51606ed402
Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
46 lines
983 B
Plaintext
46 lines
983 B
Plaintext
[Passthru]
|
|
# PT <bdev name> <vbdev name>
|
|
PT Malloc3 TestPT
|
|
|
|
[crypto]
|
|
# CRY <bdev name> <vbdev name> <key> <PMD>
|
|
# key size depends on cipher
|
|
# supported PMD names: crypto_aesni_mb, crypto_qat
|
|
# Note: QAT is experimental while test HW is being setup
|
|
CRY Malloc6 crypto_ram 0123456789123456 crypto_aesni_mb
|
|
|
|
[Malloc]
|
|
NumberOfLuns 7
|
|
LunSizeInMB 64
|
|
|
|
[Split]
|
|
# Split Malloc1 into two auto-sized halves
|
|
Split Malloc1 2
|
|
|
|
# Split Malloc2 into eight 4-megabyte pieces,
|
|
# leaving the rest of the device inaccessible
|
|
Split Malloc2 8 4
|
|
|
|
[AIO]
|
|
AIO /dev/ram0 AIO0
|
|
AIO /tmp/aiofile AIO1 2048
|
|
|
|
[QoS]
|
|
# QoS section defines limitation on performance
|
|
# metric like IOPS
|
|
#
|
|
# Format: Limit_IOPS Bdev_Name IOPS_Limit_Value
|
|
#
|
|
# IOPS limit must be 10000 or greater and be multiple
|
|
# of 10000
|
|
#
|
|
# Assign 20000 IOPS for the Malloc0 block device
|
|
Limit_IOPS Malloc0 20000
|
|
|
|
[RAID0]
|
|
Name raid0
|
|
StripSize 64
|
|
NumDevices 2
|
|
RaidLevel 0
|
|
Devices Malloc4 Malloc5
|