On Wed, Apr 14, 2010 at 8:48 AM, roy rosen <roy.1ro...@gmail.com> wrote:
> Hi All,
>
> I have implemented some vectorization features in my gcc port.
>
> In the generated code for this function I can see a scheduling problem:
>
> int xxx(int* __restrict__ a, int* __restrict__ b)
> {
> int __restrict__ i;
> for (i = 0; i < 8; i++)
> {
> a[i] = b[i];
> }
> return 0;
> }
>
> from asmcons:
>
> (insn 17 14 18 3 a.c:15 (set (reg:V4HI 105)
> (mem:V4HI (reg/v/f:SI 100 [ b ]) [2 S8 A64])) 115 {*movv4hi_load}
> (nil))
>
> (insn 18 17 19 3 a.c:15 (set (mem:V4HI (reg/v/f:SI 99 [ a ]) [2 S8 A64])
> (reg:V4HI 105)) 116 {*movv4hi_store} (expr_list:REG_DEAD (reg:V4HI 105)
> (nil)))
>
> (insn 19 18 20 3 a.c:15 (set (reg:V4HI 106)
> (mem:V4HI (plus:SI (reg/v/f:SI 100 [ b ])
> (const_int 8 [0x8])) [2 S8 A64])) 115 {*movv4hi_load}
> (expr_list:REG_DEAD (reg/v/f:SI 100 [ b ])
> (nil)))
>
> (insn 20 19 58 3 a.c:15 (set (mem:V4HI (plus:SI (reg/v/f:SI 99 [ a ])
> (const_int 8 [0x8])) [2 S8 A64])
> (reg:V4HI 106)) 116 {*movv4hi_store} (expr_list:REG_DEAD (reg:V4HI 106)
> (expr_list:REG_DEAD (reg/v/f:SI 99 [ a ])
> (nil))))
>
> from sched1 - dependecies list:
> --- Region Dependences --- b 3 bb 0
> ;; insn code bb dep prio cost reservation
> ;; ---- ---- -- --- ---- ---- -----------
> ;; 17 115 3 0 4 1 (2_slots+lsu) : 58 20 18
> ;; 18 116 3 1 3 1 (2_slots+lsu) : 58 19
> ;; 19 115 3 1 2 1 (2_slots+lsu) : 58 20
> ;; 20 116 3 2 1 1 (2_slots+lsu) : 58
> ;; 58 101 3 4 1 1 (3_slots+pcu) :
>
> As you can see, insn 19 is dependant on 18.
> I think that this should not happen since the arrays has __restrict__.
> It might have something to do with aliasing.
>
> Are you aware of any kind of problems regarding aliasing and vectors?
> Can you think of anything else that might be wrong in my code that
> leads gcc to think that there is a dependency between insn 18 and 19?
It completely depends on the GCC version you are using. The
MEM_EXPRs suggest that it is maybe 4.4 or 4.5 based? Because
they are not existant. Try recent trunk which should work fine.
Richard.
> Thanks, Roy.
>
