Re: [PATCH] EDAC/Intel: Fix shift-out-of-bounds when DIMM/NVDIMM is absent

[Kai-Heng Feng]

On Wed, May 17, 2023 at 1:13 AM Luck, Tony <tony.luck@xxxxxxxxx> wrote:
>
> >> [   13.875282] Hardware name: HP HP Z4 G5 Workstation Desktop PC/8962,
> > > BIOS U61 Ver. 01.01.15 04/19/2023
>
>
> >> When a DIMM slot is empty, the read value of mtr can be 0xffffffff, therefore
>
> > Looked like a buggy BIOS/hw that didn't set the mtr register.
> >
> > 1. Did you print the mtr register whose value was 0xffffffff?
> > 2. Can you take a dmesg log with kernel "CONFIG_EDAC_DEBUG=y" enabled?
> > 3. What was the CPU? Please take the output of "lscpu".
> > 4. Did you verify your patch that the issue was fixed on your systems?
>
> I wonder if BIOS is "hiding" some devices from the OS? The 0xffffffff return is
> the standard PCI response for reading a non-existent register. But that doesn't
> quite make sense with having a "dimm present" bit in the MTR register. If
> the register only exists if the DIMM is present, then there is no need for
> a "dimm present" bit.

I wonder if the "non-existent register" read is intended?

>
> Some "lspci" output may also be useful.

lspci can be found in [1]:

[1] https://bugzilla.kernel.org/show_bug.cgi?id=217453

Kai-Heng

>
> -Tony