Hello everyone!
(s->a = (s->b = (s->c = n + 4)));
gcc & msvc:
--> 30 / 30 / 126 // Because the return value is left
tcc
--> 254 / 30 / 126 // Because the return is the right value
Shall tcc assignment modify what? Same with gcc&msvc.
于 2014年06月22日 01:49, jiang 写道:
This is my patch (see Annex)
tcc result is correct
--> 254 / 30 / 126
1 1 1 1 1 1 1 0 -> 254
254 & 0x7f (c:7)
1 1 1 1 1 1 1 0 -> 254
0 1 1 1 1 1 1 1 -> 0x7f
---------------------&-----
0 1 1 1 1 1 1 0 -> 126
(b:5)
0 1 1 1 1 1 1 0 -> 126
0 0 0 1 1 1 1 1 -> 0x1f
---------------------&-----
0 0 0 1 1 1 1 0 -> 30
(a:9)
0 0 0 0 1 1 1 1 0 -> 30
1 1 1 1 1 1 1 1 1 -> 0x1ff
---------------------&-----
0 0 0 1 1 1 1 0 -> 30
I guess gcc & mvc repeated use of the register, so wrong
于 2014年06月21日 22:25, tinycc-devel-requ...@nongnu.org 写道:
jiang wrote:
/* bitfield store handling */
+ SValue tmp;
+ tmp = vtop[0];
[...]
+ vtop--;
+ vpushv(&tmp);
This is still not a solution. See
#include <stdio.h>
int main(int argc, char **argv)
{
struct {
unsigned a:9, b:5, c:7;
} _s, *s = &_s;
int n = 250;
s->a = s->b = s->c = n + 4;
printf("--> %d / %d / %d\n", s->a, s->b, s->c);
return 0;
}
--> 432 / 16 / 126
gcc & msvc:
--> 30 / 30 / 126
tcc release_0_9_26
--> 254 / 30 / 126
FWIW, the line above is the reason why I'm trying to investigate
this. Let's see if Mr. jiang can come up with something useful
(and how long it will take).
-- gr
_______________________________________________
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
https://lists.nongnu.org/mailman/listinfo/tinycc-devel
