July 2022 - Wine-Devel - list.winehq.org

[PATCH 0/5] MR355: win32u: Move more display mode logic out of graphics drivers.

by Rémi Bernon

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/355

July 1, 2022

Re: [PATCH vkd3d 5/5] tests: Test arithmetic int operations without constant operands.

by Zebediah Figura

On 6/29/22 09:55, Giovanni Mascellani wrote: > diff --git a/tests/arithmetic-int-uniform.shader_test b/tests/arithmetic-int-uniform.shader_test > new file mode 100644 > index 00000000..9bd7f2ac > --- /dev/null > +++ b/tests/arithmetic-int-uniform.shader_test > @@ -0,0 +1,28 @@ > +[require] > +shader model >= 4.0 Passing "uniform int" requires sm4 (well, sort of), but the actual operations should be possible with sm1. Perhaps pass the inputs as floats and then cast them to int variables before performing arithmetic? > + > +[pixel shader] > +uniform int x, y; > + > +float4 main() : SV_TARGET > +{ > + return float4(x + y, x - y, x * y, x / y); > +} > + > +[test] > +uniform 0 int4 5 15 0 0 > +todo draw quad > +probe all rgba (20.0, -10.0, 75.0, 0.0) > + > +[pixel shader] > +uniform int x, y; > + > +float4 main() : SV_TARGET > +{ > + return float4(x % y, +x, -x, y / x); > +} > + > +[test] > +uniform 0 int4 5 15 0 0 > +todo draw quad > +probe all rgba (5.0, 5.0, -5.0, 3.0)

July 1, 2022

Re: [PATCH vkd3d 1/5] vkd3d-shader/hlsl: Write SM4 casts from bool.

by Zebediah Figura

On 6/29/22 09:55, Giovanni Mascellani wrote: > Signed-off-by: Giovanni Mascellani <gmascellani(a)codeweavers.com> > --- > libs/vkd3d-shader/hlsl_sm4.c | 44 ++++++++++++++++++++++++++++-- > tests/cast-to-float.shader_test | 2 +- > tests/cast-to-int.shader_test | 4 +-- > tests/cast-to-uint.shader_test | 4 +-- > tests/hlsl-bool-cast.shader_test | 18 ++++++++++++ > tests/logic-operations.shader_test | 6 ++-- > 6 files changed, 67 insertions(+), 11 deletions(-) > > diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c > index 79027169..23678498 100644 > --- a/libs/vkd3d-shader/hlsl_sm4.c > +++ b/libs/vkd3d-shader/hlsl_sm4.c > @@ -1459,6 +1459,44 @@ static bool type_is_float(const struct hlsl_type *type) > return type->base_type == HLSL_TYPE_FLOAT || type->base_type == HLSL_TYPE_HALF; > } > > +static void write_sm4_cast_from_bool(struct hlsl_ctx *ctx, > + struct vkd3d_bytecode_buffer *buffer, const struct hlsl_ir_expr *expr, > + const struct hlsl_ir_node *arg) > +{ > + struct sm4_instruction instr; > + uint32_t c; > + > + switch (expr->node.data_type->base_type) > + { > + case HLSL_TYPE_FLOAT: Should this include HLSL_TYPE_HALF as well? > + c = 0x3f800000; > + break; Maybe not worth worrying about, but the hardcoded number could be avoided by doing something like union { uint32_t u; float f; } c; ... c.f = 1.0f; > + > + case HLSL_TYPE_INT: > + case HLSL_TYPE_UINT: > + c = 1; > + break; > + > + default: > + assert(0); We don't need to handle doubles, but I'm not sure they should result in an assert either. > + } > + > + memset(&instr, 0, sizeof(instr)); > + instr.opcode = VKD3D_SM4_OP_AND; > + > + sm4_dst_from_node(&instr.dsts[0], &expr->node); > + instr.dst_count = 1; > + > + sm4_src_from_node(&instr.srcs[0], arg, instr.dsts[0].writemask); > + instr.srcs[1].swizzle_type = VKD3D_SM4_SWIZZLE_NONE; > + instr.srcs[1].reg.type = VKD3D_SM4_RT_IMMCONST; > + instr.srcs[1].reg.dim = VKD3D_SM4_DIMENSION_SCALAR; > + instr.srcs[1].reg.immconst_uint[0] = c; > + instr.src_count = 2; > + > + write_sm4_instruction(buffer, &instr); > +} I kind of wonder if this should be a lowering pass instead. It wouldn't apply to sm1, but it would (maybe?) be a bit easier to write, and it would apply to a hypothetical sm6 backend, if we indeed write our own instead of leveraging the Microsoft compiler. Fine to leave it as is for now, I think, but food for thought. > + > static void write_sm4_cast(struct hlsl_ctx *ctx, > struct vkd3d_bytecode_buffer *buffer, const struct hlsl_ir_expr *expr) > { > @@ -1488,7 +1526,7 @@ static void write_sm4_cast(struct hlsl_ctx *ctx, > break; > > case HLSL_TYPE_BOOL: > - hlsl_fixme(ctx, &expr->node.loc, "SM4 cast from bool to float."); > + write_sm4_cast_from_bool(ctx, buffer, expr, arg1); > break; > > case HLSL_TYPE_DOUBLE: > @@ -1514,7 +1552,7 @@ static void write_sm4_cast(struct hlsl_ctx *ctx, > break; > > case HLSL_TYPE_BOOL: > - hlsl_fixme(ctx, &expr->node.loc, "SM4 cast from bool to int."); > + write_sm4_cast_from_bool(ctx, buffer, expr, arg1); > break; > > case HLSL_TYPE_DOUBLE: > @@ -1540,7 +1578,7 @@ static void write_sm4_cast(struct hlsl_ctx *ctx, > break; > > case HLSL_TYPE_BOOL: > - hlsl_fixme(ctx, &expr->node.loc, "SM4 cast from bool to uint."); > + write_sm4_cast_from_bool(ctx, buffer, expr, arg1); > break; > > case HLSL_TYPE_DOUBLE: Could we just pass the mask from here instead, and skip the switch inside of write_sm4_cast_from_bool()?

July 1, 2022

Re: [PATCH v3 0/3] MR330: Tests for properties. - closed

by Ziqing Hui (＠zhui)

This merge request was closed by Ziqing Hui. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/330

July 1, 2022

[PATCH 1/1] msvcp140_2: Add stub dll.

by Paul Gofman

From: Paul Gofman <pgofman(a)codeweavers.com> --- configure.ac | 1 + dlls/msvcp140_2/Makefile.in | 1 + dlls/msvcp140_2/msvcp140_2.spec | 44 +++++++++++++++++++++++++++++++++ 3 files changed, 46 insertions(+) create mode 100644 dlls/msvcp140_2/Makefile.in create mode 100644 dlls/msvcp140_2/msvcp140_2.spec diff --git a/configure.ac b/configure.ac index 9ad1b19164c..d5791bab7b1 100644 --- a/configure.ac +++ b/configure.ac @@ -2825,6 +2825,7 @@ WINE_CONFIG_MAKEFILE(dlls/msvcp140) WINE_CONFIG_MAKEFILE(dlls/msvcp140/tests) WINE_CONFIG_MAKEFILE(dlls/msvcp140_1) WINE_CONFIG_MAKEFILE(dlls/msvcp140_1/tests) +WINE_CONFIG_MAKEFILE(dlls/msvcp140_2) WINE_CONFIG_MAKEFILE(dlls/msvcp140_atomic_wait) WINE_CONFIG_MAKEFILE(dlls/msvcp140_atomic_wait/tests) WINE_CONFIG_MAKEFILE(dlls/msvcp60) diff --git a/dlls/msvcp140_2/Makefile.in b/dlls/msvcp140_2/Makefile.in new file mode 100644 index 00000000000..5b4865db844 --- /dev/null +++ b/dlls/msvcp140_2/Makefile.in @@ -0,0 +1 @@ +MODULE = msvcp140_2.dll diff --git a/dlls/msvcp140_2/msvcp140_2.spec b/dlls/msvcp140_2/msvcp140_2.spec new file mode 100644 index 00000000000..c59da9aba9c --- /dev/null +++ b/dlls/msvcp140_2/msvcp140_2.spec @@ -0,0 +1,44 @@ +@ stub __std_smf_assoc_laguerre +@ stub __std_smf_assoc_laguerref +@ stub __std_smf_assoc_legendre +@ stub __std_smf_assoc_legendref +@ stub __std_smf_beta +@ stub __std_smf_betaf +@ stub __std_smf_comp_ellint_1 +@ stub __std_smf_comp_ellint_1f +@ stub __std_smf_comp_ellint_2 +@ stub __std_smf_comp_ellint_2f +@ stub __std_smf_comp_ellint_3 +@ stub __std_smf_comp_ellint_3f +@ stub __std_smf_cyl_bessel_i +@ stub __std_smf_cyl_bessel_if +@ stub __std_smf_cyl_bessel_j +@ stub __std_smf_cyl_bessel_jf +@ stub __std_smf_cyl_bessel_k +@ stub __std_smf_cyl_bessel_kf +@ stub __std_smf_cyl_neumann +@ stub __std_smf_cyl_neumannf +@ stub __std_smf_ellint_1 +@ stub __std_smf_ellint_1f +@ stub __std_smf_ellint_2 +@ stub __std_smf_ellint_2f +@ stub __std_smf_ellint_3 +@ stub __std_smf_ellint_3f +@ stub __std_smf_expint +@ stub __std_smf_expintf +@ stub __std_smf_hermite +@ stub __std_smf_hermitef +@ stub __std_smf_hypot3 +@ stub __std_smf_hypot3f +@ stub __std_smf_laguerre +@ stub __std_smf_laguerref +@ stub __std_smf_legendre +@ stub __std_smf_legendref +@ stub __std_smf_riemann_zeta +@ stub __std_smf_riemann_zetaf +@ stub __std_smf_sph_bessel +@ stub __std_smf_sph_besself +@ stub __std_smf_sph_legendre +@ stub __std_smf_sph_legendref +@ stub __std_smf_sph_neumann +@ stub __std_smf_sph_neumannf -- GitLab https://gitlab.winehq.org/wine/wine/-/merge_requests/354

July 1, 2022

[PATCH 0/1] MR354: msvcp140_2: Add stub dll.

by Paul Gofman (＠gofman)

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/354

July 1, 2022