Re: [PATCH] tree-optimization/90579 - avoid STLF fail by better optimizing

2025-02-13 Thread Jeff Law
On 2/12/25 7:58 AM, Richard Biener wrote: For the testcase in question which uses a fold-left vectorized reduction of a reverse iterating loop we'd need two forwprop invocations to first bypass the permute emitted for the reverse iterating loop and then to decompose the vector load that only f

Re: [PATCH] tree-optimization/90579 - avoid STLF fail by better optimizing

2025-02-12 Thread Andrew Pinski
On Wed, Feb 12, 2025 at 6:58 AM Richard Biener wrote: > > For the testcase in question which uses a fold-left vectorized > reduction of a reverse iterating loop we'd need two forwprop > invocations to first bypass the permute emitted for the reverse > iterating loop and then to decompose the vecto

[PATCH] tree-optimization/90579 - avoid STLF fail by better optimizing

2025-02-12 Thread Richard Biener
For the testcase in question which uses a fold-left vectorized reduction of a reverse iterating loop we'd need two forwprop invocations to first bypass the permute emitted for the reverse iterating loop and then to decompose the vector load that only feeds element extracts. The following moves the