kaiser: asm/tlbflush.h handle noPGE at lower level
authorHugh Dickins <hughd@google.com>
Sun, 5 Nov 2017 01:23:24 +0000 (18:23 -0700)
committerYves-Alexis Perez <corsac@debian.org>
Thu, 4 Jan 2018 11:12:40 +0000 (11:12 +0000)
commitabcc8f2648cd8734153f65251e2cb9b44e5c7684
tree5e644511ac5139bfb0223deb9e4e3a63c3b03583
parent1e4cb51d5a3be6c179001f4f896a2bdbaeefc44b
kaiser: asm/tlbflush.h handle noPGE at lower level

I found asm/tlbflush.h too twisty, and think it safer not to avoid
__native_flush_tlb_global_irq_disabled() in the kaiser_enabled case,
but instead let it handle kaiser_enabled along with cr3: it can just
use __native_flush_tlb() for that, no harm in re-disabling preemption.

(This is not the same change as Kirill and Dave have suggested for
upstream, flipping PGE in cr4: that's neat, but needs a cpu_has_pge
check; cr3 is enough for kaiser, and thought to be cheaper than cr4.)

Also delete the X86_FEATURE_INVPCID invpcid_flush_all_nonglobals()
preference from __native_flush_tlb(): unlike the invpcid_flush_all()
preference in __native_flush_tlb_global(), it's not seen in upstream
4.14, and was recently reported to be surprisingly slow.

Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Gbp-Pq: Topic features/all/kpti
Gbp-Pq: Name kaiser-asm-tlbflush.h-handle-nopge-at-lower-level.patch
arch/x86/include/asm/tlbflush.h