Commit 9a9cde60 authored by paul luse's avatar paul luse Committed by Jim Harris
Browse files

docs: remove accel_fw doc page on batching 



Batch support has been removed from both the accel_fw and idxd
public API so remove the supporting docs.

Signed-off-by: default avatarpaul luse <paul.e.luse@intel.com>
Change-Id: I98d1ab7a80942ef69a824482beb76cf921bc3d7f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11379


Tested-by: default avatarSPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: default avatarBen Walker <benjamin.walker@intel.com>
Reviewed-by: default avatarJim Harris <james.r.harris@intel.com>
parent 847139f6

doc/accel_fw.md

+0 −19
Original line number Diff line number Diff line
@@ -141,22 +141,3 @@ The software module is enabled by default. If no hardware engine is explicitly 
enabled via startup RPC as discussed earlier, the software module will use ISA-L

if available for functions such as CRC32C. Otherwise, standard glibc calls are

used to back the framework API.
 


### Batching {#batching}



Batching is exposed by the acceleration framework and provides an interface to

batch sets of commands up and then submit them with a single command.  The public

API is consistent with the implementation however each plug-in module behaves

differently depending on its capabilities.



The DSA engine has complete support for batching all supported commands together

into one submission. This is advantageous as it reduces the overhead incurred in

the submission process to the hardware.



The software engine supports batching only to be consistent with the framework API.

In software there is no savings by batching sets of commands versus submitting them

individually.



The IOAT engine supports batching but it is only beneficial for `memmove` and `memfill`

as these are supported by the hardware.  All other commands can be batched and the

framework will manage all other commands via software.