RAS are three aspects of the dependability of a system:
Reliability: continuity of correct service
Availability: readiness for correct service
Serviceability: ability to undergo modifications and repairs
ARMv8.2 provides RAS extensions to achieve the above features in a system.
arm64_ras driver allows you to handle Correctable errors using per core
Fault Handling Interrupt (or FHI). Un-correactable errors are handled by
raising SError exception to kernel or Trusted Firmware(EL3) on CCLEX.
The driver provides an API for CPU specific RAS drivers to
register callbacks in case of FHI. When FHI occurs, the FHI
ISR goes through the list of registered callbacks and executes
them. Similarly, the driver allows registering SError callbacks
for Un-correctable errors within Core, Core-Cluster & CCPLEX.
When SError is reported, then those callbacks are scanned to
find and report error in a Core, Cluster or CCPLEX
Bug 3625675
Change-Id: I6dba7178fa7bbcf55bfa083be1b077874f4865c9
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2705396
GVS: Gerrit_Virtual_Submit
Actmon is a hardware block that can be used to track the activity of
certain hardware units. It can boost EMC clock depending
on the memory traffic among various client. It is called central actmon
as it monitors central activity for example MC activity.
Add central activity monitor driver.
Bug 3625675
Change-Id: I1a5918e7d84bc247f694f53f965c28888b773c91
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2705395
GVS: Gerrit_Virtual_Submit
Add header file for tegra-prod APIs so that driver
can use the prod APIs. The implementation is kept as
dummy i.e. not supported.
Bug 3621816
Change-Id: I0aeada74eace74770bc3c5d2964b57ba9ba8f1c5
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2704358
GVS: Gerrit_Virtual_Submit
