Commit afbab54c authored by paul luse's avatar paul luse Committed by Jim Harris
Browse files

doc/accel_fw: update docs to reflect updated configuration 



Previous patch changed config #0.  Also clarrified config #1 as
well.  A future patch will expose the priority setting mentioned in
config #1 as this is currently internal.

Signed-off-by: default avatarpaul luse <paul.e.luse@intel.com>
Change-Id: I4761324e522b24f580918a8397810b52dd1a6bd3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6008


Tested-by: default avatarSPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: default avatarShuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: default avatarJim Harris <james.r.harris@intel.com>
parent e4070ee0

doc/accel_fw.md

+8 −8
Original line number Diff line number Diff line
@@ -66,14 +66,14 @@ To use the IOAT engine, use the RPC [`ioat_scan_accel_engine`](https://spdk.io/d 
To use the DSA engine, use the RPC [`idxd_scan_accel_engine`](https://spdk.io/doc/jsonrpc.html) with an optional parameter of `-c` and provide a configuration number of either 0 or 1. These pre-defined configurations determine how the DSA engine will be setup in terms

of work queues and engines.  The DSA engine is very flexible allowing for various configurations of these elements to either account for different quality of service requirements or to isolate hardware paths where the back end media is of varying latency (i.e. persistent memory vs DRAM).  The pre-defined configurations are as follows:



0: Four separate work queues each backed with one DSA engine.  This is a generic

configuration that provides 4 portals to submit operations to each with a

single engine behind it providing some level of isolation as operations are

submitted round-robin.

0: A single work queue backed with four DSA engines.  This is a generic configuration

that enables the hardware to best determine which engine to use as it pulls in new

operations.



1: Two separate work queues each backed with two DSA engines. This is another

generic configuration that provides 2 portals to submit operations to and

lets the DSA hardware decide which engine to select based on loading.

generic configuration that is documented in the specification and allows the

application to partition submissions across two work queues. This would be useful

when different priorities might be desired per group.



There are several other configurations that are possible that include quality

of service parameters on the work queues that are not currently utilized by