linux-kernel - Re: [PATCH 2/2] mtd: spi-nand: Add read retry support

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: <20241001121724.5edbb64b@xps-13>
Date: Tue, 1 Oct 2024 12:17:24 +0200
From: Miquel Raynal <miquel.raynal@...tlin.com>
To: Cheng Ming Lin <linchengming884@...il.com>
Cc: vigneshr@...com, linux-mtd@...ts.infradead.org,
 linux-kernel@...r.kernel.org, richard@....at, alvinzhou@...c.com.tw,
 leoyu@...c.com.tw, Cheng Ming Lin <chengminglin@...c.com.tw>
Subject: Re: [PATCH 2/2] mtd: spi-nand: Add read retry support

Hi Cheng Ming,

linchengming884@...il.com wrote on Thu,  5 Sep 2024 13:53:33 +0800:

> From: Cheng Ming Lin <chengminglin@...c.com.tw>
> 
> When the host ECC fails to correct the data error of NAND device,
> there's a special read for data recovery method which host setups
> for the next read retry mode and may recover the lost data by host
> ECC again.
> 
> Signed-off-by: Cheng Ming Lin <chengminglin@...c.com.tw>
> ---
>  drivers/mtd/nand/spi/core.c | 33 +++++++++++++++++++++++++++++++--
>  1 file changed, 31 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
> index e0b6715e5dfe..2f21ea926132 100644
> --- a/drivers/mtd/nand/spi/core.c
> +++ b/drivers/mtd/nand/spi/core.c
> @@ -640,6 +640,7 @@ static int spinand_mtd_read(struct mtd_info *mtd, loff_t from,
>  	struct nand_io_iter iter;
>  	bool disable_ecc = false;
>  	bool ecc_failed = false;
> +	u8 retry_mode = 0;
>  	int ret = 0;
>  
>  	if (ops->mode == MTD_OPS_RAW || !spinand->eccinfo.ooblayout)
> @@ -657,20 +658,45 @@ static int spinand_mtd_read(struct mtd_info *mtd, loff_t from,

This will no longer apply with continuous support added in. However
please make this only in the non continuous path. I don't think it can
fly in the other.

>  		if (ret)
>  			break;
>  
> +read_retry:
>  		ret = spinand_read_page(spinand, &iter.req);
>  		if (ret < 0 && ret != -EBADMSG)
>  			break;
>  
> -		if (ret == -EBADMSG)
> +		if (ret == -EBADMSG && spinand->info->fixups) {
> +			if (spinand->read_retries && ((retry_mode + 1) < spinand->read_retries)) {

							++retry_mode?
> +				retry_mode++;

So this can be dropped.

> +				ret = spinand->info->fixups->setup_read_retry(spinand, retry_mode);
> +				if (ret < 0)
> +					break;

No, you need to set ecc_failed here.

> +
> +				/* Reset ecc_stats; retry */
> +				mtd->ecc_stats = old_stats;
> +				goto read_retry;
> +			} else {
> +				/* No more retry modes; real failure */
> +				ecc_failed = true;
> +			}
> +		} else if (ret == -EBADMSG) {
>  			ecc_failed = true;
> -		else
> +		} else {
>  			max_bitflips = max_t(unsigned int, max_bitflips, ret);
> +		}
>  
>  		ret = 0;
>  		ops->retlen += iter.req.datalen;
>  		ops->oobretlen += iter.req.ooblen;
> +
> +		/* Reset to retry mode 0*/
> +		if (retry_mode) {

			retry_mode = 0;

> +			ret = spinand->info->fixups->setup_read_retry(spinand, 0);

									retry_mode);

> +			if (ret < 0)
> +				break;

this if clause is useless.

> +			retry_mode = 0;

And then this can be dropped from here.

> +		}
>  	}
>  
> +

Spurious line

>  	if (ops->stats) {
>  		ops->stats->uncorrectable_errors +=
>  			mtd->ecc_stats.failed - old_stats.failed;
> @@ -1095,6 +1121,9 @@ int spinand_match_and_init(struct spinand_device *spinand,
>  		spinand->flags = table[i].flags;
>  		spinand->id.len = 1 + table[i].devid.len;
>  		spinand->select_target = table[i].select_target;
> +		spinand->info = info;
> +		if (spinand->info->fixups && spinand->info->fixups->init_read_retry)
> +			spinand->read_retries = spinand->info->fixups->init_read_retry(spinand);

Now I get you init. Ok, fine.

>  
>  		op = spinand_select_op_variant(spinand,
>  					       info->op_variants.read_cache);


Thanks,
Miquèl