Re: [PATCH 2/2] KVM: PPC: Book3S HV: lockless tlbie for HPT hcalls

[Paul Mackerras <paulus@ozlabs.org>]

On Fri, Apr 06, 2018 at 04:12:32PM +1000, Michael Ellerman wrote:
> Nicholas Piggin <npiggin@gmail.com> writes:
> > diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
> > index 78e6a392330f..0221a0f74f07 100644
> > --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
> > +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
> > @@ -439,6 +439,9 @@ static inline int try_lock_tlbie(unsigned int *lock)
> >  	unsigned int tmp, old;
> >  	unsigned int token = LOCK_TOKEN;
> >  
> > +	if (mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE))
> > +		return 1;
> > +
> >  	asm volatile("1:lwarx	%1,0,%2\n"
> >  		     "	cmpwi	cr0,%1,0\n"
> >  		     "	bne	2f\n"
> > @@ -452,6 +455,12 @@ static inline int try_lock_tlbie(unsigned int *lock)
> >  	return old == 0;
> >  }
> >  
> > +static inline void unlock_tlbie_after_sync(unsigned int *lock)
> > +{
> > +	if (mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE))
> > +		return;
> > +}
> 
> So this is a bit hard to follow:
> 
> #define MMU_FTRS_DEFAULT_HPTE_ARCH_V2	\
> 	MMU_FTR_HPTE_TABLE | MMU_FTR_PPCAS_ARCH_V2
> #define MMU_FTRS_POWER		MMU_FTRS_DEFAULT_HPTE_ARCH_V2
> #define MMU_FTRS_PPC970		MMU_FTRS_POWER | MMU_FTR_TLBIE_CROP_VA			// does NOT
> #define MMU_FTRS_POWER5		MMU_FTRS_POWER | MMU_FTR_LOCKLESS_TLBIE
> #define MMU_FTRS_POWER6		MMU_FTRS_POWER5 | MMU_FTR_KERNEL_RO | MMU_FTR_68_BIT_VA	// includes lockless TLBIE
> #define MMU_FTRS_POWER7		MMU_FTRS_POWER6						// includes lockless TLBIE
> #define MMU_FTRS_POWER8		MMU_FTRS_POWER6						// includes lockless TLBIE
> #define MMU_FTRS_POWER9		MMU_FTRS_POWER6						// includes lockless TLBIE
> #define MMU_FTRS_CELL		MMU_FTRS_DEFAULT_HPTE_ARCH_V2 | 			// does NOT
> 				MMU_FTR_CI_LARGE_PAGE
> #define MMU_FTRS_PA6T		MMU_FTRS_DEFAULT_HPTE_ARCH_V2 | \			// does NOT
> 				MMU_FTR_CI_LARGE_PAGE | MMU_FTR_NO_SLBIE_B
> 
> 
> So it's only 970, Cell and Pasemi that *don't* have lockless TLBIE.
> 
> And KVM HV doesn't doesn't run on any of those.
> 
> So we can just not check for the feature in the KVM HV code.
> 
> Am I right?

Yes; that code was written when we still supported HV KVM on 970,
but we ripped that out some time ago.

Paul.