RE: [PATCH v10 2/7] i3c: master: use i3c_master_register only for main master

From: Parshuram Raju Thombare <pthombar@cadence.com>
To: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Nicolas Pitre <nico@fluxnic.net>,
	"slongerbeam@gmail.com" <slongerbeam@gmail.com>,
	"vitor.soares@synopsys.com" <vitor.soares@synopsys.com>,
	"praneeth@ti.com" <praneeth@ti.com>,
	Milind Parab <mparab@cadence.com>,
	"linux-i3c@lists.infradead.org" <linux-i3c@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH v10 2/7] i3c: master: use i3c_master_register only for main master
Date: Thu, 17 Jun 2021 08:42:28 +0000	[thread overview]
Message-ID: <CY4PR07MB27570948E3536C160662AEEAC10E9@CY4PR07MB2757.namprd07.prod.outlook.com> (raw)
In-Reply-To: <YL/7XP1fs/2bRiTM@piout.net>

Hi Alexandre,

Thank you for review comments.

>On 08/12/2020 05:47:51+0000, Parshuram Raju Thombare wrote:
>> >This looks a bit confusing. Here you're rolling back detailss in
>> >i3c_primary_master_register() that were factored out in
>> >i3c_master_init(). If i3c_master_init() is successful, then you
>> >shouldn't be undoing its things openly in i3c_primary_master_register().
>> >Instead, there should be another function that does the reverse of
>> >i3c_master_init() here.
>>
>> Every function do its cleanup in case of failures.
>> And if any failure occur after successful i3c_master_init(), we have
>> function i3c_master_bus_cleanup() which does the major cleanup.
>>
>
>The point from Nicolas here was that the workqueue is allocated in
>i3c_master_init so you should have a function to destroy it instead of
>having to do that separately in i3c_primary_master_register. The same is
>true for the put_device. Or you have to ensure i3c_masterdev_release
>is called when i3c_primary_master_register fails.

Ok, IIUC, it is just that we need separate clean up function. 
i3c_master_bus_cleanup is used here to call driver cleanup function and
detach + release devices from bus, and I am not sure i3c_masterdev_release
can be used for that.
Better alternative is i3c_master_unregister(). However, it doesn't handle
the case where device_add fails, as this function unconditionally calls
device_unregister which call both device_del and put_device.