From 55f9fbb03d0413cb8fe74e5ec5d6c2dd4280933e Mon Sep 17 00:00:00 2001 From: Alexander Ivchenko Date: Fri, 11 Jul 2014 15:24:10 +0400 Subject: [4.8, 4.9] Backport of additional SLM tuning. Six patches from trunk, reg-tested via 'make check': 2014-05-07 Evgeny Stupachenko * tree-vect-data-refs.c (vect_grouped_load_supported): New check for loads group of length 3. (vect_permute_load_chain): New permutations for loads group of length 3. * tree-vect-stmts.c (vect_model_load_cost): Change cost of vec_perm_shuffle for the new permutations. 2014-04-17 Evgeny Stupachenko * config/i386/i386.c (x86_add_stmt_cost): Fix vector cost model for Silvermont. 2014-04-17 Evgeny Stupachenko * config/i386/x86-tune.def (TARGET_SLOW_PSHUFB): New tune definition. * config/i386/i386.h (TARGET_SLOW_PSHUFB): New tune flag. * config/i386/i386.c (expand_vec_perm_even_odd_1): Avoid byte shuffles for TARGET_SLOW_PSHUFB 2014-04-17 Evgeny Stupachenko * config/i386/i386.c (slm_cost): Adjust vec_to_scalar_cost. * config/i386/i386.c (intel_cost): Ditto. 2014-06-18 Evgeny Stupachenko * config/i386/i386.c (ix86_reassociation_width): Add alternative for vector case. * config/i386/i386.h (TARGET_VECTOR_PARALLEL_EXECUTION): New. * config/i386/x86-tune.def (X86_TUNE_VECTOR_PARALLEL_EXECUTION): New. * tree-vect-data-refs.c (vect_shift_permute_load_chain): New. Introduces alternative way of loads group permutaions. (vect_transform_grouped_load): Try alternative way of permutations. 2014-06-05 Evgeny Stupachenko * config/i386/sse.md (*ssse3_palignr_perm): New. * config/i386/predicates.md (palignr_operand): New. Indicates if permutation is suitable for palignr instruction. Change-Id: I5e505735ce3dc0ec3c2a1151713a119b24d712fe Signed-off-by: Alexander Ivchenko --- gcc-4.9/gcc/testsuite/gcc.dg/vect/pr52252-ld.c | 30 ++++++++++++++++++++++ .../gcc/testsuite/gcc.target/i386/pr52252-atom.c | 29 +++++++++++++++++++++ .../gcc/testsuite/gcc.target/i386/pr52252-core.c | 29 +++++++++++++++++++++ gcc-4.9/gcc/testsuite/gcc.target/i386/pr61403.c | 27 +++++++++++++++++++ 4 files changed, 115 insertions(+) create mode 100644 gcc-4.9/gcc/testsuite/gcc.dg/vect/pr52252-ld.c create mode 100644 gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-atom.c create mode 100644 gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-core.c create mode 100644 gcc-4.9/gcc/testsuite/gcc.target/i386/pr61403.c (limited to 'gcc-4.9/gcc/testsuite') diff --git a/gcc-4.9/gcc/testsuite/gcc.dg/vect/pr52252-ld.c b/gcc-4.9/gcc/testsuite/gcc.dg/vect/pr52252-ld.c new file mode 100644 index 000000000..6e3cb52b8 --- /dev/null +++ b/gcc-4.9/gcc/testsuite/gcc.dg/vect/pr52252-ld.c @@ -0,0 +1,30 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -g -ftree-vectorize -mssse3 -fdump-tree-vect-details" { target { i?86-*-* x86_64-*-* } } } */ + +#define byte unsigned char + +void +matrix_mul (byte *in, byte *out, int size) +{ + int i; + for (i = 0; i < size; i++) + { + byte in0 = in[0]; + byte in1 = in[1]; + byte in2 = in[2]; + byte out0, out1, out2, out3; + out0 = in0 + in1; + out1 = in0 + in2; + out2 = in1 + in2; + out3 = in0 + in1 + in2; + out[0] = out0; + out[1] = out1; + out[2] = out2; + out[3] = out3; + in += 3; + out += 4; + } +} + +/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" } } */ +/* { dg-final { cleanup-tree-dump "vect" } } */ diff --git a/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-atom.c b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-atom.c new file mode 100644 index 000000000..715b45943 --- /dev/null +++ b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-atom.c @@ -0,0 +1,29 @@ +/* { dg-do compile } */ +/* { dg-require-effective-target ssse3 } */ +/* { dg-options "-O2 -ftree-vectorize -mssse3 -mtune=slm" } */ +#define byte unsigned char + +void +matrix_mul (byte *in, byte *out, int size) +{ + int i; + for (i = 0; i < size; i++) + { + byte in0 = in[0]; + byte in1 = in[1]; + byte in2 = in[2]; + byte out0, out1, out2, out3; + out0 = in0 + in1; + out1 = in0 + in2; + out2 = in1 + in2; + out3 = in0 + in1 + in2; + out[0] = out0; + out[1] = out1; + out[2] = out2; + out[3] = out3; + in += 3; + out += 4; + } +} + +/* { dg-final { scan-assembler "palignr" } } */ diff --git a/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-core.c b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-core.c new file mode 100644 index 000000000..ac857a5fe --- /dev/null +++ b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr52252-core.c @@ -0,0 +1,29 @@ +/* { dg-do compile } */ +/* { dg-require-effective-target ssse3 } */ +/* { dg-options "-O2 -ftree-vectorize -mssse3 -mtune=corei7" } */ +#define byte unsigned char + +void +matrix_mul (byte *in, byte *out, int size) +{ + int i; + for (i = 0; i < size; i++) + { + byte in0 = in[0]; + byte in1 = in[1]; + byte in2 = in[2]; + byte out0, out1, out2, out3; + out0 = in0 + in1; + out1 = in0 + in2; + out2 = in1 + in2; + out3 = in0 + in1 + in2; + out[0] = out0; + out[1] = out1; + out[2] = out2; + out[3] = out3; + in += 3; + out += 4; + } +} + +/* { dg-final { scan-assembler "pshufb" } } */ diff --git a/gcc-4.9/gcc/testsuite/gcc.target/i386/pr61403.c b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr61403.c new file mode 100644 index 000000000..84cc5c5c8 --- /dev/null +++ b/gcc-4.9/gcc/testsuite/gcc.target/i386/pr61403.c @@ -0,0 +1,27 @@ +/* { dg-do compile } */ +/* { dg-require-effective-target sse4 } */ +/* { dg-options "-O2 -ffast-math -ftree-vectorize -msse4.2 -mtune=corei7" } */ + +#include + +struct XYZ +{ + float x; + float y; + float z; +}; + +void +norm (struct XYZ *in, struct XYZ *out, int size) +{ + int i; + for (i = 0; i < size; ++i) + { + float n = sqrt (in[i].x * in[i].x + in[i].y * in[i].y + in[i].z * in[i].z); + out[i].x = in[i].x / n; + out[i].y = in[i].y / n; + out[i].z = in[i].z / n; + } +} + +/* { dg-final { scan-assembler "blend" } } */ -- cgit v1.2.3