Part Number: TDA2PXEVM

Hello,

I would like to request assistance on the following issue I am encoutering when support AR0132RCCC sensor to Vision SDK Capture link:

Strong colored patches appears on white pixels in the YUV frame

I created a AR0132 sensor support following the VisionSDK_UserGuide_ISS_SensorFramework.pdf where sensor register settings used are from the reference pdk_01_10_02_07/packages/ti/drv/vps/src/devices/ar0132rccc/src/bspdrv_ar0132rcccReg.c.

When test pattern is enabled in sensor, the resulting YUV frame is as follows:

This confirms that the CSI2 lane configuration, dataformat, and BPP are correct since we can see that the colors matches the expected test pattern.

However, when test pattern is disabled and active pixels are used, there are strong red patches that appears over the white areas in the frame:

Since the camera sensor is using RCCC color array filter, it is expected to have a monochrome image with red and partial green shading.
But the strong red patches are not expected.

Unfortunately, I cannot figure out which part of the pipeline in ISS is causing this issue.
Because of this I would like to get support from experts or from someone who also faced similar issue.

Part Number: AM6548

Hi,

- Is Security Accelerator available on AM65x GP device?

- Is SA library provided on (Linux/RTOS) SDK?

Best regards,

HaTa.

Part Number: TMS320C6678

Hi,

   In the TMS320C6678 datasheet it is specified that the total visible window is 256MB and the application can define up to 64 segments. The segment size is 256B to 256MB.

   Even so, the sprugw8c.pdf manual specified for RXLEN_VAL field some values that seems to be out of range (over 256MB) probbale because of a typo in the documentation - the rxlen_val is displayed on 9 hex digits.

The rxlen_val is the Length value read or written from the Segment/Length array element indexed by The Rx
Address Segment Index Register.
If rxlen_val = 0-7, the segment size is 0;
If rxlen_val = 8, the segment size is 0x000000200
If rxlen_val = 9, the segment size is 0x000000400
If rxlen_val = 10, the segment size is 0x000000800
If rxlen_val = 11, the segment size is 0x000001000
If rxlen_val = 12, the segment size is 0x000002000
If rxlen_val = 13, the segment size is 0x000004000
If rxlen_val = 14, the segment size is 0x000008000

If rxlen_val = 15, the segment size is 0x000010000
If rxlen_val = 16, the segment size is 0x000020000
If rxlen_val = 17, the segment size is 0x000040000
If rxlen_val = 18, the segment size is 0x000080000
If rxlen_val = 19, the segment size is 0x000100000
If rxlen_val = 20, the segment size is 0x000200000
If rxlen_val = 21, the segment size is 0x000400000
If rxlen_val = 22, the segment size is 0x000800000
If rxlen_val = 23, the segment size is 0x001000000
If rxlen_val = 24, the segment size is 0x002000000
If rxlen_val = 25, the segment size is 0x004000000
If rxlen_val = 26, the segment size is 0x008000000
If rxlen_val = 27, the segment size is 0x010000000
If rxlen_val = 28, the segment size is 0x020000000 -- this hex values has 9 digits and the value is 512MBs ->0x0.2000.0000
If rxlen_val = 29, the segment size is 0x040000000 -- this hex values has 9 digits ...
If rxlen_val = 30, the segment size is 0x080000000 -- this hex values has 9 digits ...
If rxlen_val = 31, the segment size is 0x100000000 -- this hex values has 9 digits ...

Part Number: TDA2PXEVM

Hello,

I would like to request assistance in supporting AR0132RCCC sensor to Vision SDK Capture link.
Following the required changes specified in http://www.ti.com/lit/an/spraca8/spraca8.pdf to support RCCC in TDA2xx ISP,
I modified pdk_01_10_02_07/packages/ti/drv/vps/src/vpslib/hal/src/vpshal_issipipe.c. with the following:

gIssHalIpipeDefCfg is modified as follows:

• CFA Interpolation parameters
  vpsissIpipeCfaConfig_t member value:

/* CFA Configuration */
{
VPS_ISS_IPIPE_CFA_MODE_2DIRAC,
8192, 167U, 100U, 1023U, 16U, 10U, 0U,
{
16U, 25U, 25U, 20U, 500U, 20U, 24U
}
},

• Auto white balance : Disabled in ISS by setting ALGORITHMS_ISS_AEWB_MODE_NONE
• Auto exposure : Disabled in ISS by setting ALGORITHMS_ISS_AEWB_MODE_NONE
• Color Correction
  vpsissIpipeRgb2RgbConfig_t members with unity matrix:

/* RGB2RGB 1 Configuration */
{
{
{256U, 0, 0},
{0, 256U, 0},
{0, 0, 256U}
},
{0U, 0U, 0U}
},
/* RGB2RGB 2 Configuration */
{
{
{256U, 0, 0},
{0, 256U, 0},
{0, 0, 256U}
},
{0U, 0U, 0U}
},

• RGB2YUV conversion
  vpsissIpipeRgb2YuvConfig_t member value:

/* RGB2YUV Configuration */
{
0U, /* Brightness */
16U, /* Contrast */
{
{0, 256U, 0},
{256U, 0, 0},
{0, 0, 0}
},
{0U, 128U, 128U}
},

However, the resulting frame is not expected.

Using RGGB interpolations configuration and coefficients, the result is as follows:

[BEFORE] 
(Please visit the site to view this video)

Image is monochrome (expected) but colors (red and green) appears to be washed out.
Separate issue on red patches appears on white pixels reported on a separate thread here: e2e.ti.com/.../846005

[AFTER]

(Please visit the site to view this video)

The entire frame has blue overtone and the previous red patches are not blue.

Can anyone support me in configuring the ISS to convert RCCC RAW data into YUV properly?

Part Number: TMS320C6748

Hi,

I want to build the EDMA3 driver stand-alone application. And i have followed uesr guide in C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10\edma3_lld_2_12_05_30C\packages\ti\sdo\edma3\drv\docs.

This my steps:

INTERNAL_SW_ROOT ?= C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10\edma3_lld_2_12_05_30C

EXTERNAL_SW_ROOT ?= C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10

UTILS_INSTALL_DIR = C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10\xdctools_3_50_08_24_core(i am not sure about this variable)

# EDMA3 LLD
edma3_lld_PATH = C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10\edma3_lld_2_12_05_30C
include $(edma3_lld_PATH)/packages/component.mk

The output log said nothing to be done for .....

And i didn't find app in C:\ti-processor-sdk-rtos-omapl138-lcdk-05.02.00.10\edma3_lld_2_12_05_30C\examples

Please help check my configuration.

Part Number: AM3356

Tool/software: TI C/C++ Compiler

Hi，

When I run the GEL file,this error occasionally occurs.Sometimes it's normal.I always power up the board again and excute the "CPU Reset(HW)" before the load of GEL file.

CortxA8: Trouble Reading Memory Block at 0x44e10040 on Page 0 of Length 0x4: (Error -1205 @ 0x44E10040) Device memory bus has an error and may be hung. Verify that the memory address is in valid memory.

What causes this problem?

Best Regards,

Andy Zhou

Does the Ethernet issue also occur for the beaglebone industrial version by sancloud ("https://www.sancloud.co.uk/?page_id=1028&model_number=1214994? They use a different ethernet chip.

Part Number: AM5718

Hi,

I am using an AM5718 design which has most of IO rails running on 1V8 (except VDDSHV8 and VDDSHV11). I have maintained same no of decoupling caps given in EVM (TMDXIDK5718) V3_3D for V1V8_D.

I am using GPMC interface (16 bit data and 13 bit address line). I would like to know whether the decoupling provided is sufficient for 1V8 rail

As of now :

1V8  decoupling has  (VDDS18V & VDDS18V_DDR1 included)

30 X 100nF (HF cap 0201) and  6 X 10 uF

Please suggest.

Part Number: D3-3P-TDA3X-SK

Hello

I am using TDA3x RVP with OV10640 camera module.

I want to develop lane Crossing detection use case using this SoC in Vision SDK.

I have seen a default use case in Vision SDK on lane detection-

But this is using VIP (HDMI) capture. I have taken this as the reference and developed my own use case using the following links (for lane detection using ISS capture)-

IssCapture -> IssM2mIsp -> IssM2mSimcop -> Dup -> Merge_alg

Dup -> Alg_ImagePreProcess (EVE1) -> Alg_ImgPyramid -> Alg_LaneDetect (DSP1) -> Merge_alg -> Sync_alg (DSP1) -> Alg_ObjectDraw (DSP1) -> Display_Video

IssM2mIsp -> Alg_IssAewb

GrpxSrc -> Display_Grpx

The problem with the use case is that it is detecting only single side lane crossing and showing the warning on the display.

Is the reference use case developed in order to detect only single crossing?

Part Number: AM3359

Hi All,

We are planning to use DDR 1GB single chip with AM3359 processor. In all the reference design we could only 512MB DDR interface.

Could you please suggest any reference design or DDR3 1GB IC which TI has done testing?

Regards,

Hareesh.

Part Number: TDA4VMXEVM

Hi,

Env : VirtualBox , ubuntu (64bit)

Recently I try to import my caffe model , following the TI Deep Learning Library User Guide.

At the import phase , here's the import text file:

modelType          = 0
inputNetFile       = "../../test/testvecs/models/public/caffe/Model1.prototxt"
inputParamsFile    = "../../test/testvecs/models/public/caffe/Model1.caffemodel"
outputNetFile      = "../../test/testvecs/config/tidl_models/caffe/tidl_net.bin"
outputParamsFile   = "../../test/testvecs/config/tidl_models/caffe/tidl_param_"
inWidth = 256
inHeight = 192
inNumChannels = 3
inDataFormat = 1
resizeWidth = 256
resizeHeight = 192
inData = ../../test/testvecs/config/test.txt
postProcType = 0

According to the document inDataFormat = 1 (raw file).

If I want to use bmp file , which value inDataFormat should be set?

For inData the file content is

testvecs/input/aa.y testvecs/input/aa.y

"aa.y" is my raw image file with width = 256 , height = 192 ,channels=  3.

Segmentation_list.txt is like this:

testvecs/input/ti_lindau_I00000.jpg testvecs/input/ti_lindau_I00000.jpg

Here is the question : Why the jpg file is written twice?

Could you explain how the inData text file written?

Another question is after I run the import command , message showed in command line as below:

"Error in reading testvecs/input/aa.y testvecs/input/aa.y"

Neither original "airshow_mobilenet_256x256.y" files in the input folder and my raw image file showed the error message.

If I using the .bmp/.jpg file there is no error.

Do you have the document for details about inData?

The file imageNet_sample_val.txt

testvecs/input/airshow.jpg 895
testvecs/input/ti_lindau_I00000.jpg 557

The number follows .jpg = class_type?

Thank you.

Part Number: AM5726

I have some problem to create a bootable SD-card for an custom AM5726 board:

I have installed the TI SDK am57xx-evm-linux-rt-sdk-src-06.00.00.07.tar.xz ..

and startet <SDK>/bin/create-sdcard.sh

It has done the following steps:

1. found a 32 GB mmc device

2. create two partinitions

                                             Disk: /dev/sdc
                           Size: 29.8 GiB, 32010928128 bytes, 62521344 sectors
                                   Label: dos, identifier: 0x2724ff83

    Device          Boot             Start          End      Sectors      Size     Id Type
>>  /dev/sdc1       *                 2048       145407       143360       70M      c W95 FAT32 (LBA)    
    /dev/sdc2                       145408     62488575     62343168     29.7G     83 Linux
    Free space                    62488576     62521343        32768       16M

3. It had installed the pre-built images form SDK

If I start the device with SD-Card the device will restart because it found nothing to boot. The same on the TI-EVM board.

I can mount both partition and I can see the "rootfs" and "boot" with some content like MLO, u-boot.img and uEnv.txt and in "rootfs" a linux filesystem structure .. whats wrong?

Part Number: AM5728

Hi,

I run a baremetal DSP software that manage two ISR. The ARM is empty. The first one is an interrupt generated by the rising edge of the input capture ECAP1, and the second is generated by the comparaison event of the EPWM module. The ECAP ISR allows to force the synchronization of the EPWM and to measure in the same time the frequency of the input signal.

I measured the time between the rising edge and the effective input into the ISR (by toggling a GPIO), which is around 3µs. I don't understand why...

Could you help me to figure out how to minimize this delay, that should be close to zero?

Thanks,

Sylvain.

Hi All, 

I am trying to create a basic application for TDA4x using Opencv on A72 .

For that i followed the document "OpenCV 3.1 cross compilation on Linux A15" present under "Vision_sdk\ti_components\open_compute\opencv\opencv-3.1.0\docs" 

I am able to successfully cross-compile the opencv library.

My makefile for the basic demos application is :

###################################

ifeq ($(TARGET_CPU),A72)
ifeq ($(TARGET_OS),LINUX)

include $(PRELUDE)

TARGET := vx_app_MSI
TARGETTYPE := exe
#CSOURCES := $(call all-c-files)

IDIRS += $(VISION_APPS_PATH)/apps/basic_demos/app_tirtos/tirtos_linux/mpu1
IDIRS += $(VISION_APPS_PATH)/apps/basic_demos/app_tirtos/tirtos_linux/c66x_1
IDIRS += $(VISION_APPS_PATH)/apps/basic_demos/app_vx_msi/include
IDIRS += $(VISION_APPS_PATH)/apps/basic_demos/app_vx_msi
include $(VISION_APPS_PATH)/apps/concerto_a72_inc.mak
LDIRS += $(TIOVX_PATH)/out/J7/A72/LINUX/release/opencv_lib
EXT_LIBS += opencv_imgproc
EXT_LIBS += opencv_highgui
EXT_LIBS += opencv_core
CPPSOURCES := \
main_linux_arm.cpp \

include $(FINALE)

endif
endif

###########################

My main_linux_arm.cpp contains :

int main(int argc, char *argv[])
{
int status = 0;
cv::Mat m(cv::Size(12,23),CV_8UC1);

m = 120;
status = appInit();
if(status==0)
{
vx_uc_MSI_sample();
appDeInit();
}

return 0;
}

While compiling i get the following errors:

Linking /home/anshuman/psdk_rtos_auto_j7_06_00_01_00/vision_apps/out/J7/A72/LINUX/release/vx_app_MSI.out
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: /home/anshuman/psdk_rtos_auto_j7_06_00_01_00/tiovx/lib/J7/A72/LINUX/release/libvx_app_uc_MSI.a(uc_MSI.o): in function `callOpenCVGaussianBlur(_vx_image*, _vx_image*)':
uc_MSI.cpp:(.text+0x308): undefined reference to `cv::Mat::create(int, int const*, int)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x324): undefined reference to `cv::Mat::operator=(cv::Scalar_<double> const&)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x50c): undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x51c): undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x544): undefined reference to `cv::GaussianBlur(cv::_InputArray const&, cv::_OutputArray const&, cv::Size_<int>, double, double, int)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x59c): undefined reference to `cv::fastFree(void*)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x5e4): undefined reference to `cv::fastFree(void*)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x62c): undefined reference to `cv::fastFree(void*)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x6c4): undefined reference to `cv::Mat::deallocate()'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x6d0): undefined reference to `cv::Mat::deallocate()'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text+0x6dc): undefined reference to `cv::Mat::deallocate()'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: /home/anshuman/psdk_rtos_auto_j7_06_00_01_00/tiovx/lib/J7/A72/LINUX/release/libvx_app_uc_MSI.a(uc_MSI.o): in function `cv::Mat::~Mat()':
uc_MSI.cpp:(.text._ZN2cv3MatD2Ev[_ZN2cv3MatD5Ev]+0x5c): undefined reference to `cv::fastFree(void*)'
/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: uc_MSI.cpp:(.text._ZN2cv3MatD2Ev[_ZN2cv3MatD5Ev]+0x6c): undefined reference to `cv::Mat::deallocate()'
collect2: error: ld returned 1 exit status
concerto/finale.mak:204: recipe for target '/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/vision_apps/out/J7/A72/LINUX/release/vx_app_MSI.out' failed
make: [/home/anshuman/psdk_rtos_auto_j7_06_00_01_00/vision_apps/out/J7/A72/LINUX/release/vx_app_MSI.out] Error 1 (ignored)

Please suggest what i am missing.

I even tried using the prebuilt opencv library from "vision_sdk_03_06\ti_components\open_compute\opencv\opencv-3.1.0\libs_linux".

Regards,

Anshuman

Part Number: AM5728

Hi,

If power down sequence couldn’t comply the requirement described in the data manual, what does it impact AM5728 ?

In the customer’s system, there is a possibility of shut down 24V main power suddenly. In this case, no one knows how 5V and 3.3V into PMIC TPS659037 power down and output power from PMIC to AM5728 also are unpredictable.

Could you tell us what kind of risk they should consider ?

Thanks and regards,
Hideaki 

Part Number: 66AK2G12

Tool/software: Code Composer Studio

Dear Sir,

Currently, We develop our product using 66AK2G12, And select the DDR3 parts.

The DDR3 is MT41K256M16TW-107. But the parts can not select in the K2 DDR3 Register Calc v1p60.xlsx.
So please tell us about how to set the Datasheet in the K2 DDR3 Register Calc v1p60.xlsx.

The MT41K256M16 125 can select, The parts is similar.
But the parameters is different.

Best regards.

Part Number: TMS320C6678

Hi,

    1. We have the following description for TXIGMASK field of Tx Address Overlay Control Register (Base Address + 0x1c):

"The txigmask is used to create the mask that is logically anded to the incoming address to create the address sent to the remote."

   Can you please explain how a TX mask can be applied to an incoming address? Incoming from who if the mask is for TX direction? Shouldn't we speak about transmitted address?

   2. In the "C66x KeyStone Training HyperLink" document, page 27) it says that txigmask = depends on the maximum segment size and as example we have: 10 → 0x07FF_FFFF (2 segments)

   What is the relation of segment size and txigmask value with the number of segments from parentheses?

Part Number: PROCESSOR-SDK-AM335X

Hi,

I am using SKAM335x board. In what format LCD displays character? 

Suppose I want to display "LCD Testing", in what format I have to give values to buffer.

I am using TI Starterware on windows host PC.

Regards

Gaurav

Part Number: 66AK2G12

Tool/software: Code Composer Studio

Dear Sir,

Currently, We develop our product using 66AK2G12, And check about the DDR3 registers.

We select the DDR3 parts, this register automatically set value, We want to know about that  which bit need to set by software code. 

Please tell us below questions about DDR_PHY_PGCR1.

1. What is Loopback Mode. How to use this? Should we set?  The Loopback  is described on the LBMODE,LBGDQS,LBDQSS,IOLB

2. What is VT Calculation?  How to use this? Should we set?  THe VT is described on the INHVT DLDLMT

3. When the PHYHRST set  to 0? How to use this?

4. When the FDEPTH and LPFDEPTH,LPFEN use?

5. When the MDLEN use?

6. When the WLSELT and WLSTEP and WLMODE use? How to use ?

7. When the PDDISDX use?

Best regards.

Part Number: AM5728

Hello Specialist,

May I ask do we have Absolute Maximum Operating Junction Temperature for AM5728?

I can't find it in Datasheet.

Thank you.

Lin HsinYu