subject:"\[Bug tree\-optimization\/80570\] auto\-vectorizing int\->double conversion should use half\-width memory operands to avoid shuffles, instead of load\+extract"

[Bug tree-optimization/80570] auto-vectorizing int->double conversion should use half-width memory operands to avoid shuffles, instead of load+extract

2021-09-26 Thread peter at cordes dot ca via Gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80570 --- Comment #3 from Peter Cordes --- (In reply to Andrew Pinski from comment #2) > Even on aarch64: > > .L2: > ldr q0, [x1], 16 > sxtlv1.2d, v0.2s > sxtl2 v0.2d, v0.4s > scvtf v1.2d, v1.2d >

[Bug tree-optimization/80570] auto-vectorizing int->double conversion should use half-width memory operands to avoid shuffles, instead of load+extract

2021-09-26 Thread pinskia at gcc dot gnu.org via Gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80570 Andrew Pinski changed: What|Removed |Added Component|target |tree-optimization --- Comment #2 from