On Mon, Dec 16, 2019 at 06:01:58PM +0530, Aneesh Kumar K.V wrote: > Peter Zijlstra <peterz@xxxxxxxxxxxxx> writes: > > > Towards a more consistent naming scheme. > > > > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx> > > --- > > arch/Kconfig | 3 ++- > > arch/powerpc/Kconfig | 2 +- > > arch/sparc/Kconfig | 2 +- > > include/asm-generic/tlb.h | 2 +- > > mm/mmu_gather.c | 2 +- > > 5 files changed, 6 insertions(+), 5 deletions(-) > > > > --- a/arch/Kconfig > > +++ b/arch/Kconfig > > @@ -396,8 +396,9 @@ config HAVE_ARCH_JUMP_LABEL_RELATIVE > > config MMU_GATHER_RCU_TABLE_FREE > > bool > > > > -config HAVE_RCU_TABLE_NO_INVALIDATE > > +config MMU_GATHER_NO_TABLE_INVALIDATE > > bool > > + depends on MMU_GATHER_RCU_TABLE_FREE > > > Can we drop this Kernel config option instead use > MMU_GATHER_RCU_TABLE_FREE? IMHO reducing the kernel config related to > mmu_gather can reduce the complexity. I'm confused, are you saing you're happy to have PowerPC eat the extra TLB invalidates? I thought you cared about PPC performance :-)
