[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-id: <b26d966f-5acd-5d1e-0cb6-0232f84c2b31@samsung.com>
Date: Fri, 07 Apr 2017 12:26:31 +0200
From: Sylwester Nawrocki <s.nawrocki@...sung.com>
To: Smitha T Murthy <smitha.t@...sung.com>
Cc: linux-arm-kernel@...ts.infradead.org, linux-media@...r.kernel.org,
linux-kernel@...r.kernel.org, kyungmin.park@...sung.com,
kamil@...as.org, jtp.park@...sung.com, a.hajda@...sung.com,
mchehab@...nel.org, pankaj.dubey@...sung.com, krzk@...nel.org,
m.szyprowski@...sung.com
Subject: Re: [Patch v4 12/12] Documention: v4l: Documentation for HEVC CIDs
On 04/07/2017 12:03 PM, Smitha T Murthy wrote:
>>> +``V4L2_CID_MPEG_VIDEO_HEVC_LF``
>>> + Indicates loop filtering. Control ID 0 indicates loop filtering
>>> + is enabled and when set to 1 indicates no filter.
>>
>> "Setting this control to 0 enables loop filtering, setting this control
>> to 1 disables loop filtering." ?
>>
>> Couldn't the meaning be inverted, so setting the control to 0 disables
>> the loop filtering?
>>
>>>From register point of view, this control value needs be 0 to enable
> loop filtering.
OK, this is true for our specific hardware/firmware implementation.
In general, for this user space interface I would rather define that
boolean control so value 1 enables LF and value 0 disables LF.
The driver could simply negate the value when writing registers.
BTW we might need to specify type of the control here as Hans suggested
commenting on other control.
--
Regards,
Sylwester
Powered by blists - more mailing lists