> -----Original Message----- > From: Jason Gunthorpe <jgg@xxxxxxxx> > Sent: Tuesday, June 30, 2020 12:17 PM > To: Xiong, Jianxin <jianxin.xiong@xxxxxxxxx> > Cc: linux-rdma@xxxxxxxxxxxxxxx; Doug Ledford <dledford@xxxxxxxxxx>; Sumit Semwal <sumit.semwal@xxxxxxxxxx>; Leon Romanovsky > <leon@xxxxxxxxxx>; Vetter, Daniel <daniel.vetter@xxxxxxxxx>; Christian Koenig <christian.koenig@xxxxxxx>; dri- > devel@xxxxxxxxxxxxxxxxxxxxx > Subject: Re: [RFC PATCH v2 0/3] RDMA: add dma-buf support > > > > > > > On Tue, Jun 30, 2020 at 05:21:33PM +0000, Xiong, Jianxin wrote: > > > > > > Heterogeneous Memory Management (HMM) utilizes > > > > > > mmu_interval_notifier and ZONE_DEVICE to support shared > > > > > > virtual address space and page migration between system memory > > > > > > and device memory. HMM doesn't support pinning device memory > > > > > > because pages located on device must be able to migrate to > > > > > > system memory when accessed by CPU. Peer-to-peer access is > > > > > > possible if the peer can handle page fault. For RDMA, that means the NIC must support on-demand paging. > > > > > > > > > > peer-peer access is currently not possible with hmm_range_fault(). > > > > > > > > Currently hmm_range_fault() always sets the cpu access flag and > > > > device private pages are migrated to the system RAM in the fault handler. > > > > However, it's possible to have a modified code flow to keep the > > > > device private page info for use with peer to peer access. > > > > > > Sort of, but only within the same device, RDMA or anything else generic can't reach inside a DEVICE_PRIVATE and extract anything > useful. > > > > But pfn is supposed to be all that is needed. > > Needed for what? The PFN of the DEVICE_PRIVATE pages is useless for anything. Hmm. I thought the pfn corresponds to the address in the BAR range. I could be wrong here. > > > > Well, what do you want to happen here? The RDMA parts are > > > reasonable, but I don't want to add new functionality without a > > > purpose - the other parts need to be settled out first. > > > > At the RDMA side, we mainly want to check if the changes are > > acceptable. For example, the part about adding 'fd' to the device ops > > and the ioctl interface. All the previous comments are very helpful > > for us to refine the patch so that we can be ready when GPU side > > support becomes available. > > Well, I'm not totally happy with the way the umem and the fd is handled so roughly and incompletely.. Yes, this feedback is very helpful. Will work on improving the code. > > > > Hum. This is not actually so hard to do. The whole dma buf proposal > > > would make a lot more sense if the 'dma buf MR' had to be the > > > dynamic kind and the driver had to provide the faulting. It would > > > not be so hard to change mlx5 to be able to work like this, perhaps. > > > (the locking might be a bit tricky though) > > > > The main issue is that not all NICs support ODP. > > Sure, but there is lots of infrastructure work here to be done on dma buf, having a correct consumer in the form of ODP might be helpful to > advance it. Good point. Thanks. > > Jason
