Re: [PATCH 0/6] EDAC Patch Set

[Jurgen Kramer <gtm.kramer@inter.nl.net>]

On Wed, 2006-05-24 at 10:43 -0700, Doug Thompson wrote:
> This patch set applies to kernel 2.6.17-rc4 fine and to 2.6.17-rc4-mm1 with slight
> fuzz in the MAINTAINERs file.
> 
> The following set of patches to EDAC provide various cleanups of the EDAC core and
> memory controller drivers.  Most of which came from Dave Peterson and from others.
> 
> edac-pci_dep.patch
> Change MC drivers from using CVS revisions strings for their version number,
> Now each driver has its own local string.
> 
> Remove some PCI dependencies from the core EDAC module.  Most of the code
> changes here are from a patch by Dave Jiang.
> It may be best to eventually move the PCI-specific code into a separate source file.
> 
> edac-mc_numbers_1_of_2.patch
> This is part 1 of a 2-part patch set.  The code changes are split into two
> parts to make the patches more readable.
> 
> Remove add_mc_to_global_list().  In next patch, this function will be
> reimplemented with different semantics.
> 
> edac-mc_numbers_2_of_2.patch
> This is part 2 of a 2-part patch set.
> 
> - Reimplement add_mc_to_global_list() with semantics that allow the caller to
>   determine the ID number for a mem_ctl_info structure.  Then modify
>   edac_mc_add_mc() so that the caller specifies the ID number for the new
>   mem_ctl_info structure.  Platform-specific code should be able to assign the
>   ID numbers in a platform-specific manner.  For instance, on Opteron it makes
>   sense to have the ID of the mem_ctl_info structure match the ID of the node
>   that the memory controller belongs to.
> 
> - Modify callers of edac_mc_add_mc() so they use the new semantics.
> 
> edac-probe1_cleanup_1_of_2.patch
> This is part 1 of a 2-part patch set.  The code changes are split into two
> parts to make the patches more readable.
> 
> Add lower-level functions that handle various parts of the initialization done
> by the xxx_probe1() functions.  Some of the xxx_probe1() functions are much
> too long and complicated (see "Chapter 5: Functions" in
> Documentation/CodingStyle).
> This is part 2 of a 2-part patch set.
> 
> Modify the xxx_probe1() functions so they call the lower-level functions
> created by the first patch in the set.

Will this patchset fix/suppress the "Non-Fatal Error PCI Express B"
messages I see with the E7525 edac?

I am running 2.6.16 (or more specific FC5 2.6.16-1.2111) with seems to
already include this version:

MC: drivers/edac/edac_mc.c version edac_mc  Ver: 2.0.0 May  4 2006

This version still floods my syslog with "Non-Fatal Error...." messages.

Jurgen