On Wed, Jun 27, 2012 at 9:00 PM, Richard Henderson <r...@redhat.com> wrote:
> On 06/27/2012 09:07 AM, Igor Zamyatin wrote:
>> May I ask about the purpose of the following piece of change? Doesn't
>> it affect non-sse cases either?
>
> Err, no, it doesn't affect non-sse cases. All MODE_VECTOR_INT
> cases will be implemented in the xmm registers (ignoring the
> deprecated and largely ignored mmx case).
Probably I misunderstand something... This condition - GET_MODE_SIZE
(mode) < UNITS_PER_WORD - is outside the check for MODE_VECTOR_INT.
>
>
>>
>> @@ -32038,7 +32042,15 @@ ix86_rtx_costs (rtx x, int code, int
>> outer_code_i, int opno, int *total,
>> case ASHIFTRT:
>> case LSHIFTRT:
>> case ROTATERT:
>> - if (!TARGET_64BIT && GET_MODE (XEXP (x, 0)) == DImode)
>> + if (GET_MODE_CLASS (mode) == MODE_VECTOR_INT)
>> + {
>> + /* ??? Should be SSE vector operation cost. */
>> + /* At least for published AMD latencies, this really is the same
>> + as the latency for a simple fpu operation like fabs. */
>> + *total = cost->fabs;
>> + return false;
>> + }
>> + if (GET_MODE_SIZE (mode) < UNITS_PER_WORD)
>> {
>> if (CONST_INT_P (XEXP (x, 1)))
>> {
>>
>> It also seems that we reversed the condition for the code that is now
>> under if (GET_MODE_SIZE (mode) < UNITS_PER_WORD). Why do we need this?
>
> I'm not sure what you're suggesting. But we certainly don't use
> the xmm registers to implement DImode operations in 32-bit, so...
>
>
> r~
