MSI X570-A PRO (MS-7C37)

[Clay Daniels]

I'm just trying to read what bios info I can:

MSI X570-A PRO (MS-7C37)
AMD Ryzen 7 3700X
FreeBSD fbsd13 13.0-CURRENT FreeBSD 13.0-CURRENT

root@fbsd13:~ # flashrom -p internal
flashrom v1.2 on FreeBSD 13.0-CURRENT (amd64)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 4, resolution: 1ns).
Found chipset "AMD FP4".
Enabling flash write... FCH device found but SMBus revision 0x61 does not
match known values.
Please report this to [e-mail] and include this log and
the output of lspci -nnvx, thanks!.
Could not determine chipset generation.PROBLEMS, continuing anyway
No EEPROM/flash device found.
Note: flashrom can never write if the flash chip isn't found automatically.

pciconf -lvb output attached, as well as flashrom -V -p internal (the
verbose version)

My first question is: It looks to me that "AMD FP4" is just a BGA (FP4)
Socket, not a chip. Newbie to flashrom & coreboot.

Thanks,
Clay

[Ivan Ivanov]

While doing the internal flashrom operations, that BIOS chip is
situated behind a southbridge that you find by AMD FP4 name. Looks
like either its' support isn't good enough at flashrom or UEFI
firmware / EC controller somehow disturb the operation. Maybe try to
access a BIOS chip directly with the external programmer like usb
ch341a?

ср, 9 сент. 2020 г. в 10:52, Clay Daniels :
> I'm just trying to read what bios info I can:
>
> MSI X570-A PRO (MS-7C37)
> AMD Ryzen 7 3700X
> FreeBSD fbsd13 13.0-CURRENT FreeBSD 13.0-CURRENT
>
> root@fbsd13:~ # flashrom -p internal
> flashrom v1.2 on FreeBSD 13.0-CURRENT (amd64)
> flashrom is free software, get the source code at https://flashrom.org
>
> Using clock_gettime for delay loops (clk_id: 4, resolution: 1ns).
> Found chipset "AMD FP4".
> Enabling flash write... FCH device found but SMBus revision 0x61 does not match known values.
> Please report this to [e-mail] and include this log and
> the output of lspci -nnvx, thanks!.
> Could not determine chipset generation.PROBLEMS, continuing anyway
> No EEPROM/flash device found.
> Note: flashrom can never write if the flash chip isn't found automatically.
>
> pciconf -lvb output attached, as well as flashrom -V -p internal (the verbose version)
>
> My first question is: It looks to me that "AMD FP4" is just a BGA (FP4) Socket, not a chip. Newbie to flashrom & coreboot.
>
> Thanks,
> Clay
>
> flashrom mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Clay Daniels]

On Wed, Sep 16, 2020 at 11:43 AM Ivan Ivanov  wrote:
> While doing the internal flashrom operations, that BIOS chip is
> situated behind a southbridge that you find by AMD FP4 name. Looks
> like either its' support isn't good enough at flashrom or UEFI
> firmware / EC controller somehow disturb the operation. Maybe try to
> access a BIOS chip directly with the external programmer like usb
> ch341a?

Ivan, thanks for the useful info. Your explanation likely tells me why I
can't see my bios chip. I just looked and the little usb devices are
available from Newegg where I got the parts to build my Ryzen 7 machine.
Before I order one, do you or anyone else on the list have suggestions on
what to look for in a usb ch341a external programmer? Newegg has a wide
variety.

You may have noticed my https://paste.flashrom.org/  lspci file. I added a
Ubuntu Linux disk as FreeBSD doesn't do lspci, just pciconf. I also tried
(twice) to load a pciconf -lvb from FreeBSD, which is there but called
lspci ;-(

Anyway, thanks for your help,
Clay
> ср, 9 сент. 2020 г. в 10:52, Clay Daniels :
> > I'm just trying to read what bios info I can:
> >
> > MSI X570-A PRO (MS-7C37)
> > AMD Ryzen 7 3700X
> > FreeBSD fbsd13 13.0-CURRENT FreeBSD 13.0-CURRENT
> >
> > root@fbsd13:~ # flashrom -p internal
> > flashrom v1.2 on FreeBSD 13.0-CURRENT (amd64)
> > flashrom is free software, get the source code at https://flashrom.org
> >
> > Using clock_gettime for delay loops (clk_id: 4, resolution: 1ns).
> > Found chipset "AMD FP4".
> > Enabling flash write... FCH device found but SMBus revision 0x61 does
>
> not match known values.
> > Please report this to [e-mail] and include this log and
> > the output of lspci -nnvx, thanks!.
> > Could not determine chipset generation.PROBLEMS, continuing anyway
> > No EEPROM/flash device found.
> > Note: flashrom can never write if the flash chip isn't found
>
> automatically.
> > pciconf -lvb output attached, as well as flashrom -V -p internal (the
>
> verbose version)
> > My first question is: It looks to me that "AMD FP4" is just a BGA (FP4)
>
> Socket, not a chip. Newbie to flashrom & coreboot.
> > Thanks,
> > Clay
> >
> > flashrom mailing list -- [e-mail]
> > To unsubscribe send an email to [e-mail]

[Ivan Ivanov]

CH341A are really overpriced at Newegg: you can get it from AliExpress
China for less than 2 dollars + free shipping, if this project isn't
urgent and you could wait about 1 month for it to arrive. From the
posts online it seems that CH341A with a green PCB is preferable -
less likely to have a hardware design bug causing it to output 5V
instead of 3.3V: I heard that some black PCB ones suffer from it
(although possible to fix with some soldering).

The desktop boards used to have a DIP8 "shape" BIOS chip plugged into
a socket, from where you could easily remove it using a PLCC clip.
Sometimes they have a soldered SOIC-8 "shape" chip like many laptops,
but its' bearable since there are good SOIC-8 test clips available,
using which you can connect to SOIC-8 BIOS chip without soldering and
do the ISP (In-System Programming). However, in your case - your board
seems to have a WSON-8 "shape" chip, which really sucks as there are
no (good/cheap) test clip adapters last time I checked. But at least
there's a MSI JSPI1 header near this chip, so maybe you can use it for
flashing. Read more about it at flashrom wiki and elsewhere. So, you
may also need some 1.27mm 1P male - female or female-female cables to
connect a CH341A to this header (10 cm length is recommended, although
you can get longer cables and manually resolder them if it turns out
that a flashing operation isn't reliable). Additionally, USB extension
cable of ~1m length will make all this more convenient. And you'll use
another PC with some Linux loaded (either from HDD or LiveUSB) and
flashrom, to read from and write to a BIOS chip of this motherboard.

Thank you for providing a lspci file, although I'm not skilled enough
to tell what exactly caused a flashrom's internal mode to fail - maybe
someone else can help you.

ср, 16 сент. 2020 г. в 22:59, Clay Daniels :
> On Wed, Sep 16, 2020 at 11:43 AM Ivan Ivanov  wrote:
> > While doing the internal flashrom operations, that BIOS chip is
> > situated behind a southbridge that you find by AMD FP4 name. Looks
> > like either its' support isn't good enough at flashrom or UEFI
> > firmware / EC controller somehow disturb the operation. Maybe try to
> > access a BIOS chip directly with the external programmer like usb
> > ch341a?
>
> Ivan, thanks for the useful info. Your explanation likely tells me why I can't see my bios chip. I just looked and the little usb devices are available from Newegg where I got the parts to build my Ryzen 7 machine. Before I order one, do you or anyone else on the list have suggestions on what to look for in a usb ch341a external programmer? Newegg has a wide variety.
>
> You may have noticed my https://paste.flashrom.org/  lspci file. I added a Ubuntu Linux disk as FreeBSD doesn't do lspci, just pciconf. I also tried (twice) to load a pciconf -lvb from FreeBSD, which is there but called lspci ;-(
>
> Anyway, thanks for your help,
> Clay
> > ср, 9 сент. 2020 г. в 10:52, Clay Daniels :
> > > I'm just trying to read what bios info I can:
> > >
> > > MSI X570-A PRO (MS-7C37)
> > > AMD Ryzen 7 3700X
> > > FreeBSD fbsd13 13.0-CURRENT FreeBSD 13.0-CURRENT
> > >
> > > root@fbsd13:~ # flashrom -p internal
> > > flashrom v1.2 on FreeBSD 13.0-CURRENT (amd64)
> > > flashrom is free software, get the source code at https://flashrom.org
> > >
> > > Using clock_gettime for delay loops (clk_id: 4, resolution: 1ns).
> > > Found chipset "AMD FP4".
> > > Enabling flash write... FCH device found but SMBus revision 0x61 does not match known values.
> > > Please report this to [e-mail] and include this log and
> > > the output of lspci -nnvx, thanks!.
> > > Could not determine chipset generation.PROBLEMS, continuing anyway
> > > No EEPROM/flash device found.
> > > Note: flashrom can never write if the flash chip isn't found automatically.
> > >
> > > pciconf -lvb output attached, as well as flashrom -V -p internal (the verbose version)
> > >
> > > My first question is: It looks to me that "AMD FP4" is just a BGA (FP4) Socket, not a chip. Newbie to flashrom & coreboot.
> > >
> > > Thanks,
> > > Clay
> > >
> > > flashrom mailing list -- [e-mail]
> > > To unsubscribe send an email to [e-mail]

[Clay Daniels]

Ivan, thanks so much for the valuable information! It seems to me you have
found a more detailed picture of the motherboard than I have. All I can
find is marketing pictures, and a 'quick start guide', but no real picture
that shows stuff like you describe. Please do let me know the link to the
picture you are looking at.

The project is not really that urgent, and I have no trouble using the MSI
flash routine in the bios. The board is new enough that there are regular
updates to the bios, 10 versions (1 thru A) since I bought the motherboard
& built the machine the summer of 2019.

Another thing that MSI offers on the motherboard is a "flash button" and
dedicated usb port for the source bios on the back of the board. I have not
used this feature, but it may offer some direct flash route to the bios for
all I know.

The MSI bios is ok, and seems very advanced with a lot of graphics but
actually is kind of primitive in how it deals with bootable disks and
devices. Everything is an icon, and sometimes the icons don't match the
device very well. I think it's geared to a gamer who has one big disk
drive, and they really want you to use UEFI/GPT, and treat MBR as totally a
thing of the past, which may or may not be true. I have NetBSD running on a
MBR whole disk setup on my older 2014 HP Pavilion that originally came with
Windows 8.1. It works fine using MBR & NetBSD. No icons in that bios. The
machine is really nice for what it is, but is too slow for much action. The
newer home-built machine is my main hobby, I suppose.

Anyway, thanks for the info, and do send me a link to the detailed
motherboard layout you found.

Clay

On Thu, Sep 17, 2020 at 4:46 AM Ivan Ivanov  wrote:
> CH341A are really overpriced at Newegg: you can get it from AliExpress
> China for less than 2 dollars + free shipping, if this project isn't
> urgent and you could wait about 1 month for it to arrive. From the
> posts online it seems that CH341A with a green PCB is preferable -
> less likely to have a hardware design bug causing it to output 5V
> instead of 3.3V: I heard that some black PCB ones suffer from it
> (although possible to fix with some soldering).
>
> The desktop boards used to have a DIP8 "shape" BIOS chip plugged into
> a socket, from where you could easily remove it using a PLCC clip.
> Sometimes they have a soldered SOIC-8 "shape" chip like many laptops,
> but its' bearable since there are good SOIC-8 test clips available,
> using which you can connect to SOIC-8 BIOS chip without soldering and
> do the ISP (In-System Programming). However, in your case - your board
> seems to have a WSON-8 "shape" chip, which really sucks as there are
> no (good/cheap) test clip adapters last time I checked. But at least
> there's a MSI JSPI1 header near this chip, so maybe you can use it for
> flashing. Read more about it at flashrom wiki and elsewhere. So, you
> may also need some 1.27mm 1P male - female or female-female cables to
> connect a CH341A to this header (10 cm length is recommended, although
> you can get longer cables and manually resolder them if it turns out
> that a flashing operation isn't reliable). Additionally, USB extension
> cable of ~1m length will make all this more convenient. And you'll use
> another PC with some Linux loaded (either from HDD or LiveUSB) and
> flashrom, to read from and write to a BIOS chip of this motherboard.
>
> Thank you for providing a lspci file, although I'm not skilled enough
> to tell what exactly caused a flashrom's internal mode to fail - maybe
> someone else can help you.
>
> ср, 16 сент. 2020 г. в 22:59, Clay Daniels :
> > On Wed, Sep 16, 2020 at 11:43 AM Ivan Ivanov
>
> wrote:
> > > While doing the internal flashrom operations, that BIOS chip is
> > > situated behind a southbridge that you find by AMD FP4 name. Looks
> > > like either its' support isn't good enough at flashrom or UEFI
> > > firmware / EC controller somehow disturb the operation. Maybe try to
> > > access a BIOS chip directly with the external programmer like usb
> > > ch341a?
> >
> > Ivan, thanks for the useful info. Your explanation likely tells me why I
>
> can't see my bios chip. I just looked and the little usb devices are
> available from Newegg where I got the parts to build my Ryzen 7 machine.
> Before I order one, do you or anyone else on the list have suggestions on
> what to look for in a usb ch341a external programmer? Newegg has a wide
> variety.
> > You may have noticed my https://paste.flashrom.org/  lspci file. I
>
> added a Ubuntu Linux disk as FreeBSD doesn't do lspci, just pciconf. I also
> tried (twice) to load a pciconf -lvb from FreeBSD, which is there but
> called lspci ;-(
> > Anyway, thanks for your help,
> > Clay
> > > ср, 9 сент. 2020 г. в 10:52, Clay Daniels :
> > > > I'm just trying to read what bios info I can:
> > > >
> > > > MSI X570-A PRO (MS-7C37)
> > > > AMD Ryzen 7 3700X
> > > > FreeBSD fbsd13 13.0-CURRENT FreeBSD 13.0-CURRENT
> > > >
> > > > root@fbsd13:~ # flashrom -p internal
> > > > flashrom v1.2 on FreeBSD 13.0-CURRENT (amd64)
> > > > flashrom is free software, get the source code at
>
> https://flashrom.org
> > > > Using clock_gettime for delay loops (clk_id: 4, resolution: 1ns).
> > > > Found chipset "AMD FP4".
> > > > Enabling flash write... FCH device found but SMBus revision 0x61 does
>
> not match known values.
> > > > Please report this to [e-mail] and include this log and
> > > > the output of lspci -nnvx, thanks!.
> > > > Could not determine chipset generation.PROBLEMS, continuing anyway
> > > > No EEPROM/flash device found.
> > > > Note: flashrom can never write if the flash chip isn't found
>
> automatically.
> > > > pciconf -lvb output attached, as well as flashrom -V -p internal (the
>
> verbose version)
> > > > My first question is: It looks to me that "AMD FP4" is just a BGA
>
> (FP4) Socket, not a chip. Newbie to flashrom & coreboot.
> > > > Thanks,
> > > > Clay
> > > >
> > > > flashrom mailing list -- [e-mail]
> > > > To unsubscribe send an email to [e-mail]