Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1530518

Re: Enabling peer to peer device transactions for PCIe devices

From Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Newsgroups linux.kernel
Subject Re: Enabling peer to peer device transactions for PCIe devices
Date 2016-11-25 21:30 +0100
Message-ID <sHvi9-5he-1@gated-at.bofh.it> (permalink)
References (5 earlier) <sH5nH-5hZ-7@gated-at.bofh.it> <sHuvL-4HW-7@gated-at.bofh.it> <sHuvL-4HW-9@gated-at.bofh.it> <sHuvL-4HW-5@gated-at.bofh.it> <sHvi9-5he-3@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Fri, Nov 25, 2016 at 02:49:50PM -0500, Serguei Sagalovitch wrote:

> GPU could perfectly access all VRAM.  It is only issue for p2p without
> special interconnect and CPU access. Strictly speaking as long as we
> have "bus address"  we could have RDMA but  I agreed that for
> RDMA we could/should(?) always "request"  CPU address (I hope that we
> could forget about 32-bit application :-)).

At least on x86 if you have a bus address you have a CPU address. All
RDMAable VRAM has to be visible in the BAR.

> BTW/FYI: About CPU access: Some user-level API is mainly handle based
> so there is no need for CPU access by default.

You mean no need for the memory to be virtually mapped into the
process?

Do you expect to RDMA from this kind of API? How will that work?

Jason

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Re: Enabling peer to peer device transactions for PCIe devices Jason Gunthorpe <jgunthorpe@obsidianresearch.com> - 2016-11-25 20:40 +0100
  Re: Enabling peer to peer device transactions for PCIe devices Jason Gunthorpe <jgunthorpe@obsidianresearch.com> - 2016-11-25 21:30 +0100
  Re: Enabling peer to peer device transactions for PCIe devices Alex Deucher <alexdeucher@gmail.com> - 2016-11-26 00:50 +0100

csiph-web