Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Thomas Gleixner <tglx@linutronix.de>
To:     James Bottomley <James.Bottomley@HansenPartnership.com>,
        Jarkko Sakkinen <jarkko@kernel.org>,
        Jerry Snitselaar <jsnitsel@redhat.com>
Cc:     linux-kernel@vger.kernel.org, linux-integrity@vger.kernel.org,
        intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
        kernel test robot <lkp@intel.com>,
        Jason Gunthorpe <jgg@ziepe.ca>,
        Peter Huewe <peterhuewe@gmx.de>,
        Matthew Garrett <mjg59@google.com>,
        Hans de Goede <hdegoede@redhat.com>,
        Jiri Kosina <jkosina@suse.cz>,
        Yunfeng Ye <yeyunfeng@huawei.com>,
        David Airlie <airlied@linux.ie>,
        Daniel Vetter <daniel@ffwll.ch>, jani.nikula@linux.intel.com,
        joonas.lahtinen@linux.intel.com
Subject: Re: [PATCH v3 1/4] irq: export kstat_irqs
In-Reply-To: <7fe2d70b5daca57d9cd50d3877376e420658ed71.camel@HansenPartnership.com>
References: <20201205014340.148235-1-jsnitsel@redhat.com> <20201205014340.148235-2-jsnitsel@redhat.com> <20201205103954.GA17088@kernel.org> <87360i99ck.fsf@nanos.tec.linutronix.de> <7fe2d70b5daca57d9cd50d3877376e420658ed71.camel@HansenPartnership.com>
Date:   Sun, 06 Dec 2020 20:29:00 +0100
Message-ID: <87r1o27myr.fsf@nanos.tec.linutronix.de>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Sun, Dec 06 2020 at 09:40, James Bottomley wrote:
> On Sun, 2020-12-06 at 17:40 +0100, Thomas Gleixner wrote:
>> On Sat, Dec 05 2020 at 12:39, Jarkko Sakkinen wrote:
>> > On Fri, Dec 04, 2020 at 06:43:37PM -0700, Jerry Snitselaar wrote:
>> > > To try and detect potential interrupt storms that
>> > > have been occurring with tpm_tis devices it was suggested
>> > > to use kstat_irqs() to get the number of interrupts.
>> > > Since tpm_tis can be built as a module it needs kstat_irqs
>> > > exported.
>> > 
>> > I think you should also have a paragraph explicitly stating that
>> > i915_pmu.c contains a duplicate of kstat_irqs() because it is not
>> > exported as of today. It adds a lot more weight to this given that
>> > there is already existing mainline usage (kind of).
>> 
>> It's abusage and just the fact that it exists is not an argument by
>> itself.
>
> What we want is a count of the interrupts to see if we're having an
> interrupt storm from the TPM device (some seem to be wired to fire the
> interrupt even when there's no event to warrant it).  Since
> kstat_irqs_user() does the correct RCU locking, should we be using that
> instead?

If we need to export it, yes. But I still have to understand the
value. See my other reply.

Thanks,

        tglx