Date:   Fri, 10 Mar 2023 11:20:54 +0900
From:   Sergey Senozhatsky <senozhatsky@chromium.org>
To:     Ricardo Ribalda <ribalda@chromium.org>
Cc:     Mauro Carvalho Chehab <mchehab@kernel.org>,
        Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
        linux-media@vger.kernel.org,
        Sergey Senozhatsky <senozhatsky@chromium.org>,
        linux-kernel@vger.kernel.org, "hn.chen" <hn.chen@sunplusit.com>
Subject: Re: [PATCH v6 3/5] media: uvcvideo: Allow hw clock updates with
 buffers not full
Message-ID: <20230310022054.GG1045797@google.com>
References: <20220920-resend-hwtimestamp-v6-0-c7a99299ec35@chromium.org>
 <20220920-resend-hwtimestamp-v6-3-c7a99299ec35@chromium.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220920-resend-hwtimestamp-v6-3-c7a99299ec35@chromium.org>
Precedence: bulk

On (23/03/09 13:55), Ricardo Ribalda wrote:
[..]
> +++ b/drivers/media/usb/uvc/uvc_video.c
> @@ -754,10 +754,10 @@ void uvc_video_clock_update(struct uvc_streaming *stream,
>  
>  	spin_lock_irqsave(&clock->lock, flags);
>  
> -	if (clock->count < clock->size)
> +	if (clock->count < 2)
>  		goto done;
>  
> -	first = &clock->samples[clock->head];
> +	first = &clock->samples[(clock->head - clock->count) % clock->size];

This looks like a significant change: `rb->head` and `rb->head - INT`
are quite different things. May I ask for some clarifications?

[..]
> +	/*
> +	 * Have at least 1/4 of a second of timestamps before we
> +	 * try to do any calculation. Otherwise we do not have enough
> +	 * precision. This value was determined by running Android CTS
> +	 * on different devices.
> +	 */
> +	if ((y2 - y1) < (256 << 16))

256 << 16 is 1/4 of a second?