github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-09 15:08:02 -05:00
Code Issues Packages Projects Releases Wiki Activity

remove more stale stuff (#13765)

Browse Source 
* remove more stale stuff

* remove disassemblers/adreno

* stale
This commit is contained in:
George Hotz
2025-12-19 17:14:56 -04:00
committed by GitHub
parent 744af193f0
commit 45c459848d
19 changed files with 1 additions and 7965 deletions
Download Patch File Download Diff File Expand all files Collapse all files

199
extra/datasets/coco.py
View File

@@ -1,199 +0,0 @@
import json
import pathlib
import zipfile
import numpy as np
from tinygrad.helpers import fetch
import pycocotools._mask as _mask
from examples.mask_rcnn import Masker
from pycocotools.coco import COCO
from pycocotools.cocoeval import COCOeval
iou = _mask.iou
merge = _mask.merge
frPyObjects = _mask.frPyObjects
BASEDIR = pathlib.Path(__file__).parent / "COCO"
BASEDIR.mkdir(exist_ok=True)
def create_dict(key_row, val_row, rows): return {row[key_row]:row[val_row] for row in rows}
if not pathlib.Path(BASEDIR/'val2017').is_dir():
fn = fetch('http://images.cocodataset.org/zips/val2017.zip')
with zipfile.ZipFile(fn, 'r') as zip_ref:
zip_ref.extractall(BASEDIR)
fn.unlink()
if not pathlib.Path(BASEDIR/'annotations').is_dir():
fn = fetch('http://images.cocodataset.org/annotations/annotations_trainval2017.zip')
with zipfile.ZipFile(fn, 'r') as zip_ref:
zip_ref.extractall(BASEDIR)
fn.unlink()
with open(BASEDIR/'annotations/instances_val2017.json', 'r') as f:
annotations_raw = json.loads(f.read())
images = annotations_raw['images']
categories = annotations_raw['categories']
annotations = annotations_raw['annotations']
file_name_to_id = create_dict('file_name', 'id', images)
id_to_width = create_dict('id', 'width', images)
id_to_height = create_dict('id', 'height', images)
json_category_id_to_contiguous_id = {v['id']: i + 1 for i, v in enumerate(categories)}
contiguous_category_id_to_json_id = {v:k for k,v in json_category_id_to_contiguous_id.items()}
def encode(bimask):
if len(bimask.shape) == 3:
return _mask.encode(bimask)
elif len(bimask.shape) == 2:
h, w = bimask.shape
return _mask.encode(bimask.reshape((h, w, 1), order='F'))[0]
def decode(rleObjs):
if type(rleObjs) == list:
return _mask.decode(rleObjs)
else:
return _mask.decode([rleObjs])[:,:,0]
def area(rleObjs):
if type(rleObjs) == list:
return _mask.area(rleObjs)
else:
return _mask.area([rleObjs])[0]
def toBbox(rleObjs):
if type(rleObjs) == list:
return _mask.toBbox(rleObjs)
else:
return _mask.toBbox([rleObjs])[0]
def convert_prediction_to_coco_bbox(file_name, prediction):
coco_results = []
try:
original_id = file_name_to_id[file_name]
if len(prediction) == 0:
return coco_results
image_width = id_to_width[original_id]
image_height = id_to_height[original_id]
prediction = prediction.resize((image_width, image_height))
prediction = prediction.convert("xywh")
boxes = prediction.bbox.numpy().tolist()
scores = prediction.get_field("scores").numpy().tolist()
labels = prediction.get_field("labels").numpy().tolist()
mapped_labels = [contiguous_category_id_to_json_id[int(i)] for i in labels]
coco_results.extend(
[
{
"image_id": original_id,
"category_id": mapped_labels[k],
"bbox": box,
"score": scores[k],
}
for k, box in enumerate(boxes)
]
)
except Exception as e:
print(file_name, e)
return coco_results
masker = Masker(threshold=0.5, padding=1)
def convert_prediction_to_coco_mask(file_name, prediction):
coco_results = []
try:
original_id = file_name_to_id[file_name]
if len(prediction) == 0:
return coco_results
image_width = id_to_width[original_id]
image_height = id_to_height[original_id]
prediction = prediction.resize((image_width, image_height))
masks = prediction.get_field("mask")
scores = prediction.get_field("scores").numpy().tolist()
labels = prediction.get_field("labels").numpy().tolist()
masks = masker([masks], [prediction])[0].numpy()
rles = [
encode(np.array(mask[0, :, :, np.newaxis], order="F"))[0]
for mask in masks
]
for rle in rles:
rle["counts"] = rle["counts"].decode("utf-8")
mapped_labels = [contiguous_category_id_to_json_id[int(i)] for i in labels]
coco_results.extend(
[
{
"image_id": original_id,
"category_id": mapped_labels[k],
"segmentation": rle,
"score": scores[k],
}
for k, rle in enumerate(rles)
]
)
except Exception as e:
print(file_name, e)
return coco_results
def accumulate_predictions_for_coco(coco_results, json_result_file, rm=False):
path = pathlib.Path(json_result_file)
if rm and path.exists(): path.unlink()
with open(path, "a") as f:
for s in coco_results:
f.write(json.dumps(s))
f.write('\n')
def remove_dup(l):
seen = set()
seen_add = seen.add
return [x for x in l if not (x in seen or seen_add(x))]
class NpEncoder(json.JSONEncoder):
def default(self, obj):
if isinstance(obj, np.integer):
return int(obj)
if isinstance(obj, np.floating):
return float(obj)
if isinstance(obj, np.ndarray):
return obj.tolist()
return super(NpEncoder, self).default(obj)
def evaluate_predictions_on_coco(json_result_file, iou_type="bbox"):
coco_results = []
with open(json_result_file, "r") as f:
for line in f:
coco_results.append(json.loads(line))
coco_gt = COCO(str(BASEDIR/'annotations/instances_val2017.json'))
set_of_json = remove_dup([json.dumps(d, cls=NpEncoder) for d in coco_results])
unique_list = [json.loads(s) for s in set_of_json]
with open(f'{json_result_file}.flattend', "w") as f:
json.dump(unique_list, f)
coco_dt = coco_gt.loadRes(str(f'{json_result_file}.flattend'))
coco_eval = COCOeval(coco_gt, coco_dt, iou_type)
coco_eval.evaluate()
coco_eval.accumulate()
coco_eval.summarize()
return coco_eval
def iterate(files, bs=1):
batch = []
for file in files:
batch.append(file)
if len(batch) >= bs: yield batch; batch = []
if len(batch) > 0: yield batch; batch = []

1
extra/disassemblers/adreno/.gitignore vendored
View File

@@ -1 +0,0 @@
disasm.so

5
extra/disassemblers/adreno/README
View File

@@ -1,5 +0,0 @@
From the Freedreno project
https://gallium.readthedocs.io/en/latest/gallium/drivers/freedreno.html
In Mesa3D, so licensed MIT.

1431
extra/disassemblers/adreno/disasm-a3xx.c
View File

File diff suppressed because it is too large Load Diff

1119
extra/disassemblers/adreno/instr-a3xx.h
View File

File diff suppressed because it is too large Load Diff

1757
extra/disassemblers/adreno/ir3.h
View File

File diff suppressed because it is too large Load Diff

906
extra/disassemblers/adreno/shader_enums.h
View File

@@ -1,906 +0,0 @@
/*
* Mesa 3-D graphics library
*
* Copyright (C) 1999-2008 Brian Paul All Rights Reserved.
* Copyright (C) 2009 VMware, Inc. All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*/
#ifndef SHADER_ENUMS_H
#define SHADER_ENUMS_H
#include <stdbool.h>
/* Project-wide (GL and Vulkan) maximum. */
#define MAX_DRAW_BUFFERS 8
#ifdef __cplusplus
extern "C" {
#endif
/**
* Shader stages.
*
* The order must match how shaders are ordered in the pipeline.
* The GLSL linker assumes that if i<j, then the j-th shader is
* executed later than the i-th shader.
*/
typedef enum
{
MESA_SHADER_NONE = -1,
MESA_SHADER_VERTEX = 0,
MESA_SHADER_TESS_CTRL = 1,
MESA_SHADER_TESS_EVAL = 2,
MESA_SHADER_GEOMETRY = 3,
MESA_SHADER_FRAGMENT = 4,
MESA_SHADER_COMPUTE = 5,
/* must be last so it doesn't affect the GL pipeline */
MESA_SHADER_KERNEL = 6,
} gl_shader_stage;
static inline bool
gl_shader_stage_is_compute(gl_shader_stage stage)
{
return stage == MESA_SHADER_COMPUTE || stage == MESA_SHADER_KERNEL;
}
/**
* Number of STATE_* values we need to address any GL state.
* Used to dimension arrays.
*/
#define STATE_LENGTH 5
typedef short gl_state_index16; /* see enum gl_state_index */
const char *gl_shader_stage_name(gl_shader_stage stage);
/**
* Translate a gl_shader_stage to a short shader stage name for debug
* printouts and error messages.
*/
const char *_mesa_shader_stage_to_string(unsigned stage);
/**
* Translate a gl_shader_stage to a shader stage abbreviation (VS, GS, FS)
* for debug printouts and error messages.
*/
const char *_mesa_shader_stage_to_abbrev(unsigned stage);
/**
* GL related stages (not including CL)
*/
#define MESA_SHADER_STAGES (MESA_SHADER_COMPUTE + 1)
/**
* All stages
*/
#define MESA_ALL_SHADER_STAGES (MESA_SHADER_KERNEL + 1)
/**
* Indexes for vertex program attributes.
* GL_NV_vertex_program aliases generic attributes over the conventional
* attributes. In GL_ARB_vertex_program shader the aliasing is optional.
* In GL_ARB_vertex_shader / OpenGL 2.0 the aliasing is disallowed (the
* generic attributes are distinct/separate).
*/
typedef enum
{
VERT_ATTRIB_POS,
VERT_ATTRIB_NORMAL,
VERT_ATTRIB_COLOR0,
VERT_ATTRIB_COLOR1,
VERT_ATTRIB_FOG,
VERT_ATTRIB_COLOR_INDEX,
VERT_ATTRIB_EDGEFLAG,
VERT_ATTRIB_TEX0,
VERT_ATTRIB_TEX1,
VERT_ATTRIB_TEX2,
VERT_ATTRIB_TEX3,
VERT_ATTRIB_TEX4,
VERT_ATTRIB_TEX5,
VERT_ATTRIB_TEX6,
VERT_ATTRIB_TEX7,
VERT_ATTRIB_POINT_SIZE,
VERT_ATTRIB_GENERIC0,
VERT_ATTRIB_GENERIC1,
VERT_ATTRIB_GENERIC2,
VERT_ATTRIB_GENERIC3,
VERT_ATTRIB_GENERIC4,
VERT_ATTRIB_GENERIC5,
VERT_ATTRIB_GENERIC6,
VERT_ATTRIB_GENERIC7,
VERT_ATTRIB_GENERIC8,
VERT_ATTRIB_GENERIC9,
VERT_ATTRIB_GENERIC10,
VERT_ATTRIB_GENERIC11,
VERT_ATTRIB_GENERIC12,
VERT_ATTRIB_GENERIC13,
VERT_ATTRIB_GENERIC14,
VERT_ATTRIB_GENERIC15,
VERT_ATTRIB_MAX
} gl_vert_attrib;
const char *gl_vert_attrib_name(gl_vert_attrib attrib);
/**
* Symbolic constats to help iterating over
* specific blocks of vertex attributes.
*
* VERT_ATTRIB_FF
* includes all fixed function attributes as well as
* the aliased GL_NV_vertex_program shader attributes.
* VERT_ATTRIB_TEX
* include the classic texture coordinate attributes.
* Is a subset of VERT_ATTRIB_FF.
* VERT_ATTRIB_GENERIC
* include the OpenGL 2.0+ GLSL generic shader attributes.
* These alias the generic GL_ARB_vertex_shader attributes.
* VERT_ATTRIB_MAT
* include the generic shader attributes used to alias
* varying material values for the TNL shader programs.
* They are located at the end of the generic attribute
* block not to overlap with the generic 0 attribute.
*/
#define VERT_ATTRIB_FF(i) (VERT_ATTRIB_POS + (i))
#define VERT_ATTRIB_FF_MAX VERT_ATTRIB_GENERIC0
#define VERT_ATTRIB_TEX(i) (VERT_ATTRIB_TEX0 + (i))
#define VERT_ATTRIB_TEX_MAX MAX_TEXTURE_COORD_UNITS
#define VERT_ATTRIB_GENERIC(i) (VERT_ATTRIB_GENERIC0 + (i))
#define VERT_ATTRIB_GENERIC_MAX MAX_VERTEX_GENERIC_ATTRIBS
#define VERT_ATTRIB_MAT0 \
(VERT_ATTRIB_GENERIC_MAX - VERT_ATTRIB_MAT_MAX)
#define VERT_ATTRIB_MAT(i) \
VERT_ATTRIB_GENERIC((i) + VERT_ATTRIB_MAT0)
#define VERT_ATTRIB_MAT_MAX MAT_ATTRIB_MAX
/**
* Bitflags for vertex attributes.
* These are used in bitfields in many places.
*/
/*@{*/
#define VERT_BIT_POS BITFIELD_BIT(VERT_ATTRIB_POS)
#define VERT_BIT_NORMAL BITFIELD_BIT(VERT_ATTRIB_NORMAL)
#define VERT_BIT_COLOR0 BITFIELD_BIT(VERT_ATTRIB_COLOR0)
#define VERT_BIT_COLOR1 BITFIELD_BIT(VERT_ATTRIB_COLOR1)
#define VERT_BIT_FOG BITFIELD_BIT(VERT_ATTRIB_FOG)
#define VERT_BIT_COLOR_INDEX BITFIELD_BIT(VERT_ATTRIB_COLOR_INDEX)
#define VERT_BIT_EDGEFLAG BITFIELD_BIT(VERT_ATTRIB_EDGEFLAG)
#define VERT_BIT_TEX0 BITFIELD_BIT(VERT_ATTRIB_TEX0)
#define VERT_BIT_TEX1 BITFIELD_BIT(VERT_ATTRIB_TEX1)
#define VERT_BIT_TEX2 BITFIELD_BIT(VERT_ATTRIB_TEX2)
#define VERT_BIT_TEX3 BITFIELD_BIT(VERT_ATTRIB_TEX3)
#define VERT_BIT_TEX4 BITFIELD_BIT(VERT_ATTRIB_TEX4)
#define VERT_BIT_TEX5 BITFIELD_BIT(VERT_ATTRIB_TEX5)
#define VERT_BIT_TEX6 BITFIELD_BIT(VERT_ATTRIB_TEX6)
#define VERT_BIT_TEX7 BITFIELD_BIT(VERT_ATTRIB_TEX7)
#define VERT_BIT_POINT_SIZE BITFIELD_BIT(VERT_ATTRIB_POINT_SIZE)
#define VERT_BIT_GENERIC0 BITFIELD_BIT(VERT_ATTRIB_GENERIC0)
#define VERT_BIT(i) BITFIELD_BIT(i)
#define VERT_BIT_ALL BITFIELD_RANGE(0, VERT_ATTRIB_MAX)
#define VERT_BIT_FF(i) VERT_BIT(i)
#define VERT_BIT_FF_ALL BITFIELD_RANGE(0, VERT_ATTRIB_FF_MAX)
#define VERT_BIT_TEX(i) VERT_BIT(VERT_ATTRIB_TEX(i))
#define VERT_BIT_TEX_ALL \
BITFIELD_RANGE(VERT_ATTRIB_TEX(0), VERT_ATTRIB_TEX_MAX)
#define VERT_BIT_GENERIC(i) VERT_BIT(VERT_ATTRIB_GENERIC(i))
#define VERT_BIT_GENERIC_ALL \
BITFIELD_RANGE(VERT_ATTRIB_GENERIC(0), VERT_ATTRIB_GENERIC_MAX)
#define VERT_BIT_MAT(i) VERT_BIT(VERT_ATTRIB_MAT(i))
#define VERT_BIT_MAT_ALL \
BITFIELD_RANGE(VERT_ATTRIB_MAT(0), VERT_ATTRIB_MAT_MAX)
/*@}*/
#define MAX_VARYING 32 /**< number of float[4] vectors */
/**
* Indexes for vertex shader outputs, geometry shader inputs/outputs, and
* fragment shader inputs.
*
* Note that some of these values are not available to all pipeline stages.
*
* When this enum is updated, the following code must be updated too:
* - vertResults (in prog_print.c's arb_output_attrib_string())
* - fragAttribs (in prog_print.c's arb_input_attrib_string())
* - _mesa_varying_slot_in_fs()
*/
typedef enum
{
VARYING_SLOT_POS,
VARYING_SLOT_COL0, /* COL0 and COL1 must be contiguous */
VARYING_SLOT_COL1,
VARYING_SLOT_FOGC,
VARYING_SLOT_TEX0, /* TEX0-TEX7 must be contiguous */
VARYING_SLOT_TEX1,
VARYING_SLOT_TEX2,
VARYING_SLOT_TEX3,
VARYING_SLOT_TEX4,
VARYING_SLOT_TEX5,
VARYING_SLOT_TEX6,
VARYING_SLOT_TEX7,
VARYING_SLOT_PSIZ, /* Does not appear in FS */
VARYING_SLOT_BFC0, /* Does not appear in FS */
VARYING_SLOT_BFC1, /* Does not appear in FS */
VARYING_SLOT_EDGE, /* Does not appear in FS */
VARYING_SLOT_CLIP_VERTEX, /* Does not appear in FS */
VARYING_SLOT_CLIP_DIST0,
VARYING_SLOT_CLIP_DIST1,
VARYING_SLOT_CULL_DIST0,
VARYING_SLOT_CULL_DIST1,
VARYING_SLOT_PRIMITIVE_ID, /* Does not appear in VS */
VARYING_SLOT_LAYER, /* Appears as VS or GS output */
VARYING_SLOT_VIEWPORT, /* Appears as VS or GS output */
VARYING_SLOT_FACE, /* FS only */
VARYING_SLOT_PNTC, /* FS only */
VARYING_SLOT_TESS_LEVEL_OUTER, /* Only appears as TCS output. */
VARYING_SLOT_TESS_LEVEL_INNER, /* Only appears as TCS output. */
VARYING_SLOT_BOUNDING_BOX0, /* Only appears as TCS output. */
VARYING_SLOT_BOUNDING_BOX1, /* Only appears as TCS output. */
VARYING_SLOT_VIEW_INDEX,
VARYING_SLOT_VIEWPORT_MASK, /* Does not appear in FS */
VARYING_SLOT_VAR0, /* First generic varying slot */
/* the remaining are simply for the benefit of gl_varying_slot_name()
* and not to be construed as an upper bound:
*/
VARYING_SLOT_VAR1,
VARYING_SLOT_VAR2,
VARYING_SLOT_VAR3,
VARYING_SLOT_VAR4,
VARYING_SLOT_VAR5,
VARYING_SLOT_VAR6,
VARYING_SLOT_VAR7,
VARYING_SLOT_VAR8,
VARYING_SLOT_VAR9,
VARYING_SLOT_VAR10,
VARYING_SLOT_VAR11,
VARYING_SLOT_VAR12,
VARYING_SLOT_VAR13,
VARYING_SLOT_VAR14,
VARYING_SLOT_VAR15,
VARYING_SLOT_VAR16,
VARYING_SLOT_VAR17,
VARYING_SLOT_VAR18,
VARYING_SLOT_VAR19,
VARYING_SLOT_VAR20,
VARYING_SLOT_VAR21,
VARYING_SLOT_VAR22,
VARYING_SLOT_VAR23,
VARYING_SLOT_VAR24,
VARYING_SLOT_VAR25,
VARYING_SLOT_VAR26,
VARYING_SLOT_VAR27,
VARYING_SLOT_VAR28,
VARYING_SLOT_VAR29,
VARYING_SLOT_VAR30,
VARYING_SLOT_VAR31,
} gl_varying_slot;
#define VARYING_SLOT_MAX (VARYING_SLOT_VAR0 + MAX_VARYING)
#define VARYING_SLOT_PATCH0 (VARYING_SLOT_MAX)
#define VARYING_SLOT_TESS_MAX (VARYING_SLOT_PATCH0 + MAX_VARYING)
#define MAX_VARYINGS_INCL_PATCH (VARYING_SLOT_TESS_MAX - VARYING_SLOT_VAR0)
const char *gl_varying_slot_name(gl_varying_slot slot);
/**
* Bitflags for varying slots.
*/
/*@{*/
#define VARYING_BIT_POS BITFIELD64_BIT(VARYING_SLOT_POS)
#define VARYING_BIT_COL0 BITFIELD64_BIT(VARYING_SLOT_COL0)
#define VARYING_BIT_COL1 BITFIELD64_BIT(VARYING_SLOT_COL1)
#define VARYING_BIT_FOGC BITFIELD64_BIT(VARYING_SLOT_FOGC)
#define VARYING_BIT_TEX0 BITFIELD64_BIT(VARYING_SLOT_TEX0)
#define VARYING_BIT_TEX1 BITFIELD64_BIT(VARYING_SLOT_TEX1)
#define VARYING_BIT_TEX2 BITFIELD64_BIT(VARYING_SLOT_TEX2)
#define VARYING_BIT_TEX3 BITFIELD64_BIT(VARYING_SLOT_TEX3)
#define VARYING_BIT_TEX4 BITFIELD64_BIT(VARYING_SLOT_TEX4)
#define VARYING_BIT_TEX5 BITFIELD64_BIT(VARYING_SLOT_TEX5)
#define VARYING_BIT_TEX6 BITFIELD64_BIT(VARYING_SLOT_TEX6)
#define VARYING_BIT_TEX7 BITFIELD64_BIT(VARYING_SLOT_TEX7)
#define VARYING_BIT_TEX(U) BITFIELD64_BIT(VARYING_SLOT_TEX0 + (U))
#define VARYING_BITS_TEX_ANY BITFIELD64_RANGE(VARYING_SLOT_TEX0, \
MAX_TEXTURE_COORD_UNITS)
#define VARYING_BIT_PSIZ BITFIELD64_BIT(VARYING_SLOT_PSIZ)
#define VARYING_BIT_BFC0 BITFIELD64_BIT(VARYING_SLOT_BFC0)
#define VARYING_BIT_BFC1 BITFIELD64_BIT(VARYING_SLOT_BFC1)
#define VARYING_BITS_COLOR (VARYING_BIT_COL0 | \
VARYING_BIT_COL1 | \
VARYING_BIT_BFC0 | \
VARYING_BIT_BFC1)
#define VARYING_BIT_EDGE BITFIELD64_BIT(VARYING_SLOT_EDGE)
#define VARYING_BIT_CLIP_VERTEX BITFIELD64_BIT(VARYING_SLOT_CLIP_VERTEX)
#define VARYING_BIT_CLIP_DIST0 BITFIELD64_BIT(VARYING_SLOT_CLIP_DIST0)
#define VARYING_BIT_CLIP_DIST1 BITFIELD64_BIT(VARYING_SLOT_CLIP_DIST1)
#define VARYING_BIT_CULL_DIST0 BITFIELD64_BIT(VARYING_SLOT_CULL_DIST0)
#define VARYING_BIT_CULL_DIST1 BITFIELD64_BIT(VARYING_SLOT_CULL_DIST1)
#define VARYING_BIT_PRIMITIVE_ID BITFIELD64_BIT(VARYING_SLOT_PRIMITIVE_ID)
#define VARYING_BIT_LAYER BITFIELD64_BIT(VARYING_SLOT_LAYER)
#define VARYING_BIT_VIEWPORT BITFIELD64_BIT(VARYING_SLOT_VIEWPORT)
#define VARYING_BIT_FACE BITFIELD64_BIT(VARYING_SLOT_FACE)
#define VARYING_BIT_PNTC BITFIELD64_BIT(VARYING_SLOT_PNTC)
#define VARYING_BIT_TESS_LEVEL_OUTER BITFIELD64_BIT(VARYING_SLOT_TESS_LEVEL_OUTER)
#define VARYING_BIT_TESS_LEVEL_INNER BITFIELD64_BIT(VARYING_SLOT_TESS_LEVEL_INNER)
#define VARYING_BIT_BOUNDING_BOX0 BITFIELD64_BIT(VARYING_SLOT_BOUNDING_BOX0)
#define VARYING_BIT_BOUNDING_BOX1 BITFIELD64_BIT(VARYING_SLOT_BOUNDING_BOX1)
#define VARYING_BIT_VIEWPORT_MASK BITFIELD64_BIT(VARYING_SLOT_VIEWPORT_MASK)
#define VARYING_BIT_VAR(V) BITFIELD64_BIT(VARYING_SLOT_VAR0 + (V))
/*@}*/
/**
* Bitflags for system values.
*/
#define SYSTEM_BIT_SAMPLE_ID ((uint64_t)1 << SYSTEM_VALUE_SAMPLE_ID)
#define SYSTEM_BIT_SAMPLE_POS ((uint64_t)1 << SYSTEM_VALUE_SAMPLE_POS)
#define SYSTEM_BIT_SAMPLE_MASK_IN ((uint64_t)1 << SYSTEM_VALUE_SAMPLE_MASK_IN)
#define SYSTEM_BIT_LOCAL_INVOCATION_ID ((uint64_t)1 << SYSTEM_VALUE_LOCAL_INVOCATION_ID)
/**
* If the gl_register_file is PROGRAM_SYSTEM_VALUE, the register index will be
* one of these values. If a NIR variable's mode is nir_var_system_value, it
* will be one of these values.
*/
typedef enum
{
/**
* \name System values applicable to all shaders
*/
/*@{*/
/**
* Builtin variables added by GL_ARB_shader_ballot.
*/
/*@{*/
/**
* From the GL_ARB_shader-ballot spec:
*
* "A sub-group is a collection of invocations which execute in lockstep.
* The variable <gl_SubGroupSizeARB> is the maximum number of
* invocations in a sub-group. The maximum <gl_SubGroupSizeARB>
* supported in this extension is 64."
*
* The spec defines this as a uniform. However, it's highly unlikely that
* implementations actually treat it as a uniform (which is loaded from a
* constant buffer). Most likely, this is an implementation-wide constant,
* or perhaps something that depends on the shader stage.
*/
SYSTEM_VALUE_SUBGROUP_SIZE,
/**
* From the GL_ARB_shader_ballot spec:
*
* "The variable <gl_SubGroupInvocationARB> holds the index of the
* invocation within sub-group. This variable is in the range 0 to
* <gl_SubGroupSizeARB>-1, where <gl_SubGroupSizeARB> is the total
* number of invocations in a sub-group."
*/
SYSTEM_VALUE_SUBGROUP_INVOCATION,
/**
* From the GL_ARB_shader_ballot spec:
*
* "The <gl_SubGroup??MaskARB> variables provide a bitmask for all
* invocations, with one bit per invocation starting with the least
* significant bit, according to the following table,
*
* variable equation for bit values
* -------------------- ------------------------------------
* gl_SubGroupEqMaskARB bit index == gl_SubGroupInvocationARB
* gl_SubGroupGeMaskARB bit index >= gl_SubGroupInvocationARB
* gl_SubGroupGtMaskARB bit index > gl_SubGroupInvocationARB
* gl_SubGroupLeMaskARB bit index <= gl_SubGroupInvocationARB
* gl_SubGroupLtMaskARB bit index < gl_SubGroupInvocationARB
*/
SYSTEM_VALUE_SUBGROUP_EQ_MASK,
SYSTEM_VALUE_SUBGROUP_GE_MASK,
SYSTEM_VALUE_SUBGROUP_GT_MASK,
SYSTEM_VALUE_SUBGROUP_LE_MASK,
SYSTEM_VALUE_SUBGROUP_LT_MASK,
/*@}*/
/**
* Builtin variables added by VK_KHR_subgroups
*/
/*@{*/
SYSTEM_VALUE_NUM_SUBGROUPS,
SYSTEM_VALUE_SUBGROUP_ID,
/*@}*/
/*@}*/
/**
* \name Vertex shader system values
*/
/*@{*/
/**
* OpenGL-style vertex ID.
*
* Section 2.11.7 (Shader Execution), subsection Shader Inputs, of the
* OpenGL 3.3 core profile spec says:
*
* "gl_VertexID holds the integer index i implicitly passed by
* DrawArrays or one of the other drawing commands defined in section
* 2.8.3."
*
* Section 2.8.3 (Drawing Commands) of the same spec says:
*
* "The commands....are equivalent to the commands with the same base
* name (without the BaseVertex suffix), except that the ith element
* transferred by the corresponding draw call will be taken from
* element indices[i] + basevertex of each enabled array."
*
* Additionally, the overview in the GL_ARB_shader_draw_parameters spec
* says:
*
* "In unextended GL, vertex shaders have inputs named gl_VertexID and
* gl_InstanceID, which contain, respectively the index of the vertex
* and instance. The value of gl_VertexID is the implicitly passed
* index of the vertex being processed, which includes the value of
* baseVertex, for those commands that accept it."
*
* gl_VertexID gets basevertex added in. This differs from DirectX where
* SV_VertexID does \b not get basevertex added in.
*
* \note
* If all system values are available, \c SYSTEM_VALUE_VERTEX_ID will be
* equal to \c SYSTEM_VALUE_VERTEX_ID_ZERO_BASE plus
* \c SYSTEM_VALUE_BASE_VERTEX.
*
* \sa SYSTEM_VALUE_VERTEX_ID_ZERO_BASE, SYSTEM_VALUE_BASE_VERTEX
*/
SYSTEM_VALUE_VERTEX_ID,
/**
* Instanced ID as supplied to gl_InstanceID
*
* Values assigned to gl_InstanceID always begin with zero, regardless of
* the value of baseinstance.
*
* Section 11.1.3.9 (Shader Inputs) of the OpenGL 4.4 core profile spec
* says:
*
* "gl_InstanceID holds the integer instance number of the current
* primitive in an instanced draw call (see section 10.5)."
*
* Through a big chain of pseudocode, section 10.5 describes that
* baseinstance is not counted by gl_InstanceID. In that section, notice
*
* "If an enabled vertex attribute array is instanced (it has a
* non-zero divisor as specified by VertexAttribDivisor), the element
* index that is transferred to the GL, for all vertices, is given by
*
* floor(instance/divisor) + baseinstance
*
* If an array corresponding to an attribute required by a vertex
* shader is not enabled, then the corresponding element is taken from
* the current attribute state (see section 10.2)."
*
* Note that baseinstance is \b not included in the value of instance.
*/
SYSTEM_VALUE_INSTANCE_ID,
/**
* Vulkan InstanceIndex.
*
* InstanceIndex = gl_InstanceID + gl_BaseInstance
*/
SYSTEM_VALUE_INSTANCE_INDEX,
/**
* DirectX-style vertex ID.
*
* Unlike \c SYSTEM_VALUE_VERTEX_ID, this system value does \b not include
* the value of basevertex.
*
* \sa SYSTEM_VALUE_VERTEX_ID, SYSTEM_VALUE_BASE_VERTEX
*/
SYSTEM_VALUE_VERTEX_ID_ZERO_BASE,
/**
* Value of \c basevertex passed to \c glDrawElementsBaseVertex and similar
* functions.
*
* \sa SYSTEM_VALUE_VERTEX_ID, SYSTEM_VALUE_VERTEX_ID_ZERO_BASE
*/
SYSTEM_VALUE_BASE_VERTEX,
/**
* Depending on the type of the draw call (indexed or non-indexed),
* is the value of \c basevertex passed to \c glDrawElementsBaseVertex and
* similar, or is the value of \c first passed to \c glDrawArrays and
* similar.
*
* \note
* It can be used to calculate the \c SYSTEM_VALUE_VERTEX_ID as
* \c SYSTEM_VALUE_VERTEX_ID_ZERO_BASE plus \c SYSTEM_VALUE_FIRST_VERTEX.
*
* \sa SYSTEM_VALUE_VERTEX_ID_ZERO_BASE, SYSTEM_VALUE_VERTEX_ID
*/
SYSTEM_VALUE_FIRST_VERTEX,
/**
* If the Draw command used to start the rendering was an indexed draw
* or not (~0/0). Useful to calculate \c SYSTEM_VALUE_BASE_VERTEX as
* \c SYSTEM_VALUE_IS_INDEXED_DRAW & \c SYSTEM_VALUE_FIRST_VERTEX.
*/
SYSTEM_VALUE_IS_INDEXED_DRAW,
/**
* Value of \c baseinstance passed to instanced draw entry points
*
* \sa SYSTEM_VALUE_INSTANCE_ID
*/
SYSTEM_VALUE_BASE_INSTANCE,
/**
* From _ARB_shader_draw_parameters:
*
* "Additionally, this extension adds a further built-in variable,
* gl_DrawID to the shading language. This variable contains the index
* of the draw currently being processed by a Multi* variant of a
* drawing command (such as MultiDrawElements or
* MultiDrawArraysIndirect)."
*
* If GL_ARB_multi_draw_indirect is not supported, this is always 0.
*/
SYSTEM_VALUE_DRAW_ID,
/*@}*/
/**
* \name Geometry shader system values
*/
/*@{*/
SYSTEM_VALUE_INVOCATION_ID, /**< (Also in Tessellation Control shader) */
/*@}*/
/**
* \name Fragment shader system values
*/
/*@{*/
SYSTEM_VALUE_FRAG_COORD,
SYSTEM_VALUE_POINT_COORD,
SYSTEM_VALUE_FRONT_FACE,
SYSTEM_VALUE_SAMPLE_ID,
SYSTEM_VALUE_SAMPLE_POS,
SYSTEM_VALUE_SAMPLE_MASK_IN,
SYSTEM_VALUE_HELPER_INVOCATION,
SYSTEM_VALUE_COLOR0,
SYSTEM_VALUE_COLOR1,
/*@}*/
/**
* \name Tessellation Evaluation shader system values
*/
/*@{*/
SYSTEM_VALUE_TESS_COORD,
SYSTEM_VALUE_VERTICES_IN, /**< Tessellation vertices in input patch */
SYSTEM_VALUE_PRIMITIVE_ID,
SYSTEM_VALUE_TESS_LEVEL_OUTER, /**< TES input */
SYSTEM_VALUE_TESS_LEVEL_INNER, /**< TES input */
SYSTEM_VALUE_TESS_LEVEL_OUTER_DEFAULT, /**< TCS input for passthru TCS */
SYSTEM_VALUE_TESS_LEVEL_INNER_DEFAULT, /**< TCS input for passthru TCS */
/*@}*/
/**
* \name Compute shader system values
*/
/*@{*/
SYSTEM_VALUE_LOCAL_INVOCATION_ID,
SYSTEM_VALUE_LOCAL_INVOCATION_INDEX,
SYSTEM_VALUE_GLOBAL_INVOCATION_ID,
SYSTEM_VALUE_GLOBAL_INVOCATION_INDEX,
SYSTEM_VALUE_WORK_GROUP_ID,
SYSTEM_VALUE_NUM_WORK_GROUPS,
SYSTEM_VALUE_LOCAL_GROUP_SIZE,
SYSTEM_VALUE_GLOBAL_GROUP_SIZE,
SYSTEM_VALUE_WORK_DIM,
SYSTEM_VALUE_USER_DATA_AMD,
/*@}*/
/** Required for VK_KHR_device_group */
SYSTEM_VALUE_DEVICE_INDEX,
/** Required for VK_KHX_multiview */
SYSTEM_VALUE_VIEW_INDEX,
/**
* Driver internal vertex-count, used (for example) for drivers to
* calculate stride for stream-out outputs. Not externally visible.
*/
SYSTEM_VALUE_VERTEX_CNT,
/**
* Required for AMD_shader_explicit_vertex_parameter and also used for
* varying-fetch instructions.
*
* The _SIZE value is "primitive size", used to scale i/j in primitive
* space to pixel space.
*/
SYSTEM_VALUE_BARYCENTRIC_PERSP_PIXEL,
SYSTEM_VALUE_BARYCENTRIC_PERSP_SAMPLE,
SYSTEM_VALUE_BARYCENTRIC_PERSP_CENTROID,
SYSTEM_VALUE_BARYCENTRIC_PERSP_SIZE,
SYSTEM_VALUE_BARYCENTRIC_LINEAR_PIXEL,
SYSTEM_VALUE_BARYCENTRIC_LINEAR_CENTROID,
SYSTEM_VALUE_BARYCENTRIC_LINEAR_SAMPLE,
SYSTEM_VALUE_BARYCENTRIC_PULL_MODEL,
/**
* IR3 specific geometry shader and tesselation control shader system
* values that packs invocation id, thread id and vertex id. Having this
* as a nir level system value lets us do the unpacking in nir.
*/
SYSTEM_VALUE_GS_HEADER_IR3,
SYSTEM_VALUE_TCS_HEADER_IR3,
SYSTEM_VALUE_MAX /**< Number of values */
} gl_system_value;
const char *gl_system_value_name(gl_system_value sysval);
/**
* The possible interpolation qualifiers that can be applied to a fragment
* shader input in GLSL.
*
* Note: INTERP_MODE_NONE must be 0 so that memsetting the
* ir_variable data structure to 0 causes the default behavior.
*/
enum glsl_interp_mode
{
INTERP_MODE_NONE = 0,
INTERP_MODE_SMOOTH,
INTERP_MODE_FLAT,
INTERP_MODE_NOPERSPECTIVE,
INTERP_MODE_EXPLICIT,
INTERP_MODE_COUNT /**< Number of interpolation qualifiers */
};
enum glsl_interface_packing {
GLSL_INTERFACE_PACKING_STD140,
GLSL_INTERFACE_PACKING_SHARED,
GLSL_INTERFACE_PACKING_PACKED,
GLSL_INTERFACE_PACKING_STD430
};
const char *glsl_interp_mode_name(enum glsl_interp_mode qual);
/**
* Fragment program results
*/
typedef enum
{
FRAG_RESULT_DEPTH = 0,
FRAG_RESULT_STENCIL = 1,
/* If a single color should be written to all render targets, this
* register is written. No FRAG_RESULT_DATAn will be written.
*/
FRAG_RESULT_COLOR = 2,
FRAG_RESULT_SAMPLE_MASK = 3,
/* FRAG_RESULT_DATAn are the per-render-target (GLSL gl_FragData[n]
* or ARB_fragment_program fragment.color[n]) color results. If
* any are written, FRAG_RESULT_COLOR will not be written.
* FRAG_RESULT_DATA1 and up are simply for the benefit of
* gl_frag_result_name() and not to be construed as an upper bound
*/
FRAG_RESULT_DATA0 = 4,
FRAG_RESULT_DATA1,
FRAG_RESULT_DATA2,
FRAG_RESULT_DATA3,
FRAG_RESULT_DATA4,
FRAG_RESULT_DATA5,
FRAG_RESULT_DATA6,
FRAG_RESULT_DATA7,
} gl_frag_result;
const char *gl_frag_result_name(gl_frag_result result);
#define FRAG_RESULT_MAX (FRAG_RESULT_DATA0 + MAX_DRAW_BUFFERS)
/**
* \brief Layout qualifiers for gl_FragDepth.
*
* Extension AMD_conservative_depth allows gl_FragDepth to be redeclared with
* a layout qualifier.
*
* \see enum ir_depth_layout
*/
enum gl_frag_depth_layout
{
FRAG_DEPTH_LAYOUT_NONE, /**< No layout is specified. */
FRAG_DEPTH_LAYOUT_ANY,
FRAG_DEPTH_LAYOUT_GREATER,
FRAG_DEPTH_LAYOUT_LESS,
FRAG_DEPTH_LAYOUT_UNCHANGED
};
/**
* \brief Buffer access qualifiers
*/
enum gl_access_qualifier
{
ACCESS_COHERENT = (1 << 0),
ACCESS_RESTRICT = (1 << 1),
ACCESS_VOLATILE = (1 << 2),
ACCESS_NON_READABLE = (1 << 3),
ACCESS_NON_WRITEABLE = (1 << 4),
/** The access may use a non-uniform buffer or image index */
ACCESS_NON_UNIFORM = (1 << 5),
/* This has the same semantics as NIR_INTRINSIC_CAN_REORDER, only to be
* used with loads. In other words, it means that the load can be
* arbitrarily reordered, or combined with other loads to the same address.
* It is implied by ACCESS_NON_WRITEABLE together with ACCESS_RESTRICT, and
* a lack of ACCESS_COHERENT and ACCESS_VOLATILE.
*/
ACCESS_CAN_REORDER = (1 << 6),
/** Use as little cache space as possible. */
ACCESS_STREAM_CACHE_POLICY = (1 << 7),
};
/**
* \brief Blend support qualifiers
*/
enum gl_advanced_blend_mode
{
BLEND_NONE = 0x0000,
BLEND_MULTIPLY = 0x0001,
BLEND_SCREEN = 0x0002,
BLEND_OVERLAY = 0x0004,
BLEND_DARKEN = 0x0008,
BLEND_LIGHTEN = 0x0010,
BLEND_COLORDODGE = 0x0020,
BLEND_COLORBURN = 0x0040,
BLEND_HARDLIGHT = 0x0080,
BLEND_SOFTLIGHT = 0x0100,
BLEND_DIFFERENCE = 0x0200,
BLEND_EXCLUSION = 0x0400,
BLEND_HSL_HUE = 0x0800,
BLEND_HSL_SATURATION = 0x1000,
BLEND_HSL_COLOR = 0x2000,
BLEND_HSL_LUMINOSITY = 0x4000,
BLEND_ALL = 0x7fff,
};
enum blend_func
{
BLEND_FUNC_ADD,
BLEND_FUNC_SUBTRACT,
BLEND_FUNC_REVERSE_SUBTRACT,
BLEND_FUNC_MIN,
BLEND_FUNC_MAX,
};
enum blend_factor
{
BLEND_FACTOR_ZERO,
BLEND_FACTOR_SRC_COLOR,
BLEND_FACTOR_DST_COLOR,
BLEND_FACTOR_SRC_ALPHA,
BLEND_FACTOR_DST_ALPHA,
BLEND_FACTOR_CONSTANT_COLOR,
BLEND_FACTOR_CONSTANT_ALPHA,
BLEND_FACTOR_SRC_ALPHA_SATURATE,
};
enum gl_tess_spacing
{
TESS_SPACING_UNSPECIFIED,
TESS_SPACING_EQUAL,
TESS_SPACING_FRACTIONAL_ODD,
TESS_SPACING_FRACTIONAL_EVEN,
};
/**
* A compare function enum for use in compiler lowering passes. This is in
* the same order as GL's compare functions (shifted down by GL_NEVER), and is
* exactly the same as gallium's PIPE_FUNC_*.
*/
enum compare_func
{
COMPARE_FUNC_NEVER,
COMPARE_FUNC_LESS,
COMPARE_FUNC_EQUAL,
COMPARE_FUNC_LEQUAL,
COMPARE_FUNC_GREATER,
COMPARE_FUNC_NOTEQUAL,
COMPARE_FUNC_GEQUAL,
COMPARE_FUNC_ALWAYS,
};
/**
* Arrangements for grouping invocations from NV_compute_shader_derivatives.
*
* The extension provides new layout qualifiers that support two different
* arrangements of compute shader invocations for the purpose of derivative
* computation. When specifying
*
* layout(derivative_group_quadsNV) in;
*
* compute shader invocations are grouped into 2x2x1 arrays whose four local
* invocation ID values follow the pattern:
*
* +-----------------+------------------+
* | (2x+0, 2y+0, z) | (2x+1, 2y+0, z) |
* +-----------------+------------------+
* | (2x+0, 2y+1, z) | (2x+1, 2y+1, z) |
* +-----------------+------------------+
*
* where Y increases from bottom to top. When specifying
*
* layout(derivative_group_linearNV) in;
*
* compute shader invocations are grouped into 2x2x1 arrays whose four local
* invocation index values follow the pattern:
*
* +------+------+
* | 4n+0 | 4n+1 |
* +------+------+
* | 4n+2 | 4n+3 |
* +------+------+
*
* If neither layout qualifier is specified, derivatives in compute shaders
* return zero, which is consistent with the handling of built-in texture
* functions like texture() in GLSL 4.50 compute shaders.
*/
enum gl_derivative_group {
DERIVATIVE_GROUP_NONE = 0,
DERIVATIVE_GROUP_QUADS,
DERIVATIVE_GROUP_LINEAR,
};
enum float_controls
{
FLOAT_CONTROLS_DEFAULT_FLOAT_CONTROL_MODE = 0x0000,
FLOAT_CONTROLS_DENORM_PRESERVE_FP16 = 0x0001,
FLOAT_CONTROLS_DENORM_PRESERVE_FP32 = 0x0002,
FLOAT_CONTROLS_DENORM_PRESERVE_FP64 = 0x0004,
FLOAT_CONTROLS_DENORM_FLUSH_TO_ZERO_FP16 = 0x0008,
FLOAT_CONTROLS_DENORM_FLUSH_TO_ZERO_FP32 = 0x0010,
FLOAT_CONTROLS_DENORM_FLUSH_TO_ZERO_FP64 = 0x0020,
FLOAT_CONTROLS_SIGNED_ZERO_INF_NAN_PRESERVE_FP16 = 0x0040,
FLOAT_CONTROLS_SIGNED_ZERO_INF_NAN_PRESERVE_FP32 = 0x0080,
FLOAT_CONTROLS_SIGNED_ZERO_INF_NAN_PRESERVE_FP64 = 0x0100,
FLOAT_CONTROLS_ROUNDING_MODE_RTE_FP16 = 0x0200,
FLOAT_CONTROLS_ROUNDING_MODE_RTE_FP32 = 0x0400,
FLOAT_CONTROLS_ROUNDING_MODE_RTE_FP64 = 0x0800,
FLOAT_CONTROLS_ROUNDING_MODE_RTZ_FP16 = 0x1000,
FLOAT_CONTROLS_ROUNDING_MODE_RTZ_FP32 = 0x2000,
FLOAT_CONTROLS_ROUNDING_MODE_RTZ_FP64 = 0x4000,
};
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* SHADER_ENUMS_H */

326
extra/disassemblers/adreno/util/bitscan.h
View File

@@ -1,326 +0,0 @@
/**************************************************************************
*
* Copyright 2008 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
* IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
**************************************************************************/
#ifndef BITSCAN_H
#define BITSCAN_H
#include <assert.h>
#include <stdint.h>
#include <stdbool.h>
#include <string.h>
#if defined(_MSC_VER)
#include <intrin.h>
#endif
#if defined(__POPCNT__)
#include <popcntintrin.h>
#endif
//#include "c99_compat.h"
#ifdef __cplusplus
extern "C" {
#endif
/**
* Find first bit set in word. Least significant bit is 1.
* Return 0 if no bits set.
*/
#ifdef HAVE___BUILTIN_FFS
#define ffs __builtin_ffs
#elif defined(_MSC_VER) && (_M_IX86 || _M_ARM || _M_AMD64 || _M_IA64)
static inline
int ffs(int i)
{
unsigned long index;
if (_BitScanForward(&index, i))
return index + 1;
else
return 0;
}
#else
extern
int ffs(int i);
#endif
#ifdef HAVE___BUILTIN_FFSLL
#define ffsll __builtin_ffsll
#elif defined(_MSC_VER) && (_M_AMD64 || _M_ARM64 || _M_IA64)
static inline int
ffsll(long long int i)
{
unsigned long index;
if (_BitScanForward64(&index, i))
return index + 1;
else
return 0;
}
#else
extern int
ffsll(long long int val);
#endif
/* Destructively loop over all of the bits in a mask as in:
*
* while (mymask) {
* int i = u_bit_scan(&mymask);
* ... process element i
* }
*
*/
static inline int
u_bit_scan(unsigned *mask)
{
const int i = ffs(*mask) - 1;
*mask ^= (1u << i);
return i;
}
static inline int
u_bit_scan64(uint64_t *mask)
{
const int i = ffsll(*mask) - 1;
*mask ^= (((uint64_t)1) << i);
return i;
}
/* Determine if an unsigned value is a power of two.
*
* \note
* Zero is treated as a power of two.
*/
static inline bool
util_is_power_of_two_or_zero(unsigned v)
{
return (v & (v - 1)) == 0;
}
/* Determine if an uint64_t value is a power of two.
*
* \note
* Zero is treated as a power of two.
*/
static inline bool
util_is_power_of_two_or_zero64(uint64_t v)
{
return (v & (v - 1)) == 0;
}
/* Determine if an unsigned value is a power of two.
*
* \note
* Zero is \b not treated as a power of two.
*/
static inline bool
util_is_power_of_two_nonzero(unsigned v)
{
/* __POPCNT__ is different from HAVE___BUILTIN_POPCOUNT. The latter
* indicates the existence of the __builtin_popcount function. The former
* indicates that _mm_popcnt_u32 exists and is a native instruction.
*
* The other alternative is to use SSE 4.2 compile-time flags. This has
* two drawbacks. First, there is currently no build infrastructure for
* SSE 4.2 (only 4.1), so that would have to be added. Second, some AMD
* CPUs support POPCNT but not SSE 4.2 (e.g., Barcelona).
*/
#ifdef __POPCNT__
return _mm_popcnt_u32(v) == 1;
#else
return v != 0 && (v & (v - 1)) == 0;
#endif
}
/* For looping over a bitmask when you want to loop over consecutive bits
* manually, for example:
*
* while (mask) {
* int start, count, i;
*
* u_bit_scan_consecutive_range(&mask, &start, &count);
*
* for (i = 0; i < count; i++)
* ... process element (start+i)
* }
*/
static inline void
u_bit_scan_consecutive_range(unsigned *mask, int *start, int *count)
{
if (*mask == 0xffffffff) {
*start = 0;
*count = 32;
*mask = 0;
return;
}
*start = ffs(*mask) - 1;
*count = ffs(~(*mask >> *start)) - 1;
*mask &= ~(((1u << *count) - 1) << *start);
}
static inline void
u_bit_scan_consecutive_range64(uint64_t *mask, int *start, int *count)
{
if (*mask == ~0ull) {
*start = 0;
*count = 64;
*mask = 0;
return;
}
*start = ffsll(*mask) - 1;
*count = ffsll(~(*mask >> *start)) - 1;
*mask &= ~(((((uint64_t)1) << *count) - 1) << *start);
}
/**
* Find last bit set in a word. The least significant bit is 1.
* Return 0 if no bits are set.
* Essentially ffs() in the reverse direction.
*/
static inline unsigned
util_last_bit(unsigned u)
{
#if defined(HAVE___BUILTIN_CLZ)
return u == 0 ? 0 : 32 - __builtin_clz(u);
#elif defined(_MSC_VER) && (_M_IX86 || _M_ARM || _M_AMD64 || _M_IA64)
unsigned long index;
if (_BitScanReverse(&index, u))
return index + 1;
else
return 0;
#else
unsigned r = 0;
while (u) {
r++;
u >>= 1;
}
return r;
#endif
}
/**
* Find last bit set in a word. The least significant bit is 1.
* Return 0 if no bits are set.
* Essentially ffsll() in the reverse direction.
*/
static inline unsigned
util_last_bit64(uint64_t u)
{
#if defined(HAVE___BUILTIN_CLZLL)
return u == 0 ? 0 : 64 - __builtin_clzll(u);
#elif defined(_MSC_VER) && (_M_AMD64 || _M_ARM64 || _M_IA64)
unsigned long index;
if (_BitScanReverse64(&index, u))
return index + 1;
else
return 0;
#else
unsigned r = 0;
while (u) {
r++;
u >>= 1;
}
return r;
#endif
}
/**
* Find last bit in a word that does not match the sign bit. The least
* significant bit is 1.
* Return 0 if no bits are set.
*/
static inline unsigned
util_last_bit_signed(int i)
{
if (i >= 0)
return util_last_bit(i);
else
return util_last_bit(~(unsigned)i);
}
/* Returns a bitfield in which the first count bits starting at start are
* set.
*/
static inline unsigned
u_bit_consecutive(unsigned start, unsigned count)
{
assert(start + count <= 32);
if (count == 32)
return ~0;
return ((1u << count) - 1) << start;
}
static inline uint64_t
u_bit_consecutive64(unsigned start, unsigned count)
{
assert(start + count <= 64);
if (count == 64)
return ~(uint64_t)0;
return (((uint64_t)1 << count) - 1) << start;
}
/**
* Return number of bits set in n.
*/
static inline unsigned
util_bitcount(unsigned n)
{
#if defined(HAVE___BUILTIN_POPCOUNT)
return __builtin_popcount(n);
#else
/* K&R classic bitcount.
*
* For each iteration, clear the LSB from the bitfield.
* Requires only one iteration per set bit, instead of
* one iteration per bit less than highest set bit.
*/
unsigned bits;
for (bits = 0; n; bits++) {
n &= n - 1;
}
return bits;
#endif
}
static inline unsigned
util_bitcount64(uint64_t n)
{
#ifdef HAVE___BUILTIN_POPCOUNTLL
return __builtin_popcountll(n);
#else
return util_bitcount(n) + util_bitcount(n >> 32);
#endif
}
#ifdef __cplusplus
}
#endif
#endif /* BITSCAN_H */

261
extra/disassemblers/adreno/util/bitset.h
View File

@@ -1,261 +0,0 @@
/*
* Mesa 3-D graphics library
*
* Copyright (C) 2006 Brian Paul All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*/
/**
* \file bitset.h
* \brief Bitset of arbitrary size definitions.
* \author Michal Krol
*/
#ifndef BITSET_H
#define BITSET_H
//#include "util/bitscan.h"
//#include "util/macros.h"
/****************************************************************************
* generic bitset implementation
*/
#define BITSET_WORD unsigned int
#define BITSET_WORDBITS (sizeof (BITSET_WORD) * 8)
/* bitset declarations
*/
#define BITSET_WORDS(bits) (((bits) + BITSET_WORDBITS - 1) / BITSET_WORDBITS)
#define BITSET_DECLARE(name, bits) BITSET_WORD name[BITSET_WORDS(bits)]
/* bitset operations
*/
#define BITSET_COPY(x, y) memcpy( (x), (y), sizeof (x) )
#define BITSET_EQUAL(x, y) (memcmp( (x), (y), sizeof (x) ) == 0)
#define BITSET_ZERO(x) memset( (x), 0, sizeof (x) )
#define BITSET_ONES(x) memset( (x), 0xff, sizeof (x) )
#define BITSET_BITWORD(b) ((b) / BITSET_WORDBITS)
#define BITSET_BIT(b) (1u << ((b) % BITSET_WORDBITS))
/* single bit operations
*/
#define BITSET_TEST(x, b) (((x)[BITSET_BITWORD(b)] & BITSET_BIT(b)) != 0)
#define BITSET_SET(x, b) ((x)[BITSET_BITWORD(b)] |= BITSET_BIT(b))
#define BITSET_CLEAR(x, b) ((x)[BITSET_BITWORD(b)] &= ~BITSET_BIT(b))
#define BITSET_MASK(b) (((b) % BITSET_WORDBITS == 0) ? ~0 : BITSET_BIT(b) - 1)
#define BITSET_RANGE(b, e) ((BITSET_MASK((e) + 1)) & ~(BITSET_BIT(b) - 1))
/* bit range operations
*/
#define BITSET_TEST_RANGE(x, b, e) \
(BITSET_BITWORD(b) == BITSET_BITWORD(e) ? \
(((x)[BITSET_BITWORD(b)] & BITSET_RANGE(b, e)) != 0) : \
(assert (!"BITSET_TEST_RANGE: bit range crosses word boundary"), 0))
#define BITSET_SET_RANGE(x, b, e) \
(BITSET_BITWORD(b) == BITSET_BITWORD(e) ? \
((x)[BITSET_BITWORD(b)] |= BITSET_RANGE(b, e)) : \
(assert (!"BITSET_SET_RANGE: bit range crosses word boundary"), 0))
#define BITSET_CLEAR_RANGE(x, b, e) \
(BITSET_BITWORD(b) == BITSET_BITWORD(e) ? \
((x)[BITSET_BITWORD(b)] &= ~BITSET_RANGE(b, e)) : \
(assert (!"BITSET_CLEAR_RANGE: bit range crosses word boundary"), 0))
/* Get first bit set in a bitset.
*/
static inline int
__bitset_ffs(const BITSET_WORD *x, int n)
{
int i;
for (i = 0; i < n; i++) {
if (x[i])
return ffs(x[i]) + BITSET_WORDBITS * i;
}
return 0;
}
#define BITSET_FFS(x) __bitset_ffs(x, ARRAY_SIZE(x))
static inline unsigned
__bitset_next_set(unsigned i, BITSET_WORD *tmp,
const BITSET_WORD *set, unsigned size)
{
unsigned bit, word;
/* NOTE: The initial conditions for this function are very specific. At
* the start of the loop, the tmp variable must be set to *set and the
* initial i value set to 0. This way, if there is a bit set in the first
* word, we ignore the i-value and just grab that bit (so 0 is ok, even
* though 0 may be returned). If the first word is 0, then the value of
* `word` will be 0 and we will go on to look at the second word.
*/
word = BITSET_BITWORD(i);
while (*tmp == 0) {
word++;
if (word >= BITSET_WORDS(size))
return size;
*tmp = set[word];
}
/* Find the next set bit in the non-zero word */
bit = ffs(*tmp) - 1;
/* Unset the bit */
*tmp &= ~(1ull << bit);
return word * BITSET_WORDBITS + bit;
}
/**
* Iterates over each set bit in a set
*
* @param __i iteration variable, bit number
* @param __set the bitset to iterate (will not be modified)
* @param __size number of bits in the set to consider
*/
#define BITSET_FOREACH_SET(__i, __set, __size) \
for (BITSET_WORD __tmp = *(__set), *__foo = &__tmp; __foo != NULL; __foo = NULL) \
for (__i = 0; \
(__i = __bitset_next_set(__i, &__tmp, __set, __size)) < __size;)
#ifdef __cplusplus
/**
* Simple C++ wrapper of a bitset type of static size, with value semantics
* and basic bitwise arithmetic operators. The operators defined below are
* expected to have the same semantics as the same operator applied to other
* fundamental integer types. T is the name of the struct to instantiate
* it as, and N is the number of bits in the bitset.
*/
#define DECLARE_BITSET_T(T, N) struct T { \
EXPLICIT_CONVERSION \
operator bool() const \
{ \
for (unsigned i = 0; i < BITSET_WORDS(N); i++) \
if (words[i]) \
return true; \
return false; \
} \
\
T & \
operator=(int x) \
{ \
const T c = {{ (BITSET_WORD)x }}; \
return *this = c; \
} \
\
friend bool \
operator==(const T &b, const T &c) \
{ \
return BITSET_EQUAL(b.words, c.words); \
} \
\
friend bool \
operator!=(const T &b, const T &c) \
{ \
return !(b == c); \
} \
\
friend bool \
operator==(const T &b, int x) \
{ \
const T c = {{ (BITSET_WORD)x }}; \
return b == c; \
} \
\
friend bool \
operator!=(const T &b, int x) \
{ \
return !(b == x); \
} \
\
friend T \
operator~(const T &b) \
{ \
T c; \
for (unsigned i = 0; i < BITSET_WORDS(N); i++) \
c.words[i] = ~b.words[i]; \
return c; \
} \
\
T & \
operator|=(const T &b) \
{ \
for (unsigned i = 0; i < BITSET_WORDS(N); i++) \
words[i] |= b.words[i]; \
return *this; \
} \
\
friend T \
operator|(const T &b, const T &c) \
{ \
T d = b; \
d |= c; \
return d; \
} \
\
T & \
operator&=(const T &b) \
{ \
for (unsigned i = 0; i < BITSET_WORDS(N); i++) \
words[i] &= b.words[i]; \
return *this; \
} \
\
friend T \
operator&(const T &b, const T &c) \
{ \
T d = b; \
d &= c; \
return d; \
} \
\
bool \
test(unsigned i) const \
{ \
return BITSET_TEST(words, i); \
} \
\
T & \
set(unsigned i) \
{ \
BITSET_SET(words, i); \
return *this; \
} \
\
T & \
clear(unsigned i) \
{ \
BITSET_CLEAR(words, i); \
return *this; \
} \
\
BITSET_WORD words[BITSET_WORDS(N)]; \
}
#endif
#endif

262
extra/disassemblers/adreno/util/list.h
View File

@@ -1,262 +0,0 @@
/**************************************************************************
*
* Copyright 2006 VMware, Inc., Bismarck, ND. USA.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
* USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
**************************************************************************/
/**
* \file
* List macros heavily inspired by the Linux kernel
* list handling. No list looping yet.
*
* Is not threadsafe, so common operations need to
* be protected using an external mutex.
*/
#ifndef _UTIL_LIST_H_
#define _UTIL_LIST_H_
#include <stdbool.h>
#include <stddef.h>
#include <assert.h>
#ifdef DEBUG
# define list_assert(cond, msg) assert(cond && msg)
#else
# define list_assert(cond, msg) (void)(0 && (cond))
#endif
struct list_head
{
struct list_head *prev;
struct list_head *next;
};
static inline void list_inithead(struct list_head *item)
{
item->prev = item;
item->next = item;
}
static inline void list_add(struct list_head *item, struct list_head *list)
{
item->prev = list;
item->next = list->next;
list->next->prev = item;
list->next = item;
}
static inline void list_addtail(struct list_head *item, struct list_head *list)
{
item->next = list;
item->prev = list->prev;
list->prev->next = item;
list->prev = item;
}
static inline bool list_is_empty(const struct list_head *list);
static inline void list_replace(struct list_head *from, struct list_head *to)
{
if (list_is_empty(from)) {
list_inithead(to);
} else {
to->prev = from->prev;
to->next = from->next;
from->next->prev = to;
from->prev->next = to;
}
}
static inline void list_del(struct list_head *item)
{
item->prev->next = item->next;
item->next->prev = item->prev;
item->prev = item->next = NULL;
}
static inline void list_delinit(struct list_head *item)
{
item->prev->next = item->next;
item->next->prev = item->prev;
item->next = item;
item->prev = item;
}
static inline bool list_is_empty(const struct list_head *list)
{
return list->next == list;
}
/**
* Returns whether the list has exactly one element.
*/
static inline bool list_is_singular(const struct list_head *list)
{
return list->next != NULL && list->next != list && list->next->next == list;
}
static inline unsigned list_length(const struct list_head *list)
{
struct list_head *node;
unsigned length = 0;
for (node = list->next; node != list; node = node->next)
length++;
return length;
}
static inline void list_splice(struct list_head *src, struct list_head *dst)
{
if (list_is_empty(src))
return;
src->next->prev = dst;
src->prev->next = dst->next;
dst->next->prev = src->prev;
dst->next = src->next;
}
static inline void list_splicetail(struct list_head *src, struct list_head *dst)
{
if (list_is_empty(src))
return;
src->prev->next = dst;
src->next->prev = dst->prev;
dst->prev->next = src->next;
dst->prev = src->prev;
}
static inline void list_validate(const struct list_head *list)
{
struct list_head *node;
assert(list->next->prev == list && list->prev->next == list);
for (node = list->next; node != list; node = node->next)
assert(node->next->prev == node && node->prev->next == node);
}
#define LIST_ENTRY(__type, __item, __field) \
((__type *)(((char *)(__item)) - offsetof(__type, __field)))
/**
* Cast from a pointer to a member of a struct back to the containing struct.
*
* 'sample' MUST be initialized, or else the result is undefined!
*/
#ifndef container_of
#define container_of(ptr, sample, member) \
(void *)((char *)(ptr) \
- ((char *)&(sample)->member - (char *)(sample)))
#endif
#define list_first_entry(ptr, type, member) \
LIST_ENTRY(type, (ptr)->next, member)
#define list_last_entry(ptr, type, member) \
LIST_ENTRY(type, (ptr)->prev, member)
#define LIST_FOR_EACH_ENTRY(pos, head, member) \
for (pos = NULL, pos = container_of((head)->next, pos, member); \
&pos->member != (head); \
pos = container_of(pos->member.next, pos, member))
#define LIST_FOR_EACH_ENTRY_SAFE(pos, storage, head, member) \
for (pos = NULL, pos = container_of((head)->next, pos, member), \
storage = container_of(pos->member.next, pos, member); \
&pos->member != (head); \
pos = storage, storage = container_of(storage->member.next, storage, member))
#define LIST_FOR_EACH_ENTRY_SAFE_REV(pos, storage, head, member) \
for (pos = NULL, pos = container_of((head)->prev, pos, member), \
storage = container_of(pos->member.prev, pos, member); \
&pos->member != (head); \
pos = storage, storage = container_of(storage->member.prev, storage, member))
#define LIST_FOR_EACH_ENTRY_FROM(pos, start, head, member) \
for (pos = NULL, pos = container_of((start), pos, member); \
&pos->member != (head); \
pos = container_of(pos->member.next, pos, member))
#define LIST_FOR_EACH_ENTRY_FROM_REV(pos, start, head, member) \
for (pos = NULL, pos = container_of((start), pos, member); \
&pos->member != (head); \
pos = container_of(pos->member.prev, pos, member))
#define list_for_each_entry(type, pos, head, member) \
for (type *pos = LIST_ENTRY(type, (head)->next, member), \
*__next = LIST_ENTRY(type, pos->member.next, member); \
&pos->member != (head); \
pos = LIST_ENTRY(type, pos->member.next, member), \
list_assert(pos == __next, "use _safe iterator"), \
__next = LIST_ENTRY(type, __next->member.next, member))
#define list_for_each_entry_safe(type, pos, head, member) \
for (type *pos = LIST_ENTRY(type, (head)->next, member), \
*__next = LIST_ENTRY(type, pos->member.next, member); \
&pos->member != (head); \
pos = __next, \
__next = LIST_ENTRY(type, __next->member.next, member))
#define list_for_each_entry_rev(type, pos, head, member) \
for (type *pos = LIST_ENTRY(type, (head)->prev, member), \
*__prev = LIST_ENTRY(type, pos->member.prev, member); \
&pos->member != (head); \
pos = LIST_ENTRY(type, pos->member.prev, member), \
list_assert(pos == __prev, "use _safe iterator"), \
__prev = LIST_ENTRY(type, __prev->member.prev, member))
#define list_for_each_entry_safe_rev(type, pos, head, member) \
for (type *pos = LIST_ENTRY(type, (head)->prev, member), \
*__prev = LIST_ENTRY(type, pos->member.prev, member); \
&pos->member != (head); \
pos = __prev, \
__prev = LIST_ENTRY(type, __prev->member.prev, member))
#define list_for_each_entry_from(type, pos, start, head, member) \
for (type *pos = LIST_ENTRY(type, (start), member); \
&pos->member != (head); \
pos = LIST_ENTRY(type, pos->member.next, member))
#define list_for_each_entry_from_safe(type, pos, start, head, member) \
for (type *pos = LIST_ENTRY(type, (start), member), \
*__next = LIST_ENTRY(type, pos->member.next, member); \
&pos->member != (head); \
pos = __next, \
__next = LIST_ENTRY(type, __next->member.next, member))
#define list_for_each_entry_from_rev(type, pos, start, head, member) \
for (type *pos = LIST_ENTRY(type, (start), member); \
&pos->member != (head); \
pos = LIST_ENTRY(type, pos->member.prev, member))
#define list_pair_for_each_entry(type, pos1, pos2, head1, head2, member) \
for (type *pos1 = LIST_ENTRY(type, (head1)->next, member), \
*pos2 = LIST_ENTRY(type, (head2)->next, member); \
&pos1->member != (head1) && &pos2->member != (head2); \
pos1 = LIST_ENTRY(type, pos1->member.next, member), \
pos2 = LIST_ENTRY(type, pos2->member.next, member))
#endif /*_UTIL_LIST_H_*/

346
extra/disassemblers/adreno/util/macros.h
View File

@@ -1,346 +0,0 @@
/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef UTIL_MACROS_H
#define UTIL_MACROS_H
#include <assert.h>
/* Compute the size of an array */
#ifndef ARRAY_SIZE
# define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
#endif
/* For compatibility with Clang's __has_builtin() */
#ifndef __has_builtin
# define __has_builtin(x) 0
#endif
/**
* __builtin_expect macros
*/
#if !defined(HAVE___BUILTIN_EXPECT)
# define __builtin_expect(x, y) (x)
#endif
#ifndef likely
# ifdef HAVE___BUILTIN_EXPECT
# define likely(x) __builtin_expect(!!(x), 1)
# define unlikely(x) __builtin_expect(!!(x), 0)
# else
# define likely(x) (x)
# define unlikely(x) (x)
# endif
#endif
/**
* Static (compile-time) assertion.
* Basically, use COND to dimension an array. If COND is false/zero the
* array size will be -1 and we'll get a compilation error.
*/
#define STATIC_ASSERT(COND) \
do { \
(void) sizeof(char [1 - 2*!(COND)]); \
} while (0)
/**
* Unreachable macro. Useful for suppressing "control reaches end of non-void
* function" warnings.
*/
#if defined(HAVE___BUILTIN_UNREACHABLE) || __has_builtin(__builtin_unreachable)
#define unreachable(str) \
do { \
assert(!str); \
__builtin_unreachable(); \
} while (0)
#elif defined (_MSC_VER)
#define unreachable(str) \
do { \
assert(!str); \
__assume(0); \
} while (0)
#else
#define unreachable(str) assert(!str)
#endif
/**
* Assume macro. Useful for expressing our assumptions to the compiler,
* typically for purposes of silencing warnings.
*/
#if __has_builtin(__builtin_assume)
#define assume(expr) \
do { \
assert(expr); \
__builtin_assume(expr); \
} while (0)
#elif defined HAVE___BUILTIN_UNREACHABLE
#define assume(expr) ((expr) ? ((void) 0) \
: (assert(!"assumption failed"), \
__builtin_unreachable()))
#elif defined (_MSC_VER)
#define assume(expr) __assume(expr)
#else
#define assume(expr) assert(expr)
#endif
/* Attribute const is used for functions that have no effects other than their
* return value, and only rely on the argument values to compute the return
* value. As a result, calls to it can be CSEed. Note that using memory
* pointed to by the arguments is not allowed for const functions.
*/
#ifdef HAVE_FUNC_ATTRIBUTE_CONST
#define ATTRIBUTE_CONST __attribute__((__const__))
#else
#define ATTRIBUTE_CONST
#endif
#ifdef HAVE_FUNC_ATTRIBUTE_FLATTEN
#define FLATTEN __attribute__((__flatten__))
#else
#define FLATTEN
#endif
#ifdef HAVE_FUNC_ATTRIBUTE_FORMAT
#define PRINTFLIKE(f, a) __attribute__ ((format(__printf__, f, a)))
#else
#define PRINTFLIKE(f, a)
#endif
#ifdef HAVE_FUNC_ATTRIBUTE_MALLOC
#define MALLOCLIKE __attribute__((__malloc__))
#else
#define MALLOCLIKE
#endif
/* Forced function inlining */
/* Note: Clang also sets __GNUC__ (see other cases below) */
#ifndef ALWAYS_INLINE
# if defined(__GNUC__)
# define ALWAYS_INLINE inline __attribute__((always_inline))
# elif defined(_MSC_VER)
# define ALWAYS_INLINE __forceinline
# else
# define ALWAYS_INLINE inline
# endif
#endif
/* Used to optionally mark structures with misaligned elements or size as
* packed, to trade off performance for space.
*/
#ifdef HAVE_FUNC_ATTRIBUTE_PACKED
#define PACKED __attribute__((__packed__))
#else
#define PACKED
#endif
/* Attribute pure is used for functions that have no effects other than their
* return value. As a result, calls to it can be dead code eliminated.
*/
#ifdef HAVE_FUNC_ATTRIBUTE_PURE
#define ATTRIBUTE_PURE __attribute__((__pure__))
#else
#define ATTRIBUTE_PURE
#endif
#ifdef HAVE_FUNC_ATTRIBUTE_RETURNS_NONNULL
#define ATTRIBUTE_RETURNS_NONNULL __attribute__((__returns_nonnull__))
#else
#define ATTRIBUTE_RETURNS_NONNULL
#endif
#ifndef NORETURN
# ifdef _MSC_VER
# define NORETURN __declspec(noreturn)
# elif defined HAVE_FUNC_ATTRIBUTE_NORETURN
# define NORETURN __attribute__((__noreturn__))
# else
# define NORETURN
# endif
#endif
#ifdef __cplusplus
/**
* Macro function that evaluates to true if T is a trivially
* destructible type -- that is, if its (non-virtual) destructor
* performs no action and all member variables and base classes are
* trivially destructible themselves.
*/
# if (defined(__clang__) && defined(__has_feature))
# if __has_feature(has_trivial_destructor)
# define HAS_TRIVIAL_DESTRUCTOR(T) __has_trivial_destructor(T)
# endif
# elif defined(__GNUC__)
# if ((__GNUC__ > 4) || ((__GNUC__ == 4) && (__GNUC_MINOR__ >= 3)))
# define HAS_TRIVIAL_DESTRUCTOR(T) __has_trivial_destructor(T)
# endif
# elif defined(_MSC_VER) && !defined(__INTEL_COMPILER)
# define HAS_TRIVIAL_DESTRUCTOR(T) __has_trivial_destructor(T)
# endif
# ifndef HAS_TRIVIAL_DESTRUCTOR
/* It's always safe (if inefficient) to assume that a
* destructor is non-trivial.
*/
# define HAS_TRIVIAL_DESTRUCTOR(T) (false)
# endif
#endif
/**
* PUBLIC/USED macros
*
* If we build the library with gcc's -fvisibility=hidden flag, we'll
* use the PUBLIC macro to mark functions that are to be exported.
*
* We also need to define a USED attribute, so the optimizer doesn't
* inline a static function that we later use in an alias. - ajax
*/
#ifndef PUBLIC
# if defined(__GNUC__)
# define PUBLIC __attribute__((visibility("default")))
# define USED __attribute__((used))
# elif defined(_MSC_VER)
# define PUBLIC __declspec(dllexport)
# define USED
# else
# define PUBLIC
# define USED
# endif
#endif
/**
* UNUSED marks variables (or sometimes functions) that have to be defined,
* but are sometimes (or always) unused beyond that. A common case is for
* a function parameter to be used in some build configurations but not others.
* Another case is fallback vfuncs that don't do anything with their params.
*
* Note that this should not be used for identifiers used in `assert()`;
* see ASSERTED below.
*/
#ifdef HAVE_FUNC_ATTRIBUTE_UNUSED
#define UNUSED __attribute__((unused))
#else
#define UNUSED
#endif
/**
* Use ASSERTED to indicate that an identifier is unused outside of an `assert()`,
* so that assert-free builds don't get "unused variable" warnings.
*/
#ifdef NDEBUG
#define ASSERTED UNUSED
#else
#define ASSERTED
#endif
#ifdef HAVE_FUNC_ATTRIBUTE_WARN_UNUSED_RESULT
#define MUST_CHECK __attribute__((warn_unused_result))
#else
#define MUST_CHECK
#endif
#if defined(__GNUC__)
#define ATTRIBUTE_NOINLINE __attribute__((noinline))
#else
#define ATTRIBUTE_NOINLINE
#endif
/**
* Check that STRUCT::FIELD can hold MAXVAL. We use a lot of bitfields
* in Mesa/gallium. We have to be sure they're of sufficient size to
* hold the largest expected value.
* Note that with MSVC, enums are signed and enum bitfields need one extra
* high bit (always zero) to ensure the max value is handled correctly.
* This macro will detect that with MSVC, but not GCC.
*/
#define ASSERT_BITFIELD_SIZE(STRUCT, FIELD, MAXVAL) \
do { \
ASSERTED STRUCT s; \
s.FIELD = (MAXVAL); \
assert((int) s.FIELD == (MAXVAL) && "Insufficient bitfield size!"); \
} while (0)
/** Compute ceiling of integer quotient of A divided by B. */
#define DIV_ROUND_UP( A, B ) ( ((A) + (B) - 1) / (B) )
/** Clamp X to [MIN,MAX]. Turn NaN into MIN, arbitrarily. */
#define CLAMP( X, MIN, MAX ) ( (X)>(MIN) ? ((X)>(MAX) ? (MAX) : (X)) : (MIN) )
/** Minimum of two values: */
#define MIN2( A, B ) ( (A)<(B) ? (A) : (B) )
/** Maximum of two values: */
#define MAX2( A, B ) ( (A)>(B) ? (A) : (B) )
/** Minimum and maximum of three values: */
#define MIN3( A, B, C ) ((A) < (B) ? MIN2(A, C) : MIN2(B, C))
#define MAX3( A, B, C ) ((A) > (B) ? MAX2(A, C) : MAX2(B, C))
/** Align a value to a power of two */
#define ALIGN_POT(x, pot_align) (((x) + (pot_align) - 1) & ~((pot_align) - 1))
/**
* Macro for declaring an explicit conversion operator. Defaults to an
* implicit conversion if C++11 is not supported.
*/
#if __cplusplus >= 201103L
#define EXPLICIT_CONVERSION explicit
#elif defined(__cplusplus)
#define EXPLICIT_CONVERSION
#endif
/** Set a single bit */
#define BITFIELD_BIT(b) (1u << (b))
/** Set all bits up to excluding bit b */
#define BITFIELD_MASK(b) \
((b) == 32 ? (~0u) : BITFIELD_BIT((b) % 32) - 1)
/** Set count bits starting from bit b */
#define BITFIELD_RANGE(b, count) \
(BITFIELD_MASK((b) + (count)) & ~BITFIELD_MASK(b))
/** Set a single bit */
#define BITFIELD64_BIT(b) (1ull << (b))
/** Set all bits up to excluding bit b */
#define BITFIELD64_MASK(b) \
((b) == 64 ? (~0ull) : BITFIELD64_BIT(b) - 1)
/** Set count bits starting from bit b */
#define BITFIELD64_RANGE(b, count) \
(BITFIELD64_MASK((b) + (count)) & ~BITFIELD64_MASK(b))
/* TODO: In future we should try to move this to u_debug.h once header
* dependencies are reorganised to allow this.
*/
enum pipe_debug_type
{
PIPE_DEBUG_TYPE_OUT_OF_MEMORY = 1,
PIPE_DEBUG_TYPE_ERROR,
PIPE_DEBUG_TYPE_SHADER_INFO,
PIPE_DEBUG_TYPE_PERF_INFO,
PIPE_DEBUG_TYPE_INFO,
PIPE_DEBUG_TYPE_FALLBACK,
PIPE_DEBUG_TYPE_CONFORMANCE,
};
#endif /* UTIL_MACROS_H */