Lines Matching +full:per +full:- +full:processor

2 OpenCAPI (Open Coherent Accelerator Processor Interface)
6 at being low-latency and high-bandwidth. The specification is
14 OpenCAPI is known in linux as 'ocxl', as the open, processor-agnostic
20 High-level view
24 be implemented on top of a physical link. Any processor or device
29   +-----------+                         +-------------+
32   | Processor |                         |  Function   |
33   |           |  +--------+             |    Unit     |  +--------+
34   |           |--| Memory |             |    (AFU)    |--| Memory |
35   |           |  +--------+             |             |  +--------+
36   +-----------+                         +-------------+
38   +-----------+                         +-------------+
40   +-----------+                         +-------------+
42   +-----------+                         +-------------+
44   +-----------+                         +-------------+
47        +---------------------------------------+
54 OpenCAPI relies on a PCI-like configuration space, implemented on the
76 * a per-process MMIO area, which has a fixed size for each context.
85 Transaction Layer, specifying a 64-bit object handle which defines the
89 64-bit object handle, that can be passed to the AFU.
96 The driver creates one char device per AFU found on the physical
104 where <AFU name> is a max 20-character long name, as found in the
118 Documentation/ABI/testing/sysfs-class-ocxl
126 ----
135 -----
173 ----
175 A process can mmap the per-process MMIO area for interactions with the