This creates a common function that can be shared by the tgsi
and nir backends.
v2: use LLVMBuildBitCast() directly
---
src/gallium/drivers/radeonsi/si_shader.c | 61 ++++++++++++++---------
src/gallium/drivers/radeonsi/si_shader_internal.h | 6 +++
2 files changed, 44 insertions(+), 23 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader.c
b/src/gallium/drivers/radeonsi/si_shader.c
index ce90c7beb5..c2338089b3 100644
--- a/src/gallium/drivers/radeonsi/si_shader.c
+++ b/src/gallium/drivers/radeonsi/si_shader.c
@@ -1297,47 +1297,42 @@ static void store_output_tcs(struct
lp_build_tgsi_context *bld_base,
}
if (reg->Register.WriteMask == 0xF && !is_tess_factor) {
LLVMValueRef value = lp_build_gather_values(&ctx->gallivm,
values, 4);
ac_build_buffer_store_dword(&ctx->ac, buffer, value, 4,
buf_addr,
base, 0, 1, 0, true, false);
}
}
-static LLVMValueRef fetch_input_gs(
- struct lp_build_tgsi_context *bld_base,
- const struct tgsi_full_src_register *reg,
- enum tgsi_opcode_type type,
- unsigned swizzle)
+LLVMValueRef si_llvm_load_input_gs(struct ac_shader_abi *abi,
+ unsigned input_index,
+ unsigned vtx_offset_param,
+ LLVMTypeRef type,
+ unsigned swizzle)
{
- struct si_shader_context *ctx = si_shader_context(bld_base);
+ struct si_shader_context *ctx = si_shader_context_from_abi(abi);
+ struct lp_build_tgsi_context *bld_base = &ctx->bld_base;
struct si_shader *shader = ctx->shader;
struct lp_build_context *uint = &ctx->bld_base.uint_bld;
LLVMValueRef vtx_offset, soffset;
struct tgsi_shader_info *info = &shader->selector->info;
- unsigned semantic_name = info->input_semantic_name[reg->Register.Index];
- unsigned semantic_index =
info->input_semantic_index[reg->Register.Index];
+ unsigned semantic_name = info->input_semantic_name[input_index];
+ unsigned semantic_index = info->input_semantic_index[input_index];
unsigned param;
LLVMValueRef value;
- if (swizzle != ~0 && semantic_name == TGSI_SEMANTIC_PRIMID)
- return get_primitive_id(ctx, swizzle);
-
- if (!reg->Register.Dimension)
- return NULL;
-
param = si_shader_io_get_unique_index(semantic_name, semantic_index);
/* GFX9 has the ESGS ring in LDS. */
if (ctx->screen->b.chip_class >= GFX9) {
- unsigned index = reg->Dimension.Index;
+ unsigned index = vtx_offset_param;
switch (index / 2) {
case 0:
vtx_offset = unpack_param(ctx,
ctx->param_gs_vtx01_offset,
index % 2 ? 16 : 0, 16);
break;
case 1:
vtx_offset = unpack_param(ctx,
ctx->param_gs_vtx23_offset,
index % 2 ? 16 : 0, 16);
break;
@@ -1345,56 +1340,76 @@ static LLVMValueRef fetch_input_gs(
vtx_offset = unpack_param(ctx,
ctx->param_gs_vtx45_offset,
index % 2 ? 16 : 0, 16);
break;
default:
assert(0);
return NULL;
}
vtx_offset = LLVMBuildAdd(ctx->ac.builder, vtx_offset,
LLVMConstInt(ctx->i32, param * 4, 0),
"");
- return lds_load(bld_base, tgsi2llvmtype(bld_base, type),
- swizzle, vtx_offset);
+ return lds_load(bld_base, type, swizzle, vtx_offset);
}
/* GFX6: input load from the ESGS ring in memory. */
if (swizzle == ~0) {
LLVMValueRef values[TGSI_NUM_CHANNELS];
unsigned chan;
for (chan = 0; chan < TGSI_NUM_CHANNELS; chan++) {
- values[chan] = fetch_input_gs(bld_base, reg, type,
chan);
+ values[chan] = si_llvm_load_input_gs(abi, input_index,
vtx_offset_param,
+ type, chan);
}
return lp_build_gather_values(&ctx->gallivm, values,
TGSI_NUM_CHANNELS);
}
/* Get the vertex offset parameter on GFX6. */
- unsigned vtx_offset_param = reg->Dimension.Index;
LLVMValueRef gs_vtx_offset = ctx->gs_vtx_offset[vtx_offset_param];
vtx_offset = lp_build_mul_imm(uint, gs_vtx_offset, 4);
soffset = LLVMConstInt(ctx->i32, (param * 4 + swizzle) * 256, 0);
value = ac_build_buffer_load(&ctx->ac, ctx->esgs_ring, 1, ctx->i32_0,
vtx_offset, soffset, 0, 1, 0, true, false);
- if (tgsi_type_is_64bit(type)) {
+ if (llvm_type_is_64bit(ctx, type)) {
LLVMValueRef value2;
soffset = LLVMConstInt(ctx->i32, (param * 4 + swizzle + 1) *
256, 0);
value2 = ac_build_buffer_load(&ctx->ac, ctx->esgs_ring, 1,
ctx->i32_0, vtx_offset, soffset,
0, 1, 0, true, false);
- return si_llvm_emit_fetch_64bit(bld_base,
tgsi2llvmtype(bld_base, type),
- value, value2);
+ return si_llvm_emit_fetch_64bit(bld_base, type, value, value2);
}
- return bitcast(bld_base, type, value);
+ return LLVMBuildBitCast(ctx->ac.builder, value, type, "");
+}
+
+static LLVMValueRef fetch_input_gs(
+ struct lp_build_tgsi_context *bld_base,
+ const struct tgsi_full_src_register *reg,
+ enum tgsi_opcode_type type,
+ unsigned swizzle)
+{
+ struct si_shader_context *ctx = si_shader_context(bld_base);
+ struct tgsi_shader_info *info = &ctx->shader->selector->info;
+
+ unsigned semantic_name = info->input_semantic_name[reg->Register.Index];
+ if (swizzle != ~0 && semantic_name == TGSI_SEMANTIC_PRIMID)
+ return get_primitive_id(ctx, swizzle);
+
+ if (!reg->Register.Dimension)
+ return NULL;
+
+ return si_llvm_load_input_gs(&ctx->abi, reg->Register.Index,
+ reg->Dimension.Index,
+ tgsi2llvmtype(bld_base, type),
+ swizzle);
}
static int lookup_interp_param_index(unsigned interpolate, unsigned location)
{
switch (interpolate) {
case TGSI_INTERPOLATE_CONSTANT:
return 0;
case TGSI_INTERPOLATE_LINEAR:
if (location == TGSI_INTERPOLATE_LOC_SAMPLE)
diff --git a/src/gallium/drivers/radeonsi/si_shader_internal.h
b/src/gallium/drivers/radeonsi/si_shader_internal.h
index 5219aa0f1d..03cf8fa6d9 100644
--- a/src/gallium/drivers/radeonsi/si_shader_internal.h
+++ b/src/gallium/drivers/radeonsi/si_shader_internal.h
@@ -272,20 +272,26 @@ void si_llvm_optimize_module(struct si_shader_context
*ctx);
LLVMValueRef si_llvm_emit_fetch_64bit(struct lp_build_tgsi_context *bld_base,
LLVMTypeRef type,
LLVMValueRef ptr,
LLVMValueRef ptr2);
LLVMValueRef si_llvm_emit_fetch(struct lp_build_tgsi_context *bld_base,
const struct tgsi_full_src_register *reg,
enum tgsi_opcode_type type,
unsigned swizzle);
+LLVMValueRef si_llvm_load_input_gs(struct ac_shader_abi *abi,
+ unsigned input_index,
+ unsigned vtx_offset_param,
+ LLVMTypeRef type,
+ unsigned swizzle);
+
void si_llvm_emit_store(struct lp_build_tgsi_context *bld_base,
const struct tgsi_full_instruction *inst,
const struct tgsi_opcode_info *info,
unsigned index,
LLVMValueRef dst[4]);
/* Combine these with & instead of |. */
#define NOOP_WAITCNT 0xf7f
#define LGKM_CNT 0x07f
#define VM_CNT 0xf70
--
2.14.3
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
