ipc/ipcdev.git
6 years agoAdd autoconf generated file
vwan@ti.com [Tue, 17 Feb 2015 21:56:57 +0000 (13:56 -0800)]
Add autoconf generated file

Signed-off-by: VW <vwan@ti.com>
6 years agoSimplify configuration of maximum name length in NameServer
vwan@ti.com [Mon, 16 Feb 2015 22:45:58 +0000 (14:45 -0800)]
Simplify configuration of maximum name length in NameServer

This commit cleans up how the maximum name length is configured in NameServer.
Now the user only needs to configure it during instance creation, and to
configure it at the remote level in
packages/ti/ipc/namesrv/_NameServerRemoteRpmsg.h.

When a name passed to the NameServer remote exceeds the length it can support,
it now returns NameServer_E_NAMETOOLONG.

The NameServerApp is updated to test for long names that exceed the
maximum length.

This commit addresses SDOCM00111686.

Signed-off-by: VW <vwan@ti.com>
6 years agoSDOCM00103540 Incorrect number of hardware spin locks on DRA7XX
Ramsey Harris [Tue, 17 Feb 2015 21:38:35 +0000 (13:38 -0800)]
SDOCM00103540 Incorrect number of hardware spin locks on DRA7XX

Set the number of hardware spin locks to 256 in spinlockDelegate.

6 years agoSet default reserved messages queues to zero
Ramsey Harris [Tue, 17 Feb 2015 00:11:28 +0000 (16:11 -0800)]
Set default reserved messages queues to zero

The default value for reserved message queues in the LAD
configuration should be zero. I accidentally checked-in
my test copy. This commit corrects that mistake.

6 years agoSpecify Address Family for RPMessage in products.mak
Ramsey Harris [Mon, 16 Feb 2015 23:44:55 +0000 (15:44 -0800)]
Specify Address Family for RPMessage in products.mak

We use the Linux kernel version to make a guess at the value
of AF_RPMSG. However, depending on the kernel configuration,
this guess could be wrong. In this case, you must specify the
AF_RPMSG value directly. This had been done in ipc-linux.mak,
but this was too obscure. So, I moved it to products.mak.

6 years agoAndroid: Update to LADs makefile after cfg file move
Arnie Reynoso [Fri, 13 Feb 2015 16:37:55 +0000 (08:37 -0800)]
Android: Update to LADs makefile after cfg file move

LAD config files were moved into a cfg folder. Updated the Android.mk
file to reflect the move and included the addition of a MessageQCfg.c
file.

6 years agoAdded comment to explain hazards of enabling debug trace 3.35.00.04_eng
Ramsey Harris [Fri, 13 Feb 2015 01:59:26 +0000 (17:59 -0800)]
Added comment to explain hazards of enabling debug trace

SocketFxns.c is used by both LAD and TransportRpmsg. When debug
trace is enabled in LAD, it can cause data corruption. I explain
this in the comment and provide advice.

6 years agoGenerated makefile to follow up the previous commit.
Ramsey Harris [Fri, 13 Feb 2015 01:52:08 +0000 (17:52 -0800)]
Generated makefile to follow up the previous commit.

6 years agoSDOCM00114730 Reserved message queues on Linux/Keystone II
Ramsey Harris [Fri, 13 Feb 2015 01:47:46 +0000 (17:47 -0800)]
SDOCM00114730 Reserved message queues on Linux/Keystone II

Added numReservedEntries to MessageQ_Config structure; defines
how many queues to reserve. Moved all LAD config files into cfg
folder to separate configuration data from implementation code.
Added MessageQCfg.c file so users could specify MessageQ config
(such as reserved queues). New concept of queue port number: this
is computed by adding an offset to the queue index. Needed to
avoid collisions with port numbers reserved by services (such as
RPMessage, NameServer). Queue port number is embedded in queueId
and used to address messages. New macro MessageQ_getQueueIndex to
extract queue index from QID. Implemented MessageQ_openQueueId on
Linux. Added port number management to ti.sdo.ipc.MessageQ module.

LAD daemon: removed slot zero reservation for NameServer; added
queue index logic to manage reserved queues; use port number in
queueId construction; reference new MessageQ_cfg object.

TransportRpmsg: replaced queue index references with queue port.

6 years agoFix build warnings from rpmsg-rpc-stress test in QNX
vwan@ti.com [Fri, 13 Feb 2015 00:02:26 +0000 (16:02 -0800)]
Fix build warnings from rpmsg-rpc-stress test in QNX

Signed-off-by: VW <vwan@ti.com>
6 years agoClean up exported contents in QNX
vwan@ti.com [Wed, 11 Feb 2015 01:44:00 +0000 (17:44 -0800)]
Clean up exported contents in QNX

This commit modifies 'make install' to export only tests that we run. It also
puts all tests into the 'tests' subdirectory and builds them all in debug and
release modes for consistency.

Signed-off-by: VW <vwan@ti.com>
6 years agoGenerated script and makefiles to follow up the previous commit.
Ramsey Harris [Tue, 10 Feb 2015 23:50:26 +0000 (15:50 -0800)]
Generated script and makefiles to follow up the previous commit.

6 years agoAdded AF_RPMSG to products.mak to improve ease-of-use
Ramsey Harris [Tue, 10 Feb 2015 23:42:27 +0000 (15:42 -0800)]
Added AF_RPMSG to products.mak to improve ease-of-use

When you don't have access to your Linux kernel source, or you
need to override the Address Family (AF) configuration, you can
now define the AF for RPMSG in your products.mak file. Relaxed
the requirement to specify KERNEL_INSTALL_DIR for TCI6638. Fixed
up the include path to put the kernel include folder ahead of our
IPC linux/include folder.

6 years agoEliminate build warnings in QNX build
vwan@ti.com [Thu, 5 Feb 2015 19:32:48 +0000 (11:32 -0800)]
Eliminate build warnings in QNX build

This commit eliminates all build warnings in the QNX IPC build process.
It addresses SDOCM00113501.

Signed-off-by: VW <vwan@ti.com>
6 years agoReduce QNX build output verbosity
vwan@ti.com [Tue, 27 Jan 2015 22:54:59 +0000 (14:54 -0800)]
Reduce QNX build output verbosity

This commit reduces the build output verbosity, and allows it to be configured
in the qnx/quiet.mk file.

This helps address SDOCM00113501.

Signed-off-by: VW <vwan@ti.com>
6 years agoRework MessageQ_put to prioritize transport if given
Ramsey Harris [Thu, 5 Feb 2015 22:46:20 +0000 (14:46 -0800)]
Rework MessageQ_put to prioritize transport if given

In MessageQ_put, use secondary transport (if given) regardless
of destination queue. For processor local deliver, verify that
destination queue is process local, otherwise give error. Changed
secondary transport array to base interface type. Added interface
casting as needed.

6 years agoFixed run-time configuration of cluster baseId 3.35.00.03_eng
Ramsey Harris [Wed, 4 Feb 2015 03:20:17 +0000 (19:20 -0800)]
Fixed run-time configuration of cluster baseId

New method to encapsulate the run-time configuration of IPC
cluster baseId: Ipc_clusterConfig. Call this method from a
startup first function after setting the cluster baseId. Set
Ipc.entry array with valid procId when cluster baseId is known,
otherwise set to INVALIDID. Set MultiProc.clusterProcList array
as above.

6 years agoNew method for run-time configuration of cluster baseId
Ramsey Harris [Wed, 4 Feb 2015 03:07:06 +0000 (19:07 -0800)]
New method for run-time configuration of cluster baseId

Added MultiProc_setBaseIdOfCluster(). Only supported on SYS/BIOS.
This method should be called from a startup first function.

6 years agoLinux: Added option to set base cluster id in LAD
Arnie Reynoso [Tue, 3 Feb 2015 19:40:20 +0000 (11:40 -0800)]
Linux: Added option to set base cluster id in LAD

An option (-b) has been added to LAD to set the cluster's base id when
launching LAD.

6 years agoSDOCM00114696 Add method to return address of shared region entry
Ramsey Harris [Tue, 3 Feb 2015 17:40:54 +0000 (09:40 -0800)]
SDOCM00114696 Add method to return address of shared region entry

Added method to return the address of the in-memory shared region
entry structure; SharedRegion_getEntryPtr. Use this method from
a pre-startup hook function to configure the entry.

6 years agoSDOCM00114392 Interrupt module on tci6638 does not support clusters
Ramsey Harris [Tue, 3 Feb 2015 16:29:37 +0000 (08:29 -0800)]
SDOCM00114392 Interrupt module on tci6638 does not support clusters

Added segmentSize to MultiProcSetup to define the number of
processors on the device. Needed for sending interrupts to
processors in a different cluster but still on the local device.
Added cluster awareness. Added cluster support to the tci6638
package. Cleaned up Log_print statements. Replaced interface
IInterrupt references with Interrupt module references.

6 years agoFixed QNX build error due to transport header file changes
Ramsey Harris [Tue, 3 Feb 2015 18:48:45 +0000 (10:48 -0800)]
Fixed QNX build error due to transport header file changes

Updated the location of the transport header files to match
what is expected by MessageQ.h.

6 years agoAdd IPC cluster support to socket utility functions
Ramsey Harris [Tue, 3 Feb 2015 01:50:29 +0000 (17:50 -0800)]
Add IPC cluster support to socket utility functions

The socket connect and bind functions need to map procId
to clusterId in order to correctly index into the rprocList[]
and assign the correct vproc_id.

6 years agoGenerated makefiles to follow up on the previous commit.
Ramsey Harris [Wed, 28 Jan 2015 17:31:03 +0000 (09:31 -0800)]
Generated makefiles to follow up on the previous commit.

6 years agoAdd missing transport header files to Linux install command
Ramsey Harris [Wed, 28 Jan 2015 17:14:27 +0000 (09:14 -0800)]
Add missing transport header files to Linux install command

Added new transport interface header files to top-level install
goal. Promoted Std.h to top-level install goal. Also added
TransportRpmsg.h because application needs to call the create
method. Moved location of interface files and transport files to
respective folders under ti/ipc for better name recognition.

6 years agoRemove references to SysLink in QNX code base
vwan@ti.com [Tue, 27 Jan 2015 00:06:37 +0000 (16:06 -0800)]
Remove references to SysLink in QNX code base

This commit removes all references to "SysLink" in the QNX IPC code base, with
the exception that it preserves the file and directory names. The decision was
made to preserve the path names to avoid breaking any existing apps that may
be including headers and/or libraries based on the existing directory
structure, and to ease git history lookup going forward.

Note that this does introduce a break in the naming of internal features such
as SYSLINK_DEBUG and SYSLINK_DEBUG_TRACE_LEVEL.

This helps address SDOCM00108142.

Signed-off-by: VW <vwan@ti.com>
6 years agoChange max threads to 25 in both MessageQMulti app and documentation in QNX
vwan@ti.com [Thu, 22 Jan 2015 01:09:19 +0000 (17:09 -0800)]
Change max threads to 25 in both MessageQMulti app and documentation in QNX

This helps address SDOCM00104833.

Signed-off-by: VW <vwan@ti.com>
6 years agodocs: Remove BIOS-only comment from GateMP
Chris Ring [Thu, 22 Jan 2015 00:26:58 +0000 (16:26 -0800)]
docs: Remove BIOS-only comment from GateMP

GateMP is now supported on HLOS's (for some devices), so remove the
errant note that it's only supported on SYS/BIOS.

Signed-off-by: Chris Ring <cring@ti.com>
6 years agoUpdate QNX IPC loader to RIDL 2.1.0
vwan@ti.com [Wed, 21 Jan 2015 19:37:27 +0000 (11:37 -0800)]
Update QNX IPC loader to RIDL 2.1.0

This commit updates the loader in QNX IPC to the latest codebase in RIDL 2.1.0,
which can be found at https://gforge.ti.com/gf/project/ridl_oss. It also brings
in any relevant differences between SysLink 2 and RIDL 2.0.

This addresses SDOCM00101007.

Signed-off-by: VW <vwan@ti.com>
6 years agoMove 'ret' declaration inside XDC %if to avoid 'unused' warning
Robert Tivy [Fri, 23 Jan 2015 21:40:22 +0000 (13:40 -0800)]
Move 'ret' declaration inside XDC %if to avoid 'unused' warning

6 years agoGenerated new autoreconf makefile 3.35.00.02_eng
Ramsey Harris [Wed, 21 Jan 2015 01:24:43 +0000 (17:24 -0800)]
Generated new autoreconf makefile

These are the new generated autoreconf makefiles from the previous
commit which modified two makefiles.

6 years agoAndroid: Added makefile to build transportrpmsg library
Arnie Reynoso [Wed, 21 Jan 2015 19:46:29 +0000 (11:46 -0800)]
Android: Added makefile to build transportrpmsg library

Added an Android.mak file to build newly separated transportRpmsg library.
Updated the corresponding host-side test examples to link with the library.

6 years agoAdded transport factory to remove circular dependency
Ramsey Harris [Wed, 21 Jan 2015 01:14:27 +0000 (17:14 -0800)]
Added transport factory to remove circular dependency

There was a circular dependency between the ipc and transport
libraries. To break this circle, moved the transport create
method into a factory. Ipc invokes the factory methods through
a function table configured with a new Ipc_transportConfig
method called by the application. Renamed transport library to
follow our naming conventions: libtitransportrpmsg.

6 years agoSDOCM00114391 IPC cluster support is missing on Linux
Ramsey Harris [Sat, 17 Jan 2015 00:58:24 +0000 (16:58 -0800)]
SDOCM00114391 IPC cluster support is missing on Linux

Added cluster support on IPC Linux implementation. Added two new
fields to the MultiProc_Config structure: numProcsInCluster, and
baseIdOfCluster. Updated all LAD config files with new fields.
Made changes to iterate over processors in cluster instead of
iterating over all processors in the system. Mapped procId to
clusterId as needed. Added LOG3 macro to LAD.

Added new cluster fields to MultiProc config struct

Linux and QNX share the same MultiProc API implementation
but different MultiProc configuration definitions. I needed
to add the new cluster fields to the QNX config struct. This
does not add cluster support to QNX.

6 years agoCode formatting. Broke long lines into short ones.
Ramsey Harris [Fri, 16 Jan 2015 01:48:12 +0000 (17:48 -0800)]
Code formatting. Broke long lines into short ones.

6 years agoSDOCM00114518 MessageQ_setMsgPri overwrites entire flags field
Ramsey Harris [Thu, 15 Jan 2015 22:52:01 +0000 (14:52 -0800)]
SDOCM00114518 MessageQ_setMsgPri overwrites entire flags field

The macro was simply assigning the new message priority to the
flags field in the message header. Fixed macro to read, clear,
then update only the priority bits of the flags field ([1:0]).

6 years agoRemoved registerTransportId from MessageQ.xdc file
Ramsey Harris [Thu, 15 Jan 2015 18:14:42 +0000 (10:14 -0800)]
Removed registerTransportId from MessageQ.xdc file

The new register/unregister transport ID functions were in the
MessageQ XDC spec file. This caused them to be defined using the
long name (i.e. package prefix). They are intended to be called
by clients using the short name (like MessageQ_put).

6 years agoSDOCM114476 Ipc_start is not thread safe (Linux)
Ramsey Harris [Wed, 14 Jan 2015 00:21:35 +0000 (16:21 -0800)]
SDOCM114476 Ipc_start is not thread safe (Linux)

Added a Posix mutex and a reference count to serialize the Ipc_start
and Ipc_stop methods. Added a module object to hold the mutex and
reference count.

6 years agoInstance converter methods need to be inline functions.
Ramsey Harris [Thu, 8 Jan 2015 17:06:56 +0000 (09:06 -0800)]
Instance converter methods need to be inline functions.

Added missing copyright notice. Added comments. The instance
converter methods are defined in the header file. Need to be
declared as inline methods to avoid multiple implementation
link error.

6 years agoFix GateMP to allow SR0's physical address to be different from its virtual
vwan@ti.com [Tue, 6 Jan 2015 18:18:47 +0000 (10:18 -0800)]
Fix GateMP to allow SR0's physical address to be different from its virtual

This commit fixes GateMP host-support (when host is running an HLOS) to allow
SR0's physical address to be a different value from its virtual address on the
slave core, by performing a virtual to physical address translation on the
addresses of the inUse arrays prior to stashing them into a NameServer entry
on the slave.

This addresses CQ SDOCM00113149.

Signed-off-by: VW <vwan@ti.com>
6 years agovring initialization in QNX is incorrectly clearing address 0
vwan@ti.com [Tue, 6 Jan 2015 00:48:56 +0000 (16:48 -0800)]
vring initialization in QNX is incorrectly clearing address 0

This commit fixes the vring initialization code to clear the vring as opposed
to physical address 0.

This addresses SDOCM00114409.

Signed-off-by: VW <vwan@ti.com>
6 years agoSDOCM00114390 Incorrect procId in NameServer_get used by lad daemon
Ramsey Harris [Tue, 6 Jan 2015 01:33:34 +0000 (17:33 -0800)]
SDOCM00114390 Incorrect procId in NameServer_get used by lad daemon

Fixed call to NameServer_getRemote; pass in procId array element
instead of loop index value.

6 years agoRename ipc-trace paths to /dev/ipc-trace/<core name> in QNX
vwan@ti.com [Sat, 20 Dec 2014 00:24:05 +0000 (16:24 -0800)]
Rename ipc-trace paths to /dev/ipc-trace/<core name> in QNX

Renaming the trace paths to match the convention adopted by /dev/ipc-file
and /dev/ipc-state, so that MultiProc id is not required to be known ahead
of time.

Note that this introduces a break in compatibility for anyone who's been
relying on the path name /dev/ipc-trace#.

This addresses SDOCM00113638.

Signed-off-by: VW <vwan@ti.com>
6 years agoMake MessageQCopy_send less noisy in QNX
vwan@ti.com [Sat, 15 Nov 2014 00:26:22 +0000 (16:26 -0800)]
Make MessageQCopy_send less noisy in QNX

This patch downgrades a common error message from MessageQCopy_send to an
info message of GT_2CLASS. The message was appearing even during normal
operation when a slave is not attached.

This addresses SDOCM00106179.

Signed-off-by: VW <vwan@ti.com>
6 years agoRemoved unused implementation of MessageQ_Params_init
Ramsey Harris [Wed, 10 Dec 2014 19:28:41 +0000 (11:28 -0800)]
Removed unused implementation of MessageQ_Params_init

The implementation of MessageQ_Params_init is now a static inline
in the header file. Need to remove this unused implementation in
MessageQ_daemon.c (fixed multiple definition compiler error).

6 years agoSDOCM00113609 Add version support for MessageQ_Params structure
Ramsey Harris [Thu, 4 Dec 2014 19:32:12 +0000 (11:32 -0800)]
SDOCM00113609 Add version support for MessageQ_Params structure

Added a version field to the params structure; used internally
to track the structure definition changes. MessageQ_Params_init
is now a static inline function which embeds the current version
in the call site. Added new MessageQ_Params_init__S function to
initialize the given params structure. Updated MessageQ_create
to be version aware. Fixed uninitialized params structure in
the MessageQ instance object.

6 years agoAdded function hook for the MessageQ_put method
Ramsey Harris [Sat, 6 Dec 2014 00:58:52 +0000 (16:58 -0800)]
Added function hook for the MessageQ_put method

New feature where you can register a function hook to be called
near the beginning of the MessageQ_put method. New type definition
for the hook function. Runtime API to register the hook function.
Also added a configuration parameter so you can register your hook
function in the application configuration script.

6 years agoSDOCM00114110 IPC custom build failure, part 2
Ramsey Harris [Tue, 2 Dec 2014 18:43:35 +0000 (10:43 -0800)]
SDOCM00114110 IPC custom build failure, part 2

Updates from code review comments. Build include path based
on package path, not from explicit repository names. This will
ensure correct include path as new repositories are added to
the package path.

6 years agoAdd MessageQ interface header files to QNX include area
Robert Tivy [Tue, 2 Dec 2014 00:36:09 +0000 (16:36 -0800)]
Add MessageQ interface header files to QNX include area

Even though QNX doesn't implement the new MessageQTransport interface,
it is including a common header (packages/ti/ipc/MessageQ.h) that does
reference the new interfaces.

6 years agoInitial MessageQ transport implementation 3.35.00.01_eng
Robert Tivy [Wed, 19 Nov 2014 22:16:31 +0000 (14:16 -0800)]
Initial MessageQ transport implementation

6 years agoMultiple transport interface support for ti.sdo.ipc.MessageQ
Ramsey Harris [Wed, 26 Nov 2014 23:52:42 +0000 (15:52 -0800)]
Multiple transport interface support for ti.sdo.ipc.MessageQ

Added new ITransport base class interface. New INetworkTransport
interface. MessageQ is now able to register multiple transport
instances. Dummy network transport to work around RTSC bug.

6 years agoSDOCM00114110 IPC custom build failure with XDCtools core product
Ramsey Harris [Wed, 26 Nov 2014 18:29:00 +0000 (10:29 -0800)]
SDOCM00114110 IPC custom build failure with XDCtools core product

When building with the XDCtools core product, the IPC custom build
was not finding the target header files. Fixed the makefile templates
to discover the repository for the targets and to add this to the
include path.

6 years agoCleanup, removed unused local array: sharedCreateId
Ramsey Harris [Wed, 26 Nov 2014 01:37:22 +0000 (17:37 -0800)]
Cleanup, removed unused local array: sharedCreateId

This array was allocated and initialized at config time but never
used. Furthermore, the array size was always zero.

6 years agoMove cluster processor list from config to module state
Ramsey Harris [Fri, 21 Nov 2014 22:13:16 +0000 (14:13 -0800)]
Move cluster processor list from config to module state

The list of processors in the cluster used to be defined at
configuration time. This prevents reusing a single executable
image on multiple processors (i.e. one image for all DSPs).
I moved this list into the MultiProc module state, so that it
could be defined in a reset hook when reusing a single image.

6 years agoAdded transportId bit-field to flags element in MessageQ header
Ramsey Harris [Wed, 19 Nov 2014 22:53:19 +0000 (14:53 -0800)]
Added transportId bit-field to flags element in MessageQ header

In the MessageQ header, added a transportId field to the flags
element. This field is 3-bits wide. Added macros to MessageQ.h
to set and get the transportId. The transportId is part of a
new feature for supporting multiple transports in MessageQ.

6 years agoSimplify output from GateMPApp
vwan@ti.com [Fri, 14 Nov 2014 22:33:52 +0000 (14:33 -0800)]
Simplify output from GateMPApp

This commit addresses SDOCM00113915.

Signed-off-by: VW <vwan@ti.com>
6 years agoMmRpc: update rpmsg header path for Android on K3.14
Sunita Nadampalli [Thu, 13 Nov 2014 21:41:05 +0000 (15:41 -0600)]
MmRpc: update rpmsg header path for Android on K3.14

Signed-off-by: Sunita Nadampalli <sunitan@ti.com>
6 years agoChange TMS320F2837.* to F2837.* in family Settings.xs 3.30.01.12 3.35.00.00_eng
Janet Head [Thu, 6 Nov 2014 22:01:24 +0000 (14:01 -0800)]
Change TMS320F2837.* to F2837.* in family Settings.xs

6 years agoSDOCM00113753 Notify mailbox driver is clearing wrong fifo
Ramsey Harris [Fri, 31 Oct 2014 23:48:26 +0000 (16:48 -0700)]
SDOCM00113753 Notify mailbox driver is clearing wrong fifo

Fixed the notify mailbox driver to clear the inbound fifo, not the
outbound fifo. In addition, write the End-Of-Interrupt (EOI) register
after clearing the fifo; this was not being done. Finally, after
fetching inbound messages (in the notify ISR), the EOI register was
not written correctly; now it writes the corresponding interrupt index.

6 years agoAdded TDA3XX to top-level product file.
Ramsey Harris [Thu, 30 Oct 2014 16:29:41 +0000 (09:29 -0700)]
Added TDA3XX to top-level product file.

Updated the platform comment to include TDA3XX. Improved comments
for MessageQ header; fixed typo in MessageQ flags bit description.

6 years agoSDOCM00113709 NotifyDriverMbx on Vayu should not use Ipc module
Ramsey Harris [Thu, 30 Oct 2014 21:34:09 +0000 (14:34 -0700)]
SDOCM00113709 NotifyDriverMbx on Vayu should not use Ipc module

Removed all references to ti.sdo.ipc.Ipc module in NotifyDriverMbx.
The Ipc and many other modules (13 in all) are no longer brought into
the executable when building an IPC Notify only system.

6 years agoAdd fw_rsc_intmem definition to support new TYPE_INTMEM entries in QNX
vwan@ti.com [Thu, 23 Oct 2014 22:43:28 +0000 (15:43 -0700)]
Add fw_rsc_intmem definition to support new TYPE_INTMEM entries in QNX

Previously, TYPE_INTMEM resource entries were represented as fw_rsc_carveout
structures internally in QNX. Given the update of the fw_rsc_intmem
structure in BIOS (SDOCM00113567), we now need to match them up by creating
a new fw_rsc_intmem structure.

This addresses SDOCM00113568.

Signed-off-by: VW <vwan@ti.com>
6 years agoLinux/MessageQ: Close the socket upon failure
Suman Anna [Mon, 27 Oct 2014 18:23:23 +0000 (13:23 -0500)]
Linux/MessageQ: Close the socket upon failure

The socket created in transportCreateEndpoint is not freed up
immediately if the SocketBindAddr fails, and is not freed up
by the application in the corresponding MessageQ_delete as well.
These unbound sockets only get freed up as part of the process
cleanup of any open file descriptors by the Linux kernel. Close
these sockets proactively, so that there are no stale sockets
present during the execution of a process.

Signed-off-by: Suman Anna <s-anna@ti.com>
6 years agoMove Hwi.excHandlerFunc assignment into ti.deh.Deh:module$use() 3.30.01.10_eng 3.30.01.11_eng
Robert Tivy [Wed, 22 Oct 2014 22:51:53 +0000 (15:51 -0700)]
Move Hwi.excHandlerFunc assignment into ti.deh.Deh:module$use()

A previous change moved this assignment from module$static$init() to
module$meta$init(), but this no longer works since ti.ipc.ipcmgr:getLibs()
now references ti.deh.Deh:
        var deh = xdc.module("ti.deh.Deh");
and if a module has not otherwise been "used" up until this point then
the xdc.module() reference will run module$meta$init(), and since this
is getLibs() you can't "useModule" at this point (which Deh's
module$meta$init() does).

Also, some other change is causing ti.deh's getLibs() to be called even
when nothing does a useModule() on any module within the ti.deh package.
For a device which ti.deh doesn't support, such as Keystone, it would
previously throw an error and fail the whole app build.  Now, ti.deh's
getLibs() is changed to contribute nothing if none of its modules are
used for an unsupported device.

6 years agoChange fw_rsc_intmem to match Linux 3.14
Robert Tivy [Fri, 17 Oct 2014 22:50:41 +0000 (15:50 -0700)]
Change fw_rsc_intmem to match Linux 3.14

This struct was changed for Linux 3.14, so we need to match that new
one because Ipc 3.30 uses Linux 3.14.

Later we should introduce a construct that will allow the user to specify
the older struct usage.

6 years agoMerge remote-tracking branch 'origin/ipc-3.23-next' into ipc-next
Robert Tivy [Fri, 17 Oct 2014 23:01:27 +0000 (16:01 -0700)]
Merge remote-tracking branch 'origin/ipc-3.23-next' into ipc-next

6 years agoBuild libipc and libutils in QNX for dynamic linking 3.23.01.03
vwan@ti.com [Tue, 14 Oct 2014 23:59:55 +0000 (16:59 -0700)]
Build libipc and libutils in QNX for dynamic linking

This addresses SDOCM00113498.

Signed-off-by: VW <vwan@ti.com>
6 years agoProperly free previously allocated file path string and other minor cleanup
vwan@ti.com [Tue, 14 Oct 2014 19:04:02 +0000 (12:04 -0700)]
Properly free previously allocated file path string and other minor cleanup

Signed-off-by: VW <vwan@ti.com>
6 years agoAdd carriage return to strings returned by /dev/ipc-file/* and /dev/ipc-state/*
vwan@ti.com [Tue, 14 Oct 2014 16:42:15 +0000 (09:42 -0700)]
Add carriage return to strings returned by /dev/ipc-file/* and /dev/ipc-state/*

Signed-off-by: VW <vwan@ti.com>
6 years agoAllow QNX IPC resource manager to be launched without starting any core
vwan@ti.com [Mon, 13 Oct 2014 22:53:01 +0000 (15:53 -0700)]
Allow QNX IPC resource manager to be launched without starting any core

This commit allows the resource manager to be launched without specifying
any core in its core list. This is useful now that cores can be started
at a subsequent time.

Signed-off-by: VW <vwan@ti.com>
6 years agoReject all operations on DSP2 in QNX
vwan@ti.com [Mon, 13 Oct 2014 22:12:19 +0000 (15:12 -0700)]
Reject all operations on DSP2 in QNX

Since DSP2 is not supported on QNX, we need to reject all attempts at
interacting with DSP2 using the new start/stop feature.

Signed-off-by: VW <vwan@ti.com>
6 years agoPrevent SR0 owner from being stopped if host process is still using GateMP
vwan@ti.com [Sun, 12 Oct 2014 22:58:54 +0000 (15:58 -0700)]
Prevent SR0 owner from being stopped if host process is still using GateMP

Signed-off-by: VW <vwan@ti.com>
6 years agoFix an issue where disabling mailbox interrupts on DSP1 disables IPU interrupts
vwan@ti.com [Fri, 10 Oct 2014 23:44:42 +0000 (16:44 -0700)]
Fix an issue where disabling mailbox interrupts on DSP1 disables IPU interrupts

Signed-off-by: VW <vwan@ti.com>
6 years agoAdd initial support to start/stop cores after launching QNX IPC resource manager
vwan@ti.com [Fri, 10 Oct 2014 23:31:40 +0000 (16:31 -0700)]
Add initial support to start/stop cores after launching QNX IPC resource manager

This commit adds the ability to start and stop slave cores after the QNX
IPC resource manager has been previously launched, with at least a few known
caveats that would be fixed subsequently.

It addresses SDOCM00113295.

Signed-off-by: VW <vwan@ti.com>
6 years agoFix "@link @setConfig", should be "@link #setConfig" 3.23.01.02_eng
Robert Tivy [Wed, 8 Oct 2014 00:40:04 +0000 (17:40 -0700)]
Fix "@link @setConfig", should be "@link #setConfig"

6 years agoSDOCM00113338 fix: reduce number of Tasks for messageq_multi.c
Robert Tivy [Sat, 4 Oct 2014 01:00:32 +0000 (18:00 -0700)]
SDOCM00113338 fix: reduce number of Tasks for messageq_multi.c

Also reduce the maximum number of threads supported in MessageQMulti.c

The need for this fix arose when the BIOS heap size was reduced in
some .cfg files in packages/ti/ipc/tests.  The new, smaller heap was
not sufficient for allocating 55 Tasks (it failed at about the 30th
Task).

6 years agoSDOCM00113339 fix: change TMS320TCI6630 to TMS320TCI6630K2L
Robert Tivy [Fri, 3 Oct 2014 22:18:30 +0000 (15:18 -0700)]
SDOCM00113339 fix: change TMS320TCI6630 to TMS320TCI6630K2L

6 years agoSDOCM00113341 fix: restrict writing Ipc_Reserved struct to core's own area
Robert Tivy [Mon, 6 Oct 2014 21:45:29 +0000 (14:45 -0700)]
SDOCM00113341 fix: restrict writing Ipc_Reserved struct to core's own area

In Ipc_attach(), when a 'memReq' size is 0, a core can write to another
core's Ipc_Reserved struct in SR0 when it isn't supposed to *ever* write
another core's area.  The incorrect assignment needs to be done only when
MultiProc_self() < remoteProcId, similarly to how it is done in other places,
which ensures that a core is writing to only its own IpcReserved area.

6 years agoSDOCM00113171 fix: release NameServerRemoteRpmsg gate on timeout
Robert Tivy [Tue, 9 Sep 2014 22:43:54 +0000 (15:43 -0700)]
SDOCM00113171 fix: release NameServerRemoteRpmsg gate on timeout

The NameServerRemoteRpmsg_get() function has a timeout-based Semaphore_pend
that can return failure when the timeout expires.  The function returns
without releasing the gate when the timeout failure happens.  This causes
any other thread to block when calling MessageQ_open(), while subsequent
MessageQ_open() calls on the same thread can succeed.

7 years agoSDOCM00108130 IPC custom library build creates empty folder 3.30.00.07_eng 3.30.00.08_eng 3.30.00.09
Ramsey Harris [Thu, 28 Aug 2014 20:32:00 +0000 (13:32 -0700)]
SDOCM00108130 IPC custom library build creates empty folder

When not building product libraries, remove directive in makefile
which was creating output folder. The directive resulted in empty
folder named 'undefined'.

7 years agoSDOCM00112918 fix: release NameServerRemoteRpmsg gate on timeout
Robert Tivy [Tue, 9 Sep 2014 22:43:54 +0000 (15:43 -0700)]
SDOCM00112918 fix: release NameServerRemoteRpmsg gate on timeout

The NameServerRemoteRpmsg_get() function has a timeout-based Semaphore_pend
that can return failure when the timeout expires.  The function returns
without releasing the gate when the timeout failure happens.  This causes
any other thread to block when calling MessageQ_open(), while subsequent
MessageQ_open() calls on the same thread can succeed.

7 years agoSDOCM00107419 fix: add cast for OMAPL138 & Keystone devices
Robert Tivy [Wed, 10 Sep 2014 00:29:51 +0000 (17:29 -0700)]
SDOCM00107419 fix: add cast for OMAPL138 & Keystone devices

Since VirtQueue_Params is XDC-generated for OMAPL138 & Keystone devices
we need to add a (xdc_Fxn) cast for the 'callback' element assignment.
Other devices (OMAP5 & VAYU) already have the correct type for the
'callback' element in the manually-generated VirtQueue.h file.

7 years agoMove Watchdog module configuration assignments to module$use()
Robert Tivy [Sat, 6 Sep 2014 00:12:40 +0000 (17:12 -0700)]
Move Watchdog module configuration assignments to module$use()

When the ti.deh package is loaded the Watchdog module$meta$init() method
is run.  If the package is loaded by an architecture/device that is not
supported by Watchdog then module$meta$init() fails and the whole
configuration fails.  This happens even when the Watchdog module is *not*
used, such as when the ti.deh package gets loaded as a result of an
xdc.module() reference.

7 years agoFix recovery mechanism for late-attach use-case in QNX
vwan@ti.com [Thu, 28 Aug 2014 21:55:50 +0000 (14:55 -0700)]
Fix recovery mechanism for late-attach use-case in QNX

This patch adds support for recovery from MMU faults, exceptions,
and watchdog timeouts when 'late-attached' to a slave core (SDOCM00112666).

Signed-off-by: VW <vwan@ti.com>
7 years agoUse different #include path for 3.14 Linux 3.30.00.06_eng
Robert Tivy [Fri, 29 Aug 2014 01:59:57 +0000 (18:59 -0700)]
Use different #include path for 3.14 Linux

Linux 3.14 has split header files into kernel and user ones, which means
MmRpc.c needs to include rpmsg_rpc.h from user location.

7 years agoAdd Ipc.c to default source file list
Robert Tivy [Fri, 29 Aug 2014 16:25:39 +0000 (09:25 -0700)]
Add Ipc.c to default source file list

7 years agoFixes to support minimal configuration of IPC
Ramsey Harris [Thu, 28 Aug 2014 00:20:28 +0000 (17:20 -0700)]
Fixes to support minimal configuration of IPC

Validate processor ID in Notify_attach, which is the public API.
No need to re-validate in NotifyDriverMbx_Instance_init(). This
removes the Ipc dependency (needed for the assert object).

7 years agoFixes to support minimal configuration of IPC
Ramsey Harris [Wed, 27 Aug 2014 23:34:26 +0000 (16:34 -0700)]
Fixes to support minimal configuration of IPC

When generating file list for a custom build, don't include Ipc.c
unless the application actually used the module. Added --output_file
option to makefile template; other minor cleanup.

7 years agoIPC makefile template improvements
Ramsey Harris [Tue, 26 Aug 2014 22:07:58 +0000 (15:07 -0700)]
IPC makefile template improvements

For custom builds, the makefile template was computing the list
of source files twice. Now it caches the list in a local array.

7 years agovayu: add new rprocList[] setting for Linux v3.14
Robert Tivy [Tue, 26 Aug 2014 17:04:49 +0000 (10:04 -0700)]
vayu: add new rprocList[] setting for Linux v3.14

The rproc order for vayu remote cores has changed in Linux kernel v3.14,
as part of the conversion to completely DT-based configuration.

7 years agoMerge branch 'ipc-3.23-next' into ipc-next 3.30.00.04_eng 3.30.00.05_eng
Robert Tivy [Wed, 20 Aug 2014 18:16:56 +0000 (11:16 -0700)]
Merge branch 'ipc-3.23-next' into ipc-next

7 years agoSDOCM00112254 - Add IPC support for Soprano
Janet Head [Wed, 20 Aug 2014 17:12:40 +0000 (10:12 -0700)]
SDOCM00112254 - Add IPC support for Soprano

7 years agoSDOCM00112680 - IPC for Concerto has incorrect implemetations of NotifyDriverCirc_ena...
Janet Head [Wed, 20 Aug 2014 17:04:54 +0000 (10:04 -0700)]
SDOCM00112680 - IPC for Concerto has incorrect implemetations of NotifyDriverCirc_enable() and NotifyDriverCirc_enableEvent()

7 years agoApplied bug fixes found by Vision SDK team.
Ramsey Harris [Mon, 18 Aug 2014 23:55:41 +0000 (16:55 -0700)]
Applied bug fixes found by Vision SDK team.

On EVE1, fixed mailbox index computed when enabling interrupt
from DSP2. Fixed mailbox table entry for IPU1-1 to DSP1 mapping.

7 years agoUse core-specific WUGEN masks for suspend on OMAP 3.23.00.01
Robert Tivy [Wed, 13 Aug 2014 23:42:38 +0000 (16:42 -0700)]
Use core-specific WUGEN masks for suspend on OMAP

The WUGEN event masks for each core are set to values appropriate
for the architecture and core.

A new API is added that allows the application to control the WUGEN
event mask (IpcPower_{get,set}WugenEvtMask()).

Addresses CQ SDOCM00108585.

7 years agoUpdate QNX Makefile to setup environment for QNX 6.6.0
vwan@ti.com [Tue, 12 Aug 2014 23:27:54 +0000 (16:27 -0700)]
Update QNX Makefile to setup environment for QNX 6.6.0

Signed-off-by: VW <vwan@ti.com>
7 years agoLinux: GameMP inUse arrays had invalid address mappings
Arnie Reynoso [Wed, 13 Aug 2014 18:14:22 +0000 (11:14 -0700)]
Linux: GameMP inUse arrays had invalid address mappings

The GameMP address for the inUse arrays were being assigned incorrectly.
A page alignment was originally needed to properly map the arrays in
user space. After mapping, the alignment difference wasn't being taken
into consideration.

7 years agoAdd late-attach support for IPUs in QNX
vwan@ti.com [Thu, 14 Aug 2014 00:31:23 +0000 (17:31 -0700)]
Add late-attach support for IPUs in QNX

This commit addresses CQ SDOCM00108582.

Signed-off-by: VW <vwan@ti.com>
7 years agoFixed bad array reference in IPU ducati interrupt handler.
Ramsey Harris [Fri, 8 Aug 2014 23:04:22 +0000 (16:04 -0700)]
Fixed bad array reference in IPU ducati interrupt handler.

The ducati ISR was still using the vayu virtual processor IDs
instead of the new tda3xx ones.

7 years agoFixed incorrect mailbox table entry for tda3xx 3.30.00.03_eng
Ramsey Harris [Fri, 8 Aug 2014 19:57:10 +0000 (12:57 -0700)]
Fixed incorrect mailbox table entry for tda3xx

For TDA3xx, the IPU1-0 to EVE1 entry in the mailbox table was
incorrect. User should be 0 instead of 4.