Received-SPF: pass (google.com: domain of linux-kernel+bounces-196131-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: <1333039e-ec61-467e-a0ee-d3cb86e769f5@linux.intel.com>
Date: Fri, 31 May 2024 09:19:37 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Cc: baolu.lu@linux.intel.com, iommu@lists.linux.dev,
 linux-kernel@vger.kernel.org
Subject: Re: device_def_domain_type documentation header does not match
 implementation
To: Robin Murphy <robin.murphy@arm.com>,
 Diederik de Haas <didi.debian@cknow.org>,
 David Woodhouse <dwmw2@infradead.org>, Joerg Roedel <joro@8bytes.org>,
 Will Deacon <will@kernel.org>
References: <14311965.TaHA55BQu8@bagend>
 <32921840-43d6-4ad9-99eb-aac32e67e04c@arm.com>
Content-Language: en-US
From: Baolu Lu <baolu.lu@linux.intel.com>
In-Reply-To: <32921840-43d6-4ad9-99eb-aac32e67e04c@arm.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

On 5/30/24 11:49 PM, Robin Murphy wrote:
> On 30/05/2024 2:57 pm, Diederik de Haas wrote:
>> Hi,
>>
>> While looking into 
>> ``drivers/iommu/intel/iommu.c::device_def_domain_type``
>> function I noticed a discrepancy between the documentation header and the
>> implementation.
>>
>> ``@startup: true if this is during early boot``
>> 0e31a7266508 ("iommu/vt-d: Remove startup parameter from
>> device_def_domain_type()")
>> removed the ``startup`` function parameter
>>
>> returns ``IOMMU_DOMAIN_DMA: device requires a dynamic mapping domain``
>> 28b41e2c6aeb ("iommu: Move def_domain type check for untrusted device 
>> into
>> core")
>> moved the possible return of ``IOMMU_DOMAIN_DMA`` to 
>> ``drivers/iommu/iommu.c``
>>
>> But neither updated the documentation header.
> 
> TBH it could probably just be deleted now, since the 
> iommu_ops::def_domain_type callback is properly documented in iommu.h, 
> so individual implementations shouldn't need to repeat that. It's also 
> never been actual kerneldoc either, since it's a regular "/*" comment. 
> Feel free to send a patch ????

Agreed. I will make a patch to remove it later.

Best regards,
baolu