[sdiy] Some Audio DSP prototypes

sleepy_dog at gmx.de sleepy_dog at gmx.de
Wed Apr 20 21:10:04 CEST 2022


Mike Bryant:
> We actually still use CubeIDE to create the clock tree and I/O list.
> But if you look for one of the most read threads on the STM Forum it’s
> one of mine pointing out that the use of gdb with CubeIDE was borked.
> Some people get it to work, whilst for others it simply won’t.
> Something wrong with the installer possibly but STM never seem to fix
> bugs.

I've only done a couple projects now with the newer CubeIDE, only had a
couplke minor issues. I wonder what you mean by "GDB won't work". What
does that look like when it happens?
Do you have (much more than default) complicated linker scripts or
something?
(the closest I go to having problems was due to something like that,
but, well it wasn't really a problem or "doesn't work at all").

I like about CubeIDE that looking at register values, decoded bits in
registers with names form the manual, "just works" for a supported MCU,
and other little very helpful things like that.
(although I begged ST, with some others, to finally upgrade to a newer
GCC version, so we have at least C++ 17 for some months now, yay! ;-) )

> But main issue is the last time I looked STM still hadn’t integrated
> Clang into CubeIDE which makes it useless for any application where
> performance is essential.
>

Do you have anything to point to, to substantiate this idea of Clang's
general superiority?
Could it be application specific?

E.g. "For the vast majority of benchmarks the LLVM Clang vs. GCC
performance was quite close"
and "The NCNN neural network inference library from Tencent was
performing hugely better when built under the GCC compiler"
https://www.phoronix.com/scan.php?page=article&item=apple-m1-compilers

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20220420/62ae0e05/attachment.htm>


More information about the Synth-diy mailing list