From 2dd94f462b0069fc3a20c9a93a9cfe97dd079837 Mon Sep 17 00:00:00 2001 From: Jose Maria Casanova Crespo Date: Mon, 26 Feb 2018 20:28:34 +0100 Subject: [PATCH] i965/fs: shuffle_32bit_load_result_to_16bit_data now skips components This helper used to load 16bit components from 32-bits read now allows skipping components with the new parameter first_component. The semantics now skip components until we reach the first_component, and then reads the number of components passed to the function. All previous uses of the helper are updated to use 0 as first_component. This will allow read 16-bit components when the first one is not aligned 32-bit. Enabling more usages of untyped_reads with 16-bit types. v2: (Jason Ektrand) Change parameters order to first_component, num_components Reviewed-by: Jason Ekstrand --- src/intel/compiler/brw_fs.cpp | 2 +- src/intel/compiler/brw_fs.h | 1 + src/intel/compiler/brw_fs_nir.cpp | 6 ++++-- 3 files changed, 6 insertions(+), 3 deletions(-) diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp index 113f62c46c7..244c6cda03a 100644 --- a/src/intel/compiler/brw_fs.cpp +++ b/src/intel/compiler/brw_fs.cpp @@ -194,7 +194,7 @@ fs_visitor::VARYING_PULL_CONSTANT_LOAD(const fs_builder &bld, fs_reg dw = offset(vec4_result, bld, (const_offset & 0xf) / 4); switch (type_sz(dst.type)) { case 2: - shuffle_32bit_load_result_to_16bit_data(bld, dst, dw, 1); + shuffle_32bit_load_result_to_16bit_data(bld, dst, dw, 0, 1); bld.MOV(dst, subscript(dw, dst.type, (const_offset / 2) & 1)); break; case 4: diff --git a/src/intel/compiler/brw_fs.h b/src/intel/compiler/brw_fs.h index 76ad76e08b7..38e9991df73 100644 --- a/src/intel/compiler/brw_fs.h +++ b/src/intel/compiler/brw_fs.h @@ -505,6 +505,7 @@ fs_reg shuffle_64bit_data_for_32bit_write(const brw::fs_builder &bld, void shuffle_32bit_load_result_to_16bit_data(const brw::fs_builder &bld, const fs_reg &dst, const fs_reg &src, + uint32_t first_component, uint32_t components); void shuffle_16bit_data_for_32bit_write(const brw::fs_builder &bld, diff --git a/src/intel/compiler/brw_fs_nir.cpp b/src/intel/compiler/brw_fs_nir.cpp index d8300589a5e..0d1ab5b01c1 100644 --- a/src/intel/compiler/brw_fs_nir.cpp +++ b/src/intel/compiler/brw_fs_nir.cpp @@ -2316,7 +2316,7 @@ do_untyped_vector_read(const fs_builder &bld, shuffle_32bit_load_result_to_16bit_data(bld, retype(dest, BRW_REGISTER_TYPE_W), retype(read_result, BRW_REGISTER_TYPE_D), - num_components); + 0, num_components); } else { assert(num_components == 1); /* scalar 16-bit are read using one byte_scattered_read message */ @@ -4912,6 +4912,7 @@ void shuffle_32bit_load_result_to_16bit_data(const fs_builder &bld, const fs_reg &dst, const fs_reg &src, + uint32_t first_component, uint32_t components) { assert(type_sz(src.type) == 4); @@ -4926,7 +4927,8 @@ shuffle_32bit_load_result_to_16bit_data(const fs_builder &bld, for (unsigned i = 0; i < components; i++) { const fs_reg component_i = - subscript(offset(src, bld, i / 2), dst.type, i % 2); + subscript(offset(src, bld, (first_component + i) / 2), dst.type, + (first_component + i) % 2); bld.MOV(offset(tmp, bld, i % 2), component_i); -- 2.30.2