[Bug tree-optimization/82518] [8 regression] gfortran.fortran-torture/execute/in-pack.f90 fails on armeb since r252917

2018-02-16 Thread wilco.dijkstra at arm dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82518 --- Comment #49 from Wilco --- AArch64 does this: (define_expand "vec_store_lanesoi" [(set (match_operand:OI 0 "aarch64_simd_struct_operand" "=Utv") (unspec:OI [(match_operand:OI 1 "register_operand" "w") (unspec:VQ

[Bug tree-optimization/84114] global reassociation pass prevents fma usage, generates slower code

2018-02-16 Thread wilco.dijkstra at arm dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84114 Wilco changed: What|Removed |Added CC||wilco.dijkstra at arm dot com --- Comment #5

[Bug tree-optimization/82518] [8 regression] gfortran.fortran-torture/execute/in-pack.f90 fails on armeb since r252917

2018-02-16 Thread wilco.dijkstra at arm dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82518 --- Comment #47 from Wilco --- (In reply to Jakub Jelinek from comment #46) > Wonder if that: > vect_array.11[0] = vect_vec_iv_.7_45; > vect_array.11[1] = vect__4.8_48; > on armeb shouldn't have been [1] and [0] instead, otherwise we end up

[Bug tree-optimization/82518] [8 regression] gfortran.fortran-torture/execute/in-pack.f90 fails on armeb since r252917

2018-02-15 Thread wilco.dijkstra at arm dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82518 --- Comment #42 from Wilco --- Cut down example: typedef struct { int x, y; } X; void f (X *p, int n) { for (int i = 0; i < n; i++) { p[i].x = i; p[i].y = i + 1; } }

[Bug tree-optimization/82518] [8 regression] gfortran.fortran-torture/execute/in-pack.f90 fails on armeb since r252917

2018-02-15 Thread wilco.dijkstra at arm dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82518 Wilco changed: What|Removed |Added CC||wilco.dijkstra at arm dot com --- Comment #41