Received-SPF: pass (google.com: domain of linux-kernel+bounces-91362-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <e4c54951-0e7d-4307-9266-2d6b984c967a@redhat.com>
Date: Mon, 4 Mar 2024 23:26:58 +0100
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [RFC PATCH v5 1/1] fpga: add an owner and use it to take the
 low-level module's refcount
Content-Language: en-US
To: Xu Yilun <yilun.xu@linux.intel.com>
Cc: Moritz Fischer <mdf@kernel.org>, Wu Hao <hao.wu@intel.com>,
 Xu Yilun <yilun.xu@intel.com>, Tom Rix <trix@redhat.com>,
 Jonathan Corbet <corbet@lwn.net>,
 Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
 Alan Tull <atull@opensource.altera.com>, linux-kernel@vger.kernel.org,
 linux-doc@vger.kernel.org, linux-fpga@vger.kernel.org
References: <Zbh7iO9wlm9ekzB7@yilunxu-OptiPlex-7050>
 <0720eb91-72f9-4781-8558-8a1b0a3691c2@redhat.com>
 <Zb8dd9af0Ru/fzGi@yilunxu-OptiPlex-7050>
 <4aaa131a-4b64-4b86-9548-68aef63c87b3@redhat.com>
 <ZdHWaeU+/On6LmHX@yilunxu-OptiPlex-7050>
 <9a9d4018-fd65-49be-9e0a-1eecc9cbf15d@redhat.com>
 <ZdYKnZxdTCvu5THG@yilunxu-OptiPlex-7050>
 <ae3cd81b-82af-4977-91d7-fa809c6fc45a@redhat.com>
 <Zd7cff43ffbJOGNY@yilunxu-OptiPlex-7050>
 <08ba8bce-0ebf-4c8f-952d-a6665dc7fdf9@redhat.com>
 <ZeXRWDGvd8P7KhxJ@yilunxu-OptiPlex-7050>
From: Marco Pagani <marpagan@redhat.com>
In-Reply-To: <ZeXRWDGvd8P7KhxJ@yilunxu-OptiPlex-7050>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit


On 2024-03-04 14:49, Xu Yilun wrote:
>> Just to be sure that I understood correctly, you want to split the
>> changes into two patches, like:
>>
>> a) add module owner to the manager struct and take it in
>> __fpga_mgr_get(); move put_device() from __fpga_mgr_get() to
>> fpga_mgr_get() and of_fpga_mgr_get().
>>
>> b) add the mutex and the unregistered flag for protection against races.
>>
>> So that (b) can be reverted if try_module_get_safe() will be accepted?
> 
> Yes, that's what I mean.
> 

On second thought, I would prefer to send only the patches (a) for the
manager, bridge, and region for the moment. As you said, the chances of
having a race are slim (no crash has ever been reported to date). So,
after (a), I think it is worth focusing first on try_module_get_safe()
since it could also be useful for other subsystems. The (b) patches
could always be applied later if try_module_get_safe() is not accepted.
Thanks,
Marco