This patch fixes a case in which we vectorised something with a
fully-predicated loop even after the cost model had rejected it.
E.g. the loop in the testcase has the costs:
Vector inside of loop cost: 27
Vector prologue cost: 0
Vector epilogue cost: 0
Scalar iteration cost: 7
Scalar out
On Mon, Mar 18, 2019 at 10:59 AM Richard Sandiford
wrote:
>
> This patch fixes a case in which we vectorised something with a
> fully-predicated loop even after the cost model had rejected it.
> E.g. the loop in the testcase has the costs:
>
> Vector inside of loop cost: 27
> Vector prologue c
Richard Biener writes:
> On Mon, Mar 18, 2019 at 10:59 AM Richard Sandiford
> wrote:
>>
>> This patch fixes a case in which we vectorised something with a
>> fully-predicated loop even after the cost model had rejected it.
>> E.g. the loop in the testcase has the costs:
>>
>> Vector inside of l
On 18 March 2019 10:58:53 CET, Richard Sandiford
wrote:
>This patch fixes a case in which we vectorised something with a
>fully-predicated loop even after the cost model had rejected it.
>E.g. the loop in the testcase has the costs:
>
> Vector inside of loop cost: 27
> Vector prologue cost: 0
>
Bernhard Reutner-Fischer writes:
> On 18 March 2019 10:58:53 CET, Richard Sandiford
> wrote:
>>This patch fixes a case in which we vectorised something with a
>>fully-predicated loop even after the cost model had rejected it.
>>E.g. the loop in the testcase has the costs:
>>
>> Vector inside of
On Tue, 19 Mar 2019 08:47:49 +
Richard Sandiford wrote:
> > ... this identical condition, AFAICS?
> > So this second conditions else arm should be dead, shouldn't it?
>
> Yeah, that's what:
>
> /* ??? The "if" arm is written to handle all cases; see below for what
> we would do for
