Naked attribute in gcc

Kernel
1

anonymous wrote on Tuesday, April 27, 2010:

GCC documentation (http://gcc.gnu.org/onlinedocs/gcc/Function-Attributes.html) states in 6.29 Declaring Attributes of Functions

"naked

Use this attribute on the ARM, AVR, IP2K, RX and SPU ports to indicate that the specified function does not need prologue/epilogue sequences generated by the compiler. It is up to the programmer to provide these sequences. The only statements that can be safely included in naked functions are asm statements that do not have operands. All other statements, including declarations of local variables, if statements, and so forth, should be avoided. Naked functions should be used to implement the body of an assembly function, while allowing the compiler to construct the requisite function declaration for the assembler."

In the source of FreeRTOS for AVR I have seen the following:

 void vPortYield( void ) attribute ( ( naked ) );
 void vPortYield( void ) {
   portSAVE_CONTEXT();
   vTaskSwitchContext();
   portRESTORE_CONTEXT();
   asm volatile ( "ret" ); }

portSAVE_CONTEXT() and portRESTORE_CONTEXT() are macroses with asm but vTaskSwitchContext is a function

My question is: Is the port for AVR of FreeRTOS is correct, or did I miss something?

2

rtel wrote on Tuesday, April 27, 2010:

I presume your point is “the naked function makes a C function call, but the GCC manual says you should only use asm statements”.  If so, then you are correct that in this case it goes against the advice, but it works fine, as far as I am aware, at all optimisation levels.  In some ARM ports certain versions of GCC require the C function call to be replaced with a single asm statement that does exactly the same thing, but I have never known this to be the case with the AVR.

Are you having a problem with the generated code?

Regards.

3

anonymous wrote on Tuesday, April 27, 2010:

Thank you for the answer.

No. I have not problem with the generated code.

My aim was to clarify my understanding of naked functions in GCC.