netdev - Re: [PATCH tip:irq/core v1] genirq: remove auto-set of the mask when setting the hint

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 20 May 2021 20:03:38 -0400
From:   Nitesh Lal <nilal@...hat.com>
To:     Thomas Gleixner <tglx@...utronix.de>,
        Jesse Brandeburg <jesse.brandeburg@...el.com>,
        Robin Murphy <robin.murphy@....com>,
        Marcelo Tosatti <mtosatti@...hat.com>
Cc:     Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org,
        intel-wired-lan@...ts.osuosl.org, jbrandeb@...nel.org,
        "frederic@...nel.org" <frederic@...nel.org>,
        "juri.lelli@...hat.com" <juri.lelli@...hat.com>,
        Alex Belits <abelits@...vell.com>,
        "linux-api@...r.kernel.org" <linux-api@...r.kernel.org>,
        "bhelgaas@...gle.com" <bhelgaas@...gle.com>,
        "linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
        "rostedt@...dmis.org" <rostedt@...dmis.org>,
        "peterz@...radead.org" <peterz@...radead.org>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
        "sfr@...b.auug.org.au" <sfr@...b.auug.org.au>,
        "stephen@...workplumber.org" <stephen@...workplumber.org>,
        "rppt@...ux.vnet.ibm.com" <rppt@...ux.vnet.ibm.com>,
        "jinyuqi@...wei.com" <jinyuqi@...wei.com>,
        "zhangshaokun@...ilicon.com" <zhangshaokun@...ilicon.com>,
        netdev@...r.kernel.org, chris.friesen@...driver.com,
        Marc Zyngier <maz@...nel.org>,
        Neil Horman <nhorman@...driver.com>, pjwaskiewicz@...il.com
Subject: Re: [PATCH tip:irq/core v1] genirq: remove auto-set of the mask when
 setting the hint

On Thu, May 20, 2021 at 5:57 PM Nitesh Lal <nilal@...hat.com> wrote:
>
> On Mon, May 17, 2021 at 8:23 PM Nitesh Lal <nilal@...hat.com> wrote:
> >
> > On Mon, May 17, 2021 at 8:04 PM Thomas Gleixner <tglx@...utronix.de> wrote:
> > >
> > > On Mon, May 17 2021 at 18:44, Nitesh Lal wrote:
> > > > On Mon, May 17, 2021 at 4:48 PM Thomas Gleixner <tglx@...utronix.de> wrote:
> > > >> The hint was added so that userspace has a better understanding where it
> > > >> should place the interrupt. So if irqbalanced ignores it anyway, then
> > > >> what's the point of the hint? IOW, why is it still used drivers?
> > > >>
> > > > Took a quick look at the irqbalance repo and saw the following commit:
> > > >
> > > > dcc411e7bf    remove affinity_hint infrastructure
> > > >
> > > > The commit message mentions that "PJ is redesiging how affinity hinting
> > > > works in the kernel, the future model will just tell us to ignore an IRQ,
> > > > and the kernel will handle placement for us.  As such we can remove the
> > > > affinity_hint recognition entirely".
> > >
> > > No idea who PJ is. I really love useful commit messages. Maybe Neil can
> > > shed some light on that.
> > >
> > > > This does indicate that apparently, irqbalance moved away from the usage of
> > > > affinity_hint. However, the next question is what was this future
> > > > model?
> > >
> > > I might have missed something in the last 5 years, but that's the first
> > > time I hear about someone trying to cleanup that thing.
> > >
> > > > I don't know but I can surely look into it if that helps or maybe someone
> > > > here already knows about it?
> > >
> > > I CC'ed Neil :)
> >
> > Thanks, I have added PJ Waskiewicz as well who I think was referred in
> > that commit message as PJ.
> >
> > >
> > > >> Now there is another aspect to that. What happens if irqbalanced does
> > > >> not run at all and a driver relies on the side effect of the hint
> > > >> setting the initial affinity. Bah...
> > > >>
> > > >
> > > > Right, but if they only rely on this API so that the IRQs are spread across
> > > > all the CPUs then that issue is already resolved and these other drivers
> > > > should not regress because of changing this behavior. Isn't it?
> > >
> > > Is that true for all architectures?
> >
> > Unfortunately, I don't know and that's probably why we have to be careful.
>
> I think here to ensure that we are not breaking any of the drivers we have
> to first analyze all the existing drivers and understand how they are using
> this API.
> AFAIK there are three possible scenarios:
>
> - A driver use this API to spread the IRQs
>   + For this case we should be safe considering the spreading is naturally
>     done from the IRQ subsystem itself.

Forgot to mention another thing in the above case is to determine whether
it is true for all architectures or not as Thomas mentioned.

>
> - A driver use this API to actually set the hint
>   + These drivers should have no functional impact because of this revert
>
> - Driver use this API to force a certain affinity mask
>   + In this case we have to replace the API with the irq_force_affinity()
>
> I can start looking into the individual drivers, however, testing them will
> be a challenge.
>
> Any thoughts?
>
> --
> Thanks
> Nitesh



-- 
Thanks
Nitesh