[12:56] <defcon> chuck short here?
[12:58] <defcon> he replied to my posted kernel bug https://bugs.launchpad.net/ubuntu/+bug/130078 and he requested a full "oops" is oops the output of the error or the whole boot log, btw bootlogd and logd isnt working, its not logging my boot process
[03:54] <tampafl> evening all
[03:55] <tampafl> I am looking for my kernel src code.  I guess in /usr/src... but I have script that says, " Kernel source from "/lib/modules/2.6.20-16-generic/build" will be used to build the module.
[03:55] <infinity> apt-get install linux-headers-`uname -r`
[03:56] <infinity> I'm assuming you just need the headers here, if it's just to compile something against your running kernel.
[03:56] <tampafl> thanks you ;) ;)
[03:56] <tampafl> said everytrhing  up to date, 0 done
[03:57] <infinity> Err, then you have the headers installed.  You win.
[03:57] <tampafl> make -C /lib/modules/2.6.20-16-generic/build SUBDIRS=/home/bob/vpnclient modules
[03:57] <tampafl> make[1] : Entering directory `/usr/src/linux-headers-2.6.20-16-generic'
[03:57] <tampafl>   CC [M]   /home/bob/vpnclient/linuxcniapi.o
[03:58] <infinity> (note that /lib/modules/2.6.20-16-generic/build is a symlink to /usr/src/linux-headers-2.6.20-16-generic)
[03:58] <tampafl> hummm
[03:59] <tampafl> any ideas for the first make error?
[03:59] <mjg59> tampafl: You haven't pasted an error yet...
[04:01] <infinity> Yeah, I'm not seeing any errors.
[06:52] <lamont> mkdir -p debian/d-i-hppa
[06:52] <lamont> imagelist=$(cat kernel-versions | grep ^hppa | awk '{print $4}') && \
[06:52] <lamont>         for i in $imagelist; do \
[06:52] <lamont>           dpkg -x $(ls ../linux-image-$i\_*hppa.deb) \
[06:53] <lamont>                 debian/d-i-hppa; \
[06:53] <lamont>         done
[06:53] <lamont> dpkg-deb: --extract takes at most two arguments (.deb and directory)
[03:44] <mjg59> rtg: There's a (apparently) working patch for alps
[03:48] <rtg> mjg59: Where?
[03:49] <mjg59> rtg: I've attached a link to the bug
[03:50] <rtg> mjg59: Thanks.
[03:51] <rtg> mjg59: Is Vostro the model of the ALPS touchpad?
[03:52] <mjg59> No, it's a range of Dell laptops
[03:53] <rtg> Its not a term I've heard from Dell.
[03:54] <mjg59> http://www.dell.com/content/products/category.aspx/vostronb?c=us&cs=04&l=en&s=bsd
[03:55] <rtg> mjg59: Huh. Marketing stuff. All I ever deal with are the engineers and they use different terminology.
[03:59] <rtg> mjg59: I'll post a test kernel in the LP report later today (if it works).
[04:49] <Nafallo> is atheros working those days? :-)
[04:54] <Nafallo> AR5005G to be exact :-)
[05:57] <BenC> Nafallo: should be
[05:58] <Nafallo> nice. I just ordered a cheapas laptop to have until I can afford a better one :-)
[05:58] <Nafallo> has all sorts of weird hardware in it :-/
[06:02] <lamont> BenC: I won't say my tree is ready to be synced yet...  In somewhere around an hour I expect that I will say that...
[06:03] <lamont> BenC: and it would be nice if the extraction mess above had the kernel deb version in the expansion for dpkg -x, so that it quits bitching when there is cruft above
[06:19] <BenC> lamont: we don't expect cruft in a build, but I guess it makes sense :)
[06:19] <lamont> it's cruft in the parent directory, though... :-)
[06:20] <lamont> and yeah.  hence the "it'd be nice" instead of a patch.. :)
[06:20] <BenC> I should "rm -f ../*.deb" in the clean target :)
[06:20] <lamont> that'd get you a bug report, for certain. :-)
[06:20] <BenC> hehe
[06:39] <login_> hi guys. any of you know how to enable 1000Hz while compiling kernel?
[06:43] <kylem> eh?
[06:44] <kylem> make menuconfig -> processor type -> timer frequency
[06:46] <lamont> morning kylem 
[06:47] <movi> kylem -> no_Hz
[06:48] <mjg59> movi: So?
[06:49] <mjg59> It's a point, though - I don't think there's any real downside to 1000Hz on a dyntick system
[07:05] <lamont> find: debian/nic-usb-modules-2.6.22-9-hppa64-di: No such file or directory
[07:05] <lamont> nic-usb-modules-2.6.22-9-hppa64-di will be empty
[07:05] <lamont> make: *** [binary-udebs]  Error 1
[07:05] <lamont> is it trivial to tell the build to not expect that directory?
[07:06] <lamont> for just one kernel image on one arch?
[07:21] <BenC> mjg59: if that were true, wouldn't they disable that option when NOHZ is enabled?
[07:21] <mjg59> BenC: It still has an influence
[07:21] <mjg59> The most likely poor influence would be a slight drop in throughput
[07:22] <mjg59> But there's none of the power-related issues
[07:22] <BenC> what's the gain then?
[07:24] <mjg59> Reduced latency
[07:24] <lamont> timer granularity, polled-event granularity
[07:24] <mjg59> The same reason the -rt people have always wanted CONFIG_1000HZ
[07:24] <BenC> might be worth it for -generic then
[07:25] <lamont> we use 1000 on the telco stuff HP ships