Re: [PATCH] only use 48-bit lba when necessary

[Matt Mackall <mpm@selenic.com>]

On Fri, Apr 04, 2003 at 12:02:00PM -0500, Chuck Ebbert wrote:
> Juan Quintela wrote:
> 
> 
> >Reason is that:
> >
> >if (expr)
> >   var = true;
> >else
> >   var = false;
> >
> >is always a bad construct.
> >
> >var = expr;
> >
> >is a better construct to express that meaning.
> 
> 
>  Yes, but:
> 
>    if (expr1 && expr2)
>       var = true;
>    else
>       var = false;
> 
> is usually better turned into something that avoids jumps
> when it's safe to evaluate both parts unconditionally:
> 
>    var = (expr1 != 0) & (expr2 != 0);
> 
> or (if you can stand it):
> 
>    var = !!expr1 & !!expr2;

Such ugly transformations are a job for compiler writers and may
occassionally be acceptable in some critical paths. The IO path, which
is literally dozens of function calls deep from read()/write() to
driver methods, does not qualify.

FYI, GCC as of 3.2.3 doesn't yet reduce the if(...) form to branchless code
but the & and && versions come out the same with -O2.

-- 
Matt Mackall : http://www.selenic.com : of or relating to the moon