Messages in this thread | | | Date | Fri, 12 Jul 2013 13:13:40 -0300 | From | Mauro Carvalho Chehab <> | Subject | Re: BUG: key ffff880c1148c478 not in .data! (V3.10.0) |
| |
Em Fri, 12 Jul 2013 16:21:06 +0200 Borislav Petkov <bp@alien8.de> escreveu:
> On Fri, Jul 12, 2013 at 10:57:41AM -0300, Mauro Carvalho Chehab wrote: > > This will be overriding the content of the static var mc_bus every for > > every new memory controller. Are you sure that bus.name is only used > > on register, or if its contents is stored somewhere? > > bus_register does kobject_set_name which copies bus->name, for example,
Ok, so, it could be safe.
> but I didn't look exhaustively.
Did you try to remove and reinsert the edac driver a few times, on a multi-memory controller machine? The bus nodes got created properly? > > Just to be on the safe side, I should probably do a > > static const char **bus_names = { "mc0", "mc1", ..., "mc7" };
You would likely to use an array for the bus_type too, if reusing the static one is an issue.
> and use it. Are 8 enough for your edac drivers too?
With edac_ghes, I suspect that the worse case, on Intel side, is the Nehalem/Sandy Bridge/Ivy Bridge EX machines.
Tony,
What would be a reasonable maximum limit for the number of memory controllers, on a -EX machine?
Cheers, Mauro
| |