[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 12 Dec 2016 00:39:39 -0700
From: Daniele Nicolodi <daniele@...nta.net>
To: SF Markus Elfring <elfring@...rs.sourceforge.net>
Cc: linux-media@...r.kernel.org,
Alexey Khoroshilov <khoroshilov@...ras.ru>,
Hans Verkuil <hans.verkuil@...co.com>,
Mauro Carvalho Chehab <mchehab@...nel.org>,
LKML <linux-kernel@...r.kernel.org>,
kernel-janitors@...r.kernel.org
Subject: Re: [media] bt8xx: One function call less in bttv_input_init() after
error detection
On 12/12/16 00:33, SF Markus Elfring wrote:
>>> I would prefer a safer coding style for the corresponding
>>> exception handling.
>>
>> Can you please point out what is wrong in the current code
>
> Is it useful to reconsider the software situation that another memory
> allocation is attempted when it could be determined that a previous one
> failed already?
No.
> Are two successful allocations finally needed to achieve the desired task?
Yes.
>> and how the changes you propose fix the problem?
>
> I suggest to check return values immediately after each function call.
> An error situation can be detected earlier then and only the required
> clean-up functionality will be executed at the end.
Which improvement does this bring?
>> No one has expressed acceptance for the kind of change you propose with
>> this patch, or to previous patches you proposed changing similar constructs.
>
> I got a mixed impression from the acceptance statistics about my
> published patches.
Have you proposed a similar patch that was accepted? I don't find record
of it, but I may be wrong.
>> The fact that you propose over and over again a class of changes that
>> has been already vocally rejected would suggest otherwise.
>
> I dare to propose another look at results from source code search patterns.
Why?
Cheers,
Daniele
Powered by blists - more mailing lists