Re: [coreboot] Fwd: Ethernet problem in x200

[Nico Huber]

Hello Serdar, Zoran,

On 27.03.2017 16:14, Zoran Stojsavljevic wrote:
> [user@localhost projects]$ cat dmesg.txt | grep 00:19.0
> [    0.151652] pci 0000:00:19.0: *[8086:294c*] type 00 class 0x020000
> [    0.151694] pci 0000:00:19.0: reg 0x10: [mem 0xe1600000-0xe161ffff]
> [    0.151707] pci 0000:00:19.0: reg 0x14: [mem 0xe1624000-0xe1624fff]
> [    0.151721] pci 0000:00:19.0: reg 0x18: [io  0x3000-0x301f]
> [    0.151802] pci 0000:00:19.0: PME# supported from D0 D3hot D3cold
> [    1.489719] e1000e 0000:00:19.0: Interrupt Throttling Rate (ints/sec)
> set to dynamic conservative mode
> [    1.611526] e1000e 0000:00:19.0: The NVM Checksum Is Not Valid

I can't say for sure, but this looks much like the GbE and/or Firmware
Descriptor regions of the BIOS flash are corrupted. If in doubt, you can
generate a layout file for flashrom from the backup of the vendor BIOS
with `ifdtool -f` (see util/ifdtool/ in the coreboot tree). And flash
back these selected regions with `flashrom -l layoutfile -i fd -i gbe
...` from the backup.
> *[    1.635873] e1000e: probe of 0000:00:19.0 failed with error -5*
> [user@localhost projects]$

Hope that helps,
Nico

Fwd: Ethernet problem in x200

[Zoran Stojsavljevic]

*> [    1.611526] e1000e 0000:00:19.0: The NVM Checksum Is Not Valid*

I admit, I missed this one... But I'll improve, since, according to my
expectations, my mind "blacklisted" possibility that GbE descriptor is
corrupted. I am more on Linux/kernel side, and every day I read somebody's
dmesg, which come out of/based upon completely healthy BIOS. Coreboot as
BSP asset for ASUS, GIGABYTE and such vendors, in mobile, WS, DT and
Servers is not almost at all present?! :-(

Anyway, good catch, I'll add this one to my (growing) rich understanding of
dmesg logs! :-)

Thank you,
Zoran

On Mon, Mar 27, 2017 at 4:33 PM, Nico Huber  wrote:
> Hello Serdar, Zoran,
>
> On 27.03.2017 16:14, Zoran Stojsavljevic wrote:
> > [user@localhost projects]$ cat dmesg.txt | grep 00:19.0
> > [    0.151652] pci 0000:00:19.0: *[8086:294c*] type 00 class 0x020000
> > [    0.151694] pci 0000:00:19.0: reg 0x10: [mem 0xe1600000-0xe161ffff]
> > [    0.151707] pci 0000:00:19.0: reg 0x14: [mem 0xe1624000-0xe1624fff]
> > [    0.151721] pci 0000:00:19.0: reg 0x18: [io  0x3000-0x301f]
> > [    0.151802] pci 0000:00:19.0: PME# supported from D0 D3hot D3cold
> > [    1.489719] e1000e 0000:00:19.0: Interrupt Throttling Rate (ints/sec)
> > set to dynamic conservative mode
> > [    1.611526] e1000e 0000:00:19.0: The NVM Checksum Is Not Valid
>
> I can't say for sure, but this looks much like the GbE and/or Firmware
> Descriptor regions of the BIOS flash are corrupted. If in doubt, you can
> generate a layout file for flashrom from the backup of the vendor BIOS
> with `ifdtool -f` (see util/ifdtool/ in the coreboot tree). And flash
> back these selected regions with `flashrom -l layoutfile -i fd -i gbe
> ...` from the backup.
> > *[    1.635873] e1000e: probe of 0000:00:19.0 failed with error -5*
> > [user@localhost projects]$
>
> Hope that helps,
> Nico