Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20221122085724.3245078-1-milkfafa@gmail.com> <20221122085724.3245078-8-milkfafa@gmail.com>
 <78cded0f-c75d-819f-0aff-7e4f3b264f64@collabora.com>
In-Reply-To: <78cded0f-c75d-819f-0aff-7e4f3b264f64@collabora.com>
From:   Kun-Fa Lin <milkfafa@gmail.com>
Date:   Thu, 24 Nov 2022 11:30:31 +0800
Message-ID: <CADnNmFqKt_gVO8zonD3DYvM4PazKjKOgE2j_X8jZjSn9Pc9HzA@mail.gmail.com>
Subject: Re: [PATCH v7 7/7] media: nuvoton: Add driver for NPCM video capture
 and encode engine
To:     Andrzej Pietrasiewicz <andrzej.p@collabora.com>
Cc:     mchehab@kernel.org, linux-media@vger.kernel.org,
        linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
        openbmc@lists.ozlabs.org, avifishman70@gmail.com,
        tmaimon77@gmail.com, tali.perry1@gmail.com, kwliu@nuvoton.com,
        kflin@nuvoton.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Andrzej,

Thanks for the review.

> > +#define GPLLST                               0x48
> > +#define  GPLLST_PLLOTDIV1            GENMASK(2, 0)
> > +#define  GPLLST_PLLOTDIV2            GENMASK(5, 3)
> > +#define  GPLLST_GPLLFBDV109          GENMASK(7, 6)
> > +
>
> There's a bunch of register definitions. Given you're adding a dedicated
> directory for nuvoton maybe it makes sense to factor these definitions
> out to a local header file?

Agreed. I'll move these definitions out to a local header file in the
next patch.

> > +     for (i = 0; i < video->num_buffers; i++) {
> > +             head = &video->list[i];
> > +             list_for_each_safe(pos, nx, head) {
> > +                     tmp = list_entry(pos, struct rect_list, list);
>
> If we ever get here isn't pos guaranteed to be non-NULL?
> And so consequently is tmp.
>
> > +                     if (tmp) {
>
> Then this condition is always true?

Indeed the condition is always true, will remove the condition check.

> > +     video->rect = kcalloc(*num_buffers, sizeof(*video->rect), GFP_KERNEL);
>
> In practice "small allocations never fail", but what if kcalloc fails some day?
>
> > +
> > +     if (video->list) {
> > +             npcm_video_free_diff_table(video);
> > +             kfree(video->list);
> > +             video->list = NULL;
> > +     }
> > +
> > +     video->list = kzalloc(sizeof(*video->list) * *num_buffers, GFP_KERNEL);
>
> Or kzalloc?

Will add error handling.

> In this function there are 3 similar error recovery paths. Can nice "goto"s
> be introduced to handle them?

Will do it for sure.

Regards,
Marvin