Re: [Qemu-devel] [PATCH 2/4] tcg-hppa: Finish the port.

[Richard Henderson <rth@twiddle.net>]

On 04/08/2010 09:32 AM, Richard Henderson wrote:
>>> +static void tcg_out_ori(TCGContext *s, int ret, int arg, tcg_target_ulong m)
>>> +{
>>> +    if (m == 0) {
>>> +        tcg_out_mov(s, ret, arg);
>>> +    } else if (m == -1) {
>>> +        tcg_out_movi(s, TCG_TYPE_I32, ret, -1);
>>
>> Those cases are already eliminated in tcg/tcg-op.h. This code looks
>> redundant.
> 
> The cases eliminated in tcg-op.h are with immediate constants.
> There is no generic code in tcg.c to eliminate these cases 
> after constant propagation.  However, I can remove them with...

For the record, a real case that appears in linux-test-0.3 sparc:

0x435e8ac4:  andcc  %g0, %g0, %o4

 ---- 0x435e8ac4
 movi_i32 tmp19,$0x0
 movi_i32 tmp20,$0x0
 and_i32 loc4,tmp19,tmp20
 mov_i32 cc_dst,loc4
 movi_i32 cc_op,$0xb
 st_i32 loc4,regwptr,$0x10

The and_i32 there is "loc4 = 0 & 0".

I've no idea why the original sparc code uses this instruction.


r~