summaryrefslogtreecommitdiffstats
path: root/src/intel
diff options
context:
space:
mode:
authorIan Romanick <[email protected]>2018-08-22 20:31:11 -0700
committerIan Romanick <[email protected]>2018-08-28 15:35:38 -0700
commitfabe3ead57c8d0cd1ad0ac9e22a181ae3491154e (patch)
tree199472ed7a1e4f40fa0e52142bd42e9dfee23d2a /src/intel
parent41399f4bc745ef6b99081e7498b543eb3fbf82e7 (diff)
i965/fs: Emit BRW_AOP_INC or BRW_AOP_DEC for atomicAdd of +1 or -1
Funny story... a single shader was hurt for instructions, spills, fills. That same shader was also the most helped for cycles. #GPUsAreWeird No changes on any other Intel platform. v2: Refactor selection of atomic opcode to a separate function. Suggested by Jason. Haswell, Broadwell, and Skylake had similar results. (Skylake shown) total instructions in shared programs: 14304116 -> 14304261 (<.01%) instructions in affected programs: 12776 -> 12921 (1.13%) helped: 19 HURT: 1 helped stats (abs) min: 1 max: 16 x̄: 2.32 x̃: 1 helped stats (rel) min: 0.05% max: 7.27% x̄: 0.92% x̃: 0.55% HURT stats (abs) min: 189 max: 189 x̄: 189.00 x̃: 189 HURT stats (rel) min: 4.87% max: 4.87% x̄: 4.87% x̃: 4.87% 95% mean confidence interval for instructions value: -12.83 27.33 95% mean confidence interval for instructions %-change: -1.57% 0.31% Inconclusive result (value mean confidence interval includes 0). total cycles in shared programs: 527552861 -> 527531226 (<.01%) cycles in affected programs: 1459195 -> 1437560 (-1.48%) helped: 16 HURT: 2 helped stats (abs) min: 2 max: 21328 x̄: 1353.69 x̃: 6 helped stats (rel) min: 0.01% max: 5.29% x̄: 0.36% x̃: 0.03% HURT stats (abs) min: 12 max: 12 x̄: 12.00 x̃: 12 HURT stats (rel) min: 0.03% max: 0.03% x̄: 0.03% x̃: 0.03% 95% mean confidence interval for cycles value: -3699.81 1295.92 95% mean confidence interval for cycles %-change: -0.94% 0.30% Inconclusive result (value mean confidence interval includes 0). total spills in shared programs: 8025 -> 8033 (0.10%) spills in affected programs: 208 -> 216 (3.85%) helped: 1 HURT: 1 total fills in shared programs: 10989 -> 11040 (0.46%) fills in affected programs: 444 -> 495 (11.49%) helped: 1 HURT: 1 Ivy Bridge total instructions in shared programs: 11709181 -> 11709153 (<.01%) instructions in affected programs: 3505 -> 3477 (-0.80%) helped: 3 HURT: 0 helped stats (abs) min: 1 max: 23 x̄: 9.33 x̃: 4 helped stats (rel) min: 0.11% max: 1.16% x̄: 0.63% x̃: 0.61% total cycles in shared programs: 254741126 -> 254738801 (<.01%) cycles in affected programs: 919067 -> 916742 (-0.25%) helped: 3 HURT: 0 helped stats (abs) min: 21 max: 2144 x̄: 775.00 x̃: 160 helped stats (rel) min: 0.03% max: 0.90% x̄: 0.32% x̃: 0.03% total spills in shared programs: 4536 -> 4533 (-0.07%) spills in affected programs: 40 -> 37 (-7.50%) helped: 1 HURT: 0 total fills in shared programs: 4819 -> 4813 (-0.12%) fills in affected programs: 94 -> 88 (-6.38%) helped: 1 HURT: 0 Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Caio Marcelo de Oliveira Filho <[email protected]> [v1] Reviewed-by: Jason Ekstrand <[email protected]>
Diffstat (limited to 'src/intel')
-rw-r--r--src/intel/compiler/brw_fs_nir.cpp27
1 files changed, 23 insertions, 4 deletions
diff --git a/src/intel/compiler/brw_fs_nir.cpp b/src/intel/compiler/brw_fs_nir.cpp
index 62bff2a323a..1b7ed5e92cc 100644
--- a/src/intel/compiler/brw_fs_nir.cpp
+++ b/src/intel/compiler/brw_fs_nir.cpp
@@ -3604,6 +3604,21 @@ fs_visitor::nir_emit_fs_intrinsic(const fs_builder &bld,
}
}
+static int
+get_op_for_atomic_add(nir_intrinsic_instr *instr, unsigned src)
+{
+ const nir_const_value *const val = nir_src_as_const_value(instr->src[src]);
+
+ if (val != NULL) {
+ if (val->i32[0] == 1)
+ return BRW_AOP_INC;
+ else if (val->i32[0] == -1)
+ return BRW_AOP_DEC;
+ }
+
+ return BRW_AOP_ADD;
+}
+
void
fs_visitor::nir_emit_cs_intrinsic(const fs_builder &bld,
nir_intrinsic_instr *instr)
@@ -3660,7 +3675,7 @@ fs_visitor::nir_emit_cs_intrinsic(const fs_builder &bld,
}
case nir_intrinsic_shared_atomic_add:
- nir_emit_shared_atomic(bld, BRW_AOP_ADD, instr);
+ nir_emit_shared_atomic(bld, get_op_for_atomic_add(instr, 1), instr);
break;
case nir_intrinsic_shared_atomic_imin:
nir_emit_shared_atomic(bld, BRW_AOP_IMIN, instr);
@@ -4378,7 +4393,7 @@ fs_visitor::nir_emit_intrinsic(const fs_builder &bld, nir_intrinsic_instr *instr
}
case nir_intrinsic_ssbo_atomic_add:
- nir_emit_ssbo_atomic(bld, BRW_AOP_ADD, instr);
+ nir_emit_ssbo_atomic(bld, get_op_for_atomic_add(instr, 2), instr);
break;
case nir_intrinsic_ssbo_atomic_imin:
nir_emit_ssbo_atomic(bld, BRW_AOP_IMIN, instr);
@@ -4889,7 +4904,9 @@ fs_visitor::nir_emit_ssbo_atomic(const fs_builder &bld,
}
fs_reg offset = get_nir_src(instr->src[1]);
- fs_reg data1 = get_nir_src(instr->src[2]);
+ fs_reg data1;
+ if (op != BRW_AOP_INC && op != BRW_AOP_DEC && op != BRW_AOP_PREDEC)
+ data1 = get_nir_src(instr->src[2]);
fs_reg data2;
if (op == BRW_AOP_CMPWR)
data2 = get_nir_src(instr->src[3]);
@@ -4963,7 +4980,9 @@ fs_visitor::nir_emit_shared_atomic(const fs_builder &bld,
fs_reg surface = brw_imm_ud(GEN7_BTI_SLM);
fs_reg offset;
- fs_reg data1 = get_nir_src(instr->src[1]);
+ fs_reg data1;
+ if (op != BRW_AOP_INC && op != BRW_AOP_DEC && op != BRW_AOP_PREDEC)
+ data1 = get_nir_src(instr->src[1]);
fs_reg data2;
if (op == BRW_AOP_CMPWR)
data2 = get_nir_src(instr->src[2]);