Hightec TriCore toolchain sources (again)

Dec 05, 2006 3 Replies

Hello,



I would be very grateful if someone having recent sources of their toolchain would share them... Hightec-rt.com is avoiding any contact to me since I asked them to provide me the GPLed parts of their toolchain in source (GCC, binutils, ...). I respect that they've spent a lot of money and work into porting GCC/binutils/... to the TriCore architecture but still: this is GPL and should be available along with their binary-eval-versions.



A different question: is using anyone HighTec's port in a serious project? I mean is their toolchain useable beyond playing around (I'd like to use it for my graduation thesis implementation)?



Thanks for any help, Bernd



Yes, my employer (no names, no pack-drill) is using it. The compiler seems pretty solid overall. One thing to watch is, if you put on -O2 (ie optimisation), it can fall over itself, so that a bit-set operation uses the specific bit-set-mask instructions. Very reasonable, except that those instructions act on byte-wide data (appropriate for RAM), while the memory-mapped internal peripherals often require their registers are accessed as a full word. So if your code fails when optimised, look for this.

My major gripe would be GDB: very flaky, crashes often. In fairness, I don't know if this is GDB itself, or the JTAG driver module (afaik, by Infineon).

snipped-for-privacy@gmail.com wrote: ...

Thanks for the hint!

By the way: HighTec-rt has suddenly decided to reply and provide me the sources. They are, as far as I can tell, currently, fully in compliance with the GPL.

Greetings, Bernd

Another bug (reported to Hightec, but they haven't replied yet):

float func() { float a; unsigned long i = 147; a = (float)i; // So if your code fails when optimised, look for this.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required