mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Lorenzo Pieralisi <lpieralisi@kernel.org>
To: Fuad Tabba <fuad.tabba@linux.dev>
Cc: Sascha Bischoff <sascha.bischoff@arm.com>,
	Marc Zyngier <maz@kernel.org>, Thomas Gleixner <tglx@kernel.org>,
	Radu Rendec <radu@rendec.net>,
	Timothy Hayes <timothy.hayes@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Jonathan Cameron <jic23@kernel.org>,
	"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 04/14] irqchip/gic-v5: Retain the L2 IST when mapping times out
Date: Thu, 19 Nov 2026 14:32:37 +0100	[thread overview]
Message-ID: <av76dQ_dMCWlTAse@red-moon> (raw)
In-Reply-To: <CA+EHjTzPFjUOXCiccswH2xGvbiqgXPbSihov7MA0WuXn-SmrEg@mail.gmail.com>

On Thu, Nov 19, 2026 at 12:00:15PM +0000, Fuad Tabba wrote:
> Hi Sascha,
> 
> On Wed, 18 Nov 2026 17:35:33 +0000, Sascha Bischoff
> <sascha.bischoff@arm.com> wrote:
> [...]
> > diff --git a/drivers/irqchip/irq-gic-v5-irs.c b/drivers/irqchip/irq-gic-v5-irs.c
> [...]
> > @@ -217,6 +217,10 @@ int gicv5_irs_iste_alloc(const u32 lpi)
> >          if (FIELD_GET(GICV5_ISTL1E_VALID, le64_to_cpu(l1ist[index])))
> >                  return 0;
> >
> > +        /* A previous map timed out; do not overwrite its pending L1 entry. */
> > +        if (READ_ONCE(l1ist[index]))
> > +                return -EBUSY;
> 
> Sashiko's review [1] didn't go on the list, in case you missed it.
> Sashiko found that l1ist[index] is read twice, and that on a

Yes, we should read it once but see below.

> non-coherent IRS a map completing after the timeout leaves a stale
> copy of the L1 entry in the cache for later allocations. Invalidating
> the entry before it's first read would fix the latter. Invalidating it
> on the timeout path wouldn't, since a later read can refill the line
> before the map completes.

Well, we don't know when the map completes do we ? Is invalidating
just before it is READ_ONCE() really "fixing" this (actually, is what
we are trying to do minimizing the false negatives ie l1 entries
that we wrote and became valid that we would detect as not valid ?)
or we need to sync the IRS and check for idle before the invalidation ?

I will give it more thought - I don't like this complexity much for
basically fixing a very close to fatal error.

Thanks,
Lorenzo

> With those fixed:
> Reviewed-by: Fuad Tabba <fuad.tabba@linux.dev>
> 
> Cheers,
> /fuad
> 
> [1] https://sashiko.dev/#/patchset/20261118-host_gicv5_fixes-v3-0-13c1859e8aa3%40arm.com

  reply	other threads:[~2026-11-19 13:32 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-18 17:35 [PATCH v3 00/14] irqchip/gic-v5: Fix misc IRS and ITS host driver issues Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 01/14] irqchip/gic-v5: Reject SPIs not backed by an IRS Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 02/14] irqchip/gic-v5: Only set IRS DPS when 1ofN routing is supported Sascha Bischoff
2026-11-19 11:50   ` Fuad Tabba
2026-11-18 17:35 ` [PATCH v3 03/14] irqchip/gic-v5: Serialise IRS synchronisation requests Sascha Bischoff
2026-11-19 11:53   ` Fuad Tabba
2026-11-18 17:35 ` [PATCH v3 04/14] irqchip/gic-v5: Retain the L2 IST when mapping times out Sascha Bischoff
2026-11-19 12:00   ` Fuad Tabba
2026-11-19 13:32     ` Lorenzo Pieralisi [this message]
2026-11-18 17:35 ` [PATCH v3 05/14] irqchip/gic-v5: Serialise ITS invalidation commands Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 06/14] irqchip/gic-v5: Use linear DT and ITT when L2 covers all IDs Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 07/14] irqchip/gic-v5: Honour ACPI ITS noncoherent flag Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 08/14] irqchip/gic-v5: Use firmware node name in ITS warning Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 09/14] irqchip/gic-v5: Invalidate ITS device cache before freeing ITT Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 10/14] irqchip/gic-v5: Clean the allocated linear device table size Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 11/14] irqchip/gic-v5: Release ITS MMIO region on DT probe failure Sascha Bischoff
2026-11-18 17:35 ` [PATCH v3 12/14] irqchip/gic-v5: Retain the ITT when device invalidation times out Sascha Bischoff
2026-11-19 10:28   ` Lorenzo Pieralisi
2026-11-19 12:07     ` Fuad Tabba
2026-11-18 17:35 ` [PATCH v3 13/14] irqchip/gic-v5: Clean a level 2 device table before making it valid Sascha Bischoff
2026-11-19 10:19   ` Lorenzo Pieralisi
2026-11-18 17:35 ` [PATCH v3 14/14] irqchip/gic-v5: Invalidate the L1 DTE when allocating an L2 device table Sascha Bischoff
2026-11-19 10:13   ` Lorenzo Pieralisi
2026-11-19 12:32     ` Fuad Tabba
2026-11-19 13:36       ` Lorenzo Pieralisi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=av76dQ_dMCWlTAse@red-moon \
    --to=lpieralisi@kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=fuad.tabba@linux.dev \
    --cc=jic23@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maz@kernel.org \
    --cc=radu@rendec.net \
    --cc=rafael.j.wysocki@intel.com \
    --cc=sascha.bischoff@arm.com \
    --cc=tglx@kernel.org \
    --cc=timothy.hayes@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®