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 modifier on instructions).
31 For inputs which have a floating point type, both absolute value and negation
32 modifiers are supported (with absolute value being applied first).
33 TGSI_OPCODE_MOV is considered to have float input type for applying modifiers.
35 For inputs which have signed or unsigned type only the negate modifier is
42 ^^^^^^^^^^^^^^^^^^^^^^^^^
44 These opcodes are guaranteed to be available regardless of the driver being
47 .. opcode:: ARL - Address Register Load
51 dst.x = \lfloor src.x\rfloor
53 dst.y = \lfloor src.y\rfloor
55 dst.z = \lfloor src.z\rfloor
57 dst.w = \lfloor src.w\rfloor
60 .. opcode:: MOV - Move
73 .. opcode:: LIT - Light Coefficients
78 dst.y &= max(src.x, 0) \\
79 dst.z &= (src.x > 0) ? max(src.y, 0)^{clamp(src.w, -128, 128))} : 0 \\
83 .. opcode:: RCP - Reciprocal
85 This instruction replicates its result.
92 .. opcode:: RSQ - Reciprocal Square Root
94 This instruction replicates its result. The results are undefined for src <= 0.
98 dst = \frac{1}{\sqrt{src.x}}
101 .. opcode:: SQRT - Square Root
103 This instruction replicates its result. The results are undefined for src < 0.
110 .. opcode:: EXP - Approximate Exponential Base 2
114 dst.x &= 2^{\lfloor src.x\rfloor} \\
115 dst.y &= src.x - \lfloor src.x\rfloor \\
116 dst.z &= 2^{src.x} \\
120 .. opcode:: LOG - Approximate Logarithm Base 2
124 dst.x &= \lfloor\log_2{|src.x|}\rfloor \\
125 dst.y &= \frac{|src.x|}{2^{\lfloor\log_2{|src.x|}\rfloor}} \\
126 dst.z &= \log_2{|src.x|} \\
130 .. opcode:: MUL - Multiply
134 dst.x = src0.x \times src1.x
136 dst.y = src0.y \times src1.y
138 dst.z = src0.z \times src1.z
140 dst.w = src0.w \times src1.w
143 .. opcode:: ADD - Add
147 dst.x = src0.x + src1.x
149 dst.y = src0.y + src1.y
151 dst.z = src0.z + src1.z
153 dst.w = src0.w + src1.w
156 .. opcode:: DP3 - 3-component Dot Product
158 This instruction replicates its result.
162 dst = src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z
165 .. opcode:: DP4 - 4-component Dot Product
167 This instruction replicates its result.
171 dst = src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z + src0.w \times src1.w
174 .. opcode:: DST - Distance Vector
179 dst.y &= src0.y \times src1.y\\
184 .. opcode:: MIN - Minimum
188 dst.x = min(src0.x, src1.x)
190 dst.y = min(src0.y, src1.y)
192 dst.z = min(src0.z, src1.z)
194 dst.w = min(src0.w, src1.w)
197 .. opcode:: MAX - Maximum
201 dst.x = max(src0.x, src1.x)
203 dst.y = max(src0.y, src1.y)
205 dst.z = max(src0.z, src1.z)
207 dst.w = max(src0.w, src1.w)
210 .. opcode:: SLT - Set On Less Than
214 dst.x = (src0.x < src1.x) ? 1.0F : 0.0F
216 dst.y = (src0.y < src1.y) ? 1.0F : 0.0F
218 dst.z = (src0.z < src1.z) ? 1.0F : 0.0F
220 dst.w = (src0.w < src1.w) ? 1.0F : 0.0F
223 .. opcode:: SGE - Set On Greater Equal Than
227 dst.x = (src0.x >= src1.x) ? 1.0F : 0.0F
229 dst.y = (src0.y >= src1.y) ? 1.0F : 0.0F
231 dst.z = (src0.z >= src1.z) ? 1.0F : 0.0F
233 dst.w = (src0.w >= src1.w) ? 1.0F : 0.0F
236 .. opcode:: MAD - Multiply And Add
240 dst.x = src0.x \times src1.x + src2.x
242 dst.y = src0.y \times src1.y + src2.y
244 dst.z = src0.z \times src1.z + src2.z
246 dst.w = src0.w \times src1.w + src2.w
249 .. opcode:: SUB - Subtract
253 dst.x = src0.x - src1.x
255 dst.y = src0.y - src1.y
257 dst.z = src0.z - src1.z
259 dst.w = src0.w - src1.w
262 .. opcode:: LRP - Linear Interpolate
266 dst.x = src0.x \times src1.x + (1 - src0.x) \times src2.x
268 dst.y = src0.y \times src1.y + (1 - src0.y) \times src2.y
270 dst.z = src0.z \times src1.z + (1 - src0.z) \times src2.z
272 dst.w = src0.w \times src1.w + (1 - src0.w) \times src2.w
275 .. opcode:: CND - Condition
279 dst.x = (src2.x > 0.5) ? src0.x : src1.x
281 dst.y = (src2.y > 0.5) ? src0.y : src1.y
283 dst.z = (src2.z > 0.5) ? src0.z : src1.z
285 dst.w = (src2.w > 0.5) ? src0.w : src1.w
288 .. opcode:: DP2A - 2-component Dot Product And Add
292 dst.x = src0.x \times src1.x + src0.y \times src1.y + src2.x
294 dst.y = src0.x \times src1.x + src0.y \times src1.y + src2.x
296 dst.z = src0.x \times src1.x + src0.y \times src1.y + src2.x
298 dst.w = src0.x \times src1.x + src0.y \times src1.y + src2.x
301 .. opcode:: FRC - Fraction
305 dst.x = src.x - \lfloor src.x\rfloor
307 dst.y = src.y - \lfloor src.y\rfloor
309 dst.z = src.z - \lfloor src.z\rfloor
311 dst.w = src.w - \lfloor src.w\rfloor
314 .. opcode:: CLAMP - Clamp
318 dst.x = clamp(src0.x, src1.x, src2.x)
320 dst.y = clamp(src0.y, src1.y, src2.y)
322 dst.z = clamp(src0.z, src1.z, src2.z)
324 dst.w = clamp(src0.w, src1.w, src2.w)
327 .. opcode:: FLR - Floor
329 This is identical to :opcode:`ARL`.
333 dst.x = \lfloor src.x\rfloor
335 dst.y = \lfloor src.y\rfloor
337 dst.z = \lfloor src.z\rfloor
339 dst.w = \lfloor src.w\rfloor
342 .. opcode:: ROUND - Round
355 .. opcode:: EX2 - Exponential Base 2
357 This instruction replicates its result.
364 .. opcode:: LG2 - Logarithm Base 2
366 This instruction replicates its result.
373 .. opcode:: POW - Power
375 This instruction replicates its result.
379 dst = src0.x^{src1.x}
381 .. opcode:: XPD - Cross Product
385 dst.x = src0.y \times src1.z - src1.y \times src0.z
387 dst.y = src0.z \times src1.x - src1.z \times src0.x
389 dst.z = src0.x \times src1.y - src1.x \times src0.y
394 .. opcode:: ABS - Absolute
407 .. opcode:: RCC - Reciprocal Clamped
409 This instruction replicates its result.
411 XXX cleanup on aisle three
415 dst = (1 / src.x) > 0 ? clamp(1 / src.x, 5.42101e-020, 1.84467e+019) : clamp(1 / src.x, -1.84467e+019, -5.42101e-020)
418 .. opcode:: DPH - Homogeneous Dot Product
420 This instruction replicates its result.
424 dst = src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z + src1.w
427 .. opcode:: COS - Cosine
429 This instruction replicates its result.
436 .. opcode:: DDX - Derivative Relative To X
440 dst.x = partialx(src.x)
442 dst.y = partialx(src.y)
444 dst.z = partialx(src.z)
446 dst.w = partialx(src.w)
449 .. opcode:: DDY - Derivative Relative To Y
453 dst.x = partialy(src.x)
455 dst.y = partialy(src.y)
457 dst.z = partialy(src.z)
459 dst.w = partialy(src.w)
462 .. opcode:: PK2H - Pack Two 16-bit Floats
467 .. opcode:: PK2US - Pack Two Unsigned 16-bit Scalars
472 .. opcode:: PK4B - Pack Four Signed 8-bit Scalars
477 .. opcode:: PK4UB - Pack Four Unsigned 8-bit Scalars
482 .. opcode:: RFL - Reflection Vector
486 dst.x = 2 \times (src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z) / (src0.x \times src0.x + src0.y \times src0.y + src0.z \times src0.z) \times src0.x - src1.x
488 dst.y = 2 \times (src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z) / (src0.x \times src0.x + src0.y \times src0.y + src0.z \times src0.z) \times src0.y - src1.y
490 dst.z = 2 \times (src0.x \times src1.x + src0.y \times src1.y + src0.z \times src1.z) / (src0.x \times src0.x + src0.y \times src0.y + src0.z \times src0.z) \times src0.z - src1.z
496 Considered for removal.
499 .. opcode:: SEQ - Set On 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:: SFL - Set On False
514 This instruction replicates its result.
522 Considered for removal.
525 .. opcode:: SGT - Set On Greater Than
529 dst.x = (src0.x > src1.x) ? 1.0F : 0.0F
531 dst.y = (src0.y > src1.y) ? 1.0F : 0.0F
533 dst.z = (src0.z > src1.z) ? 1.0F : 0.0F
535 dst.w = (src0.w > src1.w) ? 1.0F : 0.0F
538 .. opcode:: SIN - Sine
540 This instruction replicates its result.
547 .. opcode:: SLE - Set On Less Equal Than
551 dst.x = (src0.x <= src1.x) ? 1.0F : 0.0F
553 dst.y = (src0.y <= src1.y) ? 1.0F : 0.0F
555 dst.z = (src0.z <= src1.z) ? 1.0F : 0.0F
557 dst.w = (src0.w <= src1.w) ? 1.0F : 0.0F
560 .. opcode:: SNE - Set On Not Equal
564 dst.x = (src0.x != src1.x) ? 1.0F : 0.0F
566 dst.y = (src0.y != src1.y) ? 1.0F : 0.0F
568 dst.z = (src0.z != src1.z) ? 1.0F : 0.0F
570 dst.w = (src0.w != src1.w) ? 1.0F : 0.0F
573 .. opcode:: STR - Set On True
575 This instruction replicates its result.
582 .. opcode:: TEX - Texture Lookup
584 for array textures src0.y contains the slice for 1D,
585 and src0.z contain the slice for 2D.
587 for shadow textures with no arrays, src0.z contains
590 for shadow textures with arrays, src0.z contains
591 the reference value for 1D arrays, and src0.w contains
592 the reference value for 2D arrays.
594 There is no way to pass a bias in the .w value for
595 shadow arrays, and GLSL doesn't allow this.
596 GLSL does allow cube shadows maps to take a bias value,
597 and we have to determine how this will look in TGSI.
605 dst = texture\_sample(unit, coord, bias)
607 .. opcode:: TXD - Texture Lookup with Derivatives
619 dst = texture\_sample\_deriv(unit, coord, bias, ddx, ddy)
622 .. opcode:: TXP - Projective Texture Lookup
626 coord.x = src0.x / src.w
628 coord.y = src0.y / src.w
630 coord.z = src0.z / src.w
636 dst = texture\_sample(unit, coord, bias)
639 .. opcode:: UP2H - Unpack Two 16-Bit Floats
645 Considered for removal.
647 .. opcode:: UP2US - Unpack Two Unsigned 16-Bit Scalars
653 Considered for removal.
655 .. opcode:: UP4B - Unpack Four Signed 8-Bit Values
661 Considered for removal.
663 .. opcode:: UP4UB - Unpack Four Unsigned 8-Bit Scalars
669 Considered for removal.
671 .. opcode:: X2D - 2D Coordinate Transformation
675 dst.x = src0.x + src1.x \times src2.x + src1.y \times src2.y
677 dst.y = src0.y + src1.x \times src2.z + src1.y \times src2.w
679 dst.z = src0.x + src1.x \times src2.x + src1.y \times src2.y
681 dst.w = src0.y + src1.x \times src2.z + src1.y \times src2.w
685 Considered for removal.
688 .. opcode:: ARA - Address Register Add
694 Considered for removal.
696 .. opcode:: ARR - Address Register Load With Round
709 .. opcode:: SSG - Set Sign
713 dst.x = (src.x > 0) ? 1 : (src.x < 0) ? -1 : 0
715 dst.y = (src.y > 0) ? 1 : (src.y < 0) ? -1 : 0
717 dst.z = (src.z > 0) ? 1 : (src.z < 0) ? -1 : 0
719 dst.w = (src.w > 0) ? 1 : (src.w < 0) ? -1 : 0
722 .. opcode:: CMP - Compare
726 dst.x = (src0.x < 0) ? src1.x : src2.x
728 dst.y = (src0.y < 0) ? src1.y : src2.y
730 dst.z = (src0.z < 0) ? src1.z : src2.z
732 dst.w = (src0.w < 0) ? src1.w : src2.w
735 .. opcode:: KILL_IF - Conditional Discard
737 Conditional discard. Allowed in fragment shaders only.
741 if (src.x < 0 || src.y < 0 || src.z < 0 || src.w < 0)
746 .. opcode:: KILL - Discard
748 Unconditional discard. Allowed in fragment shaders only.
751 .. opcode:: SCS - Sine Cosine
764 .. opcode:: TXB - Texture Lookup With Bias
778 dst = texture\_sample(unit, coord, bias)
781 .. opcode:: NRM - 3-component Vector Normalise
785 dst.x = src.x / (src.x \times src.x + src.y \times src.y + src.z \times src.z)
787 dst.y = src.y / (src.x \times src.x + src.y \times src.y + src.z \times src.z)
789 dst.z = src.z / (src.x \times src.x + src.y \times src.y + src.z \times src.z)
794 .. opcode:: DIV - Divide
798 dst.x = \frac{src0.x}{src1.x}
800 dst.y = \frac{src0.y}{src1.y}
802 dst.z = \frac{src0.z}{src1.z}
804 dst.w = \frac{src0.w}{src1.w}
807 .. opcode:: DP2 - 2-component Dot Product
809 This instruction replicates its result.
813 dst = src0.x \times src1.x + src0.y \times src1.y
816 .. opcode:: TXL - Texture Lookup With explicit LOD
830 dst = texture\_sample(unit, coord, lod)
833 .. opcode:: PUSHA - Push Address Register On Stack
842 Considered for cleanup.
846 Considered for removal.
848 .. opcode:: POPA - Pop Address Register From Stack
857 Considered for cleanup.
861 Considered for removal.
864 .. opcode:: BRA - Branch
870 Considered for removal.
873 .. opcode:: CALLNZ - Subroutine Call If Not Zero
879 Considered for cleanup.
883 Considered for removal.
887 ^^^^^^^^^^^^^^^^^^^^^^^^
889 These opcodes are primarily provided for special-use computational shaders.
890 Support for these opcodes indicated by a special pipe capability bit (TBD).
892 XXX doesn't look like most of the opcodes really belong here.
894 .. opcode:: CEIL - Ceiling
898 dst.x = \lceil src.x\rceil
900 dst.y = \lceil src.y\rceil
902 dst.z = \lceil src.z\rceil
904 dst.w = \lceil src.w\rceil
907 .. opcode:: TRUNC - Truncate
920 .. opcode:: MOD - Modulus
924 dst.x = src0.x \bmod src1.x
926 dst.y = src0.y \bmod src1.y
928 dst.z = src0.z \bmod src1.z
930 dst.w = src0.w \bmod src1.w
933 .. opcode:: UARL - Integer Address Register Load
935 Moves the contents of the source register, assumed to be an integer, into the
936 destination register, which is assumed to be an address (ADDR) register.
939 .. opcode:: SAD - Sum Of Absolute Differences
943 dst.x = |src0.x - src1.x| + src2.x
945 dst.y = |src0.y - src1.y| + src2.y
947 dst.z = |src0.z - src1.z| + src2.z
949 dst.w = |src0.w - src1.w| + src2.w
952 .. opcode:: TXF - Texel Fetch
954 As per NV_gpu_shader4, extract a single texel from a specified texture
955 image. The source sampler may not be a CUBE or SHADOW. src 0 is a
956 four-component signed integer vector used to identify the single texel
957 accessed. 3 components + level. src 1 is a 3 component constant signed
958 integer vector, with each component only have a range of -8..+8 (hw only
959 seems to deal with this range, interface allows for up to unsigned int).
960 TXF(uint_vec coord, int_vec offset).
963 .. opcode:: TXQ - Texture Size Query
965 As per NV_gpu_program4, retrieve the dimensions of the texture depending on
966 the target. For 1D (width), 2D/RECT/CUBE (width, height), 3D (width, height,
967 depth), 1D array (width, layers), 2D array (width, height, layers).
968 Also return the number of accessible levels (last_level - first_level + 1)
975 dst.x = texture\_width(unit, lod)
977 dst.y = texture\_height(unit, lod)
979 dst.z = texture\_depth(unit, lod)
981 dst.w = texture\_levels(unit)
983 .. opcode:: TG4 - Texture Gather
985 As per ARB_texture_gather, gathers the four texels to be used in a bi-linear
986 filtering operation and packs them into a single register. Only works with
987 2D, 2D array, cubemaps, and cubemaps arrays. For 2D textures, only the
988 addressing modes of the sampler and the top level of any mip pyramid are
989 used. Set W to zero. It behaves like the TEX instruction, but a filtered
990 sample is not generated. The four samples that contribute to filtering are
991 placed into xyzw in clockwise order, starting with the (u,v) texture
992 coordinate delta at the following locations (-, +), (+, +), (+, -), (-, -),
993 where the magnitude of the deltas are half a texel.
995 PIPE_CAP_TEXTURE_SM5 enhances this instruction to support shadow per-sample
996 depth compares, single component selection, and a non-constant offset. It
997 doesn't allow support for the GL independent offset to get i0,j0. This would
998 require another CAP is hw can do it natively. For now we lower that before
1007 dst = texture\_gather4 (unit, coord, component)
1009 (with SM5 - cube array shadow)
1017 dst = texture\_gather (uint, coord, compare)
1019 .. opcode:: LODQ - level of detail query
1021 Compute the LOD information that the texture pipe would use to access the
1022 texture. The Y component contains the computed LOD lambda_prime. The X
1023 component contains the LOD that will be accessed, based on min/max lod's
1030 dst.xy = lodq(uint, coord);
1033 ^^^^^^^^^^^^^^^^^^^^^^^^
1034 These opcodes are used for integer operations.
1035 Support for these opcodes indicated by PIPE_SHADER_CAP_INTEGERS (all of them?)
1038 .. opcode:: I2F - Signed Integer To Float
1040 Rounding is unspecified (round to nearest even suggested).
1044 dst.x = (float) src.x
1046 dst.y = (float) src.y
1048 dst.z = (float) src.z
1050 dst.w = (float) src.w
1053 .. opcode:: U2F - Unsigned Integer To Float
1055 Rounding is unspecified (round to nearest even suggested).
1059 dst.x = (float) src.x
1061 dst.y = (float) src.y
1063 dst.z = (float) src.z
1065 dst.w = (float) src.w
1068 .. opcode:: F2I - Float to Signed Integer
1070 Rounding is towards zero (truncate).
1071 Values outside signed range (including NaNs) produce undefined results.
1084 .. opcode:: F2U - Float to Unsigned Integer
1086 Rounding is towards zero (truncate).
1087 Values outside unsigned range (including NaNs) produce undefined results.
1091 dst.x = (unsigned) src.x
1093 dst.y = (unsigned) src.y
1095 dst.z = (unsigned) src.z
1097 dst.w = (unsigned) src.w
1100 .. opcode:: UADD - Integer Add
1102 This instruction works the same for signed and unsigned integers.
1103 The low 32bit of the result is returned.
1107 dst.x = src0.x + src1.x
1109 dst.y = src0.y + src1.y
1111 dst.z = src0.z + src1.z
1113 dst.w = src0.w + src1.w
1116 .. opcode:: UMAD - Integer Multiply And Add
1118 This instruction works the same for signed and unsigned integers.
1119 The multiplication returns the low 32bit (as does the result itself).
1123 dst.x = src0.x \times src1.x + src2.x
1125 dst.y = src0.y \times src1.y + src2.y
1127 dst.z = src0.z \times src1.z + src2.z
1129 dst.w = src0.w \times src1.w + src2.w
1132 .. opcode:: UMUL - Integer Multiply
1134 This instruction works the same for signed and unsigned integers.
1135 The low 32bit of the result is returned.
1139 dst.x = src0.x \times src1.x
1141 dst.y = src0.y \times src1.y
1143 dst.z = src0.z \times src1.z
1145 dst.w = src0.w \times src1.w
1148 .. opcode:: IMUL_HI - Signed Integer Multiply High Bits
1150 The high 32bits of the multiplication of 2 signed integers are returned.
1154 dst.x = (src0.x \times src1.x) >> 32
1156 dst.y = (src0.y \times src1.y) >> 32
1158 dst.z = (src0.z \times src1.z) >> 32
1160 dst.w = (src0.w \times src1.w) >> 32
1163 .. opcode:: UMUL_HI - Unsigned Integer Multiply High Bits
1165 The high 32bits of the multiplication of 2 unsigned integers are returned.
1169 dst.x = (src0.x \times src1.x) >> 32
1171 dst.y = (src0.y \times src1.y) >> 32
1173 dst.z = (src0.z \times src1.z) >> 32
1175 dst.w = (src0.w \times src1.w) >> 32
1178 .. opcode:: IDIV - Signed Integer Division
1180 TBD: behavior for division by zero.
1184 dst.x = src0.x \ src1.x
1186 dst.y = src0.y \ src1.y
1188 dst.z = src0.z \ src1.z
1190 dst.w = src0.w \ src1.w
1193 .. opcode:: UDIV - Unsigned Integer Division
1195 For division by zero, 0xffffffff is returned.
1199 dst.x = src0.x \ src1.x
1201 dst.y = src0.y \ src1.y
1203 dst.z = src0.z \ src1.z
1205 dst.w = src0.w \ src1.w
1208 .. opcode:: UMOD - Unsigned Integer Remainder
1210 If second arg is zero, 0xffffffff is returned.
1214 dst.x = src0.x \ src1.x
1216 dst.y = src0.y \ src1.y
1218 dst.z = src0.z \ src1.z
1220 dst.w = src0.w \ src1.w
1223 .. opcode:: NOT - Bitwise Not
1236 .. opcode:: AND - Bitwise And
1240 dst.x = src0.x \& src1.x
1242 dst.y = src0.y \& src1.y
1244 dst.z = src0.z \& src1.z
1246 dst.w = src0.w \& src1.w
1249 .. opcode:: OR - Bitwise Or
1253 dst.x = src0.x | src1.x
1255 dst.y = src0.y | src1.y
1257 dst.z = src0.z | src1.z
1259 dst.w = src0.w | src1.w
1262 .. opcode:: XOR - Bitwise Xor
1266 dst.x = src0.x \oplus src1.x
1268 dst.y = src0.y \oplus src1.y
1270 dst.z = src0.z \oplus src1.z
1272 dst.w = src0.w \oplus src1.w
1275 .. opcode:: IMAX - Maximum of Signed Integers
1279 dst.x = max(src0.x, src1.x)
1281 dst.y = max(src0.y, src1.y)
1283 dst.z = max(src0.z, src1.z)
1285 dst.w = max(src0.w, src1.w)
1288 .. opcode:: UMAX - Maximum of Unsigned Integers
1292 dst.x = max(src0.x, src1.x)
1294 dst.y = max(src0.y, src1.y)
1296 dst.z = max(src0.z, src1.z)
1298 dst.w = max(src0.w, src1.w)
1301 .. opcode:: IMIN - Minimum of Signed Integers
1305 dst.x = min(src0.x, src1.x)
1307 dst.y = min(src0.y, src1.y)
1309 dst.z = min(src0.z, src1.z)
1311 dst.w = min(src0.w, src1.w)
1314 .. opcode:: UMIN - Minimum of Unsigned Integers
1318 dst.x = min(src0.x, src1.x)
1320 dst.y = min(src0.y, src1.y)
1322 dst.z = min(src0.z, src1.z)
1324 dst.w = min(src0.w, src1.w)
1327 .. opcode:: SHL - Shift Left
1329 The shift count is masked with 0x1f before the shift is applied.
1333 dst.x = src0.x << (0x1f \& src1.x)
1335 dst.y = src0.y << (0x1f \& src1.y)
1337 dst.z = src0.z << (0x1f \& src1.z)
1339 dst.w = src0.w << (0x1f \& src1.w)
1342 .. opcode:: ISHR - Arithmetic Shift Right (of Signed Integer)
1344 The shift count is masked with 0x1f before the shift is applied.
1348 dst.x = src0.x >> (0x1f \& src1.x)
1350 dst.y = src0.y >> (0x1f \& src1.y)
1352 dst.z = src0.z >> (0x1f \& src1.z)
1354 dst.w = src0.w >> (0x1f \& src1.w)
1357 .. opcode:: USHR - Logical Shift Right
1359 The shift count is masked with 0x1f before the shift is applied.
1363 dst.x = src0.x >> (unsigned) (0x1f \& src1.x)
1365 dst.y = src0.y >> (unsigned) (0x1f \& src1.y)
1367 dst.z = src0.z >> (unsigned) (0x1f \& src1.z)
1369 dst.w = src0.w >> (unsigned) (0x1f \& src1.w)
1372 .. opcode:: UCMP - Integer Conditional Move
1376 dst.x = src0.x ? src1.x : src2.x
1378 dst.y = src0.y ? src1.y : src2.y
1380 dst.z = src0.z ? src1.z : src2.z
1382 dst.w = src0.w ? src1.w : src2.w
1386 .. opcode:: ISSG - Integer Set Sign
1390 dst.x = (src0.x < 0) ? -1 : (src0.x > 0) ? 1 : 0
1392 dst.y = (src0.y < 0) ? -1 : (src0.y > 0) ? 1 : 0
1394 dst.z = (src0.z < 0) ? -1 : (src0.z > 0) ? 1 : 0
1396 dst.w = (src0.w < 0) ? -1 : (src0.w > 0) ? 1 : 0
1400 .. opcode:: FSLT - Float Set On Less Than (ordered)
1402 Same comparison as SLT but returns integer instead of 1.0/0.0 float
1406 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1408 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1410 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1412 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1415 .. opcode:: ISLT - Signed Integer Set On Less Than
1419 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1421 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1423 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1425 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1428 .. opcode:: USLT - Unsigned Integer Set On Less Than
1432 dst.x = (src0.x < src1.x) ? \sim 0 : 0
1434 dst.y = (src0.y < src1.y) ? \sim 0 : 0
1436 dst.z = (src0.z < src1.z) ? \sim 0 : 0
1438 dst.w = (src0.w < src1.w) ? \sim 0 : 0
1441 .. opcode:: FSGE - Float Set On Greater Equal Than (ordered)
1443 Same comparison as SGE but returns integer instead of 1.0/0.0 float
1447 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1449 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1451 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1453 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1456 .. opcode:: ISGE - Signed Integer Set On Greater Equal Than
1460 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1462 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1464 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1466 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1469 .. opcode:: USGE - Unsigned Integer Set On Greater Equal Than
1473 dst.x = (src0.x >= src1.x) ? \sim 0 : 0
1475 dst.y = (src0.y >= src1.y) ? \sim 0 : 0
1477 dst.z = (src0.z >= src1.z) ? \sim 0 : 0
1479 dst.w = (src0.w >= src1.w) ? \sim 0 : 0
1482 .. opcode:: FSEQ - Float Set On Equal (ordered)
1484 Same comparison as SEQ but returns integer instead of 1.0/0.0 float
1488 dst.x = (src0.x == src1.x) ? \sim 0 : 0
1490 dst.y = (src0.y == src1.y) ? \sim 0 : 0
1492 dst.z = (src0.z == src1.z) ? \sim 0 : 0
1494 dst.w = (src0.w == src1.w) ? \sim 0 : 0
1497 .. opcode:: USEQ - Integer Set On Equal
1501 dst.x = (src0.x == src1.x) ? \sim 0 : 0
1503 dst.y = (src0.y == src1.y) ? \sim 0 : 0
1505 dst.z = (src0.z == src1.z) ? \sim 0 : 0
1507 dst.w = (src0.w == src1.w) ? \sim 0 : 0
1510 .. opcode:: FSNE - Float Set On Not Equal (unordered)
1512 Same comparison as SNE but returns integer instead of 1.0/0.0 float
1516 dst.x = (src0.x != src1.x) ? \sim 0 : 0
1518 dst.y = (src0.y != src1.y) ? \sim 0 : 0
1520 dst.z = (src0.z != src1.z) ? \sim 0 : 0
1522 dst.w = (src0.w != src1.w) ? \sim 0 : 0
1525 .. opcode:: USNE - Integer Set On Not Equal
1529 dst.x = (src0.x != src1.x) ? \sim 0 : 0
1531 dst.y = (src0.y != src1.y) ? \sim 0 : 0
1533 dst.z = (src0.z != src1.z) ? \sim 0 : 0
1535 dst.w = (src0.w != src1.w) ? \sim 0 : 0
1538 .. opcode:: INEG - Integer Negate
1553 .. opcode:: IABS - Integer Absolute Value
1567 These opcodes are used for bit-level manipulation of integers.
1569 .. opcode:: IBFE - Signed Bitfield Extract
1571 See SM5 instruction of the same name. Extracts a set of bits from the input,
1572 and sign-extends them if the high bit of the extracted window is set.
1576 def ibfe(value, offset, bits):
1577 offset = offset & 0x1f
1579 if bits == 0: return 0
1580 # Note: >> sign-extends
1581 if width + offset < 32:
1582 return (value << (32 - offset - bits)) >> (32 - bits)
1584 return value >> offset
1586 .. opcode:: UBFE - Unsigned Bitfield Extract
1588 See SM5 instruction of the same name. Extracts a set of bits from the input,
1589 without any sign-extension.
1593 def ubfe(value, offset, bits):
1594 offset = offset & 0x1f
1596 if bits == 0: return 0
1597 # Note: >> does not sign-extend
1598 if width + offset < 32:
1599 return (value << (32 - offset - bits)) >> (32 - bits)
1601 return value >> offset
1603 .. opcode:: BFI - Bitfield Insert
1605 See SM5 instruction of the same name. Replaces a bit region of 'base' with
1606 the low bits of 'insert'.
1610 def bfi(base, insert, offset, bits):
1611 offset = offset & 0x1f
1613 mask = ((1 << bits) - 1) << offset
1614 return ((insert << offset) & mask) | (base & ~mask)
1616 .. opcode:: BREV - Bitfield Reverse
1618 See SM5 instruction BFREV. Reverses the bits of the argument.
1620 .. opcode:: POPC - Population Count
1622 See SM5 instruction COUNTBITS. Counts the number of set bits in the argument.
1624 .. opcode:: LSB - Index of lowest set bit
1626 See SM5 instruction FIRSTBIT_LO. Computes the 0-based index of the first set
1627 bit of the argument. Returns -1 if none are set.
1629 .. opcode:: IMSB - Index of highest non-sign bit
1631 See SM5 instruction FIRSTBIT_SHI. Computes the 0-based index of the highest
1632 non-sign bit of the argument (i.e. highest 0 bit for negative numbers,
1633 highest 1 bit for positive numbers). Returns -1 if all bits are the same
1634 (i.e. for inputs 0 and -1).
1636 .. opcode:: UMSB - Index of highest set bit
1638 See SM5 instruction FIRSTBIT_HI. Computes the 0-based index of the highest
1639 set bit of the argument. Returns -1 if none are set.
1642 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1644 These opcodes are only supported in geometry shaders; they have no meaning
1645 in any other type of shader.
1647 .. opcode:: EMIT - Emit
1649 Generate a new vertex for the current primitive into the specified vertex
1650 stream using the values in the output registers.
1653 .. opcode:: ENDPRIM - End Primitive
1655 Complete the current primitive in the specified vertex stream (consisting of
1656 the emitted vertices), and start a new one.
1662 These opcodes are part of :term:`GLSL`'s opcode set. Support for these
1663 opcodes is determined by a special capability bit, ``GLSL``.
1664 Some require glsl version 1.30 (UIF/BREAKC/SWITCH/CASE/DEFAULT/ENDSWITCH).
1666 .. opcode:: CAL - Subroutine Call
1672 .. opcode:: RET - Subroutine Call Return
1677 .. opcode:: CONT - Continue
1679 Unconditionally moves the point of execution to the instruction after the
1680 last bgnloop. The instruction must appear within a bgnloop/endloop.
1684 Support for CONT is determined by a special capability bit,
1685 ``TGSI_CONT_SUPPORTED``. See :ref:`Screen` for more information.
1688 .. opcode:: BGNLOOP - Begin a Loop
1690 Start a loop. Must have a matching endloop.
1693 .. opcode:: BGNSUB - Begin Subroutine
1695 Starts definition of a subroutine. Must have a matching endsub.
1698 .. opcode:: ENDLOOP - End a Loop
1700 End a loop started with bgnloop.
1703 .. opcode:: ENDSUB - End Subroutine
1705 Ends definition of a subroutine.
1708 .. opcode:: NOP - No Operation
1713 .. opcode:: BRK - Break
1715 Unconditionally moves the point of execution to the instruction after the
1716 next endloop or endswitch. The instruction must appear within a loop/endloop
1717 or switch/endswitch.
1720 .. opcode:: BREAKC - Break Conditional
1722 Conditionally moves the point of execution to the instruction after the
1723 next endloop or endswitch. The instruction must appear within a loop/endloop
1724 or switch/endswitch.
1725 Condition evaluates to true if src0.x != 0 where src0.x is interpreted
1726 as an integer register.
1730 Considered for removal as it's quite inconsistent wrt other opcodes
1731 (could emulate with UIF/BRK/ENDIF).
1734 .. opcode:: IF - Float If
1736 Start an IF ... ELSE .. ENDIF block. Condition evaluates to true if
1740 where src0.x is interpreted as a floating point register.
1743 .. opcode:: UIF - Bitwise If
1745 Start an UIF ... ELSE .. ENDIF block. Condition evaluates to true if
1749 where src0.x is interpreted as an integer register.
1752 .. opcode:: ELSE - Else
1754 Starts an else block, after an IF or UIF statement.
1757 .. opcode:: ENDIF - End If
1759 Ends an IF or UIF block.
1762 .. opcode:: SWITCH - Switch
1764 Starts a C-style switch expression. The switch consists of one or multiple
1765 CASE statements, and at most one DEFAULT statement. Execution of a statement
1766 ends when a BRK is hit, but just like in C falling through to other cases
1767 without a break is allowed. Similarly, DEFAULT label is allowed anywhere not
1768 just as last statement, and fallthrough is allowed into/from it.
1769 CASE src arguments are evaluated at bit level against the SWITCH src argument.
1775 (some instructions here)
1778 (some instructions here)
1781 (some instructions here)
1786 .. opcode:: CASE - Switch case
1788 This represents a switch case label. The src arg must be an integer immediate.
1791 .. opcode:: DEFAULT - Switch default
1793 This represents the default case in the switch, which is taken if no other
1797 .. opcode:: ENDSWITCH - End of switch
1799 Ends a switch expression.
1802 .. opcode:: NRM4 - 4-component Vector Normalise
1804 This instruction replicates its result.
1808 dst = \frac{src.x}{src.x \times src.x + src.y \times src.y + src.z \times src.z + src.w \times src.w}
1816 The double-precision opcodes reinterpret four-component vectors into
1817 two-component vectors with doubled precision in each component.
1819 Support for these opcodes is XXX undecided. :T
1821 .. opcode:: DADD - Add
1825 dst.xy = src0.xy + src1.xy
1827 dst.zw = src0.zw + src1.zw
1830 .. opcode:: DDIV - Divide
1834 dst.xy = src0.xy / src1.xy
1836 dst.zw = src0.zw / src1.zw
1838 .. opcode:: DSEQ - Set on Equal
1842 dst.xy = src0.xy == src1.xy ? 1.0F : 0.0F
1844 dst.zw = src0.zw == src1.zw ? 1.0F : 0.0F
1846 .. opcode:: DSLT - Set on Less than
1850 dst.xy = src0.xy < src1.xy ? 1.0F : 0.0F
1852 dst.zw = src0.zw < src1.zw ? 1.0F : 0.0F
1854 .. opcode:: DFRAC - Fraction
1858 dst.xy = src.xy - \lfloor src.xy\rfloor
1860 dst.zw = src.zw - \lfloor src.zw\rfloor
1863 .. opcode:: DFRACEXP - Convert Number to Fractional and Integral Components
1865 Like the ``frexp()`` routine in many math libraries, this opcode stores the
1866 exponent of its source to ``dst0``, and the significand to ``dst1``, such that
1867 :math:`dst1 \times 2^{dst0} = src` .
1871 dst0.xy = exp(src.xy)
1873 dst1.xy = frac(src.xy)
1875 dst0.zw = exp(src.zw)
1877 dst1.zw = frac(src.zw)
1879 .. opcode:: DLDEXP - Multiply Number by Integral Power of 2
1881 This opcode is the inverse of :opcode:`DFRACEXP`.
1885 dst.xy = src0.xy \times 2^{src1.xy}
1887 dst.zw = src0.zw \times 2^{src1.zw}
1889 .. opcode:: DMIN - Minimum
1893 dst.xy = min(src0.xy, src1.xy)
1895 dst.zw = min(src0.zw, src1.zw)
1897 .. opcode:: DMAX - Maximum
1901 dst.xy = max(src0.xy, src1.xy)
1903 dst.zw = max(src0.zw, src1.zw)
1905 .. opcode:: DMUL - Multiply
1909 dst.xy = src0.xy \times src1.xy
1911 dst.zw = src0.zw \times src1.zw
1914 .. opcode:: DMAD - Multiply And Add
1918 dst.xy = src0.xy \times src1.xy + src2.xy
1920 dst.zw = src0.zw \times src1.zw + src2.zw
1923 .. opcode:: DRCP - Reciprocal
1927 dst.xy = \frac{1}{src.xy}
1929 dst.zw = \frac{1}{src.zw}
1931 .. opcode:: DSQRT - Square Root
1935 dst.xy = \sqrt{src.xy}
1937 dst.zw = \sqrt{src.zw}
1940 .. _samplingopcodes:
1942 Resource Sampling Opcodes
1943 ^^^^^^^^^^^^^^^^^^^^^^^^^
1945 Those opcodes follow very closely semantics of the respective Direct3D
1946 instructions. If in doubt double check Direct3D documentation.
1947 Note that the swizzle on SVIEW (src1) determines texel swizzling
1952 Using provided address, sample data from the specified texture using the
1953 filtering mode identified by the gven sampler. The source data may come from
1954 any resource type other than buffers.
1956 Syntax: ``SAMPLE dst, address, sampler_view, sampler``
1958 Example: ``SAMPLE TEMP[0], TEMP[1], SVIEW[0], SAMP[0]``
1960 .. opcode:: SAMPLE_I
1962 Simplified alternative to the SAMPLE instruction. Using the provided
1963 integer address, SAMPLE_I fetches data from the specified sampler view
1964 without any filtering. The source data may come from any resource type
1967 Syntax: ``SAMPLE_I dst, address, sampler_view``
1969 Example: ``SAMPLE_I TEMP[0], TEMP[1], SVIEW[0]``
1971 The 'address' is specified as unsigned integers. If the 'address' is out of
1972 range [0...(# texels - 1)] the result of the fetch is always 0 in all
1973 components. As such the instruction doesn't honor address wrap modes, in
1974 cases where that behavior is desirable 'SAMPLE' instruction should be used.
1975 address.w always provides an unsigned integer mipmap level. If the value is
1976 out of the range then the instruction always returns 0 in all components.
1977 address.yz are ignored for buffers and 1d textures. address.z is ignored
1978 for 1d texture arrays and 2d textures.
1980 For 1D texture arrays address.y provides the array index (also as unsigned
1981 integer). If the value is out of the range of available array indices
1982 [0... (array size - 1)] then the opcode always returns 0 in all components.
1983 For 2D texture arrays address.z provides the array index, otherwise it
1984 exhibits the same behavior as in the case for 1D texture arrays. The exact
1985 semantics of the source address are presented in the table below:
1987 +---------------------------+----+-----+-----+---------+
1988 | resource type | X | Y | Z | W |
1989 +===========================+====+=====+=====+=========+
1990 | ``PIPE_BUFFER`` | x | | | ignored |
1991 +---------------------------+----+-----+-----+---------+
1992 | ``PIPE_TEXTURE_1D`` | x | | | mpl |
1993 +---------------------------+----+-----+-----+---------+
1994 | ``PIPE_TEXTURE_2D`` | x | y | | mpl |
1995 +---------------------------+----+-----+-----+---------+
1996 | ``PIPE_TEXTURE_3D`` | x | y | z | mpl |
1997 +---------------------------+----+-----+-----+---------+
1998 | ``PIPE_TEXTURE_RECT`` | x | y | | mpl |
1999 +---------------------------+----+-----+-----+---------+
2000 | ``PIPE_TEXTURE_CUBE`` | not allowed as source |
2001 +---------------------------+----+-----+-----+---------+
2002 | ``PIPE_TEXTURE_1D_ARRAY`` | x | idx | | mpl |
2003 +---------------------------+----+-----+-----+---------+
2004 | ``PIPE_TEXTURE_2D_ARRAY`` | x | y | idx | mpl |
2005 +---------------------------+----+-----+-----+---------+
2007 Where 'mpl' is a mipmap level and 'idx' is the array index.
2009 .. opcode:: SAMPLE_I_MS
2011 Just like SAMPLE_I but allows fetch data from multi-sampled surfaces.
2013 Syntax: ``SAMPLE_I_MS dst, address, sampler_view, sample``
2015 .. opcode:: SAMPLE_B
2017 Just like the SAMPLE instruction with the exception that an additional bias
2018 is applied to the level of detail computed as part of the instruction
2021 Syntax: ``SAMPLE_B dst, address, sampler_view, sampler, lod_bias``
2023 Example: ``SAMPLE_B TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2].x``
2025 .. opcode:: SAMPLE_C
2027 Similar to the SAMPLE instruction but it performs a comparison filter. The
2028 operands to SAMPLE_C are identical to SAMPLE, except that there is an
2029 additional float32 operand, reference value, which must be a register with
2030 single-component, or a scalar literal. SAMPLE_C makes the hardware use the
2031 current samplers compare_func (in pipe_sampler_state) to compare reference
2032 value against the red component value for the surce resource at each texel
2033 that the currently configured texture filter covers based on the provided
2036 Syntax: ``SAMPLE_C dst, address, sampler_view.r, sampler, ref_value``
2038 Example: ``SAMPLE_C TEMP[0], TEMP[1], SVIEW[0].r, SAMP[0], TEMP[2].x``
2040 .. opcode:: SAMPLE_C_LZ
2042 Same as SAMPLE_C, but LOD is 0 and derivatives are ignored. The LZ stands
2045 Syntax: ``SAMPLE_C_LZ dst, address, sampler_view.r, sampler, ref_value``
2047 Example: ``SAMPLE_C_LZ TEMP[0], TEMP[1], SVIEW[0].r, SAMP[0], TEMP[2].x``
2050 .. opcode:: SAMPLE_D
2052 SAMPLE_D is identical to the SAMPLE opcode except that the derivatives for
2053 the source address in the x direction and the y direction are provided by
2056 Syntax: ``SAMPLE_D dst, address, sampler_view, sampler, der_x, der_y``
2058 Example: ``SAMPLE_D TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2], TEMP[3]``
2060 .. opcode:: SAMPLE_L
2062 SAMPLE_L is identical to the SAMPLE opcode except that the LOD is provided
2063 directly as a scalar value, representing no anisotropy.
2065 Syntax: ``SAMPLE_L dst, address, sampler_view, sampler, explicit_lod``
2067 Example: ``SAMPLE_L TEMP[0], TEMP[1], SVIEW[0], SAMP[0], TEMP[2].x``
2071 Gathers the four texels to be used in a bi-linear filtering operation and
2072 packs them into a single register. Only works with 2D, 2D array, cubemaps,
2073 and cubemaps arrays. For 2D textures, only the addressing modes of the
2074 sampler and the top level of any mip pyramid are used. Set W to zero. It
2075 behaves like the SAMPLE instruction, but a filtered sample is not
2076 generated. The four samples that contribute to filtering are placed into
2077 xyzw in counter-clockwise order, starting with the (u,v) texture coordinate
2078 delta at the following locations (-, +), (+, +), (+, -), (-, -), where the
2079 magnitude of the deltas are half a texel.
2082 .. opcode:: SVIEWINFO
2084 Query the dimensions of a given sampler view. dst receives width, height,
2085 depth or array size and number of mipmap levels as int4. The dst can have a
2086 writemask which will specify what info is the caller interested in.
2088 Syntax: ``SVIEWINFO dst, src_mip_level, sampler_view``
2090 Example: ``SVIEWINFO TEMP[0], TEMP[1].x, SVIEW[0]``
2092 src_mip_level is an unsigned integer scalar. If it's out of range then
2093 returns 0 for width, height and depth/array size but the total number of
2094 mipmap is still returned correctly for the given sampler view. The returned
2095 width, height and depth values are for the mipmap level selected by the
2096 src_mip_level and are in the number of texels. For 1d texture array width
2097 is in dst.x, array size is in dst.y and dst.z is 0. The number of mipmaps is
2098 still in dst.w. In contrast to d3d10 resinfo, there's no way in the tgsi
2099 instruction encoding to specify the return type (float/rcpfloat/uint), hence
2100 always using uint. Also, unlike the SAMPLE instructions, the swizzle on src1
2101 resinfo allowing swizzling dst values is ignored (due to the interaction
2102 with rcpfloat modifier which requires some swizzle handling in the state
2105 .. opcode:: SAMPLE_POS
2107 Query the position of a given sample. dst receives float4 (x, y, 0, 0)
2108 indicated where the sample is located. If the resource is not a multi-sample
2109 resource and not a render target, the result is 0.
2111 .. opcode:: SAMPLE_INFO
2113 dst receives number of samples in x. If the resource is not a multi-sample
2114 resource and not a render target, the result is 0.
2117 .. _resourceopcodes:
2119 Resource Access Opcodes
2120 ^^^^^^^^^^^^^^^^^^^^^^^
2122 .. opcode:: LOAD - Fetch data from a shader resource
2124 Syntax: ``LOAD dst, resource, address``
2126 Example: ``LOAD TEMP[0], RES[0], TEMP[1]``
2128 Using the provided integer address, LOAD fetches data
2129 from the specified buffer or texture without any
2132 The 'address' is specified as a vector of unsigned
2133 integers. If the 'address' is out of range the result
2136 Only the first mipmap level of a resource can be read
2137 from using this instruction.
2139 For 1D or 2D texture arrays, the array index is
2140 provided as an unsigned integer in address.y or
2141 address.z, respectively. address.yz are ignored for
2142 buffers and 1D textures. address.z is ignored for 1D
2143 texture arrays and 2D textures. address.w is always
2146 .. opcode:: STORE - Write data to a shader resource
2148 Syntax: ``STORE resource, address, src``
2150 Example: ``STORE RES[0], TEMP[0], TEMP[1]``
2152 Using the provided integer address, STORE writes data
2153 to the specified buffer or texture.
2155 The 'address' is specified as a vector of unsigned
2156 integers. If the 'address' is out of range the result
2159 Only the first mipmap level of a resource can be
2160 written to using this instruction.
2162 For 1D or 2D texture arrays, the array index is
2163 provided as an unsigned integer in address.y or
2164 address.z, respectively. address.yz are ignored for
2165 buffers and 1D textures. address.z is ignored for 1D
2166 texture arrays and 2D textures. address.w is always
2170 .. _threadsyncopcodes:
2172 Inter-thread synchronization opcodes
2173 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2175 These opcodes are intended for communication between threads running
2176 within the same compute grid. For now they're only valid in compute
2179 .. opcode:: MFENCE - Memory fence
2181 Syntax: ``MFENCE resource``
2183 Example: ``MFENCE RES[0]``
2185 This opcode forces strong ordering between any memory access
2186 operations that affect the specified resource. This means that
2187 previous loads and stores (and only those) will be performed and
2188 visible to other threads before the program execution continues.
2191 .. opcode:: LFENCE - Load memory fence
2193 Syntax: ``LFENCE resource``
2195 Example: ``LFENCE RES[0]``
2197 Similar to MFENCE, but it only affects the ordering of memory loads.
2200 .. opcode:: SFENCE - Store memory fence
2202 Syntax: ``SFENCE resource``
2204 Example: ``SFENCE RES[0]``
2206 Similar to MFENCE, but it only affects the ordering of memory stores.
2209 .. opcode:: BARRIER - Thread group barrier
2213 This opcode suspends the execution of the current thread until all
2214 the remaining threads in the working group reach the same point of
2215 the program. Results are unspecified if any of the remaining
2216 threads terminates or never reaches an executed BARRIER instruction.
2224 These opcodes provide atomic variants of some common arithmetic and
2225 logical operations. In this context atomicity means that another
2226 concurrent memory access operation that affects the same memory
2227 location is guaranteed to be performed strictly before or after the
2228 entire execution of the atomic operation.
2230 For the moment they're only valid in compute programs.
2232 .. opcode:: ATOMUADD - Atomic integer addition
2234 Syntax: ``ATOMUADD dst, resource, offset, src``
2236 Example: ``ATOMUADD TEMP[0], RES[0], TEMP[1], TEMP[2]``
2238 The following operation is performed atomically on each component:
2242 dst_i = resource[offset]_i
2244 resource[offset]_i = dst_i + src_i
2247 .. opcode:: ATOMXCHG - Atomic exchange
2249 Syntax: ``ATOMXCHG dst, resource, offset, src``
2251 Example: ``ATOMXCHG TEMP[0], RES[0], TEMP[1], TEMP[2]``
2253 The following operation is performed atomically on each component:
2257 dst_i = resource[offset]_i
2259 resource[offset]_i = src_i
2262 .. opcode:: ATOMCAS - Atomic compare-and-exchange
2264 Syntax: ``ATOMCAS dst, resource, offset, cmp, src``
2266 Example: ``ATOMCAS TEMP[0], RES[0], TEMP[1], TEMP[2], TEMP[3]``
2268 The following operation is performed atomically on each component:
2272 dst_i = resource[offset]_i
2274 resource[offset]_i = (dst_i == cmp_i ? src_i : dst_i)
2277 .. opcode:: ATOMAND - Atomic bitwise And
2279 Syntax: ``ATOMAND dst, resource, offset, src``
2281 Example: ``ATOMAND TEMP[0], RES[0], TEMP[1], TEMP[2]``
2283 The following operation is performed atomically on each component:
2287 dst_i = resource[offset]_i
2289 resource[offset]_i = dst_i \& src_i
2292 .. opcode:: ATOMOR - Atomic bitwise Or
2294 Syntax: ``ATOMOR dst, resource, offset, src``
2296 Example: ``ATOMOR TEMP[0], RES[0], TEMP[1], TEMP[2]``
2298 The following operation is performed atomically on each component:
2302 dst_i = resource[offset]_i
2304 resource[offset]_i = dst_i | src_i
2307 .. opcode:: ATOMXOR - Atomic bitwise Xor
2309 Syntax: ``ATOMXOR dst, resource, offset, src``
2311 Example: ``ATOMXOR TEMP[0], RES[0], TEMP[1], TEMP[2]``
2313 The following operation is performed atomically on each component:
2317 dst_i = resource[offset]_i
2319 resource[offset]_i = dst_i \oplus src_i
2322 .. opcode:: ATOMUMIN - Atomic unsigned minimum
2324 Syntax: ``ATOMUMIN dst, resource, offset, src``
2326 Example: ``ATOMUMIN TEMP[0], RES[0], TEMP[1], TEMP[2]``
2328 The following operation is performed atomically on each component:
2332 dst_i = resource[offset]_i
2334 resource[offset]_i = (dst_i < src_i ? dst_i : src_i)
2337 .. opcode:: ATOMUMAX - Atomic unsigned maximum
2339 Syntax: ``ATOMUMAX dst, resource, offset, src``
2341 Example: ``ATOMUMAX TEMP[0], RES[0], TEMP[1], TEMP[2]``
2343 The following operation is performed atomically on each component:
2347 dst_i = resource[offset]_i
2349 resource[offset]_i = (dst_i > src_i ? dst_i : src_i)
2352 .. opcode:: ATOMIMIN - Atomic signed minimum
2354 Syntax: ``ATOMIMIN dst, resource, offset, src``
2356 Example: ``ATOMIMIN TEMP[0], RES[0], TEMP[1], TEMP[2]``
2358 The following operation is performed atomically on each component:
2362 dst_i = resource[offset]_i
2364 resource[offset]_i = (dst_i < src_i ? dst_i : src_i)
2367 .. opcode:: ATOMIMAX - Atomic signed maximum
2369 Syntax: ``ATOMIMAX dst, resource, offset, src``
2371 Example: ``ATOMIMAX TEMP[0], RES[0], TEMP[1], TEMP[2]``
2373 The following operation is performed atomically on each component:
2377 dst_i = resource[offset]_i
2379 resource[offset]_i = (dst_i > src_i ? dst_i : src_i)
2383 Explanation of symbols used
2384 ------------------------------
2391 :math:`|x|` Absolute value of `x`.
2393 :math:`\lceil x \rceil` Ceiling of `x`.
2395 clamp(x,y,z) Clamp x between y and z.
2396 (x < y) ? y : (x > z) ? z : x
2398 :math:`\lfloor x\rfloor` Floor of `x`.
2400 :math:`\log_2{x}` Logarithm of `x`, base 2.
2402 max(x,y) Maximum of x and y.
2405 min(x,y) Minimum of x and y.
2408 partialx(x) Derivative of x relative to fragment's X.
2410 partialy(x) Derivative of x relative to fragment's Y.
2412 pop() Pop from stack.
2414 :math:`x^y` `x` to the power `y`.
2416 push(x) Push x on stack.
2420 trunc(x) Truncate x, i.e. drop the fraction bits.
2427 discard Discard fragment.
2431 target Label of target instruction.
2442 Declares a register that is will be referenced as an operand in Instruction
2445 File field contains register file that is being declared and is one
2448 UsageMask field specifies which of the register components can be accessed
2449 and is one of TGSI_WRITEMASK.
2451 The Local flag specifies that a given value isn't intended for
2452 subroutine parameter passing and, as a result, the implementation
2453 isn't required to give any guarantees of it being preserved across
2454 subroutine boundaries. As it's merely a compiler hint, the
2455 implementation is free to ignore it.
2457 If Dimension flag is set to 1, a Declaration Dimension token follows.
2459 If Semantic flag is set to 1, a Declaration Semantic token follows.
2461 If Interpolate flag is set to 1, a Declaration Interpolate token follows.
2463 If file is TGSI_FILE_RESOURCE, a Declaration Resource token follows.
2465 If Array flag is set to 1, a Declaration Array token follows.
2468 ^^^^^^^^^^^^^^^^^^^^^^^^
2470 Declarations can optional have an ArrayID attribute which can be referred by
2471 indirect addressing operands. An ArrayID of zero is reserved and treaded as
2472 if no ArrayID is specified.
2474 If an indirect addressing operand refers to a specific declaration by using
2475 an ArrayID only the registers in this declaration are guaranteed to be
2476 accessed, accessing any register outside this declaration results in undefined
2477 behavior. Note that for compatibility the effective index is zero-based and
2478 not relative to the specified declaration
2480 If no ArrayID is specified with an indirect addressing operand the whole
2481 register file might be accessed by this operand. This is strongly discouraged
2482 and will prevent packing of scalar/vec2 arrays and effective alias analysis.
2484 Declaration Semantic
2485 ^^^^^^^^^^^^^^^^^^^^^^^^
2487 Vertex and fragment shader input and output registers may be labeled
2488 with semantic information consisting of a name and index.
2490 Follows Declaration token if Semantic bit is set.
2492 Since its purpose is to link a shader with other stages of the pipeline,
2493 it is valid to follow only those Declaration tokens that declare a register
2494 either in INPUT or OUTPUT file.
2496 SemanticName field contains the semantic name of the register being declared.
2497 There is no default value.
2499 SemanticIndex is an optional subscript that can be used to distinguish
2500 different register declarations with the same semantic name. The default value
2503 The meanings of the individual semantic names are explained in the following
2506 TGSI_SEMANTIC_POSITION
2507 """"""""""""""""""""""
2509 For vertex shaders, TGSI_SEMANTIC_POSITION indicates the vertex shader
2510 output register which contains the homogeneous vertex position in the clip
2511 space coordinate system. After clipping, the X, Y and Z components of the
2512 vertex will be divided by the W value to get normalized device coordinates.
2514 For fragment shaders, TGSI_SEMANTIC_POSITION is used to indicate that
2515 fragment shader input contains the fragment's window position. The X
2516 component starts at zero and always increases from left to right.
2517 The Y component starts at zero and always increases but Y=0 may either
2518 indicate the top of the window or the bottom depending on the fragment
2519 coordinate origin convention (see TGSI_PROPERTY_FS_COORD_ORIGIN).
2520 The Z coordinate ranges from 0 to 1 to represent depth from the front
2521 to the back of the Z buffer. The W component contains the reciprocol
2522 of the interpolated vertex position W component.
2524 Fragment shaders may also declare an output register with
2525 TGSI_SEMANTIC_POSITION. Only the Z component is writable. This allows
2526 the fragment shader to change the fragment's Z position.
2533 For vertex shader outputs or fragment shader inputs/outputs, this
2534 label indicates that the resister contains an R,G,B,A color.
2536 Several shader inputs/outputs may contain colors so the semantic index
2537 is used to distinguish them. For example, color[0] may be the diffuse
2538 color while color[1] may be the specular color.
2540 This label is needed so that the flat/smooth shading can be applied
2541 to the right interpolants during rasterization.
2545 TGSI_SEMANTIC_BCOLOR
2546 """"""""""""""""""""
2548 Back-facing colors are only used for back-facing polygons, and are only valid
2549 in vertex shader outputs. After rasterization, all polygons are front-facing
2550 and COLOR and BCOLOR end up occupying the same slots in the fragment shader,
2551 so all BCOLORs effectively become regular COLORs in the fragment shader.
2557 Vertex shader inputs and outputs and fragment shader inputs may be
2558 labeled with TGSI_SEMANTIC_FOG to indicate that the register contains
2559 a fog coordinate. Typically, the fragment shader will use the fog coordinate
2560 to compute a fog blend factor which is used to blend the normal fragment color
2561 with a constant fog color. But fog coord really is just an ordinary vec4
2562 register like regular semantics.
2568 Vertex shader input and output registers may be labeled with
2569 TGIS_SEMANTIC_PSIZE to indicate that the register contains a point size
2570 in the form (S, 0, 0, 1). The point size controls the width or diameter
2571 of points for rasterization. This label cannot be used in fragment
2574 When using this semantic, be sure to set the appropriate state in the
2575 :ref:`rasterizer` first.
2578 TGSI_SEMANTIC_TEXCOORD
2579 """"""""""""""""""""""
2581 Only available if PIPE_CAP_TGSI_TEXCOORD is exposed !
2583 Vertex shader outputs and fragment shader inputs may be labeled with
2584 this semantic to make them replaceable by sprite coordinates via the
2585 sprite_coord_enable state in the :ref:`rasterizer`.
2586 The semantic index permitted with this semantic is limited to <= 7.
2588 If the driver does not support TEXCOORD, sprite coordinate replacement
2589 applies to inputs with the GENERIC semantic instead.
2591 The intended use case for this semantic is gl_TexCoord.
2594 TGSI_SEMANTIC_PCOORD
2595 """"""""""""""""""""
2597 Only available if PIPE_CAP_TGSI_TEXCOORD is exposed !
2599 Fragment shader inputs may be labeled with TGSI_SEMANTIC_PCOORD to indicate
2600 that the register contains sprite coordinates in the form (x, y, 0, 1), if
2601 the current primitive is a point and point sprites are enabled. Otherwise,
2602 the contents of the register are undefined.
2604 The intended use case for this semantic is gl_PointCoord.
2607 TGSI_SEMANTIC_GENERIC
2608 """""""""""""""""""""
2610 All vertex/fragment shader inputs/outputs not labeled with any other
2611 semantic label can be considered to be generic attributes. Typical
2612 uses of generic inputs/outputs are texcoords and user-defined values.
2615 TGSI_SEMANTIC_NORMAL
2616 """"""""""""""""""""
2618 Indicates that a vertex shader input is a normal vector. This is
2619 typically only used for legacy graphics APIs.
2625 This label applies to fragment shader inputs only and indicates that
2626 the register contains front/back-face information of the form (F, 0,
2627 0, 1). The first component will be positive when the fragment belongs
2628 to a front-facing polygon, and negative when the fragment belongs to a
2629 back-facing polygon.
2632 TGSI_SEMANTIC_EDGEFLAG
2633 """"""""""""""""""""""
2635 For vertex shaders, this sematic label indicates that an input or
2636 output is a boolean edge flag. The register layout is [F, x, x, x]
2637 where F is 0.0 or 1.0 and x = don't care. Normally, the vertex shader
2638 simply copies the edge flag input to the edgeflag output.
2640 Edge flags are used to control which lines or points are actually
2641 drawn when the polygon mode converts triangles/quads/polygons into
2645 TGSI_SEMANTIC_STENCIL
2646 """""""""""""""""""""
2648 For fragment shaders, this semantic label indicates that an output
2649 is a writable stencil reference value. Only the Y component is writable.
2650 This allows the fragment shader to change the fragments stencilref value.
2653 TGSI_SEMANTIC_VIEWPORT_INDEX
2654 """"""""""""""""""""""""""""
2656 For geometry shaders, this semantic label indicates that an output
2657 contains the index of the viewport (and scissor) to use.
2658 Only the X value is used.
2664 For geometry shaders, this semantic label indicates that an output
2665 contains the layer value to use for the color and depth/stencil surfaces.
2666 Only the X value is used. (Also known as rendertarget array index.)
2669 TGSI_SEMANTIC_CULLDIST
2670 """"""""""""""""""""""
2672 Used as distance to plane for performing application-defined culling
2673 of individual primitives against a plane. When components of vertex
2674 elements are given this label, these values are assumed to be a
2675 float32 signed distance to a plane. Primitives will be completely
2676 discarded if the plane distance for all of the vertices in the
2677 primitive are < 0. If a vertex has a cull distance of NaN, that
2678 vertex counts as "out" (as if its < 0);
2679 The limits on both clip and cull distances are bound
2680 by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_COUNT define which defines
2681 the maximum number of components that can be used to hold the
2682 distances and by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_ELEMENT_COUNT
2683 which specifies the maximum number of registers which can be
2684 annotated with those semantics.
2687 TGSI_SEMANTIC_CLIPDIST
2688 """"""""""""""""""""""
2690 When components of vertex elements are identified this way, these
2691 values are each assumed to be a float32 signed distance to a plane.
2692 Primitive setup only invokes rasterization on pixels for which
2693 the interpolated plane distances are >= 0. Multiple clip planes
2694 can be implemented simultaneously, by annotating multiple
2695 components of one or more vertex elements with the above specified
2696 semantic. The limits on both clip and cull distances are bound
2697 by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_COUNT define which defines
2698 the maximum number of components that can be used to hold the
2699 distances and by the PIPE_MAX_CLIP_OR_CULL_DISTANCE_ELEMENT_COUNT
2700 which specifies the maximum number of registers which can be
2701 annotated with those semantics.
2703 TGSI_SEMANTIC_SAMPLEID
2704 """"""""""""""""""""""
2706 For fragment shaders, this semantic label indicates that a system value
2707 contains the current sample id (i.e. gl_SampleID). Only the X value is used.
2709 TGSI_SEMANTIC_SAMPLEPOS
2710 """""""""""""""""""""""
2712 For fragment shaders, this semantic label indicates that a system value
2713 contains the current sample's position (i.e. gl_SamplePosition). Only the X
2714 and Y values are used.
2716 TGSI_SEMANTIC_SAMPLEMASK
2717 """"""""""""""""""""""""
2719 For fragment shaders, this semantic label indicates that an output contains
2720 the sample mask used to disable further sample processing
2721 (i.e. gl_SampleMask). Only the X value is used, up to 32x MS.
2723 TGSI_SEMANTIC_INVOCATIONID
2724 """"""""""""""""""""""""""
2726 For geometry shaders, this semantic label indicates that a system value
2727 contains the current invocation id (i.e. gl_InvocationID). Only the X value is
2730 Declaration Interpolate
2731 ^^^^^^^^^^^^^^^^^^^^^^^
2733 This token is only valid for fragment shader INPUT declarations.
2735 The Interpolate field specifes the way input is being interpolated by
2736 the rasteriser and is one of TGSI_INTERPOLATE_*.
2738 The Location field specifies the location inside the pixel that the
2739 interpolation should be done at, one of ``TGSI_INTERPOLATE_LOC_*``. Note that
2740 when per-sample shading is enabled, the implementation may choose to
2741 interpolate at the sample irrespective of the Location field.
2743 The CylindricalWrap bitfield specifies which register components
2744 should be subject to cylindrical wrapping when interpolating by the
2745 rasteriser. If TGSI_CYLINDRICAL_WRAP_X is set to 1, the X component
2746 should be interpolated according to cylindrical wrapping rules.
2749 Declaration Sampler View
2750 ^^^^^^^^^^^^^^^^^^^^^^^^
2752 Follows Declaration token if file is TGSI_FILE_SAMPLER_VIEW.
2754 DCL SVIEW[#], resource, type(s)
2756 Declares a shader input sampler view and assigns it to a SVIEW[#]
2759 resource can be one of BUFFER, 1D, 2D, 3D, 1DArray and 2DArray.
2761 type must be 1 or 4 entries (if specifying on a per-component
2762 level) out of UNORM, SNORM, SINT, UINT and FLOAT.
2765 Declaration Resource
2766 ^^^^^^^^^^^^^^^^^^^^
2768 Follows Declaration token if file is TGSI_FILE_RESOURCE.
2770 DCL RES[#], resource [, WR] [, RAW]
2772 Declares a shader input resource and assigns it to a RES[#]
2775 resource can be one of BUFFER, 1D, 2D, 3D, CUBE, 1DArray and
2778 If the RAW keyword is not specified, the texture data will be
2779 subject to conversion, swizzling and scaling as required to yield
2780 the specified data type from the physical data format of the bound
2783 If the RAW keyword is specified, no channel conversion will be
2784 performed: the values read for each of the channels (X,Y,Z,W) will
2785 correspond to consecutive words in the same order and format
2786 they're found in memory. No element-to-address conversion will be
2787 performed either: the value of the provided X coordinate will be
2788 interpreted in byte units instead of texel units. The result of
2789 accessing a misaligned address is undefined.
2791 Usage of the STORE opcode is only allowed if the WR (writable) flag
2796 ^^^^^^^^^^^^^^^^^^^^^^^^
2798 Properties are general directives that apply to the whole TGSI program.
2803 Specifies the fragment shader TGSI_SEMANTIC_POSITION coordinate origin.
2804 The default value is UPPER_LEFT.
2806 If UPPER_LEFT, the position will be (0,0) at the upper left corner and
2807 increase downward and rightward.
2808 If LOWER_LEFT, the position will be (0,0) at the lower left corner and
2809 increase upward and rightward.
2811 OpenGL defaults to LOWER_LEFT, and is configurable with the
2812 GL_ARB_fragment_coord_conventions extension.
2814 DirectX 9/10 use UPPER_LEFT.
2816 FS_COORD_PIXEL_CENTER
2817 """""""""""""""""""""
2819 Specifies the fragment shader TGSI_SEMANTIC_POSITION pixel center convention.
2820 The default value is HALF_INTEGER.
2822 If HALF_INTEGER, the fractionary part of the position will be 0.5
2823 If INTEGER, the fractionary part of the position will be 0.0
2825 Note that this does not affect the set of fragments generated by
2826 rasterization, which is instead controlled by half_pixel_center in the
2829 OpenGL defaults to HALF_INTEGER, and is configurable with the
2830 GL_ARB_fragment_coord_conventions extension.
2832 DirectX 9 uses INTEGER.
2833 DirectX 10 uses HALF_INTEGER.
2835 FS_COLOR0_WRITES_ALL_CBUFS
2836 """"""""""""""""""""""""""
2837 Specifies that writes to the fragment shader color 0 are replicated to all
2838 bound cbufs. This facilitates OpenGL's fragColor output vs fragData[0] where
2839 fragData is directed to a single color buffer, but fragColor is broadcast.
2842 """"""""""""""""""""""""""
2843 If this property is set on the program bound to the shader stage before the
2844 fragment shader, user clip planes should have no effect (be disabled) even if
2845 that shader does not write to any clip distance outputs and the rasterizer's
2846 clip_plane_enable is non-zero.
2847 This property is only supported by drivers that also support shader clip
2849 This is useful for APIs that don't have UCPs and where clip distances written
2850 by a shader cannot be disabled.
2855 Specifies the number of times a geometry shader should be executed for each
2856 input primitive. Each invocation will have a different
2857 TGSI_SEMANTIC_INVOCATIONID system value set. If not specified, assumed to
2860 VS_WINDOW_SPACE_POSITION
2861 """"""""""""""""""""""""""
2862 If this property is set on the vertex shader, the TGSI_SEMANTIC_POSITION output
2863 is assumed to contain window space coordinates.
2864 Division of X,Y,Z by W and the viewport transformation are disabled, and 1/W is
2865 directly taken from the 4-th component of the shader output.
2866 Naturally, clipping is not performed on window coordinates either.
2867 The effect of this property is undefined if a geometry or tessellation shader
2870 Texture Sampling and Texture Formats
2871 ------------------------------------
2873 This table shows how texture image components are returned as (x,y,z,w) tuples
2874 by TGSI texture instructions, such as :opcode:`TEX`, :opcode:`TXD`, and
2875 :opcode:`TXP`. For reference, OpenGL and Direct3D conventions are shown as
2878 +--------------------+--------------+--------------------+--------------+
2879 | Texture Components | Gallium | OpenGL | Direct3D 9 |
2880 +====================+==============+====================+==============+
2881 | R | (r, 0, 0, 1) | (r, 0, 0, 1) | (r, 1, 1, 1) |
2882 +--------------------+--------------+--------------------+--------------+
2883 | RG | (r, g, 0, 1) | (r, g, 0, 1) | (r, g, 1, 1) |
2884 +--------------------+--------------+--------------------+--------------+
2885 | RGB | (r, g, b, 1) | (r, g, b, 1) | (r, g, b, 1) |
2886 +--------------------+--------------+--------------------+--------------+
2887 | RGBA | (r, g, b, a) | (r, g, b, a) | (r, g, b, a) |
2888 +--------------------+--------------+--------------------+--------------+
2889 | A | (0, 0, 0, a) | (0, 0, 0, a) | (0, 0, 0, a) |
2890 +--------------------+--------------+--------------------+--------------+
2891 | L | (l, l, l, 1) | (l, l, l, 1) | (l, l, l, 1) |
2892 +--------------------+--------------+--------------------+--------------+
2893 | LA | (l, l, l, a) | (l, l, l, a) | (l, l, l, a) |
2894 +--------------------+--------------+--------------------+--------------+
2895 | I | (i, i, i, i) | (i, i, i, i) | N/A |
2896 +--------------------+--------------+--------------------+--------------+
2897 | UV | XXX TBD | (0, 0, 0, 1) | (u, v, 1, 1) |
2898 | | | [#envmap-bumpmap]_ | |
2899 +--------------------+--------------+--------------------+--------------+
2900 | Z | XXX TBD | (z, z, z, 1) | (0, z, 0, 1) |
2901 | | | [#depth-tex-mode]_ | |
2902 +--------------------+--------------+--------------------+--------------+
2903 | S | (s, s, s, s) | unknown | unknown |
2904 +--------------------+--------------+--------------------+--------------+
2906 .. [#envmap-bumpmap] http://www.opengl.org/registry/specs/ATI/envmap_bumpmap.txt
2907 .. [#depth-tex-mode] the default is (z, z, z, 1) but may also be (0, 0, 0, z)
2908 or (z, z, z, z) depending on the value of GL_DEPTH_TEXTURE_MODE.