[v4,2/2] EDAC, sb_edac: Fix reporting wrong DIMM when patrol scrubber finds error

[Borislav Petkov <bp@alien8.de>]

On Mon, Sep 10, 2018 at 02:11:45PM -0700, Luck, Tony wrote:
> From: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
> 
> The sb_edac driver sometimes reports the wrong DIMM for a memory
> error found by the patrol scrubber. It's rooted in h/w that
> only provides a 4KB page aligned address for the error case.
> This means that the EDAC driver will point at the DIMM matching
> offset 0x0 in the 4KB page, but because of interleaving across
> channels and ranks the actual DIMM involved may be different
> if the error is on some other cache line within the page.
> 
> For this case, we can reconstruct the socket/iMC/channel
> information from the "mce" structure passed to the EDAC driver.
> We cannot determine the DIMM, so pass "dimm=-1" to the EDAC core.
> It will report all the DIMMs on that channel may be affected.
> 
> [Tony: Improved comments on the functions to convert bank number
>  to memory controller number. Minor cleanup to commit comment]
> 
> Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
> Signed-off-by: Tony Luck <tony.luck@intel.com>
> ---
> 
> v4:
> 	Fixed typo s/detemine/determine/
> 	Changed the "msg" when we can't determine the "ha" to
> 	include the bank number.  I've left this in the "msg"
> 	because we do use that in the error path to the edac
> 	core code. I don't think a KERN_ERROR message will
> 	help (especially as this is a "Can't happen(TM)" error).

Fair enough.

All applied,
thx.