compress/qat: support IM buffer too small operation
authorAdam Dybkowski <adamx.dybkowski@intel.com>
Fri, 17 Apr 2020 15:44:03 +0000 (17:44 +0200)
committerAkhil Goyal <akhil.goyal@nxp.com>
Sun, 19 Apr 2020 15:15:14 +0000 (17:15 +0200)
commitc13cecf60f128cc6ae89557d9799f4f92d785cdc
tree74fbf9f582b4aa7ec22828ef90e5c69a8ce7d085
parentf5ad5016869b06aa6e03dc1e38dcf8e4d1c733bf
compress/qat: support IM buffer too small operation

This patch implements a special way of buffer handling when internal
QAT IM buffer is too small for Huffman dynamic compression operation.
Instead of falling back to fixed compression, the operation is now
split into multiple smaller dynamic compression requests (possible to
execute on QAT) and their results are then combined and copied into
the output buffer. This is not possible if any checksum calculation
was requested - in such case the code falls back to fixed compression
as before.

Signed-off-by: Adam Dybkowski <adamx.dybkowski@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
doc/guides/compressdevs/qat_comp.rst
doc/guides/cryptodevs/qat.rst
doc/guides/rel_notes/release_20_05.rst
drivers/common/qat/qat_qp.c
drivers/common/qat/qat_qp.h
drivers/compress/qat/qat_comp.c
drivers/compress/qat/qat_comp.h
drivers/compress/qat/qat_comp_pmd.c