linux-kernel - Re: [PATCH] ASoC: fsl_asrc: Add an option to select internal ratio mode

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: <20200629200813.GB27967@Asurada-Nvidia>
Date:   Mon, 29 Jun 2020 13:08:13 -0700
From:   Nicolin Chen <nicoleotsuka@...il.com>
To:     Shengjiu Wang <shengjiu.wang@....com>
Cc:     timur@...nel.org, Xiubo.Lee@...il.com, festevam@...il.com,
        broonie@...nel.org, alsa-devel@...a-project.org,
        lgirdwood@...il.com, perex@...ex.cz, tiwai@...e.com,
        linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ASoC: fsl_asrc: Add an option to select internal ratio
 mode

On Mon, Jun 29, 2020 at 09:58:35PM +0800, Shengjiu Wang wrote:
> The ASRC not only supports ideal ratio mode, but also supports
> internal ratio mode.
> 
> For internal rato mode, the rate of clock source should be divided
> with no remainder by sample rate, otherwise there is sound
> distortion.
> 
> Add function fsl_asrc_select_clk() to find proper clock source for
> internal ratio mode, if the clock source is available then internal
> ratio mode will be selected.
> 
> With change, the ideal ratio mode is not the only option for user.
> 
> Signed-off-by: Shengjiu Wang <shengjiu.wang@....com>
> ---

> +static int fsl_asrc_select_clk(struct fsl_asrc_priv *asrc_priv,
> +			       struct fsl_asrc_pair *pair,
> +			       int in_rate,
> +			       int out_rate)
> +{
> +	struct fsl_asrc_pair_priv *pair_priv = pair->private;
> +	struct asrc_config *config = pair_priv->config;
> +	int rate[2], select_clk[2]; /* Array size 2 means IN and OUT */
> +	int clk_rate, clk_index;
> +	int i = 0, j = 0;
> +	bool clk_sel[2];
> +
> +	rate[0] = in_rate;
> +	rate[1] = out_rate;
> +
> +	/* Select proper clock source for internal ratio mode */
> +	for (j = 0; j < 2; j++) {
> +		for (i = 0; i < ASRC_CLK_MAP_LEN; i++) {
> +			clk_index = asrc_priv->clk_map[j][i];
> +			clk_rate = clk_get_rate(asrc_priv->asrck_clk[clk_index]);

+			/* Only match a perfect clock source with no remainder */

> +			if (clk_rate != 0 && (clk_rate / rate[j]) <= 1024 &&
> +			    (clk_rate % rate[j]) == 0)
> +				break;
> +		}
> +
> +		if (i == ASRC_CLK_MAP_LEN) {
> +			select_clk[j] = OUTCLK_ASRCK1_CLK;
> +			clk_sel[j] = false;
> +		} else {
> +			select_clk[j] = i;
> +			clk_sel[j] = true;
> +		}
> +	}
> +
> +	/* Switch to ideal ratio mode if there is no proper clock source */
> +	if (!clk_sel[IN] || !clk_sel[OUT])
> +		select_clk[IN] = INCLK_NONE;

Could get rid of clk_set:

	for (j) {
		for (i) {
			if (match)
				break;
		}

		clk[j] = i;
	}

	if (clk[IN] == ASRC_CLK_MAP_LEN || clk[OUT] == ASRC_CLK_MAP_LEN)

And it only overrides clk[IN] setting but leaving clk[OUT] to
to the searching result. This means that clk[OUT] may be using
a clock source other than OUTCLK_ASRCK1_CLK if sel[IN] happens
to be false while sel[OUT] happens to be true. Not sure if it
is intended...but I feel it would probably be safer to use the
previous settings: INCLK_NONE + OUTCLK_ASRCK1_CLK?