| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
Fixes openbmc/openbmc#2118
Change-Id: If9e2610fe7443daa2196b0e5989f81bc544266b2
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, only thing that OCC controller does is create PassThrough
objects. However, there is a need now to create OCC Status objects
and hence some restructuring is needed to consume that.
Since OCC control now is doing more than one thing, service name is
changed to map to that.
Change-Id: I466979a873d6f14385eb59d0e9d9f3a8b3f95a9b
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
|
|
|
|
|
|
|
|
| |
Create OCC pass-through d-bus objects when corresponding CPU objects are
created in the inventory.
Resolves openbmc/openbmc#1450.
Change-Id: I8da879f51ebef8dcc3d25358def81c5e0dce0617
Signed-off-by: Deepak Kodihalli <dkodihal@in.ibm.com>
|
|
|
|
|
|
|
|
|
|
| |
Initially, the occ fifo devices were named as /dev/occfifo1..N
but it has been renamed to /dev/occ1..N.
Fixes openbmc/openbmc#1696
Change-Id: Ibf37cc3f61a85a46accd678a55e1b2b6019bd6e5
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
| |
This application interacts with OCC driver and hence its good to
terminate when there is an error during IO since it would mostly
mean a faulty hardware.
Fixes openbmc/openbmc#1428
Change-Id: I48bc7b2cf19922a7a53dbab78cdd4f2338a7431b
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
|
|
|
| |
Change-Id: I8159f71ff3dd087f10cdf3b013e1acb60d4758fb
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Communicating to OCC from BMC in PassThrough mode involves 2 drivers,
namely;
1) SBE FIFO driver --> Used to communicate with SBE
2) OCC driver --> Which uses FIFO driver to write data to OCC
at a predefined address and reads it back from a predefined
OCC address.
Device entry that is of interest here belongs to #2 above.
This will map occ0 to fifo1, occ2 to fifo2 and so on.
Until the udev rules are in place to make sure that a particular
entry always maps to correct device not caring the probing order,
this change here is a best guess that cpu0 and cpu1 would mostly in
all cases be probed in that order.
Change-Id: I485e0ea4eab1b4c6f59ee4b68c0725f9633cd933
Signed-off-by: Vishwanatha Subbanna <vishwa@linux.vnet.ibm.com>
|
|
This d-bus interface is implemented by the
open_power::occ::pass_through::PassThrough class.
Change-Id: I6bce9e609b7b977418bcfee15e97432cb7d9e6b3
Signed-off-by: Deepak Kodihalli <dkodihal@in.ibm.com>
|