On Thu, Aug 10, 2017 at 6:34 PM, Luc Van Oostenryck
<luc.vanoostenryck@xxxxxxxxx> wrote:
>
> Absolutely not.
> Anyway, currectlyeach distinct value correspond to a *single/unique*
> pseudo. IN other words, there is no need to do any kind of CSE on
> *values*, they are already kinda pre-CSEed at their creation
> (because a constant value is nothing more than this value).
I see. value_pseudo() has the hash for all constant values.
But float is not covered.
#define PI 3.1415
double a;
void foo(void)
{
a = a*PI + a*PI*5;
}
./test-linearize /tmp/pi.c
/tmp/pi.c:3:8: warning: symbol 'a' was not declared. Should it be static?
/tmp/pi.c:4:6: warning: symbol 'foo' was not declared. Should it be static?
foo:
.L0:
<entry-point>
load.64 %r1 <- 0[a]
set.64 %r2 <- 3.141500 <================
mulu.64 %r3 <- %r1, %r2
set.64 %r5 <- 3.141500 <===duplicate======
mulu.64 %r6 <- %r1, %r5
set.64 %r7 <- 5.000000
mulu.64 %r8 <- %r6, %r7
add.64 %r9 <- %r3, %r8
store.64 %r9 -> 0[a]
ret
>
> Sorry, this has been tried and discussed already. It doesn't really work.
> * a size is essentially a type, certainly for integers in sparse IR.
> * A constant is a constant, 1 is 1, when it has 1 bit or 64 bits, it's the same
> constant and follow the same mathematical rules. It's even not signed or
> unsigned, it's just 1.
> * in sparse, types are conveyed by the *operations* made on their operands.
> There are lots of good reasons for this. Trying to do the opposite for these
> constants will just break a lot of things, duplicate a lot of code
> (try it) and
> bring confusion everywhere (because of simply using the operation you
> then need at each time check the type of the operands. It's something
> we want to avoid (like for the discussion about casts we had yesterday)
> not something we want to add).
I see, thanks for the explain. Let me think about it for a while.
> * the OP_PUSH instructions do a perfect job of giving a type to functions
> arguments
> * the OP_PUSH instructions also nicely abstract away the *operation* of
> argument passing that all architectures have. It's a good thing.
I need to take a closer look at that patch.
Chris
--
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
