Quantcast
Channel: Processors forum - Recent Threads
Viewing all 17527 articles
Browse latest View live

NetworkRx to Isp (Image Processing Issue)

$
0
0

Hi

I am sending the bayer_raw image using network_tx.exe tool to a usecase for processing.

THe usecase structure is :

NetworkRx -> IssM2mIsp -> Display

IssM2mIsp -> Alg_aewb

I am able to send the image and i am seeing output over display but the image on the display is not expected one.

I have an Bayer_raw with grbg format (8 bit and 12 bit )  which will result into an green image  when i converted it into rgb888, but when i send the same image into the usecase it is producing a different colour.

I am using tda2px and vsdk3.0.3.

Please help me in processing the image correctly and also please mention the configuration which i have to setup to do this image processing.

Thanks and Regards,

Aneesh


PROCESSOR-SDK-AM437X: EtherCAT Master real-time fluctuation

$
0
0

Part Number: PROCESSOR-SDK-AM437X

Hello,

       The hardware of my side is ti's arm am4377, the software is ti's SDK 5.2, and the kernel version is preempt RT 4.14.79, as shown in the following figure:

       

      I work as the EtherCAT master station. After the master station is running, the maximum fluctuation of the real-time system monitored by the master station reaches 225138ns, which exceeds 20% of the master station's synchronization period of 1ms, resulting in the master station's warning,as shown in the following figure:

   

     How can the real-time system be so volatile? Is it not well configured? Thank you!

AMIC110: UART Configuration Problem

$
0
0

Part Number: AMIC110

Hi,

I am using AMIC110, configured as EtherCAT Slave, full stack generated by SSC Tool.

Now I need to enable UART and need to recieve data though UART. what ever the data i receive through UART should be loaded on to the Ethercat Frame.

But the problem is when i configure UART With below configuration the program is getting stuck at UART_open(...) function, why?

Is there any thing i have did wrong? what else i need to do?

Can u share me the sequence of steps i need to take care to configure UART.......

//***********************************************************************************************************************//

 #define UART_INSTANCE 0

UART_HwAttrs uart_cfg;
Board_init(BOARD_INIT_PINMUX_CONFIG | BOARD_INIT_MODULE_CLOCK | BOARD_INIT_ICSS_PINMUX | BOARD_INIT_UART_STDIO);

UART_socGetInitCfg(UART_INSTANCE, &uart_cfg);
UART_socSetInitCfg(UART_INSTANCE, &uart_cfg);


UART_init();
Board_setDigOutput(0x20);

params.baudRate = 115200;
params.writeDataMode = UART_DATA_BINARY;
params.readDataMode = UART_DATA_BINARY;
params.readReturnMode = UART_RETURN_FULL;
params.readEcho = UART_ECHO_OFF;

UART_Params_init(&params);
handle = UART_open(UART_INSTANCE, &params);    // Stuck here...
if (!handle) {
//System_printf("UART did not open");
}
const unsigned char hello[] = "Hello World\n";
int32_t ret;
ret = UART_write(handle, hello, sizeof(hello)); //..

//*********************************************************************************************************************//

Please reply as soon as possible....

TDA2EVM5777: TIDL problem

$
0
0

Part Number: TDA2EVM5777

Hi FAE:
we are using the TIDL tools in the PROCESSOR_SDK_VISION_03_07_00_00. it performs well in the Mnist test which contains only convolution layer and ReLU layer in its convolution block. but when we use the tools in our model which cotains convolution, normalization, normalization scale and ReLU layers, the error increases rapidly after a few layers. we compare the trace dump in the tempDir directory with our caffe floating result and find some different. After the normalization scale layer, a few channels are negatived compared with the caffe result. that is, the maxium point of the caffe result corresponds to the minimum point in the TIDL trace dump. if we mutiply these channels by -1(XOR them with 0xFF), the following layer results seem to be improved. we wonder if this phenomenon is related with the bad result in our model. Would you please help us to improve the result.
Our model is in the attachment. the example channel is the layer 0 channel 4, the caffe result is in the directory \caffe_result_float, it is negative to the channel 4 of trace_dump_1_427x240.y.

(Please visit the site to view this file)

EVMK2GX: EVMK2GX clock down to 600MHz

$
0
0

Part Number: EVMK2GX

Hi Experts,

A customer would evaluate the 66AK2G12 600MHz with EVMK2GX.
Could you please let me know the easiest and safer way to down the clock from 1GHz to 600GHz?

Thank you very much for your kind help.
Best regards,
Hitoshi

AM5749: TIDL model training on the fly

$
0
0

Part Number: AM5749

After the trained model from offline running on the target board, can the model keep training by itself on the fly?

[TDA4M] Connector CSI2 Expansion to MIPI Camera

$
0
0

Hi.

FPD-Link was not used.

AR0233 module was connected to CSI2 and MIPI-4lanes.

I deleted FPD-link related to iss_sensor_ar0233.c and modified I2C-ID .

I want to know what needs to be changed in the CSI2 code.

I am working on app_single_cam_main.c

Thank you .

WEBENCH® Tools/TDA2EG-17: about TDA4X

$
0
0

Part Number: TDA2EG-17

Tool/software: WEBENCH® Design Tools

Hi,

     we are planing next general product ,Could you supply some data to me about TDA4X. include SDK and BSP  ,or demoboard data ,reference design.thanks.


PROCESSOR-SDK-AM335X: Flash Application

$
0
0

Part Number: PROCESSOR-SDK-AM335X

Hi,

I am using AM335x  EVM board and SDK RTOS. 

I want to flash the bootloader image and application to NAND through RS422. How can I do it.

I am using Windows Host PC.

Regards

Gaurav

AM5728: LINUX: Unable to run X11 based applications on Weston using Xwayland

$
0
0

Part Number: AM5728

Hello,

I am using ti-processor-sdk-linux-am57xx-evm-05.03.00.07 on my AM5728 EVM. I have written a third party application that utilises SDL-X11 to draw a window.

I have successfully cross-compiled Xwayland (that is used for X clients under Wayland) for my AM5728-EVM. I have used instructions from this official website and did the cross-compilation. However, when I run the command:

root@am57xx-evm:/# /mnt/Xwayland 

a black screen appears on top of weston-display on EVM. I don't see any logs on command line but I am not able to run my X11 based application or even simple X11 demo apps (like xeyes, xclock) cross-compiled for AM5728 EVM.

As suggested in the above attached link, I even did changes to /etc/weston.ini file as - 

[core]
xwayland=true

[xwayland]
path=/mnt/Xwayland

and I restarted weston to get following error:

[02:19:21.709] weston 2.0.0
               http://wayland.freedesktop.org
               Bug reports to: bugs.freedesktop.org/enter_bug.cgi
               Build: 1.99.94-2-g4c4f13d configure.ac: bump to version 2.0.0 for the official release (2017-02-24 16:19:03 -0800)
[02:19:21.709] Command line: weston
[02:19:21.709] OS: Linux, 4.14.79-ge669d52447, #1 SMP PREEMPT Sat Apr 6 02:23:36 UTC 2019, armv7l
[02:19:21.709] Using config file '/etc//weston.ini'
[02:19:21.710] Output repaint window is 7 ms maximum.
[02:19:21.720] Loading module '/usr/lib/libweston-2/wayland-backend.so'
[02:19:21.735] Failed to load module: /usr/lib/libweston-2/wayland-backend.so: undefined symbol: weston_compositor_init
[02:19:21.735] fatal: failed to create compositor backend

I have no idea, which part I did wrong. But I'd really appreciate if you guys can help me out.

If you need any other information, please let me know.

Thanks & Regards,

Devashish

TDA2EG-17: TDA2EG-17 csi2Cal_multi_cam_view isscapture vpe link

$
0
0

Part Number: TDA2EG-17

Hello, expert:

Because of insufficient bandwidth resources, csi2cal multi cam view needs to delete VPE link. After trying to delete VPE link, the display is abnormal. Can isscapture delete VPE link? If VPE link can be deleted, how to modify it?

AM3352: Build error

$
0
0

Part Number: AM3352

linux-ti-staging_4.1.18 Kernel Error. Build errorLog data follows:
| DEBUG: Executing python function sysroot_cleansstate
| DEBUG: Python function sysroot_cleansstate finished
| DEBUG: Executing shell function do_configure
| NOTE: make -C /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work/beaglebone-oasys-linux-gnueabi/linux-ti-staging/4.1.18+gitAUTOINC+6b41ca0b94-l/git O=/home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work/beaglebone-oasys-linux-gnueabi/linux-ti-staging/4.1.18+gitAUTOINC+6b41ca0b94-l/build omap2plus_defconfig
| make: Entering directory `/home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source'
| make[1]: Entering directory `/home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work/beaglebone-oasys-linux-gnueabi/linux-ti-staging/4.1.18+gitAUTOINC+6b41ca0b94-l/build'
| GEN ./Makefile
| HOSTCC scripts/kconfig/zconf.tab.o
| In file included from scripts/kconfig/zconf.tab.c:2532:0:
| scripts/kconfig/zconf.lex.c_shipped: In function 'yy_get_next_buffer':
| scripts/kconfig/zconf.lex.c_shipped:934:16: error: 'EINTR' undeclared (first use in this function)
| scripts/kconfig/zconf.lex.c_shipped:1562:3: note: in expansion of macro 'YY_INPUT'
| scripts/kconfig/zconf.lex.c_shipped:934:16: note: each undeclared identifier is reported only once for each function it appears in
| scripts/kconfig/zconf.lex.c_shipped:1562:3: note: in expansion of macro 'YY_INPUT'
| In file included from scripts/kconfig/zconf.tab.c:2534:0:
| /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source/scripts/kconfig/confdata.c: In function 'conf_split_config':
| /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source/scripts/kconfig/confdata.c:920:17: error: 'ENOENT' undeclared (first use in this function)
| if (errno != ENOENT) {
| ^
| /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source/scripts/kconfig/confdata.c: In function 'conf_set_all_new_symbols':
| /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source/scripts/kconfig/confdata.c:1143:13: error: 'ERANGE' undeclared (first use in this function)
| errno = ERANGE;
| ^
| make[2]: *** [scripts/kconfig/zconf.tab.o] Error 1
| make[1]: *** [omap2plus_defconfig] Error 2
| make[1]: Leaving directory `/home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work/beaglebone-oasys-linux-gnueabi/linux-ti-staging/4.1.18+gitAUTOINC+6b41ca0b94-l/build'
| make: *** [sub-make] Error 2
| make: Leaving directory `/home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work-shared/beaglebone/kernel-source'
| ERROR: oe_runmake failed
| WARNING: exit code 1 from a shell command.
| ERROR: Function failed: do_configure (log file is located at /home/siva/002_Oasys_Backup/rtl8_source_poky_486/build/tmp/work/beaglebone-oasys-linux-gnueabi/linux-ti-staging/4.1.18+gitAUTOINC+6b41ca0b94-l/temp/log.do_configure.26255)
ERROR: Task 74 (/home/siva/002_Oasys_Backup/rtl8_source_poky_486/meta/recipes-kernel/linux/linux-ti-staging_4.1.bb, do_configure) failed with exit code '1'
NOTE: Tasks Summary: Attempted 6091 tasks of which 6090 didn't need to be rerun and 1 failed.
No currently running tasks (1479 of 6138)

CCS/TMDX654IDKEVM: IPC ping example

$
0
0

Part Number: TMDX654IDKEVM

Tool/software: Code Composer Studio

Hi All,

I am trying to upload and execute the ex11_ping example for the AM65X with BIOS. The example is located within C:\ti\ipc_3_50_03_05\examples\AM65XX_bios_elf. The following is true for my setup:

- CCS 910

- ipc_3_50_03_05

- pdk_am65xx_1_0_5

- processor_sdk_rtos_am65xx_6_00_00_07

Firsty, I am just attempting to upload the output binaries (built from the makefiles) onto the A53, R5f_0 and R5F_1 of the EVM, and execute it just in order to observe the expected output. For this, I know that the R5 core should be in split mode. I thus added the following line to my launch.js file:

...

...

// Connect targets
    dsDMSC_0.target.connect();
	
	//Add this to configure R5 cores in split mode
	dsDMSC_0.expression.evaluate("Change_MCUSS_to_SplitMode()");
	
    print("Loading DMSC Firmware...");

...

...

Now I can successfully upload the output binaries to the A53 and R5F_0 cores and the cores halts at the main() functions. However, when attempting to upload the output binaries onto the R5F_1 core, the core immediate goes into the "Suspended" state after the binaries have been uploaded, with some xdc error functions running:

A cant find source error is also displayed:

What can be the cause of this error? I have not yet rebuilt the IPC examples, thus the output binaries came standard with the IPC component installation. I believe the error may be with incorrect board initialization and setup? Otherwise it might be incompatible versions of XDCtools/PDK/IPC etc. Could anyone please assist? Unfortunately there is only reference for uploading this example on the AM537X IDK, which I have referred to.

I then tried to import the source of the examples into a CCS project, and tried to get it building from there (I did this successfully with the MessageQ example). This would be the ideal way to go rather than the makefiles. With this method I get all three cores to build successfully, however,I have only been successful in uploading the A53 core output onto the EVM. When uploading the output binaries onto the R5F_0 and R5F_1 cores, the application simply executes immediately and never halts at the main() function. Thus I seem to not be able to configure my project correctly for these cores. Is it possible for someone to prepare a CCS project for this example, so that this can be used as baseline for development of our IPC application? Otherwise, please advise as to what might be the cause of the core never halting at main?

Here are some steps I applied to get the project to build from within CCS:

- Imported all the .c, .h, .cmd, and .cfg files related to the specific cores into a CCS project specific to that core. I also imported the files located it the "shared" folder of the example

-added profile="debug" as cfgArgs at Project Properties->XDCtools->Advanced options

- changed the include declarations to include the header files from the correct dictory ( #include "SvrMsg.h" instead of #include "..Shared/SvrMsg.h" for example)

- within the shared configuration file, I commented out the proclist.toUpperCase.split() command, and rather defined the array manually:

/* configure processor names */
//var procNameAry = MultiProc.getDeviceProcNames();
var procNameAry = [ "HOST","R5F-0", "R5F-1" ];
//var procNameAry = cfgArgs.procList.toUpperCase().split(/\s+/);

From here the projects builds successfully, however uploading and execution fails (The MessageQ example however executes successfully). Here I have also attached the CCS projects I have managed to create thus far, with successful build:

(Please visit the site to view this file)

Any assistance would be greatly appreciated.

Regards,

Johnny

TDA2HG: Can we get the GPU load Statistics?

$
0
0

Part Number: TDA2HG

Hello,

We need to modify the CPU load for customer board ESD test based on VSDK 03 05. We can get CPU load Statistics? But, for GPU, can we get its load Statistics? If yes, how?

Thanks

Terence

[TDA4VM] About MCU RGMII Ethernet

$
0
0

Hi,

Actually, it is question about Ethernet PHY(DP83867ERGZ) and TDA4VM.

I have two qestions about RGMII RX signal (this signal goes to TDA4VM from ethernet PHY, DP83867E) of MCU ethernet PHY.

1. We have measured MCU ethernet RX signal, and this PHY signal is very unclean.

   Q) We can't upload the graph of rx signal measured, so could you please share EVM's MCU ethernet RX signal if you have?

2. We try to change the internal CLK delay, when we measured the 1G ethernet.

   Q) When we set the internal CLK delay as 2ns, this quality is not good. But, we set the internal CLK delay as 2.25ns or 2.5ns, the quallity is good.

        Is it normal? We think the margin is not enough.

Please give me the commnets about the above questions.

Thanks,

Manwoo Kim


CCS/AM5728: PRU program reload issue

$
0
0

Part Number: AM5728

Tool/software: Code Composer Studio

On Nov 21, I started the PRU program as normal, but suddenly I found PRU reload itself .(recreating rpmsg-pru:Channel)

This  happened very occasionally(not acceptable) and I can make sure the my PRU program won't restart itself(I mean software logic)

Here is the software logic 

	int ret=-1;
	Rtrunner runner;
	//Tag init(), Here I create the rpmsg-pru channel
	ret = runner.init();
	if(ret != 0)
	{

		return ret;
	}
	ret = runner.shakehands();
	if(ret != 0)
	{

		return ret;
	}
	do
	{
		ret = runner.run();
	}while(ret==0);

	return ret;

Below is the syslog. 

Nov 21 09:37:30 arm kernel: [169890.524749] ti-pruss 4b200000.pruss: unconfigured system_events = 0x0000000000030000 host_intr = 0x00000005
Nov 21 09:37:30 arm kernel: [169890.524794] remoteproc remoteproc4: stopped remote processor 4b234000.pru0
Nov 21 09:37:30 arm kernel: [169890.637111] remoteproc remoteproc4: powering up 4b234000.pru0
Nov 21 09:37:30 arm kernel: [169890.651946] remoteproc remoteproc4: Booting fw image am57xx-pru1_0-fw, size 271916
Nov 21 09:37:30 arm kernel: [169890.652491] ti-pruss 4b200000.pruss: configured system_events = 0x0000000000030000 intr_channels = 0x00000005 host_intr = 0x00000005
Nov 21 09:37:30 arm kernel: [169890.657632] virtio_rpmsg_bus virtio0: rpmsg host is online
Nov 21 09:37:30 arm kernel: [169890.657669] remoteproc remoteproc4: registered virtio0 (type 7)
Nov 21 09:37:30 arm kernel: [169890.657677] remoteproc remoteproc4: remote processor 4b234000.pru0 is now up
Nov 21 09:37:30 arm kernel: [169890.658683] virtio_rpmsg_bus virtio0: creating channel rpmsg-pru addr 0x1e
Nov 21 09:37:30 arm kernel: [169890.659612] rpmsg_pru virtio0.rpmsg-pru.-1.30: new rpmsg_pru device: /dev/rpmsg_pru30
.....
Nov 21 10:07:48 arm sh[565]: g_multi: waiting for /sys/class/net/usb1/
Nov 21 10:07:49 arm kernel: [171709.651591] virtio_rpmsg_bus virtio0: creating channel rpmsg-pru addr 0x1e Nov 21 10:07:49 arm kernel: [171709.651617] virtio_rpmsg_bus virtio0: channel rpmsg-pru:Channel 30:ffffffff:1e already exist Nov 21 10:07:49 arm kernel: [171709.660186] virtio_rpmsg_bus virtio0: rpmsg_create_channel failed Nov 21 10:07:49 arm kernel: [171709.774552] rpmsg_pru rpmsg_pru30: Device already open




AM5718: Audio clock configuration

$
0
0

Part Number: AM5718

Hi

I have custom board based on AM5718 with audio codec as tlv320aic3106.

Audio codec is operated in maser mode so bitclock and frame clock is to be generated by audio codec.

I feeds master clock (24.576MHz) through clockout2 pin. I want to operate codec at rate of 8,16,32,48KHz.

So I don't want to operate codec at PLL mode. only clock division is enough.

But when I probe the codec register using i2cdump -f -y -r 0x00-0xff 0 0x18.

PLL gets configured. How can I disable PLL ?.

For getting 48Khz reference, the Q value should be selected as 4.

How can I set Q =4?

How codec gets master clock(24.576MHz) information to compute 48KHz reference value?

is driver source(tlv320aic31xx.c) file needs modification to achieve these requirement? 

Regards

Satheesh Kumar S

TDA2PXEVM: Use AWR1642+mmWave-devpack+TSW1400 for data collection

$
0
0

Part Number: TDA2PXEVM

Hello we use AWR1642+mmWave-devpack+TSW1400 for data collection,there is a question that we can't detect target.Related pictures are as follows,

There is no target point in the display.Use case link as follwos,

NullSource -> Alg_RadarProcess_fft (DSP1) -> Alg_RadarProcess_pkDetect (DSP1) -> Alg_RadarProcess_beamForm (DSP1) -> Dup_0 (DSP1) -> Null
Dup_0 -> Alg_RadarProcess_drawObjs (DSP1) -> Display_Video

GrpxSrc -> Display_Grpx

What could be the problem? waiting for your reply

AM3352: GPMC problem

$
0
0

Part Number: AM3352

Hi champs,

customer has reached a reproducible state in which the GPMC controller does not behave in conformity with the system.

The first accesses proceed as expected correctly, only after a certain time the GMPC controller does not seem to have the need

to operate the Write_Enable signal.

Question:

What can cause the GPMC controller to go completely out of step and ignore the GPMC settings?

 

Below the representation of the error image:

 

Following is the analog measurement of the write_enable signal, the signal is not pulled low by the GPMC controller:

 

A review of the timing both before the situation and in case of error shows no manipulation of

GPMC timing what this could have been explained.

 

Another insight,

This could be avoided by adding an ARM-DSB (Data Synchronization Barrier) command.

The problem is that this is not possible in the DMA transfer.

 

Can you help us here?

AM5749: Interface with SD/HD-SDI Video Input Streams

$
0
0

Part Number: AM5749

Hello,

What is the recommended method to interface the AM5749 with SD/HD-SDI video input streams?

Example schematics or reference designs would be most welcome!

Thanks

Nick

Viewing all 17527 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>