[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <34dd7c2e-b6db-684f-f0a2-73f2e6951308@huawei.com>
Date: Thu, 9 Apr 2020 08:02:20 +0100
From: John Garry <john.garry@...wei.com>
To: Robin Murphy <robin.murphy@....com>,
"will@...nel.org" <will@...nel.org>,
"mark.rutland@....com" <mark.rutland@....com>
CC: "linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
Shameerali Kolothum Thodi
<shameerali.kolothum.thodi@...wei.com>,
"harb@...erecomputing.com" <harb@...erecomputing.com>,
"tuanphan@...amperecomputing.com" <tuanphan@...amperecomputing.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH] perf/smmuv3: Fix shared interrupt handling
On 08/04/2020 17:49, Robin Murphy wrote:
> IRQF_SHARED is dangerous, since it allows other agents to retarget the
> IRQ's affinity without migrating PMU contexts to match, breaking the way
> in which perf manages mutual exclusion for accessing events. Although
> this means it's not realistically possible to support PMU IRQs being
> shared with other drivers, we *can* handle sharing between multiple PMU
> instances with some explicit affinity bookkeeping and manual interrupt
> multiplexing.
Hi Robin,
Out of curiosity, do we even need to support shared interrupts for any
implementations today?
D06 board:
john@...ntu:~$ more /proc/interrupts | grep smmuv3-pmu
989: 0 0 0 0 ITS-pMSI 133120 Edge smmuv3-pmu
990: 0 0 0 0 ITS-pMSI 135168 Edge smmuv3-pmu
991: 0 0 0 0 ITS-pMSI 137216 Edge smmuv3-pmu
992: 0 0 0 0 ITS-pMSI 139264 Edge smmuv3-pmu
993: 0 0 0 0 ITS-pMSI 141312 Edge smmuv3-pmu
994: 0 0 0 0 ITS-pMSI 143360 Edge smmuv3-pmu
995: 0 0 0 0 ITS-pMSI 145408 Edge smmuv3-pmu
996: 0 0 0 0 ITS-pMSI 147456 Edge smmuv3-pmu
Thanks,
John
Powered by blists - more mailing lists