Hi all, I am having trouble getting the LM3s6965 evaluation kit working with Eclipse/GCC/OpenOCD and FreeRTOS. I have followed all the instructions and successfully flashed the chip however nothing happens on the display and I can’t get it to respond to a web client… It seems to debug ok insofar as I can set break points and play/resume between them. I can ping the device with out a problem and get replies back (usually in about 2-10ms). This warning (at the bottom) appears when programming the device with OpenOCD programmer which makes me wonder if it’s a memory issue… Info: configuration.c:50 configuration_output_handler(): LMI Stellaris information: Chip is class 1 LM3S6965 vA.2 Info: configuration.c:50 configuration_output_handler(): did1: 0x1073402d, arch: 0x1073402d, eproc: ARMV7M, ramsize:64k, flashsize: 256k Info: configuration.c:50 configuration_output_handler(): master clock(estimated): 50000kHz, rcc is 0x1ce0380 Info: configuration.c:50 configuration_output_handler(): pagesize: 1024, lockbits: 128 0xffffffff, pages in lock region: 2 Info: configuration.c:50 configuration_output_handler(): Info: configuration.c:50 configuration_output_handler(): erased sectors 0 through 255 on flash bank 0 in 0s 483000us Warning: target.c:562 target_alloc_working_area(): not enough working area available I’m new to FreeRTOS and have developed mostly with PICs so any help would be greatly appreciated! Thanks, Tom --
I remember getting warnings when flashing but I don’t remember that warning about working area. If you can ping the board then the flashing must have worked though, right? This makes it odd you don’t see anything on the display then. Can you set a break point at the entry to vOLEDTask() in main.c, then step through the code to check that the display initialization is working. There is a switch statement that checks to see what type of board it is and initialize the display accordingly. Check it is going through the correct case in the switch statement. I think you need to get the display working first then move on to the Ethernet.
Hi Dave, Thanks for your reply. I confirmed the flashing worked by hard coding a loop to turn one of the board LEDs on which works fine.
Hi Dave, Thanks for your reply. I confirmed the flashing worked by hard coding a loop to turn one of the board LEDs on which works OK. I’ll try setting a break point and see if the vOLEDTask() function is initialising the display (not infront of the board at the moment…). Regards.
Ok…It seems that the demo board I have was updated (according to schematic – 8/09/07) to use the 128×96 RiT display not the OSRAM128x64 which is selected by the vOLEDTask switch statement if running a LM3s6965. However, forcing the software to select the RIT128x96 still does not make the display work but this is possibly due to the RIT128x96 code being for older/different boards. I’ll try and debug to make it work for this board and post updates… Regards
Getting somewhere… Managed to get the display working by downloading the DriverLib source from Luminary Micro and replacing rit128x96x4.c/.h in the project directory with those from the driverlibboardsek-lm3s6965_revc directory. (This board is obviously a Rev C). So now the graphic appears and then displays PASS [120ns] repeatedly. Still no web server running though? I can still ping without any problems… Regards.
Ok. It seems that I can access the web server using Lynx from a Linux box without any problem however if I try and access it from a Windows box within Firefox I get the "The connection was reset" page… Any ideas? Regards
That is good information – thanks. I was not aware the boards had been changed. Regards.
Strange, I’ve only ever tested it from a Windows machine using both IE and Firefox. I need to update to the same driver lib version as you to see if I can replicate this. Regards.
Copyright (C) Amazon Web Services, Inc. or its affiliates. All rights reserved.