linux-kernel - Re: [PATCH 02/11] mmc: core: Take into account MMC_CAP_NEED_RSP

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: <6d528e02-dd27-65d3-0cd3-e764cc780a81@rock-chips.com>
Date:   Fri, 7 May 2021 09:44:30 +0800
From:   Shawn Lin <shawn.lin@...k-chips.com>
To:     Ulf Hansson <ulf.hansson@...aro.org>, linux-mmc@...r.kernel.org,
        Adrian Hunter <adrian.hunter@...el.com>
Cc:     shawn.lin@...k-chips.com, Linus Walleij <linus.walleij@...aro.org>,
        Wolfram Sang <wsa+renesas@...g-engineering.com>,
        Avri Altman <avri.altman@....com>,
        Masami Hiramatsu <masami.hiramatsu@...aro.org>,
        linux-block@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 02/11] mmc: core: Take into account MMC_CAP_NEED_RSP_BUSY
 for eMMC HPI commands

On 2021/5/5 0:12, Ulf Hansson wrote:
> In mmc_send_hpi_cmd() the host->max_busy_timeout is being validated towards
> the timeout for the eMMC HPI command, as to decide whether an R1 or R1B
> response should be used.
> 
> Although, it has turned out the some host can't cope with that conversion,
> but needs R1B, which means MMC_CAP_NEED_RSP_BUSY is set for them. Let's
> take this into account, via using the common mmc_prepare_busy_cmd() when
> doing the validation, which also avoids some open coding.

Reviewed-by: Shawn Lin <shawn.lin@...k-chips.com>

> 
> Signed-off-by: Ulf Hansson <ulf.hansson@...aro.org>
> ---
>   drivers/mmc/core/mmc_ops.c | 21 +++++----------------
>   1 file changed, 5 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
> index 025a4134d5c7..66ae699a410f 100644
> --- a/drivers/mmc/core/mmc_ops.c
> +++ b/drivers/mmc/core/mmc_ops.c
> @@ -817,28 +817,17 @@ static int mmc_send_hpi_cmd(struct mmc_card *card)
>   {
>   	unsigned int busy_timeout_ms = card->ext_csd.out_of_int_time;
>   	struct mmc_host *host = card->host;
> -	bool use_r1b_resp = true;
> +	bool use_r1b_resp = false;
>   	struct mmc_command cmd = {};
>   	int err;
>   
>   	cmd.opcode = card->ext_csd.hpi_cmd;
>   	cmd.arg = card->rca << 16 | 1;
> +	cmd.flags = MMC_RSP_R1 | MMC_CMD_AC;
>   
> -	/*
> -	 * Make sure the host's max_busy_timeout fit the needed timeout for HPI.
> -	 * In case it doesn't, let's instruct the host to avoid HW busy
> -	 * detection, by using a R1 response instead of R1B.
> -	 */
> -	if (host->max_busy_timeout && busy_timeout_ms > host->max_busy_timeout)
> -		use_r1b_resp = false;
> -
> -	if (cmd.opcode == MMC_STOP_TRANSMISSION && use_r1b_resp) {
> -		cmd.flags = MMC_RSP_R1B | MMC_CMD_AC;
> -		cmd.busy_timeout = busy_timeout_ms;
> -	} else {
> -		cmd.flags = MMC_RSP_R1 | MMC_CMD_AC;
> -		use_r1b_resp = false;
> -	}
> +	if (cmd.opcode == MMC_STOP_TRANSMISSION)
> +		use_r1b_resp = mmc_prepare_busy_cmd(host, &cmd,
> +						    busy_timeout_ms);
>   
>   	err = mmc_wait_for_cmd(host, &cmd, 0);
>   	if (err) {
>