Re: [RFC v6 20/62] powerpc: store and restore the pkey state across context switches

From	Michael Ellerman <mpe@ellerman.id.au>
Newsgroups	linux.kernel
Subject	Re: [RFC v6 20/62] powerpc: store and restore the pkey state across context switches
Date	2017-07-31 15:10 +0200
Message-ID	<u9iCm-6GI-9@gated-at.bofh.it> (permalink)
References	<u3ISR-1St-3@gated-at.bofh.it> <u3Jce-2ea-33@gated-at.bofh.it> <u7UVr-zr-1@gated-at.bofh.it> <u8JuV-Nu-1@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

Ram Pai <linuxram@us.ibm.com> writes:
> On Thu, Jul 27, 2017 at 02:32:59PM -0300, Thiago Jung Bauermann wrote:
>> Ram Pai <linuxram@us.ibm.com> writes:
>> > diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
>> > index 2ad725e..9429361 100644
>> > --- a/arch/powerpc/kernel/process.c
>> > +++ b/arch/powerpc/kernel/process.c
>> > @@ -1096,6 +1096,11 @@ static inline void save_sprs(struct thread_struct *t)
>> >  		t->tar = mfspr(SPRN_TAR);
>> >  	}
>> >  #endif
>> > +#ifdef CONFIG_PPC64_MEMORY_PROTECTION_KEYS
>> > +	t->amr = mfspr(SPRN_AMR);
>> > +	t->iamr = mfspr(SPRN_IAMR);
>> > +	t->uamor = mfspr(SPRN_UAMOR);
>> > +#endif
>> >  }
>> >
>> >  static inline void restore_sprs(struct thread_struct *old_thread,
>> > @@ -1131,6 +1136,14 @@ static inline void restore_sprs(struct thread_struct *old_thread,
>> >  			mtspr(SPRN_TAR, new_thread->tar);
>> >  	}
>> >  #endif
>> > +#ifdef CONFIG_PPC64_MEMORY_PROTECTION_KEYS
>> > +	if (old_thread->amr != new_thread->amr)
>> > +		mtspr(SPRN_AMR, new_thread->amr);
>> > +	if (old_thread->iamr != new_thread->iamr)
>> > +		mtspr(SPRN_IAMR, new_thread->iamr);
>> > +	if (old_thread->uamor != new_thread->uamor)
>> > +		mtspr(SPRN_UAMOR, new_thread->uamor);
>> > +#endif
>> >  }
>> 
>> Shouldn't the saving and restoring of the SPRs be guarded by a check for
>> whether memory protection keys are enabled? What happens when trying to
>> access these registers on a CPU which doesn't have them?
>
> Good point. need to guard it.  However; i think, these registers have been
> available since power6.

The kernel runs on CPUs much older than that.

IAMR was added on Power8.

And performance is also an issue, so we should only switch them when we
need to.

cheers

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: [RFC v6 20/62] powerpc: store and restore the pkey state across context switches Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com> - 2017-07-27 19:40 +0200
  Re: [RFC v6 20/62] powerpc: store and restore the pkey state across  context switches Ram Pai <linuxram@us.ibm.com> - 2017-07-30 01:40 +0200
    Re: [RFC v6 20/62] powerpc: store and restore the pkey state across context switches Michael Ellerman <mpe@ellerman.id.au> - 2017-07-31 15:10 +0200

csiph-web