Report on Flashing Board with Intel H67

[Peter A. Braun]

Hi,

last night I flashed my MSI H67MA-ED55(B3) with E7676IMS170. As the name
says Intel 
H67 is the 
built in chipset. Flash chip is W25Q64. The flash process worked fine for
me.

Two things are a bit strange. Currently I try to investigate the reasons
for. Perhaps you have any suggestions.

1.

Sys fans cannot longer be controlled, because according BIOS parameters are
not longer accessible. Currently I don't know how to get them back.

2.

After cold booting the disabled "Full Screen Logo Display" is enabled again.
It seems to be the default. But other boot parameter are not changed back to
default.

I attached the result of "flashrom -V" (result.txt).

I hope results are detailed enough to switch entry H67 in your hardware list

onto tested.

Best regards,

Peter

[Stefan Tauner]

On Wed, 11 Jan 2012 03:40:28 +0100
"Peter A. Braun"  wrote:
> Hi,

hello peter and thanks for your report!
please do not send html mails to mailing lists. your client seems to
also send a plain text version, but that one has way too much line
breaks in it to be readable (18 line breaks instead of 1 :)
> last night I flashed my MSI H67MA-ED55(B3) with E7676IMS170. As the name says Intel  H67 is the  built in chipset. Flash chip is W25Q64. The flash process worked fine for me.

by working fine you mean that flashrom said "VERIFIED" at the end, i
presume. "E7676IMS[.]170" is the name of the file flashed in the zip
file 7676v17.zip which is version 1.7 according to MSI.
so far so good.
> Two things are a bit strange. Currently I try to investigate the reasons for. Perhaps you have any suggestions.
>   Sys fans cannot longer be controlled, because according BIOS parameters are not longer accessible. Currently I don’t know how to get them back.
>
>   After cold booting the disabled “Full Screen Logo Display” is enabled again. It seems to be the default. But other boot parameter are not changed back to default.

the changelog of that bios version mentions "Update CPU smart fan
module". so i guess what you are experiencing is that fine new update ;)
the newer intel chipsets are quite complicated in these regards and
since intel does not specify the update process in detail publicly, it
could theoretically be that flashrom is involved too. i have
experienced similar problems with an intel board (without flashrom
involved) too, so i doubt flashrom is the culprit. try removing all
power from the board (maybe even the small battery) for at least 10
seconds. this resets the embedded controller inside the chipset that is
used for hardware monitoring and that might be powered whenever there
is a power source, even if the computer is turned "off".

one way to test if flashrom might be the problem would be to rewrite the
old version (did you make a backup?) and reinstall 1.7 with MSI's tool.
from my experience if this works it would not prove that flashrom is
the problem, but only if it does not work it would prove that flashrom
is not the problem. the reason is the problems i have seen as mentioned
above.

if you did not have problems with the old version, i would suggest
rewriting your backup with flashrom.
> I attached the result of “flashrom -V” (result.txt).
>
> I hope results are detailed enough to switch entry H67 in your hardware list  onto tested.

this was done already due to an earlier report, but i will add the
mainboard to our list with a small note referencing this thread.

[Peter A. Braun]

Hi Stefan,

thank you very much for your kind reply. Wow. I really appreciate the level of detail. I were glad flashrom-community would evaluate my humble report as useful. Thank you very much.

Now coming back to your points:
- Next mails will not come in HTML but plain text. I hope changes in my mail setup/initialization file work well. Otherwise I need a hint to correct mail systems' features.
- Your assumptions concerning the meaning of "works fine" and the interpretation of update information are correct.
- Following your proposal to reset the CMOS chip I used a jumper on my MB as stated in the MSI manual. Analyzing the BIOS afterwards I found most of parameters were reset to default, but e.g. internal BIOS time hadn't been affected. What does your experience recommend? Do you think this behavior is correct or should I remove the battery additionally?
- Using MSI tools to flash BIOS is not a good idea, because e.g. MSIHQ set off virus alarms (12 of 40 findings by virustotal.com analyzing the unpacked MSIHQ tool).
- Going back to the old version is not such a good idea. You need to know -sorry for not mentioning it earlier- I had some trouble with the USB 3.0 ports. Removing or adding HD hot plugged led sometimes to blue screens (W7). That's what I want to improve by using a newer BIOS.
- I had not generated a backup copy of the old BIOS, as I have a second board of same type and age in place. I will use its BIOS if I decide to go back to the old version with the other board. Otherwise I will back up the BIOS of this test board by a different SW. By the way, does flashrom support backing up the BIOS?

I hope you and I will find all necessary information.

Best regards, Peter

-----Ursprüngliche Nachricht-----
Von: Stefan Tauner [mailto:[e-mail]] 
Gesendet: Mittwoch, 11. Januar 2012 14:02
An: Peter A. Braun
Cc: [e-mail]
Betreff: Re: [flashrom] Report on Flashing Board with Intel H67

On Wed, 11 Jan 2012 03:40:28 +0100
"Peter A. Braun"  wrote:
> Hi,

hello peter and thanks for your report!
please do not send html mails to mailing lists. your client seems to also send a plain text version, but that one has way too much line breaks in it to be readable (18 line breaks instead of 1 :)
> last night I flashed my MSI H67MA-ED55(B3) with E7676IMS170. As the name says Intel  H67 is the  built in chipset. Flash chip is W25Q64. The flash process worked fine for me.

by working fine you mean that flashrom said "VERIFIED" at the end, i presume. "E7676IMS[.]170" is the name of the file flashed in the zip file 7676v17.zip which is version 1.7 according to MSI.
so far so good.
> Two things are a bit strange. Currently I try to investigate the reasons for. Perhaps you have any suggestions.
>   Sys fans cannot longer be controlled, because according BIOS parameters are not longer accessible. Currently I don’t know how to get them back.
>
>   After cold booting the disabled “Full Screen Logo Display” is enabled again. It seems to be the default. But other boot parameter are not changed back to default.

the changelog of that bios version mentions "Update CPU smart fan module". so i guess what you are experiencing is that fine new update ;) the newer intel chipsets are quite complicated in these regards and since intel does not specify the update process in detail publicly, it could theoretically be that flashrom is involved too. i have experienced similar problems with an intel board (without flashrom
involved) too, so i doubt flashrom is the culprit. try removing all power from the board (maybe even the small battery) for at least 10 seconds. this resets the embedded controller inside the chipset that is used for hardware monitoring and that might be powered whenever there is a power source, even if the computer is turned "off".

one way to test if flashrom might be the problem would be to rewrite the old version (did you make a backup?) and reinstall 1.7 with MSI's tool.
from my experience if this works it would not prove that flashrom is the problem, but only if it does not work it would prove that flashrom is not the problem. the reason is the problems i have seen as mentioned above.

if you did not have problems with the old version, i would suggest rewriting your backup with flashrom.
> I attached the result of “flashrom -V” (result.txt).
>
> I hope results are detailed enough to switch entry H67 in your hardware list  onto tested.

this was done already due to an earlier report, but i will add the mainboard to our list with a small note referencing this thread.

[Stefan Tauner]

On Fri, 13 Jan 2012 13:57:57 +0100
"Peter A. Braun"  wrote:
> Hi Stefan,
>
> thank you very much for your kind reply. Wow. I really appreciate the level of detail. I were glad flashrom-community would evaluate my humble report as useful. Thank you very much.
>
> Now coming back to your points:
> Next mails will not come in HTML but plain text. I hope changes in my mail setup/initialization file work well. Otherwise I need a hint to correct mail systems' features.
> Your assumptions concerning the meaning of "works fine" and the interpretation of update information are correct.
> Following your proposal to reset the CMOS chip I used a jumper on my MB as stated in the MSI manual. Analyzing the BIOS afterwards I found most of parameters were reset to default, but e.g. internal BIOS time hadn't been affected. What does your experience recommend? Do you think this behavior is correct or should I remove the battery additionally?

sounds normal, and removing the battery would most probably reset the
real time clock too. if there is other data retained by the battery...
i dont know. trying if it help/change anything, wont hurt, even if i
doubt it.
> Using MSI tools to flash BIOS is not a good idea, because e.g. MSIHQ set off virus alarms (12 of 40 findings by virustotal.com analyzing the unpacked MSIHQ tool).

might still be a false positive... i deem AV products to be snake oil.
maybe the bios can update itself from USB? f9 at startup on my intel/ami
bios.
> Going back to the old version is not such a good idea. You need to know -sorry for not mentioning it earlier- I had some trouble with the USB 3.0 ports. Removing or adding HD hot plugged led sometimes to blue screens (W7). That's what I want to improve by using a newer BIOS.

your choice what is more important to you...
> I had not generated a backup copy of the old BIOS, as I have a second board of same type and age in place. I will use its BIOS if I decide to go back to the old version with the other board. Otherwise I will back up the BIOS of this test board by a different SW. By the way, does flashrom support backing up the BIOS?

flashrom -r <backupfilename>
reads out the flash content and stores it in a file which can later be
written back.