1 // REQUIRES: amdgpu-registered-target
2 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -S -emit-llvm -o - %s | FileCheck %s
3 // RUN: %clang_cc1 -triple r600-unknown-unknown -S -emit-llvm -o - %s | FileCheck %s
5 typedef __attribute__(( ext_vector_type(2) )) char char2;
6 typedef __attribute__(( ext_vector_type(3) )) char char3;
7 typedef __attribute__(( ext_vector_type(4) )) char char4;
9 typedef __attribute__(( ext_vector_type(2) )) short short2;
10 typedef __attribute__(( ext_vector_type(3) )) short short3;
11 typedef __attribute__(( ext_vector_type(4) )) short short4;
13 typedef __attribute__(( ext_vector_type(2) )) int int2;
14 typedef __attribute__(( ext_vector_type(3) )) int int3;
15 typedef __attribute__(( ext_vector_type(4) )) int int4;
16 typedef __attribute__(( ext_vector_type(16) )) int int16;
17 typedef __attribute__(( ext_vector_type(32) )) int int32;
19 // CHECK: %struct.empty_struct = type {}
20 typedef struct empty_struct
24 // CHECK-NOT: %struct.single_element_struct_arg
25 typedef struct single_element_struct_arg
28 } single_element_struct_arg_t;
30 // CHECK-NOT: %struct.nested_single_element_struct_arg
31 typedef struct nested_single_element_struct_arg
33 single_element_struct_arg_t i;
34 } nested_single_element_struct_arg_t;
36 // CHECK: %struct.struct_arg = type { i32, float, i32 }
37 typedef struct struct_arg
44 // CHECK: %struct.struct_padding_arg = type { i8, i64 }
45 typedef struct struct_padding_arg
51 // CHECK: %struct.struct_of_arrays_arg = type { [2 x i32], float, [4 x i32], [3 x float], i32 }
52 typedef struct struct_of_arrays_arg
59 } struct_of_arrays_arg_t;
61 // CHECK: %struct.struct_of_structs_arg = type { i32, float, %struct.struct_arg, i32 }
62 typedef struct struct_of_structs_arg
68 } struct_of_structs_arg_t;
70 // CHECK: %union.transparent_u = type { i32 }
75 } transparent_u __attribute__((__transparent_union__));
77 // CHECK: %struct.single_array_element_struct_arg = type { [4 x i32] }
78 typedef struct single_array_element_struct_arg
81 } single_array_element_struct_arg_t;
83 // CHECK: %struct.single_struct_element_struct_arg = type { %struct.inner }
84 // CHECK: %struct.inner = type { i32, i64 }
85 typedef struct single_struct_element_struct_arg
91 } single_struct_element_struct_arg_t;
93 // CHECK: %struct.different_size_type_pair
94 typedef struct different_size_type_pair {
97 } different_size_type_pair;
99 // CHECK: %struct.flexible_array = type { i32, [0 x i32] }
100 typedef struct flexible_array
106 // CHECK: %struct.struct_arr16 = type { [16 x i32] }
107 typedef struct struct_arr16
112 // CHECK: %struct.struct_arr32 = type { [32 x i32] }
113 typedef struct struct_arr32
118 // CHECK: %struct.struct_arr33 = type { [33 x i32] }
119 typedef struct struct_arr33
124 // CHECK: %struct.struct_char_arr32 = type { [32 x i8] }
125 typedef struct struct_char_arr32
130 // CHECK-NOT: %struct.struct_char_x8
131 typedef struct struct_char_x8 {
136 // CHECK-NOT: %struct.struct_char_x4
137 typedef struct struct_char_x4 {
141 // CHECK-NOT: %struct.struct_char_x3
142 typedef struct struct_char_x3 {
146 // CHECK-NOT: %struct.struct_char_x2
147 typedef struct struct_char_x2 {
151 // CHECK-NOT: %struct.struct_char_x1
152 typedef struct struct_char_x1 {
156 // 4 registers from fields, 5 if padding included.
157 // CHECK: %struct.nested = type { i8, i64 }
158 // CHECK: %struct.num_regs_nested_struct = type { i32, %struct.nested }
159 typedef struct num_regs_nested_struct {
165 } num_regs_nested_struct;
167 // CHECK: %struct.double_nested = type { %struct.inner_inner }
168 // CHECK: %struct.inner_inner = type { i8, i32, i8 }
169 // CHECK: %struct.double_nested_struct = type { i32, %struct.double_nested, i16 }
170 typedef struct double_nested_struct {
172 struct double_nested {
181 } double_nested_struct;
183 // This is a large struct, but uses fewer registers than the limit.
184 // CHECK: %struct.large_struct_padding = type { i8, i32, i8, i32, i8, i8, i16, i16, [3 x i8], i64, i32, i8, i32, i16, i8 }
185 typedef struct large_struct_padding {
201 } large_struct_padding;
203 // CHECK: %struct.int3_pair = type { <3 x i32>, <3 x i32> }
204 // The number of registers computed should be 6, not 8.
205 typedef struct int3_pair {
210 // CHECK: %struct.struct_4regs = type { i32, i32, i32, i32 }
211 typedef struct struct_4regs
219 // CHECK: void @kernel_empty_struct_arg(%struct.empty_struct %s.coerce)
220 __kernel void kernel_empty_struct_arg(empty_struct s) { }
222 // CHECK: void @kernel_single_element_struct_arg(i32 %arg1.coerce)
223 __kernel void kernel_single_element_struct_arg(single_element_struct_arg_t arg1) { }
225 // CHECK: void @kernel_nested_single_element_struct_arg(i32 %arg1.coerce)
226 __kernel void kernel_nested_single_element_struct_arg(nested_single_element_struct_arg_t arg1) { }
228 // CHECK: void @kernel_struct_arg(%struct.struct_arg %arg1.coerce)
229 __kernel void kernel_struct_arg(struct_arg_t arg1) { }
231 // CHECK: void @kernel_struct_padding_arg(%struct.struct_padding_arg %arg1.coerce)
232 __kernel void kernel_struct_padding_arg(struct_padding_arg arg1) { }
234 // CHECK: void @kernel_test_struct_of_arrays_arg(%struct.struct_of_arrays_arg %arg1.coerce)
235 __kernel void kernel_test_struct_of_arrays_arg(struct_of_arrays_arg_t arg1) { }
237 // CHECK: void @kernel_struct_of_structs_arg(%struct.struct_of_structs_arg %arg1.coerce)
238 __kernel void kernel_struct_of_structs_arg(struct_of_structs_arg_t arg1) { }
240 // CHECK: void @test_kernel_transparent_union_arg(%union.transparent_u %u.coerce)
241 __kernel void test_kernel_transparent_union_arg(transparent_u u) { }
243 // CHECK: void @kernel_single_array_element_struct_arg(%struct.single_array_element_struct_arg %arg1.coerce)
244 __kernel void kernel_single_array_element_struct_arg(single_array_element_struct_arg_t arg1) { }
246 // CHECK: void @kernel_single_struct_element_struct_arg(%struct.single_struct_element_struct_arg %arg1.coerce)
247 __kernel void kernel_single_struct_element_struct_arg(single_struct_element_struct_arg_t arg1) { }
249 // CHECK: void @kernel_different_size_type_pair_arg(%struct.different_size_type_pair %arg1.coerce)
250 __kernel void kernel_different_size_type_pair_arg(different_size_type_pair arg1) { }
252 // CHECK: define void @func_f32_arg(float %arg)
253 void func_f32_arg(float arg) { }
255 // CHECK: define void @func_v2i16_arg(<2 x i16> %arg)
256 void func_v2i16_arg(short2 arg) { }
258 // CHECK: define void @func_v3i32_arg(<3 x i32> %arg)
259 void func_v3i32_arg(int3 arg) { }
261 // CHECK: define void @func_v4i32_arg(<4 x i32> %arg)
262 void func_v4i32_arg(int4 arg) { }
264 // CHECK: define void @func_v16i32_arg(<16 x i32> %arg)
265 void func_v16i32_arg(int16 arg) { }
267 // CHECK: define void @func_v32i32_arg(<32 x i32> %arg)
268 void func_v32i32_arg(int32 arg) { }
270 // CHECK: define void @func_empty_struct_arg()
271 void func_empty_struct_arg(empty_struct empty) { }
273 // CHECK: void @func_single_element_struct_arg(i32 %arg1.coerce)
274 void func_single_element_struct_arg(single_element_struct_arg_t arg1) { }
276 // CHECK: void @func_nested_single_element_struct_arg(i32 %arg1.coerce)
277 void func_nested_single_element_struct_arg(nested_single_element_struct_arg_t arg1) { }
279 // CHECK: void @func_struct_arg(i32 %arg1.coerce0, float %arg1.coerce1, i32 %arg1.coerce2)
280 void func_struct_arg(struct_arg_t arg1) { }
282 // CHECK: void @func_struct_padding_arg(i8 %arg1.coerce0, i64 %arg1.coerce1)
283 void func_struct_padding_arg(struct_padding_arg arg1) { }
285 // CHECK: define void @func_struct_char_x8([2 x i32] %arg.coerce)
286 void func_struct_char_x8(struct_char_x8 arg) { }
288 // CHECK: define void @func_struct_char_x4(i32 %arg.coerce)
289 void func_struct_char_x4(struct_char_x4 arg) { }
291 // CHECK: define void @func_struct_char_x3(i32 %arg.coerce)
292 void func_struct_char_x3(struct_char_x3 arg) { }
294 // CHECK: define void @func_struct_char_x2(i16 %arg.coerce)
295 void func_struct_char_x2(struct_char_x2 arg) { }
297 // CHECK: define void @func_struct_char_x1(i8 %arg.coerce)
298 void func_struct_char_x1(struct_char_x1 arg) { }
300 // CHECK: void @func_transparent_union_arg(i32 %u.coerce)
301 void func_transparent_union_arg(transparent_u u) { }
303 // CHECK: void @func_single_array_element_struct_arg([4 x i32] %arg1.coerce)
304 void func_single_array_element_struct_arg(single_array_element_struct_arg_t arg1) { }
306 // CHECK: void @func_single_struct_element_struct_arg(%struct.inner %arg1.coerce)
307 void func_single_struct_element_struct_arg(single_struct_element_struct_arg_t arg1) { }
309 // CHECK: void @func_different_size_type_pair_arg(i64 %arg1.coerce0, i32 %arg1.coerce1)
310 void func_different_size_type_pair_arg(different_size_type_pair arg1) { }
312 // CHECK: void @func_flexible_array_arg(%struct.flexible_array addrspace(5)* byval nocapture align 4 %arg)
313 void func_flexible_array_arg(flexible_array arg) { }
315 // CHECK: define float @func_f32_ret()
321 // CHECK: define void @func_empty_struct_ret()
322 empty_struct func_empty_struct_ret()
328 // CHECK: define i32 @single_element_struct_ret()
330 single_element_struct_arg_t single_element_struct_ret()
332 single_element_struct_arg_t s = { 0 };
336 // CHECK: define i32 @nested_single_element_struct_ret()
338 nested_single_element_struct_arg_t nested_single_element_struct_ret()
340 nested_single_element_struct_arg_t s = { 0 };
344 // CHECK: define %struct.struct_arg @func_struct_ret()
345 // CHECK: ret %struct.struct_arg zeroinitializer
346 struct_arg_t func_struct_ret()
348 struct_arg_t s = { 0 };
352 // CHECK: define %struct.struct_padding_arg @func_struct_padding_ret()
353 // CHECK: ret %struct.struct_padding_arg zeroinitializer
354 struct_padding_arg func_struct_padding_ret()
356 struct_padding_arg s = { 0 };
360 // CHECK: define [2 x i32] @func_struct_char_x8_ret()
361 // CHECK: ret [2 x i32] zeroinitializer
362 struct_char_x8 func_struct_char_x8_ret()
364 struct_char_x8 s = { 0 };
368 // CHECK: define i32 @func_struct_char_x4_ret()
370 struct_char_x4 func_struct_char_x4_ret()
372 struct_char_x4 s = { 0 };
376 // CHECK: define i32 @func_struct_char_x3_ret()
378 struct_char_x3 func_struct_char_x3_ret()
380 struct_char_x3 s = { 0 };
384 // CHECK: define i16 @func_struct_char_x2_ret()
385 struct_char_x2 func_struct_char_x2_ret()
387 struct_char_x2 s = { 0 };
391 // CHECK: define i8 @func_struct_char_x1_ret()
393 struct_char_x1 func_struct_char_x1_ret()
395 struct_char_x1 s = { 0 };
399 // CHECK: define %struct.struct_arr16 @func_ret_struct_arr16()
400 // CHECK: ret %struct.struct_arr16 zeroinitializer
401 struct_arr16 func_ret_struct_arr16()
403 struct_arr16 s = { 0 };
407 // CHECK: define void @func_ret_struct_arr32(%struct.struct_arr32 addrspace(5)* noalias nocapture sret %agg.result)
408 struct_arr32 func_ret_struct_arr32()
410 struct_arr32 s = { 0 };
414 // CHECK: define void @func_ret_struct_arr33(%struct.struct_arr33 addrspace(5)* noalias nocapture sret %agg.result)
415 struct_arr33 func_ret_struct_arr33()
417 struct_arr33 s = { 0 };
421 // CHECK: define %struct.struct_char_arr32 @func_ret_struct_char_arr32()
422 struct_char_arr32 func_ret_struct_char_arr32()
424 struct_char_arr32 s = { 0 };
428 // CHECK: define i32 @func_transparent_union_ret() local_unnamed_addr #1 {
430 transparent_u func_transparent_union_ret()
432 transparent_u u = { 0 };
436 // CHECK: define %struct.different_size_type_pair @func_different_size_type_pair_ret()
437 different_size_type_pair func_different_size_type_pair_ret()
439 different_size_type_pair s = { 0 };
443 // CHECK: define void @func_flexible_array_ret(%struct.flexible_array addrspace(5)* noalias nocapture sret %agg.result)
444 flexible_array func_flexible_array_ret()
446 flexible_array s = { 0 };
450 // CHECK: define void @func_reg_state_lo(<4 x i32> %arg0, <4 x i32> %arg1, <4 x i32> %arg2, i32 %arg3, i32 %s.coerce0, float %s.coerce1, i32 %s.coerce2)
451 void func_reg_state_lo(int4 arg0, int4 arg1, int4 arg2, int arg3, struct_arg_t s) { }
453 // CHECK: define void @func_reg_state_hi(<4 x i32> %arg0, <4 x i32> %arg1, <4 x i32> %arg2, i32 %arg3, i32 %arg4, %struct.struct_arg addrspace(5)* byval nocapture align 4 %s)
454 void func_reg_state_hi(int4 arg0, int4 arg1, int4 arg2, int arg3, int arg4, struct_arg_t s) { }
456 // XXX - Why don't the inner structs flatten?
457 // CHECK: define void @func_reg_state_num_regs_nested_struct(<4 x i32> %arg0, i32 %arg1, i32 %arg2.coerce0, %struct.nested %arg2.coerce1, i32 %arg3.coerce0, %struct.nested %arg3.coerce1, %struct.num_regs_nested_struct addrspace(5)* byval nocapture align 8 %arg4)
458 void func_reg_state_num_regs_nested_struct(int4 arg0, int arg1, num_regs_nested_struct arg2, num_regs_nested_struct arg3, num_regs_nested_struct arg4) { }
460 // CHECK: define void @func_double_nested_struct_arg(<4 x i32> %arg0, i32 %arg1, i32 %arg2.coerce0, %struct.double_nested %arg2.coerce1, i16 %arg2.coerce2)
461 void func_double_nested_struct_arg(int4 arg0, int arg1, double_nested_struct arg2) { }
463 // CHECK: define %struct.double_nested_struct @func_double_nested_struct_ret(<4 x i32> %arg0, i32 %arg1)
464 double_nested_struct func_double_nested_struct_ret(int4 arg0, int arg1) {
465 double_nested_struct s = { 0 };
469 // CHECK: define void @func_large_struct_padding_arg_direct(i8 %arg.coerce0, i32 %arg.coerce1, i8 %arg.coerce2, i32 %arg.coerce3, i8 %arg.coerce4, i8 %arg.coerce5, i16 %arg.coerce6, i16 %arg.coerce7, [3 x i8] %arg.coerce8, i64 %arg.coerce9, i32 %arg.coerce10, i8 %arg.coerce11, i32 %arg.coerce12, i16 %arg.coerce13, i8 %arg.coerce14)
470 void func_large_struct_padding_arg_direct(large_struct_padding arg) { }
472 // CHECK: define void @func_large_struct_padding_arg_store(%struct.large_struct_padding addrspace(1)* nocapture %out, %struct.large_struct_padding addrspace(5)* byval nocapture readonly align 8 %arg)
473 void func_large_struct_padding_arg_store(global large_struct_padding* out, large_struct_padding arg) {
477 // CHECK: define void @v3i32_reg_count(<3 x i32> %arg1, <3 x i32> %arg2, <3 x i32> %arg3, <3 x i32> %arg4, i32 %arg5.coerce0, float %arg5.coerce1, i32 %arg5.coerce2)
478 void v3i32_reg_count(int3 arg1, int3 arg2, int3 arg3, int3 arg4, struct_arg_t arg5) { }
480 // Function signature from blender, nothing should be passed byval. The v3i32
481 // should not count as 4 passed registers.
482 // CHECK: define void @v3i32_pair_reg_count(%struct.int3_pair addrspace(5)* nocapture %arg0, <3 x i32> %arg1.coerce0, <3 x i32> %arg1.coerce1, <3 x i32> %arg2, <3 x i32> %arg3.coerce0, <3 x i32> %arg3.coerce1, <3 x i32> %arg4, float %arg5)
483 void v3i32_pair_reg_count(int3_pair *arg0, int3_pair arg1, int3 arg2, int3_pair arg3, int3 arg4, float arg5) { }
485 // Each short4 should fit pack into 2 registers.
486 // CHECK: define void @v4i16_reg_count(<4 x i16> %arg0, <4 x i16> %arg1, <4 x i16> %arg2, <4 x i16> %arg3, <4 x i16> %arg4, <4 x i16> %arg5, i32 %arg6.coerce0, i32 %arg6.coerce1, i32 %arg6.coerce2, i32 %arg6.coerce3)
487 void v4i16_reg_count(short4 arg0, short4 arg1, short4 arg2, short4 arg3,
488 short4 arg4, short4 arg5, struct_4regs arg6) { }
490 // CHECK: define void @v4i16_pair_reg_count_over(<4 x i16> %arg0, <4 x i16> %arg1, <4 x i16> %arg2, <4 x i16> %arg3, <4 x i16> %arg4, <4 x i16> %arg5, <4 x i16> %arg6, %struct.struct_4regs addrspace(5)* byval nocapture align 4 %arg7)
491 void v4i16_pair_reg_count_over(short4 arg0, short4 arg1, short4 arg2, short4 arg3,
492 short4 arg4, short4 arg5, short4 arg6, struct_4regs arg7) { }
494 // CHECK: define void @v3i16_reg_count(<3 x i16> %arg0, <3 x i16> %arg1, <3 x i16> %arg2, <3 x i16> %arg3, <3 x i16> %arg4, <3 x i16> %arg5, i32 %arg6.coerce0, i32 %arg6.coerce1, i32 %arg6.coerce2, i32 %arg6.coerce3)
495 void v3i16_reg_count(short3 arg0, short3 arg1, short3 arg2, short3 arg3,
496 short3 arg4, short3 arg5, struct_4regs arg6) { }
498 // CHECK: define void @v3i16_reg_count_over(<3 x i16> %arg0, <3 x i16> %arg1, <3 x i16> %arg2, <3 x i16> %arg3, <3 x i16> %arg4, <3 x i16> %arg5, <3 x i16> %arg6, %struct.struct_4regs addrspace(5)* byval nocapture align 4 %arg7)
499 void v3i16_reg_count_over(short3 arg0, short3 arg1, short3 arg2, short3 arg3,
500 short3 arg4, short3 arg5, short3 arg6, struct_4regs arg7) { }
502 // CHECK: define void @v2i16_reg_count(<2 x i16> %arg0, <2 x i16> %arg1, <2 x i16> %arg2, <2 x i16> %arg3, <2 x i16> %arg4, <2 x i16> %arg5, <2 x i16> %arg6, <2 x i16> %arg7, <2 x i16> %arg8, <2 x i16> %arg9, <2 x i16> %arg10, <2 x i16> %arg11, i32 %arg13.coerce0, i32 %arg13.coerce1, i32 %arg13.coerce2, i32 %arg13.coerce3)
503 void v2i16_reg_count(short2 arg0, short2 arg1, short2 arg2, short2 arg3,
504 short2 arg4, short2 arg5, short2 arg6, short2 arg7,
505 short2 arg8, short2 arg9, short2 arg10, short2 arg11,
506 struct_4regs arg13) { }
508 // CHECK: define void @v2i16_reg_count_over(<2 x i16> %arg0, <2 x i16> %arg1, <2 x i16> %arg2, <2 x i16> %arg3, <2 x i16> %arg4, <2 x i16> %arg5, <2 x i16> %arg6, <2 x i16> %arg7, <2 x i16> %arg8, <2 x i16> %arg9, <2 x i16> %arg10, <2 x i16> %arg11, <2 x i16> %arg12, %struct.struct_4regs addrspace(5)* byval nocapture align 4 %arg13)
509 void v2i16_reg_count_over(short2 arg0, short2 arg1, short2 arg2, short2 arg3,
510 short2 arg4, short2 arg5, short2 arg6, short2 arg7,
511 short2 arg8, short2 arg9, short2 arg10, short2 arg11,
512 short2 arg12, struct_4regs arg13) { }
514 // CHECK: define void @v2i8_reg_count(<2 x i8> %arg0, <2 x i8> %arg1, <2 x i8> %arg2, <2 x i8> %arg3, <2 x i8> %arg4, <2 x i8> %arg5, i32 %arg6.coerce0, i32 %arg6.coerce1, i32 %arg6.coerce2, i32 %arg6.coerce3)
515 void v2i8_reg_count(char2 arg0, char2 arg1, char2 arg2, char2 arg3,
516 char2 arg4, char2 arg5, struct_4regs arg6) { }
518 // CHECK: define void @v2i8_reg_count_over(<2 x i8> %arg0, <2 x i8> %arg1, <2 x i8> %arg2, <2 x i8> %arg3, <2 x i8> %arg4, <2 x i8> %arg5, i32 %arg6, %struct.struct_4regs addrspace(5)* byval nocapture align 4 %arg7)
519 void v2i8_reg_count_over(char2 arg0, char2 arg1, char2 arg2, char2 arg3,
520 char2 arg4, char2 arg5, int arg6, struct_4regs arg7) { }
522 // CHECK: define void @num_regs_left_64bit_aggregate(<4 x i32> %arg0, <4 x i32> %arg1, <4 x i32> %arg2, <3 x i32> %arg3, [2 x i32] %arg4.coerce, i32 %arg5)
523 void num_regs_left_64bit_aggregate(int4 arg0, int4 arg1, int4 arg2, int3 arg3, struct_char_x8 arg4, int arg5) { }