[coreboot] anyone know what happened here?

Nico Huber nico.h at gmx.de
Fri Dec 22 15:17:38 CET 2017


On 22.12.2017 04:36, Matt DeVillier wrote:
> from what I recall, the driver was trying to be responsible and lock SPI
> write access by default,

Yep, I remember something like that too. But in the usual UEFI case
there is nothing you could lock without breaking UEFI runtime at
least ;) These MTD drivers seem to assume that they are the only
entity accessing the chip. I doubt they saw the bigger picture at
Intel when they upstreamed that driver. Maybe it's only meant for
non-UEFI embedded use cases? Also, the Kconfig says (from the very
beginning):

  Say N here unless you know what you are doing. Overwriting the
  SPI flash may render the system unbootable.

I wonder why distros pick it up.

> Nico -- did this ever get fixed in the upstream kernel?  From what I saw,
> the "fixed" Ubuntu ISO simply omitted the driver in the kernel config/build

Yes, there was a train of fixes that took care of this, too. At least I
hope it was the off-by-one fixed here:

  9d63f17661e2 spi-nor: intel-spi: Fix broken software sequencing codes

Nico



More information about the coreboot mailing list