So, I'm getting started with the CLA co-processor and I'm running into a strange issue.
I've built a .cla file into an existing project. But it doesn't seem to compile correctly in my project but it will in the sample project.
For instance:
Converting a long (stored in the CPU to CLA memory space).
float fTemp = 0;
__mdebugstop();
fTemp = 0;
fTemp = (float)xAdcInput;
tempF = tempF * 3.0;
xResult= (int)tempF;
In my program, the dis-assembly looks like this:
00009010: 78400000 MMOVIZ MR0, #0x0
00009012: 74C08C00 MMOV32 @0x8c00, MR0
00009014: 73C01506 MMOV32 MR0, @0x1506, UNCF
00009016: 74C08C00 MMOV32 @0x8c00, MR0
0000902a: 77804040 MMPYF32 MR0, #0x4040, MR0
0000902c: 74C08C00 MMOV32 @0x8c00, MR0
00009040: 7D600000 MF32TOI32 MR0, MR0
00009042: 74C01480 MMOV32 @0x1480, MR0
But in the example program (built from the fir32 example for the 28035 v1.30 ), the dis-assembly looks like the following:
0000900e: 74401500 MI32TOF32 MR0, @0x1500
00009028: 77804040 MMPYF32 MR0, #0x4040, MR0
0000902a: 7D600000 MF32TOI32 MR0, MR0
0000902c: 74C01480 MMOV32 @0x1480, MR0
Why is the same code different after compiling... I've gone through the project... the compiler rev is the same (TI v6.4.9) the cla0 options are the same.
I don't get why this is happening. What's wrong here? Why isn't my project using the MI32TOF32 assembly among other issues?
Thank you,
Kyle
