Issue:
Observed HSI diagnostic timer execution is taking
more than MAX_PROC_TIME due to OSI_CMD_READ_MMC
execution time. As this is reading 151 MMC
registers for MGBE and 91 registers for EQOS.
Fix:
As this OSI_CMD_READ_MMC command execution is not
meeting the MAX_PROC_TIME, add new OSI_CMD_READ_HSI_ERR
command only to read 5 MMC error registers.
Bug 4069585
(cherry picked from commit 163e9cf67c)
Change-Id: I38a10b7f09ac7614d548b5caa4203f1c94889908
Signed-off-by: Mark Mendez <mmendez@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/2905534
Reviewed-by: Vipin Kumar <vipink@nvidia.com>