Abit NF-M2 nView upgrade problem

[Grzesiek Pycia]

Hi

I'm trying to upgrade bios on Abit NF-M2 nView mainboard which is marked
as supported but it is impossible, because it fails on erasing the chip,
so I'm sending log as advised by flashrom.
If there are some other information needed or testing please let me know
I'll be happy to help out.

Regards
Grzegorz

[Stefan Tauner]

On Tue, 10 Jan 2012 22:43:30 +0100
Grzesiek Pycia  wrote:
> Hi
>
> I'm trying to upgrade bios on Abit NF-M2 nView mainboard which is marked
> as supported but it is impossible, because it fails on erasing the chip,
> so I'm sending log as advised by flashrom.
> If there are some other information needed or testing please let me know
> I'll be happy to help out.

hello,

your board needs a small piece of code to make the flash rom writable.
actually we already have it in the source and it is just not triggered
on your board. the question is why. we identify mainboards via PCI IDs
(and sometimes SMBIOS data) and apparently yours does not match our
pattern. please provide the output of
lspci -nn
lspci -nnxxxvvv

[Grzesiek]

On 11.01.2012 00:54, Stefan Tauner wrote:
> On Tue, 10 Jan 2012 22:43:30 +0100
> Grzesiek Pycia  wrote:
> > Hi
> >
> > I'm trying to upgrade bios on Abit NF-M2 nView mainboard which is marked
> > as supported but it is impossible, because it fails on erasing the chip,
> > so I'm sending log as advised by flashrom.
> > If there are some other information needed or testing please let me know
> > I'll be happy to help out.
>
> hello,
>
> your board needs a small piece of code to make the flash rom writable.
> actually we already have it in the source and it is just not triggered
> on your board. the question is why. we identify mainboards via PCI IDs
> (and sometimes SMBIOS data) and apparently yours does not match our
> pattern. please provide the output of
> lspci -nn
> lspci -nnxxxvvv

There it goes.

[Stefan Tauner]

On Wed, 11 Jan 2012 08:45:39 +0100
Grzesiek  wrote:
> On 11.01.2012 00:54, Stefan Tauner wrote:
> > On Tue, 10 Jan 2012 22:43:30 +0100
> > Grzesiek Pycia  wrote:
> > > Hi
> > >
> > > I'm trying to upgrade bios on Abit NF-M2 nView mainboard which is marked
> > > as supported but it is impossible, because it fails on erasing the chip,
> > > so I'm sending log as advised by flashrom.
> > > If there are some other information needed or testing please let me know
> > > I'll be happy to help out.
> >
> > hello,
> >
> > your board needs a small piece of code to make the flash rom writable.
> > actually we already have it in the source and it is just not triggered
> > on your board. the question is why. we identify mainboards via PCI IDs
> > (and sometimes SMBIOS data) and apparently yours does not match our
> > pattern. please provide the output of
> > lspci -nn
> > lspci -nnxxxvvv
>
> There it goes.

thanks!
your log shows that you have added a GeForce 8600 GT discrete graphics
card. normally we try to avoid to include pci devices in our pattern
that can easily be changed or disabled in the firmware, but in the
existing board enable code the onboard card was specified. this is a
counter example of what should have been done obviously.

please change the source as specified in the patch snippet below and
recompile. it should then detect your mainboard reliably. it is a
one-line change in board_enable.c that removes the VGA IDs and adds the
IDs from the LPC controller, which should always be there.

sorry for this! please report back if it has worked. i will commit it
then to our repository with other small changes. thanks!

diff --git a/board_enable.c b/board_enable.c
index 2229f8b..e9a630b 100644
@@ -2090,7 +2090,7 @@ const struct board_match board_matches[] = {
    {0x8086, 0x2930, 0x147b, 0x1083,  0x10ec, 0x8167, 0x147b, 0x1083, NULL,         NULL, NULL,           P3, "abit",        "IP35 Pro",              0,   OK, intel_ich_gpio16_raise},
    {0x10de, 0x0050, 0x147b, 0x1c1a,       0,      0,      0,      0, NULL,         NULL, NULL,           P3, "abit",        "KN8 Ultra",             0,   NT, nvidia_mcp_gpio2_lower},
    {0x10de, 0x01e0, 0x147b, 0x1c00,  0x10de, 0x0060, 0x147B, 0x1c00, NULL,         NULL, NULL,           P3, "abit",        "NF7-S",                 0,   OK, nvidia_mcp_gpio8_raise},
-	{0x10de, 0x02f0, 0x147b, 0x1c26,  0x10de, 0x0240, 0x10de, 0x0222, NULL,         NULL, NULL,           P3, "abit",        "NF-M2 nView",           0,   OK, nvidia_mcp_gpio4_lower},
+	{0x10de, 0x02f0, 0x147b, 0x1c26,  0x10de, 0x0260, 0x147b, 0x1c26, NULL,         NULL, NULL,           P3, "abit",        "NF-M2 nView",           0,   OK, nvidia_mcp_gpio4_lower},
    {0x1106, 0x0691,      0,      0,  0x1106, 0x3057,      0,      0, "(VA6)$",     NULL, NULL,           P3, "abit",        "VA6",                   0,   OK, via_apollo_gpo4_lower},
    {0x1106, 0x0691,      0,      0,  0x1106, 0x3057,      0,      0, NULL,         "abit", "vt6x4",      P3, "abit",        "VT6X4",                 0,   OK, via_apollo_gpo4_lower},
    {0x105a, 0x0d30, 0x105a, 0x4d33,  0x8086, 0x1130, 0x8086,      0, NULL,         NULL, NULL,           P3, "Acorp",       "6A815EPD",              0,   OK, board_acorp_6a815epd},

[Grzesiek]

On 11.01.2012 13:27, Stefan Tauner wrote:
> On Wed, 11 Jan 2012 08:45:39 +0100
> Grzesiek  wrote:
> > On 11.01.2012 00:54, Stefan Tauner wrote:
> > > On Tue, 10 Jan 2012 22:43:30 +0100
> > > Grzesiek Pycia   wrote:
> > > > Hi
> > > >
> > > > I'm trying to upgrade bios on Abit NF-M2 nView mainboard which is marked
> > > > as supported but it is impossible, because it fails on erasing the chip,
> > > > so I'm sending log as advised by flashrom.
> > > > If there are some other information needed or testing please let me know
> > > > I'll be happy to help out.
> > >
> > > hello,
> > >
> > > your board needs a small piece of code to make the flash rom writable.
> > > actually we already have it in the source and it is just not triggered
> > > on your board. the question is why. we identify mainboards via PCI IDs
> > > (and sometimes SMBIOS data) and apparently yours does not match our
> > > pattern. please provide the output of
> > > lspci -nn
> > > lspci -nnxxxvvv
> >
> > There it goes.
>
> thanks!
> your log shows that you have added a GeForce 8600 GT discrete graphics
> card. normally we try to avoid to include pci devices in our pattern
> that can easily be changed or disabled in the firmware, but in the
> existing board enable code the onboard card was specified. this is a
> counter example of what should have been done obviously.
>
> please change the source as specified in the patch snippet below and
> recompile. it should then detect your mainboard reliably. it is a
> one-line change in board_enable.c that removes the VGA IDs and adds the
> IDs from the LPC controller, which should always be there.
>
> sorry for this! please report back if it has worked. i will commit it
> then to our repository with other small changes. thanks!

Seems to work, update and verification was successful.
I'll report back in few hours if the board does work after reboot, 
because it is at remote location.

Regards

[Grzesiek Pycia]

W dniu 11.01.2012 14:16, Grzesiek pisze:
> On 11.01.2012 13:27, Stefan Tauner wrote:
> > On Wed, 11 Jan 2012 08:45:39 +0100
> > Grzesiek  wrote:
> >
> > sorry for this! please report back if it has worked. i will commit it
> > then to our repository with other small changes. thanks!
>
> Seems to work, update and verification was successful.
> I'll report back in few hours if the board does work after reboot,
> because it is at remote location.

Upgrade was successful, after reboot the mainboard works fine and bios
is updated so You can commit changes.

Thanks for help.

Regards
Grzegorz