xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 0/6] Support calling functions on dedicated physical cpu
@ 2016-03-11 11:59 Juergen Gross
  0 siblings, 0 replies; 5+ messages in thread
From: Juergen Gross @ 2016-03-11 11:59 UTC (permalink / raw)
  To: linux-kernel, xen-devel
  Cc: Juergen Gross, jdelvare, peterz, hpa, x86, mingo, david.vrabel,
	Douglas_Warzecha, pali.rohar, boris.ostrovsky, tglx, linux

Some hardware (e.g. Dell Studio laptops) require special functions to
be called on physical cpu 0 in order to avoid occasional hangs. When
running as dom0 under Xen this could be achieved only via special boot
parameters (vcpu pinning) limiting the hypervisor in it's scheduling
decisions.

This patch series is adding a generic function to be able to temporarily
pin a (virtual) cpu to a dedicated physical cpu for executing above
mentioned functions on that specific cpu. The drivers (dcdbas and i8k)
requiring this functionality are modified accordingly.

Juergen Gross (6):
  xen: sync xen header
  sched: add function to execute a function synchronously on a physical
    cpu
  dcdbas: make use of call_sync_on_phys_cpu()
  hwmon: use call_sync_on_phys_cpu() for dell-smm i8k
  virt, sched: add cpu pinning to call_sync_on_phys_cpu()
  xen: add xen_pin_vcpu() to support calling functions on a dedicated
    pcpu

 arch/x86/include/asm/hypervisor.h |   9 ++++
 arch/x86/xen/enlighten.c          |  40 +++++++++++++++
 drivers/firmware/dcdbas.c         |  46 ++++++++----------
 drivers/hwmon/dell-smm-hwmon.c    |  27 +++++-----
 include/linux/sched.h             |  31 ++++++++++++
 include/xen/interface/sched.h     | 100 +++++++++++++++++++++++++++++++-------
 kernel/sched/core.c               |  30 ++++++++++++
 7 files changed, 224 insertions(+), 59 deletions(-)

-- 
2.6.2


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/6] Support calling functions on dedicated physical cpu
  2016-03-11 13:15   ` One Thousand Gnomes
@ 2016-03-11 13:19     ` Peter Zijlstra
  0 siblings, 0 replies; 5+ messages in thread
From: Peter Zijlstra @ 2016-03-11 13:19 UTC (permalink / raw)
  To: One Thousand Gnomes
  Cc: Juergen Gross, x86, jdelvare, hpa, linux-kernel, mingo,
	david.vrabel, Douglas_Warzecha, pali.rohar, xen-devel,
	boris.ostrovsky, tglx, linux

On Fri, Mar 11, 2016 at 01:15:04PM +0000, One Thousand Gnomes wrote:
> On Fri, 11 Mar 2016 13:25:14 +0100
> Peter Zijlstra <peterz@infradead.org> wrote:
> 
> > On Fri, Mar 11, 2016 at 12:59:28PM +0100, Juergen Gross wrote:
> > > Some hardware (e.g. Dell Studio laptops) require special functions to
> > > be called on physical cpu 0 in order to avoid occasional hangs. When
> > > running as dom0 under Xen this could be achieved only via special boot
> > > parameters (vcpu pinning) limiting the hypervisor in it's scheduling
> > > decisions.  
> > 
> > So instead of telling Dell to get their act together and fix their damn
> > firmware, we're going to add the most horrid gunk to the kernel? How
> > does that make sense?
> 
> It's been normal forever. The convention with a lot of older BIOS crap
> was always that it should be called on the boot CPU (APM. PnPBIOS etc).
> 
> It's a stretch pre-EFI to even call it a "bug"

Yeah, I knew about the APM/PnP muck, but I was under the impression this
was about new hardware.


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/6] Support calling functions on dedicated physical cpu
       [not found] ` <20160311122514.GA6344@twins.programming.kicks-ass.net>
  2016-03-11 12:28   ` Pali Rohár
@ 2016-03-11 13:15   ` One Thousand Gnomes
  2016-03-11 13:19     ` Peter Zijlstra
  1 sibling, 1 reply; 5+ messages in thread
From: One Thousand Gnomes @ 2016-03-11 13:15 UTC (permalink / raw)
  To: Peter Zijlstra
  Cc: Juergen Gross, x86, jdelvare, hpa, linux-kernel, mingo,
	david.vrabel, Douglas_Warzecha, pali.rohar, xen-devel,
	boris.ostrovsky, tglx, linux

On Fri, 11 Mar 2016 13:25:14 +0100
Peter Zijlstra <peterz@infradead.org> wrote:

> On Fri, Mar 11, 2016 at 12:59:28PM +0100, Juergen Gross wrote:
> > Some hardware (e.g. Dell Studio laptops) require special functions to
> > be called on physical cpu 0 in order to avoid occasional hangs. When
> > running as dom0 under Xen this could be achieved only via special boot
> > parameters (vcpu pinning) limiting the hypervisor in it's scheduling
> > decisions.  
> 
> So instead of telling Dell to get their act together and fix their damn
> firmware, we're going to add the most horrid gunk to the kernel? How
> does that make sense?

It's been normal forever. The convention with a lot of older BIOS crap
was always that it should be called on the boot CPU (APM. PnPBIOS etc).

It's a stretch pre-EFI to even call it a "bug"

Alan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/6] Support calling functions on dedicated physical cpu
       [not found] ` <20160311122514.GA6344@twins.programming.kicks-ass.net>
@ 2016-03-11 12:28   ` Pali Rohár
  2016-03-11 13:15   ` One Thousand Gnomes
  1 sibling, 0 replies; 5+ messages in thread
From: Pali Rohár @ 2016-03-11 12:28 UTC (permalink / raw)
  To: Peter Zijlstra
  Cc: Juergen Gross, jdelvare, x86, linux-kernel, mingo, david.vrabel,
	Douglas_Warzecha, hpa, xen-devel, boris.ostrovsky, tglx, linux

On Friday 11 March 2016 13:25:14 Peter Zijlstra wrote:
> On Fri, Mar 11, 2016 at 12:59:28PM +0100, Juergen Gross wrote:
> > Some hardware (e.g. Dell Studio laptops) require special functions to
> > be called on physical cpu 0 in order to avoid occasional hangs. When
> > running as dom0 under Xen this could be achieved only via special boot
> > parameters (vcpu pinning) limiting the hypervisor in it's scheduling
> > decisions.
> 
> So instead of telling Dell to get their act together and fix their damn
> firmware, we're going to add the most horrid gunk to the kernel? How
> does that make sense?

Hi Peter! That problem is for old laptops and I doubt that Dell would
fix and distribute patches for 5-15 years old laptops...

-- 
Pali Rohár
pali.rohar@gmail.com

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 0/6] Support calling functions on dedicated physical cpu
       [not found] <1457697574-6710-1-git-send-email-jgross@suse.com>
@ 2016-03-11 12:25 ` Peter Zijlstra
       [not found] ` <20160311122514.GA6344@twins.programming.kicks-ass.net>
  1 sibling, 0 replies; 5+ messages in thread
From: Peter Zijlstra @ 2016-03-11 12:25 UTC (permalink / raw)
  To: Juergen Gross
  Cc: x86, jdelvare, hpa, linux-kernel, mingo, david.vrabel,
	Douglas_Warzecha, pali.rohar, xen-devel, boris.ostrovsky, tglx,
	linux

On Fri, Mar 11, 2016 at 12:59:28PM +0100, Juergen Gross wrote:
> Some hardware (e.g. Dell Studio laptops) require special functions to
> be called on physical cpu 0 in order to avoid occasional hangs. When
> running as dom0 under Xen this could be achieved only via special boot
> parameters (vcpu pinning) limiting the hypervisor in it's scheduling
> decisions.

So instead of telling Dell to get their act together and fix their damn
firmware, we're going to add the most horrid gunk to the kernel? How
does that make sense?

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2016-03-11 13:19 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-03-11 11:59 [PATCH 0/6] Support calling functions on dedicated physical cpu Juergen Gross
     [not found] <1457697574-6710-1-git-send-email-jgross@suse.com>
2016-03-11 12:25 ` Peter Zijlstra
     [not found] ` <20160311122514.GA6344@twins.programming.kicks-ass.net>
2016-03-11 12:28   ` Pali Rohár
2016-03-11 13:15   ` One Thousand Gnomes
2016-03-11 13:19     ` Peter Zijlstra

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).