Received-SPF: pass (google.com: domain of linux-kernel+bounces-329-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Message-ID: <a121f7ee-6993-4784-a199-e747643d441a@linux.intel.com>
Date: Fri, 15 Dec 2023 09:51:10 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 2/2] iommu/vt-d: don's issue devTLB flush request when
 device is disconnected
To: Baolu Lu <baolu.lu@linux.intel.com>, bhelgaas@google.com,
 dwmw2@infradead.org, will@kernel.org, robin.murphy@arm.com
Cc: linux-pci@vger.kernel.org, iommu@lists.linux.dev,
 linux-kernel@vger.kernel.org, Haorong Ye <yehaorong@bytedance.com>
References: <20231213034637.2603013-1-haifeng.zhao@linux.intel.com>
 <20231213034637.2603013-3-haifeng.zhao@linux.intel.com>
 <96051115-c928-4f3c-bd65-4f3f8e83ca9c@linux.intel.com>
 <e6fddbb0-fa8f-43d4-8a5e-d177e9465912@linux.intel.com>
 <8c8a1e49-5447-4d61-b3ba-782873cc0f4a@linux.intel.com>
 <2f49219e-601d-4ddc-a7b7-6ea8968a2f80@linux.intel.com>
From: Ethan Zhao <haifeng.zhao@linux.intel.com>
In-Reply-To: <2f49219e-601d-4ddc-a7b7-6ea8968a2f80@linux.intel.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit


On 12/15/2023 9:34 AM, Baolu Lu wrote:
> On 2023/12/15 9:03, Ethan Zhao wrote:
>>
>> 2. supprise_removal
>>
>>   Users remove the devece directly or bring the device link down/turn 
>> off
>>
>>   device power first by setting pci config space, link-down/not-present/
>>
>>   power-off are all handled by pciehp the same way 
>> "supprise_removal", in
>>
>>   such case, pciehp_ist() will flag the device as "disconnected" 
>> first, then
>>
>>   unconfig the devcie, unload driver, iommu release device(issing 
>> devTLB flush)
>>
>>   delete device. so we checking the device state could work such cases.
>
> If so, then it is fine for the iommu driver. As Robin said, if the
> device needs more cleanup, the iommu core should register a right
> callback to the driver core and handle it before the device goes away.
>
> Disabling PCI features seems to be a reasonable device cleanup. This
> gives us another reason to move ATS enabling/disabling out from the

For supprise_removal, device was already removed, powered-off, iommu

device-release got notification  or cleanup callback is  invoked to disable

ATS to not-present device etc ,

I didn't get the meaning to do so, perhaps I misunderstand ?

Thanks,

Ethan

> iommu subsystem. Once this is done, the device driver will enable ATS
> during its probe and disable it during its release. There will be no
> such workaround in the iommu driver anymore.
>
> Best regards,
> baolu