Philippe Gerum wrote:
Jan Kiszka wrote:
Philippe Gerum wrote:
Philippe Gerum wrote:
Bernhard Walle wrote:
Hello,
I have a kernel 2.6.15.6 with
http://download.gna.org/adeos/patches/v2.6/i386/adeos-ipipe-2.6.15-i386-1.2-03.patch
patch. In this configuration, CONFIG_PREEMPT doesn't work any more,
i.e. the kernel is not preempted.
How do you check this?
Forget about this question. I'm going to give a look at your testcase
when time allows. Thanks.
I did, and I can confirm this (2.6.16, original ipipe-1.2-03 applied,
ipipe on vs. off). Maybe something related to the TASK_ATOMICSWITCH
changes?
Don't know yet, I'm rebuilding an instrumented kernel, but considering
that we don't even have local echo of keystrokes, I'd rather think of
IRQs lagging in the pipeline log. More news later.
Er, the softlockup detector triggers while the module holds the CPU, so
the timer top-half must be running properly. Something is possibly fishy
in the IRQ return path for x86 from resume_kernel and beyond.
With the kernel without ADEOS patch, it works fine. A test program is
at http://www.bwalle.de/temp/kern-preempt.tar.bz2.
Do you know what's wrong?
Regards,
Bernhard
Jan
------------------------------------------------------------------------
_______________________________________________
Adeos-main mailing list
[email protected]
https://mail.gna.org/listinfo/adeos-main
--
Philippe.
_______________________________________________
Adeos-main mailing list
[email protected]
https://mail.gna.org/listinfo/adeos-main
