Received-SPF: pass (google.com: domain of linux-kernel+bounces-12308-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <64c77298-dd3e-4102-a9d3-0433708d33ac@linux.intel.com>
Date: Thu, 28 Dec 2023 07:31:35 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [RFC PATCH v8 4/5] iommu/vt-d: don't issue device-TLB invalidate
 request when device is disconnected
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: bhelgaas@google.com, baolu.lu@linux.intel.com, dwmw2@infradead.org,
 will@kernel.org, robin.murphy@arm.com, lukas@wunner.de,
 linux-pci@vger.kernel.org, iommu@lists.linux.dev,
 linux-kernel@vger.kernel.org
References: <20231227131151.GA1499234@bhelgaas>
From: Ethan Zhao <haifeng.zhao@linux.intel.com>
In-Reply-To: <20231227131151.GA1499234@bhelgaas>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit


On 12/27/2023 9:11 PM, Bjorn Helgaas wrote:
> I suggest using "ATS Invalidate Request" in the subject as well.
> Otherwise we have to figure out whether "device-TLB invalidate
> request" is the same as "ATS Invalidate Request".
>
> If they are the same, just use the same words.
>
> On Tue, Dec 26, 2023 at 09:59:22PM -0500, Ethan Zhao wrote:
>> Except those aggressive hotplug cases - surprise remove a hotplug device
>> while its safe removal is requested and handled in process by:
>>
>> 1. pull it out directly.
>> 2. turn off its power.
>> 3. bring the link down.
>> 4. just died there that moment.
>>
>> etc, in a word, 'gone' or 'disconnected'.
>>
>> Mostly are regular normal safe removal and surprise removal unplug.
>> these hot unplug handling process could be optimized for fix the ATS
>> invalidation hang issue by calling pci_dev_is_disconnected() in function
>> devtlb_invalidation_with_pasid() to check target device state to avoid
>> sending meaningless ATS invalidation request to iommu when device is gone.
>> (see IMPLEMENTATION NOTE in PCIe spec r6.1 section 10.3.1)
> Suggest "ATS Invalidate Request", capitalized exactly that way so we
> know it's a specific name of something defined in the PCIe spec.
>
>> For safe removal, device wouldn't be removed untill the whole software
>> handling process is done, it wouldn't trigger the hard lock up issue
>> caused by too long ATS invalidation timeout wait. in safe removal path,
> Ditto.
>
> Capitalize "In the safe removal ..." since it starts a new sentence.
>
>> device state isn't set to pci_channel_io_perm_failure in
>> pciehp_unconfigure_device() by checking 'presence' parameter, calling
>> pci_dev_is_disconnected() in devtlb_invalidation_with_pasid() will return
>> false there, wouldn't break the function.
>>
>> For surprise removal, device state is set to pci_channel_io_perm_failure in
>> pciehp_unconfigure_device(), means device is already gone (disconnected)
>> call pci_dev_is_disconnected() in devtlb_invalidation_with_pasid() will
>> return true to break the function not to send ATS invalidation request to
> Ditto.

Okay.


Thanks,

Ethan

>> the disconnected device blindly, thus avoid the further long time waiting
>> triggers the hard lockup.
>>
>> safe removal & surprise removal
>>
>> pciehp_ist()
>>     pciehp_handle_presence_or_link_change()
>>       pciehp_disable_slot()
>>         remove_board()
>>           pciehp_unconfigure_device(presence)