Re: [PATCH 2/2] amd/iommu: Use raw locks on atomic context paths

2018-02-13 Thread Joerg Roedel
On Sun, Jan 21, 2018 at 03:28:54AM -0600, Scott Wood wrote: > Several functions in this driver are called from atomic context, > and thus raw locks must be used in order to be safe on PREEMPT_RT. > > This includes paths that must wait for command completion, which is > a potential PREEMPT_RT

[PATCH 2/2] amd/iommu: Use raw locks on atomic context paths

2018-01-21 Thread Scott Wood
Several functions in this driver are called from atomic context, and thus raw locks must be used in order to be safe on PREEMPT_RT. This includes paths that must wait for command completion, which is a potential PREEMPT_RT latency concern but not easily avoidable. Signed-off-by: Scott Wood