x86-64: Don't apply destructive erratum workaround on unaffected CPUs
commite05139f2569ecf699b229a6473a86cdffed62956
authorJan Beulich <JBeulich@suse.com>
Wed, 28 Sep 2011 15:56:51 +0000 (28 16:56 +0100)
committerIngo Molnar <mingo@elte.hu>
Wed, 28 Sep 2011 17:04:48 +0000 (28 19:04 +0200)
tree58882e5b22d9f309c8748f72d25e92ee8de883b7
parenta102a9ece5489e1718cd7543aa079082450ac3a2
x86-64: Don't apply destructive erratum workaround on unaffected CPUs

Erratum 93 applies to AMD K8 CPUs only, and its workaround
(forcing the upper 32 bits of %rip to all get set under certain
conditions) is actually getting in the way of analyzing page
faults occurring during EFI physical mode runtime calls (in
particular the page table walk shown is completely unrelated to
the actual fault). This is because typically EFI runtime code
lives in the space between 2G and 4G, which - modulo the above
manipulation - is likely to overlap with the kernel or modules
area.

While even for the other errata workarounds their taking effect
could be limited to just the affected CPUs, none of them appears
to be destructive, and they're generally getting called only
outside of performance critical paths, so they're being left
untouched.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Link: http://lkml.kernel.org/r/4E835FE30200007800058464@nat28.tlf.novell.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/mm/fault.c