Re: Code optimization: warning for code that hangs

[David Brown <david dot brown at hesbynett dot no>]

On 24/06/12 23:54, Robert Dewar wrote:
> On 6/24/2012 12:09 PM, Ángel González wrote:
> > "Peter A. Felvegi" writes:
> > > > > My question is: wouldn't it be possible to print a warning when a jmp
> > > > > to itself or trivial infinite recursion is generated? The code
> > > > > compiled fine w/ -Wall -Wextra -Werror w/ 4.6 and 4.7.
> > Note that if the target architecture is a microcontroller, an endless
> > loop can be a legitimate way to finish / abort the program.
>
> But not an infinite recursion! And an endless loop is such a rare
> case that it deserves a warning, it's a false positive in this case,
> so what?

An endless loop is not rare - as Ángel says, it is standard practice in 
embedded systems.  Typical systems can have more than one such loop - 
the main loop is usually endless, and you often have some sort of "halt" 
or "wait for watchdog reset" endless loop.

Most such loops are not simple "jump to yourself" instructions - but 
some are (with other actions being triggered by interrupts).

In the majority of cases, however, it should be possible to mark the 
function in question as "noreturn".

In fact, when I think about it, perhaps a "-Wsuggest-attribute=noreturn" 
will give people the warning they want?  After all, if the compiler can 
work out that the best code is a "jump to yourself" instruction, it 
should also have noted that the function does not return.