/external/llvm-project/polly/test/GPGPU/ |
D | scalar-parameter.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 19 ; CODE: Code 20 ; CODE-NEXT: ==== 21 ; CODE-NEXT: # host 22 ; CODE-NEXT: { 23 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (1024) * sizeof(float), cudaMemcp… 24 ; CODE-NEXT: { 25 ; CODE-NEXT: dim3 k0_dimBlock(32); 26 ; CODE-NEXT: dim3 k0_dimGrid(32); 27 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A, MemRef_b); [all …]
|
D | non-read-only-scalars.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 34 ; CODE: dim3 k0_dimBlock(32); 35 ; CODE-NEXT: dim3 k0_dimGrid(1); 36 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A); 37 ; CODE-NEXT: cudaCheckKernel(); 38 ; CODE-NEXT: } 40 ; CODE: { 41 ; CODE-NEXT: dim3 k1_dimBlock; 42 ; CODE-NEXT: dim3 k1_dimGrid; 43 ; CODE-NEXT: kernel1 <<<k1_dimGrid, k1_dimBlock>>> (dev_MemRef_sum_0__phi); [all …]
|
D | mostly-sequential.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 16 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (128) * sizeof(float), cudaMemcpy… 17 ; CODE-NEXT: { 18 ; CODE-NEXT: dim3 k0_dimBlock(32); 19 ; CODE-NEXT: dim3 k0_dimGrid(4); 20 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A); 21 ; CODE-NEXT: cudaCheckKernel(); 22 ; CODE-NEXT: } 24 ; CODE: { 25 ; CODE-NEXT: dim3 k1_dimBlock; [all …]
|
D | parametric-loop-bound.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 16 ; CODE: if (n >= 1) { 17 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (n) * sizeof(i64), cudaMemcpyHost… 18 ; CODE-NEXT: { 19 ; CODE-NEXT: dim3 k0_dimBlock(32); 20 ; CODE-NEXT: dim3 k0_dimGrid(n >= 1048545 ? 32768 : (n + 31) / 32); 21 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A, n); 22 ; CODE-NEXT: cudaCheckKernel(); 23 ; CODE-NEXT: } 25 ; CODE: cudaCheckReturn(cudaMemcpy(MemRef_A, dev_MemRef_A, (n) * sizeof(i64), cudaMemcpyDevi… [all …]
|
D | host-statement.ll | 4 ; RUN: FileCheck -check-prefix=CODE %s 21 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (512) * (512) * sizeof(double), c… 22 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_R, MemRef_R, (p_0 + 1) * (512) * sizeof(double… 23 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_Q, MemRef_Q, (512) * (512) * sizeof(double), c… 24 ; CODE-NEXT: { 25 ; CODE-NEXT: dim3 k0_dimBlock(32); 26 ; CODE-NEXT: dim3 k0_dimGrid(16); 27 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A, dev_MemRef_R, dev_MemRef_Q, p… 28 ; CODE-NEXT: cudaCheckKernel(); 29 ; CODE-NEXT: } [all …]
|
D | size-cast.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 12 ; CODE: if (arg >= 1 && arg1 == 0) { 13 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_arg3, MemRef_arg3, (arg) * sizeof(double), cud… 14 ; CODE-NEXT: { 15 ; CODE-NEXT: dim3 k0_dimBlock(32); 16 ; CODE-NEXT: dim3 k0_dimGrid(arg >= 1048545 ? 32768 : (arg + 31) / 32); 17 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_arg3, dev_MemRef_arg2, arg, arg1… 18 ; CODE-NEXT: cudaCheckKernel(); 19 ; CODE-NEXT: } 21 ; CODE: cudaCheckReturn(cudaMemcpy(MemRef_arg2, dev_MemRef_arg2, (arg) * sizeof(double), cudaMemc… [all …]
|
D | non-zero-array-offset.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 10 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_B, MemRef_B, (16) * sizeof(float), cudaMemcpyHos… 11 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (8) * sizeof(float), cudaMemcpyHost… 13 ; CODE: dim3 k0_dimBlock(8); 14 ; CODE-NEXT: dim3 k0_dimGrid(1); 15 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A); 16 ; CODE-NEXT: cudaCheckKernel(); 17 ; CODE-NEXT: } 19 ; CODE: { 20 ; CODE-NEXT: dim3 k1_dimBlock(8); [all …]
|
D | scheduler-timeout.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 30 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (4096) * (4096) * sizeof(float), … 31 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_B, MemRef_B, (4096) * (4096) * sizeof(float), … 32 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_D, MemRef_D, (4096) * (4096) * sizeof(float), … 33 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_C, MemRef_C, (4096) * (4096) * sizeof(float), … 34 ; CODE-NEXT: { 35 ; CODE-NEXT: dim3 k0_dimBlock(16, 32); 36 ; CODE-NEXT: dim3 k0_dimGrid(128, 128); 37 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_tmp, dev_MemRef_A, MemRef_alpha,… 38 ; CODE-NEXT: cudaCheckKernel(); [all …]
|
D | managed-pointers-preparation.ll | 8 ; RUN: -polly-acc-dump-code < %s | FileCheck %s -check-prefix=CODE 20 ; CODE: if (p_0_loaded_from___data_runcontrol_MOD_lmulti_layer == 0) { 21 ; CODE-NEXT: { 22 ; CODE-NEXT: dim3 k0_dimBlock; 23 ; CODE-NEXT: dim3 k0_dimGrid; 24 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef__pn__phi, p_0_loaded_from___dat… 25 ; CODE-NEXT: cudaCheckKernel(); 26 ; CODE-NEXT: } 28 ; CODE: } else { 29 ; CODE-NEXT: { [all …]
|
D | invalid-kernel.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 23 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_B, MemRef_B, (1024) * sizeof(i64), cudaMemcpyH… 24 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (1024) * sizeof(i64), cudaMemcpyH… 25 ; CODE-NEXT: { 26 ; CODE-NEXT: dim3 k0_dimBlock(32); 27 ; CODE-NEXT: dim3 k0_dimGrid(32); 28 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_B, dev_MemRef_A); 29 ; CODE-NEXT: cudaCheckKernel(); 30 ; CODE-NEXT: } 32 ; CODE: cudaCheckReturn(cudaMemcpy(MemRef_A, dev_MemRef_A, (1024) * sizeof(i64), cudaMemcpyDevice… [all …]
|
D | region-stmt.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 8 ; CODE: cudaCheckReturn(cudaMemcpy(dev_MemRef_A, MemRef_A, (128) * sizeof(float), cudaMemcpy… 9 ; CODE-NEXT: cudaCheckReturn(cudaMemcpy(dev_MemRef_B, MemRef_B, (128) * sizeof(float), cudaMemcpy… 10 ; CODE-NEXT: { 11 ; CODE-NEXT: dim3 k0_dimBlock(32); 12 ; CODE-NEXT: dim3 k0_dimGrid(4); 13 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_A, dev_MemRef_B); 14 ; CODE-NEXT: cudaCheckKernel(); 15 ; CODE-NEXT: } 17 ; CODE: cudaCheckReturn(cudaMemcpy(MemRef_B, dev_MemRef_B, (128) * sizeof(float), cudaMemcpyDevic… [all …]
|
D | shared-memory-two-dimensional.ll | 4 ; RUN: FileCheck -check-prefix=CODE %s 21 ; CODE: # kernel0 22 ; CODE-NEXT: { 23 ; CODE-NEXT: if (t0 <= 7) 24 ; CODE-NEXT: for (int c0 = 0; c0 <= 15; c0 += 1) 25 ; CODE-NEXT: read(c0, t0); 26 ; CODE-NEXT: read(t0); 27 ; CODE-NEXT: sync0(); 28 ; CODE-NEXT: for (int c3 = 0; c3 <= 15; c3 += 1) 29 ; CODE-NEXT: for (int c4 = 0; c4 <= 7; c4 += 1) [all …]
|
D | phi-nodes-in-kernel.ll | 3 ; RUN: FileCheck -check-prefix=CODE %s 35 ; CODE: cudaCheckReturn(cudaMalloc((void **) &dev_MemRef_c, (50) * sizeof(i32))); 37 ; CODE: { 38 ; CODE-NEXT: dim3 k0_dimBlock(32); 39 ; CODE-NEXT: dim3 k0_dimGrid(2); 40 ; CODE-NEXT: kernel0 <<<k0_dimGrid, k0_dimBlock>>> (dev_MemRef_c); 41 ; CODE-NEXT: cudaCheckKernel(); 42 ; CODE-NEXT: } 44 ; CODE: cudaCheckReturn(cudaMemcpy(MemRef_c, dev_MemRef_c, (50) * sizeof(i32), cudaMemcpyDevi… 45 ; CODE-NEXT: cudaCheckReturn(cudaFree(dev_MemRef_c)); [all …]
|
/external/llvm-project/llvm/test/Analysis/CostModel/AArch64/ |
D | min-max.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 7 ; CODE-LABEL: umin.v8i8 8 ; CODE: bb.0 9 ; CODE-NEXT: umin v{{.*}}.8b, v{{.*}}.8b, v{{.*}}.8b 10 ; CODE-NEXT: ret 21 ; CODE-LABEL: umin.v9i8 22 ; CODE: bb.0 23 ; CODE-NEXT: umin v{{.*}}.16b, v{{.*}}.16b, v{{.*}}.16b 24 ; CODE-NEXT: ret 35 ; CODE-LABEL: umin.v4i16 [all …]
|
D | vector-select.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 8 ; CODE-LABEL: v8i8_select_eq 9 ; CODE: bb.0 10 ; CODE-NEXT: cmeq v{{.+}}.8b, v{{.+}}.8b, v{{.+}}.8b 11 ; CODE-NEXT: bif v{{.+}}.8b, v{{.+}}.8b, v{{.+}}.8b 12 ; CODE-NEXT: ret 24 ; CODE-LABEL: v16i8_select_sgt 25 ; CODE: bb.0 26 ; CODE-NEXT: cmgt v{{.+}}.16b, v{{.+}}.16b, v{{.+}}.16b 27 ; CODE-NEXT: bif v{{.+}}.16b, v{{.+}}.16b, v{{.+}}.16b [all …]
|
D | vector-reduce.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 6 ; CODE-LABEL: add.i8.v8i8 7 ; CODE: addv b0, v0.8b 15 ; CODE-LABEL: add.i8.v16i8 16 ; CODE: addv b0, v0.16b 24 ; CODE-LABEL: add.i16.v4i16 25 ; CODE: addv h0, v0.4h 33 ; CODE-LABEL: add.i16.v8i16 34 ; CODE: addv h0, v0.8h 42 ; CODE-LABEL: add.i32.v4i32 [all …]
|
D | free-widening-casts.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 7 ; CODE-LABEL: uaddl_8h 8 ; CODE: uaddl v0.8h, v0.8b, v1.8b 19 ; CODE-LABEL: uaddl_4s 20 ; CODE: uaddl v0.4s, v0.4h, v1.4h 31 ; CODE-LABEL: uaddl_2d 32 ; CODE: uaddl v0.2d, v0.2s, v1.2s 43 ; CODE-LABEL: uaddl2_8h 44 ; CODE: uaddl2 v2.8h, v0.16b, v1.16b 45 ; CODE-NEXT: uaddl v0.8h, v0.8b, v1.8b [all …]
|
D | shuffle-transpose.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 6 ; CODE-LABEL: trn1.v8i8 7 ; CODE: trn1 v0.8b, v0.8b, v1.8b 15 ; CODE-LABEL: trn2.v8i8 16 ; CODE: trn2 v0.8b, v0.8b, v1.8b 24 ; CODE-LABEL: trn1.v16i8 25 ; CODE: trn1 v0.16b, v0.16b, v1.16b 33 ; CODE-LABEL: trn2.v16i8 34 ; CODE: trn2 v0.16b, v0.16b, v1.16b 42 ; CODE-LABEL: trn1.v4i16 [all …]
|
D | shuffle-select.ll | 2 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE 6 ; CODE-LABEL: sel.v8i8 7 ; CODE: tbl v0.8b, { v0.16b }, v2.8b 15 ; CODE-LABEL: sel.v16i8 16 ; CODE: tbl v0.16b, { v0.16b, v1.16b }, v2.16b 24 ; CODE-LABEL: sel.v4i16 25 ; CODE: rev32 v0.4h, v0.4h 26 ; CODE: trn2 v0.4h, v0.4h, v1.4h 34 ; CODE-LABEL: sel.v8i16 35 ; CODE: tbl v0.16b, { v0.16b, v1.16b }, v2.16b [all …]
|
/external/llvm-project/llvm/test/CodeGen/ARM/Windows/ |
D | vla.ll | 2 ; RUN: | FileCheck %s -check-prefix CHECK-SMALL-CODE 4 ; RUN: | FileCheck %s -check-prefix CHECK-LARGE-CODE 6 ; RUN: | FileCheck %s -check-prefix CHECK-SMALL-CODE 16 ; CHECK-SMALL-CODE: adds [[R4:r[0-9]+]], #7 17 ; CHECK-SMALL-CODE: bic [[R4]], [[R4]], #4 18 ; CHECK-SMALL-CODE: lsrs r4, [[R4]], #2 19 ; CHECK-SMALL-CODE: bl __chkstk 20 ; CHECK-SMALL-CODE: sub.w sp, sp, r4 22 ; CHECK-LARGE-CODE: adds [[R4:r[0-9]+]], #7 23 ; CHECK-LARGE-CODE: bic [[R4]], [[R4]], #4 [all …]
|
D | chkstk.ll | 2 ; RUN: | FileCheck -check-prefix CHECK-DEFAULT-CODE-MODEL %s 5 ; RUN: | FileCheck -check-prefix CHECK-LARGE-CODE-MODEL %s 13 ; CHECK-DEFAULT-CODE-MODEL: check_watermark: 14 ; CHECK-DEFAULT-CODE-MODEL: movw r4, #1024 15 ; CHECK-DEFAULT-CODE-MODEL: bl __chkstk 16 ; CHECK-DEFAULT-CODE-MODEL: sub.w sp, sp, r4 18 ; CHECK-LARGE-CODE-MODEL: check_watermark: 19 ; CHECK-LARGE-CODE-MODEL-DAG: movw r12, :lower16:__chkstk 20 ; CHECK-LARGE-CODE-MODEL-DAG: movt r12, :upper16:__chkstk 21 ; CHECK-LARGE-CODE-MODEL-DAG: movw r4, #1024 [all …]
|
D | stack-probe-non-default.ll | 2 ; RUN: | FileCheck %s -check-prefix CHECK-DEFAULT-CODE-MODEL 5 ; RUN: | FileCheck %s -check-prefix CHECK-LARGE-CODE-MODEL 21 ; CHECK-DEFAULT-CODE-MODEL-NOT: __chkstk 22 ; CHECK-DEFAULT-CODE-MODEL: sub.w sp, sp, #4096 24 ; CHECK-LARGE-CODE-MODEL-NOT: movw r12, :lower16:__chkstk 25 ; CHECK-LARGE-CODE-MODEL-NOT: movt r12, :upper16:__chkstk 26 ; CHECK-LARGE-CODE-MODEL: sub.w sp, sp, #4096
|
/external/llvm/test/CodeGen/ARM/Windows/ |
D | vla.ll | 2 ; RUN: | FileCheck %s -check-prefix CHECK-SMALL-CODE 4 ; RUN: | FileCheck %s -check-prefix CHECK-LARGE-CODE 16 ; CHECK-SMALL-CODE: adds [[R4:r[0-9]+]], #7 17 ; CHECK-SMALL-CODE: bic [[R4]], [[R4]], #7 18 ; CHECK-SMALL-CODE: lsrs r4, [[R4]], #2 19 ; CHECK-SMALL-CODE: bl __chkstk 20 ; CHECK-SMALL-CODE: sub.w sp, sp, r4 22 ; CHECK-LARGE-CODE: adds [[R4:r[0-9]+]], #7 23 ; CHECK-LARGE-CODE: bic [[R4]], [[R4]], #7 24 ; CHECK-LARGE-CODE: lsrs r4, [[R4]], #2 [all …]
|
D | chkstk.ll | 2 ; RUN: | FileCheck -check-prefix CHECK-DEFAULT-CODE-MODEL %s 5 ; RUN: | FileCheck -check-prefix CHECK-LARGE-CODE-MODEL %s 13 ; CHECK-DEFAULT-CODE-MODEL: check_watermark: 14 ; CHECK-DEFAULT-CODE-MODEL: movw r4, #1024 15 ; CHECK-DEFAULT-CODE-MODEL: bl __chkstk 16 ; CHECK-DEFAULT-CODE-MODEL: sub.w sp, sp, r4 18 ; CHECK-LARGE-CODE-MODEL: check_watermark: 19 ; CHECK-LARGE-CODE-MODEL: movw r12, :lower16:__chkstk 20 ; CHECK-LARGE-CODE-MODEL: movt r12, :upper16:__chkstk 21 ; CHECK-LARGE-CODE-MODEL: movw r4, #1024 [all …]
|
/external/llvm-project/llvm/test/CodeGen/AArch64/ |
D | chkstk.ll | 2 ; RUN: | FileCheck -check-prefix CHECK-DEFAULT-CODE-MODEL %s 7 ; RUN: | FileCheck -check-prefix CHECK-LARGE-CODE-MODEL %s 17 ; CHECK-DEFAULT-CODE-MODEL: check_watermark: 18 ; CHECK-DEFAULT-CODE-MODEL-DAG: stp x29, x30, [sp 19 ; CHECK-DEFAULT-CODE-MODEL-DAG: mov x15, #256 20 ; CHECK-DEFAULT-CODE-MODEL: bl __chkstk 21 ; CHECK-DEFAULT-CODE-MODEL: sub sp, sp, x15, lsl #4 25 ; CHECK-LARGE-CODE-MODEL: check_watermark: 26 ; CHECK-LARGE-CODE-MODEL-DAG: stp x29, x30, [sp 27 ; CHECK-LARGE-CODE-MODEL-DAG: mov x15, #256 [all …]
|