gallium: add TGSI opcodes TEX_LZ and TXF_LZ

for better code generation in radeonsi

4 files changed, 39 insertions, 5 deletions

diff --git a/src/gallium/auxiliary/tgsi/tgsi_info.c b/src/gallium/auxiliary/tgsi/tgsi_info.c
index ec1726595d..ae6a499324 100644
--- a/src/gallium/auxiliary/tgsi/tgsi_info.c
+++ b/src/gallium/auxiliary/tgsi/tgsi_info.c
@@ -54,7 +54,7 @@ static const struct tgsi_opcode_info opcode_info[TGSI_OPCODE_LAST] =
    { 1, 2, 0, 0, 0, 0, 0, COMP, "SLT", TGSI_OPCODE_SLT },
    { 1, 2, 0, 0, 0, 0, 0, COMP, "SGE", TGSI_OPCODE_SGE },
    { 1, 3, 0, 0, 0, 0, 0, COMP, "MAD", TGSI_OPCODE_MAD },
-   { 1, 2, 0, 0, 0, 0, 0, COMP, "", 17 }, /* removed */
+   { 1, 2, 1, 0, 0, 0, 0, OTHR, "TEX_LZ", TGSI_OPCODE_TEX_LZ },
    { 1, 3, 0, 0, 0, 0, 0, COMP, "LRP", TGSI_OPCODE_LRP },
    { 1, 3, 0, 0, 0, 0, 0, COMP, "FMA", TGSI_OPCODE_FMA },
    { 1, 1, 0, 0, 0, 0, 0, REPL, "SQRT", TGSI_OPCODE_SQRT },
@@ -62,7 +62,7 @@ static const struct tgsi_opcode_info opcode_info[TGSI_OPCODE_LAST] =
    { 1, 1, 0, 0, 0, 0, 0, COMP, "F2U64", TGSI_OPCODE_F2U64 },
    { 1, 1, 0, 0, 0, 0, 0, COMP, "F2I64", TGSI_OPCODE_F2I64 },
    { 1, 1, 0, 0, 0, 0, 0, COMP, "FRC", TGSI_OPCODE_FRC },
-   { 1, 3, 0, 0, 0, 0, 0, COMP, "", 25 }, /* removed */
+   { 1, 2, 1, 0, 0, 0, 0, OTHR, "TXF_LZ", TGSI_OPCODE_TXF_LZ },
    { 1, 1, 0, 0, 0, 0, 0, COMP, "FLR", TGSI_OPCODE_FLR },
    { 1, 1, 0, 0, 0, 0, 0, COMP, "ROUND", TGSI_OPCODE_ROUND },
    { 1, 1, 0, 0, 0, 0, 0, REPL, "EX2", TGSI_OPCODE_EX2 },
@@ -478,6 +478,7 @@ tgsi_opcode_infer_src_type( uint opcode )
    switch (opcode) {
    case TGSI_OPCODE_UIF:
    case TGSI_OPCODE_TXF:
+   case TGSI_OPCODE_TXF_LZ:
    case TGSI_OPCODE_BREAKC:
    case TGSI_OPCODE_U2F:
    case TGSI_OPCODE_U2D:
diff --git a/src/gallium/auxiliary/tgsi/tgsi_scan.c b/src/gallium/auxiliary/tgsi/tgsi_scan.c
index 99799fa16e..bf614db806 100644
--- a/src/gallium/auxiliary/tgsi/tgsi_scan.c
+++ b/src/gallium/auxiliary/tgsi/tgsi_scan.c
@@ -87,6 +87,8 @@ computes_derivative(unsigned opcode)
       return opcode != TGSI_OPCODE_TG4 &&
              opcode != TGSI_OPCODE_TXD &&
              opcode != TGSI_OPCODE_TXF &&
+             opcode != TGSI_OPCODE_TXF_LZ &&
+             opcode != TGSI_OPCODE_TEX_LZ &&
              opcode != TGSI_OPCODE_TXL &&
              opcode != TGSI_OPCODE_TXL2 &&
              opcode != TGSI_OPCODE_TXQ &&
diff --git a/src/gallium/docs/source/tgsi.rst b/src/gallium/docs/source/tgsi.rst
index 32ec4ef2ac..18b42fba74 100644
--- a/src/gallium/docs/source/tgsi.rst
+++ b/src/gallium/docs/source/tgsi.rst
@@ -755,6 +755,29 @@ This instruction replicates its result.
   dst = src0.x \times src1.x + src0.y \times src1.y
 
 
+.. opcode:: TEX_LZ - Texture Lookup With LOD = 0
+
+  This is the same as TXL with LOD = 0. Like every texture opcode, it obeys
+  pipe_sampler_view::u.tex.first_level and pipe_sampler_state::min_lod.
+  There is no way to override those two in shaders.
+
+.. math::
+
+  coord.x = src0.x
+
+  coord.y = src0.y
+
+  coord.z = src0.z
+
+  coord.w = none
+
+  lod = 0
+
+  unit = src1
+
+  dst = texture\_sample(unit, coord, lod)
+
+
 .. opcode:: TXL - Texture Lookup With explicit LOD
 
   for cube map array textures, the explicit lod value
@@ -918,10 +941,18 @@ XXX doesn't look like most of the opcodes really belong here.
   four-component signed integer vector used to identify the single texel
   accessed. 3 components + level.  Just like texture instructions, an optional
   offset vector is provided, which is subject to various driver restrictions
-  (regarding range, source of offsets).
+  (regarding range, source of offsets). This instruction ignores the sampler
+  state.
+
   TXF(uint_vec coord, int_vec offset).
 
 
+.. opcode:: TXF_LZ - Texel Fetch
+
+  This is the same as TXF with level = 0. Like TXF, it obeys
+  pipe_sampler_view::u.tex.first_level.
+
+
 .. opcode:: TXQ - Texture Size Query
 
   As per NV_gpu_program4, retrieve the dimensions of the texture depending on
diff --git a/src/gallium/include/pipe/p_shader_tokens.h b/src/gallium/include/pipe/p_shader_tokens.h
index 11186047a6..6a3fb9876b 100644
--- a/src/gallium/include/pipe/p_shader_tokens.h
+++ b/src/gallium/include/pipe/p_shader_tokens.h
@@ -346,7 +346,7 @@ struct tgsi_property_data {
 #define TGSI_OPCODE_SLT                 14
 #define TGSI_OPCODE_SGE                 15
 #define TGSI_OPCODE_MAD                 16
-/* gap */
+#define TGSI_OPCODE_TEX_LZ              17
 #define TGSI_OPCODE_LRP                 18
 #define TGSI_OPCODE_FMA                 19
 #define TGSI_OPCODE_SQRT                20
@@ -354,7 +354,7 @@ struct tgsi_property_data {
 #define TGSI_OPCODE_F2U64               22
 #define TGSI_OPCODE_F2I64               23
 #define TGSI_OPCODE_FRC                 24
-/* gap */
+#define TGSI_OPCODE_TXF_LZ              25
 #define TGSI_OPCODE_FLR                 26
 #define TGSI_OPCODE_ROUND               27
 #define TGSI_OPCODE_EX2                 28