linux-kernel - Re: [PATCH 2/2] mmc: core: Add no single read retries

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 17 Feb 2021 07:46:54 +0200
From:   Adrian Hunter <adrian.hunter@...el.com>
To:     DooHyun Hwang <dh0421.hwang@...sung.com>,
        linux-mmc@...r.kernel.org, linux-kernel@...r.kernel.org,
        devicetree@...r.kernel.org, ulf.hansson@...aro.org,
        robh+dt@...nel.org, axboe@...nel.dk, satyat@...gle.com,
        ebiggers@...gle.com, gustavoars@...nel.org
Cc:     grant.jung@...sung.com, jt77.jang@...sung.com,
        junwoo80.lee@...sung.com, jangsub.yi@...sung.com,
        sh043.lee@...sung.com, cw9316.lee@...sung.com,
        sh8267.baek@...sung.com, wkon.kim@...sung.com
Subject: Re: [PATCH 2/2] mmc: core: Add no single read retries

On 17/02/21 7:22 am, DooHyun Hwang wrote:
> This makes to handle read errors faster by not retrying
> multiple block read(CMD18) errors with single block reads(CMD17).
> 
> On some bad SD Cards that have problem with read operations,
> it is not helpful to retry multiple block read errors with
> several single block reads, and it is delayed to treat read
> operations as I/O error as much as retrying single block reads.

If the issue is that it takes too long, then maybe it would be better to get
mmc_blk_read_single() to give up after a certain amount of time.

> 
> Signed-off-by: DooHyun Hwang <dh0421.hwang@...sung.com>
> ---
>  drivers/mmc/core/block.c | 3 ++-
>  drivers/mmc/core/host.c  | 6 ++++++
>  include/linux/mmc/host.h | 3 +++
>  3 files changed, 11 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
> index d666e24fbe0e..e25aaf8fca34 100644
> --- a/drivers/mmc/core/block.c
> +++ b/drivers/mmc/core/block.c
> @@ -1809,7 +1809,8 @@ static void mmc_blk_mq_rw_recovery(struct mmc_queue *mq, struct request *req)
>  
>  	/* FIXME: Missing single sector read for large sector size */
>  	if (!mmc_large_sector(card) && rq_data_dir(req) == READ &&
> -	    brq->data.blocks > 1) {
> +	    brq->data.blocks > 1 &&
> +	    !card->host->no_single_read_retry) {
>  		/* Read one sector at a time */
>  		mmc_blk_read_single(mq, req);
>  		return;
> diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
> index 9b89a91b6b47..3bf5b2fc111b 100644
> --- a/drivers/mmc/core/host.c
> +++ b/drivers/mmc/core/host.c
> @@ -352,6 +352,12 @@ int mmc_of_parse(struct mmc_host *host)
>  	if (device_property_read_bool(dev, "no-mmc"))
>  		host->caps2 |= MMC_CAP2_NO_MMC;
>  
> +	if (device_property_read_bool(dev, "no-single-read-retry")) {
> +		dev_info(host->parent,
> +			"Single block read retrying is not supported\n");
> +		host->no_single_read_retry = true;
> +	}
> +
>  	/* Must be after "non-removable" check */
>  	if (device_property_read_u32(dev, "fixed-emmc-driver-type", &drv_type) == 0) {
>  		if (host->caps & MMC_CAP_NONREMOVABLE)
> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
> index 26a3c7bc29ae..faec55035a63 100644
> --- a/include/linux/mmc/host.h
> +++ b/include/linux/mmc/host.h
> @@ -502,6 +502,9 @@ struct mmc_host {
>  	/* Host Software Queue support */
>  	bool			hsq_enabled;
>  
> +	/* Do not retry multi block read as single block read */
> +	bool			no_single_read_retry;
> +
>  	unsigned long		private[] ____cacheline_aligned;
>  };
>  
>