PROBLEM: tickless scheduling
in [Kernel]
|
Prev: staging: hv: Optimize adj_guesttime function and add more detailed comments
Next: s3c-fb: added patch series for s5pv210 and some features.
From: john stultz on 12 May 2010 17:00 On Wed, 2010-05-12 at 15:54 -0400, Donald Allen wrote: > Before doing what you asked, I ran my home-brew backup script, which > tars up the whole machine, save my home directory, to a big sata drive > in a usb shoebox. I did so by booting the most recent Arch Linux > install/live CD (while I use Slackware, the Slackware install CDs are > not suitable for this sort of thing, having very old versions of > things like tar). While doing so, I observed exactly the same symptoms > I did with the Slackware 13.1 install, which I described in the bug > report. So rather than messing with the hard-won custom kernel that I > now have installed on the netbook, I am attaching the various things > from /proc gathered with the Arch kernel running. Yes, it's a somewhat > older kernel (2.6.30), but I'm hoping that things haven't changed much > in tickless land. If that's not the case, then I will attempt to > reproduce this with the newer kernel on the Slackware 13.1 install > DVD. Let me know if you need me to do this. The attached tar file is > bzip2-compressed, so you want xjf to extract. Hmm.. Sorry, but I have another quick request. Could you send your /proc/interrupts output from the kernel having the problem? thanks -john -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Donald Allen on 15 May 2010 13:20 On Wed, May 12, 2010 at 4:52 PM, john stultz <johnstul(a)us.ibm.com> wrote: > On Wed, 2010-05-12 at 15:54 -0400, Donald Allen wrote: >> Before doing what you asked, I ran my home-brew backup script, which >> tars up the whole machine, save my home directory, to a big sata drive >> in a usb shoebox. I did so by booting the most recent Arch Linux >> install/live CD (while I use Slackware, the Slackware install CDs are >> not suitable for this sort of thing, having very old versions of >> things like tar). While doing so, I observed exactly the same symptoms >> I did with the Slackware 13.1 install, which I described in the bug >> report. So rather than messing with the hard-won custom kernel that I >> now have installed on the netbook, I am attaching the various things >> from /proc gathered with the Arch kernel running. Yes, it's a somewhat >> older kernel (2.6.30), but I'm hoping that things haven't changed much >> in tickless land. If that's not the case, then I will attempt to >> reproduce this with the newer kernel on the Slackware 13.1 install >> DVD. Let me know if you need me to do this. The attached tar file is >> bzip2-compressed, so you want xjf to extract. > > Hmm.. Sorry, but I have another quick request. Could you send > your /proc/interrupts output from the kernel having the problem? Attached. This is from the 2.6.30 kernel on the Arch Linux install cd. Here's another bit of data. As I've said previously, the problems I'm reporting were observed on a Toshiba NB310-305 netbook with a single-core Atom 450 processor. I just built myself a mini-ITX system using the Intel D510MO motherboard, which provides a dual-core D510 Atom processor. The other hardware on the board is similar to the Toshiba. I installed the same Slackware snapshot I used on the Toshiba, and did the home directory transfer without any problem at all with the default tickless kernel. The hardware isn't identical, and while I don't know the internals of the Linux kernel at all, my gut, backed up by many years of OS development work in scheduling and memory management, is telling me that the key difference is dual- vs. single-core. Just a guess. Hope this helps -- /Don > > thanks > -john > > >
From: Stefan Biereigel on 16 May 2010 12:20 > Attached. This is from the 2.6.30 kernel on the Arch Linux install cd. > > Here's another bit of data. As I've said previously, the problems I'm > reporting were observed on a Toshiba NB310-305 netbook with a > single-core Atom 450 processor. I just built myself a mini-ITX system > using the Intel D510MO motherboard, which provides a dual-core D510 > Atom processor. The other hardware on the board is similar to the > Toshiba. I installed the same Slackware snapshot I used on the > Toshiba, and did the home directory transfer without any problem at > all with the default tickless kernel. The hardware isn't identical, > and while I don't know the internals of the Linux kernel at all, my > gut, backed up by many years of OS development work in scheduling and > memory management, is telling me that the key difference is dual- vs. > single-core. Just a guess. > > Hope this helps -- > > /Don > Hello everyone, I hope I can add something here, because I am experiencing the exact same Problem as Don describes. I'm running a PackardBell EasyNote MB89 featuring a Core 2 Duo, 4 GB of RAM, Intel Chipset (Santa Rosa), SATA HDD. My Machine even hangs at boot, absolutely doing nothing until i wiggle the touchpad. This is definately reproduceable, but I think it doesn't occur that often in X-Window-System, if X is off I can just wait a couple of seconds and there I go. I compiled other kernels myself with tickless disabled and Ticks set to various values (250, 1000) which completely resolved my problem. If you want, I can get you some Output/logs/whatever because I'm fixed to using this kernel ATM (which doesn't really hurt because I use X and son't shut down my Notebook that often). Even though I don't know what exact kernel-version this problem brings (using another machine ATM from vacation) I can say that I existed since Opensuse 11.1 I guess, so maybe 2.6.30 and above. Hope I can help -- Stefan -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Arjan van de Ven on 16 May 2010 16:40 On Sun, 16 May 2010 18:14:07 +0200 Stefan Biereigel <security(a)biereigel-wb.de> wrote: > Hello everyone, > > I hope I can add something here, because I am experiencing the exact > same Problem as Don describes. I'm running a PackardBell EasyNote > MB89 featuring a Core 2 Duo, 4 GB of RAM, Intel Chipset (Santa Rosa), > SATA HDD. My Machine even hangs at boot, absolutely doing nothing > until i wiggle the touchpad. This is definately reproduceable, but I > think it doesn't occur that often in X-Window-System, if X is off I > can just wait a couple of seconds and there I go. this has nothing to do with tickless or not... you're losing interrupts. if you're not tickless, there's just so many regular interrupts happening that you don't notice that other interrupts are lost. try disabling MSI (iirc pci=nomsi) to see if that helps.. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Thomas Gleixner on 16 May 2010 19:40
Donald, On Sat, 15 May 2010, Donald Allen wrote: > Attached. This is from the 2.6.30 kernel on the Arch Linux install cd. > > Here's another bit of data. As I've said previously, the problems I'm > reporting were observed on a Toshiba NB310-305 netbook with a > single-core Atom 450 processor. I just built myself a mini-ITX system > using the Intel D510MO motherboard, which provides a dual-core D510 > Atom processor. The other hardware on the board is similar to the > Toshiba. I installed the same Slackware snapshot I used on the > Toshiba, and did the home directory transfer without any problem at > all with the default tickless kernel. The hardware isn't identical, > and while I don't know the internals of the Linux kernel at all, my > gut, backed up by many years of OS development work in scheduling and > memory management, is telling me that the key difference is dual- vs. > single-core. Just a guess. I fear you are wrong. The key difference is almost certainly that the BIOS of your netbook tries to be overly clever vs. power management and is not aware of the fact that the Linux kernel uses timer hardware in a very different way than the other OS which comes preinstalled on that machine. The overly clever BIOS power management which works nicely with the vendor provided "drivers" for the other OS is just interfering with the kernels way of dealing with the problem. Can you please boot with "hpet=disable" on the kernel command line ? Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |