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

From: Zhuo, Qiuxu
Date: Thu Jul 06 2023 - 10:19:22 EST


> From: Kai-Heng Feng <kai.heng.feng@xxxxxxxxxxxxx>
> ...
> > > > Some "lspci" output may also be useful.
> > >
> > > lspci can be found in [1]:
> > >
> > > [1] https://bugzilla.kernel.org/show_bug.cgi?id=217453
> >
> > A gentle ping...
>
> Another gentle ping...

The same issue was also reported recently by Koba Ko:
https://lore.kernel.org/linux-edac/CAJB-X+XtfBm0a4btt6NT9rvdrxETNLNMVQ3G=u513Nh8RKwjWw@xxxxxxxxxxxxxx/T/#t

The root cause of this issue was that the absent memory controllers still
appeared as PCIe devices that fooled the i10nm_edac driver.

Could you please verify the new fix patch below on your machine
and feedback on the testing results? Thanks!
https://lore.kernel.org/linux-edac/20230706134216.37044-1-qiuxu.zhuo@xxxxxxxxx/T/#u

- Qiuxu