Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Mime-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset=UTF-8
Date:   Tue, 06 Jun 2023 19:42:03 +0300
Message-Id: <CT5PU1RV2X20.3AYQOV5CF0EIH@suppilovahvero>
From:   "Jarkko Sakkinen" <jarkko@kernel.org>
To:     "Lino Sanfilippo" <LinoSanfilippo@gmx.de>,
        =?utf-8?q?P=C3=A9ter_Ujfalusi?= <peter.ujfalusi@linux.intel.com>,
        "Lino Sanfilippo" <l.sanfilippo@kunbus.com>, <peterhuewe@gmx.de>,
        <jgg@ziepe.ca>
Cc:     <jsnitsel@redhat.com>, <hdegoede@redhat.com>,
        <oe-lkp@lists.linux.dev>, <lkp@intel.com>, <peterz@infradead.org>,
        <linux@mniewoehner.de>, <linux-integrity@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>, <lukas@wunner.de>,
        <p.rosenberger@kunbus.com>
Subject: Re: [PATCH 1/2] tpm, tpm_tis: Handle interrupt storm
References: <20230522143105.8617-1-LinoSanfilippo@gmx.de>
 <CSTVVFNKUVJW.P69FKI6IF3ZN@suppilovahvero>
 <da435e0d-5f22-fac7-bc10-96a0fd4c6d54@kunbus.com>
 <a84c447f-cdfb-d33c-62cb-bb5d9aa8510b@linux.intel.com>
 <3d350827-795e-8277-a209-1c1d33ca57fa@gmx.de>
In-Reply-To: <3d350827-795e-8277-a209-1c1d33ca57fa@gmx.de>
Precedence: bulk

On Mon May 29, 2023 at 4:15 PM EEST, Lino Sanfilippo wrote:
> > Except that if the line got the spurious designation in core, the
> > interrupt line will be disabled while the TPM driver will think that it
> > is still using IRQ mode and will not switch to polling.
>
> In the case that an interrupt storm cant be detected (since there might n=
ot even
> be one) I am fine with adding a quirk.

Speaking of generic vs quirk (if storm can be detected): detection
should be eager as ever possible. Too eager does not cause systems
failing.

This way I think we can converge to stability fast as possible.

Then, later on, if a system where interrupts should work but the
detection disables interrupts we can either make the fixed threshold
less eager, or add a tuning parameter to sysfs.

BR, Jarkko