CLA accepts the C28x pragmas except for FAST_FUNC_CALL.
The far and ioport keywords are not recognized.
Access to the 'MMOV32 MSTF,mem32' and 'MMOV32 mem32,MSTF' instructions is provided using the cregister keyword. To access these MSTF instructions include the following declaration:
extern cregister volatile unsigned int MSTF;
The intrinsics listed in Table 10-1 are supported. Additionally, the run-time library functions abs() and fabs() are implemented as intrinsics.
| Intrinsic | Assembly Instruction(s) | Description |
|---|---|---|
| __mdebugstop() | MDEBUGSTOP | Debug stop |
| __meallow() | MEALLOW | Enable write access to EALLOW registers |
| __medis() | MEDIS | Disable write access to EALLOW registers |
| float __meinvf32(float x); | MEINVF32x | Estimate 1/x to about 8 bits of precision. |
| float __meisqrtf32(float x); | MEISQRTF32x | Estimate the square root of 1/x to about 8 bits of precision. The Newton-Raphson method is used to approximate the square root. |
| short __mf32toi16r(float src); | MF32TOI16Rdst,src | Convert double to int and round. |
| unsigned short __mf32toui16r(float src); | MF32TOUI16Rdst,src | Convert double to unsigned int and round. |
| float __mfracf32(float src); | MFRACF32dst,src | Return the fractional portion of src. |
| short __mgeq( signed int x, signed int y); | MCMP32dst,src | Return 1 if x is greater than or equal to y. Otherwise, return 0. |
| short __mgequ( unsigned int x, unsigned int y); | MCMP32dst,src | Return 1 if x is greater than or equal to y. Otherwise, return 0. |
| short __mgt( signed int x, signed int y); | MCMP32dst,src | Return 1 if x is greater than y. Otherwise, return 0. |
| short __mgtu( unsigned int x, unsigned int y); | MCMP32dst,src | Return 1 if x is greater than y. Otherwise, return 0. |
| short __mleq( signed int x, signed int y); | MCMP32dst,src | Return 1 if x is less than or equal to y. Otherwise, return 0. |
| short __mlequ( unsigned int x, unsigned int y); | MCMP32dst,src | Return 1 if x is less than or equal to y. Otherwise, return 0. |
| short __mlt( signed int x, signed int y); | MCMP32dst,src | Return 1 if x is less than y. Otherwise, return 0. |
| short __mltu( unsigned int x, unsigned int y); | MCMP32dst,src | Return 1 if x is less than y. Otherwise, return 0. |
| float __mmaxf32( float x, float y); | MMAXF32dst, src | Return the maximum of two 32-bit floating point values. If src>dst, copy src to dst. |
| float __mminf32( float x, float y); | MMINF32dst, src | Return the minimum of two 32-bit floating point values. If src<dst, copy src to dst. |
| __mnop() | MNOP | CLA no operation |
| __msetflg(unsigned short flag, unsigned short value) | MSETFLGflag,value | Set/clear flag(s) in the MSTF register. The flag is a bit mask to indicate which bits to modify. The value provides the values to assign to these bits. See the CLA Reference Guide (SPRUGE6) for details about the MSETFLG instruction and the MSET register.
This example sets the RNDF32 flag (bit 7) to 0, the TF flag (bit 6) to 0, and the NF flag (bit 2) to 1. The 0b prefix is a GCC language extension indicating that these are binary numbers. __msetflg(0b11000100, 0b00000100);
|
| void __mswapf(float &a, float &b); | MSWAPFa, b | Swap the contents of a and b. |
| float __sqrt(float x); | MEISQRTF32x | Estimate the square root of 1/x to about 8 bits of precision. The Newton-Raphson method is used to approximate the square root. This is an alias for the __meisqrtf32 intrinsic. |
Comparison intrinsics, such as __mlt and __mleq are provided for CLA to avoid possible overflows when performing signed integer comparisons.
The following intrinsics are available only within CLA2 background tasks. The background task has the lowest priority task and so will always be interrupted when another interrupt is available. These intrinsics can be used to temporarily prevent such interrupts from suspending the background task.