Opinions on it87spi.c hack to support IT8686E?

Hi,

I have a a GIGABYTE GA-H270N-WIFI motherboard with DualBIOS feature.
Flashrom 1.2 is unable to select between the two flash chips with
dualbiosindex parameter, as the Super I/O chip (ITE IT8686E) is
unsupported:

"Super I/O ID 0x8686 is not on the list of flash-capable controllers."

In a naive attempt, I just tried to add the chip ID to the list:


This modification made flashrom able to select between the flash chips:

# flashrom -p internal:dualbiosindex=
DualBIOS: Selected chip: 0

# flashrom -p internal:dualbiosindex=1
DualBIOS: Selected chip: 1

# flashrom -p internal:dualbiosindex=
DualBIOS: Selected chip: 1

Reading the flash content also seems to work:

# flashrom -p internal:dualbiosindex=0 -r M_BIOS.bin
Enabling flash write... SPI Configuration is locked down.
Enabling hardware sequencing because some important opcode is locked.
OK.
DualBIOS: Selected chip: 0
Found Programmer flash chip "Opaque flash chip" (16384 kB,
Programmer-specific) mapped at physical address 0x0000000000000000.
Reading flash... done.

# flashrom -p internal:dualbiosindex=1 -r B_BIOS.bin
Enabling flash write... SPI Configuration is locked down.
Enabling hardware sequencing because some important opcode is locked.
OK.
DualBIOS: Selected chip: 1
Found Programmer flash chip "Opaque flash chip" (16384 kB,
Programmer-specific) mapped at physical address 0x0000000000000000.
Reading flash... done.

Comparing the two firmware dumps, I find the only difference is that
M_BIOS.bin contains my current BIOS settings which B_BIOS.bin does not,
so I'm quite sure that these dumps are actually from two different
flash chips.

However, if enabling verbose (-VV) output while reading flash, I also
see this message, which I don't know if normal or not:

"No IT87* serial flash segment enabled."

I have not yet tried writing to any of the flash chips, as I would
really appreciate if someone with more experience than me could comment
on this modification first, and maybe highlight problems I'm not
currently aware of?

Thanks in advance.


Best regards,
Pete Smith

Hey Pete,

Send the patch to gerrit, some support seems better than no support imho.

Kind regards,
Edward.

On Sat, 9 Jan 2021 at 08:06, Pete Smith via flashrom
<flashrom@flashrom.org> wrote:
>
> Hi,
>
> I have a a GIGABYTE GA-H270N-WIFI motherboard with DualBIOS feature.
> Flashrom 1.2 is unable to select between the two flash chips with
> dualbiosindex parameter, as the Super I/O chip (ITE IT8686E) is
> unsupported:
>
> "Super I/O ID 0x8686 is not on the list of flash-capable controllers."
>
> In a naive attempt, I just tried to add the chip ID to the list:
>
> --- a/it87spi.c
> +++ b/it87spi.c
> @@ -256,6 +256,7 @@ int init_superio_ite(void)
>                 case 0x8718:
>                 case 0x8720:
>                 case 0x8728:
> +               case 0x8686:
>                         ret |= it87spi_probe(superios[i].port);
>                         break;
>                 default:
>
> This modification made flashrom able to select between the flash chips:
>
> # flashrom -p internal:dualbiosindex=
> DualBIOS: Selected chip: 0
>
> # flashrom -p internal:dualbiosindex=1
> DualBIOS: Selected chip: 1
>
> # flashrom -p internal:dualbiosindex=
> DualBIOS: Selected chip: 1
>
> Reading the flash content also seems to work:
>
> # flashrom -p internal:dualbiosindex=0 -r M_BIOS.bin
> Enabling flash write... SPI Configuration is locked down.
> Enabling hardware sequencing because some important opcode is locked.
> OK.
> DualBIOS: Selected chip: 0
> Found Programmer flash chip "Opaque flash chip" (16384 kB,
> Programmer-specific) mapped at physical address 0x0000000000000000.
> Reading flash... done.
>
> # flashrom -p internal:dualbiosindex=1 -r B_BIOS.bin
> Enabling flash write... SPI Configuration is locked down.
> Enabling hardware sequencing because some important opcode is locked.
> OK.
> DualBIOS: Selected chip: 1
> Found Programmer flash chip "Opaque flash chip" (16384 kB,
> Programmer-specific) mapped at physical address 0x0000000000000000.
> Reading flash... done.
>
> Comparing the two firmware dumps, I find the only difference is that
> M_BIOS.bin contains my current BIOS settings which B_BIOS.bin does not,
> so I'm quite sure that these dumps are actually from two different
> flash chips.
>
> However, if enabling verbose (-VV) output while reading flash, I also
> see this message, which I don't know if normal or not:
>
> "No IT87* serial flash segment enabled."
>
> I have not yet tried writing to any of the flash chips, as I would
> really appreciate if someone with more experience than me could comment
> on this modification first, and maybe highlight problems I'm not
> currently aware of?
>
> Thanks in advance.
>
>
> Best regards,
> Pete Smith
> _______________________________________________
> flashrom mailing list -- flashrom@flashrom.org
> To unsubscribe send an email to flashrom-leave@flashrom.org