This patch correctly assigns the self_id and proc_id values for MPU and Ducati.
Signed-off-by: Jayan John <x00jayan@ti.com>

This patch resolves the issue in the missing padding bits in the Notify shared memory structures.
Signed-off-by: Jayan John <x00jayan@ti.com>

This patch resolves the issue of multiple ISR
from the MPU side being fired on receiving a
mailbox interrupt from Ducati Side.
Signed-off-by: C A Subramaniam <subramaniam.ca@ti.com>

This patch removes white spaces in function headers
of the notify ducati module
Signed-off-by: C A Subramaniam <subramaniam.ca@ti.com>

This patch moves all the notify shm data
structures from the notify header to the
notify shm header.Also, shm structure header
names changed.
Signed-off-by: C A Subramaniam <subramaniam.ca@ti.com>

This patch fixes the bugs found during heapbuf sample validation.
The following are the main changes.
	- object creation attributes are assigned conditionally
		based on the create flag passed in _heapbuf_create
	- object lock has been commented out.
	- gt_traces are removed.

These fixes does not resolve the issues seen with heapbuf_destroy
in the sample.
Signed-off-by: Arun M G <arunmg@ti.com>

This patch replaces the linux mutex with gatepeterson for shared memory
protection in all the listmp_sharedmemory functions that deal with
listmp elements.
Signed-off-by: Suman Anna <s-anna@ti.com>

listmp_params has an extra additional member heap_handle, which is not
defined in the user-space. This has been removed.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes the issues within listmp_sharedmemory module that are
causing the heapbuf_open and heapbuf_create to fail. listmp_sharedmemory module
uses the gate passed onto it from heapbuf. The module currently uses the linux
mutex apis with this gate, whereas the actual gate passed is a gatepeterson object.
The linux mutex apis are replaced with gatepeterson apis in
listmp_sharedmemort_put_tail function.
Signed-off-by: Jayan John <x00jayan@ti.com>

This patch fixes for the listmp element pointer corruption in
listmp_sharedmemory_destroy function. The corruption is caused
by usage of the linux kernel list macros within the
listmp_sharedmemory_delete function.

The patch also assigns SysM3 as the default remote owner processor for
lists being opened.
Signed-off-by: Simon Que <sque@ti.com>

This patch fixes the following bugs found during ListMP sample application
testing.
	- listmp_sharedmemory_ioctl_open has an out-of-place
		listmp_sharedmemory_create function call. This is removed.
	- cargs->api_status is not returning a failure value in case of
		premature exit for listmp_sharedmemory_get_head,
		listmp_sharedmemory_get_tail, listmp_sharedmemory_put_head,
		listmp_sharedmemory_put_tail functions.

There are other places where cargs->api_status needs to be fixed and this will
be done in a subsequent patch.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch provides fixes for minor bugs in gatepeterson driver
and sharedregion modules. gatepeterson driver has an incorrect
parameter check. Error traces are slightly adjusted in the
sharedregion module. Double semi-colons are corrected.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch adds notify dependancy in ipc module
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch includes couple of minor bug fixes in listmp_sharedmemory module.
NULL config parameter is supported in the listmp_sharedmemory_setup function,
and couple of additional error checks have been added.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes coding style issues.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes coding style issues.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes coding style issues.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes coding style issues
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This Patch fixes coding format in notify
driver.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes the issues with the status returned from kernel
for gatepeterson_open. The following are the changes.
1. gatepeterson_open - changed the ENOENT to ENXIO, as ENOENT might
   be returned from nameserver and will translate to a different error
   code in user-space
2. gatepeterson_ioctl_open - returning the status of gatepeterson_open
   to the user-level in api_status.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes coding style issues in
proc4430 module.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes the DMM for user space buffer.
The Ducati virtual address that was mapped was
initialized to 0 instead of using the one that is
passed to the map function. Also, this includes
the fix for VM_IO part where teh va was incremented
twice.
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>

Signed-off-by:

The physical address of a memory entry is stored into the user
virtual address entry in proc4430_attach so that the user-side
can remap using this address in ProcMgrDrv_Open. The current mapping
uses the Slave address and assumes that the Slave address is same as
the physical address.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes the default element size for Ducati
virtual address mapping. It is changed from 16bit to
32bits. Also, temporary changed the start address of
Ducati VA to 0xa0000000 from 0x80000000
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>

Added the MODULE_EXPORT for the multiproc functions.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch handles the following
	- Adds checks for null pointers in proc attach
	- fixes the proc4430_proc_info to return correct
	  mem entries
	- fixes compile warnings in previous patch
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>

This patch fixes a missing copy_from_user for the embedded pointer for
proc4430_mem_entry passed in the params object of the proc4430_cmd_args_create.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes the issues with messageq_setup and messageq_transportshm_setup.
The issue is caused by incorrect BUG_ON and unlikely statements.
Signed-off-by: Suman Anna <s-anna@ti.com>

This patch fixes coding style issues and compile time
warnings.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch removes unused file
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes coding style issues and compile time
warnings.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes coding style issues and compile
time warnings.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

In proc manager open, the address translation (ducti -> mpu kernel,
ducati -> mpu user) were not populated from proc 4430 get info
function.
In this fix the adress translation tables are populated directly
from the proc instance handle memory entry tables in proc get info
function, rather than going in to proc4430 get info function for
the same. This fix is based on the approach followed in syslink code.
Signed-off-by: Arun M G <arunmg@ti.com>

This patch adds proc4430 Attach and Detach functionality.
The Attach functions does ioremap of the Ducati physical
address, sets the slave virtual address same as the
physical address. This values are returned to the caller
in the params parameter passed to the Attach.
The Detach function does the reverse of Attach, where it
does the iounmap of the Master Virtual address.
Also, fixed the compile warning in procmgr.
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>

This patch fixes the issue with getting mem_entries for
a proc instance.
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>

This will remove multiple header inclusion in some of the
ioctl files of ipc
Signed-off-by: Arun M G <arunmg@ti.com>

Some  variables that makes a osal feel were renamed
Signed-off-by: Arun M G <arunmg@ti.com>

This patch fizes checkpatch errors in
ducati driver files.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch fixes check patch errors in
dispatcher driver files.
Signed-off-by: Ramesh Gupta G <grgupta@ti.com>

This patch handles the following.
   - Calling the proc4430 setup multiple twice.
   - Fixes the bugs in proc4430 create function that were
     causing segmentation fault.
   - Handled wrong check for return value of kmalloc.
Signed-off-by: Hari Kanigeri <h-kanigeri2@ti.com>
Signed-off-by: Arun Radhakrishnan <x0051460@ti.com>