5. PSDK QNX Components

5.1. QNX BSP release

The QNX BSP package must be downloaded from QNX Software Center.

Note

Refer Release Notes software dependencies section for more information.

For the purposes of creating a consolidated build process and in turn allowing for support of scripts to create SD content, the BSP needs to be extracted to the folder within the PSDK QNX build environment as show below.

For QNX SDP 7.1:

# Unzip the BSP
mkdir -p ${PSDK_RTOS_PATH}/psdkqa/qnx/bsp
cd ${PSDK_RTOS_PATH}/psdkqa/qnx/bsp
unzip ${QNX_BASE}/bsp/BSP_ti-j7200--evm_br-mainline_be-710_<version>.zip

5.1.1. TI Modifications to the BSP

Memory Carveout

Memory sections with pre-defined physical addresses must be set aside in the QNX BSP IFS build file so that the memory is not given to other programs and can be used solely by the remote cores

Specify a section to be set aside by modifying the startup line to use the “-r” option. For example, to reserve 0xA0000000 bytes, at physical address 0x5C000000, on J7200 QNX BSP, the build file the arguments would be (highlighted below):

[+keeplinked] startup-j721e-evm -v -r0xA0000000,0x5C000000,1

Reference TI build file

The reference TI build file is provide in the qnx/scripts/bsp/<BSP_REVISION> directory.

# Copy in TI specific build scripts and settings to allow building of a QNX-IFS which supports
# vision_apps demos
cp ${PSDK_RTOS_PATH}/psdkqa/qnx/scripts/bsp/<BSP_REVISION>/j7200-evm-ti.build ${PSDK_RTOS_PATH}/psdkqa/qnx/bsp/images/

Note

Other build file deltas between QNX and TI PSDK QNX environment may be present. TI modifies build scripts for ease of use on TI EVM and running of demonstration software

5.2. QNX 7.1 Dependency Packages

The QNX 7.1 dependency packages must be downloaded from QNX Software Center.

Note

Refer Release Notes software dependencies section for more information.

5.3. SCI Client Resource Manager

The SCI Client Resource Manager (tisci-mgr) provides support for multiple users to make use of the sciclient library from PSDK RTOS without interfering with each other requests. It provides a mechanism to serialize the transactions to the DMSC.

Note

The tisci-mgr logs the SYSFW information to the slogger.

Note

Refer PSDK RTOS Components for more information about each individual component.

5.4. IPC Resource Manager

5.4.1. Overview

The IPC resource manager (tiipc-mgr) provides a processor-agnostic API which can be used for communication between processors in a multi-processor environment.

The IPC resource manager provides a user library (tiipc-usr) that exposes the IPC LLD API to QNX applications for communication through the IPC resource manager. The IPC LLD API is described in detail in the PDK documentation.

5.4.2. Example Application

An IPC example application (ipc_test) is provided to test the IPC communication with remote cores running the IPC echo test remote core firmware images.. To use this application, the appropriate remote core firmware images need to be built and placed on the target filesystem. The following remote core firmware images are needed:

Firmware Name

Core(s)

ipc_qnx_echo_testb_freertos

use for mcu1_0

ipc_qnx_echo_test_freertos

use for mcu2_0 & mcu2_1

For remote core firmware build instructions, please refer to the IPC LLD PDK documentation.

Note

For the ipc_test to work with ipc echo test remote core firmware images, The IPC resource manager (tiipc-mgr) need to be built with “VISION_APPS_BUILD_FLAGS_MAK=0” build flag, so that the correct VRING_BASE_ADDRESS is used.

Note

The below step to copy the remote core firmware to the rootfs partition of target filesystem is applicable only to the SPL-UBOOT bootflow. For the SBL-BootApp bootflow, the ipc echo test image need to be bundled into the Appimage. Please refer to the MCUSW documentation for details.

Once the remote core firmware is built, copy the images to the target filesystem in the rootfs partition. Existing firmware binaries should be backed up as required. The ex02* firmware binaries should be renamed to the firmware name expected by the bootloader. For example:

cp ipc_qnx_echo_test_freertos_mcu2_1_release.xer5f ${ROOTFS}/lib/firmware/j7200-main-r5f0_1-fw
cp ipc_qnx_echo_test_freertos_mcu2_0_release.xer5f ${ROOTFS}/lib/firmware/j7200-main-r5f0_0-fw

Note

As seen above, the mcu1_0 firmware image is not copied to the target filesystem in the rootfs partition. Instead it need to be built into the tispl.bin as part of the SPL-UBOOT boot binaries.

For testing ipc with mcu1_0 image, the PSDK Linux package is required to be installed. The default SPL-UBOOT binaries include the ipc test mcu1_0 image that work with Linux only. Hence for QNX, we need to rebuilt the SPL-UBOOT to include the “ipc_qnx_echo_testb_freertos_mcu1_0_release” image.

Below are the steps to do this:

  1. Build “ipc_qnx_echo_testb_freertos_mcu1_0_release” image.

cd ${PSDK_RTOS_PATH}/pdk_j7200_{version}/packages/ti/build
make -s ipc_qnx_echo_testb_freertos BOARD=j7200_evm CORE=mcu1_0 -j2
  1. Copy the genreate mcu1_0 firmware image to the PSDK Linux path mentioned below and rebuild UBOOT.

cp ${PSDK_RTOS_PATH}/pdk_j7200_{version}/packages/ti/binary/ipc_qnx_echo_testb_freertos/bin/j7200_evm/ipc_qnx_echo_testb_freertos_mcu1_0_release_strip.xer5f ${PSDK_LINUX_PATH}/board-support/prebuilt-images/ipc_echo_testb_mcu1_0_release_strip.xer5f
cd ${PSDK_LINUX_PATH}
make u-boot_clean
make u-boot
  1. Copy the newly generated UBOOT file to the SD card boot partition

cd ${PSDK_LINUX_PATH}/board-support/u-boot_build/r5/tiboot3.bin ${BOOTFS}
cd ${PSDK_LINUX_PATH}/board-support/u-boot_build/a72/tispl.bin  ${BOOTFS}
cd ${PSDK_LINUX_PATH}/board-support/u-boot_build/a72/u-boot.img  ${BOOTFS}

After copying the firmware and booting the target, the ipc_test can be run from the command line (example output given below):

J7EVM@QNX:/# /sd/tibin/ipc_test
IPC_echo_test (core : mpu1_0) .....
responderFxn will stay active. Please use ctrl-c to exit the test when finished.
SendTask1: mpu1_0 <--> mcu1_0, Ping- 10, pong - 10 completed
SendTask3: mpu1_0 <--> mcu2_0, Ping- 10, pong - 10 completed
SendTask4: mpu1_0 <--> mcu2_1, Ping- 10, pong - 10 completed

Make sure that the tiipc-mgr is running before executing the ipc_test. Note that the test app will not exit. Press “ctrl+c” to exit

Note

If SPL-UBOOT boot flow is used, the ipc_test will not be able to communicate with mcu1_1. All other core will work. Note that mcu1_1 is not loaded with any firmware image.

Note

If SBL / BootApp boot flow is used, the ipc_test will not be able to communicate with mcu1_0, & mcu1_1. All other core will work.

Note

Refer PSDK RTOS Components for more information about each individual component.

5.5. UDMA Resource Manager

The UDMA resource manager (tiudma-mgr) provides support for multiple users to make use of the UDMA functionality without interfering with each other requests.

Note

Refer PSDK RTOS Components for more information about each individual component.

5.6. Shared Memory Allocator

The Shared Memory Allocator resource manager (shmemallocator) provides support for multiple users to allocate memory from the shared memory region. This shared memory region is carvedout of the QNX memory as part of the QNX Startup parameters.

5.7. CPSW2G DEVNP driver

5.7.1. Overview

The CPSW2G DEVNP network driver can be viewed as the “glue” between the underlying cpsw2g lowlevel driver, and the software infrastructure of io-pkt, the protocol stack above it. The “bottom half” of the driver is coded specifically to interact with the PDK’s cpsw & udma low-level drivers, and the “top half” of the driver is coded specifically for io-pkt.

Note

Refer PSDK RTOS Components for more information about each individual component.

Note

The cpsw2g driver’s CPTS interrupt thread must always be higher priority compared to the RX and TX threads of the driver. By default, the network stack thread is at priority 21. Hence the RX and TX thread are created at this priority level. The CPTS interrupt thread is create at priority 22. These priority values can be changed by providing them as command-line parameters. See driver use command.

5.7.2. Running

Boot the board with the SDcard. At the QNX prompt, run below, to mount the SD card and then launch the executable:

tisci-mgr
tiudma-mgr
io-pkt-v6-hc -d cpsw2g
dhclient -nw am0

The cpsw2g driver configures all the driver created threads to have the runmask as 0x2. This is to make sure all the threads run on both cores of A72. In case, the driver has to use a one of the cores only, please use the run_mask_cpu= command-line option.

For debug traces, run the below command before starting the DEVNP driver and start the driver with increased verbose parameter (ex: verbose=0x3ff) this will show all the drivers slog messages

slog2info –c
slog2info -w &

5.7.3. Additonal steps

  • Run “if_up -p am0” to check if the interface is ready

  • Run “ifconfig am0 up” to bring UP the link.

  • Run “dhclient -nw am0” for DHCP server provide the IP address

  • Run “ifconfig -v” to check the assigned IP address and status

  • Run “tcpdump -e” to look at the tcp traffic

5.7.4. Usage of devnp-cpsw2g.so

We can run the below command to get the usage of the DEVNP driver

$ use devnp-cpsw2g.so

devnp-cpsw2g.so j721e cpsw2g ethernet driver based on enet low level driver

Syntax:
  io-pkt-v6-hc -d cpsw2g [option[,option ...]] ...

Options (to override autodetected defaults):
  verbose=num             Set verbosity level (default: 0).
  mac-to-mac=1            Set for mac-to-mac mode (default: 0)
  speed=100|1000          Media data rate for port 0 in Mb/s. (default: 1000)
  p0mac=XXXXXXXXXXXX      Custom MAC address to use on port 0.
  ptp=0|1                 1 to enable PTP support (default: 0)
  promiscuous=0|1         1 to enable promiscuous mode (default: 0)
  typed_mem=name          Set the typed memory
  udma_chnum=val          Set the preferred udma channel to use (default: dynamic allocation)
  tx_freeq_threshold=val  Set the tx free Q threshold value (default: 120), must be less than no of TX descriptors.
  tx_descriptor_cnt=val   Set the tx descriptors count value (default: 128, Max: 256)
  rx_descriptor_cnt=val   Set the rx descriptors count value (default: 128, Max: 256)
  run_mask_cpu=val        Set the run mask - cpu core where threads are scheduled (default:2) (0 - core0, 1 - core1, 2 - both core)
  poll_phy_ms=val         Set the frequency in ms, to poll the phy status, and poll for management of resources (default: 2000)
  cache_ops=val           Set the cache operation to on/off (default:0  -> cache-coherency set), 1 - Turn on cache off -> cache-coherency off)
  smmu=0|1                1 to enable smmu support (default: 0)
  virt_id=val             Set the virt_id to use for the dma channel when smmu is enabled
  hw_csum=1               1 to enable hw csum (default: 0)
  joinvlan="1;2;3..."     List of VLANs to join
  rx_intr_prio=val        Set the rx interrupt thread priority (default: 21)
  tx_intr_prio=val        Set the tx interrupt thread priority (default: 21)
  cpts_intr_prio=val      Set the cpts interrupt thread priority (default: 22)
  no_stack_thread=val     1 to disable using the stack thread needed for bridge and fastforward to work  (default: 0)

Examples:
  # Start io-pkt using the driver:
  io-pkt-v6-hc -d cpsw2g
  ifconfig am0 192.0.2.1

  # Start io-pkt using the driver for mac-to-mac mode in gitbit speed
  io-pkt-v6-hc -d cpsw2g verbose=1,mac-to-mac=1,speed=1000
  ifconfig am0 192.0.2.1

  # Start io-pkt using the driver with custom mac address
  io-pkt-v6-hc -d cpsw2g p0mac=001122334455
  ifconfig am0 192.0.2.1

  # Start io-pkt using the driver with typed memory "ram":
  io-pkt-v6-hc -d cpsw2g typed_mem=ram -ptcpip pkt_typed_mem=ram
  ifconfig am0 192.0.2.1

  # Start io-pkt using the driver with typed memory "ram",preferred udma channel "24" and smmu enabled:
  io-pkt-v6-hc -d cpsw2g ptp=1,typed_mem=ram,udma_chnum=24,smmu=1 -ptcpip pkt_typed_mem=ram
  ifconfig am0 192.0.2.1

  # Start io-pkt using the driver with hw csum enabled
  io-pkt-v6-hc -d cpsw2g hw_csum=1
  ifconfig am0 tcp4csum udp4csum tcp6csum udp6csum
  ifconfig am0 192.0.2.1

5.7.5. Starting driver with other options

To run the cpsw2g devnp driver with full debug log (run “slog2info” to see the debug log):

io-pkt-v6-hc –d cpsw2g verbose=0xff

To run the cpsw2g devnp driver in mac-to-mac mode with 1Gbps:

io-pkt-v6-hc –d cpsw2g mac-to-mac=1,speed=1000

To run the cpsw2g devnp driver in mac-to-mac mode with 1Gbps with gPTP:

io-pkt-v6-hc –d cpsw2g ptp=1,mac-to-mac=1,speed=1000

To get a dynamic IP address for the cpsw2g port:

dhclient -nw am0

To get a static IP address for the cpsw2g port:

ifconfig am0 up
ifconfig am0 <static_ip_address>

To enabled HW CSUM offloading support:

io-pkt-v6-hc -d cpsw2g hw_csum=1
ifconfig am0 tcp4csum udp4csum tcp6csum udp6csum
dhclient -nw am0

5.8. CPSW5G DEVNP driver

5.8.1. Overview

The CPSW5G DEVNP network driver implemented as “virtual” driver that communicates with the Ethernet Firmware Switch firmware running on the R5 core. The control message transfer is done via IPC. The RX and TX data packets are passed to the CPSW5G port using the UDMA.

CPSW5G Firmware on MCU2_0 implemented TimeSync Module utilizing CPTS timer for PTP support to sync with Master clock on the network. TimeSync module configured PTP stack with the following default properties.

  • Ordinary Clock

  • IEEE 802.3 Transport

  • Peer Delay Mechanism (P2P)

For TimeSync Module design, please refer to EthFW documentation for details. CPSW5G DEVNP driver has included an option to return CPTS timer timestamp via devctl() with PTP_GET_TIME command. Please refer to the example, ptp_test, to see how to use PTP_GET_TIME command to get CPTS timer timestamp from cpsw5g devnp driver.

Note

Refer PSDK RTOS Components for more information about EthFW component.

5.8.2. Running

Boot the board with the SDcard. At the QNX prompt, run below, to mount the SD card and then launch the executable:

tisci-mgr
tiipc-mgr
tiudma-mgr
io-pkt-v6-hc -d cpsw5g
dhclient -nw an0

The cpsw5g driver configures all the driver created threads to have the runmask as 0x2. This is to make sure all the threads run on both cores of A72. In case, the driver has to use a one of the cores only, please use the run_mask_cpu= command-line option.

For debug traces, run the below command before starting the DEVNP driver and start the driver with increased verbose parameter (ex: verbose=0x3ff) this will show all the drivers slog messages

slog2info –c
slog2info -w &

5.8.3. Additional steps

  • Run “if_up -p an0” to check if the interface is ready

  • Run “ifconfig an0 up” to bring UP the link.

  • Run “dhclient -nw an0” for DHCP server provide the IP address

  • Run “ifconfig -v” to check the assigned IP address and status

  • Run “tcpdump -e” to look at the tcp traffic

5.8.4. Usage of devnp-cpsw5g.so

We can run the below command to get the usage of the DEVNP driver

$ use devnp-cpsw5g.so

devnp-cpsw5g.so j721e cpsw 5G virtual ethernet driver based on enet low level driver

Syntax:
  io-pkt-v6-hc -d cpsw5g [option[,option ...]] ...

Options (to override autodetected defaults):
  verbose=num             Set verbosity level (default: 0).
  ptp=0|1                 1 to enable PTP support (default: 0)
  typed_mem=name          Set the typed memory
  udma_chnum=val          Set the preferred udma channel to use (default: dynamic allocation)
  tx_freeq_threshold=val  Set the tx free Q threshold value (default: 120), must be less than no of TX descriptors.
  tx_descriptor_cnt=val   Set the tx desciptors count value (default: 128, Max: 256)
  rx_descriptor_cnt=val   Set the rx desciptors count value (default: 128, Max: 256)
  run_mask_cpu=val        Set the run mask - cpu core where threads are scheduled (default:2) (0 - core0, 1 - core1, 2 - both core)
  poll_phy_ms=val         Set the frequency in ms, to poll the phy status, and poll for management of resources (default: 10000)
  cache_ops=val           Set the cache operation to on/off (default:0  -> cache-coherency set), 1 - Turn on cache off -> cache-coherency off)
  smmu=0|1                1 to enable smmu support (default: 0)
  virt_id=val             Set the virt_id to use for the dma channel when smmu is enabled
  mac-to-mac=1            Set for mac-to-mac mode (default: 0) (only when EthFW is setup for mac-to-mac mode)
  speed=100|1000          Media data rate for link in Mb/s. (default: 1000)
  joinvlan="1;2;3..."     List of VLANs to join
  rx_intr_prio=val        Set the rx interrupt thread priority (default: 21)
  tx_intr_prio=val        Set the tx interrupt thread priority (default: 21)
  no_stack_thread=val     1 to disable using the stack thread needed for bridge and fast-forward to work  (default: 0)
  hw_csum=1               1 to enable hw csum (default: 0)
  p0mac=XXXXXXXXXXXX      Custom MAC address to use on virtual port

Examples:
  # Start io-pkt using the driver:
  io-pkt-v6-hc -d cpsw5g
  ifconfig an0 192.0.2.1

  # Start io-pkt using the driver with typed memory "ram":
  io-pkt-v6-hc -d cpsw5g typed_mem=ram -ptcpip pkt_typed_mem=ram
  ifconfig an0 192.0.2.1

  # Start io-pkt using the driver with typed memory "ram",preferred udma channel "24" and smmu enabled:
  io-pkt-v6-hc -d cpsw5g ptp=1,typed_mem=ram,udma_chnum=24,smmu=1 -ptcpip pkt_typed_mem=ram
  ifconfig an0 192.0.2.1

  # Start io-pkt using the driver for mac-to-mac mode in gitbit speed
  io-pkt-v6-hc -d cpsw5g mac-to-mac=1,speed=1000
  ifconfig am0 192.0.2.1

5.8.5. Starting driver with other options

To run the cpsw5g devnp driver with gPTP:

io-pkt-v6-hc –d cpsw5g ptp=1

Run "ptp_test" utility to  verfiy/get the synchronized time from EthFW.

To run the cpsw5g devnp driver in mac-to-mac mode with 1Gbps:

io-pkt-v6-hc –d cpsw5g mac-to-mac=1

5.9. K3conf QNX utility

K3CONF is a QNX port of a standalone application designed to provide a quick’n easy way to dynamically diagnose Texas Instruments’ K3 architecture based processors. K3CONF is intended to provide similar experience to that of OMAPCONF that runs on legacy TI platforms.

Note

WARNING: This is work in progress! Please don’t expect things to be complete in any dimension. Use at your own risk. And keep the reset button in reach.

To get more details on how to use k3conf utility, run below:

k3conf --help