* Christian Leber <[email protected]> wrote: > I wasted a day to track it down, unfortunally it's for me completly > unclear how this commit s related to IDE, it could be some timing > issue, but that is just guessing. yeah, i too suspect that it's timing related. > With 77ba89c5cf28d5d98a3cae17f67a3e42b102cc25 (linux-2.6 tree) i don't > have a problem (i booted it 21 times until now and it allways worked) > and b29739f902ee76a05493fb7d2303490fc75364f4 is the first bad commit. but in case it's not timing related, could you do some more testing, ontop of the 77ba89c5cf28d5d98a3cae17f67a3e42b102cc25 tree? The b29739f902ee76a05493fb7d2303490fc75364f4 patch is quite large, so i have created a finegrained, functional splitup of it: sched-cleanups.patch sched-add-task-rq-lock-ops.patch sched-add-rt-mutex-setprio.patch sched-pi-lock.patch sched-add-new-macros.patch sched-add-normal-prio.patch sched-use-has-rt-policy.patch sched-set-user-nice-fix.patch sched-use-normal-prio.patch and have attached the resulting tarball. (Just extract the tarball into the known-good 77ba89c5cf28d5d98a3cae17f67a3e42b102cc25 tree, it will create a patches/ directory which includes a series file and the patches. If you install quilt then you can do 'quilt push' / 'quilt pop' to navigate in the patch-queue easily. Use 'quilt applied' to see the current patch-stack.) i have ordered the patches within the splitup in a way so that the patches with more impact are at the end. The last patch (sched-use-normal-prio) is the one most likely to cause the problem (but sched-set-user-nice-fix and sched-use-has-rt-policy are possible candidates too). I have verified that every intermediate step builds and boots cleanly as well. Could you try this patch-queue and figure which patch causes the failure you are seeing? That would reduce the search space quite significantly. Ingo
Attachment:
patches.tar.gz
Description: GNU Zip compressed data
- Follow-Ups:
- Re: [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting / pi futex
- From: Christian Leber <[email protected]>
- Re: [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting / pi futex
- References:
- [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting (non-deterministic)
- From: Christian Leber <[email protected]>
- Re: [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting / pi futex
- From: Christian Leber <[email protected]>
- [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting (non-deterministic)
- Prev by Date: Re: [ckrm-tech] [PATCH] BC: resource beancounters (v4) (added user memory)
- Next by Date: Re: kernel BUGs when removing largish files with the SLOB allocator
- Previous by thread: Re: [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting / pi futex
- Next by thread: Re: [BUG] 2.6.18-rc6: hda is allready "IN USE" when booting / pi futex
- Index(es):