/srv/irclogs.ubuntu.com/2010/05/31/#ubuntu-kernel.txt

smbMorning to the part of the world without bank holiday08:16
RAOFWhich part of the world *has* a bank holiday, and how do I apply? :)08:16
smbRAOF, US and UK ant least. How you apply I don'T know. :)08:17
smbMy part of the world has its holiday Thursday08:17
RAOFAh.  That might be why it's been unusually quiet.08:18
smbYep, time to get work done *cough, cough*08:18
* smb waves to ikepanhc 08:21
jk-hey smb & RAOF08:22
ikepanhcsmb: good morning08:22
smbHi jk- 08:23
* amitk waves08:25
* cooloney waves back to amitk, smb, ikepanhc, jk- and RAOF 08:27
RAOFo/08:27
smbcooloney, \o08:27
* ikepanhc waves08:30
stentenCan someone help me decide the importance of Bug #587136?08:35
ubot2Launchpad bug 587136 in linux (Ubuntu) "2nd Resume from Suspend results in reboot on Toshiba Satellite U400. Fixed in 2.6.34 Mainline. (affects: 2) (dups: 1) (heat: 20)" [Undecided,Confirmed] https://launchpad.net/bugs/58713608:35
stentenBased off the other linux bugs, I'm thinking High.08:35
smbstenten, I would agree08:36
stentensmb: Thank you kindly.08:37
TeTeTsmb: Hi Stefan, any chance to get the Lenovo x201 WWAN module into lbm? seems that mjg59 refactored it, according to the last mail09:35
smbTeTeT, Its on the list to look at, now the other things have been done. I am not sure how far I get with all of it today and unfortunately I did not heve time to look at it again, yet09:37
TeTeTsmb: the roll out for the customer is not before july 18th, so anything that lands for 10.04.1 is good enough. 09:39
TeTeTsmb: Should I escalate this driver support issue through Zaid, so it get's on your "official" todo list? or is it good enough as is?09:40
smbTeTeT, Ok. Let me look at it to decide how much work it might be. I would ping you if I think it needs more work and thus a more official backing.09:42
TeTeTsmb: thanks a lot, please let me know in any case 09:43
smbTeTeT, Sure09:43
yzhaocooloney: hi, bryan~10:23
cooloneyyzhao: welcome, yingying10:32
* smb waves to yzhao 10:32
ericm|ubuntuyzhao, yingying!10:41
ericm|ubuntusmb, do you know who's maintaining libusb? I wish I know where to send a relevant patch10:41
smbericm|ubuntu, Not from my head. Would need to ask checkmaintainer.pl too. Theoreticall Greg and Alsan Stern might be suspect. But I would see what the script says.10:42
smbs/Alsan/Alan/10:43
ericm|ubuntusmb, ah - not drivers/usb/, but a user space libusb library10:44
ericm|ubuntuit's in main repo, yet seems to be unmaintained by any of canonical member?10:44
smbericm|ubuntu, oh, user-space... eek. :-P10:45
ericm|ubuntuyet the bzr branch is progressing by <archive@ubuntu.com>, .....10:45
smbMaintainer: Ubuntu Developers <ubuntu-devel-discuss@lists.ubuntu.com>10:46
smbOriginal-Maintainer: Aurelien Jarno <aurel32@debian.org>10:46
smbericm|ubuntu, So I'd post to both. The second probably is the debian maintainer10:47
ericm|ubuntusmb, how did you find this out? I executed dpkg-query -p libusb-0.1-4, only find Aurelien Jarno10:48
ericm|ubuntubut yeah, I guess these two mail addresses will work10:48
smbapt-cache show libusb-1.0-0 (not on maverick, yet :-P)10:49
smbOh, wait libusb-0.1... ?10:50
ericm|ubuntusmb, yes - legacy and unfortunately upower seems to be linked with it10:51
smbI guess I am just confused by one thing finding a libusb-1.0 and the other number being a 0.110:52
smbericm|ubuntu, But luckily the mail addresses are the same for both10:53
ericm|ubuntusmb, heh - since 0.1 is crap10:53
ericm|ubuntuthat's why it's broken on my Mac10:54
smbHeh, and somehow I got the impression both are produced by the same source package10:55
smb... no wrong...10:56
smbericm|ubuntu, So there is a libusb and a libusb-1.010:56
ericm|ubuntusmb, they are two separate10:57
smbericm|ubuntu, Which you probably were already saying by legacy lib linked agains upower10:57
=== cooloney is now known as cooloney-afk
=== smb is now known as smb-afk
ericm|ubuntuseems emerald.pgraner is down again11:15
=== popey_ is now known as popey
=== smb-afk is now known as smb
diwicHi there, how do I find out what options are enabled in a standard Lucid kernel?12:30
ograin /boot/config-$(uname -r)12:32
amitkdiwic: 'less /boot/config-2.6.32-*12:32
amitk'12:32
amitkheh12:32
ogra:)12:32
diwicthanks :-)12:33
diwicis there also an online list for different kernel versions than the one I have currently installed? 12:35
=== ghostcube_ is now known as ghostcube
lapionHello,13:51
lapionhow can I cross- search for processor-chipset combinations in launchpad13:52
hrwhi14:08
hrwis it possible to crossbuild linux-libc-dev from linux source package?14:09
smbhrw, I am not sure I understand what you mean with crossbuild. Build on another architecture?14:36
hrwsmb: on amd64 build package with headers for armel14:36
ericm|ubuntusmb, do you mind have a look at my git pull request a bit time ago (just forwarded that mail to you)?14:38
smbhrw, If you have the toolchain installed (though that is only available for i386 iirc) you could cross compile the whole package. Other I don't know14:39
hrwok14:39
ericm|ubuntuhrw, ask in ubuntu-arm, they may possibly know14:39
hrwthx for info14:39
smbmpoirier, Wouid you know something better?14:39
hrwericm|ubuntu: will14:39
ericm|ubuntuhrw, not every package is cross-build-able14:39
smbericm|ubuntu, Ok, will do14:40
ericm|ubuntusmb, let me know if it needs a SRU process - it was actually sent before -L release maybe (or happened to be the freeze period maybe)14:40
mpoiriersmb: what exactly do you want to know ?14:42
smbericm|ubuntu, Everything we now upload needs at least a little more process. Not yet looked into the mail, but one thing good to have would be a lp bug report that tracks/explains things14:42
ericm|ubuntusmb, LP bug report already there14:42
smbmpoirier, It was about the cross complile question from hrw 14:42
ericm|ubuntuand I've updated every commit with that info14:42
smbericm|ubuntu, Sounds good14:42
ericm|ubuntusmb, thanks man14:43
smbmpoirier, I don't do many arm builds but the standard cross compiles of the whole kernel package. But I don't think there is a way to get the libc-dev package without a complete cross compile env.14:44
mpoiriersmb: not that I know of...14:45
smbmpoirier, Ok, thanks for the confirmation. 14:46
mpoiriersmb: the cross-compilation environment is changing for maverick - for ARM at least.14:46
mpoiriersmb: for OMAP4 I should say.14:46
smbmpoirier, Something I probably need to inform myself sooner or later14:47
smbor saying educate myself14:47
mpoiriersmb: I wouldn't rush too quicly on that one...14:48
mpoirierI think it is still maturing.14:48
amitkmpoirier: the cross-build enviroment changing? how?14:48
mpoirierChanging is probably not the word I should have used.14:48
smbmpoirier, Oh, now worries, I am seldom rushing. :)14:48
smbgos no worries14:49
hrwamitk: I am working on making 'cross-toolchain' source package possible14:49
* smb cannot type today14:49
mpoirierI was referring to Bryan's sbuild stuff, which differs from what was being done in Lucid.14:49
hrwsmb: "ARCH=arm make headers_install" does not require cross toolchain14:49
hrwmpoirier: ah14:49
mpoirierMind you, I am still plowing through that one.14:50
amitkmpoirier: there are two different things (sbuild and cross)14:50
amitk*they14:50
smbhrw, That can be true. I am just not sure the debian build environment we use allows to produce the lib package without going through the architecture specific parts which also does the kernel image. 14:54
smbBut then, I cannot say I am an arm guy...14:54
hrwsmb: s/arm/anyotherarch/ even14:56
hrwI recently hacked binutils packaging and built cross binutils for alpha/hppa/etc just to test does it work14:57
smbOK, so I should say when I did compile I either completely cross compiled or used native builds.14:58
hrwcorrect me if I'm wrong but does not 'linux' source package has broken builddepends?15:17
hrwBuild-Depends: dpkg (>= 1.13.19), debhelper (>= 5), gawk15:17
hrwkernel-wedge was required to let me start building, then it broke due to lack of libefl-dev...15:18
amitklibefl to build a kernel package?15:18
hrwMakefile:504: *** No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel and glibc-dev[el].  Stop.15:19
hrwmake[1]: Leaving directory `/home/hrw/devel/canonical/2010/05-cross-compilers/ubuntu/maverick/source/linux-2.6.34/debian/build/tools/tools/perf'15:19
hrwmake: *** [install-tools] Błąd 215:19
hrwdpkg-buildpackage: błąd: fakeroot debian/rules binary zwrócił status błędu 215:19
amitkaah, libelf, _not_ libefl15:19
hrwargh typos15:20
amitkhrw: I am assuming you're cross compiling, what is your exact command?15:20
hrwdpkg-buildpackage -b -uc -us -aarmel 15:20
hrwwith proper 'export CROSS_COMPILE' before it15:21
smbHm, in my lucid tree (master)15:22
smbBuild-Depends: debhelper (>= 5), cpio, module-init-tools, kernel-wedge (>= 2.24ubuntu1), makedumpfile [amd64 i386 lpia], device-tree-compiler [powerpc], libelf-dev, binutils-dev, rsync15:22
amitkhrw: and what kernel are you trying to compile15:23
hrwamitk: linux (2.6.34-4.11) maverick; urgency=low15:24
amitkhrw: how soon does the build break? I've started one now, no problems15:34
hrwamitk: did not checked but took some time - kernel got built first15:37
amitk(building modules now)15:37
amitkhrw: in any case, it is a problem on your side since we've got work arm kernels built on buildds in the archive15:38
amitk*working15:38
hrwok15:40
hrwamitk: sure, cross != native so problems can exists15:40
hrwand I am fine with it15:40
amitkhrw: we'll know soon enough, I'm building cross using your commands above15:41
mpoiriersmb: https://wiki.ubuntu.com/StableReleaseUpdates15:44
=== ghostcube is now known as babyrobbe
=== babyrobbe is now known as ghostcube
=== ghostcube is now known as sash_mobile
=== sash_mobile is now known as sash_mobil
=== sash_mobil is now known as ghostcube
amitkhrw: confirmed the problem you are seeing with libelf with cross-compiling16:33
smbmpoirier, https://edge.launchpad.net/ubuntu/+source/linux/2.6.34-5.12/+build/176218716:38
smbhttps://edge.launchpad.net/ubuntu/+source/linux/+publishinghistory16:39
TeTeTsmb: thanks for your quick answer, I'll see if we need to escalate this or can solve it with the customers repo17:06
smbTeTeT, Ok, thanks. I think at least we should be aware of the implications beside the driver module17:07
hrwhave a nice rest of day17:07
=== hrw is now known as hrw|gone
TeTeTsmb: the customer has a blob from Lenovo for the firmware and they can use that internally. I'll talk to my Lenovo contact and see if there's a chance to ship it17:08
smbTeTeT, I would also keep Pete informed as he kind of tracks the legal side17:09
TeTeTsmb: no worries, if we pursue this, it's going through the proper escalation channels17:10
smbOk, cool17:11
=== emma_ is now known as emma

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!