4 TGSI, Tungsten Graphics Shader Infrastructure, is an intermediate language
5 for describing shaders. Since Gallium is inherently shaderful, shaders are
6 an important part of the API. TGSI is the only intermediate representation
12 All TGSI instructions, known as *opcodes*, operate on arbitrary-precision
13 floating-point four-component vectors. An opcode may have up to one
14 destination register, known as *dst*, and between zero and three source
15 registers, called *src0* through *src2*, or simply *src* if there is only
18 Some instructions, like :opcode:`I2F`, permit re-interpretation of vector
19 components as integers. Other instructions permit using registers as
20 two-component vectors with double precision; see :ref:`doubleopcodes`.
22 When an instruction has a scalar result, the result is usually copied into
23 each of the components of *dst*. When this happens, the result is said to be
24 *replicated* to *dst*. :opcode:`RCP` is one such instruction.
29 TGSI supports modifiers on inputs (as well as saturate and precise modifier
32 For arithmetic instruction having a precise modifier certain optimizations
33 which may alter the result are disallowed. Example: *add(mul(a,b),c)* can't be
34 optimized to TGSI_OPCODE_MAD, because some hardware only supports the fused
37 For inputs which have a floating point type, both absolute value and
38 negation modifiers are supported (with absolute value being applied
39 first). The only source of TGSI_OPCODE_MOV and the second and third
40 sources of TGSI_OPCODE_UCMP are considered to have float type for
43 For inputs which have signed or unsigned type only the negate modifier is
50 ^^^^^^^^^^^^^^^^^^^^^^^^^
52 These opcodes are guaranteed to be available regardless of the driver being
55 .. opcode:: ARL - Address Register Load
59 dst.x = (int) \lfloor src.x\rfloor
61 dst.y = (int) \lfloor src.y\rfloor
63 dst.z = (int) \lfloor src.z\rfloor
65 dst.w = (int) \lfloor src.w\rfloor
68 .. opcode:: MOV - Move
81 .. opcode:: LIT - Light Coefficients
86 dst.y &= max(src.x, 0) \\
87 dst.z &= (src.x > 0) ? max(src.y, 0)^{clamp(src.w, -128, 128))} : 0 \\
91 .. opcode:: RCP - Reciprocal
93 This instruction replicates its result.
100 .. opcode:: RSQ - Reciprocal Square Root
102 This instruction replicates its result. The results are undefined for src <= 0.
106 dst = \frac{1}{\sqrt{src.x}}
109 .. opcode:: SQRT - Square Root
111 This instruction replicates its result. The results are undefined for src < 0.
118 .. opcode:: EXP - Approximate Exponential Base 2
122 dst.x &= 2^{\lfloor src.x\rfloor} \\
123 dst.y &= src.x - \lfloor src.x\rfloor \\
124 dst.z &= 2^{src.x} \\
128 .. opcode:: LOG - Approximate Logarithm Base 2
132 dst.x &= \lfloor\log_2{|src.x|}\rfloor \\
133 dst.y &= \frac{|src.x|}{2^{\lfloor\log_2{|src.x|}\rfloor}} \\
134 dst.z &= \log_2{|src.x|} \\
138 .. opcode:: MUL - Multiply
142 dst.x = src0.x \times src1.x
144 dst.y = src0.y \times src1.y
146 dst.z = src0.z \times src1.z
148 dst.w = src0.w \times src1.w
151 .. opcode:: ADD - Add
155 dst.x = src0.x + src1.x
157 dst.y = src0.y + src1.y
159 dst.z = src0.z + src1.z
161 dst.w = src0.w + src1.w
164 .. opcode:: DP3 - 3-component Dot Product
166 This instruction replicates its result.
170 dst = src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z
173 .. opcode:: DP4 - 4-component Dot Product
175 This instruction replicates its result.
179 dst = src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z + src0.w \times src1.w
182 .. opcode:: DST - Distance Vector
187 dst.y &= src0.y \times src1.y\\
192 .. opcode:: MIN - Minimum
196 dst.x = min(src0.x, src1.x)
198 dst.y = min(src0.y, src1.y)
200 dst.z = min(src0.z, src1.z)
202 dst.w = min(src0.w, src1.w)
205 .. opcode:: MAX - Maximum
209 dst.x = max(src0.x, src1.x)
211 dst.y = max(src0.y, src1.y)
213 dst.z = max(src0.z, src1.z)
215 dst.w = max(src0.w, src1.w)
218 .. opcode:: SLT - Set On Less Than
222 dst.x = (src0.x < src1.x) ? 1.0F : 0.0F
224 dst.y = (src0.y < src1.y) ? 1.0F : 0.0F
226 dst.z = (src0.z < src1.z) ? 1.0F : 0.0F
228 dst.w = (src0.w < src1.w) ? 1.0F : 0.0F
231 .. opcode:: SGE - Set On Greater Equal Than
235 dst.x = (src0.x >= src1.x) ? 1.0F : 0.0F
237 dst.y = (src0.y >= src1.y) ? 1.0F : 0.0F
239 dst.z = (src0.z >= src1.z) ? 1.0F : 0.0F
241 dst.w = (src0.w >= src1.w) ? 1.0F : 0.0F
244 .. opcode:: MAD - Multiply And Add
246 Perform a * b + c. The implementation is free to decide whether there is an
247 intermediate rounding step or not.
251 dst.x = src0.x \times src1.x + src2.x
253 dst.y = src0.y \times src1.y + src2.y
255 dst.z = src0.z \times src1.z + src2.z
257 dst.w = src0.w \times src1.w + src2.w
260 .. opcode:: LRP - Linear Interpolate
264 dst.x = src0.x \times src1.x + (1 - src0.x) \times src2.x
266 dst.y = src0.y \times src1.y + (1 - src0.y) \times src2.y
268 dst.z = src0.z \times src1.z + (1 - src0.z) \times src2.z
270 dst.w = src0.w \times src1.w + (1 - src0.w) \times src2.w
273 .. opcode:: FMA - Fused Multiply-Add
275 Perform a * b + c with no intermediate rounding step.
279 dst.x = src0.x \times src1.x + src2.x
281 dst.y = src0.y \times src1.y + src2.y
283 dst.z = src0.z \times src1.z + src2.z
285 dst.w = src0.w \times src1.w + src2.w
288 .. opcode:: FRC - Fraction
292 dst.x = src.x - \lfloor src.x\rfloor
294 dst.y = src.y - \lfloor src.y\rfloor
296 dst.z = src.z - \lfloor src.z\rfloor
298 dst.w = src.w - \lfloor src.w\rfloor
301 .. opcode:: FLR - Floor
305 dst.x = \lfloor src.x\rfloor
307 dst.y = \lfloor src.y\rfloor
309 dst.z = \lfloor src.z\rfloor
311 dst.w = \lfloor src.w\rfloor
314 .. opcode:: ROUND - Round
327 .. opcode:: EX2 - Exponential Base 2
329 This instruction replicates its result.
336 .. opcode:: LG2 - Logarithm Base 2
338 This instruction replicates its result.
345 .. opcode:: POW - Power
347 This instruction replicates its result.
351 dst = src0.x^{src1.x}
354 .. opcode:: LDEXP - Multiply Number by Integral Power of 2
360 dst.x = src0.x * 2^{src1.x}
361 dst.y = src0.y * 2^{src1.y}
362 dst.z = src0.z * 2^{src1.z}
363 dst.w = src0.w * 2^{src1.w}
366 .. opcode:: COS - Cosine
368 This instruction replicates its result.
375 .. opcode:: DDX, DDX_FINE - Derivative Relative To X
377 The fine variant is only used when ``PIPE_CAP_TGSI_FS_FINE_DERIVATIVE`` is
378 advertised. When it is, the fine version guarantees one derivative per row
379 while DDX is allowed to be the same for the entire 2x2 quad.
383 dst.x = partialx(src.x)
385 dst.y = partialx(src.y)
387 dst.z = partialx(src.z)
389 dst.w = partialx(src.w)
392 .. opcode:: DDY, DDY_FINE - Derivative Relative To Y
394 The fine variant is only used when ``PIPE_CAP_TGSI_FS_FINE_DERIVATIVE`` is
395 advertised. When it is, the fine version guarantees one derivative per column
396 while DDY is allowed to be the same for the entire 2x2 quad.
400 dst.x = partialy(src.x)
402 dst.y = partialy(src.y)
404 dst.z = partialy(src.z)
406 dst.w = partialy(src.w)
409 .. opcode:: PK2H - Pack Two 16-bit Floats
411 This instruction replicates its result.
415 dst = f32\_to\_f16(src.x) | f32\_to\_f16(src.y) << 16
418 .. opcode:: PK2US - Pack Two Unsigned 16-bit Scalars
420 This instruction replicates its result.
424 dst = f32\_to\_unorm16(src.x) | f32\_to\_unorm16(src.y) << 16
427 .. opcode:: PK4B - Pack Four Signed 8-bit Scalars
429 This instruction replicates its result.
433 dst = f32\_to\_snorm8(src.x) |
434 (f32\_to\_snorm8(src.y) << 8) |
435 (f32\_to\_snorm8(src.z) << 16) |
436 (f32\_to\_snorm8(src.w) << 24)
439 .. opcode:: PK4UB - Pack Four Unsigned 8-bit Scalars
441 This instruction replicates its result.
445 dst = f32\_to\_unorm8(src.x) |
446 (f32\_to\_unorm8(src.y) << 8) |
447 (f32\_to\_unorm8(src.z) << 16) |
448 (f32\_to\_unorm8(src.w) << 24)
451 .. opcode:: SEQ - Set On Equal
455 dst.x = (src0.x == src1.x) ? 1.0F : 0.0F
457 dst.y = (src0.y == src1.y) ? 1.0F : 0.0F
459 dst.z = (src0.z == src1.z) ? 1.0F : 0.0F
461 dst.w = (src0.w == src1.w) ? 1.0F : 0.0F
464 .. opcode:: SGT - Set On Greater Than
468 dst.x = (src0.x > src1.x) ? 1.0F : 0.0F
470 dst.y = (src0.y > src1.y) ? 1.0F : 0.0F
472 dst.z = (src0.z > src1.z) ? 1.0F : 0.0F
474 dst.w = (src0.w > src1.w) ? 1.0F : 0.0F
477 .. opcode:: SIN - Sine
479 This instruction replicates its result.
486 .. opcode:: SLE - Set On Less Equal Than
490 dst.x = (src0.x <= src1.x) ? 1.0F : 0.0F
492 dst.y = (src0.y <= src1.y) ? 1.0F : 0.0F
494 dst.z = (src0.z <= src1.z) ? 1.0F : 0.0F
496 dst.w = (src0.w <= src1.w) ? 1.0F : 0.0F
499 .. opcode:: SNE - Set On Not Equal
503 dst.x = (src0.x != src1.x) ? 1.0F : 0.0F
505 dst.y = (src0.y != src1.y) ? 1.0F : 0.0F
507 dst.z = (src0.z != src1.z) ? 1.0F : 0.0F
509 dst.w = (src0.w != src1.w) ? 1.0F : 0.0F
512 .. opcode:: TEX - Texture Lookup
514 for array textures src0.y contains the slice for 1D,
515 and src0.z contain the slice for 2D.
517 for shadow textures with no arrays (and not cube map),
518 src0.z contains the reference value.
520 for shadow textures with arrays, src0.z contains
521 the reference value for 1D arrays, and src0.w contains
522 the reference value for 2D arrays and cube maps.
524 for cube map array shadow textures, the reference value
525 cannot be passed in src0.w, and TEX2 must be used instead.
531 shadow_ref = src0.z or src0.w (optional)
535 dst = texture\_sample(unit, coord, shadow_ref)
538 .. opcode:: TEX2 - Texture Lookup (for shadow cube map arrays only)
540 this is the same as TEX, but uses another reg to encode the
551 dst = texture\_sample(unit, coord, shadow_ref)
556 .. opcode:: TXD - Texture Lookup with Derivatives
568 dst = texture\_sample\_deriv(unit, coord, ddx, ddy)
571 .. opcode:: TXP - Projective Texture Lookup
575 coord.x = src0.x / src0.w
577 coord.y = src0.y / src0.w
579 coord.z = src0.z / src0.w
585 dst = texture\_sample(unit, coord)
588 .. opcode:: UP2H - Unpack Two 16-Bit Floats
592 dst.x = f16\_to\_f32(src0.x \& 0xffff)
594 dst.y = f16\_to\_f32(src0.x >> 16)
596 dst.z = f16\_to\_f32(src0.x \& 0xffff)
598 dst.w = f16\_to\_f32(src0.x >> 16)
602 Considered for removal.
604 .. opcode:: UP2US - Unpack Two Unsigned 16-Bit Scalars
610 Considered for removal.
612 .. opcode:: UP4B - Unpack Four Signed 8-Bit Values
618 Considered for removal.
620 .. opcode:: UP4UB - Unpack Four Unsigned 8-Bit Scalars
626 Considered for removal.
629 .. opcode:: ARR - Address Register Load With Round
633 dst.x = (int) round(src.x)
635 dst.y = (int) round(src.y)
637 dst.z = (int) round(src.z)
639 dst.w = (int) round(src.w)
642 .. opcode:: SSG - Set Sign
646 dst.x = (src.x > 0) ? 1 : (src.x < 0) ? -1 : 0
648 dst.y = (src.y > 0) ? 1 : (src.y < 0) ? -1 : 0
650 dst.z = (src.z > 0) ? 1 : (src.z < 0) ? -1 : 0
652 dst.w = (src.w > 0) ? 1 : (src.w < 0) ? -1 : 0
655 .. opcode:: CMP - Compare
659 dst.x = (src0.x < 0) ? src1.x : src2.x
661 dst.y = (src0.y < 0) ? src1.y : src2.y
663 dst.z = (src0.z < 0) ? src1.z : src2.z
665 dst.w = (src0.w < 0) ? src1.w : src2.w
668 .. opcode:: KILL_IF - Conditional Discard
670 Conditional discard. Allowed in fragment shaders only.
674 if (src.x < 0 || src.y < 0 || src.z < 0 || src.w < 0)
679 .. opcode:: KILL - Discard
681 Unconditional discard. Allowed in fragment shaders only.
684 .. opcode:: DEMOTE - Demote Invocation to a Helper
686 This demotes the current invocation to a helper, but continues
687 execution (while KILL may or may not terminate the
688 invocation). After this runs, all the usual helper invocation rules
689 apply about discarding buffer and render target writes. This is
690 useful for having accurate derivatives in the other invocations
691 which have not been demoted.
693 Allowed in fragment shaders only.
696 .. opcode:: READ_HELPER - Reads Invocation Helper Status
698 This is identical to ``TGSI_SEMANTIC_HELPER_INVOCATION``, except
699 this will read the current value, which might change as a result of
700 a ``DEMOTE`` instruction.
702 Allowed in fragment shaders only.
705 .. opcode:: TXB - Texture Lookup With Bias
707 for cube map array textures and shadow cube maps, the bias value
708 cannot be passed in src0.w, and TXB2 must be used instead.
710 if the target is a shadow texture, the reference value is always
711 in src.z (this prevents shadow 3d and shadow 2d arrays from
712 using this instruction, but this is not needed).
728 dst = texture\_sample(unit, coord, bias)
731 .. opcode:: TXB2 - Texture Lookup With Bias (some cube maps only)
733 this is the same as TXB, but uses another reg to encode the
734 lod bias value for cube map arrays and shadow cube maps.
735 Presumably shadow 2d arrays and shadow 3d targets could use
736 this encoding too, but this is not legal.
738 if the target is a shadow cube map array, the reference value is in
749 dst = texture\_sample(unit, coord, bias)
752 .. opcode:: DIV - Divide
756 dst.x = \frac{src0.x}{src1.x}
758 dst.y = \frac{src0.y}{src1.y}
760 dst.z = \frac{src0.z}{src1.z}
762 dst.w = \frac{src0.w}{src1.w}
765 .. opcode:: DP2 - 2-component Dot Product
767 This instruction replicates its result.
771 dst = src0.x \times src1.x + src0.y \times src1.y
774 .. opcode:: TEX_LZ - Texture Lookup With LOD = 0
776 This is the same as TXL with LOD = 0. Like every texture opcode, it obeys
777 pipe_sampler_view::u.tex.first_level and pipe_sampler_state::min_lod.
778 There is no way to override those two in shaders.
794 dst = texture\_sample(unit, coord, lod)
797 .. opcode:: TXL - Texture Lookup With explicit LOD
799 for cube map array textures, the explicit lod value
800 cannot be passed in src0.w, and TXL2 must be used instead.
802 if the target is a shadow texture, the reference value is always
803 in src.z (this prevents shadow 3d / 2d array / cube targets from
804 using this instruction, but this is not needed).
820 dst = texture\_sample(unit, coord, lod)
823 .. opcode:: TXL2 - Texture Lookup With explicit LOD (for cube map arrays only)
825 this is the same as TXL, but uses another reg to encode the
827 Presumably shadow 3d / 2d array / cube targets could use
828 this encoding too, but this is not legal.
830 if the target is a shadow cube map array, the reference value is in
841 dst = texture\_sample(unit, coord, lod)
845 ^^^^^^^^^^^^^^^^^^^^^^^^
847 These opcodes are primarily provided for special-use computational shaders.
848 Support for these opcodes indicated by a special pipe capability bit (TBD).
850 XXX doesn't look like most of the opcodes really belong here.
852 .. opcode:: CEIL - Ceiling
856 dst.x = \lceil src.x\rceil
858 dst.y = \lceil src.y\rceil
860 dst.z = \lceil src.z\rceil
862 dst.w = \lceil src.w\rceil
865 .. opcode:: TRUNC - Truncate
878 .. opcode:: MOD - Modulus
882 dst.x = src0.x \bmod src1.x
884 dst.y = src0.y \bmod src1.y
886 dst.z = src0.z \bmod src1.z
888 dst.w = src0.w \bmod src1.w
891 .. opcode:: UARL - Integer Address Register Load
893 Moves the contents of the source register, assumed to be an integer, into the
894 destination register, which is assumed to be an address (ADDR) register.
897 .. opcode:: TXF - Texel Fetch
899 As per NV_gpu_shader4, extract a single texel from a specified texture
900 image or PIPE_BUFFER resource. The source sampler may not be a CUBE or
902 four-component signed integer vector used to identify the single texel
903 accessed. 3 components + level. If the texture is multisampled, then
904 the fourth component indicates the sample, not the mipmap level.
905 Just like texture instructions, an optional
906 offset vector is provided, which is subject to various driver restrictions
907 (regarding range, source of offsets). This instruction ignores the sampler
910 TXF(uint_vec coord, int_vec offset).
913 .. opcode:: TXQ - Texture Size Query
915 As per NV_gpu_program4, retrieve the dimensions of the texture depending on
916 the target. For 1D (width), 2D/RECT/CUBE (width, height), 3D (width, height,
917 depth), 1D array (width, layers), 2D array (width, height, layers).
918 Also return the number of accessible levels (last_level - first_level + 1)
921 For components which don't return a resource dimension, their value
928 dst.x = texture\_width(unit, lod)
930 dst.y = texture\_height(unit, lod)
932 dst.z = texture\_depth(unit, lod)
934 dst.w = texture\_levels(unit)
937 .. opcode:: TXQS - Texture Samples Query
939 This retrieves the number of samples in the texture, and stores it
940 into the x component as an unsigned integer. The other components are
941 undefined. If the texture is not multisampled, this function returns
942 (1, undef, undef, undef).
946 dst.x = texture\_samples(unit)
949 .. opcode:: TG4 - Texture Gather
951 As per ARB_texture_gather, gathers the four texels to be used in a bi-linear
952 filtering operation and packs them into a single register. Only works with
953 2D, 2D array, cubemaps, and cubemaps arrays. For 2D textures, only the
954 addressing modes of the sampler and the top level of any mip pyramid are
955 used. Set W to zero. It behaves like the TEX instruction, but a filtered
956 sample is not generated. The four samples that contribute to filtering are
957 placed into xyzw in clockwise order, starting with the (u,v) texture
958 coordinate delta at the following locations (-, +), (+, +), (+, -), (-, -),
959 where the magnitude of the deltas are half a texel.
961 PIPE_CAP_TEXTURE_SM5 enhances this instruction to support shadow per-sample
962 depth compares, single component selection, and a non-constant offset. It
963 doesn't allow support for the GL independent offset to get i0,j0. This would
964 require another CAP is hw can do it natively. For now we lower that before
967 PIPE_CAP_TGSI_TG4_COMPONENT_IN_SWIZZLE changes the encoding so that component
968 is stored in the sampler source swizzle x.
974 (without TGSI_TG4_COMPONENT_IN_SWIZZLE)
977 dst = texture\_gather4 (unit, coord, component)
979 (with TGSI_TG4_COMPONENT_IN_SWIZZLE)
980 dst = texture\_gather4 (unit, coord)
981 component is encoded in sampler swizzle.
983 (with SM5 - cube array shadow)
991 dst = texture\_gather (uint, coord, compare)
993 .. opcode:: LODQ - level of detail query
995 Compute the LOD information that the texture pipe would use to access the
996 texture. The Y component contains the computed LOD lambda_prime. The X
997 component contains the LOD that will be accessed, based on min/max lod's
1004 dst.xy = lodq(uint, coord);
1006 .. opcode:: CLOCK - retrieve the current shader time
1008 Invoking this instruction multiple times in the same shader should
1009 cause monotonically increasing values to be returned. The values
1010 are implicitly 64-bit, so if fewer than 64 bits of precision are
1011 available, to provide expected wraparound semantics, the value
1012 should be shifted up so that the most significant bit of the time
1013 is the most significant bit of the 64-bit value.
1021 ^^^^^^^^^^^^^^^^^^^^^^^^
1022 These opcodes are used for integer operations.
1023 Support for these opcodes indicated by PIPE_SHADER_CAP_INTEGERS (all of them?)
1026 .. opcode:: I2F - Signed Integer To Float
1028 Rounding is unspecified (round to nearest even suggested).
1032 dst.x = (float) src.x
1034 dst.y = (float) src.y
1036 dst.z = (float) src.z
1038 dst.w = (float) src.w
1041 .. opcode:: U2F - Unsigned Integer To Float
1043 Rounding is unspecified (round to nearest even suggested).
1047 dst.x = (float) src.x
1049 dst.y = (float) src.y
1051 dst.z = (float) src.z
1053 dst.w = (float) src.w
1056 .. opcode:: F2I - Float to Signed Integer
1058 Rounding is towards zero (truncate).
1059 Values outside signed range (including NaNs) produce undefined results.
1072 .. opcode:: F2U - Float to Unsigned Integer
1074 Rounding is towards zero (truncate).
1075 Values outside unsigned range (including NaNs) produce undefined results.
1079 dst.x = (unsigned) src.x
1081 dst.y = (unsigned) src.y
1083 dst.z = (unsigned) src.z
1085 dst.w = (unsigned) src.w
1088 .. opcode:: UADD - Integer Add
1090 This instruction works the same for signed and unsigned integers.
1091 The low 32bit of the result is returned.
1095 dst.x = src0.x + src1.x
1097 dst.y = src0.y + src1.y
1099 dst.z = src0.z + src1.z
1101 dst.w = src0.w + src1.w
1104 .. opcode:: UMAD - Integer Multiply And Add
1106 This instruction works the same for signed and unsigned integers.
1107 The multiplication returns the low 32bit (as does the result itself).
1111 dst.x = src0.x \times src1.x + src2.x
1113 dst.y = src0.y \times src1.y + src2.y
1115 dst.z = src0.z \times src1.z + src2.z
1117 dst.w = src0.w \times src1.w + src2.w
1120 .. opcode:: UMUL - Integer Multiply
1122 This instruction works the same for signed and unsigned integers.
1123 The low 32bit of the result is returned.
1127 dst.x = src0.x \times src1.x
1129 dst.y = src0.y \times src1.y
1131 dst.z = src0.z \times src1.z
1133 dst.w = src0.w \times src1.w
1136 .. opcode:: IMUL_HI - Signed Integer Multiply High Bits
1138 The high 32bits of the multiplication of 2 signed integers are returned.
1142 dst.x = (src0.x \times src1.x) >> 32
1144 dst.y = (src0.y \times src1.y) >> 32
1146 dst.z = (src0.z \times src1.z) >> 32
1148 dst.w = (src0.w \times src1.w) >> 32
1151 .. opcode:: UMUL_HI - Unsigned Integer Multiply High Bits
1153 The high 32bits of the multiplication of 2 unsigned integers are returned.
1157 dst.x = (src0.x \times src1.x) >> 32
1159 dst.y = (src0.y \times src1.y) >> 32
1161 dst.z = (src0.z \times src1.z) >> 32
1163 dst.w = (src0.w \times src1.w) >> 32
1166 .. opcode:: IDIV - Signed Integer Division
1168 TBD: behavior for division by zero.
1172 dst.x = \frac{src0.x}{src1.x}
1174 dst.y = \frac{src0.y}{src1.y}
1176 dst.z = \frac{src0.z}{src1.z}
1178 dst.w = \frac{src0.w}{src1.w}
1181 .. opcode:: UDIV - Unsigned Integer Division
1183 For division by zero, 0xffffffff is returned.
1187 dst.x = \frac{src0.x}{src1.x}
1189 dst.y = \frac{src0.y}{src1.y}
1191 dst.z = \frac{src0.z}{src1.z}
1193 dst.w = \frac{src0.w}{src1.w}
1196 .. opcode:: UMOD - Unsigned Integer Remainder
1198 If second arg is zero, 0xffffffff is returned.
1202 dst.x = src0.x \bmod src1.x
1204 dst.y = src0.y \bmod src1.y
1206 dst.z = src0.z \bmod src1.z
1208 dst.w = src0.w \bmod src1.w
1211 .. opcode:: NOT - Bitwise Not
1224 .. opcode:: AND - Bitwise And
1228 dst.x = src0.x \& src1.x
1230 dst.y = src0.y \& src1.y
1232 dst.z = src0.z \& src1.z
1234 dst.w = src0.w \& src1.w
1237 .. opcode:: OR - Bitwise Or
1241 dst.x = src0.x | src1.x
1243 dst.y = src0.y | src1.y
1245 dst.z = src0.z | src1.z
1247 dst.w = src0.w | src1.w
1250 .. opcode:: XOR - Bitwise Xor
1254 dst.x = src0.x \oplus src1.x
1256 dst.y = src0.y \oplus src1.y
1258 dst.z = src0.z \oplus src1.z
1260 dst.w = src0.w \oplus src1.w
1263 .. opcode:: IMAX - Maximum of Signed Integers
1267 dst.x = max(src0.x, src1.x)
1269 dst.y = max(src0.y, src1.y)
1271 dst.z = max(src0.z, src1.z)
1273 dst.w = max(src0.w, src1.w)
1276 .. opcode:: UMAX - Maximum of Unsigned Integers
1280 dst.x = max(src0.x, src1.x)
1282 dst.y = max(src0.y, src1.y)
1284 dst.z = max(src0.z, src1.z)
1286 dst.w = max(src0.w, src1.w)
1289 .. opcode:: IMIN - Minimum of Signed Integers
1293 dst.x = min(src0.x, src1.x)
1295 dst.y = min(src0.y, src1.y)
1297 dst.z = min(src0.z, src1.z)
1299 dst.w = min(src0.w, src1.w)
1302 .. opcode:: UMIN - Minimum of Unsigned Integers
1306 dst.x = min(src0.x, src1.x)
1308 dst.y = min(src0.y, src1.y)
1310 dst.z = min(src0.z, src1.z)
1312 dst.w = min(src0.w, src1.w)
1315 .. opcode:: SHL - Shift Left
1317 The shift count is masked with 0x1f before the shift is applied.
1321 dst.x = src0.x << (0x1f \& src1.x)
1323 dst.y = src0.y << (0x1f \& src1.y)
1325 dst.z = src0.z << (0x1f \& src1.z)
1327 dst.w = src0.w << (0x1f \& src1.w)
1330 .. opcode:: ISHR - Arithmetic Shift Right (of Signed Integer)
1332 The shift count is masked with 0x1f before the shift is applied.
1336 dst.x = src0.x >> (0x1f \& src1.x)
1338 dst.y = src0.y >> (0x1f \& src1.y)
1340 dst.z = src0.z >> (0x1f \& src1.z)
1342 dst.w = src0.w >> (0x1f \& src1.w)
1345 .. opcode:: USHR - Logical Shift Right
1347 The shift count is masked with 0x1f before the shift is applied.
1351 dst.x = src0.x >> (unsigned) (0x1f \& src1.x)
1353 dst.y = src0.y >> (unsigned) (0x1f \& src1.y)
1355 dst.z = src0.z >> (unsigned) (0x1f \& src1.z)
1357 dst.w = src0.w >> (unsigned) (0x1f \& src1.w)
1360 .. opcode:: UCMP - Integer Conditional Move
1364 dst.x = src0.x ? src1.x : src2.x
1366 dst.y = src0.y ? src1.y : src2.y
1368 dst.z = src0.z ? src1.z : src2.z
1370 dst.w = src0.w ? src1.w : src2.w
1374 .. opcode:: ISSG - Integer Set Sign
1378 dst.x = (src0.x < 0) ? -1 : (src0.x > 0) ? 1 : 0
1380 dst.y = (src0.y < 0) ? -1 : (src0.y > 0) ? 1 : 0
1382 dst.z = (src0.z < 0) ? -1 : (src0.z > 0) ? 1 : 0
1384 dst.w = (src0.w < 0) ? -1 : (src0.w > 0) ? 1 : 0
1388 .. opcode:: FSLT - Float Set On Less Than (ordered)
1390 Same comparison as SLT but returns integer instead of 1.0/0.0 float
1394 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1396 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1398 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1400 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1403 .. opcode:: ISLT - Signed Integer Set On Less Than
1407 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1409 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1411 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1413 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1416 .. opcode:: USLT - Unsigned Integer Set On Less Than
1420 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1422 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1424 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1426 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1429 .. opcode:: FSGE - Float Set On Greater Equal Than (ordered)
1431 Same comparison as SGE but returns integer instead of 1.0/0.0 float
1435 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1437 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1439 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1441 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1444 .. opcode:: ISGE - Signed Integer Set On Greater Equal Than
1448 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1450 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1452 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1454 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1457 .. opcode:: USGE - Unsigned Integer Set On Greater Equal Than
1461 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1463 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1465 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1467 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1470 .. opcode:: FSEQ - Float Set On Equal (ordered)
1472 Same comparison as SEQ but returns integer instead of 1.0/0.0 float
1476 dst.x = (src0.x == src1.x) ? \sim 0 : 0
1478 dst.y = (src0.y == src1.y) ? \sim 0 : 0
1480 dst.z = (src0.z == src1.z) ? \sim 0 : 0
1482 dst.w = (src0.w == src1.w) ? \sim 0 : 0
1485 .. opcode:: USEQ - Integer Set On Equal
1489 dst.x = (src0.x == src1.x) ? \sim 0 : 0
1491 dst.y = (src0.y == src1.y) ? \sim 0 : 0
1493 dst.z = (src0.z == src1.z) ? \sim 0 : 0
1495 dst.w = (src0.w == src1.w) ? \sim 0 : 0
1498 .. opcode:: FSNE - Float Set On Not Equal (unordered)
1500 Same comparison as SNE but returns integer instead of 1.0/0.0 float
1504 dst.x = (src0.x != src1.x) ? \sim 0 : 0
1506 dst.y = (src0.y != src1.y) ? \sim 0 : 0
1508 dst.z = (src0.z != src1.z) ? \sim 0 : 0
1510 dst.w = (src0.w != src1.w) ? \sim 0 : 0
1513 .. opcode:: USNE - Integer Set On Not Equal
1517 dst.x = (src0.x != src1.x) ? \sim 0 : 0
1519 dst.y = (src0.y != src1.y) ? \sim 0 : 0
1521 dst.z = (src0.z != src1.z) ? \sim 0 : 0
1523 dst.w = (src0.w != src1.w) ? \sim 0 : 0
1526 .. opcode:: INEG - Integer Negate
1541 .. opcode:: IABS - Integer Absolute Value
1555 These opcodes are used for bit-level manipulation of integers.
1557 .. opcode:: IBFE - Signed Bitfield Extract
1559 Like GLSL bitfieldExtract. Extracts a set of bits from the input, and
1560 sign-extends them if the high bit of the extracted window is set.
1564 def ibfe(value, offset, bits):
1565 if offset < 0 or bits < 0 or offset + bits > 32:
1567 if bits == 0: return 0
1568 # Note: >> sign-extends
1569 return (value << (32 - offset - bits)) >> (32 - bits)
1571 .. opcode:: UBFE - Unsigned Bitfield Extract
1573 Like GLSL bitfieldExtract. Extracts a set of bits from the input, without
1578 def ubfe(value, offset, bits):
1579 if offset < 0 or bits < 0 or offset + bits > 32:
1581 if bits == 0: return 0
1582 # Note: >> does not sign-extend
1583 return (value << (32 - offset - bits)) >> (32 - bits)
1585 .. opcode:: BFI - Bitfield Insert
1587 Like GLSL bitfieldInsert. Replaces a bit region of 'base' with the low bits
1592 def bfi(base, insert, offset, bits):
1593 if offset < 0 or bits < 0 or offset + bits > 32:
1595 # << defined such that mask == ~0 when bits == 32, offset == 0
1596 mask = ((1 << bits) - 1) << offset
1597 return ((insert << offset) & mask) | (base & ~mask)
1599 .. opcode:: BREV - Bitfield Reverse
1601 See SM5 instruction BFREV. Reverses the bits of the argument.
1603 .. opcode:: POPC - Population Count
1605 See SM5 instruction COUNTBITS. Counts the number of set bits in the argument.
1607 .. opcode:: LSB - Index of lowest set bit
1609 See SM5 instruction FIRSTBIT_LO. Computes the 0-based index of the first set
1610 bit of the argument. Returns -1 if none are set.
1612 .. opcode:: IMSB - Index of highest non-sign bit
1614 See SM5 instruction FIRSTBIT_SHI. Computes the 0-based index of the highest
1615 non-sign bit of the argument (i.e. highest 0 bit for negative numbers,
1616 highest 1 bit for positive numbers). Returns -1 if all bits are the same
1617 (i.e. for inputs 0 and -1).
1619 .. opcode:: UMSB - Index of highest set bit
1621 See SM5 instruction FIRSTBIT_HI. Computes the 0-based index of the highest
1622 set bit of the argument. Returns -1 if none are set.
1625 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1627 These opcodes are only supported in geometry shaders; they have no meaning
1628 in any other type of shader.
1630 .. opcode:: EMIT - Emit
1632 Generate a new vertex for the current primitive into the specified vertex
1633 stream using the values in the output registers.
1636 .. opcode:: ENDPRIM - End Primitive
1638 Complete the current primitive in the specified vertex stream (consisting of
1639 the emitted vertices), and start a new one.
1645 These opcodes are part of :term:`GLSL`'s opcode set. Support for these
1646 opcodes is determined by a special capability bit, ``GLSL``.
1647 Some require glsl version 1.30 (UIF/SWITCH/CASE/DEFAULT/ENDSWITCH).
1649 .. opcode:: CAL - Subroutine Call
1655 .. opcode:: RET - Subroutine Call Return
1660 .. opcode:: CONT - Continue
1662 Unconditionally moves the point of execution to the instruction after the
1663 last bgnloop. The instruction must appear within a bgnloop/endloop.
1667 Support for CONT is determined by a special capability bit,
1668 ``TGSI_CONT_SUPPORTED``. See :ref:`Screen` for more information.
1671 .. opcode:: BGNLOOP - Begin a Loop
1673 Start a loop. Must have a matching endloop.
1676 .. opcode:: BGNSUB - Begin Subroutine
1678 Starts definition of a subroutine. Must have a matching endsub.
1681 .. opcode:: ENDLOOP - End a Loop
1683 End a loop started with bgnloop.
1686 .. opcode:: ENDSUB - End Subroutine
1688 Ends definition of a subroutine.
1691 .. opcode:: NOP - No Operation
1696 .. opcode:: BRK - Break
1698 Unconditionally moves the point of execution to the instruction after the
1699 next endloop or endswitch. The instruction must appear within a loop/endloop
1700 or switch/endswitch.
1703 .. opcode:: IF - Float If
1705 Start an IF ... ELSE .. ENDIF block. Condition evaluates to true if
1709 where src0.x is interpreted as a floating point register.
1712 .. opcode:: UIF - Bitwise If
1714 Start an UIF ... ELSE .. ENDIF block. Condition evaluates to true if
1718 where src0.x is interpreted as an integer register.
1721 .. opcode:: ELSE - Else
1723 Starts an else block, after an IF or UIF statement.
1726 .. opcode:: ENDIF - End If
1728 Ends an IF or UIF block.
1731 .. opcode:: SWITCH - Switch
1733 Starts a C-style switch expression. The switch consists of one or multiple
1734 CASE statements, and at most one DEFAULT statement. Execution of a statement
1735 ends when a BRK is hit, but just like in C falling through to other cases
1736 without a break is allowed. Similarly, DEFAULT label is allowed anywhere not
1737 just as last statement, and fallthrough is allowed into/from it.
1738 CASE src arguments are evaluated at bit level against the SWITCH src argument.
1744 (some instructions here)
1747 (some instructions here)
1750 (some instructions here)
1755 .. opcode:: CASE - Switch case
1757 This represents a switch case label. The src arg must be an integer immediate.
1760 .. opcode:: DEFAULT - Switch default
1762 This represents the default case in the switch, which is taken if no other
1766 .. opcode:: ENDSWITCH - End of switch
1768 Ends a switch expression.
1774 The interpolation instructions allow an input to be interpolated in a
1775 different way than its declaration. This corresponds to the GLSL 4.00
1776 interpolateAt* functions. The first argument of each of these must come from
1777 ``TGSI_FILE_INPUT``.
1779 .. opcode:: INTERP_CENTROID - Interpolate at the centroid
1781 Interpolates the varying specified by src0 at the centroid
1783 .. opcode:: INTERP_SAMPLE - Interpolate at the specified sample
1785 Interpolates the varying specified by src0 at the sample id specified by
1786 src1.x (interpreted as an integer)
1788 .. opcode:: INTERP_OFFSET - Interpolate at the specified offset
1790 Interpolates the varying specified by src0 at the offset src1.xy from the
1791 pixel center (interpreted as floats)
1799 The double-precision opcodes reinterpret four-component vectors into
1800 two-component vectors with doubled precision in each component.
1802 .. opcode:: DABS - Absolute
1810 .. opcode:: DADD - Add
1814 dst.xy = src0.xy + src1.xy
1816 dst.zw = src0.zw + src1.zw
1818 .. opcode:: DSEQ - Set on Equal
1822 dst.x = src0.xy == src1.xy ? \sim 0 : 0
1824 dst.z = src0.zw == src1.zw ? \sim 0 : 0
1826 .. opcode:: DSNE - Set on Not Equal
1830 dst.x = src0.xy != src1.xy ? \sim 0 : 0
1832 dst.z = src0.zw != src1.zw ? \sim 0 : 0
1834 .. opcode:: DSLT - Set on Less than
1838 dst.x = src0.xy < src1.xy ? \sim 0 : 0
1840 dst.z = src0.zw < src1.zw ? \sim 0 : 0
1842 .. opcode:: DSGE - Set on Greater equal
1846 dst.x = src0.xy >= src1.xy ? \sim 0 : 0
1848 dst.z = src0.zw >= src1.zw ? \sim 0 : 0
1850 .. opcode:: DFRAC - Fraction
1854 dst.xy = src.xy - \lfloor src.xy\rfloor
1856 dst.zw = src.zw - \lfloor src.zw\rfloor
1858 .. opcode:: DTRUNC - Truncate
1862 dst.xy = trunc(src.xy)
1864 dst.zw = trunc(src.zw)
1866 .. opcode:: DCEIL - Ceiling
1870 dst.xy = \lceil src.xy\rceil
1872 dst.zw = \lceil src.zw\rceil
1874 .. opcode:: DFLR - Floor
1878 dst.xy = \lfloor src.xy\rfloor
1880 dst.zw = \lfloor src.zw\rfloor
1882 .. opcode:: DROUND - Fraction
1886 dst.xy = round(src.xy)
1888 dst.zw = round(src.zw)
1890 .. opcode:: DSSG - Set Sign
1894 dst.xy = (src.xy > 0) ? 1.0 : (src.xy < 0) ? -1.0 : 0.0
1896 dst.zw = (src.zw > 0) ? 1.0 : (src.zw < 0) ? -1.0 : 0.0
1898 .. opcode:: DFRACEXP - Convert Number to Fractional and Integral Components
1900 Like the ``frexp()`` routine in many math libraries, this opcode stores the
1901 exponent of its source to ``dst0``, and the significand to ``dst1``, such that
1902 :math:`dst1 \times 2^{dst0} = src` . The results are replicated across
1907 dst0.xy = dst.zw = frac(src.xy)
1912 .. opcode:: DLDEXP - Multiply Number by Integral Power of 2
1914 This opcode is the inverse of :opcode:`DFRACEXP`. The second
1915 source is an integer.
1919 dst.xy = src0.xy \times 2^{src1.x}
1921 dst.zw = src0.zw \times 2^{src1.z}
1923 .. opcode:: DMIN - Minimum
1927 dst.xy = min(src0.xy, src1.xy)
1929 dst.zw = min(src0.zw, src1.zw)
1931 .. opcode:: DMAX - Maximum
1935 dst.xy = max(src0.xy, src1.xy)
1937 dst.zw = max(src0.zw, src1.zw)
1939 .. opcode:: DMUL - Multiply
1943 dst.xy = src0.xy \times src1.xy
1945 dst.zw = src0.zw \times src1.zw
1948 .. opcode:: DMAD - Multiply And Add
1952 dst.xy = src0.xy \times src1.xy + src2.xy
1954 dst.zw = src0.zw \times src1.zw + src2.zw
1957 .. opcode:: DFMA - Fused Multiply-Add
1959 Perform a * b + c with no intermediate rounding step.
1963 dst.xy = src0.xy \times src1.xy + src2.xy
1965 dst.zw = src0.zw \times src1.zw + src2.zw
1968 .. opcode:: DDIV - Divide
1972 dst.xy = \frac{src0.xy}{src1.xy}
1974 dst.zw = \frac{src0.zw}{src1.zw}
1977 .. opcode:: DRCP - Reciprocal
1981 dst.xy = \frac{1}{src.xy}
1983 dst.zw = \frac{1}{src.zw}
1985 .. opcode:: DSQRT - Square Root
1989 dst.xy = \sqrt{src.xy}
1991 dst.zw = \sqrt{src.zw}
1993 .. opcode:: DRSQ - Reciprocal Square Root
1997 dst.xy = \frac{1}{\sqrt{src.xy}}
1999 dst.zw = \frac{1}{\sqrt{src.zw}}
2001 .. opcode:: F2D - Float to Double
2005 dst.xy = double(src0.x)
2007 dst.zw = double(src0.y)
2009 .. opcode:: D2F - Double to Float
2013 dst.x = float(src0.xy)
2015 dst.y = float(src0.zw)
2017 .. opcode:: I2D - Int to Double
2021 dst.xy = double(src0.x)
2023 dst.zw = double(src0.y)
2025 .. opcode:: D2I - Double to Int
2029 dst.x = int(src0.xy)
2031 dst.y = int(src0.zw)
2033 .. opcode:: U2D - Unsigned Int to Double
2037 dst.xy = double(src0.x)
2039 dst.zw = double(src0.y)
2041 .. opcode:: D2U - Double to Unsigned Int
2045 dst.x = unsigned(src0.xy)
2047 dst.y = unsigned(src0.zw)
2052 The 64-bit integer opcodes reinterpret four-component vectors into
2053 two-component vectors with 64-bits in each component.
2055 .. opcode:: I64ABS - 64-bit Integer Absolute Value
2063 .. opcode:: I64NEG - 64-bit Integer Negate
2073 .. opcode:: I64SSG - 64-bit Integer Set Sign
2077 dst.xy = (src0.xy < 0) ? -1 : (src0.xy > 0) ? 1 : 0
2079 dst.zw = (src0.zw < 0) ? -1 : (src0.zw > 0) ? 1 : 0
2081 .. opcode:: U64ADD - 64-bit Integer Add
2085 dst.xy = src0.xy + src1.xy
2087 dst.zw = src0.zw + src1.zw
2089 .. opcode:: U64MUL - 64-bit Integer Multiply
2093 dst.xy = src0.xy * src1.xy
2095 dst.zw = src0.zw * src1.zw
2097 .. opcode:: U64SEQ - 64-bit Integer Set on Equal
2101 dst.x = src0.xy == src1.xy ? \sim 0 : 0
2103 dst.z = src0.zw == src1.zw ? \sim 0 : 0
2105 .. opcode:: U64SNE - 64-bit Integer Set on Not Equal
2109 dst.x = src0.xy != src1.xy ? \sim 0 : 0
2111 dst.z = src0.zw != src1.zw ? \sim 0 : 0
2113 .. opcode:: U64SLT - 64-bit Unsigned Integer Set on Less Than
2117 dst.x = src0.xy < src1.xy ? \sim 0 : 0
2119 dst.z = src0.zw < src1.zw ? \sim 0 : 0
2121 .. opcode:: U64SGE - 64-bit Unsigned Integer Set on Greater Equal
2125 dst.x = src0.xy >= src1.xy ? \sim 0 : 0
2127 dst.z = src0.zw >= src1.zw ? \sim 0 : 0
2129 .. opcode:: I64SLT - 64-bit Signed Integer Set on Less Than
2133 dst.x = src0.xy < src1.xy ? \sim 0 : 0
2135 dst.z = src0.zw < src1.zw ? \sim 0 : 0
2137 .. opcode:: I64SGE - 64-bit Signed Integer Set on Greater Equal
2141 dst.x = src0.xy >= src1.xy ? \sim 0 : 0
2143 dst.z = src0.zw >= src1.zw ? \sim 0 : 0
2145 .. opcode:: I64MIN - Minimum of 64-bit Signed Integers
2149 dst.xy = min(src0.xy, src1.xy)
2151 dst.zw = min(src0.zw, src1.zw)
2153 .. opcode:: U64MIN - Minimum of 64-bit Unsigned Integers
2157 dst.xy = min(src0.xy, src1.xy)
2159 dst.zw = min(src0.zw, src1.zw)
2161 .. opcode:: I64MAX - Maximum of 64-bit Signed Integers
2165 dst.xy = max(src0.xy, src1.xy)
2167 dst.zw = max(src0.zw, src1.zw)
2169 .. opcode:: U64MAX - Maximum of 64-bit Unsigned Integers
2173 dst.xy = max(src0.xy, src1.xy)
2175 dst.zw = max(src0.zw, src1.zw)
2177 .. opcode:: U64SHL - Shift Left 64-bit Unsigned Integer
2179 The shift count is masked with 0x3f before the shift is applied.
2183 dst.xy = src0.xy << (0x3f \& src1.x)
2185 dst.zw = src0.zw << (0x3f \& src1.y)
2187 .. opcode:: I64SHR - Arithmetic Shift Right (of 64-bit Signed Integer)
2189 The shift count is masked with 0x3f before the shift is applied.
2193 dst.xy = src0.xy >> (0x3f \& src1.x)
2195 dst.zw = src0.zw >> (0x3f \& src1.y)
2197 .. opcode:: U64SHR - Logical Shift Right (of 64-bit Unsigned Integer)
2199 The shift count is masked with 0x3f before the shift is applied.
2203 dst.xy = src0.xy >> (unsigned) (0x3f \& src1.x)
2205 dst.zw = src0.zw >> (unsigned) (0x3f \& src1.y)
2207 .. opcode:: I64DIV - 64-bit Signed Integer Division
2211 dst.xy = \frac{src0.xy}{src1.xy}
2213 dst.zw = \frac{src0.zw}{src1.zw}
2215 .. opcode:: U64DIV - 64-bit Unsigned Integer Division
2219 dst.xy = \frac{src0.xy}{src1.xy}
2221 dst.zw = \frac{src0.zw}{src1.zw}
2223 .. opcode:: U64MOD - 64-bit Unsigned Integer Remainder
2227 dst.xy = src0.xy \bmod src1.xy
2229 dst.zw = src0.zw \bmod src1.zw
2231 .. opcode:: I64MOD - 64-bit Signed Integer Remainder
2235 dst.xy = src0.xy \bmod src1.xy
2237 dst.zw = src0.zw \bmod src1.zw
2239 .. opcode:: F2U64 - Float to 64-bit Unsigned Int
2243 dst.xy = (uint64_t) src0.x
2245 dst.zw = (uint64_t) src0.y
2247 .. opcode:: F2I64 - Float to 64-bit Int
2251 dst.xy = (int64_t) src0.x
2253 dst.zw = (int64_t) src0.y
2255 .. opcode:: U2I64 - Unsigned Integer to 64-bit Integer
2257 This is a zero extension.
2261 dst.xy = (int64_t) src0.x
2263 dst.zw = (int64_t) src0.y
2265 .. opcode:: I2I64 - Signed Integer to 64-bit Integer
2267 This is a sign extension.
2271 dst.xy = (int64_t) src0.x
2273 dst.zw = (int64_t) src0.y
2275 .. opcode:: D2U64 - Double to 64-bit Unsigned Int
2279 dst.xy = (uint64_t) src0.xy
2281 dst.zw = (uint64_t) src0.zw
2283 .. opcode:: D2I64 - Double to 64-bit Int
2287 dst.xy = (int64_t) src0.xy
2289 dst.zw = (int64_t) src0.zw
2291 .. opcode:: U642F - 64-bit unsigned integer to float
2295 dst.x = (float) src0.xy
2297 dst.y = (float) src0.zw
2299 .. opcode:: I642F - 64-bit Int to Float
2303 dst.x = (float) src0.xy
2305 dst.y = (float) src0.zw
2307 .. opcode:: U642D - 64-bit unsigned integer to double
2311 dst.xy = (double) src0.xy
2313 dst.zw = (double) src0.zw
2315 .. opcode:: I642D - 64-bit Int to double
2319 dst.xy = (double) src0.xy
2321 dst.zw = (double) src0.zw
2323 .. _samplingopcodes:
2325 Resource Sampling Opcodes
2326 ^^^^^^^^^^^^^^^^^^^^^^^^^
2328 Those opcodes follow very closely semantics of the respective Direct3D
2329 instructions. If in doubt double check Direct3D documentation.
2330 Note that the swizzle on SVIEW (src1) determines texel swizzling
2335 Using provided address, sample data from the specified texture using the
2336 filtering mode identified by the given sampler. The source data may come from
2337 any resource type other than buffers.
2339 Syntax: ``SAMPLE dst, address, sampler_view, sampler``
2341 Example: ``SAMPLE TEMP[0], TEMP[1], SVIEW[0], SAMP[0]``
2343 .. opcode:: SAMPLE_I
2345 Simplified alternative to the SAMPLE instruction. Using the provided
2346 integer address, SAMPLE_I fetches data from the specified sampler view
2347 without any filtering. The source data may come from any resource type
2350 Syntax: ``SAMPLE_I dst, address, sampler_view``
2352 Example: ``SAMPLE_I TEMP[0], TEMP[1], SVIEW[0]``
2354 The 'address' is specified as unsigned integers. If the 'address' is out of
2355 range [0...(# texels - 1)] the result of the fetch is always 0 in all
2356 components. As such the instruction doesn't honor address wrap modes, in
2357 cases where that behavior is desirable 'SAMPLE' instruction should be used.
2358 address.w always provides an unsigned integer mipmap level. If the value is
2359 out of the range then the instruction always returns 0 in all components.
2360 address.yz are ignored for buffers and 1d textures. address.z is ignored
2361 for 1d texture arrays and 2d textures.
2363 For 1D texture arrays address.y provides the array index (also as unsigned
2364 integer). If the value is out of the range of available array indices
2365 [0... (array size - 1)] then the opcode always returns 0 in all components.
2366 For 2D texture arrays address.z provides the array index, otherwise it
2367 exhibits the same behavior as in the case for 1D texture arrays. The exact
2368 semantics of the source address are presented in the table below:
2370 +---------------------------+----+-----+-----+---------+
2371 | resource type | X | Y | Z | W |
2372 +===========================+====+=====+=====+=========+
2373 | ``PIPE_BUFFER`` | x | | | ignored |
2374 +---------------------------+----+-----+-----+---------+
2375 | ``PIPE_TEXTURE_1D`` | x | | | mpl |
2376 +---------------------------+----+-----+-----+---------+
2377 | ``PIPE_TEXTURE_2D`` | x | y | | mpl |
2378 +---------------------------+----+-----+-----+---------+
2379 | ``PIPE_TEXTURE_3D`` | x | y | z | mpl |
2380 +---------------------------+----+-----+-----+---------+
2381 | ``PIPE_TEXTURE_RECT`` | x | y | | mpl |
2382 +---------------------------+----+-----+-----+---------+
2383 | ``PIPE_TEXTURE_CUBE`` | not allowed as source |
2384 +---------------------------+----+-----+-----+---------+
2385 | ``PIPE_TEXTURE_1D_ARRAY`` | x | idx | | mpl |
2386 +---------------------------+----+-----+-----+---------+
2387 | ``PIPE_TEXTURE_2D_ARRAY`` | x | y | idx | mpl |
2388 +---------------------------+----+-----+-----+---------+
2390 Where 'mpl' is a mipmap level and 'idx' is the array index.
2392 .. opcode:: SAMPLE_I_MS
2394 Just like SAMPLE_I but allows fetch data from multi-sampled surfaces.
2396 Syntax: ``SAMPLE_I_MS dst, address, sampler_view, sample``
2398 .. opcode:: SAMPLE_B
2400 Just like the SAMPLE instruction with the exception that an additional bias
2401 is applied to the level of detail computed as part of the instruction
2404 Syntax: ``SAMPLE_B dst, address, sampler_view, sampler, lod_bias``
2406 Example: ``SAMPLE_B TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2].x``
2408 .. opcode:: SAMPLE_C
2410 Similar to the SAMPLE instruction but it performs a comparison filter. The
2411 operands to SAMPLE_C are identical to SAMPLE, except that there is an
2412 additional float32 operand, reference value, which must be a register with
2413 single-component, or a scalar literal. SAMPLE_C makes the hardware use the
2414 current samplers compare_func (in pipe_sampler_state) to compare reference
2415 value against the red component value for the surce resource at each texel
2416 that the currently configured texture filter covers based on the provided
2419 Syntax: ``SAMPLE_C dst, address, sampler_view.r, sampler, ref_value``
2421 Example: ``SAMPLE_C TEMP[0], TEMP[1], SVIEW[0].r, SAMP[0], TEMP[2].x``
2423 .. opcode:: SAMPLE_C_LZ
2425 Same as SAMPLE_C, but LOD is 0 and derivatives are ignored. The LZ stands
2428 Syntax: ``SAMPLE_C_LZ dst, address, sampler_view.r, sampler, ref_value``
2430 Example: ``SAMPLE_C_LZ TEMP[0], TEMP[1], SVIEW[0].r, SAMP[0], TEMP[2].x``
2433 .. opcode:: SAMPLE_D
2435 SAMPLE_D is identical to the SAMPLE opcode except that the derivatives for
2436 the source address in the x direction and the y direction are provided by
2439 Syntax: ``SAMPLE_D dst, address, sampler_view, sampler, der_x, der_y``
2441 Example: ``SAMPLE_D TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2], TEMP[3]``
2443 .. opcode:: SAMPLE_L
2445 SAMPLE_L is identical to the SAMPLE opcode except that the LOD is provided
2446 directly as a scalar value, representing no anisotropy.
2448 Syntax: ``SAMPLE_L dst, address, sampler_view, sampler, explicit_lod``
2450 Example: ``SAMPLE_L TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2].x``
2454 Gathers the four texels to be used in a bi-linear filtering operation and
2455 packs them into a single register. Only works with 2D, 2D array, cubemaps,
2456 and cubemaps arrays. For 2D textures, only the addressing modes of the
2457 sampler and the top level of any mip pyramid are used. Set W to zero. It
2458 behaves like the SAMPLE instruction, but a filtered sample is not
2459 generated. The four samples that contribute to filtering are placed into
2460 xyzw in counter-clockwise order, starting with the (u,v) texture coordinate
2461 delta at the following locations (-, +), (+, +), (+, -), (-, -), where the
2462 magnitude of the deltas are half a texel.
2465 .. opcode:: SVIEWINFO
2467 Query the dimensions of a given sampler view. dst receives width, height,
2468 depth or array size and number of mipmap levels as int4. The dst can have a
2469 writemask which will specify what info is the caller interested in.
2471 Syntax: ``SVIEWINFO dst, src_mip_level, sampler_view``
2473 Example: ``SVIEWINFO TEMP[0], TEMP[1].x, SVIEW[0]``
2475 src_mip_level is an unsigned integer scalar. If it's out of range then
2476 returns 0 for width, height and depth/array size but the total number of
2477 mipmap is still returned correctly for the given sampler view. The returned
2478 width, height and depth values are for the mipmap level selected by the
2479 src_mip_level and are in the number of texels. For 1d texture array width
2480 is in dst.x, array size is in dst.y and dst.z is 0. The number of mipmaps is
2481 still in dst.w. In contrast to d3d10 resinfo, there's no way in the tgsi
2482 instruction encoding to specify the return type (float/rcpfloat/uint), hence
2483 always using uint. Also, unlike the SAMPLE instructions, the swizzle on src1
2484 resinfo allowing swizzling dst values is ignored (due to the interaction
2485 with rcpfloat modifier which requires some swizzle handling in the state
2488 .. opcode:: SAMPLE_POS
2490 Query the position of a sample in the given resource or render target
2491 when per-sample fragment shading is in effect.
2493 Syntax: ``SAMPLE_POS dst, source, sample_index``
2495 dst receives float4 (x, y, undef, undef) indicated where the sample is
2496 located. Sample locations are in the range [0, 1] where 0.5 is the center
2499 source is either a sampler view (to indicate a shader resource) or temp
2500 register (to indicate the render target). The source register may have
2501 an optional swizzle to apply to the returned result
2503 sample_index is an integer scalar indicating which sample position is to
2506 If per-sample shading is not in effect or the source resource or render
2507 target is not multisampled, the result is (0.5, 0.5, undef, undef).
2509 NOTE: no driver has implemented this opcode yet (and no gallium frontend
2510 emits it). This information is subject to change.
2512 .. opcode:: SAMPLE_INFO
2514 Query the number of samples in a multisampled resource or render target.
2516 Syntax: ``SAMPLE_INFO dst, source``
2518 dst receives int4 (n, 0, 0, 0) where n is the number of samples in a
2519 resource or the render target.
2521 source is either a sampler view (to indicate a shader resource) or temp
2522 register (to indicate the render target). The source register may have
2523 an optional swizzle to apply to the returned result
2525 If per-sample shading is not in effect or the source resource or render
2526 target is not multisampled, the result is (1, 0, 0, 0).
2528 NOTE: no driver has implemented this opcode yet (and no gallium frontend
2529 emits it). This information is subject to change.
2531 .. opcode:: LOD - level of detail
2533 Same syntax as the SAMPLE opcode but instead of performing an actual
2534 texture lookup/filter, return the computed LOD information that the
2535 texture pipe would use to access the texture. The Y component contains
2536 the computed LOD lambda_prime. The X component contains the LOD that will
2537 be accessed, based on min/max lod's and mipmap filters.
2538 The Z and W components are set to 0.
2540 Syntax: ``LOD dst, address, sampler_view, sampler``
2543 .. _resourceopcodes:
2545 Resource Access Opcodes
2546 ^^^^^^^^^^^^^^^^^^^^^^^
2548 For these opcodes, the resource can be a BUFFER, IMAGE, or MEMORY.
2550 .. opcode:: LOAD - Fetch data from a shader buffer or image
2552 Syntax: ``LOAD dst, resource, address``
2554 Example: ``LOAD TEMP[0], BUFFER[0], TEMP[1]``
2556 Using the provided integer address, LOAD fetches data
2557 from the specified buffer or texture without any
2560 The 'address' is specified as a vector of unsigned
2561 integers. If the 'address' is out of range the result
2564 Only the first mipmap level of a resource can be read
2565 from using this instruction.
2567 For 1D or 2D texture arrays, the array index is
2568 provided as an unsigned integer in address.y or
2569 address.z, respectively. address.yz are ignored for
2570 buffers and 1D textures. address.z is ignored for 1D
2571 texture arrays and 2D textures. address.w is always
2574 A swizzle suffix may be added to the resource argument
2575 this will cause the resource data to be swizzled accordingly.
2577 .. opcode:: STORE - Write data to a shader resource
2579 Syntax: ``STORE resource, address, src``
2581 Example: ``STORE BUFFER[0], TEMP[0], TEMP[1]``
2583 Using the provided integer address, STORE writes data
2584 to the specified buffer or texture.
2586 The 'address' is specified as a vector of unsigned
2587 integers. If the 'address' is out of range the result
2590 Only the first mipmap level of a resource can be
2591 written to using this instruction.
2593 For 1D or 2D texture arrays, the array index is
2594 provided as an unsigned integer in address.y or
2595 address.z, respectively. address.yz are ignored for
2596 buffers and 1D textures. address.z is ignored for 1D
2597 texture arrays and 2D textures. address.w is always
2600 .. opcode:: RESQ - Query information about a resource
2602 Syntax: ``RESQ dst, resource``
2604 Example: ``RESQ TEMP[0], BUFFER[0]``
2606 Returns information about the buffer or image resource. For buffer
2607 resources, the size (in bytes) is returned in the x component. For
2608 image resources, .xyz will contain the width/height/layers of the
2609 image, while .w will contain the number of samples for multi-sampled
2612 .. opcode:: FBFETCH - Load data from framebuffer
2614 Syntax: ``FBFETCH dst, output``
2616 Example: ``FBFETCH TEMP[0], OUT[0]``
2618 This is only valid on ``COLOR`` semantic outputs. Returns the color
2619 of the current position in the framebuffer from before this fragment
2620 shader invocation. May return the same value from multiple calls for
2621 a particular output within a single invocation. Note that result may
2622 be undefined if a fragment is drawn multiple times without a blend
2626 .. _bindlessopcodes:
2631 These opcodes are for working with bindless sampler or image handles and
2632 require PIPE_CAP_BINDLESS_TEXTURE.
2634 .. opcode:: IMG2HND - Get a bindless handle for a image
2636 Syntax: ``IMG2HND dst, image``
2638 Example: ``IMG2HND TEMP[0], IMAGE[0]``
2640 Sets 'dst' to a bindless handle for 'image'.
2642 .. opcode:: SAMP2HND - Get a bindless handle for a sampler
2644 Syntax: ``SAMP2HND dst, sampler``
2646 Example: ``SAMP2HND TEMP[0], SAMP[0]``
2648 Sets 'dst' to a bindless handle for 'sampler'.
2651 .. _threadsyncopcodes:
2653 Inter-thread synchronization opcodes
2654 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2656 These opcodes are intended for communication between threads running
2657 within the same compute grid. For now they're only valid in compute
2660 .. opcode:: BARRIER - Thread group barrier
2664 This opcode suspends the execution of the current thread until all
2665 the remaining threads in the working group reach the same point of
2666 the program. Results are unspecified if any of the remaining
2667 threads terminates or never reaches an executed BARRIER instruction.
2669 .. opcode:: MEMBAR - Memory barrier
2673 This opcode waits for the completion of all memory accesses based on
2674 the type passed in. The type is an immediate bitfield with the following
2677 Bit 0: Shader storage buffers
2678 Bit 1: Atomic buffers
2680 Bit 3: Shared memory
2683 These may be passed in in any combination. An implementation is free to not
2684 distinguish between these as it sees fit. However these map to all the
2685 possibilities made available by GLSL.
2692 These opcodes provide atomic variants of some common arithmetic and
2693 logical operations. In this context atomicity means that another
2694 concurrent memory access operation that affects the same memory
2695 location is guaranteed to be performed strictly before or after the
2696 entire execution of the atomic operation. The resource may be a BUFFER,
2697 IMAGE, HWATOMIC, or MEMORY. In the case of an image, the offset works
2698 the same as for ``LOAD`` and ``STORE``, specified above. For atomic
2699 counters, the offset is an immediate index to the base hw atomic
2700 counter for this operation.
2701 These atomic operations may only be used with 32-bit integer image formats.
2703 .. opcode:: ATOMUADD - Atomic integer addition
2705 Syntax: ``ATOMUADD dst, resource, offset, src``
2707 Example: ``ATOMUADD TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2709 The following operation is performed atomically:
2713 dst_x = resource[offset]
2715 resource[offset] = dst_x + src_x
2718 .. opcode:: ATOMFADD - Atomic floating point addition
2720 Syntax: ``ATOMFADD dst, resource, offset, src``
2722 Example: ``ATOMFADD TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2724 The following operation is performed atomically:
2728 dst_x = resource[offset]
2730 resource[offset] = dst_x + src_x
2733 .. opcode:: ATOMXCHG - Atomic exchange
2735 Syntax: ``ATOMXCHG dst, resource, offset, src``
2737 Example: ``ATOMXCHG TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2739 The following operation is performed atomically:
2743 dst_x = resource[offset]
2745 resource[offset] = src_x
2748 .. opcode:: ATOMCAS - Atomic compare-and-exchange
2750 Syntax: ``ATOMCAS dst, resource, offset, cmp, src``
2752 Example: ``ATOMCAS TEMP[0], BUFFER[0], TEMP[1], TEMP[2], TEMP[3]``
2754 The following operation is performed atomically:
2758 dst_x = resource[offset]
2760 resource[offset] = (dst_x == cmp_x ? src_x : dst_x)
2763 .. opcode:: ATOMAND - Atomic bitwise And
2765 Syntax: ``ATOMAND dst, resource, offset, src``
2767 Example: ``ATOMAND TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2769 The following operation is performed atomically:
2773 dst_x = resource[offset]
2775 resource[offset] = dst_x \& src_x
2778 .. opcode:: ATOMOR - Atomic bitwise Or
2780 Syntax: ``ATOMOR dst, resource, offset, src``
2782 Example: ``ATOMOR TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2784 The following operation is performed atomically:
2788 dst_x = resource[offset]
2790 resource[offset] = dst_x | src_x
2793 .. opcode:: ATOMXOR - Atomic bitwise Xor
2795 Syntax: ``ATOMXOR dst, resource, offset, src``
2797 Example: ``ATOMXOR TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2799 The following operation is performed atomically:
2803 dst_x = resource[offset]
2805 resource[offset] = dst_x \oplus src_x
2808 .. opcode:: ATOMUMIN - Atomic unsigned minimum
2810 Syntax: ``ATOMUMIN dst, resource, offset, src``
2812 Example: ``ATOMUMIN TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2814 The following operation is performed atomically:
2818 dst_x = resource[offset]
2820 resource[offset] = (dst_x < src_x ? dst_x : src_x)
2823 .. opcode:: ATOMUMAX - Atomic unsigned maximum
2825 Syntax: ``ATOMUMAX dst, resource, offset, src``
2827 Example: ``ATOMUMAX TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2829 The following operation is performed atomically:
2833 dst_x = resource[offset]
2835 resource[offset] = (dst_x > src_x ? dst_x : src_x)
2838 .. opcode:: ATOMIMIN - Atomic signed minimum
2840 Syntax: ``ATOMIMIN dst, resource, offset, src``
2842 Example: ``ATOMIMIN TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2844 The following operation is performed atomically:
2848 dst_x = resource[offset]
2850 resource[offset] = (dst_x < src_x ? dst_x : src_x)
2853 .. opcode:: ATOMIMAX - Atomic signed maximum
2855 Syntax: ``ATOMIMAX dst, resource, offset, src``
2857 Example: ``ATOMIMAX TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2859 The following operation is performed atomically:
2863 dst_x = resource[offset]
2865 resource[offset] = (dst_x > src_x ? dst_x : src_x)
2868 .. opcode:: ATOMINC_WRAP - Atomic increment + wrap around
2870 Syntax: ``ATOMINC_WRAP dst, resource, offset, src``
2872 Example: ``ATOMINC_WRAP TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2874 The following operation is performed atomically:
2878 dst_x = resource[offset] + 1
2880 resource[offset] = dst_x <= src_x ? dst_x : 0
2883 .. opcode:: ATOMDEC_WRAP - Atomic decrement + wrap around
2885 Syntax: ``ATOMDEC_WRAP dst, resource, offset, src``
2887 Example: ``ATOMDEC_WRAP TEMP[0], BUFFER[0], TEMP[1], TEMP[2]``
2889 The following operation is performed atomically:
2893 dst_x = resource[offset]
2895 resource[offset] = (dst_x > 0 && dst_x < src_x) ? dst_x - 1 : 0
2898 .. _interlaneopcodes:
2903 These opcodes reduce the given value across the shader invocations
2904 running in the current SIMD group. Every thread in the subgroup will receive
2905 the same result. The BALLOT operations accept a single-channel argument that
2906 is treated as a boolean and produce a 64-bit value.
2908 .. opcode:: VOTE_ANY - Value is set in any of the active invocations
2910 Syntax: ``VOTE_ANY dst, value``
2912 Example: ``VOTE_ANY TEMP[0].x, TEMP[1].x``
2915 .. opcode:: VOTE_ALL - Value is set in all of the active invocations
2917 Syntax: ``VOTE_ALL dst, value``
2919 Example: ``VOTE_ALL TEMP[0].x, TEMP[1].x``
2922 .. opcode:: VOTE_EQ - Value is the same in all of the active invocations
2924 Syntax: ``VOTE_EQ dst, value``
2926 Example: ``VOTE_EQ TEMP[0].x, TEMP[1].x``
2929 .. opcode:: BALLOT - Lanemask of whether the value is set in each active
2932 Syntax: ``BALLOT dst, value``
2934 Example: ``BALLOT TEMP[0].xy, TEMP[1].x``
2936 When the argument is a constant true, this produces a bitmask of active
2937 invocations. In fragment shaders, this can include helper invocations
2938 (invocations whose outputs and writes to memory are discarded, but which
2939 are used to compute derivatives).
2942 .. opcode:: READ_FIRST - Broadcast the value from the first active
2943 invocation to all active lanes
2945 Syntax: ``READ_FIRST dst, value``
2947 Example: ``READ_FIRST TEMP[0], TEMP[1]``
2950 .. opcode:: READ_INVOC - Retrieve the value from the given invocation
2951 (need not be uniform)
2953 Syntax: ``READ_INVOC dst, value, invocation``
2955 Example: ``READ_INVOC TEMP[0].xy, TEMP[1].xy, TEMP[2].x``
2957 invocation.x controls the invocation number to read from for all channels.
2958 The invocation number must be the same across all active invocations in a
2959 sub-group; otherwise, the results are undefined.
2962 Explanation of symbols used
2963 ------------------------------
2970 :math:`|x|` Absolute value of `x`.
2972 :math:`\lceil x \rceil` Ceiling of `x`.
2974 clamp(x,y,z) Clamp x between y and z.
2975 (x < y) ? y : (x > z) ? z : x
2977 :math:`\lfloor x\rfloor` Floor of `x`.
2979 :math:`\log_2{x}` Logarithm of `x`, base 2.
2981 max(x,y) Maximum of x and y.
2984 min(x,y) Minimum of x and y.
2987 partialx(x) Derivative of x relative to fragment's X.
2989 partialy(x) Derivative of x relative to fragment's Y.
2991 pop() Pop from stack.
2993 :math:`x^y` `x` to the power `y`.
2995 push(x) Push x on stack.
2999 trunc(x) Truncate x, i.e. drop the fraction bits.
3006 discard Discard fragment.
3010 target Label of target instruction.
3021 Declares a register that is will be referenced as an operand in Instruction
3024 File field contains register file that is being declared and is one
3027 UsageMask field specifies which of the register components can be accessed
3028 and is one of TGSI_WRITEMASK.
3030 The Local flag specifies that a given value isn't intended for
3031 subroutine parameter passing and, as a result, the implementation
3032 isn't required to give any guarantees of it being preserved across
3033 subroutine boundaries. As it's merely a compiler hint, the
3034 implementation is free to ignore it.
3036 If Dimension flag is set to 1, a Declaration Dimension token follows.
3038 If Semantic flag is set to 1, a Declaration Semantic token follows.
3040 If Interpolate flag is set to 1, a Declaration Interpolate token follows.
3042 If file is TGSI_FILE_RESOURCE, a Declaration Resource token follows.
3044 If Array flag is set to 1, a Declaration Array token follows.
3047 ^^^^^^^^^^^^^^^^^^^^^^^^
3049 Declarations can optional have an ArrayID attribute which can be referred by
3050 indirect addressing operands. An ArrayID of zero is reserved and treated as
3051 if no ArrayID is specified.
3053 If an indirect addressing operand refers to a specific declaration by using
3054 an ArrayID only the registers in this declaration are guaranteed to be
3055 accessed, accessing any register outside this declaration results in undefined
3056 behavior. Note that for compatibility the effective index is zero-based and
3057 not relative to the specified declaration
3059 If no ArrayID is specified with an indirect addressing operand the whole
3060 register file might be accessed by this operand. This is strongly discouraged
3061 and will prevent packing of scalar/vec2 arrays and effective alias analysis.
3062 This is only legal for TEMP and CONST register files.
3064 Declaration Semantic
3065 ^^^^^^^^^^^^^^^^^^^^^^^^
3067 Vertex and fragment shader input and output registers may be labeled
3068 with semantic information consisting of a name and index.
3070 Follows Declaration token if Semantic bit is set.
3072 Since its purpose is to link a shader with other stages of the pipeline,
3073 it is valid to follow only those Declaration tokens that declare a register
3074 either in INPUT or OUTPUT file.
3076 SemanticName field contains the semantic name of the register being declared.
3077 There is no default value.
3079 SemanticIndex is an optional subscript that can be used to distinguish
3080 different register declarations with the same semantic name. The default value
3083 The meanings of the individual semantic names are explained in the following
3086 TGSI_SEMANTIC_POSITION
3087 """"""""""""""""""""""
3089 For vertex shaders, TGSI_SEMANTIC_POSITION indicates the vertex shader
3090 output register which contains the homogeneous vertex position in the clip
3091 space coordinate system. After clipping, the X, Y and Z components of the
3092 vertex will be divided by the W value to get normalized device coordinates.
3094 For fragment shaders, TGSI_SEMANTIC_POSITION is used to indicate that
3095 fragment shader input (or system value, depending on which one is
3096 supported by the driver) contains the fragment's window position. The X
3097 component starts at zero and always increases from left to right.
3098 The Y component starts at zero and always increases but Y=0 may either
3099 indicate the top of the window or the bottom depending on the fragment
3100 coordinate origin convention (see TGSI_PROPERTY_FS_COORD_ORIGIN).
3101 The Z coordinate ranges from 0 to 1 to represent depth from the front
3102 to the back of the Z buffer. The W component contains the interpolated
3103 reciprocal of the vertex position W component (corresponding to gl_Fragcoord,
3104 but unlike d3d10 which interpolates the same 1/w but then gives back
3105 the reciprocal of the interpolated value).
3107 Fragment shaders may also declare an output register with
3108 TGSI_SEMANTIC_POSITION. Only the Z component is writable. This allows
3109 the fragment shader to change the fragment's Z position.
3116 For vertex shader outputs or fragment shader inputs/outputs, this
3117 label indicates that the register contains an R,G,B,A color.
3119 Several shader inputs/outputs may contain colors so the semantic index
3120 is used to distinguish them. For example, color[0] may be the diffuse
3121 color while color[1] may be the specular color.
3123 This label is needed so that the flat/smooth shading can be applied
3124 to the right interpolants during rasterization.
3128 TGSI_SEMANTIC_BCOLOR
3129 """"""""""""""""""""
3131 Back-facing colors are only used for back-facing polygons, and are only valid
3132 in vertex shader outputs. After rasterization, all polygons are front-facing
3133 and COLOR and BCOLOR end up occupying the same slots in the fragment shader,
3134 so all BCOLORs effectively become regular COLORs in the fragment shader.
3140 Vertex shader inputs and outputs and fragment shader inputs may be
3141 labeled with TGSI_SEMANTIC_FOG to indicate that the register contains
3142 a fog coordinate. Typically, the fragment shader will use the fog coordinate
3143 to compute a fog blend factor which is used to blend the normal fragment color
3144 with a constant fog color. But fog coord really is just an ordinary vec4
3145 register like regular semantics.
3151 Vertex shader input and output registers may be labeled with
3152 TGIS_SEMANTIC_PSIZE to indicate that the register contains a point size
3153 in the form (S, 0, 0, 1). The point size controls the width or diameter
3154 of points for rasterization. This label cannot be used in fragment
3157 When using this semantic, be sure to set the appropriate state in the
3158 :ref:`rasterizer` first.
3161 TGSI_SEMANTIC_TEXCOORD
3162 """"""""""""""""""""""
3164 Only available if PIPE_CAP_TGSI_TEXCOORD is exposed !
3166 Vertex shader outputs and fragment shader inputs may be labeled with
3167 this semantic to make them replaceable by sprite coordinates via the
3168 sprite_coord_enable state in the :ref:`rasterizer`.
3169 The semantic index permitted with this semantic is limited to <= 7.
3171 If the driver does not support TEXCOORD, sprite coordinate replacement
3172 applies to inputs with the GENERIC semantic instead.
3174 The intended use case for this semantic is gl_TexCoord.
3177 TGSI_SEMANTIC_PCOORD
3178 """"""""""""""""""""
3180 Only available if PIPE_CAP_TGSI_TEXCOORD is exposed !
3182 Fragment shader inputs may be labeled with TGSI_SEMANTIC_PCOORD to indicate
3183 that the register contains sprite coordinates in the form (x, y, 0, 1), if
3184 the current primitive is a point and point sprites are enabled. Otherwise,
3185 the contents of the register are undefined.
3187 The intended use case for this semantic is gl_PointCoord.
3190 TGSI_SEMANTIC_GENERIC
3191 """""""""""""""""""""
3193 All vertex/fragment shader inputs/outputs not labeled with any other
3194 semantic label can be considered to be generic attributes. Typical
3195 uses of generic inputs/outputs are texcoords and user-defined values.
3198 TGSI_SEMANTIC_NORMAL
3199 """"""""""""""""""""
3201 Indicates that a vertex shader input is a normal vector. This is
3202 typically only used for legacy graphics APIs.
3208 This label applies to fragment shader inputs (or system values,
3209 depending on which one is supported by the driver) and indicates that
3210 the register contains front/back-face information.
3212 If it is an input, it will be a floating-point vector in the form (F, 0, 0, 1),
3213 where F will be positive when the fragment belongs to a front-facing polygon,
3214 and negative when the fragment belongs to a back-facing polygon.
3216 If it is a system value, it will be an integer vector in the form (F, 0, 0, 1),
3217 where F is 0xffffffff when the fragment belongs to a front-facing polygon and
3218 0 when the fragment belongs to a back-facing polygon.
3221 TGSI_SEMANTIC_EDGEFLAG
3222 """"""""""""""""""""""
3224 For vertex shaders, this sematic label indicates that an input or
3225 output is a boolean edge flag. The register layout is [F, x, x, x]
3226 where F is 0.0 or 1.0 and x = don't care. Normally, the vertex shader
3227 simply copies the edge flag input to the edgeflag output.
3229 Edge flags are used to control which lines or points are actually
3230 drawn when the polygon mode converts triangles/quads/polygons into
3234 TGSI_SEMANTIC_STENCIL
3235 """""""""""""""""""""
3237 For fragment shaders, this semantic label indicates that an output
3238 is a writable stencil reference value. Only the Y component is writable.
3239 This allows the fragment shader to change the fragments stencilref value.
3242 TGSI_SEMANTIC_VIEWPORT_INDEX
3243 """"""""""""""""""""""""""""
3245 For geometry shaders, this semantic label indicates that an output
3246 contains the index of the viewport (and scissor) to use.
3247 This is an integer value, and only the X component is used.
3249 If PIPE_CAP_TGSI_VS_LAYER_VIEWPORT or PIPE_CAP_TGSI_TES_LAYER_VIEWPORT is
3250 supported, then this semantic label can also be used in vertex or
3251 tessellation evaluation shaders, respectively. Only the value written in the
3252 last vertex processing stage is used.
3258 For geometry shaders, this semantic label indicates that an output
3259 contains the layer value to use for the color and depth/stencil surfaces.
3260 This is an integer value, and only the X component is used.
3261 (Also known as rendertarget array index.)
3263 If PIPE_CAP_TGSI_VS_LAYER_VIEWPORT or PIPE_CAP_TGSI_TES_LAYER_VIEWPORT is
3264 supported, then this semantic label can also be used in vertex or
3265 tessellation evaluation shaders, respectively. Only the value written in the
3266 last vertex processing stage is used.
3269 TGSI_SEMANTIC_CLIPDIST
3270 """"""""""""""""""""""
3272 Note this covers clipping and culling distances.
3274 When components of vertex elements are identified this way, these
3275 values are each assumed to be a float32 signed distance to a plane.
3278 Primitive setup only invokes rasterization on pixels for which
3279 the interpolated plane distances are >= 0.
3282 Primitives will be completely discarded if the plane distance
3283 for all of the vertices in the primitive are < 0.
3284 If a vertex has a cull distance of NaN, that vertex counts as "out"
3287 Multiple clip/cull planes can be implemented simultaneously, by
3288 annotating multiple components of one or more vertex elements with
3289 the above specified semantic.
3290 The limits on both clip and cull distances are bound
3291 by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_COUNT define which defines
3292 the maximum number of components that can be used to hold the
3293 distances and by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_ELEMENT_COUNT
3294 which specifies the maximum number of registers which can be
3295 annotated with those semantics.
3296 The properties NUM_CLIPDIST_ENABLED and NUM_CULLDIST_ENABLED
3297 are used to divide up the 2 x vec4 space between clipping and culling.
3299 TGSI_SEMANTIC_SAMPLEID
3300 """"""""""""""""""""""
3302 For fragment shaders, this semantic label indicates that a system value
3303 contains the current sample id (i.e. gl_SampleID) as an unsigned int.
3304 Only the X component is used. If per-sample shading is not enabled,
3305 the result is (0, undef, undef, undef).
3307 Note that if the fragment shader uses this system value, the fragment
3308 shader is automatically executed at per sample frequency.
3310 TGSI_SEMANTIC_SAMPLEPOS
3311 """""""""""""""""""""""
3313 For fragment shaders, this semantic label indicates that a system
3314 value contains the current sample's position as float4(x, y, undef, undef)
3315 in the render target (i.e. gl_SamplePosition) when per-fragment shading
3316 is in effect. Position values are in the range [0, 1] where 0.5 is
3317 the center of the fragment.
3319 Note that if the fragment shader uses this system value, the fragment
3320 shader is automatically executed at per sample frequency.
3322 TGSI_SEMANTIC_SAMPLEMASK
3323 """"""""""""""""""""""""
3325 For fragment shaders, this semantic label can be applied to either a
3326 shader system value input or output.
3328 For a system value, the sample mask indicates the set of samples covered by
3329 the current primitive. If MSAA is not enabled, the value is (1, 0, 0, 0).
3331 For an output, the sample mask is used to disable further sample processing.
3333 For both, the register type is uint[4] but only the X component is used
3334 (i.e. gl_SampleMask[0]). Each bit corresponds to one sample position (up
3335 to 32x MSAA is supported).
3337 TGSI_SEMANTIC_INVOCATIONID
3338 """"""""""""""""""""""""""
3340 For geometry shaders, this semantic label indicates that a system value
3341 contains the current invocation id (i.e. gl_InvocationID).
3342 This is an integer value, and only the X component is used.
3344 TGSI_SEMANTIC_INSTANCEID
3345 """"""""""""""""""""""""
3347 For vertex shaders, this semantic label indicates that a system value contains
3348 the current instance id (i.e. gl_InstanceID). It does not include the base
3349 instance. This is an integer value, and only the X component is used.
3351 TGSI_SEMANTIC_VERTEXID
3352 """"""""""""""""""""""
3354 For vertex shaders, this semantic label indicates that a system value contains
3355 the current vertex id (i.e. gl_VertexID). It does (unlike in d3d10) include the
3356 base vertex. This is an integer value, and only the X component is used.
3358 TGSI_SEMANTIC_VERTEXID_NOBASE
3359 """""""""""""""""""""""""""""""
3361 For vertex shaders, this semantic label indicates that a system value contains
3362 the current vertex id without including the base vertex (this corresponds to
3363 d3d10 vertex id, so TGSI_SEMANTIC_VERTEXID_NOBASE + TGSI_SEMANTIC_BASEVERTEX
3364 == TGSI_SEMANTIC_VERTEXID). This is an integer value, and only the X component
3367 TGSI_SEMANTIC_BASEVERTEX
3368 """"""""""""""""""""""""
3370 For vertex shaders, this semantic label indicates that a system value contains
3371 the base vertex (i.e. gl_BaseVertex). Note that for non-indexed draw calls,
3372 this contains the first (or start) value instead.
3373 This is an integer value, and only the X component is used.
3375 TGSI_SEMANTIC_PRIMID
3376 """"""""""""""""""""
3378 For geometry and fragment shaders, this semantic label indicates the value
3379 contains the primitive id (i.e. gl_PrimitiveID). This is an integer value,
3380 and only the X component is used.
3381 FIXME: This right now can be either a ordinary input or a system value...
3387 For tessellation evaluation/control shaders, this semantic label indicates a
3388 generic per-patch attribute. Such semantics will not implicitly be per-vertex
3391 TGSI_SEMANTIC_TESSCOORD
3392 """""""""""""""""""""""
3394 For tessellation evaluation shaders, this semantic label indicates the
3395 coordinates of the vertex being processed. This is available in XYZ; W is
3398 TGSI_SEMANTIC_TESSOUTER
3399 """""""""""""""""""""""
3401 For tessellation evaluation/control shaders, this semantic label indicates the
3402 outer tessellation levels of the patch. Isoline tessellation will only have XY
3403 defined, triangle will have XYZ and quads will have XYZW defined. This
3404 corresponds to gl_TessLevelOuter.
3406 TGSI_SEMANTIC_TESSINNER
3407 """""""""""""""""""""""
3409 For tessellation evaluation/control shaders, this semantic label indicates the
3410 inner tessellation levels of the patch. The X value is only defined for
3411 triangle tessellation, while quads will have XY defined. This is entirely
3412 undefined for isoline tessellation.
3414 TGSI_SEMANTIC_VERTICESIN
3415 """"""""""""""""""""""""
3417 For tessellation evaluation/control shaders, this semantic label indicates the
3418 number of vertices provided in the input patch. Only the X value is defined.
3420 TGSI_SEMANTIC_HELPER_INVOCATION
3421 """""""""""""""""""""""""""""""
3423 For fragment shaders, this semantic indicates whether the current
3424 invocation is covered or not. Helper invocations are created in order
3425 to properly compute derivatives, however it may be desirable to skip
3426 some of the logic in those cases. See ``gl_HelperInvocation`` documentation.
3428 TGSI_SEMANTIC_BASEINSTANCE
3429 """"""""""""""""""""""""""
3431 For vertex shaders, the base instance argument supplied for this
3432 draw. This is an integer value, and only the X component is used.
3434 TGSI_SEMANTIC_DRAWID
3435 """"""""""""""""""""
3437 For vertex shaders, the zero-based index of the current draw in a
3438 ``glMultiDraw*`` invocation. This is an integer value, and only the X
3442 TGSI_SEMANTIC_WORK_DIM
3443 """"""""""""""""""""""
3445 For compute shaders started via opencl this retrieves the work_dim
3446 parameter to the clEnqueueNDRangeKernel call with which the shader
3450 TGSI_SEMANTIC_GRID_SIZE
3451 """""""""""""""""""""""
3453 For compute shaders, this semantic indicates the maximum (x, y, z) dimensions
3454 of a grid of thread blocks.
3457 TGSI_SEMANTIC_BLOCK_ID
3458 """"""""""""""""""""""
3460 For compute shaders, this semantic indicates the (x, y, z) coordinates of the
3461 current block inside of the grid.
3464 TGSI_SEMANTIC_BLOCK_SIZE
3465 """"""""""""""""""""""""
3467 For compute shaders, this semantic indicates the maximum (x, y, z) dimensions
3468 of a block in threads.
3471 TGSI_SEMANTIC_THREAD_ID
3472 """""""""""""""""""""""
3474 For compute shaders, this semantic indicates the (x, y, z) coordinates of the
3475 current thread inside of the block.
3478 TGSI_SEMANTIC_SUBGROUP_SIZE
3479 """""""""""""""""""""""""""
3481 This semantic indicates the subgroup size for the current invocation. This is
3482 an integer of at most 64, as it indicates the width of lanemasks. It does not
3483 depend on the number of invocations that are active.
3486 TGSI_SEMANTIC_SUBGROUP_INVOCATION
3487 """""""""""""""""""""""""""""""""
3489 The index of the current invocation within its subgroup.
3492 TGSI_SEMANTIC_SUBGROUP_EQ_MASK
3493 """"""""""""""""""""""""""""""
3495 A bit mask of ``bit index == TGSI_SEMANTIC_SUBGROUP_INVOCATION``, i.e.
3496 ``1 << subgroup_invocation`` in arbitrary precision arithmetic.
3499 TGSI_SEMANTIC_SUBGROUP_GE_MASK
3500 """"""""""""""""""""""""""""""
3502 A bit mask of ``bit index >= TGSI_SEMANTIC_SUBGROUP_INVOCATION``, i.e.
3503 ``((1 << (subgroup_size - subgroup_invocation)) - 1) << subgroup_invocation``
3504 in arbitrary precision arithmetic.
3507 TGSI_SEMANTIC_SUBGROUP_GT_MASK
3508 """"""""""""""""""""""""""""""
3510 A bit mask of ``bit index > TGSI_SEMANTIC_SUBGROUP_INVOCATION``, i.e.
3511 ``((1 << (subgroup_size - subgroup_invocation - 1)) - 1) << (subgroup_invocation + 1)``
3512 in arbitrary precision arithmetic.
3515 TGSI_SEMANTIC_SUBGROUP_LE_MASK
3516 """"""""""""""""""""""""""""""
3518 A bit mask of ``bit index <= TGSI_SEMANTIC_SUBGROUP_INVOCATION``, i.e.
3519 ``(1 << (subgroup_invocation + 1)) - 1`` in arbitrary precision arithmetic.
3522 TGSI_SEMANTIC_SUBGROUP_LT_MASK
3523 """"""""""""""""""""""""""""""
3525 A bit mask of ``bit index < TGSI_SEMANTIC_SUBGROUP_INVOCATION``, i.e.
3526 ``(1 << subgroup_invocation) - 1`` in arbitrary precision arithmetic.
3529 TGSI_SEMANTIC_VIEWPORT_MASK
3530 """""""""""""""""""""""""""
3532 A bit mask of viewports to broadcast the current primitive to. See
3533 GL_NV_viewport_array2 for more details.
3536 TGSI_SEMANTIC_TESS_DEFAULT_OUTER_LEVEL
3537 """"""""""""""""""""""""""""""""""""""
3539 A system value equal to the default_outer_level array set via set_tess_level.
3542 TGSI_SEMANTIC_TESS_DEFAULT_INNER_LEVEL
3543 """"""""""""""""""""""""""""""""""""""
3545 A system value equal to the default_inner_level array set via set_tess_level.
3548 Declaration Interpolate
3549 ^^^^^^^^^^^^^^^^^^^^^^^
3551 This token is only valid for fragment shader INPUT declarations.
3553 The Interpolate field specifes the way input is being interpolated by
3554 the rasteriser and is one of TGSI_INTERPOLATE_*.
3556 The Location field specifies the location inside the pixel that the
3557 interpolation should be done at, one of ``TGSI_INTERPOLATE_LOC_*``. Note that
3558 when per-sample shading is enabled, the implementation may choose to
3559 interpolate at the sample irrespective of the Location field.
3561 The CylindricalWrap bitfield specifies which register components
3562 should be subject to cylindrical wrapping when interpolating by the
3563 rasteriser. If TGSI_CYLINDRICAL_WRAP_X is set to 1, the X component
3564 should be interpolated according to cylindrical wrapping rules.
3567 Declaration Sampler View
3568 ^^^^^^^^^^^^^^^^^^^^^^^^
3570 Follows Declaration token if file is TGSI_FILE_SAMPLER_VIEW.
3572 DCL SVIEW[#], resource, type(s)
3574 Declares a shader input sampler view and assigns it to a SVIEW[#]
3577 resource can be one of BUFFER, 1D, 2D, 3D, 1DArray and 2DArray.
3579 type must be 1 or 4 entries (if specifying on a per-component
3580 level) out of UNORM, SNORM, SINT, UINT and FLOAT.
3582 For TEX\* style texture sample opcodes (as opposed to SAMPLE\* opcodes
3583 which take an explicit SVIEW[#] source register), there may be optionally
3584 SVIEW[#] declarations. In this case, the SVIEW index is implied by the
3585 SAMP index, and there must be a corresponding SVIEW[#] declaration for
3586 each SAMP[#] declaration. Drivers are free to ignore this if they wish.
3587 But note in particular that some drivers need to know the sampler type
3588 (float/int/unsigned) in order to generate the correct code, so cases
3589 where integer textures are sampled, SVIEW[#] declarations should be
3592 NOTE: It is NOT legal to mix SAMPLE\* style opcodes and TEX\* opcodes
3595 Declaration Resource
3596 ^^^^^^^^^^^^^^^^^^^^
3598 Follows Declaration token if file is TGSI_FILE_RESOURCE.
3600 DCL RES[#], resource [, WR] [, RAW]
3602 Declares a shader input resource and assigns it to a RES[#]
3605 resource can be one of BUFFER, 1D, 2D, 3D, CUBE, 1DArray and
3608 If the RAW keyword is not specified, the texture data will be
3609 subject to conversion, swizzling and scaling as required to yield
3610 the specified data type from the physical data format of the bound
3613 If the RAW keyword is specified, no channel conversion will be
3614 performed: the values read for each of the channels (X,Y,Z,W) will
3615 correspond to consecutive words in the same order and format
3616 they're found in memory. No element-to-address conversion will be
3617 performed either: the value of the provided X coordinate will be
3618 interpreted in byte units instead of texel units. The result of
3619 accessing a misaligned address is undefined.
3621 Usage of the STORE opcode is only allowed if the WR (writable) flag
3624 Hardware Atomic Register File
3625 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
3627 Hardware atomics are declared as a 2D array with an optional array id.
3629 The first member of the dimension is the buffer resource the atomic
3631 The second member is a range into the buffer resource, either for
3632 one or multiple counters. If this is an array, the declaration will have
3635 Each counter is 4 bytes in size, and index and ranges are in counters not bytes.
3639 This declares two atomics, one at the start of the buffer and one in the
3644 DCL HWATOMIC[1][1..3], ARRAY(1)
3646 This declares 5 atomics, one in buffer 0 at 0,
3647 one in buffer 1 at 0, and an array of 3 atomics in
3648 the buffer 1, starting at 1.
3651 ^^^^^^^^^^^^^^^^^^^^^^^^
3653 Properties are general directives that apply to the whole TGSI program.
3658 Specifies the fragment shader TGSI_SEMANTIC_POSITION coordinate origin.
3659 The default value is UPPER_LEFT.
3661 If UPPER_LEFT, the position will be (0,0) at the upper left corner and
3662 increase downward and rightward.
3663 If LOWER_LEFT, the position will be (0,0) at the lower left corner and
3664 increase upward and rightward.
3666 OpenGL defaults to LOWER_LEFT, and is configurable with the
3667 GL_ARB_fragment_coord_conventions extension.
3669 DirectX 9/10 use UPPER_LEFT.
3671 FS_COORD_PIXEL_CENTER
3672 """""""""""""""""""""
3674 Specifies the fragment shader TGSI_SEMANTIC_POSITION pixel center convention.
3675 The default value is HALF_INTEGER.
3677 If HALF_INTEGER, the fractionary part of the position will be 0.5
3678 If INTEGER, the fractionary part of the position will be 0.0
3680 Note that this does not affect the set of fragments generated by
3681 rasterization, which is instead controlled by half_pixel_center in the
3684 OpenGL defaults to HALF_INTEGER, and is configurable with the
3685 GL_ARB_fragment_coord_conventions extension.
3687 DirectX 9 uses INTEGER.
3688 DirectX 10 uses HALF_INTEGER.
3690 FS_COLOR0_WRITES_ALL_CBUFS
3691 """"""""""""""""""""""""""
3692 Specifies that writes to the fragment shader color 0 are replicated to all
3693 bound cbufs. This facilitates OpenGL's fragColor output vs fragData[0] where
3694 fragData is directed to a single color buffer, but fragColor is broadcast.
3697 """"""""""""""""""""""""""
3698 If this property is set on the program bound to the shader stage before the
3699 fragment shader, user clip planes should have no effect (be disabled) even if
3700 that shader does not write to any clip distance outputs and the rasterizer's
3701 clip_plane_enable is non-zero.
3702 This property is only supported by drivers that also support shader clip
3704 This is useful for APIs that don't have UCPs and where clip distances written
3705 by a shader cannot be disabled.
3710 Specifies the number of times a geometry shader should be executed for each
3711 input primitive. Each invocation will have a different
3712 TGSI_SEMANTIC_INVOCATIONID system value set. If not specified, assumed to
3715 VS_WINDOW_SPACE_POSITION
3716 """"""""""""""""""""""""""
3717 If this property is set on the vertex shader, the TGSI_SEMANTIC_POSITION output
3718 is assumed to contain window space coordinates.
3719 Division of X,Y,Z by W and the viewport transformation are disabled, and 1/W is
3720 directly taken from the 4-th component of the shader output.
3721 Naturally, clipping is not performed on window coordinates either.
3722 The effect of this property is undefined if a geometry or tessellation shader
3728 The number of vertices written by the tessellation control shader. This
3729 effectively defines the patch input size of the tessellation evaluation shader
3735 This sets the tessellation primitive mode, one of ``PIPE_PRIM_TRIANGLES``,
3736 ``PIPE_PRIM_QUADS``, or ``PIPE_PRIM_LINES``. (Unlike in GL, there is no
3737 separate isolines settings, the regular lines is assumed to mean isolines.)
3742 This sets the spacing mode of the tessellation generator, one of
3743 ``PIPE_TESS_SPACING_*``.
3748 This sets the vertex order to be clockwise if the value is 1, or
3749 counter-clockwise if set to 0.
3754 If set to a non-zero value, this turns on point mode for the tessellator,
3755 which means that points will be generated instead of primitives.
3757 NUM_CLIPDIST_ENABLED
3758 """"""""""""""""""""
3760 How many clip distance scalar outputs are enabled.
3762 NUM_CULLDIST_ENABLED
3763 """"""""""""""""""""
3765 How many cull distance scalar outputs are enabled.
3767 FS_EARLY_DEPTH_STENCIL
3768 """"""""""""""""""""""
3770 Whether depth test, stencil test, and occlusion query should run before
3771 the fragment shader (regardless of fragment shader side effects). Corresponds
3772 to GLSL early_fragment_tests.
3777 Which shader stage will MOST LIKELY follow after this shader when the shader
3778 is bound. This is only a hint to the driver and doesn't have to be precise.
3779 Only set for VS and TES.
3781 CS_FIXED_BLOCK_WIDTH / HEIGHT / DEPTH
3782 """""""""""""""""""""""""""""""""""""
3784 Threads per block in each dimension, if known at compile time. If the block size
3785 is known all three should be at least 1. If it is unknown they should all be set
3791 The MUL TGSI operation (FP32 multiplication) will return 0 if either
3792 of the operands are equal to 0. That means that 0 * Inf = 0. This
3793 should be set the same way for an entire pipeline. Note that this
3794 applies not only to the literal MUL TGSI opcode, but all FP32
3795 multiplications implied by other operations, such as MAD, FMA, DP2,
3796 DP3, DP4, DST, LOG, LRP, and possibly others. If there is a
3797 mismatch between shaders, then it is unspecified whether this behavior
3800 FS_POST_DEPTH_COVERAGE
3801 """"""""""""""""""""""
3803 When enabled, the input for TGSI_SEMANTIC_SAMPLEMASK will exclude samples
3804 that have failed the depth/stencil tests. This is only valid when
3805 FS_EARLY_DEPTH_STENCIL is also specified.
3807 LAYER_VIEWPORT_RELATIVE
3808 """""""""""""""""""""""
3810 When enabled, the TGSI_SEMATNIC_LAYER output value is relative to the
3811 current viewport. This is especially useful in conjunction with
3812 TGSI_SEMANTIC_VIEWPORT_MASK.
3815 Texture Sampling and Texture Formats
3816 ------------------------------------
3818 This table shows how texture image components are returned as (x,y,z,w) tuples
3819 by TGSI texture instructions, such as :opcode:`TEX`, :opcode:`TXD`, and
3820 :opcode:`TXP`. For reference, OpenGL and Direct3D conventions are shown as
3823 +--------------------+--------------+--------------------+--------------+
3824 | Texture Components | Gallium | OpenGL | Direct3D 9 |
3825 +====================+==============+====================+==============+
3826 | R | (r, 0, 0, 1) | (r, 0, 0, 1) | (r, 1, 1, 1) |
3827 +--------------------+--------------+--------------------+--------------+
3828 | RG | (r, g, 0, 1) | (r, g, 0, 1) | (r, g, 1, 1) |
3829 +--------------------+--------------+--------------------+--------------+
3830 | RGB | (r, g, b, 1) | (r, g, b, 1) | (r, g, b, 1) |
3831 +--------------------+--------------+--------------------+--------------+
3832 | RGBA | (r, g, b, a) | (r, g, b, a) | (r, g, b, a) |
3833 +--------------------+--------------+--------------------+--------------+
3834 | A | (0, 0, 0, a) | (0, 0, 0, a) | (0, 0, 0, a) |
3835 +--------------------+--------------+--------------------+--------------+
3836 | L | (l, l, l, 1) | (l, l, l, 1) | (l, l, l, 1) |
3837 +--------------------+--------------+--------------------+--------------+
3838 | LA | (l, l, l, a) | (l, l, l, a) | (l, l, l, a) |
3839 +--------------------+--------------+--------------------+--------------+
3840 | I | (i, i, i, i) | (i, i, i, i) | N/A |
3841 +--------------------+--------------+--------------------+--------------+
3842 | UV | XXX TBD | (0, 0, 0, 1) | (u, v, 1, 1) |
3843 | | | [#envmap-bumpmap]_ | |
3844 +--------------------+--------------+--------------------+--------------+
3845 | Z | XXX TBD | (z, z, z, 1) | (0, z, 0, 1) |
3846 | | | [#depth-tex-mode]_ | |
3847 +--------------------+--------------+--------------------+--------------+
3848 | S | (s, s, s, s) | unknown | unknown |
3849 +--------------------+--------------+--------------------+--------------+
3851 .. [#envmap-bumpmap] http://www.opengl.org/registry/specs/ATI/envmap_bumpmap.txt
3852 .. [#depth-tex-mode] the default is (z, z, z, 1) but may also be (0, 0, 0, z)
3853 or (z, z, z, z) depending on the value of GL_DEPTH_TEXTURE_MODE.