mtd: spi-nor: parsers questions

mtd: spi-nor: parsers questions

[Mauro Lima]

Hey all!
Looking for some help with mtd.
I'm working on some changes for Intel's spi controller. We have a
"BIOS" partition, composed of many parts with different offsets and
could be spread within two chips in some cases (not necessarily of the
same size).
Right now the driver uses one partition that, in my system, is bigger
than the first chip's size, so I'm getting the "size truncated to
0x..." log.
All drivers using mtd_partitions know the partition information at
compile time. This is not my case, we need to read the first portion
of the chip to get the partition parts' offsets. So I found parsers,
and the drivers using parsers make partitions within the chip's size.
So my question is whether parsers could solve my truncation issue and
whether parsers support this kind of multiple chip case.
Thanks!

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: mtd: spi-nor: parsers questions

[Miquel Raynal]

Hi Mauro,

mauro.lima@eclypsium.com wrote on Wed, 14 Jun 2023 18:19:45 -0300:

> Hey all!
> Looking for some help with mtd.
> I'm working on some changes for Intel's spi controller. We have a
> "BIOS" partition, composed of many parts with different offsets and
> could be spread within two chips in some cases (not necessarily of the
> same size).
> Right now the driver uses one partition that, in my system, is bigger
> than the first chip's size, so I'm getting the "size truncated to
> 0x..." log.
> All drivers using mtd_partitions know the partition information at
> compile time. This is not my case, we need to read the first portion
> of the chip to get the partition parts' offsets. So I found parsers,
> and the drivers using parsers make partitions within the chip's size.
> So my question is whether parsers could solve my truncation issue and
> whether parsers support this kind of multiple chip case.
> Thanks!

There is mtdconcat in the kernel but I don't think it will fit your
purpose. FYI I tried a few years ago to bring support for something
like this, unsuccessfully.

Link: https://lore.kernel.org/linux-devicetree/20191127105522.31445-1-miquel.raynal@bootlin.com/

You would need something like that and on top of it a dynamic partition
parser.

Cheers,
Miquèl

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/