[flashrom] "ERASED FAILED!" messages for ICH2 / SST49LF0002A/B

Carl-Daniel Hailfinger c-d.hailfinger.devel.2006 at gmx.net
Wed Sep 30 01:12:09 CEST 2009


Hi Allan,

important things first:
I looked at the messages you got and also at the great debugging you did
(old dump of the BIOS is identical to the new dup of the BIOS) and I'm
very sure your machine is OK and can be rebooted anytime.


On 29.09.2009 23:31, Allan Bjorklund wrote:
> A Sony Vaio (PCV-RX650(UC)), with an ASUSTeK mother board (P4B266LM
> Rev: 1.xx), running Linux kernel 2.6.28.
>   

Flashrom interacts non-trivially with the embedded controllers in
laptops, so we recommend not to use flashrom on any laptop.


> flashrom v0.9.1-runknown
> No coreboot table found.
> Found chipset "Intel ICH2", enabling flash write... OK.
> Found chip "SST SST49LF002A/B" (256 KB, FWH) at physical address 0xfffc0000.
>   

Apparently your laptop is among those which have the flash chip attached
directly to the southbridge, which is a good thing (makes using flashrom
possible).


> I saved a copy of my existing BIOS image with: flashrom -r

Good decision.


> flashrom -w original_bios.bin, and here is the output:
>
> flashrom v0.9.1-runknown
> Found chipset "Intel ICH2", enabling flash write... OK.
> Found chip "SST SST49LF002A/B" (256 KB, FWH) at physical address 0xfffc0000.
> Flash image seems to be a legacy BIOS. Disabling checks.
> Writing flash chip... 
> ERASE FAILED at 0x00000000! Expected=0xff, Read=0x25, failed byte count from 0x00000000-0x00003fff: 0x3fa4
> ERASE FAILED!
> ERASE FAILED!
> ERASE FAILED!
> ERASE FAILED!
> FAILED!
>   

Since erase failed at 0x00000000 (first address), this is a good
indicator the erase had no effect at all.


> If I do a flashrom -v original_bios.bin, I get:
>
> flashrom v0.9.1-runknown
> No coreboot table found.
> Found chipset "Intel ICH2", enabling flash write... OK.
> This chipset supports the following protocols: Non-SPI.
> Calibrating delay loop... OK.
> Found chip "SST SST49LF002A/B" (256 KB, FWH) at physical address 0xfffc0000.
> Flash image seems to be a legacy BIOS. Disabling checks.
> Verifying flash... VERIFIED.
>
> And reading a second copy and doing a binary diff shows no
> differences, so my BIOS checks out as not being damaged.
>   

Absolutely correct.


> So, any advice on what to look for/change to make this work?
>   

Let me give you two advices, and select the one which fits you best:
- Don't use flashrom on any laptop. The interactions with the embedded
controller can lead to poweroff, stopped battery charging and laptop
overheating. It is technically impossible to probe for most of these
problematic embedded controllers, so there is nothing flashrom can do to
tell the user about this.
- Ask libv in #flashrom on freenode to develop a board enable routine
for your board. This is quite some work (not less than half an hour of
experienced developer time) and on laptops this may have the side
effects mentioned above. Luc (libv) needs the following data from you:
"lspci -nnvvvxxx", "superiotool -dV" (both run as root, output attached
to a reply mail in this thread), and a link to the BIOS file download
(please don't post the BIOS file itself to this list). Refer him to the
mail which started this thread (which has flashrom output and the board
name) and to the mail with the data I requested from you. That should
give him the info you need.

Regards,
Carl-Daniel

-- 
http://www.hailfinger.org/





More information about the flashrom mailing list