Re: [Xen-devel] Re: [PATCH] irq: Exclude percpu IRQs from being fixed up

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Thu, 2011-02-17 at 07:52 +0000, Jan Beulich wrote:
> >>> On 17.02.11 at 01:15, Fengzhe Zhang <fengzhe.zhang@xxxxxxxxx> wrote:
> > IRQ_PER_CPU switch is not found in current Kconfig. I'm not sure if this 
> 
> kernel/irq/Kconfig (introduced as a generic option in 2.6.38-rc2). In
> prior kernel you'd have to add a respective Kconfig item in
> drivers/xen/Kconfig.

Also this should be fixed in mainline _before_ being considered for
backporting to the xen/stable-2.6.32.x branch, otherwise it will simply
come back one day when the stable branch moves forward...

> > feature is going to be brought back in the short term. I remove the 
> > ifdef to set IRQ_PER_CPU flag in desc by default but still leave the IRQ 
> > handling logic unchanged. This is a temporary solution to fix system 
> > crash on poweroff. And this is the fix with minimum impact among the 
> > several solutions we tried.
> 
> But it's more a hack than a fix.

Agreed, it seems to take a very narrow approach to a specific failure
without looking at the bigger picture.

>  And making per-CPU IRQs properly
> treated as such isn't a bad idea in any case, I would say.

In addition changing generic code, which also has an impact on native,
in this way needs a lot more rationale in the commit message as to why
it is correct for both Xen and native. Most importantly it needs to go
via the x86 maintainers and not the Xen maintainers.

It also need to be made very clear why the semantics which are required
for this specific lock (lock_kicker_irq) are correct and desirable for
_every_  IRQ_PER_CPU (aka IRQF_PERCPU) lock on x86. The description of
this patch does not do this.

How does this change tie in with the existing mainline IRQF_NO_SUSPEND
flag (which Xen uses on these IPI IRQs) and the IRQF_FORCE_RESUME flag
currently in the tip tree (intended for 2.6.39, I believe)?

Ian.


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel