Re: [PATCH v8 13/14] IB/core: use MAP_DIRECT to fix / enable RDMA to DAX mappings

2017-10-11 Thread Dan Williams
On Wed, Oct 11, 2017 at 4:54 AM, Joerg Roedel wrote: > On Tue, Oct 10, 2017 at 07:50:12AM -0700, Dan Williams wrote: >> +static void ib_umem_lease_break(void *__umem) >> +{ >> + struct ib_umem *umem = umem; >> + struct ib_device *idev = umem->context->device; >> +

Re: [PATCH v8 13/14] IB/core: use MAP_DIRECT to fix / enable RDMA to DAX mappings

2017-10-11 Thread Joerg Roedel
On Tue, Oct 10, 2017 at 07:50:12AM -0700, Dan Williams wrote: > +static void ib_umem_lease_break(void *__umem) > +{ > + struct ib_umem *umem = umem; > + struct ib_device *idev = umem->context->device; > + struct device *dev = idev->dma_device; > + struct scatterlist *sgl =

[PATCH v8 13/14] IB/core: use MAP_DIRECT to fix / enable RDMA to DAX mappings

2017-10-10 Thread Dan Williams
Currently the ibverbs core in the kernel is completely unaware of the dangers of filesystem-DAX mappings. Specifically, the filesystem is free to move file blocks at will. In the case of DAX, it means that RDMA to a given file offset can dynamically switch to another file offset, another file, or