BSEC 2.0.6.1 - Cortex-M33F binary size of libalgobsec.a

[knieriem]

Hi,

looking at the binary sizes of libalgobsec.a, compiled for different architectures using GCC, I noticed that the size for Cortex_M33F (49453 B) is notably larger than the size for the comparable Cortex_M4F (27841). Both architectures support an FPU, and similar instruction sets, from what I can see on Wikipedia and in the  ARM architectures and processors forum; so I would expect similar compact code.

In case of the M4F, the size increase, in comparison to the non-FPU version, is only some hundred bytes, in case of the M33F it is around 12 KiB:

Cortex_M33    27337
Cortex_M33F 49453
Cortex_M4       27225
Cortex_M4F     27841

Is there a difference in GCC's compiler options (like, a different optimization choice) or some other conditions,  that explain the binary size increase on M33F?

[knieriem]

Hi BSTRobin,

yes, I'm using the M33F library on an EFM32PG22 target, without any problem so far.

So, if the size of the M33F-lib.a could be made more similar to the size of the other Cortex-M libs,
that would be nice-to-have -- the current size is not a release blocker.

[BSTRobin]

Hi knieriem,

We add -O2 to compiling options, the size of M33F library decreases about 27KB, close to other platform libraries.

Could you run attached library and give feedback?

[BSTRobin]

Hi knieriem,

Could you give feedback with optimized BSEC library if it run well or not on your side?

[knieriem]

Hi BSTRobin,

thank you for enabling optimization and for providing the adjusted library.

I have integrated the attached library into our firmware. So far it works without any problem, the software is reaching IAQ accuracy 3 as before,
and values look plausible.

Is it already planned when there will be an official release?

[BSTRobin]

Hi knieriem,

Thanks for your feedback. Now it on going to release to official website.