Re: [PATCH 0/3] cpu freq: fix problems with work_on_cpu usage in acpi-cpufreq [PULL request]

[Mike Travis <travis@sgi.com>]

Mike Travis wrote:
> This set of patches improves the work_on_cpu() function to
> eliminate circular lock dependencies from occurring by requiring
> the caller to insure the cpu does not go offline.  This effectively
> models the previous behaviour using set_cpus_allowed().
> 
> The other fix is to use a separate work queue for work_on_cpu so
> it does not clash with kevent items that might already be on the
> queue from the same caller.
> 
> This fixes the boot up and suspend/resume to disk problems previously
> seen.
> 
> Signed-off-by: Mike Travis <travis@sgi.com>
> Cc: Dieter Ries <clip2@gmx.de>
> Cc: Maciej Rutecki <maciej.rutecki@gmail.com>
> Cc: Dave Jones <davej@redhat.com>
> Cc: <cpufreq@vger.kernel.org>
> 

Hi Ingo,

If you are ready for these, I've pushed them to the cpus4096-for-ingo
tree.  (Note it's been renamed.)

Thanks!
Mike
--- 

The following changes since commit c99dbbe9f8f6b3e9383e64710217e873431d1c31:
  Mike Travis (1):
        sched: fix warning on ia64

are available in the git repository at:

  ssh://master.kernel.org/pub/scm/linux/kernel/git/travis/linux-2.6-cpus4096-for-ingo.git master

Mike Travis (1):
      cpufreq: use work_on_cpu in acpi-cpufreq.c for drv_read and drv_write

Rusty Russell (2):
      work_on_cpu: don't try to get_online_cpus() in work_on_cpu.
      work_on_cpu: Use our own workqueue.

 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c |   22 +++++++++-------------
 kernel/workqueue.c                         |   20 ++++++++++----------
 2 files changed, 19 insertions(+), 23 deletions(-)