192 -> 6.6.193)
09:17 < CIA-8> susan * r25400 /trunk/package/xorg/xgamma/xgamma.desc: * updated xgamma (1.0.1 -> 1.0.2)
09:17 < CIA-8> susan * r25401 /trunk/package/xorg/xmodmap/xmodmap.desc: * updated xmodmap (1.0.2 -> 1.0.3)
09:17 < CIA-8> susan * r25402 /trunk/package/contrib/915resolution/915resolution.desc: * updated 915resolution (0.5.2 -> 0.5.3)
09:22 < CIA-8> susan * r25403 /trunk/package/editors/tea/tea.desc: * updated tea (17.1.1 -> 17.1.4)
09:22 < CIA-8> susan * r25404 /trunk/package/network/mod_security/mod_security.desc: * updated mod_security (2.1.1 -> 2.1.2)
09:24 < CIA-8> susan * r25405 /trunk/package/scientific/comedi/comedi.desc: * updated comedi (0.7.22 -> 0.8.0)
09:24 < rxr> DeeKey: so there went your comedia update, hm ?
09:25 < DeeKey> I concluded that i need to include autoconf to the build
09:25 < sepp_> moin :)
09:25 < DeeKey> so that it will be possible to run autogen.sh
09:25 < DeeKey> hi sepp
09:25 < sepp_> hi DeeKey
09:26 < DeeKey> but what actually bothers me is the problem with busybox's init...
09:26 < rxr> DeeKey: with the release tarballs you probably will not need autoconf
09:26 < rxr> how much of rtai compiles in the branch, now?
09:27 < DeeKey> everything exept comedi-driver
09:27 < rxr> ok - maybe it is time to merge the branch
09:27 < rxr> as otherwise merging work on the comedia parts will get conflicts with the update of comedia now in trunk
09:31 < DeeKey> about comedi and tar ball... i have read a lot of docs about installing comedi and comedi-lib. There were written, that it is better t extract from cvs.. But just now I have checked the comedi download page nad can see that the latest version is dated august the first... So now i took decision to use tarballs...
09:32 < DeeKey> i have problems with init... I cannot get to command prompt after booting !
09:33 < DeeKey> And besides there is conflict if one decides to use both busybox and other init like minit or else
09:33 < DeeKey> Busybox package needs more code..
09:33 < DeeKey> to handle the configuration issue..
09:35 < rxr> so far I had no problems
09:35 < rxr> I have target that just use busybox init as well as others using minit + busybox tools
09:36 < DeeKey> hmmm... When I include minit, then during busybox compilation there is an error: found other package and a sign" : "
09:40 < DeeKey> but anyway, i need to investigate the problem with init before merging the branches... It is not good to add targets which cannot be compiled!
09:40 < DeeKey> and used
09:40 < DeeKey> at least by one user :)
09:42 < CIA-8> susan * r25406 /trunk/package/scientific/comedi-driver/comedi-driver.desc: * updated comedi-driver (0.7.72 -> 0.7.74)
10:11 -!- write_erase [n=olivier@81.80.156.36] has joined #t2
10:25 < rxr> I finally installed a trunk/live amd64 flavour on my MacBook, so it finally is gcc-4.2 as well
11:07 < rxr> re
11:07 < DeeKey> hi
11:08 < rxr> T2 SDE 7.0-trunk (2007/07/29)
11:10 < rxr> valentin: parts of the hald problems on your MacBook that you cloned from mine is that the hald I had on it wants inotify stuff in the glibc that the glibc on that version doesn
11:10 < rxr> 't provide yet
11:10 < rxr> on trunk it works better
11:10 < rxr> I polish the remaining edges
11:50 < CIA-8> rene * r25407 /trunk/package/network/dbus-qt/ (. dbus-qt.cache dbus-qt.desc): * added dbus-qt (0.70) - Qt3 binding for the desktop bus message bus system
12:33 -!- Baldzius [n=Baldzius@87.198.192.226] has joined #t2
12:33 < Baldzius> moin
13:17 < rxr> moin Baldzius
13:20 < Baldzius> hi rxr
13:28 < rxr> valentin: with trunk and the new dbus-qt and kdebase rebuild plug'n play media detection in KDE starts to work again
13:38 < rxr> yep - works
13:43 < valentin> rxr: nice
13:45 < rxr> known issues:
13:45 < rxr> su does not work
13:45 < rxr> even not after pam tweak, need to investigate
13:45 < rxr> and the touchpad (MacBook) does not work, maybe aignment / padding with gcc-4.2
13:46 < rxr> totally wrong co-ordinate movements
13:46 < rxr> probably both fixable in an hour or so
13:46 < rxr> but guess have to continue other stuff, now
13:54 < valentin> rxr: i will not sync today, so take your time ;-)
13:54 < rxr> saw the new hardware areas ?
13:55 < valentin> sure
13:55 < valentin> you should fill the intermediate pages like
13:56 < valentin> http://t2-project.org/hardware/workstation/
13:56 < valentin> theese look very deserted otherwise
13:56 < valentin> maybe include links to the sub-items in the main frame ?
13:57 < rxr> well - the menu already has them
13:57 < rxr> probably a describtive text talking a little what the section is about
14:02 < rxr> http://golem.de/0708/53962.html
14:04 < rxr>
14:04 < rxr> Suns T2: Schnellste CPU der Welt - bei 1,4 GHz?
14:05 < mqueiros> rxr: Hello! I agree with valentin. It feels empty... it may even feel that there's no information available. The user might not even look at the menu in the left...
14:21 < rxr> I thought about overview texts such as:
14:21 < rxr> http://www.t2-project.org/hardware/console/
14:21 < rxr> http://www.t2-project.org/hardware/embedded/
14:21 < rxr> (just added)what do you think?
14:23 < DeeKey> Are there any small embedded targets in t2?
14:23 < DeeKey> just 2 MB of ROM and RAM
14:24 < rxr> the embedded w/ dietlibc comes near
14:24 < rxr> but our company did targets of that size that are in production :-)
14:24 < mqueiros> I think it's much better now.
14:24 < rxr> mqueiros: ok - I do the remaining then portable, workstaiton and server
14:45 < CIA-8> aldas * r25408 /branches/6.0/package/network/tor/tor.desc:
14:45 < CIA-8> * merged r25370 from trunk to branches/6.0:
14:45 < CIA-8> updated tor (0.1.2.15 -> 0.1.2.16)
14:45 < CIA-8> security update: CVE-2007-4174
14:46 < rxr> now "only" the individual hardware sites need to be filled
14:46 < rxr> (as well as the overview texts reviewed and extended
14:46 < rxr> )
15:57 -!- yokoy [n=yokoy@e178070141.adsl.alicedsl.de] has joined #t2
16:08 < mqueiros> rxr: small typo in hardware/worstation: "markget"
16:09 < mqueiros> rxr: and I believe "ordeneray" should be "ordinary"
16:10 < rxr> done
16:10 < rxr> thanks
16:12 < mqueiros> yesterday the T2 build got to an end... (there where some errors in kernel modules, that I have disabled for now). When I was trying to create de ISO for my live CD I got ".../TOOLCHAIN/isofs/ is too big!"
16:13 < rxr> -size 9999 and burn on DVD :-)
16:14 < mqueiros> ok... that simple... :)
16:14 < rxr> libtiff gets on my nerves, ...
16:15 < rxr> mqueiros: default is normal 74m CD size ...
16:15 < rxr> if you want a CD you just have to trim your target .-)
16:15 < rxr> so - I'm a little of - cu later
16:16 < mqueiros> rxr: I thought so... but it seemed that there wheren't so many packages selected...
16:20 < rxr> how big is your iso ?
16:22 < mqueiros> 719396864
16:23 < mqueiros> 687 MB...
16:24 < rxr> I think that fits on the 80m CDs ...
16:25 < mqueiros> Yes... will try it...
--- Log opened Tue Aug 07 19:57:35 2007
19:57 < rxr> heh - selected the wrong logger
19:57 < rxr> ok - looks better:
19:57 < rxr> --- Log opened Tue Aug 07 19:57:35 2007
19:57 < rxr> 19:57 < rxr> heh - selected the wrong logger
19:59 < rxr> so - new slim iMacs are hout, hm ?
20:02 < rxr> http://www.engadget.com/2007/08/07/apples-new-imac/
20:05 < Ragnarin> neat
20:05 < rxr> saw the keyboard ?
20:05 < Ragnarin> yea
20:05 < rxr> but what I'm really looking for is a ultraportable MacBook
20:05 < rxr> I really already brought an IBM X61 (but has sucky display) or a Sony TZ (is just expensive) and both have no digital video out ..
20:06 < rxr> those ultra portable MacBooks are rumoured since last december / january, now
20:06 < rxr> damnit
20:07 < rxr> http://rene.rebe.name/2007-07-30/sony-ngw-tz-the-pretty-subnotebook-that-doesnt-wanna-be/
20:27 < Ragnarin> wow
20:27 * Ragnarin wants a T2 based server
20:27 < Ragnarin> http://www.heise.de/newsticker/meldung/93999
20:30 < rxr> Ragnarin: yeah
20:30 < Ragnarin> any idea what the most basic one is going to sell for?
20:30 < rxr> depending on price and available either that or a AMD Barcelona quad ccore wwill become our new office server
20:30 < rxr> Ragnarin: no iea, the current T1, T1000 starts around 3000 EUR or so
20:31 < Ragnarin> complete system?
20:31 < rxr> yes
20:31 < rxr> the T1, 6 core == 24 threads 2 G of RAM, some disk or so
20:32 -!- d-marc [n=d-marc@HSI-KBW-091-089-000-221.hsi2.kabelbw.de] has joined #t2
20:32 -!- d-marc [n=d-marc@HSI-KBW-091-089-000-221.hsi2.kabelbw.de] has quit [Remote closed the connection]
20:33 < rxr> Ragnarin: I had a T1000 here for testing with T2
20:33 < rxr> unfortunately Sun did not wanted to donate it ...
20:33 < rxr> (permanently)
20:33 < rxr> the T1000 starts at Your price is discounted based on this contract:
20:33 < rxr> $ 3,995.00
20:34 < rxr> no idea if the T2 will replace the T1 or if they introduce a "over the T1 performance line" ...
20:34 < rxr> but I'm waiting for those
20:34 < rxr> and I hope we do not get sued from Sun ...
20:35 < rxr> (I should have registered T2 as trademark earlier ...)
20:35 < rxr> now I hope for the best and cross my fingers
20:35 < rxr> but there aso was this Terratec T2 DVB-T receiver, soo ...
20:36 < Ragnarin> well the proc is hardware, we are software
20:36 < Ragnarin> it shouldn't really collide
20:37 < rxr> there is no diifferentiation betwen software and hardware
20:37 < Ragnarin> and since they leased you the T1, they knew we were using T2 as a name before, so we shouldn't stand too bad :p
20:37 < rxr> just electronic something whatever computer related
20:37 < rxr> I know as we recently registered something else
20:38 < rxr> the Sun site is strange
20:38 < rxr> they markeg this T2 lunch - but do not have a system to sell
20:38 < Ragnarin> ?
20:38 < rxr> strange "lunch"
20:39 < Ragnarin> lol
20:39 < rxr> launch even
20:39 < Ragnarin> in a month, heise says
20:39 < Ragnarin> but 4000$ is a bit steep ....
20:40 < rxr> in a month ++ the time David S. Miller needs to adapt the Linux kernel for it ...
20:40 < Ragnarin> lol
20:41 < Ragnarin> yea that may need quite a bit of adapting
20:41 < Ragnarin> to use the builtin multiple FPUs and crypto untis
20:41 < Ragnarin> units*
20:42 < rxr> the FPU probably needs no adaption
20:42 < rxr> as the same V9 instruction set is used
20:43 < rxr> it is just that unlike on the T1, which only has 1 FP for all running threads, the fpu instructions do not stall so much ...
20:44 < rxr> the adaptions necessariy are probably: MMU handling (if they changed it since T1), new PCI-E devices, and the crypto unit
20:44 < Ragnarin> yea but the T1 only has one FPU while the T2 has 8
20:44 < Ragnarin> oh nvm lol
20:45 < rxr> yesh but the fpu stuff is not directly visible, it is just that floating point instructinos are normally fast and do not stall the execution of the threads that much
20:45 < rxr> nvm ?
20:45 < Ragnarin> yea okay
20:45 < Ragnarin> I thought there'll still be need to save the FPU regs on a context switch, for example
20:47 < rxr> yes, but thats the same on the T1, the registers windows where there as well, per core and thread, like on the hyperthreaded P4
20:47 < rxr> it is just that there was no 1:1 core on the T1, but 24 or 32 : 1 ratio to the ALU
20:48 < rxr> while the register windows map - what was it for the T2 ? - 64 : 8 or so
20:48 < Ragnarin> ah ok ....
20:48 < Ragnarin> I think I understand what you mean
20:48 < rxr> the whole thing about this "extreme threading" of the T1 and T2 is:
20:49 < rxr> when a thread is stalled, due to memory load or store, the core continous "round robin" executing of the instructions of the other threads
20:49 < rxr> just like it was on thsi hyper-threaded Intel Pentium 4's
20:49 < Ragnarin> yea
20:49 < rxr> to hide memory latency and improve thruput of the ALUs
20:50 < rxr> it's "just" that the Sun T1/T2 make _massive_ use of this SMT setup, by providing 4 threads of execution per core
20:52 < rxr> btw the T1000 was very very very loud and noisy
20:52 < rxr> it was too noisy for a normal office room
20:53 < rxr> quite a pitty with just 72W consumed by the CPU
20:53 < rxr> I guess so that the system even works when placed in a desert or so ...
20:53 < Ragnarin> o_o what made it so noisy?
20:53 < rxr> the many fans in the 1H case ...
20:53 < rxr> but I doubt they where all needed
20:53 < rxr> I guess just to make sure it works whern the aircon in the server room is down or so ...
20:54 < rxr> worst case plaing from sun ...
20:54 < rxr> planing
20:54 -!- mqueiros_ [n=mqueiros@c-217-70-66-47.bragatel.pt] has joined #t2
20:55 < Ragnarin> yea ... good provision, that
20:55 < rxr> probable those boxes are 99% intented for big data centeres anyway
20:55 < rxr> but for our "normal" office space here it sucks ...
20:55 < Ragnarin> ya
20:58 < Ragnarin> well we have a server room here
21:23 -!- Dallur [n=earl@fire-out.ru.is] has joined #t2
21:38 [Users #t2]
21:38 [@ChanServ] [ cyda ] [ Fatal ] [ mtr ] [ renderfull] [ TobiX ]
21:38 [ cbrake ] [ Dallur ] [ LMJ ] [ pdinoto ] [ rxr ] [ valentin]
21:38 [ CIA-8 ] [ Enqlave] [ mqueiros_] [ Ragnarin] [ sepp ] [ yokoy ]
21:38 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
21:44 -!- kic- [i=rfoote@adsl-074-238-088-208.sip.bct.bellsouth.net] has joined #t2
21:44 < kic-> re
21:44 < kic-> /identme
21:44 < kic-> oops
22:49 -!- d-marc [n=d-marc@HSI-KBW-091-089-000-221.hsi2.kabelbw.de] has joined #t2
22:49 -!- d-marc [n=d-marc@HSI-KBW-091-089-000-221.hsi2.kabelbw.de] has left #t2 ["Time makes no sense"]
23:13 -!- wbx [n=wbx@xdsl-87-78-176-83.netcologne.de] has joined #t2
--- Log closed Wed Aug 08 00:00:22 2007