Messages in this thread | | | Date | Thu, 1 Sep 2022 16:03:36 +0100 | Subject | Re: [PATCH v4 1/2] iommu/s390: Fix race with release_device ops | From | Robin Murphy <> |
| |
On 2022-09-01 15:34, Jason Gunthorpe wrote: > On Thu, Sep 01, 2022 at 03:29:16PM +0100, Robin Murphy wrote: > >> Right, the next step would be to bridge that gap to iommu-dma to dump the >> flush queue when IOVA allocation failure implies we've reached the >> "rollover" point, and perhaps not use the timer at all. By that point a >> dedicated domain type, or at least some definite internal flag, for this >> alternate behaviour seems like the logical way to go. > > At least on this direction, I've been thinking it would be nice to > replace the domain type _FQ with a flag inside the domain, maybe the > ops, saying how the domain wants the common DMA API to operate. If it > wants FQ mode or other tuning parameters
Compare the not-necessarily-obvious matrix of "strict" and "passthrough" command-line parameters with the nice understandable kconfig and sysfs controls for a reminder of why I moved things *from* that paradigm in the first place ;)
This idea still fits perfectly into the the "continuum of strictness" notion underlying that domain type rework, since it potentially leaves a lot more address space mapped for a much longer time than our current FQ implementation. I would agree that exposing FQ tuneables in their own right may well have some potential value, much like John's equivalent idea for the IOVA cache layer, but I for one have no desire to bring back DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE, much less any further mess of disjoint properties at that level.
Thanks, Robin.
| |