- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Hi,
I work with M35 Concerto processor using CCS 6.1.2 and recently installed 6.2.
For me, the CCS always had and still has now strange problem when debugging C28 core: if I define a local float variable inside the function, during debug it appears in the Variables window, but cannot be evaluated. I ignored it for a while, but decided to fix it.
The local float variable is shown with the Type = unknown and the Value = unknown register 'R4HL'.
NOTE because a variable is defined as a local (inside a function) float, the R4 is (or supposed to be) a FPU register.
Checking Registers->FPU->R4H shows correct Value=0.0(Float Real). So, there is no register R4HL.
This name (ending on HL = High Low?) might be in the 64 bit DSPs, I guess?
Is there a setting in CCS allowing to correctly show local floats (i.e. "processor has 32-bit FPU") or it is a CCS bug?
Thanks, Igor
Hi, AartiG,
sorry it took me a while to create small demo. I finally found out that appearance of local floats depends on optimization setting
When optimization at default settings, local floats are shown correctly
I created new workspace and imported project into it; the zipped workspace with the project is attached.
No optimization – local floats are shown correctly
When use this optimization, local floats are not shown correctly
