Re: [PATCH 0/3] mmu_notifier: Allow to manage CPU external TLBs

2014-07-25 Thread Joerg Roedel
Hi Andrew, On Thu, Jul 24, 2014 at 04:33:03PM -0700, Andrew Morton wrote: > On Thu, 24 Jul 2014 16:35:38 +0200 Joerg Roedel wrote: > > > > Any comments and review appreciated! > > It looks pretty simple and harmless. > > I assume the AMD IOMMUv2 driver actually uses this and it's all > tested

Re: [PATCH 0/3] mmu_notifier: Allow to manage CPU external TLBs

2014-07-24 Thread Sander, Ben
Yes. AMD has tested this with the iommuv2 driver and verified it works correctly. There is a corresponding change in the iommuv2 driver to use the new API. > On Jul 24, 2014, at 6:33 PM, "Andrew Morton" > wrote: > >> On Thu, 24 Jul 2014 16:35:38 +0200 Joerg Roedel wrote: >> >> here is a p

Re: [PATCH 0/3] mmu_notifier: Allow to manage CPU external TLBs

2014-07-24 Thread Andrew Morton
On Thu, 24 Jul 2014 16:35:38 +0200 Joerg Roedel wrote: > here is a patch-set to extend the mmu_notifiers in the Linux > kernel to allow managing CPU external TLBs. Those TLBs may > be implemented in IOMMUs or any other external device, e.g. > ATS/PRI capable PCI devices. > > The problem with man

Re: [PATCH 0/3] mmu_notifier: Allow to manage CPU external TLBs

2014-07-24 Thread Andrea Arcangeli
On Thu, Jul 24, 2014 at 04:35:38PM +0200, Joerg Roedel wrote: > To solve this situation I wrote a patch-set to introduce a > new notifier call-back: mmu_notifer_invalidate_range(). This > notifier lifts the strict requirements that no new > references are taken in the range between _start() and > _

[PATCH 0/3] mmu_notifier: Allow to manage CPU external TLBs

2014-07-24 Thread Joerg Roedel
Hi, here is a patch-set to extend the mmu_notifiers in the Linux kernel to allow managing CPU external TLBs. Those TLBs may be implemented in IOMMUs or any other external device, e.g. ATS/PRI capable PCI devices. The problem with managing these TLBs are the semantics of the invalidate_range_start