Lines Matching refs:rsrc
90 define amdgpu_ps float @buffer_load_x1(<4 x i32> inreg %rsrc, i32 %idx, i32 %ofs) {
92 %data = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 %idx, i32 %ofs, i1 0, i1 0)
99 define amdgpu_ps <2 x float> @buffer_load_x2(<4 x i32> inreg %rsrc, i32 %idx, i32 %ofs) {
101 …%data = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 %idx, i32 %ofs, i1 0,…
120 define amdgpu_ps float @buffer_load_mmo(<4 x i32> inreg %rsrc, float addrspace(3)* %lds) {
123 %val = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 0, i1 0, i1 0)
134 define amdgpu_ps void @buffer_load_x1_offen_merged(<4 x i32> inreg %rsrc, i32 %a) {
142 %r1 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a1, i1 0, i1 0)
143 %r2 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a2, i1 0, i1 0)
144 %r3 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a3, i1 0, i1 0)
145 %r4 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a4, i1 0, i1 0)
146 %r5 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a5, i1 0, i1 0)
147 %r6 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a6, i1 0, i1 0)
159 define amdgpu_ps void @buffer_load_x1_offen_merged_glc_slc(<4 x i32> inreg %rsrc, i32 %a) {
167 %r1 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a1, i1 0, i1 0)
168 %r2 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a2, i1 0, i1 0)
169 %r3 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a3, i1 1, i1 0)
170 %r4 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a4, i1 1, i1 0)
171 %r5 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a5, i1 1, i1 1)
172 %r6 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a6, i1 1, i1 1)
182 define amdgpu_ps void @buffer_load_x2_offen_merged(<4 x i32> inreg %rsrc, i32 %a) {
186 …%vr1 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 %a1, i1 0, i1 0)
187 …%vr2 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 %a2, i1 0, i1 0)
200 define amdgpu_ps void @buffer_load_x3_offen_merged(<4 x i32> inreg %rsrc, i32 %a) {
204 …%vr1 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 %a1, i1 0, i1 0)
205 %r3 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 %a2, i1 0, i1 0)
217 define amdgpu_ps void @buffer_load_x1_offset_merged(<4 x i32> inreg %rsrc) {
219 %r1 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 4, i1 0, i1 0)
220 %r2 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
221 %r3 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 12, i1 0, i1 0)
222 %r4 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 16, i1 0, i1 0)
223 %r5 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 28, i1 0, i1 0)
224 %r6 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 32, i1 0, i1 0)
234 define amdgpu_ps void @buffer_load_x2_offset_merged(<4 x i32> inreg %rsrc) {
236 %vr1 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 4, i1 0, i1 0)
237 %vr2 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 12, i1 0, i1 0)
250 define amdgpu_ps void @buffer_load_x3_offset_merged(<4 x i32> inreg %rsrc) {
252 %vr1 = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 0, i32 4, i1 0, i1 0)
253 %r3 = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 0, i32 12, i1 0, i1 0)
266 define amdgpu_ps float @buffer_load_ubyte(<4 x i32> inreg %rsrc) {
268 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
279 define amdgpu_ps float @buffer_load_ushort(<4 x i32> inreg %rsrc) {
281 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 16, i1 0, i1 0)
293 define amdgpu_ps float @buffer_load_sbyte(<4 x i32> inreg %rsrc) {
295 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
307 define amdgpu_ps float @buffer_load_sshort(<4 x i32> inreg %rsrc) {
309 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 16, i1 0, i1 0)
320 define amdgpu_ps float @buffer_load_ubyte_bitcast(<4 x i32> inreg %rsrc) {
322 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
333 define amdgpu_ps float @buffer_load_ushort_bitcast(<4 x i32> inreg %rsrc) {
335 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
346 define amdgpu_ps float @buffer_load_sbyte_bitcast(<4 x i32> inreg %rsrc) {
348 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
359 define amdgpu_ps float @buffer_load_sshort_bitcast(<4 x i32> inreg %rsrc) {
361 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
373 define amdgpu_ps float @buffer_load_ubyte_mul_bitcast(<4 x i32> inreg %rsrc) {
375 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
388 define amdgpu_ps float @buffer_load_ushort_mul_bitcast(<4 x i32> inreg %rsrc) {
390 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
403 define amdgpu_ps float @buffer_load_sbyte_mul_bitcast(<4 x i32> inreg %rsrc) {
405 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
418 define amdgpu_ps float @buffer_load_sshort_mul_bitcast(<4 x i32> inreg %rsrc) {
420 %tmp = call i16 @llvm.amdgcn.buffer.load.i16(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
433 define amdgpu_ps float @buffer_load_sbyte_type_check(<4 x i32> inreg %rsrc) {
435 %tmp = call i8 @llvm.amdgcn.buffer.load.i8(<4 x i32> %rsrc, i32 0, i32 8, i1 0, i1 0)
449 define amdgpu_ps float @no_fold_fi_imm_soffset(<4 x i32> inreg %rsrc) {
453 …%ret.val = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 %alloca.cast, i32 0, i1 fa…
461 define amdgpu_ps float @no_fold_fi_reg_soffset(<4 x i32> inreg %rsrc, i32 inreg %soffset) {
465 …%ret.val = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 %alloca.cast, i32 %soffset…