[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <5f6f0c575aa00e61ec2a6a3b90df8e20a71b5900@linux.dev>
Date: Wed, 12 Mar 2025 10:28:30 +0000
From: "Yajun Deng" <yajun.deng@...ux.dev>
To: "Thomas Gleixner" <tglx@...utronix.de>
Cc: linux-kernel@...r.kernel.org
Subject: Re: [PATCH] genirq: Keep affinity_hint unchanged if it has a value
March 12, 2025 at 5:10 PM, "Thomas Gleixner" <tglx@...utronix.de mailto:tglx@...utronix.de?to=%22Thomas%20Gleixner%22%20%3Ctglx%40linutronix.de%3E > wrote:
>
> On Wed, Mar 12 2025 at 03:20, Yajun Deng wrote:
>
> >
> > March 11, 2025 at 10:14 PM, "Thomas Gleixner" <tglx@...utronix.de> wrote:
> >
> > >
> > > > For example, if affinity_hint points to smp_affinity, smp_affinity
> > > > is changed by the user, and affinity_hint would chang. affinity_hint
> > > > couldn't as a hint to the user, it should keep the value if it has.
> > >
> > > What is 'smp_affinity'?
> > >
> > It's 'desc->irq_common_data.affinity'.
> >
> Which has no business to end up in the affinity hint.
>
>
> >
> > >
> > > I really fail to understand the problem you are trying to solve.
> > >
> > irq_update_affinity_hint(irq, desc->irq_common_data.affinity);
> >
> > As in this code, this means affinity_hint points to smp_affinity.
> >
> This code does not exist and no driver should ever invoke
> irq_update_affinity_hint() that way. The hint is a cpu mask managed by
> the driver and not something the driver pulls out of a data structure
> which it has no business to fiddle with.
>
Okay, I got it. Thank you!
> Thanks
>
> tglx
>
Powered by blists - more mailing lists