linux-kernel - Re: [PATCH 1/1] mtd: spi-nor: everspin: add em004lxb entry

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]

Date: Thu, 01 Feb 2024 14:47:34 +0100
From: Michael Walle <mwalle@...nel.org>
To: Flavio Suligoi <f.suligoi@...m.it>
Cc: Tudor Ambarus <tudor.ambarus@...aro.org>, Pratyush Yadav
 <pratyush@...nel.org>, Miquel Raynal <miquel.raynal@...tlin.com>, Richard
 Weinberger <richard@....at>, Vignesh Raghavendra <vigneshr@...com>,
 linux-mtd@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] mtd: spi-nor: everspin: add em004lxb entry

Hi,

> Add the Everspin EM0004LXB 4Mb (512KB) Industrial STT-MRAM Persistent
> Memory.

Out of curiosity, what is your use case here? Usually, I push back on
this small MRAM devices in SPI-NOR in favor of the at25 driver. But
this datasheet lists octal dtr with 200mhz, which seems a bit ridiculous
for 512kB. The at25 driver only supports single bit SPI of course.

I'm not sure in which mode you are using this device, though. The DS
shows a non-volatile configuration register (Table 10, offset 0) and
it's default value is single bit SPI.

> This device is JEDEC compatible (JESD251 and JESD251-1), but it is not
> able to provide SFDP information.
> 
> Link: https://www.everspin.com/file/158244/download

No newline.

> Signed-off-by: Flavio Suligoi <f.suligoi@...m.it>
> ---
>  drivers/mtd/spi-nor/everspin.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/mtd/spi-nor/everspin.c 
> b/drivers/mtd/spi-nor/everspin.c
> index 5f321e24ae7d..4741930ce9a8 100644
> --- a/drivers/mtd/spi-nor/everspin.c
> +++ b/drivers/mtd/spi-nor/everspin.c
> @@ -31,6 +31,14 @@ static const struct flash_info everspin_nor_parts[] 
> = {
>  		.size = SZ_512K,
>  		.sector_size = SZ_512K,
>  		.flags = SPI_NOR_NO_ERASE | SPI_NOR_NO_FR,
> +	}, {
> +		.id = SNOR_ID(0x6b, 0xbb, 0x13),
> +		.name = "em004lxb",

No name. We prefer to only have the ID for parts which has JEDED IDs.

> +		.size = SZ_512K,
> +		.sector_size = SZ_512K,

This should probably be removed (and then default to the 64k erase
size).

> +		.flags = SPI_NOR_NO_ERASE | SPI_NOR_NO_FR | SPI_NOR_HAS_LOCK |

Don't use SPI_NOR_NO_ERASE for new devices. Eventually, I like to get 
rid of
this flag. This device is emulating the erase instruction, so it should 
work
without.

SPI_NOR_NO_FR is wrong here. The DS says it supports fast read.

Please also have a look at [1] for the required tests.

-michael

> +			 SPI_NOR_HAS_TB | SPI_NOR_4BIT_BP | SPI_NOR_BP3_SR_BIT6,
> +		.no_sfdp_flags = SPI_NOR_SKIP_SFDP,
>  	}
>  };

[1] https://docs.kernel.org/driver-api/mtd/spi-nor.html