Re: [1/6] powerpc/64s: Add barrier_nospec

2018-06-04 Thread Michael Ellerman
On Tue, 2018-04-24 at 04:15:54 UTC, Michael Ellerman wrote: > From: Michal Suchanek > > A no-op form of ori (or immediate of 0 into r31 and the result stored > in r31) has been re-tasked as a speculation barrier. The instruction > only acts as a barrier on newer machines with appropriate

Re: [1/6] powerpc/64s: Add barrier_nospec

2018-06-04 Thread Michael Ellerman
On Tue, 2018-04-24 at 04:15:54 UTC, Michael Ellerman wrote: > From: Michal Suchanek > > A no-op form of ori (or immediate of 0 into r31 and the result stored > in r31) has been re-tasked as a speculation barrier. The instruction > only acts as a barrier on newer machines with appropriate

Re: [PATCH 1/6] powerpc/64s: Add barrier_nospec

2018-04-23 Thread Nicholas Piggin
On Tue, 24 Apr 2018 14:15:54 +1000 Michael Ellerman wrote: > From: Michal Suchanek > > A no-op form of ori (or immediate of 0 into r31 and the result stored > in r31) has been re-tasked as a speculation barrier. The instruction > only acts as a barrier

Re: [PATCH 1/6] powerpc/64s: Add barrier_nospec

2018-04-23 Thread Nicholas Piggin
On Tue, 24 Apr 2018 14:15:54 +1000 Michael Ellerman wrote: > From: Michal Suchanek > > A no-op form of ori (or immediate of 0 into r31 and the result stored > in r31) has been re-tasked as a speculation barrier. The instruction > only acts as a barrier on newer machines with appropriate

[PATCH 1/6] powerpc/64s: Add barrier_nospec

2018-04-23 Thread Michael Ellerman
From: Michal Suchanek A no-op form of ori (or immediate of 0 into r31 and the result stored in r31) has been re-tasked as a speculation barrier. The instruction only acts as a barrier on newer machines with appropriate firmware support. On older CPUs it remains a harmless

[PATCH 1/6] powerpc/64s: Add barrier_nospec

2018-04-23 Thread Michael Ellerman
From: Michal Suchanek A no-op form of ori (or immediate of 0 into r31 and the result stored in r31) has been re-tasked as a speculation barrier. The instruction only acts as a barrier on newer machines with appropriate firmware support. On older CPUs it remains a harmless no-op. Implement