Part Number: AM6548 Tool/software: Linux Hello, I need to get an RGMII connection between PRU0/ICSSG1 of the AM6548 and a DP83TC811 PHY to work. I'm running a debian Linux with kernel version 4.14.79 on the AM654x EVM. I'm using the PHY driver which can be found here: https://github.com/torvalds/linux/blob/00f553660a7f524a3b913cb5cf6c5f33f9638f09/drivers/net/phy/dp83tc811.c I can't get the RGMII connection up and runnig with 25 MHz clock/100 Mbit. No matter what connection speed I enforce within the driver, I always get 125 MHz RGMII clock from the AM6548 . When manually changing the RGMII0_GIG_IN bit of the ICSSG_RGMII_CFG register to 0, the RGMII output clock of the AM6548 is set to 25 MHz, but there is no communication going on towards the PHY. When the PHY on the other hand receives an ethernet packet, it can be correctly received by the linux OS. I'm currently running out of ideas how to fix this. It would be great if anyone got some useful tips how to solve this problem. Best Regards, Björn
↧
Forum Post: Linux/AM6548: AM6548 PRU, RGMII with 100 MBit for DP83TC811 PHY won't work
↧
Forum Post: TDA2PXEVM: Configure HDMI pins (Documentation for HDMI_PAD_CONFIG_CONTROL)
Part Number: TDA2PXEVM With the TDA2PXEVM I'm giving a HDMI signal (HDMI OUT) to a custom converter board (DVI to PAL). The converter board detects every other HDMI source i have but not the TDA2PXEVM . The TDA2PXEVM output is displayed by different diplays and grabbers. So, both of them work, but not togehter. I tried different modes (CEA, VESA) and resolutions. I want to know what possibilities I have to configure the HDMI output pins (e.g. voltage difference for the differential signal). Is there a documentation for this register: HDMI_PAD_CONFIG_CONTROL 0x5804030C? I can find it being used in the code (pdk_01_09_00_17, vpshal_dssHdmi.c) but not in Technical Reference Manual (Literature Number: SPRUIF0B May 2017–Revised September 2018).
↧
↧
Forum Post: Linux/AM4372: Linux Boot crashes after : "Started D-Bus System Message Bus"
Part Number: AM4372 Tool/software: Linux Hello, we are testing our custom-board on different temperatures, At 75 °C the Kernel-Boot crashes after the Message "Started D-Bus System Message Bus" or after "TI DP83822 10/100 Mbps PHY 4a101000.mdio:01: attached PHY driver [TI DP83822 10/100 Mbps PHY] (mii_bus:phy_addr=4a101000.mdio:01, irq=POLL)". I have appended the Logfiles. (Please visit the site to view this file)(Please visit the site to view this file) Do you have any suggestions what can be reviewed? Thanks four your help. M.Ucar
↧
Forum Post: RE: AM5708: SD mode boot
Hello user4205571, Have you managed to solve this issue? Best regards, Kemal
↧
Forum Post: RE: CCS/OMAP-L137: JTAG connect issue with DSP
Are you using a custom board or the EVM kit?
↧
↧
Forum Post: RE: Linux/AM5706: About Package Installing on Linux PSDK v05.02.00.10
Hello J-breeze, I’ve not heard back from you in a month. I hope we were able to help. I’m going to close the thread now and mark the issue resolved. If you have any follow-up questions, please reply to this thread or start a new thread. Best regards, Kemal
↧
Forum Post: RE: Linux: AM335x Quad-USB-to-UART Bridge (CP2108) Communcation Freeze
Hi Ben, I tested the same kernel with your .config, but was unable to see the problem. Here is the test script I used, ran for almost 30min until now. #!/bin/sh while true; do for i in 0 1 2 3; do cat /dev/ttyUSB${i} || exit 1 & sleep 1 kill $(pidof cat) || exit 2 sleep 1 done done By the way, attached below is a kernel patch fixing a cppi driver bug. I don't think it is related, but can you please give it a try? (Please visit the site to view this file)
↧
Forum Post: RTOS/TDA2SX: How to change Capture and Display format from 30fps to 60fps
Part Number: TDA2SX Tool/software: TI-RTOS Hi, I am having TDA2xx EVM and OV10635. I can able to capture and display in 30fps frame rate with 1280 x 720 resolution , Now i want to change format for capture and display in 60fps . According to OV10635 datasheet, i can able to capture 60fps frame rate in 640x400 / CIF / QVGA format only. I am trying to do in 640 x 400 resolution. let me know, what are the places have to change to achieve 60fps. Capture : OV10635 camera Display : 640 x 400 resolution Thanks in advance, Regards, ARUNKUMAR
↧
Forum Post: RE: Linux: AM335x Quad-USB-to-UART Bridge (CP2108) Communcation Freeze
Do you have a USB hub on your custom board?
↧
↧
Forum Post: RE: RTOS/TDA2SX: How to change Capture and Display format from 30fps to 60fps
Hi Arun, I am sorry i did not get the question. Are you saying you are able to capture 640x400@60fps? Rgds, Brijesh
↧
Forum Post: RE: Linux/AM5708: Qt 5.12 Web engine High CPU usage
Hi Manisha, Thanks for the reply. I am going through patches. Meanwhile i am having below query, I am not using wayland and weston. instead i am using eglfs and it is directly displaying in full screen mode. I hope it will not cause any issues with GPU usage. There is no any limitation like GPU only working with wayland right? Thanks, vivek panchal
↧
Forum Post: RE: RTOS/AM3352: Reading USB DevCtl crashes system
Kevin, So, program 1 without CDC, program 2 with CDC - Both programs 1 and 2 functioning well - Program 1: USB0 Core registers starting at 0x47401400 ARE readable in Code Composer's memory browser window - Program 2: USB0 Core registers starting at 0x47401400 ARE read as ?????? in Code Composer's memory browser window Do you have MMU setup for 0x47401400 region in program 2? Just double check? Regards, Eric
↧
Forum Post: RE: TMS320C6655: EDMA CC missed events
Jim, Any chance to test the suggestions? Regards, Eric
↧
↧
Forum Post: RE: Linux/AM5708: Qt 5.12 Web engine High CPU usage
Hi Vivek, Yes, you don't have to use Wayland/Weston to use GPU. eglfs enables you to offload the graphics acceleration on GPUs. Regards, Manisha
↧
Forum Post: RE: TMS320C6678: C6678 missed completion interrupts for concurrent EDMA transfers
Roman, Thanks for the update! Waiting more info from Wolfgang when he back to office ... Regards, Eric
↧
Forum Post: RE: CCS/AM5708: TSCL and TSCH do not work on DSP
Hi, when I run the code on DSP board (AM57), TSCL and TSCH are not work .>>>>> Do you mean they are always zero? int imagesize = 10; unsigned char * p1 = new unsigned char[imagesize]; unsigned char * p2 = new unsigned char[imagesize]; beg_count = _itoll(TSCH,TSCL); end_count = _itoll(TSCH,TSCL); overhead = end_count - beg_count; You need to "initialize" it to let it running. To do this, TSCL = 0; beg_count = _itoll(TSCH,TSCL); end_count = _itoll(TSCH,TSCL); Regards, Eric
↧
Forum Post: RE: Linux/DRA77P: ULPI USB interface
Hi Padmesh, Thanks for the device tree files. The entries need for usb3 and usb4 for ulpi support are clean. No issue there.. The only suspect here is the code update done to the phy-tusb1210.c file. Have you ruled out the possibility of the module getting loaded only for the first instance of the usb that useds ulpi? Have you checked if the code is trying to write to both the tusb1210 chip to set the ULPI_OTG_CTRL register.. Also, Do you know why the probe function is called twice? [ 95.528841] TUSB1210_Probe_Function Called! [ 95.534921] TUSB1210_SetMode Function Called! = 352781393 [ 95.541258] ULPI_OTG_CTRL = 103 [ 95.547200] TUSB1210_Probe_Function Called! [ 95.553055] TUSB1210_SetMode Function Called! = 352781393 [ 95.559391] ULPI_OTG_CTRL = 103 >We are not able to build the .config with CONFIG_PHY_ TUSB1210 and CONFIG_USB_ULPI_BUS as y as it changes to m when make linux command is run. Are you manually editing the .config file? Did you try "make menuconfig" to set the above? I see that you have created a new thread: https://e2e.ti.com/support/processors/f/791/t/793747 . Since the original query was answered and you are able to get it work with at least one instance, lets close this thread and continue the discussion on the new thread. Please respond to above query on the new thread. -Praveen
↧
↧
Forum Post: RE: RTOS/AM3352: Reading USB DevCtl crashes system
Eric, Where are the MMU registers and where can I find the documentation on them? Searching the Technical Reference presents nothing useful when searching for 'Memory Management Unit' but searching for 'mmu' produces too many hits for words like 'communication'. Kevin
↧
Forum Post: RE: Linux/DRA77P: Identifying two ulpi phys
Old thread: e2e.ti.com/.../793747 Hi Padmesh, Thanks for the device tree files. The entries need for usb3 and usb4 for ulpi support are clean. No issue there.. The only suspect here is the code update done to the phy-tusb1210.c file. Have you ruled out the possibility of the module getting loaded only for the first instance of the usb that uses ulpi i/f ? Have you checked if the code is trying to write to both the tusb1210 chip to set the ULPI_OTG_CTRL register.. Note that the current support for setting the ULPI_OTG_CTRL register is done via the phy-ulpi.c file. But as we understand, looks like it is supported for dr_mode-=otg. Can you give a try setting the mode to OTG for both usb3 and usb4 and see what is the behaviour? Also, Do you know why the probe function is called twice? [ 95.528841] TUSB1210 _Probe_Function Called! [ 95.534921] TUSB1210 _SetMode Function Called! = 352781393 [ 95.541258] ULPI_OTG_CTRL = 103 [ 95.547200] TUSB1210 _Probe_Function Called! [ 95.553055] TUSB1210 _SetMode Function Called! = 352781393 [ 95.559391] ULPI_OTG_CTRL = 103 Are you manually editing the .config file? Did you try "make menuconfig" to set the above? -Praveen
↧
Forum Post: RE: TMS320C6657: Cache coherence
Hi, You may refer to the Processor SDK RTOS C6657: pdk_c665x_2_0_XX\packages\ti\drv\pcie\example\sample\src\pcie_sample.c, void cache_invalidate (void *ptr, int size): let say you have another master writes into the DDR physical memory, your CPU tries to read those new data, it only reads from its cache and it always get the old data. So you need to call this API to invalidate this to mark the cache line dirty, then CPU will read from the physical memory to get the new data. void cache_writeback (void *ptr, int size): let say your CPU wants to write some data into DDR for another master to pick it up. It only writes into the cache, you need to call this API, then data will be written into the physical memory. Regards, Eric
↧