Re: Linux 2.4.10-pre11 -- __builtin_expect

[Andreas Schwab]

James Antill <[e-mail]> writes:

|> Andreas Schwab <[e-mail]> writes:
|> 
|> > James Antill <[e-mail]> writes:
|> > 
|> > |>  unlikely() also needs to be...
|> > |> 
|> > |> #define unlikely(x)  __builtin_expect(!(x), 1) 
|> > |> 
|> > |> ...or...
|> > |> 
|> > |> #define unlikely(x)  __builtin_expect(!!(x), 0) 
|> > 
|> > This is not needed, since only 0 is the likely value and !! does not
|> > change that.
|> 
|>  Yes it is, given the code...
|> 
|> struct blah *ptr = NULL;
|> 
|> if (unlikely(ptr))
|> 
|> ...you'll get a warning from gcc because you are implicitly converting
|> from a pointer to a long.

You're right, seems like __builtin_expect is really only defined for pure
boolean values.

Andreas.

-- 
Andreas Schwab                                  "And now for something
[e-mail]				completely different."
SuSE Labs, SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5