On 11/20/23 01:39, Richard Biener wrote:
On Mon, 20 Nov 2023, Jakub Jelinek wrote:
On Mon, Nov 20, 2023 at 07:54:54AM +, Richard Biener wrote:
On Fri, 17 Nov 2023, Jakub Jelinek wrote:
Per the earlier discussions on this PR, the following patch folds
popcount (x) == 1 (and != 1) into
On Mon, Nov 20, 2023 at 08:39:42AM +, Richard Biener wrote:
> I think it's fine as you posted (and Jeff approved), I'm just wondering
> if we should rename that pass somehow ;) Note ISEL is more for
> required pre-expansion stuff and widen-mul is for expansion related but
> optimization
On Mon, 20 Nov 2023, Jakub Jelinek wrote:
> On Mon, Nov 20, 2023 at 07:54:54AM +, Richard Biener wrote:
> > On Fri, 17 Nov 2023, Jakub Jelinek wrote:
> > > Per the earlier discussions on this PR, the following patch folds
> > > popcount (x) == 1 (and != 1) into (x ^ (x - 1)) > x - 1 (or <=)
>
On Mon, Nov 20, 2023 at 07:54:54AM +, Richard Biener wrote:
> On Fri, 17 Nov 2023, Jakub Jelinek wrote:
> > Per the earlier discussions on this PR, the following patch folds
> > popcount (x) == 1 (and != 1) into (x ^ (x - 1)) > x - 1 (or <=)
> > if the corresponding popcount optab isn't
On Fri, 17 Nov 2023, Jakub Jelinek wrote:
> Hi!
>
> Per the earlier discussions on this PR, the following patch folds
> popcount (x) == 1 (and != 1) into (x ^ (x - 1)) > x - 1 (or <=)
> if the corresponding popcount optab isn't implemented (I think any
> double-word popcount or call will be
On 11/17/23 07:01, Jakub Jelinek wrote:
Hi!
Per the earlier discussions on this PR, the following patch folds
popcount (x) == 1 (and != 1) into (x ^ (x - 1)) > x - 1 (or <=)
if the corresponding popcount optab isn't implemented (I think any
double-word popcount or call will be necessarily
Hi!
Per the earlier discussions on this PR, the following patch folds
popcount (x) == 1 (and != 1) into (x ^ (x - 1)) > x - 1 (or <=)
if the corresponding popcount optab isn't implemented (I think any
double-word popcount or call will be necessarily slower than the
above cheap 3 op check and even