New Windows version of OpenOCD r93 available

Development Boards MicroMod
1

Hello,

a new compiled Windows version is available here:

http://www.yagarto.de/howto/openocd/index.html

Here you will find a small tutorial and the link for the OpenOCD.

Best regards,

Michael

2

Hi Michael.

Your YAGARTO is just great! The name does sound like one of Godzilla’s opponents. :lol:

I have the latest YAGARTO installed, with your update to OpenOCD, and have been testing it as well as SDK4ARM from Amontec and GCC For Dummies from Olimex. They all work well but YAGARTO certainly has a perceptible compilation speed advantage, which may become significant when a very large application is compiled.

I’ve just noticed that breakpoints set in assembler source files do not work. The Eclipse flags the line with the breakpoint icon but when that line is executed; it doesn’t stop the execution. I know this because I have a FIQ interrupt routine that flashes a LED if you push an on-board button.

C language source line breakpoints work just fine, in RAM and FLASH applications. When I get one of these breakpoints and then single-step into an assembler routine, I can set a breakpoint a few instructions ahead in the Eclipse disassembler window and this breakpoint will work.

I went back to Eclipse 3.1 and saw the same behavior. This is not a show-stopper but I’d sure like to get to the bottom of this.

Do you have an application that has an assembler routine wherein you can test my observation?

Cheers,

Jim Lynch

3

Hello Jim,

the comparison with Godzilla is very nice

Perhaps I should change the logo? :smiley:

If you found a speed difference between sdk4arm and YAGARTO,

this could be possible because sdk4arm use other binutils and

an other version of GCC. This could be a reason if you

find here a differences.

GCC for dummies use the Cygwin version of GNUARM, is this

correct? I have heard from other, that the MinGW/MSYS build

of YAGARTO is a little faster than GNUARM (Cygwin).

Do you have an application that has an assembler routine wherein you can test my observation?

Now, please can you send me your example?

Have you seen the problem with Insight too, or

is this only an Eclipse problem?

Best regards,

Michael