lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4d2347e4-7732-1352-5009-791362596e0b@collabora.com>
Date:   Mon, 16 Jan 2023 11:27:57 +0100
From:   Andrzej Pietrasiewicz <andrzej.p@...labora.com>
To:     Sebastian Fricke <sebastian.fricke@...labora.com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Ezequiel Garcia <ezequiel@...guardiasur.com.ar>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-rockchip@...ts.infradead.org, linux-staging@...ts.linux.dev,
        Jonas Karlman <jonas@...boo.se>,
        Alex Bee <knaerzche@...il.com>,
        Nicolas Dufresne <nicolas.dufresne@...labora.com>,
        Collabora Kernel-domain <kernel@...labora.com>,
        Robert Beckett <bob.beckett@...labora.com>,
        Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        Benjamin Gaignard <benjamin.gaignard@...labora.com>
Subject: Re: [PATCH v2 07/12] staging: media: rkvdec: Add a routine to fetch
 SPS attributes as a callback

Hi Sebastian,

W dniu 12.01.2023 o 13:56, Sebastian Fricke pisze:
> Add a callback for each codec variant, that fetches basic information
> like resolution, bit-depth and sub-sampling from a SPS structure. This
> data is used to verify whether a new SPS structure is valid.
> 
> Signed-off-by: Sebastian Fricke <sebastian.fricke@...labora.com>
> ---
>   drivers/staging/media/rkvdec/rkvdec.c | 10 ++++++++++
>   drivers/staging/media/rkvdec/rkvdec.h | 10 ++++++++++
>   2 files changed, 20 insertions(+)
> 
> diff --git a/drivers/staging/media/rkvdec/rkvdec.c b/drivers/staging/media/rkvdec/rkvdec.c
> index a46f918926a2..e8c750a7343a 100644
> --- a/drivers/staging/media/rkvdec/rkvdec.c
> +++ b/drivers/staging/media/rkvdec/rkvdec.c
> @@ -52,6 +52,16 @@ static int rkvdec_get_valid_fmt(struct rkvdec_ctx *ctx)
>   	return ctx->coded_fmt_desc->decoded_fmts[0];
>   }
>   
> +static int rkvdec_get_sps_attributes(struct rkvdec_ctx *ctx, void *sps,
> +				     struct sps_attributes *attributes)
> +{
> +	const struct rkvdec_coded_fmt_desc *coded_desc = ctx->coded_fmt_desc;
> +
> +	if (coded_desc->ops->get_sps_attributes)
> +		return coded_desc->ops->get_sps_attributes(ctx, sps, attributes);
> +	return 0;

It seems that if there's no ->get_sps_attributes(), then even though
"attributes" won't be filled in, the result is a success (as suggested
by the returned 0). That's maybe confusing to potential users, especially,
if they don't e.g. memset the attributes struct to zero - the function will
report "success" but the struct will contain random data.

In PATCH 09/12 you call the implementation directly (not through the ops
struct), and there you actually ignore the return value.

Regards,

Andrzej

> +}
> +
>   static int rkvdec_try_ctrl(struct v4l2_ctrl *ctrl)
>   {
>   	struct rkvdec_ctx *ctx = container_of(ctrl->handler, struct rkvdec_ctx, ctrl_hdl);
> diff --git a/drivers/staging/media/rkvdec/rkvdec.h b/drivers/staging/media/rkvdec/rkvdec.h
> index e353a4403e5b..7b6702c360fd 100644
> --- a/drivers/staging/media/rkvdec/rkvdec.h
> +++ b/drivers/staging/media/rkvdec/rkvdec.h
> @@ -63,10 +63,20 @@ vb2_to_rkvdec_decoded_buf(struct vb2_buffer *buf)
>   			    base.vb.vb2_buf);
>   }
>   
> +struct sps_attributes {
> +	unsigned int width;
> +	unsigned int height;
> +	unsigned int luma_bitdepth;
> +	unsigned int chroma_bitdepth;
> +	unsigned int subsampling;
> +};
> +
>   struct rkvdec_coded_fmt_ops {
>   	int (*adjust_fmt)(struct rkvdec_ctx *ctx,
>   			  struct v4l2_format *f);
>   	u32 (*valid_fmt)(struct rkvdec_ctx *ctx);
> +	int (*get_sps_attributes)(struct rkvdec_ctx *ctx, void *sps,
> +				  struct sps_attributes *attributes);
>   	int (*start)(struct rkvdec_ctx *ctx);
>   	void (*stop)(struct rkvdec_ctx *ctx);
>   	int (*run)(struct rkvdec_ctx *ctx);
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ