[PATCH vkd3d 1/5] vkd3d-shader/hlsl: Write SM4 multiplication instructions.
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_sm4.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index 32a4eda1..af1660f1 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -957,6 +957,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break; + case HLSL_OP2_MUL: + write_sm4_binary_op(buffer, VKD3D_SM4_OP_MUL, &expr->node, arg1, arg2); + break; + default: hlsl_fixme(ctx, expr->node.loc, "SM4 float \"%s\" expression.", debug_hlsl_expr_op(expr->op)); break; -- 2.33.0
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_sm4.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index af1660f1..ace2eb66 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -957,6 +957,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break; + case HLSL_OP2_DIV: + write_sm4_binary_op(buffer, VKD3D_SM4_OP_DIV, &expr->node, arg1, arg2); + break; + case HLSL_OP2_MUL: write_sm4_binary_op(buffer, VKD3D_SM4_OP_MUL, &expr->node, arg1, arg2); break; -- 2.33.0
Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> Il 30/08/21 19:21, Zebediah Figura ha scritto:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_sm4.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index af1660f1..ace2eb66 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -957,6 +957,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break;
+ case HLSL_OP2_DIV: + write_sm4_binary_op(buffer, VKD3D_SM4_OP_DIV, &expr->node, arg1, arg2); + break; + case HLSL_OP2_MUL: write_sm4_binary_op(buffer, VKD3D_SM4_OP_MUL, &expr->node, arg1, arg2); break;
Signed-off-by: Matteo Bruni <mbruni(a)codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet(a)codeweavers.com>
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- Makefile.am | 5 +---- libs/vkd3d-shader/hlsl_sm4.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 4 deletions(-) diff --git a/Makefile.am b/Makefile.am index c66f7b50..8c775c4d 100644 --- a/Makefile.am +++ b/Makefile.am @@ -277,10 +277,7 @@ XFAIL_TESTS = \ tests/hlsl-vector-indexing-uniform.shader_test \ tests/math.shader_test \ tests/trigonometry.shader_test \ - tests/writemask-assignop-0.shader_test \ - tests/writemask-assignop-1.shader_test \ - tests/writemask-assignop-2.shader_test \ - tests/writemask-assignop-3.shader_test + tests/writemask-assignop-1.shader_test endif if BUILD_DEMOS diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index ace2eb66..160da47f 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -561,6 +561,7 @@ struct sm4_register unsigned int idx_count; enum vkd3d_sm4_dimension dim; uint32_t immconst_uint[4]; + unsigned int mod; }; struct sm4_instruction @@ -710,6 +711,8 @@ static uint32_t sm4_register_order(const struct sm4_register *reg) if (reg->type == VKD3D_SM4_RT_IMMCONST) order += reg->dim == VKD3D_SM4_DIMENSION_VEC4 ? 4 : 1; order += reg->idx_count; + if (reg->mod) + ++order; return order; } @@ -743,8 +746,13 @@ static void write_sm4_instruction(struct vkd3d_bytecode_buffer *buffer, const st token = sm4_encode_register(&instr->srcs[i].reg); token |= sm4_swizzle_type(instr->srcs[i].reg.type) << VKD3D_SM4_SWIZZLE_TYPE_SHIFT; token |= instr->srcs[i].swizzle << VKD3D_SM4_SWIZZLE_SHIFT; + if (instr->srcs[i].reg.mod) + token |= VKD3D_SM4_REGISTER_MODIFIER; put_u32(buffer, token); + if (instr->srcs[i].reg.mod) + put_u32(buffer, instr->srcs[i].reg.mod); + for (j = 0; j < instr->srcs[i].reg.idx_count; ++j) put_u32(buffer, instr->srcs[i].reg.idx[j]); @@ -894,6 +902,26 @@ static void write_sm4_ret(struct vkd3d_bytecode_buffer *buffer) write_sm4_instruction(buffer, &instr); } +static void write_sm4_unary_op(struct vkd3d_bytecode_buffer *buffer, enum vkd3d_sm4_opcode opcode, + const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src, unsigned int src_mod) +{ + struct sm4_instruction instr; + unsigned int writemask; + + memset(&instr, 0, sizeof(instr)); + instr.opcode = opcode; + + sm4_register_from_node(&instr.dst.reg, &instr.dst.writemask, dst); + instr.has_dst = 1; + + sm4_register_from_node(&instr.srcs[0].reg, &writemask, src); + instr.srcs[0].swizzle = hlsl_map_swizzle(hlsl_swizzle_from_writemask(writemask), instr.dst.writemask); + instr.srcs[0].reg.mod = src_mod; + instr.src_count = 1; + + write_sm4_instruction(buffer, &instr); +} + static void write_sm4_binary_op(struct vkd3d_bytecode_buffer *buffer, enum vkd3d_sm4_opcode opcode, const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src1, const struct hlsl_ir_node *src2) { @@ -953,6 +981,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, { switch (expr->op) { + case HLSL_OP1_NEG: + write_sm4_unary_op(buffer, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3D_SM4_REGISTER_MODIFIER_NEGATE); + break; + case HLSL_OP2_ADD: write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break; -- 2.33.0
Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> Il 30/08/21 19:21, Zebediah Figura ha scritto:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- Makefile.am | 5 +---- libs/vkd3d-shader/hlsl_sm4.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 4 deletions(-)
diff --git a/Makefile.am b/Makefile.am index c66f7b50..8c775c4d 100644 --- a/Makefile.am +++ b/Makefile.am @@ -277,10 +277,7 @@ XFAIL_TESTS = \ tests/hlsl-vector-indexing-uniform.shader_test \ tests/math.shader_test \ tests/trigonometry.shader_test \ - tests/writemask-assignop-0.shader_test \ - tests/writemask-assignop-1.shader_test \ - tests/writemask-assignop-2.shader_test \ - tests/writemask-assignop-3.shader_test + tests/writemask-assignop-1.shader_test endif
if BUILD_DEMOS diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index ace2eb66..160da47f 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -561,6 +561,7 @@ struct sm4_register unsigned int idx_count; enum vkd3d_sm4_dimension dim; uint32_t immconst_uint[4]; + unsigned int mod; };
struct sm4_instruction @@ -710,6 +711,8 @@ static uint32_t sm4_register_order(const struct sm4_register *reg) if (reg->type == VKD3D_SM4_RT_IMMCONST) order += reg->dim == VKD3D_SM4_DIMENSION_VEC4 ? 4 : 1; order += reg->idx_count; + if (reg->mod) + ++order; return order; }
@@ -743,8 +746,13 @@ static void write_sm4_instruction(struct vkd3d_bytecode_buffer *buffer, const st token = sm4_encode_register(&instr->srcs[i].reg); token |= sm4_swizzle_type(instr->srcs[i].reg.type) << VKD3D_SM4_SWIZZLE_TYPE_SHIFT; token |= instr->srcs[i].swizzle << VKD3D_SM4_SWIZZLE_SHIFT; + if (instr->srcs[i].reg.mod) + token |= VKD3D_SM4_REGISTER_MODIFIER; put_u32(buffer, token);
+ if (instr->srcs[i].reg.mod) + put_u32(buffer, instr->srcs[i].reg.mod); + for (j = 0; j < instr->srcs[i].reg.idx_count; ++j) put_u32(buffer, instr->srcs[i].reg.idx[j]);
@@ -894,6 +902,26 @@ static void write_sm4_ret(struct vkd3d_bytecode_buffer *buffer) write_sm4_instruction(buffer, &instr); }
+static void write_sm4_unary_op(struct vkd3d_bytecode_buffer *buffer, enum vkd3d_sm4_opcode opcode, + const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src, unsigned int src_mod) +{ + struct sm4_instruction instr; + unsigned int writemask; + + memset(&instr, 0, sizeof(instr)); + instr.opcode = opcode; + + sm4_register_from_node(&instr.dst.reg, &instr.dst.writemask, dst); + instr.has_dst = 1; + + sm4_register_from_node(&instr.srcs[0].reg, &writemask, src); + instr.srcs[0].swizzle = hlsl_map_swizzle(hlsl_swizzle_from_writemask(writemask), instr.dst.writemask); + instr.srcs[0].reg.mod = src_mod; + instr.src_count = 1; + + write_sm4_instruction(buffer, &instr); +} + static void write_sm4_binary_op(struct vkd3d_bytecode_buffer *buffer, enum vkd3d_sm4_opcode opcode, const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src1, const struct hlsl_ir_node *src2) { @@ -953,6 +981,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, { switch (expr->op) { + case HLSL_OP1_NEG: + write_sm4_unary_op(buffer, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3D_SM4_REGISTER_MODIFIER_NEGATE); + break; + case HLSL_OP2_ADD: write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break;
Signed-off-by: Matteo Bruni <mbruni(a)codeweavers.com>
On Mon, Aug 30, 2021 at 7:21 PM Zebediah Figura <zfigura(a)codeweavers.com> wrote:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- Makefile.am | 5 +---- libs/vkd3d-shader/hlsl_sm4.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 4 deletions(-)
diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index ace2eb66..160da47f 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -561,6 +561,7 @@ struct sm4_register unsigned int idx_count; enum vkd3d_sm4_dimension dim; uint32_t immconst_uint[4]; + unsigned int mod; };
struct sm4_instruction @@ -710,6 +711,8 @@ static uint32_t sm4_register_order(const struct sm4_register *reg) if (reg->type == VKD3D_SM4_RT_IMMCONST) order += reg->dim == VKD3D_SM4_DIMENSION_VEC4 ? 4 : 1; order += reg->idx_count; + if (reg->mod) + ++order; return order; }
Something that came to my mind at least twice while looking at that function... Do you know e.g. by random looks to DXC's sources what the size of a register in the bytecode is supposed to be called? "Order" feels like a curious choice for a name. No need to go investigate, just wondering if you happened to see anything.
On 8/31/21 6:33 PM, Matteo Bruni wrote:
On Mon, Aug 30, 2021 at 7:21 PM Zebediah Figura <zfigura(a)codeweavers.com> wrote:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- Makefile.am | 5 +---- libs/vkd3d-shader/hlsl_sm4.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 4 deletions(-)
diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index ace2eb66..160da47f 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -561,6 +561,7 @@ struct sm4_register unsigned int idx_count; enum vkd3d_sm4_dimension dim; uint32_t immconst_uint[4]; + unsigned int mod; };
struct sm4_instruction @@ -710,6 +711,8 @@ static uint32_t sm4_register_order(const struct sm4_register *reg) if (reg->type == VKD3D_SM4_RT_IMMCONST) order += reg->dim == VKD3D_SM4_DIMENSION_VEC4 ? 4 : 1; order += reg->idx_count; + if (reg->mod) + ++order; return order; }
Something that came to my mind at least twice while looking at that function... Do you know e.g. by random looks to DXC's sources what the size of a register in the bytecode is supposed to be called? "Order" feels like a curious choice for a name.
No need to go investigate, just wondering if you happened to see anything.
I recently discovered that the SM4 format is effectively documented in headers released with the DDK, namely "d3d10tokenizedprogramformat.hpp" and similar. Anyway, their version of SM4_REGISTER_ORDER_* is D3D10_SB_OPERAND_INDEX_DIMENSION_*.
On Wed, Sep 1, 2021 at 3:24 AM Zebediah Figura (she/her) <zfigura(a)codeweavers.com> wrote:
On 8/31/21 6:33 PM, Matteo Bruni wrote:
On Mon, Aug 30, 2021 at 7:21 PM Zebediah Figura <zfigura(a)codeweavers.com> wrote:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- Makefile.am | 5 +---- libs/vkd3d-shader/hlsl_sm4.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 4 deletions(-)
diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index ace2eb66..160da47f 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -561,6 +561,7 @@ struct sm4_register unsigned int idx_count; enum vkd3d_sm4_dimension dim; uint32_t immconst_uint[4]; + unsigned int mod; };
struct sm4_instruction @@ -710,6 +711,8 @@ static uint32_t sm4_register_order(const struct sm4_register *reg) if (reg->type == VKD3D_SM4_RT_IMMCONST) order += reg->dim == VKD3D_SM4_DIMENSION_VEC4 ? 4 : 1; order += reg->idx_count; + if (reg->mod) + ++order; return order; }
Something that came to my mind at least twice while looking at that function... Do you know e.g. by random looks to DXC's sources what the size of a register in the bytecode is supposed to be called? "Order" feels like a curious choice for a name.
No need to go investigate, just wondering if you happened to see anything.
I recently discovered that the SM4 format is effectively documented in headers released with the DDK, namely "d3d10tokenizedprogramformat.hpp" and similar. Anyway, their version of SM4_REGISTER_ORDER_* is D3D10_SB_OPERAND_INDEX_DIMENSION_*.
Oh, excellent! I'll have a look at some point. No reason to change everything now, but going forward it's probably a good idea to make more informed decisions so to use the same naming principles as the original.
On Wed, 1 Sept 2021 at 10:27, Matteo Bruni <matteo.mystral(a)gmail.com> wrote:
On Wed, Sep 1, 2021 at 3:24 AM Zebediah Figura (she/her) <zfigura(a)codeweavers.com> wrote:
I recently discovered that the SM4 format is effectively documented in headers released with the DDK, namely "d3d10tokenizedprogramformat.hpp" and similar. Anyway, their version of SM4_REGISTER_ORDER_* is D3D10_SB_OPERAND_INDEX_DIMENSION_*.
Oh, excellent! I'll have a look at some point. No reason to change everything now, but going forward it's probably a good idea to make more informed decisions so to use the same naming principles as the original.
This has been around for a long time. We have generally avoided looking at the DDK, and I can't think of any cases where looking at it would have helped much.
On Wed, Sep 1, 2021 at 10:40 AM Henri Verbeet <hverbeet(a)gmail.com> wrote:
On Wed, 1 Sept 2021 at 10:27, Matteo Bruni <matteo.mystral(a)gmail.com> wrote:
On Wed, Sep 1, 2021 at 3:24 AM Zebediah Figura (she/her) <zfigura(a)codeweavers.com> wrote:
I recently discovered that the SM4 format is effectively documented in headers released with the DDK, namely "d3d10tokenizedprogramformat.hpp" and similar. Anyway, their version of SM4_REGISTER_ORDER_* is D3D10_SB_OPERAND_INDEX_DIMENSION_*.
Oh, excellent! I'll have a look at some point. No reason to change everything now, but going forward it's probably a good idea to make more informed decisions so to use the same naming principles as the original.
This has been around for a long time. We have generally avoided looking at the DDK, and I can't think of any cases where looking at it would have helped much.
If it's dangerous then sure. I'm certain we don't NEED anything that might be in there. Otherwise, I don't see the problem with at least using the correct names. It might just be that I recently discovered the effect framework's "preshaders" are supposed to be called FXLVM and I liked it a bit too much...
On 9/1/21 3:57 AM, Matteo Bruni wrote:
On Wed, Sep 1, 2021 at 10:40 AM Henri Verbeet <hverbeet(a)gmail.com> wrote:
On Wed, 1 Sept 2021 at 10:27, Matteo Bruni <matteo.mystral(a)gmail.com> wrote:
On Wed, Sep 1, 2021 at 3:24 AM Zebediah Figura (she/her) <zfigura(a)codeweavers.com> wrote:
I recently discovered that the SM4 format is effectively documented in headers released with the DDK, namely "d3d10tokenizedprogramformat.hpp" and similar. Anyway, their version of SM4_REGISTER_ORDER_* is D3D10_SB_OPERAND_INDEX_DIMENSION_*.
Oh, excellent! I'll have a look at some point. No reason to change everything now, but going forward it's probably a good idea to make more informed decisions so to use the same naming principles as the original.
This has been around for a long time. We have generally avoided looking at the DDK, and I can't think of any cases where looking at it would have helped much.
If it's dangerous then sure. I'm certain we don't NEED anything that might be in there.
Otherwise, I don't see the problem with at least using the correct names. It might just be that I recently discovered the effect framework's "preshaders" are supposed to be called FXLVM and I liked it a bit too much...
For what it's worth, the d3d12tokenizedprogramformat.hpp also from the DDK has the same definitions, and seems to be the same file as is shipped with the DXC source [1], and that at least is explicitly under an open license. [1] https://github.com/microsoft/DirectXShaderCompiler/blob/cb485263b75495015e2a...
Signed-off-by: Henri Verbeet <hverbeet(a)codeweavers.com>
It's certainly invalid to declare a vector of non-scalars, but it doesn't need to result in a syntax error. Indeed native d3dcompiler does not emit a syntax error in this case. Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl.y | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 7906356f..6a472c10 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -1731,7 +1731,6 @@ static struct list *declare_vars(struct hlsl_ctx *ctx, struct hlsl_type *basic_t %type <semantic> semantic -%type <type> base_type %type <type> field_type %type <type> named_struct_spec %type <type> unnamed_struct_spec @@ -2145,11 +2144,7 @@ input_mod: } type: - base_type - { - $$ = $1; - } - | KW_VECTOR '<' base_type ',' C_INTEGER '>' + KW_VECTOR '<' type ',' C_INTEGER '>' { if ($3->type != HLSL_CLASS_SCALAR) { @@ -2171,7 +2166,7 @@ type: $$ = hlsl_new_type(ctx, NULL, HLSL_CLASS_VECTOR, $3->base_type, $5, 1); } - | KW_MATRIX '<' base_type ',' C_INTEGER ',' C_INTEGER '>' + | KW_MATRIX '<' type ',' C_INTEGER ',' C_INTEGER '>' { if ($3->type != HLSL_CLASS_SCALAR) { @@ -2199,9 +2194,7 @@ type: $$ = hlsl_new_type(ctx, NULL, HLSL_CLASS_MATRIX, $3->base_type, $7, $5); } - -base_type: - KW_VOID + | KW_VOID { $$ = ctx->builtin_types.Void; } -- 2.33.0
Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> Il 30/08/21 19:21, Zebediah Figura ha scritto:
It's certainly invalid to declare a vector of non-scalars, but it doesn't need to result in a syntax error. Indeed native d3dcompiler does not emit a syntax error in this case.
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl.y | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-)
diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 7906356f..6a472c10 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -1731,7 +1731,6 @@ static struct list *declare_vars(struct hlsl_ctx *ctx, struct hlsl_type *basic_t
%type <semantic> semantic
-%type <type> base_type %type <type> field_type %type <type> named_struct_spec %type <type> unnamed_struct_spec @@ -2145,11 +2144,7 @@ input_mod: }
type: - base_type - { - $$ = $1; - } - | KW_VECTOR '<' base_type ',' C_INTEGER '>' + KW_VECTOR '<' type ',' C_INTEGER '>' { if ($3->type != HLSL_CLASS_SCALAR) { @@ -2171,7 +2166,7 @@ type:
$$ = hlsl_new_type(ctx, NULL, HLSL_CLASS_VECTOR, $3->base_type, $5, 1); } - | KW_MATRIX '<' base_type ',' C_INTEGER ',' C_INTEGER '>' + | KW_MATRIX '<' type ',' C_INTEGER ',' C_INTEGER '>' { if ($3->type != HLSL_CLASS_SCALAR) { @@ -2199,9 +2194,7 @@ type:
$$ = hlsl_new_type(ctx, NULL, HLSL_CLASS_MATRIX, $3->base_type, $7, $5); } - -base_type: - KW_VOID + | KW_VOID { $$ = ctx->builtin_types.Void; }
Signed-off-by: Matteo Bruni <mbruni(a)codeweavers.com> --- I guess the original choice was mostly a "defensive" one: it's generally better to start with a restrictive grammar and then remove constraints as necessary rather than the other way around.
On 8/31/21 6:31 PM, Matteo Bruni wrote:
Signed-off-by: Matteo Bruni <mbruni(a)codeweavers.com> --- I guess the original choice was mostly a "defensive" one: it's generally better to start with a restrictive grammar and then remove constraints as necessary rather than the other way around.
Sure. On the other hand, it's probably better to have less things result in a syntax error, so that you can emit more errors per run. But I don't mean to imply that the original code was unreasonable ;-)
Signed-off-by: Henri Verbeet <hverbeet(a)codeweavers.com>
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl.y | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 6a472c10..26ba4695 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -2836,7 +2836,7 @@ unary_expr: if ($2) { - hlsl_error(ctx, @3, VKD3D_SHADER_ERROR_HLSL_INVALID_MODIFIER, + hlsl_error(ctx, @2, VKD3D_SHADER_ERROR_HLSL_INVALID_MODIFIER, "Modifiers are not allowed on casts."); YYABORT; } -- 2.33.0
Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> Il 30/08/21 19:21, Zebediah Figura ha scritto:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl.y | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 6a472c10..26ba4695 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -2836,7 +2836,7 @@ unary_expr:
if ($2) { - hlsl_error(ctx, @3, VKD3D_SHADER_ERROR_HLSL_INVALID_MODIFIER, + hlsl_error(ctx, @2, VKD3D_SHADER_ERROR_HLSL_INVALID_MODIFIER, "Modifiers are not allowed on casts."); YYABORT; }
Signed-off-by: Matteo Bruni <mbruni(a)codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet(a)codeweavers.com>
Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> Il 30/08/21 19:21, Zebediah Figura ha scritto:
Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_sm4.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index 32a4eda1..af1660f1 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -957,6 +957,10 @@ static void write_sm4_expr(struct hlsl_ctx *ctx, write_sm4_binary_op(buffer, VKD3D_SM4_OP_ADD, &expr->node, arg1, arg2); break;
+ case HLSL_OP2_MUL: + write_sm4_binary_op(buffer, VKD3D_SM4_OP_MUL, &expr->node, arg1, arg2); + break; + default: hlsl_fixme(ctx, expr->node.loc, "SM4 float \"%s\" expression.", debug_hlsl_expr_op(expr->op)); break;
participants (7)
-
Giovanni Mascellani -
Henri Verbeet -
Henri Verbeet -
Matteo Bruni -
Matteo Bruni -
Zebediah Figura -
Zebediah Figura (she/her)