summaryrefslogtreecommitdiff
path: root/thirdparty
diff options
context:
space:
mode:
Diffstat (limited to 'thirdparty')
-rw-r--r--thirdparty/README.md10
-rw-r--r--thirdparty/misc/open-simplex-noise-LICENSE25
-rw-r--r--thirdparty/misc/open-simplex-noise-no-allocate.patch133
-rw-r--r--thirdparty/misc/open-simplex-noise.c2255
-rw-r--r--thirdparty/misc/open-simplex-noise.h58
-rw-r--r--thirdparty/noise/FastNoise-LICENSE22
-rw-r--r--thirdparty/noise/FastNoiseLite.h2589
-rw-r--r--thirdparty/noise/patches/FastNoiseLite.patch18
-rw-r--r--thirdparty/vulkan/patches/VMA-use-volk.patch2
-rw-r--r--thirdparty/vulkan/vk_mem_alloc.h1744
10 files changed, 3811 insertions, 3045 deletions
diff --git a/thirdparty/README.md b/thirdparty/README.md
index ced1165d69..9cf86dda25 100644
--- a/thirdparty/README.md
+++ b/thirdparty/README.md
@@ -424,10 +424,10 @@ Collection of single-file libraries used in Godot components.
* Upstream: https://archive.blender.org/wiki/index.php/Dev:Shading/Tangent_Space_Normal_Maps/
* Version: 1.0 (2011)
* License: zlib
-- `open-simplex-noise.{c,h}`
- * Upstream: https://github.com/smcameron/open-simplex-noise-in-c
- * Version: git (826f1dd1724e6fb3ff45f58e48c0fbae864c3403, 2020) + custom changes
- * License: Public Domain or Unlicense
+- `FastNoiseLite.h}`
+ * Upstream: https://github.com/Auburn/FastNoiseLite
+ * Version: git (6be3d6bf7fb408de341285f9ee8a29b67fd953f1, 2022) + custom changes
+ * License: MIT
- `pcg.{cpp,h}`
* Upstream: http://www.pcg-random.org
* Version: minimal C implementation, http://www.pcg-random.org/download.html
@@ -688,7 +688,7 @@ Files extracted from upstream source:
SDK release: https://github.com/KhronosGroup/Vulkan-ValidationLayers/blob/master/layers/generated/vk_enum_string_helper.h
`vk_mem_alloc.h` is taken from https://github.com/GPUOpen-LibrariesAndSDKs/VulkanMemoryAllocator
-Version: 3.0.0-development (2022-02-24), commit `dc3f6bb9159df22ceed69c7765ddfb4fbb1b6ed0`
+Version: 3.0.1-development (2022-03-28), commit `5b598e0a359381d7e2a94149210a1b7642024ae5`
`vk_mem_alloc.cpp` is a Godot file and should be preserved on updates.
Patches in the `patches` directory should be re-applied after updates.
diff --git a/thirdparty/misc/open-simplex-noise-LICENSE b/thirdparty/misc/open-simplex-noise-LICENSE
deleted file mode 100644
index a84c395662..0000000000
--- a/thirdparty/misc/open-simplex-noise-LICENSE
+++ /dev/null
@@ -1,25 +0,0 @@
-This is free and unencumbered software released into the public domain.
-
-Anyone is free to copy, modify, publish, use, compile, sell, or
-distribute this software, either in source code form or as a compiled
-binary, for any purpose, commercial or non-commercial, and by any
-means.
-
-In jurisdictions that recognize copyright laws, the author or authors
-of this software dedicate any and all copyright interest in the
-software to the public domain. We make this dedication for the benefit
-of the public at large and to the detriment of our heirs and
-successors. We intend this dedication to be an overt act of
-relinquishment in perpetuity of all present and future rights to this
-software under copyright law.
-
-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 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.
-
-For more information, please refer to <http://unlicense.org>
-
diff --git a/thirdparty/misc/open-simplex-noise-no-allocate.patch b/thirdparty/misc/open-simplex-noise-no-allocate.patch
deleted file mode 100644
index fc3abe7d00..0000000000
--- a/thirdparty/misc/open-simplex-noise-no-allocate.patch
+++ /dev/null
@@ -1,133 +0,0 @@
-diff -u orig/open-simplex-noise.c misc/open-simplex-noise.c
---- orig/open-simplex-noise.c 2018-09-14 11:11:40.049810000 +0200
-+++ misc/open-simplex-noise.c 2018-09-14 11:09:39.726457000 +0200
-@@ -13,6 +13,11 @@
- * of any particular randomization library, so results
- * will be the same when ported to other languages.
- */
-+
-+// -- GODOT start --
-+// Modified to work without allocating memory, also removed some unused function.
-+// -- GODOT end --
-+
- #include <math.h>
- #include <stdlib.h>
- #include <stdint.h>
-@@ -34,11 +39,12 @@
-
- #define DEFAULT_SEED (0LL)
-
--struct osn_context {
-+// -- GODOT start --
-+/*struct osn_context {
- int16_t *perm;
- int16_t *permGradIndex3D;
--};
--
-+};*/
-+// -- GODOT end --
- #define ARRAYSIZE(x) (sizeof((x)) / sizeof((x)[0]))
-
- /*
-@@ -126,7 +132,9 @@
- int xi = (int) x;
- return x < xi ? xi - 1 : xi;
- }
--
-+
-+// -- GODOT start --
-+/*
- static int allocate_perm(struct osn_context *ctx, int nperm, int ngrad)
- {
- if (ctx->perm)
-@@ -154,18 +162,21 @@
- memcpy(ctx->perm, p, sizeof(*ctx->perm) * nelements);
-
- for (i = 0; i < 256; i++) {
-- /* Since 3D has 24 gradients, simple bitmask won't work, so precompute modulo array. */
-+ // Since 3D has 24 gradients, simple bitmask won't work, so precompute modulo array.
- ctx->permGradIndex3D[i] = (int16_t)((ctx->perm[i] % (ARRAYSIZE(gradients3D) / 3)) * 3);
- }
- return 0;
- }
-+*/
-+// -- GODOT end --
-
- /*
- * Initializes using a permutation array generated from a 64-bit seed.
- * Generates a proper permutation (i.e. doesn't merely perform N successive pair
- * swaps on a base array). Uses a simple 64-bit LCG.
- */
--int open_simplex_noise(int64_t seed, struct osn_context **ctx)
-+// -- GODOT start --
-+int open_simplex_noise(int64_t seed, struct osn_context *ctx)
- {
- int rc;
- int16_t source[256];
-@@ -174,20 +185,9 @@
- int16_t *permGradIndex3D;
- int r;
-
-- *ctx = (struct osn_context *) malloc(sizeof(**ctx));
-- if (!(*ctx))
-- return -ENOMEM;
-- (*ctx)->perm = NULL;
-- (*ctx)->permGradIndex3D = NULL;
--
-- rc = allocate_perm(*ctx, 256, 256);
-- if (rc) {
-- free(*ctx);
-- return rc;
-- }
--
-- perm = (*ctx)->perm;
-- permGradIndex3D = (*ctx)->permGradIndex3D;
-+ perm = ctx->perm;
-+ permGradIndex3D = ctx->permGradIndex3D;
-+// -- GODOT end --
-
- for (i = 0; i < 256; i++)
- source[i] = (int16_t) i;
-@@ -206,6 +206,8 @@
- return 0;
- }
-
-+// -- GODOT start --
-+/*
- void open_simplex_noise_free(struct osn_context *ctx)
- {
- if (!ctx)
-@@ -220,6 +222,8 @@
- }
- free(ctx);
- }
-+*/
-+// -- GODOT end --
-
- /* 2D OpenSimplex (Simplectic) Noise. */
- double open_simplex_noise2(struct osn_context *ctx, double x, double y)
-diff -u orig/open-simplex-noise.h misc/open-simplex-noise.h
---- orig/open-simplex-noise.h 2018-09-14 11:11:19.659807000 +0200
-+++ misc/open-simplex-noise.h 2018-09-14 11:10:05.006460000 +0200
-@@ -35,11 +35,18 @@
- extern "C" {
- #endif
-
--struct osn_context;
-+// -- GODOT start --
-+// Modified to work without allocating memory, also removed some unused function.
-
--int open_simplex_noise(int64_t seed, struct osn_context **ctx);
-+struct osn_context {
-+ int16_t perm[256];
-+ int16_t permGradIndex3D[256];
-+};
-+
-+int open_simplex_noise(int64_t seed, struct osn_context *ctx);
-+//int open_simplex_noise_init_perm(struct osn_context *ctx, int16_t p[], int nelements);
-+// -- GODOT end --
- void open_simplex_noise_free(struct osn_context *ctx);
--int open_simplex_noise_init_perm(struct osn_context *ctx, int16_t p[], int nelements);
- double open_simplex_noise2(struct osn_context *ctx, double x, double y);
- double open_simplex_noise3(struct osn_context *ctx, double x, double y, double z);
- double open_simplex_noise4(struct osn_context *ctx, double x, double y, double z, double w);
diff --git a/thirdparty/misc/open-simplex-noise.c b/thirdparty/misc/open-simplex-noise.c
deleted file mode 100644
index 44a072cad1..0000000000
--- a/thirdparty/misc/open-simplex-noise.c
+++ /dev/null
@@ -1,2255 +0,0 @@
-/*
- * OpenSimplex (Simplectic) Noise in C.
- * Ported by Stephen M. Cameron from Kurt Spencer's java implementation
- *
- * v1.1 (October 5, 2014)
- * - Added 2D and 4D implementations.
- * - Proper gradient sets for all dimensions, from a
- * dimensionally-generalizable scheme with an actual
- * rhyme and reason behind it.
- * - Removed default permutation array in favor of
- * default seed.
- * - Changed seed-based constructor to be independent
- * of any particular randomization library, so results
- * will be the same when ported to other languages.
- */
-
-// -- GODOT start --
-// Modified to work without allocating memory, also removed some unused function.
-// -- GODOT end --
-
-#include <math.h>
-#include <stdlib.h>
-#include <stdint.h>
-#include <string.h>
-#include <errno.h>
-
-#include "open-simplex-noise.h"
-
-#define STRETCH_CONSTANT_2D (-0.211324865405187) /* (1 / sqrt(2 + 1) - 1 ) / 2; */
-#define SQUISH_CONSTANT_2D (0.366025403784439) /* (sqrt(2 + 1) -1) / 2; */
-#define STRETCH_CONSTANT_3D (-1.0 / 6.0) /* (1 / sqrt(3 + 1) - 1) / 3; */
-#define SQUISH_CONSTANT_3D (1.0 / 3.0) /* (sqrt(3+1)-1)/3; */
-#define STRETCH_CONSTANT_4D (-0.138196601125011) /* (1 / sqrt(4 + 1) - 1) / 4; */
-#define SQUISH_CONSTANT_4D (0.309016994374947) /* (sqrt(4 + 1) - 1) / 4; */
-
-#define NORM_CONSTANT_2D (47.0)
-#define NORM_CONSTANT_3D (103.0)
-#define NORM_CONSTANT_4D (30.0)
-
-#define DEFAULT_SEED (0LL)
-
-// -- GODOT start --
-/*struct osn_context {
- int16_t *perm;
- int16_t *permGradIndex3D;
-};*/
-// -- GODOT end --
-#define ARRAYSIZE(x) (sizeof((x)) / sizeof((x)[0]))
-
-/*
- * Gradients for 2D. They approximate the directions to the
- * vertices of an octagon from the center.
- */
-static const int8_t gradients2D[] = {
- 5, 2, 2, 5,
- -5, 2, -2, 5,
- 5, -2, 2, -5,
- -5, -2, -2, -5,
-};
-
-/*
- * Gradients for 3D. They approximate the directions to the
- * vertices of a rhombicuboctahedron from the center, skewed so
- * that the triangular and square facets can be inscribed inside
- * circles of the same radius.
- */
-static const signed char gradients3D[] = {
- -11, 4, 4, -4, 11, 4, -4, 4, 11,
- 11, 4, 4, 4, 11, 4, 4, 4, 11,
- -11, -4, 4, -4, -11, 4, -4, -4, 11,
- 11, -4, 4, 4, -11, 4, 4, -4, 11,
- -11, 4, -4, -4, 11, -4, -4, 4, -11,
- 11, 4, -4, 4, 11, -4, 4, 4, -11,
- -11, -4, -4, -4, -11, -4, -4, -4, -11,
- 11, -4, -4, 4, -11, -4, 4, -4, -11,
-};
-
-/*
- * Gradients for 4D. They approximate the directions to the
- * vertices of a disprismatotesseractihexadecachoron from the center,
- * skewed so that the tetrahedral and cubic facets can be inscribed inside
- * spheres of the same radius.
- */
-static const signed char gradients4D[] = {
- 3, 1, 1, 1, 1, 3, 1, 1, 1, 1, 3, 1, 1, 1, 1, 3,
- -3, 1, 1, 1, -1, 3, 1, 1, -1, 1, 3, 1, -1, 1, 1, 3,
- 3, -1, 1, 1, 1, -3, 1, 1, 1, -1, 3, 1, 1, -1, 1, 3,
- -3, -1, 1, 1, -1, -3, 1, 1, -1, -1, 3, 1, -1, -1, 1, 3,
- 3, 1, -1, 1, 1, 3, -1, 1, 1, 1, -3, 1, 1, 1, -1, 3,
- -3, 1, -1, 1, -1, 3, -1, 1, -1, 1, -3, 1, -1, 1, -1, 3,
- 3, -1, -1, 1, 1, -3, -1, 1, 1, -1, -3, 1, 1, -1, -1, 3,
- -3, -1, -1, 1, -1, -3, -1, 1, -1, -1, -3, 1, -1, -1, -1, 3,
- 3, 1, 1, -1, 1, 3, 1, -1, 1, 1, 3, -1, 1, 1, 1, -3,
- -3, 1, 1, -1, -1, 3, 1, -1, -1, 1, 3, -1, -1, 1, 1, -3,
- 3, -1, 1, -1, 1, -3, 1, -1, 1, -1, 3, -1, 1, -1, 1, -3,
- -3, -1, 1, -1, -1, -3, 1, -1, -1, -1, 3, -1, -1, -1, 1, -3,
- 3, 1, -1, -1, 1, 3, -1, -1, 1, 1, -3, -1, 1, 1, -1, -3,
- -3, 1, -1, -1, -1, 3, -1, -1, -1, 1, -3, -1, -1, 1, -1, -3,
- 3, -1, -1, -1, 1, -3, -1, -1, 1, -1, -3, -1, 1, -1, -1, -3,
- -3, -1, -1, -1, -1, -3, -1, -1, -1, -1, -3, -1, -1, -1, -1, -3,
-};
-
-static double extrapolate2(const struct osn_context *ctx, int xsb, int ysb, double dx, double dy)
-{
- const int16_t *perm = ctx->perm;
- int index = perm[(perm[xsb & 0xFF] + ysb) & 0xFF] & 0x0E;
- return gradients2D[index] * dx
- + gradients2D[index + 1] * dy;
-}
-
-static double extrapolate3(const struct osn_context *ctx, int xsb, int ysb, int zsb, double dx, double dy, double dz)
-{
- const int16_t *perm = ctx->perm;
- const int16_t *permGradIndex3D = ctx->permGradIndex3D;
- int index = permGradIndex3D[(perm[(perm[xsb & 0xFF] + ysb) & 0xFF] + zsb) & 0xFF];
- return gradients3D[index] * dx
- + gradients3D[index + 1] * dy
- + gradients3D[index + 2] * dz;
-}
-
-static double extrapolate4(const struct osn_context *ctx, int xsb, int ysb, int zsb, int wsb, double dx, double dy, double dz, double dw)
-{
- const int16_t *perm = ctx->perm;
- int index = perm[(perm[(perm[(perm[xsb & 0xFF] + ysb) & 0xFF] + zsb) & 0xFF] + wsb) & 0xFF] & 0xFC;
- return gradients4D[index] * dx
- + gradients4D[index + 1] * dy
- + gradients4D[index + 2] * dz
- + gradients4D[index + 3] * dw;
-}
-
-static INLINE int fastFloor(double x) {
- int xi = (int) x;
- return x < xi ? xi - 1 : xi;
-}
-
-// -- GODOT start --
-/*
-static int allocate_perm(struct osn_context *ctx, int nperm, int ngrad)
-{
- if (ctx->perm)
- free(ctx->perm);
- if (ctx->permGradIndex3D)
- free(ctx->permGradIndex3D);
- ctx->perm = (int16_t *) malloc(sizeof(*ctx->perm) * nperm);
- if (!ctx->perm)
- return -ENOMEM;
- ctx->permGradIndex3D = (int16_t *) malloc(sizeof(*ctx->permGradIndex3D) * ngrad);
- if (!ctx->permGradIndex3D) {
- free(ctx->perm);
- return -ENOMEM;
- }
- return 0;
-}
-
-int open_simplex_noise_init_perm(struct osn_context *ctx, int16_t p[], int nelements)
-{
- int i, rc;
-
- rc = allocate_perm(ctx, nelements, 256);
- if (rc)
- return rc;
- memcpy(ctx->perm, p, sizeof(*ctx->perm) * nelements);
-
- for (i = 0; i < 256; i++) {
- // Since 3D has 24 gradients, simple bitmask won't work, so precompute modulo array.
- ctx->permGradIndex3D[i] = (int16_t)((ctx->perm[i] % (ARRAYSIZE(gradients3D) / 3)) * 3);
- }
- return 0;
-}
-*/
-// -- GODOT end --
-
-/*
- * Initializes using a permutation array generated from a 64-bit seed.
- * Generates a proper permutation (i.e. doesn't merely perform N successive pair
- * swaps on a base array). Uses a simple 64-bit LCG.
- */
-// -- GODOT start --
-int open_simplex_noise(int64_t seed, struct osn_context *ctx)
-{
- int rc;
- int16_t source[256];
- int i;
- int16_t *perm;
- int16_t *permGradIndex3D;
- int r;
-
- perm = ctx->perm;
- permGradIndex3D = ctx->permGradIndex3D;
-// -- GODOT end --
-
- uint64_t seedU = seed;
- for (i = 0; i < 256; i++)
- source[i] = (int16_t) i;
- seedU = seedU * 6364136223846793005ULL + 1442695040888963407ULL;
- seedU = seedU * 6364136223846793005ULL + 1442695040888963407ULL;
- seedU = seedU * 6364136223846793005ULL + 1442695040888963407ULL;
- for (i = 255; i >= 0; i--) {
- seedU = seedU * 6364136223846793005ULL + 1442695040888963407ULL;
- r = (int)((seedU + 31) % (i + 1));
- if (r < 0)
- r += (i + 1);
- perm[i] = source[r];
- permGradIndex3D[i] = (short)((perm[i] % (ARRAYSIZE(gradients3D) / 3)) * 3);
- source[r] = source[i];
- }
- return 0;
-}
-
-// -- GODOT start --
-/*
-void open_simplex_noise_free(struct osn_context *ctx)
-{
- if (!ctx)
- return;
- if (ctx->perm) {
- free(ctx->perm);
- ctx->perm = NULL;
- }
- if (ctx->permGradIndex3D) {
- free(ctx->permGradIndex3D);
- ctx->permGradIndex3D = NULL;
- }
- free(ctx);
-}
-*/
-// -- GODOT end --
-
-/* 2D OpenSimplex (Simplectic) Noise. */
-double open_simplex_noise2(const struct osn_context *ctx, double x, double y)
-{
-
- /* Place input coordinates onto grid. */
- double stretchOffset = (x + y) * STRETCH_CONSTANT_2D;
- double xs = x + stretchOffset;
- double ys = y + stretchOffset;
-
- /* Floor to get grid coordinates of rhombus (stretched square) super-cell origin. */
- int xsb = fastFloor(xs);
- int ysb = fastFloor(ys);
-
- /* Skew out to get actual coordinates of rhombus origin. We'll need these later. */
- double squishOffset = (xsb + ysb) * SQUISH_CONSTANT_2D;
- double xb = xsb + squishOffset;
- double yb = ysb + squishOffset;
-
- /* Compute grid coordinates relative to rhombus origin. */
- double xins = xs - xsb;
- double yins = ys - ysb;
-
- /* Sum those together to get a value that determines which region we're in. */
- double inSum = xins + yins;
-
- /* Positions relative to origin point. */
- double dx0 = x - xb;
- double dy0 = y - yb;
-
- /* We'll be defining these inside the next block and using them afterwards. */
- double dx_ext, dy_ext;
- int xsv_ext, ysv_ext;
-
- double dx1;
- double dy1;
- double attn1;
- double dx2;
- double dy2;
- double attn2;
- double zins;
- double attn0;
- double attn_ext;
-
- double value = 0;
-
- /* Contribution (1,0) */
- dx1 = dx0 - 1 - SQUISH_CONSTANT_2D;
- dy1 = dy0 - 0 - SQUISH_CONSTANT_2D;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate2(ctx, xsb + 1, ysb + 0, dx1, dy1);
- }
-
- /* Contribution (0,1) */
- dx2 = dx0 - 0 - SQUISH_CONSTANT_2D;
- dy2 = dy0 - 1 - SQUISH_CONSTANT_2D;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate2(ctx, xsb + 0, ysb + 1, dx2, dy2);
- }
-
- if (inSum <= 1) { /* We're inside the triangle (2-Simplex) at (0,0) */
- zins = 1 - inSum;
- if (zins > xins || zins > yins) { /* (0,0) is one of the closest two triangular vertices */
- if (xins > yins) {
- xsv_ext = xsb + 1;
- ysv_ext = ysb - 1;
- dx_ext = dx0 - 1;
- dy_ext = dy0 + 1;
- } else {
- xsv_ext = xsb - 1;
- ysv_ext = ysb + 1;
- dx_ext = dx0 + 1;
- dy_ext = dy0 - 1;
- }
- } else { /* (1,0) and (0,1) are the closest two vertices. */
- xsv_ext = xsb + 1;
- ysv_ext = ysb + 1;
- dx_ext = dx0 - 1 - 2 * SQUISH_CONSTANT_2D;
- dy_ext = dy0 - 1 - 2 * SQUISH_CONSTANT_2D;
- }
- } else { /* We're inside the triangle (2-Simplex) at (1,1) */
- zins = 2 - inSum;
- if (zins < xins || zins < yins) { /* (0,0) is one of the closest two triangular vertices */
- if (xins > yins) {
- xsv_ext = xsb + 2;
- ysv_ext = ysb + 0;
- dx_ext = dx0 - 2 - 2 * SQUISH_CONSTANT_2D;
- dy_ext = dy0 + 0 - 2 * SQUISH_CONSTANT_2D;
- } else {
- xsv_ext = xsb + 0;
- ysv_ext = ysb + 2;
- dx_ext = dx0 + 0 - 2 * SQUISH_CONSTANT_2D;
- dy_ext = dy0 - 2 - 2 * SQUISH_CONSTANT_2D;
- }
- } else { /* (1,0) and (0,1) are the closest two vertices. */
- dx_ext = dx0;
- dy_ext = dy0;
- xsv_ext = xsb;
- ysv_ext = ysb;
- }
- xsb += 1;
- ysb += 1;
- dx0 = dx0 - 1 - 2 * SQUISH_CONSTANT_2D;
- dy0 = dy0 - 1 - 2 * SQUISH_CONSTANT_2D;
- }
-
- /* Contribution (0,0) or (1,1) */
- attn0 = 2 - dx0 * dx0 - dy0 * dy0;
- if (attn0 > 0) {
- attn0 *= attn0;
- value += attn0 * attn0 * extrapolate2(ctx, xsb, ysb, dx0, dy0);
- }
-
- /* Extra Vertex */
- attn_ext = 2 - dx_ext * dx_ext - dy_ext * dy_ext;
- if (attn_ext > 0) {
- attn_ext *= attn_ext;
- value += attn_ext * attn_ext * extrapolate2(ctx, xsv_ext, ysv_ext, dx_ext, dy_ext);
- }
-
- return value / NORM_CONSTANT_2D;
-}
-
-/*
- * 3D OpenSimplex (Simplectic) Noise
- */
-double open_simplex_noise3(const struct osn_context *ctx, double x, double y, double z)
-{
-
- /* Place input coordinates on simplectic honeycomb. */
- double stretchOffset = (x + y + z) * STRETCH_CONSTANT_3D;
- double xs = x + stretchOffset;
- double ys = y + stretchOffset;
- double zs = z + stretchOffset;
-
- /* Floor to get simplectic honeycomb coordinates of rhombohedron (stretched cube) super-cell origin. */
- int xsb = fastFloor(xs);
- int ysb = fastFloor(ys);
- int zsb = fastFloor(zs);
-
- /* Skew out to get actual coordinates of rhombohedron origin. We'll need these later. */
- double squishOffset = (xsb + ysb + zsb) * SQUISH_CONSTANT_3D;
- double xb = xsb + squishOffset;
- double yb = ysb + squishOffset;
- double zb = zsb + squishOffset;
-
- /* Compute simplectic honeycomb coordinates relative to rhombohedral origin. */
- double xins = xs - xsb;
- double yins = ys - ysb;
- double zins = zs - zsb;
-
- /* Sum those together to get a value that determines which region we're in. */
- double inSum = xins + yins + zins;
-
- /* Positions relative to origin point. */
- double dx0 = x - xb;
- double dy0 = y - yb;
- double dz0 = z - zb;
-
- /* We'll be defining these inside the next block and using them afterwards. */
- double dx_ext0, dy_ext0, dz_ext0;
- double dx_ext1, dy_ext1, dz_ext1;
- int xsv_ext0, ysv_ext0, zsv_ext0;
- int xsv_ext1, ysv_ext1, zsv_ext1;
-
- double wins;
- int8_t c, c1, c2;
- int8_t aPoint, bPoint;
- double aScore, bScore;
- int aIsFurtherSide;
- int bIsFurtherSide;
- double p1, p2, p3;
- double score;
- double attn0, attn1, attn2, attn3, attn4, attn5, attn6;
- double dx1, dy1, dz1;
- double dx2, dy2, dz2;
- double dx3, dy3, dz3;
- double dx4, dy4, dz4;
- double dx5, dy5, dz5;
- double dx6, dy6, dz6;
- double attn_ext0, attn_ext1;
-
- double value = 0;
- if (inSum <= 1) { /* We're inside the tetrahedron (3-Simplex) at (0,0,0) */
-
- /* Determine which two of (0,0,1), (0,1,0), (1,0,0) are closest. */
- aPoint = 0x01;
- aScore = xins;
- bPoint = 0x02;
- bScore = yins;
- if (aScore >= bScore && zins > bScore) {
- bScore = zins;
- bPoint = 0x04;
- } else if (aScore < bScore && zins > aScore) {
- aScore = zins;
- aPoint = 0x04;
- }
-
- /* Now we determine the two lattice points not part of the tetrahedron that may contribute.
- This depends on the closest two tetrahedral vertices, including (0,0,0) */
- wins = 1 - inSum;
- if (wins > aScore || wins > bScore) { /* (0,0,0) is one of the closest two tetrahedral vertices. */
- c = (bScore > aScore ? bPoint : aPoint); /* Our other closest vertex is the closest out of a and b. */
-
- if ((c & 0x01) == 0) {
- xsv_ext0 = xsb - 1;
- xsv_ext1 = xsb;
- dx_ext0 = dx0 + 1;
- dx_ext1 = dx0;
- } else {
- xsv_ext0 = xsv_ext1 = xsb + 1;
- dx_ext0 = dx_ext1 = dx0 - 1;
- }
-
- if ((c & 0x02) == 0) {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0;
- if ((c & 0x01) == 0) {
- ysv_ext1 -= 1;
- dy_ext1 += 1;
- } else {
- ysv_ext0 -= 1;
- dy_ext0 += 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1;
- }
-
- if ((c & 0x04) == 0) {
- zsv_ext0 = zsb;
- zsv_ext1 = zsb - 1;
- dz_ext0 = dz0;
- dz_ext1 = dz0 + 1;
- } else {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz_ext1 = dz0 - 1;
- }
- } else { /* (0,0,0) is not one of the closest two tetrahedral vertices. */
- c = (int8_t)(aPoint | bPoint); /* Our two extra vertices are determined by the closest two. */
-
- if ((c & 0x01) == 0) {
- xsv_ext0 = xsb;
- xsv_ext1 = xsb - 1;
- dx_ext0 = dx0 - 2 * SQUISH_CONSTANT_3D;
- dx_ext1 = dx0 + 1 - SQUISH_CONSTANT_3D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb + 1;
- dx_ext0 = dx0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dx_ext1 = dx0 - 1 - SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x02) == 0) {
- ysv_ext0 = ysb;
- ysv_ext1 = ysb - 1;
- dy_ext0 = dy0 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 + 1 - SQUISH_CONSTANT_3D;
- } else {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 1 - SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x04) == 0) {
- zsv_ext0 = zsb;
- zsv_ext1 = zsb - 1;
- dz_ext0 = dz0 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 + 1 - SQUISH_CONSTANT_3D;
- } else {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 1 - SQUISH_CONSTANT_3D;
- }
- }
-
- /* Contribution (0,0,0) */
- attn0 = 2 - dx0 * dx0 - dy0 * dy0 - dz0 * dz0;
- if (attn0 > 0) {
- attn0 *= attn0;
- value += attn0 * attn0 * extrapolate3(ctx, xsb + 0, ysb + 0, zsb + 0, dx0, dy0, dz0);
- }
-
- /* Contribution (1,0,0) */
- dx1 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy1 = dy0 - 0 - SQUISH_CONSTANT_3D;
- dz1 = dz0 - 0 - SQUISH_CONSTANT_3D;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate3(ctx, xsb + 1, ysb + 0, zsb + 0, dx1, dy1, dz1);
- }
-
- /* Contribution (0,1,0) */
- dx2 = dx0 - 0 - SQUISH_CONSTANT_3D;
- dy2 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz2 = dz1;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate3(ctx, xsb + 0, ysb + 1, zsb + 0, dx2, dy2, dz2);
- }
-
- /* Contribution (0,0,1) */
- dx3 = dx2;
- dy3 = dy1;
- dz3 = dz0 - 1 - SQUISH_CONSTANT_3D;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate3(ctx, xsb + 0, ysb + 0, zsb + 1, dx3, dy3, dz3);
- }
- } else if (inSum >= 2) { /* We're inside the tetrahedron (3-Simplex) at (1,1,1) */
-
- /* Determine which two tetrahedral vertices are the closest, out of (1,1,0), (1,0,1), (0,1,1) but not (1,1,1). */
- aPoint = 0x06;
- aScore = xins;
- bPoint = 0x05;
- bScore = yins;
- if (aScore <= bScore && zins < bScore) {
- bScore = zins;
- bPoint = 0x03;
- } else if (aScore > bScore && zins < aScore) {
- aScore = zins;
- aPoint = 0x03;
- }
-
- /* Now we determine the two lattice points not part of the tetrahedron that may contribute.
- This depends on the closest two tetrahedral vertices, including (1,1,1) */
- wins = 3 - inSum;
- if (wins < aScore || wins < bScore) { /* (1,1,1) is one of the closest two tetrahedral vertices. */
- c = (bScore < aScore ? bPoint : aPoint); /* Our other closest vertex is the closest out of a and b. */
-
- if ((c & 0x01) != 0) {
- xsv_ext0 = xsb + 2;
- xsv_ext1 = xsb + 1;
- dx_ext0 = dx0 - 2 - 3 * SQUISH_CONSTANT_3D;
- dx_ext1 = dx0 - 1 - 3 * SQUISH_CONSTANT_3D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb;
- dx_ext0 = dx_ext1 = dx0 - 3 * SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x02) != 0) {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1 - 3 * SQUISH_CONSTANT_3D;
- if ((c & 0x01) != 0) {
- ysv_ext1 += 1;
- dy_ext1 -= 1;
- } else {
- ysv_ext0 += 1;
- dy_ext0 -= 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0 - 3 * SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x04) != 0) {
- zsv_ext0 = zsb + 1;
- zsv_ext1 = zsb + 2;
- dz_ext0 = dz0 - 1 - 3 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 - 3 * SQUISH_CONSTANT_3D;
- } else {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz_ext1 = dz0 - 3 * SQUISH_CONSTANT_3D;
- }
- } else { /* (1,1,1) is not one of the closest two tetrahedral vertices. */
- c = (int8_t)(aPoint & bPoint); /* Our two extra vertices are determined by the closest two. */
-
- if ((c & 0x01) != 0) {
- xsv_ext0 = xsb + 1;
- xsv_ext1 = xsb + 2;
- dx_ext0 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dx_ext1 = dx0 - 2 - 2 * SQUISH_CONSTANT_3D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb;
- dx_ext0 = dx0 - SQUISH_CONSTANT_3D;
- dx_ext1 = dx0 - 2 * SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x02) != 0) {
- ysv_ext0 = ysb + 1;
- ysv_ext1 = ysb + 2;
- dy_ext0 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 - 2 * SQUISH_CONSTANT_3D;
- } else {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy0 - SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 * SQUISH_CONSTANT_3D;
- }
-
- if ((c & 0x04) != 0) {
- zsv_ext0 = zsb + 1;
- zsv_ext1 = zsb + 2;
- dz_ext0 = dz0 - 1 - SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 - 2 * SQUISH_CONSTANT_3D;
- } else {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz0 - SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 * SQUISH_CONSTANT_3D;
- }
- }
-
- /* Contribution (1,1,0) */
- dx3 = dx0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dy3 = dy0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dz3 = dz0 - 0 - 2 * SQUISH_CONSTANT_3D;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate3(ctx, xsb + 1, ysb + 1, zsb + 0, dx3, dy3, dz3);
- }
-
- /* Contribution (1,0,1) */
- dx2 = dx3;
- dy2 = dy0 - 0 - 2 * SQUISH_CONSTANT_3D;
- dz2 = dz0 - 1 - 2 * SQUISH_CONSTANT_3D;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate3(ctx, xsb + 1, ysb + 0, zsb + 1, dx2, dy2, dz2);
- }
-
- /* Contribution (0,1,1) */
- dx1 = dx0 - 0 - 2 * SQUISH_CONSTANT_3D;
- dy1 = dy3;
- dz1 = dz2;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate3(ctx, xsb + 0, ysb + 1, zsb + 1, dx1, dy1, dz1);
- }
-
- /* Contribution (1,1,1) */
- dx0 = dx0 - 1 - 3 * SQUISH_CONSTANT_3D;
- dy0 = dy0 - 1 - 3 * SQUISH_CONSTANT_3D;
- dz0 = dz0 - 1 - 3 * SQUISH_CONSTANT_3D;
- attn0 = 2 - dx0 * dx0 - dy0 * dy0 - dz0 * dz0;
- if (attn0 > 0) {
- attn0 *= attn0;
- value += attn0 * attn0 * extrapolate3(ctx, xsb + 1, ysb + 1, zsb + 1, dx0, dy0, dz0);
- }
- } else { /* We're inside the octahedron (Rectified 3-Simplex) in between.
- Decide between point (0,0,1) and (1,1,0) as closest */
- p1 = xins + yins;
- if (p1 > 1) {
- aScore = p1 - 1;
- aPoint = 0x03;
- aIsFurtherSide = 1;
- } else {
- aScore = 1 - p1;
- aPoint = 0x04;
- aIsFurtherSide = 0;
- }
-
- /* Decide between point (0,1,0) and (1,0,1) as closest */
- p2 = xins + zins;
- if (p2 > 1) {
- bScore = p2 - 1;
- bPoint = 0x05;
- bIsFurtherSide = 1;
- } else {
- bScore = 1 - p2;
- bPoint = 0x02;
- bIsFurtherSide = 0;
- }
-
- /* The closest out of the two (1,0,0) and (0,1,1) will replace the furthest out of the two decided above, if closer. */
- p3 = yins + zins;
- if (p3 > 1) {
- score = p3 - 1;
- if (aScore <= bScore && aScore < score) {
- aScore = score;
- aPoint = 0x06;
- aIsFurtherSide = 1;
- } else if (aScore > bScore && bScore < score) {
- bScore = score;
- bPoint = 0x06;
- bIsFurtherSide = 1;
- }
- } else {
- score = 1 - p3;
- if (aScore <= bScore && aScore < score) {
- aScore = score;
- aPoint = 0x01;
- aIsFurtherSide = 0;
- } else if (aScore > bScore && bScore < score) {
- bScore = score;
- bPoint = 0x01;
- bIsFurtherSide = 0;
- }
- }
-
- /* Where each of the two closest points are determines how the extra two vertices are calculated. */
- if (aIsFurtherSide == bIsFurtherSide) {
- if (aIsFurtherSide) { /* Both closest points on (1,1,1) side */
-
- /* One of the two extra points is (1,1,1) */
- dx_ext0 = dx0 - 1 - 3 * SQUISH_CONSTANT_3D;
- dy_ext0 = dy0 - 1 - 3 * SQUISH_CONSTANT_3D;
- dz_ext0 = dz0 - 1 - 3 * SQUISH_CONSTANT_3D;
- xsv_ext0 = xsb + 1;
- ysv_ext0 = ysb + 1;
- zsv_ext0 = zsb + 1;
-
- /* Other extra point is based on the shared axis. */
- c = (int8_t)(aPoint & bPoint);
- if ((c & 0x01) != 0) {
- dx_ext1 = dx0 - 2 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 * SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb + 2;
- ysv_ext1 = ysb;
- zsv_ext1 = zsb;
- } else if ((c & 0x02) != 0) {
- dx_ext1 = dx0 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 * SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb;
- ysv_ext1 = ysb + 2;
- zsv_ext1 = zsb;
- } else {
- dx_ext1 = dx0 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 - 2 * SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb;
- ysv_ext1 = ysb;
- zsv_ext1 = zsb + 2;
- }
- } else { /* Both closest points on (0,0,0) side */
-
- /* One of the two extra points is (0,0,0) */
- dx_ext0 = dx0;
- dy_ext0 = dy0;
- dz_ext0 = dz0;
- xsv_ext0 = xsb;
- ysv_ext0 = ysb;
- zsv_ext0 = zsb;
-
- /* Other extra point is based on the omitted axis. */
- c = (int8_t)(aPoint | bPoint);
- if ((c & 0x01) == 0) {
- dx_ext1 = dx0 + 1 - SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 1 - SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb - 1;
- ysv_ext1 = ysb + 1;
- zsv_ext1 = zsb + 1;
- } else if ((c & 0x02) == 0) {
- dx_ext1 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 + 1 - SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 1 - SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb + 1;
- ysv_ext1 = ysb - 1;
- zsv_ext1 = zsb + 1;
- } else {
- dx_ext1 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 + 1 - SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb + 1;
- ysv_ext1 = ysb + 1;
- zsv_ext1 = zsb - 1;
- }
- }
- } else { /* One point on (0,0,0) side, one point on (1,1,1) side */
- if (aIsFurtherSide) {
- c1 = aPoint;
- c2 = bPoint;
- } else {
- c1 = bPoint;
- c2 = aPoint;
- }
-
- /* One contribution is a permutation of (1,1,-1) */
- if ((c1 & 0x01) == 0) {
- dx_ext0 = dx0 + 1 - SQUISH_CONSTANT_3D;
- dy_ext0 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz_ext0 = dz0 - 1 - SQUISH_CONSTANT_3D;
- xsv_ext0 = xsb - 1;
- ysv_ext0 = ysb + 1;
- zsv_ext0 = zsb + 1;
- } else if ((c1 & 0x02) == 0) {
- dx_ext0 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy_ext0 = dy0 + 1 - SQUISH_CONSTANT_3D;
- dz_ext0 = dz0 - 1 - SQUISH_CONSTANT_3D;
- xsv_ext0 = xsb + 1;
- ysv_ext0 = ysb - 1;
- zsv_ext0 = zsb + 1;
- } else {
- dx_ext0 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy_ext0 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz_ext0 = dz0 + 1 - SQUISH_CONSTANT_3D;
- xsv_ext0 = xsb + 1;
- ysv_ext0 = ysb + 1;
- zsv_ext0 = zsb - 1;
- }
-
- /* One contribution is a permutation of (0,0,2) */
- dx_ext1 = dx0 - 2 * SQUISH_CONSTANT_3D;
- dy_ext1 = dy0 - 2 * SQUISH_CONSTANT_3D;
- dz_ext1 = dz0 - 2 * SQUISH_CONSTANT_3D;
- xsv_ext1 = xsb;
- ysv_ext1 = ysb;
- zsv_ext1 = zsb;
- if ((c2 & 0x01) != 0) {
- dx_ext1 -= 2;
- xsv_ext1 += 2;
- } else if ((c2 & 0x02) != 0) {
- dy_ext1 -= 2;
- ysv_ext1 += 2;
- } else {
- dz_ext1 -= 2;
- zsv_ext1 += 2;
- }
- }
-
- /* Contribution (1,0,0) */
- dx1 = dx0 - 1 - SQUISH_CONSTANT_3D;
- dy1 = dy0 - 0 - SQUISH_CONSTANT_3D;
- dz1 = dz0 - 0 - SQUISH_CONSTANT_3D;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate3(ctx, xsb + 1, ysb + 0, zsb + 0, dx1, dy1, dz1);
- }
-
- /* Contribution (0,1,0) */
- dx2 = dx0 - 0 - SQUISH_CONSTANT_3D;
- dy2 = dy0 - 1 - SQUISH_CONSTANT_3D;
- dz2 = dz1;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate3(ctx, xsb + 0, ysb + 1, zsb + 0, dx2, dy2, dz2);
- }
-
- /* Contribution (0,0,1) */
- dx3 = dx2;
- dy3 = dy1;
- dz3 = dz0 - 1 - SQUISH_CONSTANT_3D;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate3(ctx, xsb + 0, ysb + 0, zsb + 1, dx3, dy3, dz3);
- }
-
- /* Contribution (1,1,0) */
- dx4 = dx0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dy4 = dy0 - 1 - 2 * SQUISH_CONSTANT_3D;
- dz4 = dz0 - 0 - 2 * SQUISH_CONSTANT_3D;
- attn4 = 2 - dx4 * dx4 - dy4 * dy4 - dz4 * dz4;
- if (attn4 > 0) {
- attn4 *= attn4;
- value += attn4 * attn4 * extrapolate3(ctx, xsb + 1, ysb + 1, zsb + 0, dx4, dy4, dz4);
- }
-
- /* Contribution (1,0,1) */
- dx5 = dx4;
- dy5 = dy0 - 0 - 2 * SQUISH_CONSTANT_3D;
- dz5 = dz0 - 1 - 2 * SQUISH_CONSTANT_3D;
- attn5 = 2 - dx5 * dx5 - dy5 * dy5 - dz5 * dz5;
- if (attn5 > 0) {
- attn5 *= attn5;
- value += attn5 * attn5 * extrapolate3(ctx, xsb + 1, ysb + 0, zsb + 1, dx5, dy5, dz5);
- }
-
- /* Contribution (0,1,1) */
- dx6 = dx0 - 0 - 2 * SQUISH_CONSTANT_3D;
- dy6 = dy4;
- dz6 = dz5;
- attn6 = 2 - dx6 * dx6 - dy6 * dy6 - dz6 * dz6;
- if (attn6 > 0) {
- attn6 *= attn6;
- value += attn6 * attn6 * extrapolate3(ctx, xsb + 0, ysb + 1, zsb + 1, dx6, dy6, dz6);
- }
- }
-
- /* First extra vertex */
- attn_ext0 = 2 - dx_ext0 * dx_ext0 - dy_ext0 * dy_ext0 - dz_ext0 * dz_ext0;
- if (attn_ext0 > 0)
- {
- attn_ext0 *= attn_ext0;
- value += attn_ext0 * attn_ext0 * extrapolate3(ctx, xsv_ext0, ysv_ext0, zsv_ext0, dx_ext0, dy_ext0, dz_ext0);
- }
-
- /* Second extra vertex */
- attn_ext1 = 2 - dx_ext1 * dx_ext1 - dy_ext1 * dy_ext1 - dz_ext1 * dz_ext1;
- if (attn_ext1 > 0)
- {
- attn_ext1 *= attn_ext1;
- value += attn_ext1 * attn_ext1 * extrapolate3(ctx, xsv_ext1, ysv_ext1, zsv_ext1, dx_ext1, dy_ext1, dz_ext1);
- }
-
- return value / NORM_CONSTANT_3D;
-}
-
-/*
- * 4D OpenSimplex (Simplectic) Noise.
- */
-double open_simplex_noise4(const struct osn_context *ctx, double x, double y, double z, double w)
-{
- double uins;
- double dx1, dy1, dz1, dw1;
- double dx2, dy2, dz2, dw2;
- double dx3, dy3, dz3, dw3;
- double dx4, dy4, dz4, dw4;
- double dx5, dy5, dz5, dw5;
- double dx6, dy6, dz6, dw6;
- double dx7, dy7, dz7, dw7;
- double dx8, dy8, dz8, dw8;
- double dx9, dy9, dz9, dw9;
- double dx10, dy10, dz10, dw10;
- double attn0, attn1, attn2, attn3, attn4;
- double attn5, attn6, attn7, attn8, attn9, attn10;
- double attn_ext0, attn_ext1, attn_ext2;
- int8_t c, c1, c2;
- int8_t aPoint, bPoint;
- double aScore, bScore;
- int aIsBiggerSide;
- int bIsBiggerSide;
- double p1, p2, p3, p4;
- double score;
-
- /* Place input coordinates on simplectic honeycomb. */
- double stretchOffset = (x + y + z + w) * STRETCH_CONSTANT_4D;
- double xs = x + stretchOffset;
- double ys = y + stretchOffset;
- double zs = z + stretchOffset;
- double ws = w + stretchOffset;
-
- /* Floor to get simplectic honeycomb coordinates of rhombo-hypercube super-cell origin. */
- int xsb = fastFloor(xs);
- int ysb = fastFloor(ys);
- int zsb = fastFloor(zs);
- int wsb = fastFloor(ws);
-
- /* Skew out to get actual coordinates of stretched rhombo-hypercube origin. We'll need these later. */
- double squishOffset = (xsb + ysb + zsb + wsb) * SQUISH_CONSTANT_4D;
- double xb = xsb + squishOffset;
- double yb = ysb + squishOffset;
- double zb = zsb + squishOffset;
- double wb = wsb + squishOffset;
-
- /* Compute simplectic honeycomb coordinates relative to rhombo-hypercube origin. */
- double xins = xs - xsb;
- double yins = ys - ysb;
- double zins = zs - zsb;
- double wins = ws - wsb;
-
- /* Sum those together to get a value that determines which region we're in. */
- double inSum = xins + yins + zins + wins;
-
- /* Positions relative to origin point. */
- double dx0 = x - xb;
- double dy0 = y - yb;
- double dz0 = z - zb;
- double dw0 = w - wb;
-
- /* We'll be defining these inside the next block and using them afterwards. */
- double dx_ext0, dy_ext0, dz_ext0, dw_ext0;
- double dx_ext1, dy_ext1, dz_ext1, dw_ext1;
- double dx_ext2, dy_ext2, dz_ext2, dw_ext2;
- int xsv_ext0, ysv_ext0, zsv_ext0, wsv_ext0;
- int xsv_ext1, ysv_ext1, zsv_ext1, wsv_ext1;
- int xsv_ext2, ysv_ext2, zsv_ext2, wsv_ext2;
-
- double value = 0;
- if (inSum <= 1) { /* We're inside the pentachoron (4-Simplex) at (0,0,0,0) */
-
- /* Determine which two of (0,0,0,1), (0,0,1,0), (0,1,0,0), (1,0,0,0) are closest. */
- aPoint = 0x01;
- aScore = xins;
- bPoint = 0x02;
- bScore = yins;
- if (aScore >= bScore && zins > bScore) {
- bScore = zins;
- bPoint = 0x04;
- } else if (aScore < bScore && zins > aScore) {
- aScore = zins;
- aPoint = 0x04;
- }
- if (aScore >= bScore && wins > bScore) {
- bScore = wins;
- bPoint = 0x08;
- } else if (aScore < bScore && wins > aScore) {
- aScore = wins;
- aPoint = 0x08;
- }
-
- /* Now we determine the three lattice points not part of the pentachoron that may contribute.
- This depends on the closest two pentachoron vertices, including (0,0,0,0) */
- uins = 1 - inSum;
- if (uins > aScore || uins > bScore) { /* (0,0,0,0) is one of the closest two pentachoron vertices. */
- c = (bScore > aScore ? bPoint : aPoint); /* Our other closest vertex is the closest out of a and b. */
- if ((c & 0x01) == 0) {
- xsv_ext0 = xsb - 1;
- xsv_ext1 = xsv_ext2 = xsb;
- dx_ext0 = dx0 + 1;
- dx_ext1 = dx_ext2 = dx0;
- } else {
- xsv_ext0 = xsv_ext1 = xsv_ext2 = xsb + 1;
- dx_ext0 = dx_ext1 = dx_ext2 = dx0 - 1;
- }
-
- if ((c & 0x02) == 0) {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb;
- dy_ext0 = dy_ext1 = dy_ext2 = dy0;
- if ((c & 0x01) == 0x01) {
- ysv_ext0 -= 1;
- dy_ext0 += 1;
- } else {
- ysv_ext1 -= 1;
- dy_ext1 += 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb + 1;
- dy_ext0 = dy_ext1 = dy_ext2 = dy0 - 1;
- }
-
- if ((c & 0x04) == 0) {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb;
- dz_ext0 = dz_ext1 = dz_ext2 = dz0;
- if ((c & 0x03) != 0) {
- if ((c & 0x03) == 0x03) {
- zsv_ext0 -= 1;
- dz_ext0 += 1;
- } else {
- zsv_ext1 -= 1;
- dz_ext1 += 1;
- }
- } else {
- zsv_ext2 -= 1;
- dz_ext2 += 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb + 1;
- dz_ext0 = dz_ext1 = dz_ext2 = dz0 - 1;
- }
-
- if ((c & 0x08) == 0) {
- wsv_ext0 = wsv_ext1 = wsb;
- wsv_ext2 = wsb - 1;
- dw_ext0 = dw_ext1 = dw0;
- dw_ext2 = dw0 + 1;
- } else {
- wsv_ext0 = wsv_ext1 = wsv_ext2 = wsb + 1;
- dw_ext0 = dw_ext1 = dw_ext2 = dw0 - 1;
- }
- } else { /* (0,0,0,0) is not one of the closest two pentachoron vertices. */
- c = (int8_t)(aPoint | bPoint); /* Our three extra vertices are determined by the closest two. */
-
- if ((c & 0x01) == 0) {
- xsv_ext0 = xsv_ext2 = xsb;
- xsv_ext1 = xsb - 1;
- dx_ext0 = dx0 - 2 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 + 1 - SQUISH_CONSTANT_4D;
- dx_ext2 = dx0 - SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsv_ext2 = xsb + 1;
- dx_ext0 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx_ext2 = dx0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x02) == 0) {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb;
- dy_ext0 = dy0 - 2 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy_ext2 = dy0 - SQUISH_CONSTANT_4D;
- if ((c & 0x01) == 0x01) {
- ysv_ext1 -= 1;
- dy_ext1 += 1;
- } else {
- ysv_ext2 -= 1;
- dy_ext2 += 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb + 1;
- dy_ext0 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy_ext2 = dy0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x04) == 0) {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb;
- dz_ext0 = dz0 - 2 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz_ext2 = dz0 - SQUISH_CONSTANT_4D;
- if ((c & 0x03) == 0x03) {
- zsv_ext1 -= 1;
- dz_ext1 += 1;
- } else {
- zsv_ext2 -= 1;
- dz_ext2 += 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb + 1;
- dz_ext0 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz_ext2 = dz0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x08) == 0) {
- wsv_ext0 = wsv_ext1 = wsb;
- wsv_ext2 = wsb - 1;
- dw_ext0 = dw0 - 2 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 + 1 - SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsv_ext2 = wsb + 1;
- dw_ext0 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw_ext2 = dw0 - 1 - SQUISH_CONSTANT_4D;
- }
- }
-
- /* Contribution (0,0,0,0) */
- attn0 = 2 - dx0 * dx0 - dy0 * dy0 - dz0 * dz0 - dw0 * dw0;
- if (attn0 > 0) {
- attn0 *= attn0;
- value += attn0 * attn0 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 0, wsb + 0, dx0, dy0, dz0, dw0);
- }
-
- /* Contribution (1,0,0,0) */
- dx1 = dx0 - 1 - SQUISH_CONSTANT_4D;
- dy1 = dy0 - 0 - SQUISH_CONSTANT_4D;
- dz1 = dz0 - 0 - SQUISH_CONSTANT_4D;
- dw1 = dw0 - 0 - SQUISH_CONSTANT_4D;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1 - dw1 * dw1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 0, wsb + 0, dx1, dy1, dz1, dw1);
- }
-
- /* Contribution (0,1,0,0) */
- dx2 = dx0 - 0 - SQUISH_CONSTANT_4D;
- dy2 = dy0 - 1 - SQUISH_CONSTANT_4D;
- dz2 = dz1;
- dw2 = dw1;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2 - dw2 * dw2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 0, wsb + 0, dx2, dy2, dz2, dw2);
- }
-
- /* Contribution (0,0,1,0) */
- dx3 = dx2;
- dy3 = dy1;
- dz3 = dz0 - 1 - SQUISH_CONSTANT_4D;
- dw3 = dw1;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3 - dw3 * dw3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 1, wsb + 0, dx3, dy3, dz3, dw3);
- }
-
- /* Contribution (0,0,0,1) */
- dx4 = dx2;
- dy4 = dy1;
- dz4 = dz1;
- dw4 = dw0 - 1 - SQUISH_CONSTANT_4D;
- attn4 = 2 - dx4 * dx4 - dy4 * dy4 - dz4 * dz4 - dw4 * dw4;
- if (attn4 > 0) {
- attn4 *= attn4;
- value += attn4 * attn4 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 0, wsb + 1, dx4, dy4, dz4, dw4);
- }
- } else if (inSum >= 3) { /* We're inside the pentachoron (4-Simplex) at (1,1,1,1)
- Determine which two of (1,1,1,0), (1,1,0,1), (1,0,1,1), (0,1,1,1) are closest. */
- aPoint = 0x0E;
- aScore = xins;
- bPoint = 0x0D;
- bScore = yins;
- if (aScore <= bScore && zins < bScore) {
- bScore = zins;
- bPoint = 0x0B;
- } else if (aScore > bScore && zins < aScore) {
- aScore = zins;
- aPoint = 0x0B;
- }
- if (aScore <= bScore && wins < bScore) {
- bScore = wins;
- bPoint = 0x07;
- } else if (aScore > bScore && wins < aScore) {
- aScore = wins;
- aPoint = 0x07;
- }
-
- /* Now we determine the three lattice points not part of the pentachoron that may contribute.
- This depends on the closest two pentachoron vertices, including (0,0,0,0) */
- uins = 4 - inSum;
- if (uins < aScore || uins < bScore) { /* (1,1,1,1) is one of the closest two pentachoron vertices. */
- c = (bScore < aScore ? bPoint : aPoint); /* Our other closest vertex is the closest out of a and b. */
-
- if ((c & 0x01) != 0) {
- xsv_ext0 = xsb + 2;
- xsv_ext1 = xsv_ext2 = xsb + 1;
- dx_ext0 = dx0 - 2 - 4 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx_ext2 = dx0 - 1 - 4 * SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsv_ext2 = xsb;
- dx_ext0 = dx_ext1 = dx_ext2 = dx0 - 4 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x02) != 0) {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb + 1;
- dy_ext0 = dy_ext1 = dy_ext2 = dy0 - 1 - 4 * SQUISH_CONSTANT_4D;
- if ((c & 0x01) != 0) {
- ysv_ext1 += 1;
- dy_ext1 -= 1;
- } else {
- ysv_ext0 += 1;
- dy_ext0 -= 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb;
- dy_ext0 = dy_ext1 = dy_ext2 = dy0 - 4 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x04) != 0) {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb + 1;
- dz_ext0 = dz_ext1 = dz_ext2 = dz0 - 1 - 4 * SQUISH_CONSTANT_4D;
- if ((c & 0x03) != 0x03) {
- if ((c & 0x03) == 0) {
- zsv_ext0 += 1;
- dz_ext0 -= 1;
- } else {
- zsv_ext1 += 1;
- dz_ext1 -= 1;
- }
- } else {
- zsv_ext2 += 1;
- dz_ext2 -= 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb;
- dz_ext0 = dz_ext1 = dz_ext2 = dz0 - 4 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x08) != 0) {
- wsv_ext0 = wsv_ext1 = wsb + 1;
- wsv_ext2 = wsb + 2;
- dw_ext0 = dw_ext1 = dw0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 2 - 4 * SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsv_ext2 = wsb;
- dw_ext0 = dw_ext1 = dw_ext2 = dw0 - 4 * SQUISH_CONSTANT_4D;
- }
- } else { /* (1,1,1,1) is not one of the closest two pentachoron vertices. */
- c = (int8_t)(aPoint & bPoint); /* Our three extra vertices are determined by the closest two. */
-
- if ((c & 0x01) != 0) {
- xsv_ext0 = xsv_ext2 = xsb + 1;
- xsv_ext1 = xsb + 2;
- dx_ext0 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - 2 - 3 * SQUISH_CONSTANT_4D;
- dx_ext2 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsv_ext2 = xsb;
- dx_ext0 = dx0 - 2 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx_ext2 = dx0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x02) != 0) {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb + 1;
- dy_ext0 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy_ext2 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c & 0x01) != 0) {
- ysv_ext2 += 1;
- dy_ext2 -= 1;
- } else {
- ysv_ext1 += 1;
- dy_ext1 -= 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysv_ext2 = ysb;
- dy_ext0 = dy0 - 2 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy_ext2 = dy0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x04) != 0) {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb + 1;
- dz_ext0 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz_ext2 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c & 0x03) != 0) {
- zsv_ext2 += 1;
- dz_ext2 -= 1;
- } else {
- zsv_ext1 += 1;
- dz_ext1 -= 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsv_ext2 = zsb;
- dz_ext0 = dz0 - 2 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz_ext2 = dz0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x08) != 0) {
- wsv_ext0 = wsv_ext1 = wsb + 1;
- wsv_ext2 = wsb + 2;
- dw_ext0 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 2 - 3 * SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsv_ext2 = wsb;
- dw_ext0 = dw0 - 2 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw_ext2 = dw0 - 3 * SQUISH_CONSTANT_4D;
- }
- }
-
- /* Contribution (1,1,1,0) */
- dx4 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dy4 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dz4 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw4 = dw0 - 3 * SQUISH_CONSTANT_4D;
- attn4 = 2 - dx4 * dx4 - dy4 * dy4 - dz4 * dz4 - dw4 * dw4;
- if (attn4 > 0) {
- attn4 *= attn4;
- value += attn4 * attn4 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 1, wsb + 0, dx4, dy4, dz4, dw4);
- }
-
- /* Contribution (1,1,0,1) */
- dx3 = dx4;
- dy3 = dy4;
- dz3 = dz0 - 3 * SQUISH_CONSTANT_4D;
- dw3 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3 - dw3 * dw3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 0, wsb + 1, dx3, dy3, dz3, dw3);
- }
-
- /* Contribution (1,0,1,1) */
- dx2 = dx4;
- dy2 = dy0 - 3 * SQUISH_CONSTANT_4D;
- dz2 = dz4;
- dw2 = dw3;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2 - dw2 * dw2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 1, wsb + 1, dx2, dy2, dz2, dw2);
- }
-
- /* Contribution (0,1,1,1) */
- dx1 = dx0 - 3 * SQUISH_CONSTANT_4D;
- dz1 = dz4;
- dy1 = dy4;
- dw1 = dw3;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1 - dw1 * dw1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 1, wsb + 1, dx1, dy1, dz1, dw1);
- }
-
- /* Contribution (1,1,1,1) */
- dx0 = dx0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dy0 = dy0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dz0 = dz0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dw0 = dw0 - 1 - 4 * SQUISH_CONSTANT_4D;
- attn0 = 2 - dx0 * dx0 - dy0 * dy0 - dz0 * dz0 - dw0 * dw0;
- if (attn0 > 0) {
- attn0 *= attn0;
- value += attn0 * attn0 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 1, wsb + 1, dx0, dy0, dz0, dw0);
- }
- } else if (inSum <= 2) { /* We're inside the first dispentachoron (Rectified 4-Simplex) */
- aIsBiggerSide = 1;
- bIsBiggerSide = 1;
-
- /* Decide between (1,1,0,0) and (0,0,1,1) */
- if (xins + yins > zins + wins) {
- aScore = xins + yins;
- aPoint = 0x03;
- } else {
- aScore = zins + wins;
- aPoint = 0x0C;
- }
-
- /* Decide between (1,0,1,0) and (0,1,0,1) */
- if (xins + zins > yins + wins) {
- bScore = xins + zins;
- bPoint = 0x05;
- } else {
- bScore = yins + wins;
- bPoint = 0x0A;
- }
-
- /* Closer between (1,0,0,1) and (0,1,1,0) will replace the further of a and b, if closer. */
- if (xins + wins > yins + zins) {
- score = xins + wins;
- if (aScore >= bScore && score > bScore) {
- bScore = score;
- bPoint = 0x09;
- } else if (aScore < bScore && score > aScore) {
- aScore = score;
- aPoint = 0x09;
- }
- } else {
- score = yins + zins;
- if (aScore >= bScore && score > bScore) {
- bScore = score;
- bPoint = 0x06;
- } else if (aScore < bScore && score > aScore) {
- aScore = score;
- aPoint = 0x06;
- }
- }
-
- /* Decide if (1,0,0,0) is closer. */
- p1 = 2 - inSum + xins;
- if (aScore >= bScore && p1 > bScore) {
- bScore = p1;
- bPoint = 0x01;
- bIsBiggerSide = 0;
- } else if (aScore < bScore && p1 > aScore) {
- aScore = p1;
- aPoint = 0x01;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (0,1,0,0) is closer. */
- p2 = 2 - inSum + yins;
- if (aScore >= bScore && p2 > bScore) {
- bScore = p2;
- bPoint = 0x02;
- bIsBiggerSide = 0;
- } else if (aScore < bScore && p2 > aScore) {
- aScore = p2;
- aPoint = 0x02;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (0,0,1,0) is closer. */
- p3 = 2 - inSum + zins;
- if (aScore >= bScore && p3 > bScore) {
- bScore = p3;
- bPoint = 0x04;
- bIsBiggerSide = 0;
- } else if (aScore < bScore && p3 > aScore) {
- aScore = p3;
- aPoint = 0x04;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (0,0,0,1) is closer. */
- p4 = 2 - inSum + wins;
- if (aScore >= bScore && p4 > bScore) {
- bScore = p4;
- bPoint = 0x08;
- bIsBiggerSide = 0;
- } else if (aScore < bScore && p4 > aScore) {
- aScore = p4;
- aPoint = 0x08;
- aIsBiggerSide = 0;
- }
-
- /* Where each of the two closest points are determines how the extra three vertices are calculated. */
- if (aIsBiggerSide == bIsBiggerSide) {
- if (aIsBiggerSide) { /* Both closest points on the bigger side */
- c1 = (int8_t)(aPoint | bPoint);
- c2 = (int8_t)(aPoint & bPoint);
- if ((c1 & 0x01) == 0) {
- xsv_ext0 = xsb;
- xsv_ext1 = xsb - 1;
- dx_ext0 = dx0 - 3 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 + 1 - 2 * SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb + 1;
- dx_ext0 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x02) == 0) {
- ysv_ext0 = ysb;
- ysv_ext1 = ysb - 1;
- dy_ext0 = dy0 - 3 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy0 + 1 - 2 * SQUISH_CONSTANT_4D;
- } else {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x04) == 0) {
- zsv_ext0 = zsb;
- zsv_ext1 = zsb - 1;
- dz_ext0 = dz0 - 3 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz0 + 1 - 2 * SQUISH_CONSTANT_4D;
- } else {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x08) == 0) {
- wsv_ext0 = wsb;
- wsv_ext1 = wsb - 1;
- dw_ext0 = dw0 - 3 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 + 1 - 2 * SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsb + 1;
- dw_ext0 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- }
-
- /* One combination is a permutation of (0,0,0,2) based on c2 */
- xsv_ext2 = xsb;
- ysv_ext2 = ysb;
- zsv_ext2 = zsb;
- wsv_ext2 = wsb;
- dx_ext2 = dx0 - 2 * SQUISH_CONSTANT_4D;
- dy_ext2 = dy0 - 2 * SQUISH_CONSTANT_4D;
- dz_ext2 = dz0 - 2 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 2 * SQUISH_CONSTANT_4D;
- if ((c2 & 0x01) != 0) {
- xsv_ext2 += 2;
- dx_ext2 -= 2;
- } else if ((c2 & 0x02) != 0) {
- ysv_ext2 += 2;
- dy_ext2 -= 2;
- } else if ((c2 & 0x04) != 0) {
- zsv_ext2 += 2;
- dz_ext2 -= 2;
- } else {
- wsv_ext2 += 2;
- dw_ext2 -= 2;
- }
-
- } else { /* Both closest points on the smaller side */
- /* One of the two extra points is (0,0,0,0) */
- xsv_ext2 = xsb;
- ysv_ext2 = ysb;
- zsv_ext2 = zsb;
- wsv_ext2 = wsb;
- dx_ext2 = dx0;
- dy_ext2 = dy0;
- dz_ext2 = dz0;
- dw_ext2 = dw0;
-
- /* Other two points are based on the omitted axes. */
- c = (int8_t)(aPoint | bPoint);
-
- if ((c & 0x01) == 0) {
- xsv_ext0 = xsb - 1;
- xsv_ext1 = xsb;
- dx_ext0 = dx0 + 1 - SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb + 1;
- dx_ext0 = dx_ext1 = dx0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x02) == 0) {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0 - SQUISH_CONSTANT_4D;
- if ((c & 0x01) == 0x01)
- {
- ysv_ext0 -= 1;
- dy_ext0 += 1;
- } else {
- ysv_ext1 -= 1;
- dy_ext1 += 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x04) == 0) {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz_ext1 = dz0 - SQUISH_CONSTANT_4D;
- if ((c & 0x03) == 0x03)
- {
- zsv_ext0 -= 1;
- dz_ext0 += 1;
- } else {
- zsv_ext1 -= 1;
- dz_ext1 += 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz_ext1 = dz0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x08) == 0)
- {
- wsv_ext0 = wsb;
- wsv_ext1 = wsb - 1;
- dw_ext0 = dw0 - SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 + 1 - SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsb + 1;
- dw_ext0 = dw_ext1 = dw0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- }
- } else { /* One point on each "side" */
- if (aIsBiggerSide) {
- c1 = aPoint;
- c2 = bPoint;
- } else {
- c1 = bPoint;
- c2 = aPoint;
- }
-
- /* Two contributions are the bigger-sided point with each 0 replaced with -1. */
- if ((c1 & 0x01) == 0) {
- xsv_ext0 = xsb - 1;
- xsv_ext1 = xsb;
- dx_ext0 = dx0 + 1 - SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb + 1;
- dx_ext0 = dx_ext1 = dx0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x02) == 0) {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0 - SQUISH_CONSTANT_4D;
- if ((c1 & 0x01) == 0x01) {
- ysv_ext0 -= 1;
- dy_ext0 += 1;
- } else {
- ysv_ext1 -= 1;
- dy_ext1 += 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x04) == 0) {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz_ext1 = dz0 - SQUISH_CONSTANT_4D;
- if ((c1 & 0x03) == 0x03) {
- zsv_ext0 -= 1;
- dz_ext0 += 1;
- } else {
- zsv_ext1 -= 1;
- dz_ext1 += 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz_ext1 = dz0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x08) == 0) {
- wsv_ext0 = wsb;
- wsv_ext1 = wsb - 1;
- dw_ext0 = dw0 - SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 + 1 - SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsb + 1;
- dw_ext0 = dw_ext1 = dw0 - 1 - SQUISH_CONSTANT_4D;
- }
-
- /* One contribution is a permutation of (0,0,0,2) based on the smaller-sided point */
- xsv_ext2 = xsb;
- ysv_ext2 = ysb;
- zsv_ext2 = zsb;
- wsv_ext2 = wsb;
- dx_ext2 = dx0 - 2 * SQUISH_CONSTANT_4D;
- dy_ext2 = dy0 - 2 * SQUISH_CONSTANT_4D;
- dz_ext2 = dz0 - 2 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 2 * SQUISH_CONSTANT_4D;
- if ((c2 & 0x01) != 0) {
- xsv_ext2 += 2;
- dx_ext2 -= 2;
- } else if ((c2 & 0x02) != 0) {
- ysv_ext2 += 2;
- dy_ext2 -= 2;
- } else if ((c2 & 0x04) != 0) {
- zsv_ext2 += 2;
- dz_ext2 -= 2;
- } else {
- wsv_ext2 += 2;
- dw_ext2 -= 2;
- }
- }
-
- /* Contribution (1,0,0,0) */
- dx1 = dx0 - 1 - SQUISH_CONSTANT_4D;
- dy1 = dy0 - 0 - SQUISH_CONSTANT_4D;
- dz1 = dz0 - 0 - SQUISH_CONSTANT_4D;
- dw1 = dw0 - 0 - SQUISH_CONSTANT_4D;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1 - dw1 * dw1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 0, wsb + 0, dx1, dy1, dz1, dw1);
- }
-
- /* Contribution (0,1,0,0) */
- dx2 = dx0 - 0 - SQUISH_CONSTANT_4D;
- dy2 = dy0 - 1 - SQUISH_CONSTANT_4D;
- dz2 = dz1;
- dw2 = dw1;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2 - dw2 * dw2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 0, wsb + 0, dx2, dy2, dz2, dw2);
- }
-
- /* Contribution (0,0,1,0) */
- dx3 = dx2;
- dy3 = dy1;
- dz3 = dz0 - 1 - SQUISH_CONSTANT_4D;
- dw3 = dw1;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3 - dw3 * dw3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 1, wsb + 0, dx3, dy3, dz3, dw3);
- }
-
- /* Contribution (0,0,0,1) */
- dx4 = dx2;
- dy4 = dy1;
- dz4 = dz1;
- dw4 = dw0 - 1 - SQUISH_CONSTANT_4D;
- attn4 = 2 - dx4 * dx4 - dy4 * dy4 - dz4 * dz4 - dw4 * dw4;
- if (attn4 > 0) {
- attn4 *= attn4;
- value += attn4 * attn4 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 0, wsb + 1, dx4, dy4, dz4, dw4);
- }
-
- /* Contribution (1,1,0,0) */
- dx5 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy5 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz5 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw5 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn5 = 2 - dx5 * dx5 - dy5 * dy5 - dz5 * dz5 - dw5 * dw5;
- if (attn5 > 0) {
- attn5 *= attn5;
- value += attn5 * attn5 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 0, wsb + 0, dx5, dy5, dz5, dw5);
- }
-
- /* Contribution (1,0,1,0) */
- dx6 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy6 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz6 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw6 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn6 = 2 - dx6 * dx6 - dy6 * dy6 - dz6 * dz6 - dw6 * dw6;
- if (attn6 > 0) {
- attn6 *= attn6;
- value += attn6 * attn6 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 1, wsb + 0, dx6, dy6, dz6, dw6);
- }
-
- /* Contribution (1,0,0,1) */
- dx7 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy7 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz7 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw7 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn7 = 2 - dx7 * dx7 - dy7 * dy7 - dz7 * dz7 - dw7 * dw7;
- if (attn7 > 0) {
- attn7 *= attn7;
- value += attn7 * attn7 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 0, wsb + 1, dx7, dy7, dz7, dw7);
- }
-
- /* Contribution (0,1,1,0) */
- dx8 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy8 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz8 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw8 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn8 = 2 - dx8 * dx8 - dy8 * dy8 - dz8 * dz8 - dw8 * dw8;
- if (attn8 > 0) {
- attn8 *= attn8;
- value += attn8 * attn8 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 1, wsb + 0, dx8, dy8, dz8, dw8);
- }
-
- /* Contribution (0,1,0,1) */
- dx9 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy9 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz9 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw9 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn9 = 2 - dx9 * dx9 - dy9 * dy9 - dz9 * dz9 - dw9 * dw9;
- if (attn9 > 0) {
- attn9 *= attn9;
- value += attn9 * attn9 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 0, wsb + 1, dx9, dy9, dz9, dw9);
- }
-
- /* Contribution (0,0,1,1) */
- dx10 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy10 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz10 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw10 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn10 = 2 - dx10 * dx10 - dy10 * dy10 - dz10 * dz10 - dw10 * dw10;
- if (attn10 > 0) {
- attn10 *= attn10;
- value += attn10 * attn10 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 1, wsb + 1, dx10, dy10, dz10, dw10);
- }
- } else { /* We're inside the second dispentachoron (Rectified 4-Simplex) */
- aIsBiggerSide = 1;
- bIsBiggerSide = 1;
-
- /* Decide between (0,0,1,1) and (1,1,0,0) */
- if (xins + yins < zins + wins) {
- aScore = xins + yins;
- aPoint = 0x0C;
- } else {
- aScore = zins + wins;
- aPoint = 0x03;
- }
-
- /* Decide between (0,1,0,1) and (1,0,1,0) */
- if (xins + zins < yins + wins) {
- bScore = xins + zins;
- bPoint = 0x0A;
- } else {
- bScore = yins + wins;
- bPoint = 0x05;
- }
-
- /* Closer between (0,1,1,0) and (1,0,0,1) will replace the further of a and b, if closer. */
- if (xins + wins < yins + zins) {
- score = xins + wins;
- if (aScore <= bScore && score < bScore) {
- bScore = score;
- bPoint = 0x06;
- } else if (aScore > bScore && score < aScore) {
- aScore = score;
- aPoint = 0x06;
- }
- } else {
- score = yins + zins;
- if (aScore <= bScore && score < bScore) {
- bScore = score;
- bPoint = 0x09;
- } else if (aScore > bScore && score < aScore) {
- aScore = score;
- aPoint = 0x09;
- }
- }
-
- /* Decide if (0,1,1,1) is closer. */
- p1 = 3 - inSum + xins;
- if (aScore <= bScore && p1 < bScore) {
- bScore = p1;
- bPoint = 0x0E;
- bIsBiggerSide = 0;
- } else if (aScore > bScore && p1 < aScore) {
- aScore = p1;
- aPoint = 0x0E;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (1,0,1,1) is closer. */
- p2 = 3 - inSum + yins;
- if (aScore <= bScore && p2 < bScore) {
- bScore = p2;
- bPoint = 0x0D;
- bIsBiggerSide = 0;
- } else if (aScore > bScore && p2 < aScore) {
- aScore = p2;
- aPoint = 0x0D;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (1,1,0,1) is closer. */
- p3 = 3 - inSum + zins;
- if (aScore <= bScore && p3 < bScore) {
- bScore = p3;
- bPoint = 0x0B;
- bIsBiggerSide = 0;
- } else if (aScore > bScore && p3 < aScore) {
- aScore = p3;
- aPoint = 0x0B;
- aIsBiggerSide = 0;
- }
-
- /* Decide if (1,1,1,0) is closer. */
- p4 = 3 - inSum + wins;
- if (aScore <= bScore && p4 < bScore) {
- bScore = p4;
- bPoint = 0x07;
- bIsBiggerSide = 0;
- } else if (aScore > bScore && p4 < aScore) {
- aScore = p4;
- aPoint = 0x07;
- aIsBiggerSide = 0;
- }
-
- /* Where each of the two closest points are determines how the extra three vertices are calculated. */
- if (aIsBiggerSide == bIsBiggerSide) {
- if (aIsBiggerSide) { /* Both closest points on the bigger side */
- c1 = (int8_t)(aPoint & bPoint);
- c2 = (int8_t)(aPoint | bPoint);
-
- /* Two contributions are permutations of (0,0,0,1) and (0,0,0,2) based on c1 */
- xsv_ext0 = xsv_ext1 = xsb;
- ysv_ext0 = ysv_ext1 = ysb;
- zsv_ext0 = zsv_ext1 = zsb;
- wsv_ext0 = wsv_ext1 = wsb;
- dx_ext0 = dx0 - SQUISH_CONSTANT_4D;
- dy_ext0 = dy0 - SQUISH_CONSTANT_4D;
- dz_ext0 = dz0 - SQUISH_CONSTANT_4D;
- dw_ext0 = dw0 - SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - 2 * SQUISH_CONSTANT_4D;
- dy_ext1 = dy0 - 2 * SQUISH_CONSTANT_4D;
- dz_ext1 = dz0 - 2 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - 2 * SQUISH_CONSTANT_4D;
- if ((c1 & 0x01) != 0) {
- xsv_ext0 += 1;
- dx_ext0 -= 1;
- xsv_ext1 += 2;
- dx_ext1 -= 2;
- } else if ((c1 & 0x02) != 0) {
- ysv_ext0 += 1;
- dy_ext0 -= 1;
- ysv_ext1 += 2;
- dy_ext1 -= 2;
- } else if ((c1 & 0x04) != 0) {
- zsv_ext0 += 1;
- dz_ext0 -= 1;
- zsv_ext1 += 2;
- dz_ext1 -= 2;
- } else {
- wsv_ext0 += 1;
- dw_ext0 -= 1;
- wsv_ext1 += 2;
- dw_ext1 -= 2;
- }
-
- /* One contribution is a permutation of (1,1,1,-1) based on c2 */
- xsv_ext2 = xsb + 1;
- ysv_ext2 = ysb + 1;
- zsv_ext2 = zsb + 1;
- wsv_ext2 = wsb + 1;
- dx_ext2 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy_ext2 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz_ext2 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- if ((c2 & 0x01) == 0) {
- xsv_ext2 -= 2;
- dx_ext2 += 2;
- } else if ((c2 & 0x02) == 0) {
- ysv_ext2 -= 2;
- dy_ext2 += 2;
- } else if ((c2 & 0x04) == 0) {
- zsv_ext2 -= 2;
- dz_ext2 += 2;
- } else {
- wsv_ext2 -= 2;
- dw_ext2 += 2;
- }
- } else { /* Both closest points on the smaller side */
- /* One of the two extra points is (1,1,1,1) */
- xsv_ext2 = xsb + 1;
- ysv_ext2 = ysb + 1;
- zsv_ext2 = zsb + 1;
- wsv_ext2 = wsb + 1;
- dx_ext2 = dx0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dy_ext2 = dy0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dz_ext2 = dz0 - 1 - 4 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 1 - 4 * SQUISH_CONSTANT_4D;
-
- /* Other two points are based on the shared axes. */
- c = (int8_t)(aPoint & bPoint);
-
- if ((c & 0x01) != 0) {
- xsv_ext0 = xsb + 2;
- xsv_ext1 = xsb + 1;
- dx_ext0 = dx0 - 2 - 3 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb;
- dx_ext0 = dx_ext1 = dx0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x02) != 0) {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c & 0x01) == 0)
- {
- ysv_ext0 += 1;
- dy_ext0 -= 1;
- } else {
- ysv_ext1 += 1;
- dy_ext1 -= 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x04) != 0) {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz_ext1 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c & 0x03) == 0)
- {
- zsv_ext0 += 1;
- dz_ext0 -= 1;
- } else {
- zsv_ext1 += 1;
- dz_ext1 -= 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz_ext1 = dz0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c & 0x08) != 0)
- {
- wsv_ext0 = wsb + 1;
- wsv_ext1 = wsb + 2;
- dw_ext0 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - 2 - 3 * SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsb;
- dw_ext0 = dw_ext1 = dw0 - 3 * SQUISH_CONSTANT_4D;
- }
- }
- } else { /* One point on each "side" */
- if (aIsBiggerSide) {
- c1 = aPoint;
- c2 = bPoint;
- } else {
- c1 = bPoint;
- c2 = aPoint;
- }
-
- /* Two contributions are the bigger-sided point with each 1 replaced with 2. */
- if ((c1 & 0x01) != 0) {
- xsv_ext0 = xsb + 2;
- xsv_ext1 = xsb + 1;
- dx_ext0 = dx0 - 2 - 3 * SQUISH_CONSTANT_4D;
- dx_ext1 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- } else {
- xsv_ext0 = xsv_ext1 = xsb;
- dx_ext0 = dx_ext1 = dx0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x02) != 0) {
- ysv_ext0 = ysv_ext1 = ysb + 1;
- dy_ext0 = dy_ext1 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c1 & 0x01) == 0) {
- ysv_ext0 += 1;
- dy_ext0 -= 1;
- } else {
- ysv_ext1 += 1;
- dy_ext1 -= 1;
- }
- } else {
- ysv_ext0 = ysv_ext1 = ysb;
- dy_ext0 = dy_ext1 = dy0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x04) != 0) {
- zsv_ext0 = zsv_ext1 = zsb + 1;
- dz_ext0 = dz_ext1 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- if ((c1 & 0x03) == 0) {
- zsv_ext0 += 1;
- dz_ext0 -= 1;
- } else {
- zsv_ext1 += 1;
- dz_ext1 -= 1;
- }
- } else {
- zsv_ext0 = zsv_ext1 = zsb;
- dz_ext0 = dz_ext1 = dz0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- if ((c1 & 0x08) != 0) {
- wsv_ext0 = wsb + 1;
- wsv_ext1 = wsb + 2;
- dw_ext0 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw_ext1 = dw0 - 2 - 3 * SQUISH_CONSTANT_4D;
- } else {
- wsv_ext0 = wsv_ext1 = wsb;
- dw_ext0 = dw_ext1 = dw0 - 3 * SQUISH_CONSTANT_4D;
- }
-
- /* One contribution is a permutation of (1,1,1,-1) based on the smaller-sided point */
- xsv_ext2 = xsb + 1;
- ysv_ext2 = ysb + 1;
- zsv_ext2 = zsb + 1;
- wsv_ext2 = wsb + 1;
- dx_ext2 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy_ext2 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz_ext2 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw_ext2 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- if ((c2 & 0x01) == 0) {
- xsv_ext2 -= 2;
- dx_ext2 += 2;
- } else if ((c2 & 0x02) == 0) {
- ysv_ext2 -= 2;
- dy_ext2 += 2;
- } else if ((c2 & 0x04) == 0) {
- zsv_ext2 -= 2;
- dz_ext2 += 2;
- } else {
- wsv_ext2 -= 2;
- dw_ext2 += 2;
- }
- }
-
- /* Contribution (1,1,1,0) */
- dx4 = dx0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dy4 = dy0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dz4 = dz0 - 1 - 3 * SQUISH_CONSTANT_4D;
- dw4 = dw0 - 3 * SQUISH_CONSTANT_4D;
- attn4 = 2 - dx4 * dx4 - dy4 * dy4 - dz4 * dz4 - dw4 * dw4;
- if (attn4 > 0) {
- attn4 *= attn4;
- value += attn4 * attn4 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 1, wsb + 0, dx4, dy4, dz4, dw4);
- }
-
- /* Contribution (1,1,0,1) */
- dx3 = dx4;
- dy3 = dy4;
- dz3 = dz0 - 3 * SQUISH_CONSTANT_4D;
- dw3 = dw0 - 1 - 3 * SQUISH_CONSTANT_4D;
- attn3 = 2 - dx3 * dx3 - dy3 * dy3 - dz3 * dz3 - dw3 * dw3;
- if (attn3 > 0) {
- attn3 *= attn3;
- value += attn3 * attn3 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 0, wsb + 1, dx3, dy3, dz3, dw3);
- }
-
- /* Contribution (1,0,1,1) */
- dx2 = dx4;
- dy2 = dy0 - 3 * SQUISH_CONSTANT_4D;
- dz2 = dz4;
- dw2 = dw3;
- attn2 = 2 - dx2 * dx2 - dy2 * dy2 - dz2 * dz2 - dw2 * dw2;
- if (attn2 > 0) {
- attn2 *= attn2;
- value += attn2 * attn2 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 1, wsb + 1, dx2, dy2, dz2, dw2);
- }
-
- /* Contribution (0,1,1,1) */
- dx1 = dx0 - 3 * SQUISH_CONSTANT_4D;
- dz1 = dz4;
- dy1 = dy4;
- dw1 = dw3;
- attn1 = 2 - dx1 * dx1 - dy1 * dy1 - dz1 * dz1 - dw1 * dw1;
- if (attn1 > 0) {
- attn1 *= attn1;
- value += attn1 * attn1 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 1, wsb + 1, dx1, dy1, dz1, dw1);
- }
-
- /* Contribution (1,1,0,0) */
- dx5 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy5 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz5 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw5 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn5 = 2 - dx5 * dx5 - dy5 * dy5 - dz5 * dz5 - dw5 * dw5;
- if (attn5 > 0) {
- attn5 *= attn5;
- value += attn5 * attn5 * extrapolate4(ctx, xsb + 1, ysb + 1, zsb + 0, wsb + 0, dx5, dy5, dz5, dw5);
- }
-
- /* Contribution (1,0,1,0) */
- dx6 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy6 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz6 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw6 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn6 = 2 - dx6 * dx6 - dy6 * dy6 - dz6 * dz6 - dw6 * dw6;
- if (attn6 > 0) {
- attn6 *= attn6;
- value += attn6 * attn6 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 1, wsb + 0, dx6, dy6, dz6, dw6);
- }
-
- /* Contribution (1,0,0,1) */
- dx7 = dx0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dy7 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz7 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw7 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn7 = 2 - dx7 * dx7 - dy7 * dy7 - dz7 * dz7 - dw7 * dw7;
- if (attn7 > 0) {
- attn7 *= attn7;
- value += attn7 * attn7 * extrapolate4(ctx, xsb + 1, ysb + 0, zsb + 0, wsb + 1, dx7, dy7, dz7, dw7);
- }
-
- /* Contribution (0,1,1,0) */
- dx8 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy8 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz8 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw8 = dw0 - 0 - 2 * SQUISH_CONSTANT_4D;
- attn8 = 2 - dx8 * dx8 - dy8 * dy8 - dz8 * dz8 - dw8 * dw8;
- if (attn8 > 0) {
- attn8 *= attn8;
- value += attn8 * attn8 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 1, wsb + 0, dx8, dy8, dz8, dw8);
- }
-
- /* Contribution (0,1,0,1) */
- dx9 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy9 = dy0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dz9 = dz0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dw9 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn9 = 2 - dx9 * dx9 - dy9 * dy9 - dz9 * dz9 - dw9 * dw9;
- if (attn9 > 0) {
- attn9 *= attn9;
- value += attn9 * attn9 * extrapolate4(ctx, xsb + 0, ysb + 1, zsb + 0, wsb + 1, dx9, dy9, dz9, dw9);
- }
-
- /* Contribution (0,0,1,1) */
- dx10 = dx0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dy10 = dy0 - 0 - 2 * SQUISH_CONSTANT_4D;
- dz10 = dz0 - 1 - 2 * SQUISH_CONSTANT_4D;
- dw10 = dw0 - 1 - 2 * SQUISH_CONSTANT_4D;
- attn10 = 2 - dx10 * dx10 - dy10 * dy10 - dz10 * dz10 - dw10 * dw10;
- if (attn10 > 0) {
- attn10 *= attn10;
- value += attn10 * attn10 * extrapolate4(ctx, xsb + 0, ysb + 0, zsb + 1, wsb + 1, dx10, dy10, dz10, dw10);
- }
- }
-
- /* First extra vertex */
- attn_ext0 = 2 - dx_ext0 * dx_ext0 - dy_ext0 * dy_ext0 - dz_ext0 * dz_ext0 - dw_ext0 * dw_ext0;
- if (attn_ext0 > 0)
- {
- attn_ext0 *= attn_ext0;
- value += attn_ext0 * attn_ext0 * extrapolate4(ctx, xsv_ext0, ysv_ext0, zsv_ext0, wsv_ext0, dx_ext0, dy_ext0, dz_ext0, dw_ext0);
- }
-
- /* Second extra vertex */
- attn_ext1 = 2 - dx_ext1 * dx_ext1 - dy_ext1 * dy_ext1 - dz_ext1 * dz_ext1 - dw_ext1 * dw_ext1;
- if (attn_ext1 > 0)
- {
- attn_ext1 *= attn_ext1;
- value += attn_ext1 * attn_ext1 * extrapolate4(ctx, xsv_ext1, ysv_ext1, zsv_ext1, wsv_ext1, dx_ext1, dy_ext1, dz_ext1, dw_ext1);
- }
-
- /* Third extra vertex */
- attn_ext2 = 2 - dx_ext2 * dx_ext2 - dy_ext2 * dy_ext2 - dz_ext2 * dz_ext2 - dw_ext2 * dw_ext2;
- if (attn_ext2 > 0)
- {
- attn_ext2 *= attn_ext2;
- value += attn_ext2 * attn_ext2 * extrapolate4(ctx, xsv_ext2, ysv_ext2, zsv_ext2, wsv_ext2, dx_ext2, dy_ext2, dz_ext2, dw_ext2);
- }
-
- return value / NORM_CONSTANT_4D;
-}
-
diff --git a/thirdparty/misc/open-simplex-noise.h b/thirdparty/misc/open-simplex-noise.h
deleted file mode 100644
index fd9248c3a1..0000000000
--- a/thirdparty/misc/open-simplex-noise.h
+++ /dev/null
@@ -1,58 +0,0 @@
-#ifndef OPEN_SIMPLEX_NOISE_H__
-#define OPEN_SIMPLEX_NOISE_H__
-
-/*
- * OpenSimplex (Simplectic) Noise in C.
- * Ported to C from Kurt Spencer's java implementation by Stephen M. Cameron
- *
- * v1.1 (October 6, 2014)
- * - Ported to C
- *
- * v1.1 (October 5, 2014)
- * - Added 2D and 4D implementations.
- * - Proper gradient sets for all dimensions, from a
- * dimensionally-generalizable scheme with an actual
- * rhyme and reason behind it.
- * - Removed default permutation array in favor of
- * default seed.
- * - Changed seed-based constructor to be independent
- * of any particular randomization library, so results
- * will be the same when ported to other languages.
- */
-
-#if ((__GNUC_STDC_INLINE__) || (__STDC_VERSION__ >= 199901L))
- #include <stdint.h>
- #define INLINE inline
-#elif (defined (_MSC_VER) || defined (__GNUC_GNU_INLINE__))
- #include <stdint.h>
- #define INLINE __inline
-#else
- /* ANSI C doesn't have inline or stdint.h. */
- #define INLINE
-#endif
-
-#ifdef __cplusplus
- extern "C" {
-#endif
-
-// -- GODOT start --
-// Modified to work without allocating memory, also removed some unused function.
-
-struct osn_context {
- int16_t perm[256];
- int16_t permGradIndex3D[256];
-};
-
-int open_simplex_noise(int64_t seed, struct osn_context *ctx);
-//int open_simplex_noise_init_perm(struct osn_context *ctx, int16_t p[], int nelements);
-// -- GODOT end --
-void open_simplex_noise_free(struct osn_context *ctx);
-double open_simplex_noise2(const struct osn_context *ctx, double x, double y);
-double open_simplex_noise3(const struct osn_context *ctx, double x, double y, double z);
-double open_simplex_noise4(const struct osn_context *ctx, double x, double y, double z, double w);
-
-#ifdef __cplusplus
- }
-#endif
-
-#endif
diff --git a/thirdparty/noise/FastNoise-LICENSE b/thirdparty/noise/FastNoise-LICENSE
new file mode 100644
index 0000000000..dd6df2c160
--- /dev/null
+++ b/thirdparty/noise/FastNoise-LICENSE
@@ -0,0 +1,22 @@
+MIT License
+
+Copyright(c) 2020 Jordan Peck (jordan.me2@gmail.com)
+Copyright(c) 2020 Contributors
+
+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. \ No newline at end of file
diff --git a/thirdparty/noise/FastNoiseLite.h b/thirdparty/noise/FastNoiseLite.h
new file mode 100644
index 0000000000..3db344c149
--- /dev/null
+++ b/thirdparty/noise/FastNoiseLite.h
@@ -0,0 +1,2589 @@
+// MIT License
+//
+// Copyright(c) 2020 Jordan Peck (jordan.me2@gmail.com)
+// Copyright(c) 2020 Contributors
+//
+// 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.
+//
+// .'',;:cldxkO00KKXXNNWWWNNXKOkxdollcc::::::;:::ccllloooolllllllllooollc:,'... ...........',;cldxkO000Okxdlc::;;;,,;;;::cclllllll
+// ..',;:ldxO0KXXNNNNNNNNXXK0kxdolcc::::::;;;,,,,,,;;;;;;;;;;:::cclllllc:;'.... ...........',;:ldxO0KXXXK0Okxdolc::;;;;::cllodddddo
+// ...',:loxO0KXNNNNNXXKK0Okxdolc::;::::::::;;;,,'''''.....''',;:clllllc:;,'............''''''''',;:loxO0KXNNNNNXK0Okxdollccccllodxxxxxxd
+// ....';:ldkO0KXXXKK00Okxdolcc:;;;;;::cclllcc:;;,''..... ....',;clooddolcc:;;;;,,;;;;;::::;;;;;;:cloxk0KXNWWWWWWNXKK0Okxddoooddxxkkkkkxx
+// .....';:ldxkOOOOOkxxdolcc:;;;,,,;;:cllooooolcc:;'... ..,:codxkkkxddooollloooooooollcc:::::clodkO0KXNWWWWWWNNXK00Okxxxxxxxxkkkkxxx
+// . ....';:cloddddo___________,,,,;;:clooddddoolc:,... ..,:ldx__00OOOkkk___kkkkkkxxdollc::::cclodkO0KXXNNNNNNXXK0OOkxxxxxxxxxxxxddd
+// .......',;:cccc:| |,,,;;:cclooddddoll:;'.. ..';cox| \KKK000| |KK00OOkxdocc___;::clldxxkO0KKKKK00Okkxdddddddddddddddoo
+// .......'',,,,,''| ________|',,;;::cclloooooolc:;'......___:ldk| \KK000| |XKKK0Okxolc| |;;::cclodxxkkkkxxdoolllcclllooodddooooo
+// ''......''''....| | ....'',,,,;;;::cclloooollc:;,''.'| |oxk| \OOO0| |KKK00Oxdoll|___|;;;;;::ccllllllcc::;;,,;;;:cclloooooooo
+// ;;,''.......... | |_____',,;;;____:___cllo________.___| |___| \xkk| |KK_______ool___:::;________;;;_______...'',;;:ccclllloo
+// c:;,''......... | |:::/ ' |lo/ | | \dx| |0/ \d| |cc/ |'/ \......',,;;:ccllo
+// ol:;,'..........| _____|ll/ __ |o/ ______|____ ___| | \o| |/ ___ \| |o/ ______|/ ___ \ .......'',;:clo
+// dlc;,...........| |::clooo| / | |x\___ \KXKKK0| |dol| |\ \| | | | | |d\___ \..| | / / ....',:cl
+// xoc;'... .....'| |llodddd| \__| |_____\ \KKK0O| |lc:| |'\ | |___| | |_____\ \.| |_/___/... ...',;:c
+// dlc;'... ....',;| |oddddddo\ | |Okkx| |::;| |..\ |\ /| | | \ |... ....',;:c
+// ol:,'.......',:c|___|xxxddollc\_____,___|_________/ddoll|___|,,,|___|...\_____|:\ ______/l|___|_________/...\________|'........',;::cc
+// c:;'.......';:codxxkkkkxxolc::;::clodxkOO0OOkkxdollc::;;,,''''',,,,''''''''''',,'''''',;:loxkkOOkxol:;,'''',,;:ccllcc:;,'''''',;::ccll
+// ;,'.......',:codxkOO0OOkxdlc:;,,;;:cldxxkkxxdolc:;;,,''.....'',;;:::;;,,,'''''........,;cldkO0KK0Okdoc::;;::cloodddoolc:;;;;;::ccllooo
+// .........',;:lodxOO0000Okdoc:,,',,;:clloddoolc:;,''.......'',;:clooollc:;;,,''.......',:ldkOKXNNXX0Oxdolllloddxxxxxxdolccccccllooodddd
+// . .....';:cldxkO0000Okxol:;,''',,;::cccc:;,,'.......'',;:cldxxkkxxdolc:;;,'.......';coxOKXNWWWNXKOkxddddxxkkkkkkxdoollllooddxxxxkkk
+// ....',;:codxkO000OOxdoc:;,''',,,;;;;,''.......',,;:clodkO00000Okxolc::;,,''..',;:ldxOKXNWWWNNK0OkkkkkkkkkkkxxddooooodxxkOOOOO000
+// ....',;;clodxkkOOOkkdolc:;,,,,,,,,'..........,;:clodxkO0KKXKK0Okxdolcc::;;,,,;;:codkO0XXNNNNXKK0OOOOOkkkkxxdoollloodxkO0KKKXXXXX
+//
+// VERSION: 1.0.1
+// https://github.com/Auburn/FastNoise
+
+#ifndef FASTNOISELITE_H
+#define FASTNOISELITE_H
+
+#include <cmath>
+
+namespace fastnoiselite{
+
+class FastNoiseLite
+{
+public:
+ enum NoiseType
+ {
+ NoiseType_OpenSimplex2,
+ NoiseType_OpenSimplex2S,
+ NoiseType_Cellular,
+ NoiseType_Perlin,
+ NoiseType_ValueCubic,
+ NoiseType_Value
+ };
+
+ enum RotationType3D
+ {
+ RotationType3D_None,
+ RotationType3D_ImproveXYPlanes,
+ RotationType3D_ImproveXZPlanes
+ };
+
+ enum FractalType
+ {
+ FractalType_None,
+ FractalType_FBm,
+ FractalType_Ridged,
+ FractalType_PingPong,
+ FractalType_DomainWarpProgressive,
+ FractalType_DomainWarpIndependent
+ };
+
+ enum CellularDistanceFunction
+ {
+ CellularDistanceFunction_Euclidean,
+ CellularDistanceFunction_EuclideanSq,
+ CellularDistanceFunction_Manhattan,
+ CellularDistanceFunction_Hybrid
+ };
+
+ enum CellularReturnType
+ {
+ CellularReturnType_CellValue,
+ CellularReturnType_Distance,
+ CellularReturnType_Distance2,
+ CellularReturnType_Distance2Add,
+ CellularReturnType_Distance2Sub,
+ CellularReturnType_Distance2Mul,
+ CellularReturnType_Distance2Div
+ };
+
+ enum DomainWarpType
+ {
+ DomainWarpType_OpenSimplex2,
+ DomainWarpType_OpenSimplex2Reduced,
+ DomainWarpType_BasicGrid
+ };
+
+ /// <summary>
+ /// Create new FastNoise object with optional seed
+ /// </summary>
+ FastNoiseLite(int seed = 1337)
+ {
+ mSeed = seed;
+ mFrequency = 0.01f;
+ mNoiseType = NoiseType_OpenSimplex2;
+ mRotationType3D = RotationType3D_None;
+ mTransformType3D = TransformType3D_DefaultOpenSimplex2;
+
+ mFractalType = FractalType_None;
+ mOctaves = 3;
+ mLacunarity = 2.0f;
+ mGain = 0.5f;
+ mWeightedStrength = 0.0f;
+ mPingPongStrength = 2.0f;
+
+ mFractalBounding = 1 / 1.75f;
+
+ mCellularDistanceFunction = CellularDistanceFunction_EuclideanSq;
+ mCellularReturnType = CellularReturnType_Distance;
+ mCellularJitterModifier = 1.0f;
+
+ mDomainWarpType = DomainWarpType_OpenSimplex2;
+ mWarpTransformType3D = TransformType3D_DefaultOpenSimplex2;
+ mDomainWarpAmp = 1.0f;
+ }
+
+ /// <summary>
+ /// Sets seed used for all noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: 1337
+ /// </remarks>
+ void SetSeed(int seed) { mSeed = seed; }
+
+ /// <summary>
+ /// Sets frequency for all noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: 0.01
+ /// </remarks>
+ void SetFrequency(float frequency) { mFrequency = frequency; }
+
+ /// <summary>
+ /// Sets noise algorithm used for GetNoise(...)
+ /// </summary>
+ /// <remarks>
+ /// Default: OpenSimplex2
+ /// </remarks>
+ void SetNoiseType(NoiseType noiseType)
+ {
+ mNoiseType = noiseType;
+ UpdateTransformType3D();
+ }
+
+ /// <summary>
+ /// Sets domain rotation type for 3D Noise and 3D DomainWarp.
+ /// Can aid in reducing directional artifacts when sampling a 2D plane in 3D
+ /// </summary>
+ /// <remarks>
+ /// Default: None
+ /// </remarks>
+ void SetRotationType3D(RotationType3D rotationType3D)
+ {
+ mRotationType3D = rotationType3D;
+ UpdateTransformType3D();
+ UpdateWarpTransformType3D();
+ }
+
+ /// <summary>
+ /// Sets method for combining octaves in all fractal noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: None
+ /// Note: FractalType_DomainWarp... only affects DomainWarp(...)
+ /// </remarks>
+ void SetFractalType(FractalType fractalType) { mFractalType = fractalType; }
+
+ /// <summary>
+ /// Sets octave count for all fractal noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: 3
+ /// </remarks>
+ void SetFractalOctaves(int octaves)
+ {
+ mOctaves = octaves;
+ CalculateFractalBounding();
+ }
+
+ /// <summary>
+ /// Sets octave lacunarity for all fractal noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: 2.0
+ /// </remarks>
+ void SetFractalLacunarity(float lacunarity) { mLacunarity = lacunarity; }
+
+ /// <summary>
+ /// Sets octave gain for all fractal noise types
+ /// </summary>
+ /// <remarks>
+ /// Default: 0.5
+ /// </remarks>
+ void SetFractalGain(float gain)
+ {
+ mGain = gain;
+ CalculateFractalBounding();
+ }
+
+ /// <summary>
+ /// Sets octave weighting for all none DomainWarp fratal types
+ /// </summary>
+ /// <remarks>
+ /// Default: 0.0
+ /// Note: Keep between 0...1 to maintain -1...1 output bounding
+ /// </remarks>
+ void SetFractalWeightedStrength(float weightedStrength) { mWeightedStrength = weightedStrength; }
+
+ /// <summary>
+ /// Sets strength of the fractal ping pong effect
+ /// </summary>
+ /// <remarks>
+ /// Default: 2.0
+ /// </remarks>
+ void SetFractalPingPongStrength(float pingPongStrength) { mPingPongStrength = pingPongStrength; }
+
+
+ /// <summary>
+ /// Sets distance function used in cellular noise calculations
+ /// </summary>
+ /// <remarks>
+ /// Default: Distance
+ /// </remarks>
+ void SetCellularDistanceFunction(CellularDistanceFunction cellularDistanceFunction) { mCellularDistanceFunction = cellularDistanceFunction; }
+
+ /// <summary>
+ /// Sets return type from cellular noise calculations
+ /// </summary>
+ /// <remarks>
+ /// Default: EuclideanSq
+ /// </remarks>
+ void SetCellularReturnType(CellularReturnType cellularReturnType) { mCellularReturnType = cellularReturnType; }
+
+ /// <summary>
+ /// Sets the maximum distance a cellular point can move from it's grid position
+ /// </summary>
+ /// <remarks>
+ /// Default: 1.0
+ /// Note: Setting this higher than 1 will cause artifacts
+ /// </remarks>
+ void SetCellularJitter(float cellularJitter) { mCellularJitterModifier = cellularJitter; }
+
+
+ /// <summary>
+ /// Sets the warp algorithm when using DomainWarp(...)
+ /// </summary>
+ /// <remarks>
+ /// Default: OpenSimplex2
+ /// </remarks>
+ void SetDomainWarpType(DomainWarpType domainWarpType)
+ {
+ mDomainWarpType = domainWarpType;
+ UpdateWarpTransformType3D();
+ }
+
+
+ /// <summary>
+ /// Sets the maximum warp distance from original position when using DomainWarp(...)
+ /// </summary>
+ /// <remarks>
+ /// Default: 1.0
+ /// </remarks>
+ void SetDomainWarpAmp(float domainWarpAmp) { mDomainWarpAmp = domainWarpAmp; }
+
+
+ /// <summary>
+ /// 2D noise at given position using current settings
+ /// </summary>
+ /// <returns>
+ /// Noise output bounded between -1...1
+ /// </returns>
+ template <typename FNfloat>
+ float GetNoise(FNfloat x, FNfloat y)
+ {
+ Arguments_must_be_floating_point_values<FNfloat>();
+
+ TransformNoiseCoordinate(x, y);
+
+ switch (mFractalType)
+ {
+ default:
+ return GenNoiseSingle(mSeed, x, y);
+ case FractalType_FBm:
+ return GenFractalFBm(x, y);
+ case FractalType_Ridged:
+ return GenFractalRidged(x, y);
+ case FractalType_PingPong:
+ return GenFractalPingPong(x, y);
+ }
+ }
+
+ /// <summary>
+ /// 3D noise at given position using current settings
+ /// </summary>
+ /// <returns>
+ /// Noise output bounded between -1...1
+ /// </returns>
+ template <typename FNfloat>
+ float GetNoise(FNfloat x, FNfloat y, FNfloat z)
+ {
+ Arguments_must_be_floating_point_values<FNfloat>();
+
+ TransformNoiseCoordinate(x, y, z);
+
+ switch (mFractalType)
+ {
+ default:
+ return GenNoiseSingle(mSeed, x, y, z);
+ case FractalType_FBm:
+ return GenFractalFBm(x, y, z);
+ case FractalType_Ridged:
+ return GenFractalRidged(x, y, z);
+ case FractalType_PingPong:
+ return GenFractalPingPong(x, y, z);
+ }
+ }
+
+
+ /// <summary>
+ /// 2D warps the input position using current domain warp settings
+ /// </summary>
+ /// <example>
+ /// Example usage with GetNoise
+ /// <code>DomainWarp(x, y)
+ /// noise = GetNoise(x, y)</code>
+ /// </example>
+ template <typename FNfloat>
+ void DomainWarp(FNfloat& x, FNfloat& y)
+ {
+ Arguments_must_be_floating_point_values<FNfloat>();
+
+ switch (mFractalType)
+ {
+ default:
+ DomainWarpSingle(x, y);
+ break;
+ case FractalType_DomainWarpProgressive:
+ DomainWarpFractalProgressive(x, y);
+ break;
+ case FractalType_DomainWarpIndependent:
+ DomainWarpFractalIndependent(x, y);
+ break;
+ }
+ }
+
+ /// <summary>
+ /// 3D warps the input position using current domain warp settings
+ /// </summary>
+ /// <example>
+ /// Example usage with GetNoise
+ /// <code>DomainWarp(x, y, z)
+ /// noise = GetNoise(x, y, z)</code>
+ /// </example>
+ template <typename FNfloat>
+ void DomainWarp(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ Arguments_must_be_floating_point_values<FNfloat>();
+
+ switch (mFractalType)
+ {
+ default:
+ DomainWarpSingle(x, y, z);
+ break;
+ case FractalType_DomainWarpProgressive:
+ DomainWarpFractalProgressive(x, y, z);
+ break;
+ case FractalType_DomainWarpIndependent:
+ DomainWarpFractalIndependent(x, y, z);
+ break;
+ }
+ }
+
+private:
+ template <typename T>
+ struct Arguments_must_be_floating_point_values;
+
+ enum TransformType3D
+ {
+ TransformType3D_None,
+ TransformType3D_ImproveXYPlanes,
+ TransformType3D_ImproveXZPlanes,
+ TransformType3D_DefaultOpenSimplex2
+ };
+
+ int mSeed;
+ float mFrequency;
+ NoiseType mNoiseType;
+ RotationType3D mRotationType3D;
+ TransformType3D mTransformType3D;
+
+ FractalType mFractalType;
+ int mOctaves;
+ float mLacunarity;
+ float mGain;
+ float mWeightedStrength;
+ float mPingPongStrength;
+
+ float mFractalBounding;
+
+ CellularDistanceFunction mCellularDistanceFunction;
+ CellularReturnType mCellularReturnType;
+ float mCellularJitterModifier;
+
+ DomainWarpType mDomainWarpType;
+ TransformType3D mWarpTransformType3D;
+ float mDomainWarpAmp;
+
+
+ template <typename T>
+ struct Lookup
+ {
+ static const T Gradients2D[];
+ static const T Gradients3D[];
+ static const T RandVecs2D[];
+ static const T RandVecs3D[];
+ };
+
+ static float FastMin(float a, float b) { return a < b ? a : b; }
+
+ static float FastMax(float a, float b) { return a > b ? a : b; }
+
+ static float FastAbs(float f) { return f < 0 ? -f : f; }
+
+ static float FastSqrt(float f) { return sqrtf(f); }
+
+ template <typename FNfloat>
+ static int FastFloor(FNfloat f) { return f >= 0 ? (int)f : (int)f - 1; }
+
+ template <typename FNfloat>
+ static int FastRound(FNfloat f) { return f >= 0 ? (int)(f + 0.5f) : (int)(f - 0.5f); }
+
+ static float Lerp(float a, float b, float t) { return a + t * (b - a); }
+
+ static float InterpHermite(float t) { return t * t * (3 - 2 * t); }
+
+ static float InterpQuintic(float t) { return t * t * t * (t * (t * 6 - 15) + 10); }
+
+ static float CubicLerp(float a, float b, float c, float d, float t)
+ {
+ float p = (d - c) - (a - b);
+ return t * t * t * p + t * t * ((a - b) - p) + t * (c - a) + b;
+ }
+
+ static float PingPong(float t)
+ {
+ t -= (int)(t * 0.5f) * 2;
+ return t < 1 ? t : 2 - t;
+ }
+
+ void CalculateFractalBounding()
+ {
+ float gain = FastAbs(mGain);
+ float amp = gain;
+ float ampFractal = 1.0f;
+ for (int i = 1; i < mOctaves; i++)
+ {
+ ampFractal += amp;
+ amp *= gain;
+ }
+ mFractalBounding = 1 / ampFractal;
+ }
+
+ // Hashing
+ static const int PrimeX = 501125321;
+ static const int PrimeY = 1136930381;
+ static const int PrimeZ = 1720413743;
+
+ static int Hash(int seed, int xPrimed, int yPrimed)
+ {
+ int hash = seed ^ xPrimed ^ yPrimed;
+
+ hash *= 0x27d4eb2d;
+ return hash;
+ }
+
+
+ static int Hash(int seed, int xPrimed, int yPrimed, int zPrimed)
+ {
+ int hash = seed ^ xPrimed ^ yPrimed ^ zPrimed;
+
+ hash *= 0x27d4eb2d;
+ return hash;
+ }
+
+
+ static float ValCoord(int seed, int xPrimed, int yPrimed)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+
+ hash *= hash;
+ hash ^= hash << 19;
+ return hash * (1 / 2147483648.0f);
+ }
+
+
+ static float ValCoord(int seed, int xPrimed, int yPrimed, int zPrimed)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+
+ hash *= hash;
+ hash ^= hash << 19;
+ return hash * (1 / 2147483648.0f);
+ }
+
+
+ float GradCoord(int seed, int xPrimed, int yPrimed, float xd, float yd)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+ hash ^= hash >> 15;
+ hash &= 127 << 1;
+
+ float xg = Lookup<float>::Gradients2D[hash];
+ float yg = Lookup<float>::Gradients2D[hash | 1];
+
+ return xd * xg + yd * yg;
+ }
+
+
+ float GradCoord(int seed, int xPrimed, int yPrimed, int zPrimed, float xd, float yd, float zd)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+ hash ^= hash >> 15;
+ hash &= 63 << 2;
+
+ float xg = Lookup<float>::Gradients3D[hash];
+ float yg = Lookup<float>::Gradients3D[hash | 1];
+ float zg = Lookup<float>::Gradients3D[hash | 2];
+
+ return xd * xg + yd * yg + zd * zg;
+ }
+
+
+ void GradCoordOut(int seed, int xPrimed, int yPrimed, float& xo, float& yo)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed) & (255 << 1);
+
+ xo = Lookup<float>::RandVecs2D[hash];
+ yo = Lookup<float>::RandVecs2D[hash | 1];
+ }
+
+
+ void GradCoordOut(int seed, int xPrimed, int yPrimed, int zPrimed, float& xo, float& yo, float& zo)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed) & (255 << 2);
+
+ xo = Lookup<float>::RandVecs3D[hash];
+ yo = Lookup<float>::RandVecs3D[hash | 1];
+ zo = Lookup<float>::RandVecs3D[hash | 2];
+ }
+
+
+ void GradCoordDual(int seed, int xPrimed, int yPrimed, float xd, float yd, float& xo, float& yo)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+ int index1 = hash & (127 << 1);
+ int index2 = (hash >> 7) & (255 << 1);
+
+ float xg = Lookup<float>::Gradients2D[index1];
+ float yg = Lookup<float>::Gradients2D[index1 | 1];
+ float value = xd * xg + yd * yg;
+
+ float xgo = Lookup<float>::RandVecs2D[index2];
+ float ygo = Lookup<float>::RandVecs2D[index2 | 1];
+
+ xo = value * xgo;
+ yo = value * ygo;
+ }
+
+
+ void GradCoordDual(int seed, int xPrimed, int yPrimed, int zPrimed, float xd, float yd, float zd, float& xo, float& yo, float& zo)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+ int index1 = hash & (63 << 2);
+ int index2 = (hash >> 6) & (255 << 2);
+
+ float xg = Lookup<float>::Gradients3D[index1];
+ float yg = Lookup<float>::Gradients3D[index1 | 1];
+ float zg = Lookup<float>::Gradients3D[index1 | 2];
+ float value = xd * xg + yd * yg + zd * zg;
+
+ float xgo = Lookup<float>::RandVecs3D[index2];
+ float ygo = Lookup<float>::RandVecs3D[index2 | 1];
+ float zgo = Lookup<float>::RandVecs3D[index2 | 2];
+
+ xo = value * xgo;
+ yo = value * ygo;
+ zo = value * zgo;
+ }
+
+
+ // Generic noise gen
+
+ template <typename FNfloat>
+ float GenNoiseSingle(int seed, FNfloat x, FNfloat y)
+ {
+ switch (mNoiseType)
+ {
+ case NoiseType_OpenSimplex2:
+ return SingleSimplex(seed, x, y);
+ case NoiseType_OpenSimplex2S:
+ return SingleOpenSimplex2S(seed, x, y);
+ case NoiseType_Cellular:
+ return SingleCellular(seed, x, y);
+ case NoiseType_Perlin:
+ return SinglePerlin(seed, x, y);
+ case NoiseType_ValueCubic:
+ return SingleValueCubic(seed, x, y);
+ case NoiseType_Value:
+ return SingleValue(seed, x, y);
+ default:
+ return 0;
+ }
+ }
+
+ template <typename FNfloat>
+ float GenNoiseSingle(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ switch (mNoiseType)
+ {
+ case NoiseType_OpenSimplex2:
+ return SingleOpenSimplex2(seed, x, y, z);
+ case NoiseType_OpenSimplex2S:
+ return SingleOpenSimplex2S(seed, x, y, z);
+ case NoiseType_Cellular:
+ return SingleCellular(seed, x, y, z);
+ case NoiseType_Perlin:
+ return SinglePerlin(seed, x, y, z);
+ case NoiseType_ValueCubic:
+ return SingleValueCubic(seed, x, y, z);
+ case NoiseType_Value:
+ return SingleValue(seed, x, y, z);
+ default:
+ return 0;
+ }
+ }
+
+
+ // Noise Coordinate Transforms (frequency, and possible skew or rotation)
+
+ template <typename FNfloat>
+ void TransformNoiseCoordinate(FNfloat& x, FNfloat& y)
+ {
+ x *= mFrequency;
+ y *= mFrequency;
+
+ switch (mNoiseType)
+ {
+ case NoiseType_OpenSimplex2:
+ case NoiseType_OpenSimplex2S:
+ {
+ const FNfloat SQRT3 = (FNfloat)1.7320508075688772935274463415059;
+ const FNfloat F2 = 0.5f * (SQRT3 - 1);
+ FNfloat t = (x + y) * F2;
+ x += t;
+ y += t;
+ }
+ break;
+ default:
+ break;
+ }
+ }
+
+ template <typename FNfloat>
+ void TransformNoiseCoordinate(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ x *= mFrequency;
+ y *= mFrequency;
+ z *= mFrequency;
+
+ switch (mTransformType3D)
+ {
+ case TransformType3D_ImproveXYPlanes:
+ {
+ FNfloat xy = x + y;
+ FNfloat s2 = xy * -(FNfloat)0.211324865405187;
+ z *= (FNfloat)0.577350269189626;
+ x += s2 - z;
+ y = y + s2 - z;
+ z += xy * (FNfloat)0.577350269189626;
+ }
+ break;
+ case TransformType3D_ImproveXZPlanes:
+ {
+ FNfloat xz = x + z;
+ FNfloat s2 = xz * -(FNfloat)0.211324865405187;
+ y *= (FNfloat)0.577350269189626;
+ x += s2 - y;
+ z += s2 - y;
+ y += xz * (FNfloat)0.577350269189626;
+ }
+ break;
+ case TransformType3D_DefaultOpenSimplex2:
+ {
+ const FNfloat R3 = (FNfloat)(2.0 / 3.0);
+ FNfloat r = (x + y + z) * R3; // Rotation, not skew
+ x = r - x;
+ y = r - y;
+ z = r - z;
+ }
+ break;
+ default:
+ break;
+ }
+ }
+
+ void UpdateTransformType3D()
+ {
+ switch (mRotationType3D)
+ {
+ case RotationType3D_ImproveXYPlanes:
+ mTransformType3D = TransformType3D_ImproveXYPlanes;
+ break;
+ case RotationType3D_ImproveXZPlanes:
+ mTransformType3D = TransformType3D_ImproveXZPlanes;
+ break;
+ default:
+ switch (mNoiseType)
+ {
+ case NoiseType_OpenSimplex2:
+ case NoiseType_OpenSimplex2S:
+ mTransformType3D = TransformType3D_DefaultOpenSimplex2;
+ break;
+ default:
+ mTransformType3D = TransformType3D_None;
+ break;
+ }
+ break;
+ }
+ }
+
+
+ // Domain Warp Coordinate Transforms
+
+ template <typename FNfloat>
+ void TransformDomainWarpCoordinate(FNfloat& x, FNfloat& y)
+ {
+ switch (mDomainWarpType)
+ {
+ case DomainWarpType_OpenSimplex2:
+ case DomainWarpType_OpenSimplex2Reduced:
+ {
+ const FNfloat SQRT3 = (FNfloat)1.7320508075688772935274463415059;
+ const FNfloat F2 = 0.5f * (SQRT3 - 1);
+ FNfloat t = (x + y) * F2;
+ x += t;
+ y += t;
+ }
+ break;
+ default:
+ break;
+ }
+ }
+
+ template <typename FNfloat>
+ void TransformDomainWarpCoordinate(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ switch (mWarpTransformType3D)
+ {
+ case TransformType3D_ImproveXYPlanes:
+ {
+ FNfloat xy = x + y;
+ FNfloat s2 = xy * -(FNfloat)0.211324865405187;
+ z *= (FNfloat)0.577350269189626;
+ x += s2 - z;
+ y = y + s2 - z;
+ z += xy * (FNfloat)0.577350269189626;
+ }
+ break;
+ case TransformType3D_ImproveXZPlanes:
+ {
+ FNfloat xz = x + z;
+ FNfloat s2 = xz * -(FNfloat)0.211324865405187;
+ y *= (FNfloat)0.577350269189626;
+ x += s2 - y;
+ z += s2 - y;
+ y += xz * (FNfloat)0.577350269189626;
+ }
+ break;
+ case TransformType3D_DefaultOpenSimplex2:
+ {
+ const FNfloat R3 = (FNfloat)(2.0 / 3.0);
+ FNfloat r = (x + y + z) * R3; // Rotation, not skew
+ x = r - x;
+ y = r - y;
+ z = r - z;
+ }
+ break;
+ default:
+ break;
+ }
+ }
+
+ void UpdateWarpTransformType3D()
+ {
+ switch (mRotationType3D)
+ {
+ case RotationType3D_ImproveXYPlanes:
+ mWarpTransformType3D = TransformType3D_ImproveXYPlanes;
+ break;
+ case RotationType3D_ImproveXZPlanes:
+ mWarpTransformType3D = TransformType3D_ImproveXZPlanes;
+ break;
+ default:
+ switch (mDomainWarpType)
+ {
+ case DomainWarpType_OpenSimplex2:
+ case DomainWarpType_OpenSimplex2Reduced:
+ mWarpTransformType3D = TransformType3D_DefaultOpenSimplex2;
+ break;
+ default:
+ mWarpTransformType3D = TransformType3D_None;
+ break;
+ }
+ break;
+ }
+ }
+
+
+ // Fractal FBm
+
+ template <typename FNfloat>
+ float GenFractalFBm(FNfloat x, FNfloat y)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = GenNoiseSingle(seed++, x, y);
+ sum += noise * amp;
+ amp *= Lerp(1.0f, FastMin(noise + 1, 2) * 0.5f, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+ template <typename FNfloat>
+ float GenFractalFBm(FNfloat x, FNfloat y, FNfloat z)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = GenNoiseSingle(seed++, x, y, z);
+ sum += noise * amp;
+ amp *= Lerp(1.0f, (noise + 1) * 0.5f, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ z *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+
+ // Fractal Ridged
+
+ template <typename FNfloat>
+ float GenFractalRidged(FNfloat x, FNfloat y)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = FastAbs(GenNoiseSingle(seed++, x, y));
+ sum += (noise * -2 + 1) * amp;
+ amp *= Lerp(1.0f, 1 - noise, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+ template <typename FNfloat>
+ float GenFractalRidged(FNfloat x, FNfloat y, FNfloat z)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = FastAbs(GenNoiseSingle(seed++, x, y, z));
+ sum += (noise * -2 + 1) * amp;
+ amp *= Lerp(1.0f, 1 - noise, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ z *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+
+ // Fractal PingPong
+
+ template <typename FNfloat>
+ float GenFractalPingPong(FNfloat x, FNfloat y)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = PingPong((GenNoiseSingle(seed++, x, y) + 1) * mPingPongStrength);
+ sum += (noise - 0.5f) * 2 * amp;
+ amp *= Lerp(1.0f, noise, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+ template <typename FNfloat>
+ float GenFractalPingPong(FNfloat x, FNfloat y, FNfloat z)
+ {
+ int seed = mSeed;
+ float sum = 0;
+ float amp = mFractalBounding;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ float noise = PingPong((GenNoiseSingle(seed++, x, y, z) + 1) * mPingPongStrength);
+ sum += (noise - 0.5f) * 2 * amp;
+ amp *= Lerp(1.0f, noise, mWeightedStrength);
+
+ x *= mLacunarity;
+ y *= mLacunarity;
+ z *= mLacunarity;
+ amp *= mGain;
+ }
+
+ return sum;
+ }
+
+
+ // Simplex/OpenSimplex2 Noise
+
+ template <typename FNfloat>
+ float SingleSimplex(int seed, FNfloat x, FNfloat y)
+ {
+ // 2D OpenSimplex2 case uses the same algorithm as ordinary Simplex.
+
+ const float SQRT3 = 1.7320508075688772935274463415059f;
+ const float G2 = (3 - SQRT3) / 6;
+
+ /*
+ * --- Skew moved to TransformNoiseCoordinate method ---
+ * const FNfloat F2 = 0.5f * (SQRT3 - 1);
+ * FNfloat s = (x + y) * F2;
+ * x += s; y += s;
+ */
+
+ int i = FastFloor(x);
+ int j = FastFloor(y);
+ float xi = (float)(x - i);
+ float yi = (float)(y - j);
+
+ float t = (xi + yi) * G2;
+ float x0 = (float)(xi - t);
+ float y0 = (float)(yi - t);
+
+ i *= PrimeX;
+ j *= PrimeY;
+
+ float n0, n1, n2;
+
+ float a = 0.5f - x0 * x0 - y0 * y0;
+ if (a <= 0) n0 = 0;
+ else
+ {
+ n0 = (a * a) * (a * a) * GradCoord(seed, i, j, x0, y0);
+ }
+
+ float c = (float)(2 * (1 - 2 * G2) * (1 / G2 - 2)) * t + ((float)(-2 * (1 - 2 * G2) * (1 - 2 * G2)) + a);
+ if (c <= 0) n2 = 0;
+ else
+ {
+ float x2 = x0 + (2 * (float)G2 - 1);
+ float y2 = y0 + (2 * (float)G2 - 1);
+ n2 = (c * c) * (c * c) * GradCoord(seed, i + PrimeX, j + PrimeY, x2, y2);
+ }
+
+ if (y0 > x0)
+ {
+ float x1 = x0 + (float)G2;
+ float y1 = y0 + ((float)G2 - 1);
+ float b = 0.5f - x1 * x1 - y1 * y1;
+ if (b <= 0) n1 = 0;
+ else
+ {
+ n1 = (b * b) * (b * b) * GradCoord(seed, i, j + PrimeY, x1, y1);
+ }
+ }
+ else
+ {
+ float x1 = x0 + ((float)G2 - 1);
+ float y1 = y0 + (float)G2;
+ float b = 0.5f - x1 * x1 - y1 * y1;
+ if (b <= 0) n1 = 0;
+ else
+ {
+ n1 = (b * b) * (b * b) * GradCoord(seed, i + PrimeX, j, x1, y1);
+ }
+ }
+
+ return (n0 + n1 + n2) * 99.83685446303647f;
+ }
+
+ template <typename FNfloat>
+ float SingleOpenSimplex2(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ // 3D OpenSimplex2 case uses two offset rotated cube grids.
+
+ /*
+ * --- Rotation moved to TransformNoiseCoordinate method ---
+ * const FNfloat R3 = (FNfloat)(2.0 / 3.0);
+ * FNfloat r = (x + y + z) * R3; // Rotation, not skew
+ * x = r - x; y = r - y; z = r - z;
+ */
+
+ int i = FastRound(x);
+ int j = FastRound(y);
+ int k = FastRound(z);
+ float x0 = (float)(x - i);
+ float y0 = (float)(y - j);
+ float z0 = (float)(z - k);
+
+ int xNSign = (int)(-1.0f - x0) | 1;
+ int yNSign = (int)(-1.0f - y0) | 1;
+ int zNSign = (int)(-1.0f - z0) | 1;
+
+ float ax0 = xNSign * -x0;
+ float ay0 = yNSign * -y0;
+ float az0 = zNSign * -z0;
+
+ i *= PrimeX;
+ j *= PrimeY;
+ k *= PrimeZ;
+
+ float value = 0;
+ float a = (0.6f - x0 * x0) - (y0 * y0 + z0 * z0);
+
+ for (int l = 0; ; l++)
+ {
+ if (a > 0)
+ {
+ value += (a * a) * (a * a) * GradCoord(seed, i, j, k, x0, y0, z0);
+ }
+
+ float b = a + 1;
+ int i1 = i;
+ int j1 = j;
+ int k1 = k;
+ float x1 = x0;
+ float y1 = y0;
+ float z1 = z0;
+
+ if (ax0 >= ay0 && ax0 >= az0)
+ {
+ x1 += xNSign;
+ b -= xNSign * 2 * x1;
+ i1 -= xNSign * PrimeX;
+ }
+ else if (ay0 > ax0 && ay0 >= az0)
+ {
+ y1 += yNSign;
+ b -= yNSign * 2 * y1;
+ j1 -= yNSign * PrimeY;
+ }
+ else
+ {
+ z1 += zNSign;
+ b -= zNSign * 2 * z1;
+ k1 -= zNSign * PrimeZ;
+ }
+
+ if (b > 0)
+ {
+ value += (b * b) * (b * b) * GradCoord(seed, i1, j1, k1, x1, y1, z1);
+ }
+
+ if (l == 1) break;
+
+ ax0 = 0.5f - ax0;
+ ay0 = 0.5f - ay0;
+ az0 = 0.5f - az0;
+
+ x0 = xNSign * ax0;
+ y0 = yNSign * ay0;
+ z0 = zNSign * az0;
+
+ a += (0.75f - ax0) - (ay0 + az0);
+
+ i += (xNSign >> 1) & PrimeX;
+ j += (yNSign >> 1) & PrimeY;
+ k += (zNSign >> 1) & PrimeZ;
+
+ xNSign = -xNSign;
+ yNSign = -yNSign;
+ zNSign = -zNSign;
+
+ seed = ~seed;
+ }
+
+ return value * 32.69428253173828125f;
+ }
+
+
+ // OpenSimplex2S Noise
+
+ template <typename FNfloat>
+ float SingleOpenSimplex2S(int seed, FNfloat x, FNfloat y)
+ {
+ // 2D OpenSimplex2S case is a modified 2D simplex noise.
+
+ const FNfloat SQRT3 = (FNfloat)1.7320508075688772935274463415059;
+ const FNfloat G2 = (3 - SQRT3) / 6;
+
+ /*
+ * --- Skew moved to TransformNoiseCoordinate method ---
+ * const FNfloat F2 = 0.5f * (SQRT3 - 1);
+ * FNfloat s = (x + y) * F2;
+ * x += s; y += s;
+ */
+
+ int i = FastFloor(x);
+ int j = FastFloor(y);
+ float xi = (float)(x - i);
+ float yi = (float)(y - j);
+
+ i *= PrimeX;
+ j *= PrimeY;
+ int i1 = i + PrimeX;
+ int j1 = j + PrimeY;
+
+ float t = (xi + yi) * (float)G2;
+ float x0 = xi - t;
+ float y0 = yi - t;
+
+ float a0 = (2.0f / 3.0f) - x0 * x0 - y0 * y0;
+ float value = (a0 * a0) * (a0 * a0) * GradCoord(seed, i, j, x0, y0);
+
+ float a1 = (float)(2 * (1 - 2 * G2) * (1 / G2 - 2)) * t + ((float)(-2 * (1 - 2 * G2) * (1 - 2 * G2)) + a0);
+ float x1 = x0 - (float)(1 - 2 * G2);
+ float y1 = y0 - (float)(1 - 2 * G2);
+ value += (a1 * a1) * (a1 * a1) * GradCoord(seed, i1, j1, x1, y1);
+
+ // Nested conditionals were faster than compact bit logic/arithmetic.
+ float xmyi = xi - yi;
+ if (t > G2)
+ {
+ if (xi + xmyi > 1)
+ {
+ float x2 = x0 + (float)(3 * G2 - 2);
+ float y2 = y0 + (float)(3 * G2 - 1);
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i + (PrimeX << 1), j + PrimeY, x2, y2);
+ }
+ }
+ else
+ {
+ float x2 = x0 + (float)G2;
+ float y2 = y0 + (float)(G2 - 1);
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i, j + PrimeY, x2, y2);
+ }
+ }
+
+ if (yi - xmyi > 1)
+ {
+ float x3 = x0 + (float)(3 * G2 - 1);
+ float y3 = y0 + (float)(3 * G2 - 2);
+ float a3 = (2.0f / 3.0f) - x3 * x3 - y3 * y3;
+ if (a3 > 0)
+ {
+ value += (a3 * a3) * (a3 * a3) * GradCoord(seed, i + PrimeX, j + (PrimeY << 1), x3, y3);
+ }
+ }
+ else
+ {
+ float x3 = x0 + (float)(G2 - 1);
+ float y3 = y0 + (float)G2;
+ float a3 = (2.0f / 3.0f) - x3 * x3 - y3 * y3;
+ if (a3 > 0)
+ {
+ value += (a3 * a3) * (a3 * a3) * GradCoord(seed, i + PrimeX, j, x3, y3);
+ }
+ }
+ }
+ else
+ {
+ if (xi + xmyi < 0)
+ {
+ float x2 = x0 + (float)(1 - G2);
+ float y2 = y0 - (float)G2;
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i - PrimeX, j, x2, y2);
+ }
+ }
+ else
+ {
+ float x2 = x0 + (float)(G2 - 1);
+ float y2 = y0 + (float)G2;
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i + PrimeX, j, x2, y2);
+ }
+ }
+
+ if (yi < xmyi)
+ {
+ float x2 = x0 - (float)G2;
+ float y2 = y0 - (float)(G2 - 1);
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i, j - PrimeY, x2, y2);
+ }
+ }
+ else
+ {
+ float x2 = x0 + (float)G2;
+ float y2 = y0 + (float)(G2 - 1);
+ float a2 = (2.0f / 3.0f) - x2 * x2 - y2 * y2;
+ if (a2 > 0)
+ {
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed, i, j + PrimeY, x2, y2);
+ }
+ }
+ }
+
+ return value * 18.24196194486065f;
+ }
+
+ template <typename FNfloat>
+ float SingleOpenSimplex2S(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ // 3D OpenSimplex2S case uses two offset rotated cube grids.
+
+ /*
+ * --- Rotation moved to TransformNoiseCoordinate method ---
+ * const FNfloat R3 = (FNfloat)(2.0 / 3.0);
+ * FNfloat r = (x + y + z) * R3; // Rotation, not skew
+ * x = r - x; y = r - y; z = r - z;
+ */
+
+ int i = FastFloor(x);
+ int j = FastFloor(y);
+ int k = FastFloor(z);
+ float xi = (float)(x - i);
+ float yi = (float)(y - j);
+ float zi = (float)(z - k);
+
+ i *= PrimeX;
+ j *= PrimeY;
+ k *= PrimeZ;
+ int seed2 = seed + 1293373;
+
+ int xNMask = (int)(-0.5f - xi);
+ int yNMask = (int)(-0.5f - yi);
+ int zNMask = (int)(-0.5f - zi);
+
+ float x0 = xi + xNMask;
+ float y0 = yi + yNMask;
+ float z0 = zi + zNMask;
+ float a0 = 0.75f - x0 * x0 - y0 * y0 - z0 * z0;
+ float value = (a0 * a0) * (a0 * a0) * GradCoord(seed,
+ i + (xNMask & PrimeX), j + (yNMask & PrimeY), k + (zNMask & PrimeZ), x0, y0, z0);
+
+ float x1 = xi - 0.5f;
+ float y1 = yi - 0.5f;
+ float z1 = zi - 0.5f;
+ float a1 = 0.75f - x1 * x1 - y1 * y1 - z1 * z1;
+ value += (a1 * a1) * (a1 * a1) * GradCoord(seed2,
+ i + PrimeX, j + PrimeY, k + PrimeZ, x1, y1, z1);
+
+ float xAFlipMask0 = ((xNMask | 1) << 1) * x1;
+ float yAFlipMask0 = ((yNMask | 1) << 1) * y1;
+ float zAFlipMask0 = ((zNMask | 1) << 1) * z1;
+ float xAFlipMask1 = (-2 - (xNMask << 2)) * x1 - 1.0f;
+ float yAFlipMask1 = (-2 - (yNMask << 2)) * y1 - 1.0f;
+ float zAFlipMask1 = (-2 - (zNMask << 2)) * z1 - 1.0f;
+
+ bool skip5 = false;
+ float a2 = xAFlipMask0 + a0;
+ if (a2 > 0)
+ {
+ float x2 = x0 - (xNMask | 1);
+ float y2 = y0;
+ float z2 = z0;
+ value += (a2 * a2) * (a2 * a2) * GradCoord(seed,
+ i + (~xNMask & PrimeX), j + (yNMask & PrimeY), k + (zNMask & PrimeZ), x2, y2, z2);
+ }
+ else
+ {
+ float a3 = yAFlipMask0 + zAFlipMask0 + a0;
+ if (a3 > 0)
+ {
+ float x3 = x0;
+ float y3 = y0 - (yNMask | 1);
+ float z3 = z0 - (zNMask | 1);
+ value += (a3 * a3) * (a3 * a3) * GradCoord(seed,
+ i + (xNMask & PrimeX), j + (~yNMask & PrimeY), k + (~zNMask & PrimeZ), x3, y3, z3);
+ }
+
+ float a4 = xAFlipMask1 + a1;
+ if (a4 > 0)
+ {
+ float x4 = (xNMask | 1) + x1;
+ float y4 = y1;
+ float z4 = z1;
+ value += (a4 * a4) * (a4 * a4) * GradCoord(seed2,
+ i + (xNMask & (PrimeX * 2)), j + PrimeY, k + PrimeZ, x4, y4, z4);
+ skip5 = true;
+ }
+ }
+
+ bool skip9 = false;
+ float a6 = yAFlipMask0 + a0;
+ if (a6 > 0)
+ {
+ float x6 = x0;
+ float y6 = y0 - (yNMask | 1);
+ float z6 = z0;
+ value += (a6 * a6) * (a6 * a6) * GradCoord(seed,
+ i + (xNMask & PrimeX), j + (~yNMask & PrimeY), k + (zNMask & PrimeZ), x6, y6, z6);
+ }
+ else
+ {
+ float a7 = xAFlipMask0 + zAFlipMask0 + a0;
+ if (a7 > 0)
+ {
+ float x7 = x0 - (xNMask | 1);
+ float y7 = y0;
+ float z7 = z0 - (zNMask | 1);
+ value += (a7 * a7) * (a7 * a7) * GradCoord(seed,
+ i + (~xNMask & PrimeX), j + (yNMask & PrimeY), k + (~zNMask & PrimeZ), x7, y7, z7);
+ }
+
+ float a8 = yAFlipMask1 + a1;
+ if (a8 > 0)
+ {
+ float x8 = x1;
+ float y8 = (yNMask | 1) + y1;
+ float z8 = z1;
+ value += (a8 * a8) * (a8 * a8) * GradCoord(seed2,
+ i + PrimeX, j + (yNMask & (PrimeY << 1)), k + PrimeZ, x8, y8, z8);
+ skip9 = true;
+ }
+ }
+
+ bool skipD = false;
+ float aA = zAFlipMask0 + a0;
+ if (aA > 0)
+ {
+ float xA = x0;
+ float yA = y0;
+ float zA = z0 - (zNMask | 1);
+ value += (aA * aA) * (aA * aA) * GradCoord(seed,
+ i + (xNMask & PrimeX), j + (yNMask & PrimeY), k + (~zNMask & PrimeZ), xA, yA, zA);
+ }
+ else
+ {
+ float aB = xAFlipMask0 + yAFlipMask0 + a0;
+ if (aB > 0)
+ {
+ float xB = x0 - (xNMask | 1);
+ float yB = y0 - (yNMask | 1);
+ float zB = z0;
+ value += (aB * aB) * (aB * aB) * GradCoord(seed,
+ i + (~xNMask & PrimeX), j + (~yNMask & PrimeY), k + (zNMask & PrimeZ), xB, yB, zB);
+ }
+
+ float aC = zAFlipMask1 + a1;
+ if (aC > 0)
+ {
+ float xC = x1;
+ float yC = y1;
+ float zC = (zNMask | 1) + z1;
+ value += (aC * aC) * (aC * aC) * GradCoord(seed2,
+ i + PrimeX, j + PrimeY, k + (zNMask & (PrimeZ << 1)), xC, yC, zC);
+ skipD = true;
+ }
+ }
+
+ if (!skip5)
+ {
+ float a5 = yAFlipMask1 + zAFlipMask1 + a1;
+ if (a5 > 0)
+ {
+ float x5 = x1;
+ float y5 = (yNMask | 1) + y1;
+ float z5 = (zNMask | 1) + z1;
+ value += (a5 * a5) * (a5 * a5) * GradCoord(seed2,
+ i + PrimeX, j + (yNMask & (PrimeY << 1)), k + (zNMask & (PrimeZ << 1)), x5, y5, z5);
+ }
+ }
+
+ if (!skip9)
+ {
+ float a9 = xAFlipMask1 + zAFlipMask1 + a1;
+ if (a9 > 0)
+ {
+ float x9 = (xNMask | 1) + x1;
+ float y9 = y1;
+ float z9 = (zNMask | 1) + z1;
+ value += (a9 * a9) * (a9 * a9) * GradCoord(seed2,
+ i + (xNMask & (PrimeX * 2)), j + PrimeY, k + (zNMask & (PrimeZ << 1)), x9, y9, z9);
+ }
+ }
+
+ if (!skipD)
+ {
+ float aD = xAFlipMask1 + yAFlipMask1 + a1;
+ if (aD > 0)
+ {
+ float xD = (xNMask | 1) + x1;
+ float yD = (yNMask | 1) + y1;
+ float zD = z1;
+ value += (aD * aD) * (aD * aD) * GradCoord(seed2,
+ i + (xNMask & (PrimeX << 1)), j + (yNMask & (PrimeY << 1)), k + PrimeZ, xD, yD, zD);
+ }
+ }
+
+ return value * 9.046026385208288f;
+ }
+
+
+ // Cellular Noise
+
+ template <typename FNfloat>
+ float SingleCellular(int seed, FNfloat x, FNfloat y)
+ {
+ int xr = FastRound(x);
+ int yr = FastRound(y);
+
+ float distance0 = 1e10f;
+ float distance1 = 1e10f;
+ int closestHash = 0;
+
+ float cellularJitter = 0.43701595f * mCellularJitterModifier;
+
+ int xPrimed = (xr - 1) * PrimeX;
+ int yPrimedBase = (yr - 1) * PrimeY;
+
+ switch (mCellularDistanceFunction)
+ {
+ default:
+ case CellularDistanceFunction_Euclidean:
+ case CellularDistanceFunction_EuclideanSq:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+ int idx = hash & (255 << 1);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs2D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs2D[idx | 1] * cellularJitter;
+
+ float newDistance = vecX * vecX + vecY * vecY;
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ case CellularDistanceFunction_Manhattan:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+ int idx = hash & (255 << 1);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs2D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs2D[idx | 1] * cellularJitter;
+
+ float newDistance = FastAbs(vecX) + FastAbs(vecY);
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ case CellularDistanceFunction_Hybrid:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed);
+ int idx = hash & (255 << 1);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs2D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs2D[idx | 1] * cellularJitter;
+
+ float newDistance = (FastAbs(vecX) + FastAbs(vecY)) + (vecX * vecX + vecY * vecY);
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ }
+
+ if (mCellularDistanceFunction == CellularDistanceFunction_Euclidean && mCellularReturnType >= CellularReturnType_Distance)
+ {
+ distance0 = FastSqrt(distance0);
+
+ if (mCellularReturnType >= CellularReturnType_Distance2)
+ {
+ distance1 = FastSqrt(distance1);
+ }
+ }
+
+ switch (mCellularReturnType)
+ {
+ case CellularReturnType_CellValue:
+ return closestHash * (1 / 2147483648.0f);
+ case CellularReturnType_Distance:
+ return distance0 - 1;
+ case CellularReturnType_Distance2:
+ return distance1 - 1;
+ case CellularReturnType_Distance2Add:
+ return (distance1 + distance0) * 0.5f - 1;
+ case CellularReturnType_Distance2Sub:
+ return distance1 - distance0 - 1;
+ case CellularReturnType_Distance2Mul:
+ return distance1 * distance0 * 0.5f - 1;
+ case CellularReturnType_Distance2Div:
+ return distance0 / distance1 - 1;
+ default:
+ return 0;
+ }
+ }
+
+ template <typename FNfloat>
+ float SingleCellular(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ int xr = FastRound(x);
+ int yr = FastRound(y);
+ int zr = FastRound(z);
+
+ float distance0 = 1e10f;
+ float distance1 = 1e10f;
+ int closestHash = 0;
+
+ float cellularJitter = 0.39614353f * mCellularJitterModifier;
+
+ int xPrimed = (xr - 1) * PrimeX;
+ int yPrimedBase = (yr - 1) * PrimeY;
+ int zPrimedBase = (zr - 1) * PrimeZ;
+
+ switch (mCellularDistanceFunction)
+ {
+ case CellularDistanceFunction_Euclidean:
+ case CellularDistanceFunction_EuclideanSq:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int zPrimed = zPrimedBase;
+
+ for (int zi = zr - 1; zi <= zr + 1; zi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+ int idx = hash & (255 << 2);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs3D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs3D[idx | 1] * cellularJitter;
+ float vecZ = (float)(zi - z) + Lookup<float>::RandVecs3D[idx | 2] * cellularJitter;
+
+ float newDistance = vecX * vecX + vecY * vecY + vecZ * vecZ;
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ zPrimed += PrimeZ;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ case CellularDistanceFunction_Manhattan:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int zPrimed = zPrimedBase;
+
+ for (int zi = zr - 1; zi <= zr + 1; zi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+ int idx = hash & (255 << 2);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs3D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs3D[idx | 1] * cellularJitter;
+ float vecZ = (float)(zi - z) + Lookup<float>::RandVecs3D[idx | 2] * cellularJitter;
+
+ float newDistance = FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ);
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ zPrimed += PrimeZ;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ case CellularDistanceFunction_Hybrid:
+ for (int xi = xr - 1; xi <= xr + 1; xi++)
+ {
+ int yPrimed = yPrimedBase;
+
+ for (int yi = yr - 1; yi <= yr + 1; yi++)
+ {
+ int zPrimed = zPrimedBase;
+
+ for (int zi = zr - 1; zi <= zr + 1; zi++)
+ {
+ int hash = Hash(seed, xPrimed, yPrimed, zPrimed);
+ int idx = hash & (255 << 2);
+
+ float vecX = (float)(xi - x) + Lookup<float>::RandVecs3D[idx] * cellularJitter;
+ float vecY = (float)(yi - y) + Lookup<float>::RandVecs3D[idx | 1] * cellularJitter;
+ float vecZ = (float)(zi - z) + Lookup<float>::RandVecs3D[idx | 2] * cellularJitter;
+
+ float newDistance = (FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ)) + (vecX * vecX + vecY * vecY + vecZ * vecZ);
+
+ distance1 = FastMax(FastMin(distance1, newDistance), distance0);
+ if (newDistance < distance0)
+ {
+ distance0 = newDistance;
+ closestHash = hash;
+ }
+ zPrimed += PrimeZ;
+ }
+ yPrimed += PrimeY;
+ }
+ xPrimed += PrimeX;
+ }
+ break;
+ default:
+ break;
+ }
+
+ if (mCellularDistanceFunction == CellularDistanceFunction_Euclidean && mCellularReturnType >= CellularReturnType_Distance)
+ {
+ distance0 = FastSqrt(distance0);
+
+ if (mCellularReturnType >= CellularReturnType_Distance2)
+ {
+ distance1 = FastSqrt(distance1);
+ }
+ }
+
+ switch (mCellularReturnType)
+ {
+ case CellularReturnType_CellValue:
+ return closestHash * (1 / 2147483648.0f);
+ case CellularReturnType_Distance:
+ return distance0 - 1;
+ case CellularReturnType_Distance2:
+ return distance1 - 1;
+ case CellularReturnType_Distance2Add:
+ return (distance1 + distance0) * 0.5f - 1;
+ case CellularReturnType_Distance2Sub:
+ return distance1 - distance0 - 1;
+ case CellularReturnType_Distance2Mul:
+ return distance1 * distance0 * 0.5f - 1;
+ case CellularReturnType_Distance2Div:
+ return distance0 / distance1 - 1;
+ default:
+ return 0;
+ }
+ }
+
+
+ // Perlin Noise
+
+ template <typename FNfloat>
+ float SinglePerlin(int seed, FNfloat x, FNfloat y)
+ {
+ int x0 = FastFloor(x);
+ int y0 = FastFloor(y);
+
+ float xd0 = (float)(x - x0);
+ float yd0 = (float)(y - y0);
+ float xd1 = xd0 - 1;
+ float yd1 = yd0 - 1;
+
+ float xs = InterpQuintic(xd0);
+ float ys = InterpQuintic(yd0);
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+
+ float xf0 = Lerp(GradCoord(seed, x0, y0, xd0, yd0), GradCoord(seed, x1, y0, xd1, yd0), xs);
+ float xf1 = Lerp(GradCoord(seed, x0, y1, xd0, yd1), GradCoord(seed, x1, y1, xd1, yd1), xs);
+
+ return Lerp(xf0, xf1, ys) * 1.4247691104677813f;
+ }
+
+ template <typename FNfloat>
+ float SinglePerlin(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ int x0 = FastFloor(x);
+ int y0 = FastFloor(y);
+ int z0 = FastFloor(z);
+
+ float xd0 = (float)(x - x0);
+ float yd0 = (float)(y - y0);
+ float zd0 = (float)(z - z0);
+ float xd1 = xd0 - 1;
+ float yd1 = yd0 - 1;
+ float zd1 = zd0 - 1;
+
+ float xs = InterpQuintic(xd0);
+ float ys = InterpQuintic(yd0);
+ float zs = InterpQuintic(zd0);
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ z0 *= PrimeZ;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+ int z1 = z0 + PrimeZ;
+
+ float xf00 = Lerp(GradCoord(seed, x0, y0, z0, xd0, yd0, zd0), GradCoord(seed, x1, y0, z0, xd1, yd0, zd0), xs);
+ float xf10 = Lerp(GradCoord(seed, x0, y1, z0, xd0, yd1, zd0), GradCoord(seed, x1, y1, z0, xd1, yd1, zd0), xs);
+ float xf01 = Lerp(GradCoord(seed, x0, y0, z1, xd0, yd0, zd1), GradCoord(seed, x1, y0, z1, xd1, yd0, zd1), xs);
+ float xf11 = Lerp(GradCoord(seed, x0, y1, z1, xd0, yd1, zd1), GradCoord(seed, x1, y1, z1, xd1, yd1, zd1), xs);
+
+ float yf0 = Lerp(xf00, xf10, ys);
+ float yf1 = Lerp(xf01, xf11, ys);
+
+ return Lerp(yf0, yf1, zs) * 0.964921414852142333984375f;
+ }
+
+
+ // Value Cubic Noise
+
+ template <typename FNfloat>
+ float SingleValueCubic(int seed, FNfloat x, FNfloat y)
+ {
+ int x1 = FastFloor(x);
+ int y1 = FastFloor(y);
+
+ float xs = (float)(x - x1);
+ float ys = (float)(y - y1);
+
+ x1 *= PrimeX;
+ y1 *= PrimeY;
+ int x0 = x1 - PrimeX;
+ int y0 = y1 - PrimeY;
+ int x2 = x1 + PrimeX;
+ int y2 = y1 + PrimeY;
+ int x3 = x1 + (int)((long)PrimeX << 1);
+ int y3 = y1 + (int)((long)PrimeY << 1);
+
+ return CubicLerp(
+ CubicLerp(ValCoord(seed, x0, y0), ValCoord(seed, x1, y0), ValCoord(seed, x2, y0), ValCoord(seed, x3, y0),
+ xs),
+ CubicLerp(ValCoord(seed, x0, y1), ValCoord(seed, x1, y1), ValCoord(seed, x2, y1), ValCoord(seed, x3, y1),
+ xs),
+ CubicLerp(ValCoord(seed, x0, y2), ValCoord(seed, x1, y2), ValCoord(seed, x2, y2), ValCoord(seed, x3, y2),
+ xs),
+ CubicLerp(ValCoord(seed, x0, y3), ValCoord(seed, x1, y3), ValCoord(seed, x2, y3), ValCoord(seed, x3, y3),
+ xs),
+ ys) * (1 / (1.5f * 1.5f));
+ }
+
+ template <typename FNfloat>
+ float SingleValueCubic(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ int x1 = FastFloor(x);
+ int y1 = FastFloor(y);
+ int z1 = FastFloor(z);
+
+ float xs = (float)(x - x1);
+ float ys = (float)(y - y1);
+ float zs = (float)(z - z1);
+
+ x1 *= PrimeX;
+ y1 *= PrimeY;
+ z1 *= PrimeZ;
+
+ int x0 = x1 - PrimeX;
+ int y0 = y1 - PrimeY;
+ int z0 = z1 - PrimeZ;
+ int x2 = x1 + PrimeX;
+ int y2 = y1 + PrimeY;
+ int z2 = z1 + PrimeZ;
+ int x3 = x1 + (int)((long)PrimeX << 1);
+ int y3 = y1 + (int)((long)PrimeY << 1);
+ int z3 = z1 + (int)((long)PrimeZ << 1);
+
+
+ return CubicLerp(
+ CubicLerp(
+ CubicLerp(ValCoord(seed, x0, y0, z0), ValCoord(seed, x1, y0, z0), ValCoord(seed, x2, y0, z0), ValCoord(seed, x3, y0, z0), xs),
+ CubicLerp(ValCoord(seed, x0, y1, z0), ValCoord(seed, x1, y1, z0), ValCoord(seed, x2, y1, z0), ValCoord(seed, x3, y1, z0), xs),
+ CubicLerp(ValCoord(seed, x0, y2, z0), ValCoord(seed, x1, y2, z0), ValCoord(seed, x2, y2, z0), ValCoord(seed, x3, y2, z0), xs),
+ CubicLerp(ValCoord(seed, x0, y3, z0), ValCoord(seed, x1, y3, z0), ValCoord(seed, x2, y3, z0), ValCoord(seed, x3, y3, z0), xs),
+ ys),
+ CubicLerp(
+ CubicLerp(ValCoord(seed, x0, y0, z1), ValCoord(seed, x1, y0, z1), ValCoord(seed, x2, y0, z1), ValCoord(seed, x3, y0, z1), xs),
+ CubicLerp(ValCoord(seed, x0, y1, z1), ValCoord(seed, x1, y1, z1), ValCoord(seed, x2, y1, z1), ValCoord(seed, x3, y1, z1), xs),
+ CubicLerp(ValCoord(seed, x0, y2, z1), ValCoord(seed, x1, y2, z1), ValCoord(seed, x2, y2, z1), ValCoord(seed, x3, y2, z1), xs),
+ CubicLerp(ValCoord(seed, x0, y3, z1), ValCoord(seed, x1, y3, z1), ValCoord(seed, x2, y3, z1), ValCoord(seed, x3, y3, z1), xs),
+ ys),
+ CubicLerp(
+ CubicLerp(ValCoord(seed, x0, y0, z2), ValCoord(seed, x1, y0, z2), ValCoord(seed, x2, y0, z2), ValCoord(seed, x3, y0, z2), xs),
+ CubicLerp(ValCoord(seed, x0, y1, z2), ValCoord(seed, x1, y1, z2), ValCoord(seed, x2, y1, z2), ValCoord(seed, x3, y1, z2), xs),
+ CubicLerp(ValCoord(seed, x0, y2, z2), ValCoord(seed, x1, y2, z2), ValCoord(seed, x2, y2, z2), ValCoord(seed, x3, y2, z2), xs),
+ CubicLerp(ValCoord(seed, x0, y3, z2), ValCoord(seed, x1, y3, z2), ValCoord(seed, x2, y3, z2), ValCoord(seed, x3, y3, z2), xs),
+ ys),
+ CubicLerp(
+ CubicLerp(ValCoord(seed, x0, y0, z3), ValCoord(seed, x1, y0, z3), ValCoord(seed, x2, y0, z3), ValCoord(seed, x3, y0, z3), xs),
+ CubicLerp(ValCoord(seed, x0, y1, z3), ValCoord(seed, x1, y1, z3), ValCoord(seed, x2, y1, z3), ValCoord(seed, x3, y1, z3), xs),
+ CubicLerp(ValCoord(seed, x0, y2, z3), ValCoord(seed, x1, y2, z3), ValCoord(seed, x2, y2, z3), ValCoord(seed, x3, y2, z3), xs),
+ CubicLerp(ValCoord(seed, x0, y3, z3), ValCoord(seed, x1, y3, z3), ValCoord(seed, x2, y3, z3), ValCoord(seed, x3, y3, z3), xs),
+ ys),
+ zs) * (1 / (1.5f * 1.5f * 1.5f));
+ }
+
+
+ // Value Noise
+
+ template <typename FNfloat>
+ float SingleValue(int seed, FNfloat x, FNfloat y)
+ {
+ int x0 = FastFloor(x);
+ int y0 = FastFloor(y);
+
+ float xs = InterpHermite((float)(x - x0));
+ float ys = InterpHermite((float)(y - y0));
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+
+ float xf0 = Lerp(ValCoord(seed, x0, y0), ValCoord(seed, x1, y0), xs);
+ float xf1 = Lerp(ValCoord(seed, x0, y1), ValCoord(seed, x1, y1), xs);
+
+ return Lerp(xf0, xf1, ys);
+ }
+
+ template <typename FNfloat>
+ float SingleValue(int seed, FNfloat x, FNfloat y, FNfloat z)
+ {
+ int x0 = FastFloor(x);
+ int y0 = FastFloor(y);
+ int z0 = FastFloor(z);
+
+ float xs = InterpHermite((float)(x - x0));
+ float ys = InterpHermite((float)(y - y0));
+ float zs = InterpHermite((float)(z - z0));
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ z0 *= PrimeZ;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+ int z1 = z0 + PrimeZ;
+
+ float xf00 = Lerp(ValCoord(seed, x0, y0, z0), ValCoord(seed, x1, y0, z0), xs);
+ float xf10 = Lerp(ValCoord(seed, x0, y1, z0), ValCoord(seed, x1, y1, z0), xs);
+ float xf01 = Lerp(ValCoord(seed, x0, y0, z1), ValCoord(seed, x1, y0, z1), xs);
+ float xf11 = Lerp(ValCoord(seed, x0, y1, z1), ValCoord(seed, x1, y1, z1), xs);
+
+ float yf0 = Lerp(xf00, xf10, ys);
+ float yf1 = Lerp(xf01, xf11, ys);
+
+ return Lerp(yf0, yf1, zs);
+ }
+
+
+ // Domain Warp
+
+ template <typename FNfloat>
+ void DoSingleDomainWarp(int seed, float amp, float freq, FNfloat x, FNfloat y, FNfloat& xr, FNfloat& yr)
+ {
+ switch (mDomainWarpType)
+ {
+ case DomainWarpType_OpenSimplex2:
+ SingleDomainWarpSimplexGradient(seed, amp * 38.283687591552734375f, freq, x, y, xr, yr, false);
+ break;
+ case DomainWarpType_OpenSimplex2Reduced:
+ SingleDomainWarpSimplexGradient(seed, amp * 16.0f, freq, x, y, xr, yr, true);
+ break;
+ case DomainWarpType_BasicGrid:
+ SingleDomainWarpBasicGrid(seed, amp, freq, x, y, xr, yr);
+ break;
+ }
+ }
+
+ template <typename FNfloat>
+ void DoSingleDomainWarp(int seed, float amp, float freq, FNfloat x, FNfloat y, FNfloat z, FNfloat& xr, FNfloat& yr, FNfloat& zr)
+ {
+ switch (mDomainWarpType)
+ {
+ case DomainWarpType_OpenSimplex2:
+ SingleDomainWarpOpenSimplex2Gradient(seed, amp * 32.69428253173828125f, freq, x, y, z, xr, yr, zr, false);
+ break;
+ case DomainWarpType_OpenSimplex2Reduced:
+ SingleDomainWarpOpenSimplex2Gradient(seed, amp * 7.71604938271605f, freq, x, y, z, xr, yr, zr, true);
+ break;
+ case DomainWarpType_BasicGrid:
+ SingleDomainWarpBasicGrid(seed, amp, freq, x, y, z, xr, yr, zr);
+ break;
+ }
+ }
+
+
+ // Domain Warp Single Wrapper
+
+ template <typename FNfloat>
+ void DomainWarpSingle(FNfloat& x, FNfloat& y)
+ {
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ FNfloat xs = x;
+ FNfloat ys = y;
+ TransformDomainWarpCoordinate(xs, ys);
+
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, x, y);
+ }
+
+ template <typename FNfloat>
+ void DomainWarpSingle(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ FNfloat xs = x;
+ FNfloat ys = y;
+ FNfloat zs = z;
+ TransformDomainWarpCoordinate(xs, ys, zs);
+
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, zs, x, y, z);
+ }
+
+
+ // Domain Warp Fractal Progressive
+
+ template <typename FNfloat>
+ void DomainWarpFractalProgressive(FNfloat& x, FNfloat& y)
+ {
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ FNfloat xs = x;
+ FNfloat ys = y;
+ TransformDomainWarpCoordinate(xs, ys);
+
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, x, y);
+
+ seed++;
+ amp *= mGain;
+ freq *= mLacunarity;
+ }
+ }
+
+ template <typename FNfloat>
+ void DomainWarpFractalProgressive(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ FNfloat xs = x;
+ FNfloat ys = y;
+ FNfloat zs = z;
+ TransformDomainWarpCoordinate(xs, ys, zs);
+
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, zs, x, y, z);
+
+ seed++;
+ amp *= mGain;
+ freq *= mLacunarity;
+ }
+ }
+
+
+ // Domain Warp Fractal Independant
+
+ template <typename FNfloat>
+ void DomainWarpFractalIndependent(FNfloat& x, FNfloat& y)
+ {
+ FNfloat xs = x;
+ FNfloat ys = y;
+ TransformDomainWarpCoordinate(xs, ys);
+
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, x, y);
+
+ seed++;
+ amp *= mGain;
+ freq *= mLacunarity;
+ }
+ }
+
+ template <typename FNfloat>
+ void DomainWarpFractalIndependent(FNfloat& x, FNfloat& y, FNfloat& z)
+ {
+ FNfloat xs = x;
+ FNfloat ys = y;
+ FNfloat zs = z;
+ TransformDomainWarpCoordinate(xs, ys, zs);
+
+ int seed = mSeed;
+ float amp = mDomainWarpAmp * mFractalBounding;
+ float freq = mFrequency;
+
+ for (int i = 0; i < mOctaves; i++)
+ {
+ DoSingleDomainWarp(seed, amp, freq, xs, ys, zs, x, y, z);
+
+ seed++;
+ amp *= mGain;
+ freq *= mLacunarity;
+ }
+ }
+
+
+ // Domain Warp Basic Grid
+
+ template <typename FNfloat>
+ void SingleDomainWarpBasicGrid(int seed, float warpAmp, float frequency, FNfloat x, FNfloat y, FNfloat& xr, FNfloat& yr)
+ {
+ FNfloat xf = x * frequency;
+ FNfloat yf = y * frequency;
+
+ int x0 = FastFloor(xf);
+ int y0 = FastFloor(yf);
+
+ float xs = InterpHermite((float)(xf - x0));
+ float ys = InterpHermite((float)(yf - y0));
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+
+ int hash0 = Hash(seed, x0, y0) & (255 << 1);
+ int hash1 = Hash(seed, x1, y0) & (255 << 1);
+
+ float lx0x = Lerp(Lookup<float>::RandVecs2D[hash0], Lookup<float>::RandVecs2D[hash1], xs);
+ float ly0x = Lerp(Lookup<float>::RandVecs2D[hash0 | 1], Lookup<float>::RandVecs2D[hash1 | 1], xs);
+
+ hash0 = Hash(seed, x0, y1) & (255 << 1);
+ hash1 = Hash(seed, x1, y1) & (255 << 1);
+
+ float lx1x = Lerp(Lookup<float>::RandVecs2D[hash0], Lookup<float>::RandVecs2D[hash1], xs);
+ float ly1x = Lerp(Lookup<float>::RandVecs2D[hash0 | 1], Lookup<float>::RandVecs2D[hash1 | 1], xs);
+
+ xr += Lerp(lx0x, lx1x, ys) * warpAmp;
+ yr += Lerp(ly0x, ly1x, ys) * warpAmp;
+ }
+
+ template <typename FNfloat>
+ void SingleDomainWarpBasicGrid(int seed, float warpAmp, float frequency, FNfloat x, FNfloat y, FNfloat z, FNfloat& xr, FNfloat& yr, FNfloat& zr)
+ {
+ FNfloat xf = x * frequency;
+ FNfloat yf = y * frequency;
+ FNfloat zf = z * frequency;
+
+ int x0 = FastFloor(xf);
+ int y0 = FastFloor(yf);
+ int z0 = FastFloor(zf);
+
+ float xs = InterpHermite((float)(xf - x0));
+ float ys = InterpHermite((float)(yf - y0));
+ float zs = InterpHermite((float)(zf - z0));
+
+ x0 *= PrimeX;
+ y0 *= PrimeY;
+ z0 *= PrimeZ;
+ int x1 = x0 + PrimeX;
+ int y1 = y0 + PrimeY;
+ int z1 = z0 + PrimeZ;
+
+ int hash0 = Hash(seed, x0, y0, z0) & (255 << 2);
+ int hash1 = Hash(seed, x1, y0, z0) & (255 << 2);
+
+ float lx0x = Lerp(Lookup<float>::RandVecs3D[hash0], Lookup<float>::RandVecs3D[hash1], xs);
+ float ly0x = Lerp(Lookup<float>::RandVecs3D[hash0 | 1], Lookup<float>::RandVecs3D[hash1 | 1], xs);
+ float lz0x = Lerp(Lookup<float>::RandVecs3D[hash0 | 2], Lookup<float>::RandVecs3D[hash1 | 2], xs);
+
+ hash0 = Hash(seed, x0, y1, z0) & (255 << 2);
+ hash1 = Hash(seed, x1, y1, z0) & (255 << 2);
+
+ float lx1x = Lerp(Lookup<float>::RandVecs3D[hash0], Lookup<float>::RandVecs3D[hash1], xs);
+ float ly1x = Lerp(Lookup<float>::RandVecs3D[hash0 | 1], Lookup<float>::RandVecs3D[hash1 | 1], xs);
+ float lz1x = Lerp(Lookup<float>::RandVecs3D[hash0 | 2], Lookup<float>::RandVecs3D[hash1 | 2], xs);
+
+ float lx0y = Lerp(lx0x, lx1x, ys);
+ float ly0y = Lerp(ly0x, ly1x, ys);
+ float lz0y = Lerp(lz0x, lz1x, ys);
+
+ hash0 = Hash(seed, x0, y0, z1) & (255 << 2);
+ hash1 = Hash(seed, x1, y0, z1) & (255 << 2);
+
+ lx0x = Lerp(Lookup<float>::RandVecs3D[hash0], Lookup<float>::RandVecs3D[hash1], xs);
+ ly0x = Lerp(Lookup<float>::RandVecs3D[hash0 | 1], Lookup<float>::RandVecs3D[hash1 | 1], xs);
+ lz0x = Lerp(Lookup<float>::RandVecs3D[hash0 | 2], Lookup<float>::RandVecs3D[hash1 | 2], xs);
+
+ hash0 = Hash(seed, x0, y1, z1) & (255 << 2);
+ hash1 = Hash(seed, x1, y1, z1) & (255 << 2);
+
+ lx1x = Lerp(Lookup<float>::RandVecs3D[hash0], Lookup<float>::RandVecs3D[hash1], xs);
+ ly1x = Lerp(Lookup<float>::RandVecs3D[hash0 | 1], Lookup<float>::RandVecs3D[hash1 | 1], xs);
+ lz1x = Lerp(Lookup<float>::RandVecs3D[hash0 | 2], Lookup<float>::RandVecs3D[hash1 | 2], xs);
+
+ xr += Lerp(lx0y, Lerp(lx0x, lx1x, ys), zs) * warpAmp;
+ yr += Lerp(ly0y, Lerp(ly0x, ly1x, ys), zs) * warpAmp;
+ zr += Lerp(lz0y, Lerp(lz0x, lz1x, ys), zs) * warpAmp;
+ }
+
+
+ // Domain Warp Simplex/OpenSimplex2
+
+ template <typename FNfloat>
+ void SingleDomainWarpSimplexGradient(int seed, float warpAmp, float frequency, FNfloat x, FNfloat y, FNfloat& xr, FNfloat& yr, bool outGradOnly)
+ {
+ const float SQRT3 = 1.7320508075688772935274463415059f;
+ const float G2 = (3 - SQRT3) / 6;
+
+ x *= frequency;
+ y *= frequency;
+
+ /*
+ * --- Skew moved to TransformNoiseCoordinate method ---
+ * const FNfloat F2 = 0.5f * (SQRT3 - 1);
+ * FNfloat s = (x + y) * F2;
+ * x += s; y += s;
+ */
+
+ int i = FastFloor(x);
+ int j = FastFloor(y);
+ float xi = (float)(x - i);
+ float yi = (float)(y - j);
+
+ float t = (xi + yi) * G2;
+ float x0 = (float)(xi - t);
+ float y0 = (float)(yi - t);
+
+ i *= PrimeX;
+ j *= PrimeY;
+
+ float vx, vy;
+ vx = vy = 0;
+
+ float a = 0.5f - x0 * x0 - y0 * y0;
+ if (a > 0)
+ {
+ float aaaa = (a * a) * (a * a);
+ float xo, yo;
+ if (outGradOnly)
+ GradCoordOut(seed, i, j, xo, yo);
+ else
+ GradCoordDual(seed, i, j, x0, y0, xo, yo);
+ vx += aaaa * xo;
+ vy += aaaa * yo;
+ }
+
+ float c = (float)(2 * (1 - 2 * G2) * (1 / G2 - 2)) * t + ((float)(-2 * (1 - 2 * G2) * (1 - 2 * G2)) + a);
+ if (c > 0)
+ {
+ float x2 = x0 + (2 * (float)G2 - 1);
+ float y2 = y0 + (2 * (float)G2 - 1);
+ float cccc = (c * c) * (c * c);
+ float xo, yo;
+ if (outGradOnly)
+ GradCoordOut(seed, i + PrimeX, j + PrimeY, xo, yo);
+ else
+ GradCoordDual(seed, i + PrimeX, j + PrimeY, x2, y2, xo, yo);
+ vx += cccc * xo;
+ vy += cccc * yo;
+ }
+
+ if (y0 > x0)
+ {
+ float x1 = x0 + (float)G2;
+ float y1 = y0 + ((float)G2 - 1);
+ float b = 0.5f - x1 * x1 - y1 * y1;
+ if (b > 0)
+ {
+ float bbbb = (b * b) * (b * b);
+ float xo, yo;
+ if (outGradOnly)
+ GradCoordOut(seed, i, j + PrimeY, xo, yo);
+ else
+ GradCoordDual(seed, i, j + PrimeY, x1, y1, xo, yo);
+ vx += bbbb * xo;
+ vy += bbbb * yo;
+ }
+ }
+ else
+ {
+ float x1 = x0 + ((float)G2 - 1);
+ float y1 = y0 + (float)G2;
+ float b = 0.5f - x1 * x1 - y1 * y1;
+ if (b > 0)
+ {
+ float bbbb = (b * b) * (b * b);
+ float xo, yo;
+ if (outGradOnly)
+ GradCoordOut(seed, i + PrimeX, j, xo, yo);
+ else
+ GradCoordDual(seed, i + PrimeX, j, x1, y1, xo, yo);
+ vx += bbbb * xo;
+ vy += bbbb * yo;
+ }
+ }
+
+ xr += vx * warpAmp;
+ yr += vy * warpAmp;
+ }
+
+ template <typename FNfloat>
+ void SingleDomainWarpOpenSimplex2Gradient(int seed, float warpAmp, float frequency, FNfloat x, FNfloat y, FNfloat z, FNfloat& xr, FNfloat& yr, FNfloat& zr, bool outGradOnly)
+ {
+ x *= frequency;
+ y *= frequency;
+ z *= frequency;
+
+ /*
+ * --- Rotation moved to TransformDomainWarpCoordinate method ---
+ * const FNfloat R3 = (FNfloat)(2.0 / 3.0);
+ * FNfloat r = (x + y + z) * R3; // Rotation, not skew
+ * x = r - x; y = r - y; z = r - z;
+ */
+
+ int i = FastRound(x);
+ int j = FastRound(y);
+ int k = FastRound(z);
+ float x0 = (float)x - i;
+ float y0 = (float)y - j;
+ float z0 = (float)z - k;
+
+ int xNSign = (int)(-x0 - 1.0f) | 1;
+ int yNSign = (int)(-y0 - 1.0f) | 1;
+ int zNSign = (int)(-z0 - 1.0f) | 1;
+
+ float ax0 = xNSign * -x0;
+ float ay0 = yNSign * -y0;
+ float az0 = zNSign * -z0;
+
+ i *= PrimeX;
+ j *= PrimeY;
+ k *= PrimeZ;
+
+ float vx, vy, vz;
+ vx = vy = vz = 0;
+
+ float a = (0.6f - x0 * x0) - (y0 * y0 + z0 * z0);
+ for (int l = 0; l < 2; l++)
+ {
+ if (a > 0)
+ {
+ float aaaa = (a * a) * (a * a);
+ float xo, yo, zo;
+ if (outGradOnly)
+ GradCoordOut(seed, i, j, k, xo, yo, zo);
+ else
+ GradCoordDual(seed, i, j, k, x0, y0, z0, xo, yo, zo);
+ vx += aaaa * xo;
+ vy += aaaa * yo;
+ vz += aaaa * zo;
+ }
+
+ float b = a + 1;
+ int i1 = i;
+ int j1 = j;
+ int k1 = k;
+ float x1 = x0;
+ float y1 = y0;
+ float z1 = z0;
+
+ if (ax0 >= ay0 && ax0 >= az0)
+ {
+ x1 += xNSign;
+ b -= xNSign * 2 * x1;
+ i1 -= xNSign * PrimeX;
+ }
+ else if (ay0 > ax0 && ay0 >= az0)
+ {
+ y1 += yNSign;
+ b -= yNSign * 2 * y1;
+ j1 -= yNSign * PrimeY;
+ }
+ else
+ {
+ z1 += zNSign;
+ b -= zNSign * 2 * z1;
+ k1 -= zNSign * PrimeZ;
+ }
+
+ if (b > 0)
+ {
+ float bbbb = (b * b) * (b * b);
+ float xo, yo, zo;
+ if (outGradOnly)
+ GradCoordOut(seed, i1, j1, k1, xo, yo, zo);
+ else
+ GradCoordDual(seed, i1, j1, k1, x1, y1, z1, xo, yo, zo);
+ vx += bbbb * xo;
+ vy += bbbb * yo;
+ vz += bbbb * zo;
+ }
+
+ if (l == 1) break;
+
+ ax0 = 0.5f - ax0;
+ ay0 = 0.5f - ay0;
+ az0 = 0.5f - az0;
+
+ x0 = xNSign * ax0;
+ y0 = yNSign * ay0;
+ z0 = zNSign * az0;
+
+ a += (0.75f - ax0) - (ay0 + az0);
+
+ i += (xNSign >> 1) & PrimeX;
+ j += (yNSign >> 1) & PrimeY;
+ k += (zNSign >> 1) & PrimeZ;
+
+ xNSign = -xNSign;
+ yNSign = -yNSign;
+ zNSign = -zNSign;
+
+ seed += 1293373;
+ }
+
+ xr += vx * warpAmp;
+ yr += vy * warpAmp;
+ zr += vz * warpAmp;
+ }
+};
+
+template <>
+struct FastNoiseLite::Arguments_must_be_floating_point_values<float> {};
+template <>
+struct FastNoiseLite::Arguments_must_be_floating_point_values<double> {};
+template <>
+struct FastNoiseLite::Arguments_must_be_floating_point_values<long double> {};
+
+template <typename T>
+const T FastNoiseLite::Lookup<T>::Gradients2D[] =
+{
+ 0.130526192220052f, 0.99144486137381f, 0.38268343236509f, 0.923879532511287f, 0.608761429008721f, 0.793353340291235f, 0.793353340291235f, 0.608761429008721f,
+ 0.923879532511287f, 0.38268343236509f, 0.99144486137381f, 0.130526192220051f, 0.99144486137381f, -0.130526192220051f, 0.923879532511287f, -0.38268343236509f,
+ 0.793353340291235f, -0.60876142900872f, 0.608761429008721f, -0.793353340291235f, 0.38268343236509f, -0.923879532511287f, 0.130526192220052f, -0.99144486137381f,
+ -0.130526192220052f, -0.99144486137381f, -0.38268343236509f, -0.923879532511287f, -0.608761429008721f, -0.793353340291235f, -0.793353340291235f, -0.608761429008721f,
+ -0.923879532511287f, -0.38268343236509f, -0.99144486137381f, -0.130526192220052f, -0.99144486137381f, 0.130526192220051f, -0.923879532511287f, 0.38268343236509f,
+ -0.793353340291235f, 0.608761429008721f, -0.608761429008721f, 0.793353340291235f, -0.38268343236509f, 0.923879532511287f, -0.130526192220052f, 0.99144486137381f,
+ 0.130526192220052f, 0.99144486137381f, 0.38268343236509f, 0.923879532511287f, 0.608761429008721f, 0.793353340291235f, 0.793353340291235f, 0.608761429008721f,
+ 0.923879532511287f, 0.38268343236509f, 0.99144486137381f, 0.130526192220051f, 0.99144486137381f, -0.130526192220051f, 0.923879532511287f, -0.38268343236509f,
+ 0.793353340291235f, -0.60876142900872f, 0.608761429008721f, -0.793353340291235f, 0.38268343236509f, -0.923879532511287f, 0.130526192220052f, -0.99144486137381f,
+ -0.130526192220052f, -0.99144486137381f, -0.38268343236509f, -0.923879532511287f, -0.608761429008721f, -0.793353340291235f, -0.793353340291235f, -0.608761429008721f,
+ -0.923879532511287f, -0.38268343236509f, -0.99144486137381f, -0.130526192220052f, -0.99144486137381f, 0.130526192220051f, -0.923879532511287f, 0.38268343236509f,
+ -0.793353340291235f, 0.608761429008721f, -0.608761429008721f, 0.793353340291235f, -0.38268343236509f, 0.923879532511287f, -0.130526192220052f, 0.99144486137381f,
+ 0.130526192220052f, 0.99144486137381f, 0.38268343236509f, 0.923879532511287f, 0.608761429008721f, 0.793353340291235f, 0.793353340291235f, 0.608761429008721f,
+ 0.923879532511287f, 0.38268343236509f, 0.99144486137381f, 0.130526192220051f, 0.99144486137381f, -0.130526192220051f, 0.923879532511287f, -0.38268343236509f,
+ 0.793353340291235f, -0.60876142900872f, 0.608761429008721f, -0.793353340291235f, 0.38268343236509f, -0.923879532511287f, 0.130526192220052f, -0.99144486137381f,
+ -0.130526192220052f, -0.99144486137381f, -0.38268343236509f, -0.923879532511287f, -0.608761429008721f, -0.793353340291235f, -0.793353340291235f, -0.608761429008721f,
+ -0.923879532511287f, -0.38268343236509f, -0.99144486137381f, -0.130526192220052f, -0.99144486137381f, 0.130526192220051f, -0.923879532511287f, 0.38268343236509f,
+ -0.793353340291235f, 0.608761429008721f, -0.608761429008721f, 0.793353340291235f, -0.38268343236509f, 0.923879532511287f, -0.130526192220052f, 0.99144486137381f,
+ 0.130526192220052f, 0.99144486137381f, 0.38268343236509f, 0.923879532511287f, 0.608761429008721f, 0.793353340291235f, 0.793353340291235f, 0.608761429008721f,
+ 0.923879532511287f, 0.38268343236509f, 0.99144486137381f, 0.130526192220051f, 0.99144486137381f, -0.130526192220051f, 0.923879532511287f, -0.38268343236509f,
+ 0.793353340291235f, -0.60876142900872f, 0.608761429008721f, -0.793353340291235f, 0.38268343236509f, -0.923879532511287f, 0.130526192220052f, -0.99144486137381f,
+ -0.130526192220052f, -0.99144486137381f, -0.38268343236509f, -0.923879532511287f, -0.608761429008721f, -0.793353340291235f, -0.793353340291235f, -0.608761429008721f,
+ -0.923879532511287f, -0.38268343236509f, -0.99144486137381f, -0.130526192220052f, -0.99144486137381f, 0.130526192220051f, -0.923879532511287f, 0.38268343236509f,
+ -0.793353340291235f, 0.608761429008721f, -0.608761429008721f, 0.793353340291235f, -0.38268343236509f, 0.923879532511287f, -0.130526192220052f, 0.99144486137381f,
+ 0.130526192220052f, 0.99144486137381f, 0.38268343236509f, 0.923879532511287f, 0.608761429008721f, 0.793353340291235f, 0.793353340291235f, 0.608761429008721f,
+ 0.923879532511287f, 0.38268343236509f, 0.99144486137381f, 0.130526192220051f, 0.99144486137381f, -0.130526192220051f, 0.923879532511287f, -0.38268343236509f,
+ 0.793353340291235f, -0.60876142900872f, 0.608761429008721f, -0.793353340291235f, 0.38268343236509f, -0.923879532511287f, 0.130526192220052f, -0.99144486137381f,
+ -0.130526192220052f, -0.99144486137381f, -0.38268343236509f, -0.923879532511287f, -0.608761429008721f, -0.793353340291235f, -0.793353340291235f, -0.608761429008721f,
+ -0.923879532511287f, -0.38268343236509f, -0.99144486137381f, -0.130526192220052f, -0.99144486137381f, 0.130526192220051f, -0.923879532511287f, 0.38268343236509f,
+ -0.793353340291235f, 0.608761429008721f, -0.608761429008721f, 0.793353340291235f, -0.38268343236509f, 0.923879532511287f, -0.130526192220052f, 0.99144486137381f,
+ 0.38268343236509f, 0.923879532511287f, 0.923879532511287f, 0.38268343236509f, 0.923879532511287f, -0.38268343236509f, 0.38268343236509f, -0.923879532511287f,
+ -0.38268343236509f, -0.923879532511287f, -0.923879532511287f, -0.38268343236509f, -0.923879532511287f, 0.38268343236509f, -0.38268343236509f, 0.923879532511287f,
+};
+
+template <typename T>
+const T FastNoiseLite::Lookup<T>::RandVecs2D[] =
+{
+ -0.2700222198f, -0.9628540911f, 0.3863092627f, -0.9223693152f, 0.04444859006f, -0.999011673f, -0.5992523158f, -0.8005602176f, -0.7819280288f, 0.6233687174f, 0.9464672271f, 0.3227999196f, -0.6514146797f, -0.7587218957f, 0.9378472289f, 0.347048376f,
+ -0.8497875957f, -0.5271252623f, -0.879042592f, 0.4767432447f, -0.892300288f, -0.4514423508f, -0.379844434f, -0.9250503802f, -0.9951650832f, 0.0982163789f, 0.7724397808f, -0.6350880136f, 0.7573283322f, -0.6530343002f, -0.9928004525f, -0.119780055f,
+ -0.0532665713f, 0.9985803285f, 0.9754253726f, -0.2203300762f, -0.7665018163f, 0.6422421394f, 0.991636706f, 0.1290606184f, -0.994696838f, 0.1028503788f, -0.5379205513f, -0.84299554f, 0.5022815471f, -0.8647041387f, 0.4559821461f, -0.8899889226f,
+ -0.8659131224f, -0.5001944266f, 0.0879458407f, -0.9961252577f, -0.5051684983f, 0.8630207346f, 0.7753185226f, -0.6315704146f, -0.6921944612f, 0.7217110418f, -0.5191659449f, -0.8546734591f, 0.8978622882f, -0.4402764035f, -0.1706774107f, 0.9853269617f,
+ -0.9353430106f, -0.3537420705f, -0.9992404798f, 0.03896746794f, -0.2882064021f, -0.9575683108f, -0.9663811329f, 0.2571137995f, -0.8759714238f, -0.4823630009f, -0.8303123018f, -0.5572983775f, 0.05110133755f, -0.9986934731f, -0.8558373281f, -0.5172450752f,
+ 0.09887025282f, 0.9951003332f, 0.9189016087f, 0.3944867976f, -0.2439375892f, -0.9697909324f, -0.8121409387f, -0.5834613061f, -0.9910431363f, 0.1335421355f, 0.8492423985f, -0.5280031709f, -0.9717838994f, -0.2358729591f, 0.9949457207f, 0.1004142068f,
+ 0.6241065508f, -0.7813392434f, 0.662910307f, 0.7486988212f, -0.7197418176f, 0.6942418282f, -0.8143370775f, -0.5803922158f, 0.104521054f, -0.9945226741f, -0.1065926113f, -0.9943027784f, 0.445799684f, -0.8951327509f, 0.105547406f, 0.9944142724f,
+ -0.992790267f, 0.1198644477f, -0.8334366408f, 0.552615025f, 0.9115561563f, -0.4111755999f, 0.8285544909f, -0.5599084351f, 0.7217097654f, -0.6921957921f, 0.4940492677f, -0.8694339084f, -0.3652321272f, -0.9309164803f, -0.9696606758f, 0.2444548501f,
+ 0.08925509731f, -0.996008799f, 0.5354071276f, -0.8445941083f, -0.1053576186f, 0.9944343981f, -0.9890284586f, 0.1477251101f, 0.004856104961f, 0.9999882091f, 0.9885598478f, 0.1508291331f, 0.9286129562f, -0.3710498316f, -0.5832393863f, -0.8123003252f,
+ 0.3015207509f, 0.9534596146f, -0.9575110528f, 0.2883965738f, 0.9715802154f, -0.2367105511f, 0.229981792f, 0.9731949318f, 0.955763816f, -0.2941352207f, 0.740956116f, 0.6715534485f, -0.9971513787f, -0.07542630764f, 0.6905710663f, -0.7232645452f,
+ -0.290713703f, -0.9568100872f, 0.5912777791f, -0.8064679708f, -0.9454592212f, -0.325740481f, 0.6664455681f, 0.74555369f, 0.6236134912f, 0.7817328275f, 0.9126993851f, -0.4086316587f, -0.8191762011f, 0.5735419353f, -0.8812745759f, -0.4726046147f,
+ 0.9953313627f, 0.09651672651f, 0.9855650846f, -0.1692969699f, -0.8495980887f, 0.5274306472f, 0.6174853946f, -0.7865823463f, 0.8508156371f, 0.52546432f, 0.9985032451f, -0.05469249926f, 0.1971371563f, -0.9803759185f, 0.6607855748f, -0.7505747292f,
+ -0.03097494063f, 0.9995201614f, -0.6731660801f, 0.739491331f, -0.7195018362f, -0.6944905383f, 0.9727511689f, 0.2318515979f, 0.9997059088f, -0.0242506907f, 0.4421787429f, -0.8969269532f, 0.9981350961f, -0.061043673f, -0.9173660799f, -0.3980445648f,
+ -0.8150056635f, -0.5794529907f, -0.8789331304f, 0.4769450202f, 0.0158605829f, 0.999874213f, -0.8095464474f, 0.5870558317f, -0.9165898907f, -0.3998286786f, -0.8023542565f, 0.5968480938f, -0.5176737917f, 0.8555780767f, -0.8154407307f, -0.5788405779f,
+ 0.4022010347f, -0.9155513791f, -0.9052556868f, -0.4248672045f, 0.7317445619f, 0.6815789728f, -0.5647632201f, -0.8252529947f, -0.8403276335f, -0.5420788397f, -0.9314281527f, 0.363925262f, 0.5238198472f, 0.8518290719f, 0.7432803869f, -0.6689800195f,
+ -0.985371561f, -0.1704197369f, 0.4601468731f, 0.88784281f, 0.825855404f, 0.5638819483f, 0.6182366099f, 0.7859920446f, 0.8331502863f, -0.553046653f, 0.1500307506f, 0.9886813308f, -0.662330369f, -0.7492119075f, -0.668598664f, 0.743623444f,
+ 0.7025606278f, 0.7116238924f, -0.5419389763f, -0.8404178401f, -0.3388616456f, 0.9408362159f, 0.8331530315f, 0.5530425174f, -0.2989720662f, -0.9542618632f, 0.2638522993f, 0.9645630949f, 0.124108739f, -0.9922686234f, -0.7282649308f, -0.6852956957f,
+ 0.6962500149f, 0.7177993569f, -0.9183535368f, 0.3957610156f, -0.6326102274f, -0.7744703352f, -0.9331891859f, -0.359385508f, -0.1153779357f, -0.9933216659f, 0.9514974788f, -0.3076565421f, -0.08987977445f, -0.9959526224f, 0.6678496916f, 0.7442961705f,
+ 0.7952400393f, -0.6062947138f, -0.6462007402f, -0.7631674805f, -0.2733598753f, 0.9619118351f, 0.9669590226f, -0.254931851f, -0.9792894595f, 0.2024651934f, -0.5369502995f, -0.8436138784f, -0.270036471f, -0.9628500944f, -0.6400277131f, 0.7683518247f,
+ -0.7854537493f, -0.6189203566f, 0.06005905383f, -0.9981948257f, -0.02455770378f, 0.9996984141f, -0.65983623f, 0.751409442f, -0.6253894466f, -0.7803127835f, -0.6210408851f, -0.7837781695f, 0.8348888491f, 0.5504185768f, -0.1592275245f, 0.9872419133f,
+ 0.8367622488f, 0.5475663786f, -0.8675753916f, -0.4973056806f, -0.2022662628f, -0.9793305667f, 0.9399189937f, 0.3413975472f, 0.9877404807f, -0.1561049093f, -0.9034455656f, 0.4287028224f, 0.1269804218f, -0.9919052235f, -0.3819600854f, 0.924178821f,
+ 0.9754625894f, 0.2201652486f, -0.3204015856f, -0.9472818081f, -0.9874760884f, 0.1577687387f, 0.02535348474f, -0.9996785487f, 0.4835130794f, -0.8753371362f, -0.2850799925f, -0.9585037287f, -0.06805516006f, -0.99768156f, -0.7885244045f, -0.6150034663f,
+ 0.3185392127f, -0.9479096845f, 0.8880043089f, 0.4598351306f, 0.6476921488f, -0.7619021462f, 0.9820241299f, 0.1887554194f, 0.9357275128f, -0.3527237187f, -0.8894895414f, 0.4569555293f, 0.7922791302f, 0.6101588153f, 0.7483818261f, 0.6632681526f,
+ -0.7288929755f, -0.6846276581f, 0.8729032783f, -0.4878932944f, 0.8288345784f, 0.5594937369f, 0.08074567077f, 0.9967347374f, 0.9799148216f, -0.1994165048f, -0.580730673f, -0.8140957471f, -0.4700049791f, -0.8826637636f, 0.2409492979f, 0.9705377045f,
+ 0.9437816757f, -0.3305694308f, -0.8927998638f, -0.4504535528f, -0.8069622304f, 0.5906030467f, 0.06258973166f, 0.9980393407f, -0.9312597469f, 0.3643559849f, 0.5777449785f, 0.8162173362f, -0.3360095855f, -0.941858566f, 0.697932075f, -0.7161639607f,
+ -0.002008157227f, -0.9999979837f, -0.1827294312f, -0.9831632392f, -0.6523911722f, 0.7578824173f, -0.4302626911f, -0.9027037258f, -0.9985126289f, -0.05452091251f, -0.01028102172f, -0.9999471489f, -0.4946071129f, 0.8691166802f, -0.2999350194f, 0.9539596344f,
+ 0.8165471961f, 0.5772786819f, 0.2697460475f, 0.962931498f, -0.7306287391f, -0.6827749597f, -0.7590952064f, -0.6509796216f, -0.907053853f, 0.4210146171f, -0.5104861064f, -0.8598860013f, 0.8613350597f, 0.5080373165f, 0.5007881595f, -0.8655698812f,
+ -0.654158152f, 0.7563577938f, -0.8382755311f, -0.545246856f, 0.6940070834f, 0.7199681717f, 0.06950936031f, 0.9975812994f, 0.1702942185f, -0.9853932612f, 0.2695973274f, 0.9629731466f, 0.5519612192f, -0.8338697815f, 0.225657487f, -0.9742067022f,
+ 0.4215262855f, -0.9068161835f, 0.4881873305f, -0.8727388672f, -0.3683854996f, -0.9296731273f, -0.9825390578f, 0.1860564427f, 0.81256471f, 0.5828709909f, 0.3196460933f, -0.9475370046f, 0.9570913859f, 0.2897862643f, -0.6876655497f, -0.7260276109f,
+ -0.9988770922f, -0.047376731f, -0.1250179027f, 0.992154486f, -0.8280133617f, 0.560708367f, 0.9324863769f, -0.3612051451f, 0.6394653183f, 0.7688199442f, -0.01623847064f, -0.9998681473f, -0.9955014666f, -0.09474613458f, -0.81453315f, 0.580117012f,
+ 0.4037327978f, -0.9148769469f, 0.9944263371f, 0.1054336766f, -0.1624711654f, 0.9867132919f, -0.9949487814f, -0.100383875f, -0.6995302564f, 0.7146029809f, 0.5263414922f, -0.85027327f, -0.5395221479f, 0.841971408f, 0.6579370318f, 0.7530729462f,
+ 0.01426758847f, -0.9998982128f, -0.6734383991f, 0.7392433447f, 0.639412098f, -0.7688642071f, 0.9211571421f, 0.3891908523f, -0.146637214f, -0.9891903394f, -0.782318098f, 0.6228791163f, -0.5039610839f, -0.8637263605f, -0.7743120191f, -0.6328039957f,
+};
+
+template <typename T>
+const T FastNoiseLite::Lookup<T>::Gradients3D[] =
+{
+ 0, 1, 1, 0, 0,-1, 1, 0, 0, 1,-1, 0, 0,-1,-1, 0,
+ 1, 0, 1, 0, -1, 0, 1, 0, 1, 0,-1, 0, -1, 0,-1, 0,
+ 1, 1, 0, 0, -1, 1, 0, 0, 1,-1, 0, 0, -1,-1, 0, 0,
+ 0, 1, 1, 0, 0,-1, 1, 0, 0, 1,-1, 0, 0,-1,-1, 0,
+ 1, 0, 1, 0, -1, 0, 1, 0, 1, 0,-1, 0, -1, 0,-1, 0,
+ 1, 1, 0, 0, -1, 1, 0, 0, 1,-1, 0, 0, -1,-1, 0, 0,
+ 0, 1, 1, 0, 0,-1, 1, 0, 0, 1,-1, 0, 0,-1,-1, 0,
+ 1, 0, 1, 0, -1, 0, 1, 0, 1, 0,-1, 0, -1, 0,-1, 0,
+ 1, 1, 0, 0, -1, 1, 0, 0, 1,-1, 0, 0, -1,-1, 0, 0,
+ 0, 1, 1, 0, 0,-1, 1, 0, 0, 1,-1, 0, 0,-1,-1, 0,
+ 1, 0, 1, 0, -1, 0, 1, 0, 1, 0,-1, 0, -1, 0,-1, 0,
+ 1, 1, 0, 0, -1, 1, 0, 0, 1,-1, 0, 0, -1,-1, 0, 0,
+ 0, 1, 1, 0, 0,-1, 1, 0, 0, 1,-1, 0, 0,-1,-1, 0,
+ 1, 0, 1, 0, -1, 0, 1, 0, 1, 0,-1, 0, -1, 0,-1, 0,
+ 1, 1, 0, 0, -1, 1, 0, 0, 1,-1, 0, 0, -1,-1, 0, 0,
+ 1, 1, 0, 0, 0,-1, 1, 0, -1, 1, 0, 0, 0,-1,-1, 0
+};
+
+template <typename T>
+const T FastNoiseLite::Lookup<T>::RandVecs3D[] =
+{
+ -0.7292736885f, -0.6618439697f, 0.1735581948f, 0, 0.790292081f, -0.5480887466f, -0.2739291014f, 0, 0.7217578935f, 0.6226212466f, -0.3023380997f, 0, 0.565683137f, -0.8208298145f, -0.0790000257f, 0, 0.760049034f, -0.5555979497f, -0.3370999617f, 0, 0.3713945616f, 0.5011264475f, 0.7816254623f, 0, -0.1277062463f, -0.4254438999f, -0.8959289049f, 0, -0.2881560924f, -0.5815838982f, 0.7607405838f, 0,
+ 0.5849561111f, -0.662820239f, -0.4674352136f, 0, 0.3307171178f, 0.0391653737f, 0.94291689f, 0, 0.8712121778f, -0.4113374369f, -0.2679381538f, 0, 0.580981015f, 0.7021915846f, 0.4115677815f, 0, 0.503756873f, 0.6330056931f, -0.5878203852f, 0, 0.4493712205f, 0.601390195f, 0.6606022552f, 0, -0.6878403724f, 0.09018890807f, -0.7202371714f, 0, -0.5958956522f, -0.6469350577f, 0.475797649f, 0,
+ -0.5127052122f, 0.1946921978f, -0.8361987284f, 0, -0.9911507142f, -0.05410276466f, -0.1212153153f, 0, -0.2149721042f, 0.9720882117f, -0.09397607749f, 0, -0.7518650936f, -0.5428057603f, 0.3742469607f, 0, 0.5237068895f, 0.8516377189f, -0.02107817834f, 0, 0.6333504779f, 0.1926167129f, -0.7495104896f, 0, -0.06788241606f, 0.3998305789f, 0.9140719259f, 0, -0.5538628599f, -0.4729896695f, -0.6852128902f, 0,
+ -0.7261455366f, -0.5911990757f, 0.3509933228f, 0, -0.9229274737f, -0.1782808786f, 0.3412049336f, 0, -0.6968815002f, 0.6511274338f, 0.3006480328f, 0, 0.9608044783f, -0.2098363234f, -0.1811724921f, 0, 0.06817146062f, -0.9743405129f, 0.2145069156f, 0, -0.3577285196f, -0.6697087264f, -0.6507845481f, 0, -0.1868621131f, 0.7648617052f, -0.6164974636f, 0, -0.6541697588f, 0.3967914832f, 0.6439087246f, 0,
+ 0.6993340405f, -0.6164538506f, 0.3618239211f, 0, -0.1546665739f, 0.6291283928f, 0.7617583057f, 0, -0.6841612949f, -0.2580482182f, -0.6821542638f, 0, 0.5383980957f, 0.4258654885f, 0.7271630328f, 0, -0.5026987823f, -0.7939832935f, -0.3418836993f, 0, 0.3202971715f, 0.2834415347f, 0.9039195862f, 0, 0.8683227101f, -0.0003762656404f, -0.4959995258f, 0, 0.791120031f, -0.08511045745f, 0.6057105799f, 0,
+ -0.04011016052f, -0.4397248749f, 0.8972364289f, 0, 0.9145119872f, 0.3579346169f, -0.1885487608f, 0, -0.9612039066f, -0.2756484276f, 0.01024666929f, 0, 0.6510361721f, -0.2877799159f, -0.7023778346f, 0, -0.2041786351f, 0.7365237271f, 0.644859585f, 0, -0.7718263711f, 0.3790626912f, 0.5104855816f, 0, -0.3060082741f, -0.7692987727f, 0.5608371729f, 0, 0.454007341f, -0.5024843065f, 0.7357899537f, 0,
+ 0.4816795475f, 0.6021208291f, -0.6367380315f, 0, 0.6961980369f, -0.3222197429f, 0.641469197f, 0, -0.6532160499f, -0.6781148932f, 0.3368515753f, 0, 0.5089301236f, -0.6154662304f, -0.6018234363f, 0, -0.1635919754f, -0.9133604627f, -0.372840892f, 0, 0.52408019f, -0.8437664109f, 0.1157505864f, 0, 0.5902587356f, 0.4983817807f, -0.6349883666f, 0, 0.5863227872f, 0.494764745f, 0.6414307729f, 0,
+ 0.6779335087f, 0.2341345225f, 0.6968408593f, 0, 0.7177054546f, -0.6858979348f, 0.120178631f, 0, -0.5328819713f, -0.5205125012f, 0.6671608058f, 0, -0.8654874251f, -0.0700727088f, -0.4960053754f, 0, -0.2861810166f, 0.7952089234f, 0.5345495242f, 0, -0.04849529634f, 0.9810836427f, -0.1874115585f, 0, -0.6358521667f, 0.6058348682f, 0.4781800233f, 0, 0.6254794696f, -0.2861619734f, 0.7258696564f, 0,
+ -0.2585259868f, 0.5061949264f, -0.8227581726f, 0, 0.02136306781f, 0.5064016808f, -0.8620330371f, 0, 0.200111773f, 0.8599263484f, 0.4695550591f, 0, 0.4743561372f, 0.6014985084f, -0.6427953014f, 0, 0.6622993731f, -0.5202474575f, -0.5391679918f, 0, 0.08084972818f, -0.6532720452f, 0.7527940996f, 0, -0.6893687501f, 0.0592860349f, 0.7219805347f, 0, -0.1121887082f, -0.9673185067f, 0.2273952515f, 0,
+ 0.7344116094f, 0.5979668656f, -0.3210532909f, 0, 0.5789393465f, -0.2488849713f, 0.7764570201f, 0, 0.6988182827f, 0.3557169806f, -0.6205791146f, 0, -0.8636845529f, -0.2748771249f, -0.4224826141f, 0, -0.4247027957f, -0.4640880967f, 0.777335046f, 0, 0.5257722489f, -0.8427017621f, 0.1158329937f, 0, 0.9343830603f, 0.316302472f, -0.1639543925f, 0, -0.1016836419f, -0.8057303073f, -0.5834887393f, 0,
+ -0.6529238969f, 0.50602126f, -0.5635892736f, 0, -0.2465286165f, -0.9668205684f, -0.06694497494f, 0, -0.9776897119f, -0.2099250524f, -0.007368825344f, 0, 0.7736893337f, 0.5734244712f, 0.2694238123f, 0, -0.6095087895f, 0.4995678998f, 0.6155736747f, 0, 0.5794535482f, 0.7434546771f, 0.3339292269f, 0, -0.8226211154f, 0.08142581855f, 0.5627293636f, 0, -0.510385483f, 0.4703667658f, 0.7199039967f, 0,
+ -0.5764971849f, -0.07231656274f, -0.8138926898f, 0, 0.7250628871f, 0.3949971505f, -0.5641463116f, 0, -0.1525424005f, 0.4860840828f, -0.8604958341f, 0, -0.5550976208f, -0.4957820792f, 0.667882296f, 0, -0.1883614327f, 0.9145869398f, 0.357841725f, 0, 0.7625556724f, -0.5414408243f, -0.3540489801f, 0, -0.5870231946f, -0.3226498013f, -0.7424963803f, 0, 0.3051124198f, 0.2262544068f, -0.9250488391f, 0,
+ 0.6379576059f, 0.577242424f, -0.5097070502f, 0, -0.5966775796f, 0.1454852398f, -0.7891830656f, 0, -0.658330573f, 0.6555487542f, -0.3699414651f, 0, 0.7434892426f, 0.2351084581f, 0.6260573129f, 0, 0.5562114096f, 0.8264360377f, -0.0873632843f, 0, -0.3028940016f, -0.8251527185f, 0.4768419182f, 0, 0.1129343818f, -0.985888439f, -0.1235710781f, 0, 0.5937652891f, -0.5896813806f, 0.5474656618f, 0,
+ 0.6757964092f, -0.5835758614f, -0.4502648413f, 0, 0.7242302609f, -0.1152719764f, 0.6798550586f, 0, -0.9511914166f, 0.0753623979f, -0.2992580792f, 0, 0.2539470961f, -0.1886339355f, 0.9486454084f, 0, 0.571433621f, -0.1679450851f, -0.8032795685f, 0, -0.06778234979f, 0.3978269256f, 0.9149531629f, 0, 0.6074972649f, 0.733060024f, -0.3058922593f, 0, -0.5435478392f, 0.1675822484f, 0.8224791405f, 0,
+ -0.5876678086f, -0.3380045064f, -0.7351186982f, 0, -0.7967562402f, 0.04097822706f, -0.6029098428f, 0, -0.1996350917f, 0.8706294745f, 0.4496111079f, 0, -0.02787660336f, -0.9106232682f, -0.4122962022f, 0, -0.7797625996f, -0.6257634692f, 0.01975775581f, 0, -0.5211232846f, 0.7401644346f, -0.4249554471f, 0, 0.8575424857f, 0.4053272873f, -0.3167501783f, 0, 0.1045223322f, 0.8390195772f, -0.5339674439f, 0,
+ 0.3501822831f, 0.9242524096f, -0.1520850155f, 0, 0.1987849858f, 0.07647613266f, 0.9770547224f, 0, 0.7845996363f, 0.6066256811f, -0.1280964233f, 0, 0.09006737436f, -0.9750989929f, -0.2026569073f, 0, -0.8274343547f, -0.542299559f, 0.1458203587f, 0, -0.3485797732f, -0.415802277f, 0.840000362f, 0, -0.2471778936f, -0.7304819962f, -0.6366310879f, 0, -0.3700154943f, 0.8577948156f, 0.3567584454f, 0,
+ 0.5913394901f, -0.548311967f, -0.5913303597f, 0, 0.1204873514f, -0.7626472379f, -0.6354935001f, 0, 0.616959265f, 0.03079647928f, 0.7863922953f, 0, 0.1258156836f, -0.6640829889f, -0.7369967419f, 0, -0.6477565124f, -0.1740147258f, -0.7417077429f, 0, 0.6217889313f, -0.7804430448f, -0.06547655076f, 0, 0.6589943422f, -0.6096987708f, 0.4404473475f, 0, -0.2689837504f, -0.6732403169f, -0.6887635427f, 0,
+ -0.3849775103f, 0.5676542638f, 0.7277093879f, 0, 0.5754444408f, 0.8110471154f, -0.1051963504f, 0, 0.9141593684f, 0.3832947817f, 0.131900567f, 0, -0.107925319f, 0.9245493968f, 0.3654593525f, 0, 0.377977089f, 0.3043148782f, 0.8743716458f, 0, -0.2142885215f, -0.8259286236f, 0.5214617324f, 0, 0.5802544474f, 0.4148098596f, -0.7008834116f, 0, -0.1982660881f, 0.8567161266f, -0.4761596756f, 0,
+ -0.03381553704f, 0.3773180787f, -0.9254661404f, 0, -0.6867922841f, -0.6656597827f, 0.2919133642f, 0, 0.7731742607f, -0.2875793547f, -0.5652430251f, 0, -0.09655941928f, 0.9193708367f, -0.3813575004f, 0, 0.2715702457f, -0.9577909544f, -0.09426605581f, 0, 0.2451015704f, -0.6917998565f, -0.6792188003f, 0, 0.977700782f, -0.1753855374f, 0.1155036542f, 0, -0.5224739938f, 0.8521606816f, 0.02903615945f, 0,
+ -0.7734880599f, -0.5261292347f, 0.3534179531f, 0, -0.7134492443f, -0.269547243f, 0.6467878011f, 0, 0.1644037271f, 0.5105846203f, -0.8439637196f, 0, 0.6494635788f, 0.05585611296f, 0.7583384168f, 0, -0.4711970882f, 0.5017280509f, -0.7254255765f, 0, -0.6335764307f, -0.2381686273f, -0.7361091029f, 0, -0.9021533097f, -0.270947803f, -0.3357181763f, 0, -0.3793711033f, 0.872258117f, 0.3086152025f, 0,
+ -0.6855598966f, -0.3250143309f, 0.6514394162f, 0, 0.2900942212f, -0.7799057743f, -0.5546100667f, 0, -0.2098319339f, 0.85037073f, 0.4825351604f, 0, -0.4592603758f, 0.6598504336f, -0.5947077538f, 0, 0.8715945488f, 0.09616365406f, -0.4807031248f, 0, -0.6776666319f, 0.7118504878f, -0.1844907016f, 0, 0.7044377633f, 0.312427597f, 0.637304036f, 0, -0.7052318886f, -0.2401093292f, -0.6670798253f, 0,
+ 0.081921007f, -0.7207336136f, -0.6883545647f, 0, -0.6993680906f, -0.5875763221f, -0.4069869034f, 0, -0.1281454481f, 0.6419895885f, 0.7559286424f, 0, -0.6337388239f, -0.6785471501f, -0.3714146849f, 0, 0.5565051903f, -0.2168887573f, -0.8020356851f, 0, -0.5791554484f, 0.7244372011f, -0.3738578718f, 0, 0.1175779076f, -0.7096451073f, 0.6946792478f, 0, -0.6134619607f, 0.1323631078f, 0.7785527795f, 0,
+ 0.6984635305f, -0.02980516237f, -0.715024719f, 0, 0.8318082963f, -0.3930171956f, 0.3919597455f, 0, 0.1469576422f, 0.05541651717f, -0.9875892167f, 0, 0.708868575f, -0.2690503865f, 0.6520101478f, 0, 0.2726053183f, 0.67369766f, -0.68688995f, 0, -0.6591295371f, 0.3035458599f, -0.6880466294f, 0, 0.4815131379f, -0.7528270071f, 0.4487723203f, 0, 0.9430009463f, 0.1675647412f, -0.2875261255f, 0,
+ 0.434802957f, 0.7695304522f, -0.4677277752f, 0, 0.3931996188f, 0.594473625f, 0.7014236729f, 0, 0.7254336655f, -0.603925654f, 0.3301814672f, 0, 0.7590235227f, -0.6506083235f, 0.02433313207f, 0, -0.8552768592f, -0.3430042733f, 0.3883935666f, 0, -0.6139746835f, 0.6981725247f, 0.3682257648f, 0, -0.7465905486f, -0.5752009504f, 0.3342849376f, 0, 0.5730065677f, 0.810555537f, -0.1210916791f, 0,
+ -0.9225877367f, -0.3475211012f, -0.167514036f, 0, -0.7105816789f, -0.4719692027f, -0.5218416899f, 0, -0.08564609717f, 0.3583001386f, 0.929669703f, 0, -0.8279697606f, -0.2043157126f, 0.5222271202f, 0, 0.427944023f, 0.278165994f, 0.8599346446f, 0, 0.5399079671f, -0.7857120652f, -0.3019204161f, 0, 0.5678404253f, -0.5495413974f, -0.6128307303f, 0, -0.9896071041f, 0.1365639107f, -0.04503418428f, 0,
+ -0.6154342638f, -0.6440875597f, 0.4543037336f, 0, 0.1074204368f, -0.7946340692f, 0.5975094525f, 0, -0.3595449969f, -0.8885529948f, 0.28495784f, 0, -0.2180405296f, 0.1529888965f, 0.9638738118f, 0, -0.7277432317f, -0.6164050508f, -0.3007234646f, 0, 0.7249729114f, -0.00669719484f, 0.6887448187f, 0, -0.5553659455f, -0.5336586252f, 0.6377908264f, 0, 0.5137558015f, 0.7976208196f, -0.3160000073f, 0,
+ -0.3794024848f, 0.9245608561f, -0.03522751494f, 0, 0.8229248658f, 0.2745365933f, -0.4974176556f, 0, -0.5404114394f, 0.6091141441f, 0.5804613989f, 0, 0.8036581901f, -0.2703029469f, 0.5301601931f, 0, 0.6044318879f, 0.6832968393f, 0.4095943388f, 0, 0.06389988817f, 0.9658208605f, -0.2512108074f, 0, 0.1087113286f, 0.7402471173f, -0.6634877936f, 0, -0.713427712f, -0.6926784018f, 0.1059128479f, 0,
+ 0.6458897819f, -0.5724548511f, -0.5050958653f, 0, -0.6553931414f, 0.7381471625f, 0.159995615f, 0, 0.3910961323f, 0.9188871375f, -0.05186755998f, 0, -0.4879022471f, -0.5904376907f, 0.6429111375f, 0, 0.6014790094f, 0.7707441366f, -0.2101820095f, 0, -0.5677173047f, 0.7511360995f, 0.3368851762f, 0, 0.7858573506f, 0.226674665f, 0.5753666838f, 0, -0.4520345543f, -0.604222686f, -0.6561857263f, 0,
+ 0.002272116345f, 0.4132844051f, -0.9105991643f, 0, -0.5815751419f, -0.5162925989f, 0.6286591339f, 0, -0.03703704785f, 0.8273785755f, 0.5604221175f, 0, -0.5119692504f, 0.7953543429f, -0.3244980058f, 0, -0.2682417366f, -0.9572290247f, -0.1084387619f, 0, -0.2322482736f, -0.9679131102f, -0.09594243324f, 0, 0.3554328906f, -0.8881505545f, 0.2913006227f, 0, 0.7346520519f, -0.4371373164f, 0.5188422971f, 0,
+ 0.9985120116f, 0.04659011161f, -0.02833944577f, 0, -0.3727687496f, -0.9082481361f, 0.1900757285f, 0, 0.91737377f, -0.3483642108f, 0.1925298489f, 0, 0.2714911074f, 0.4147529736f, -0.8684886582f, 0, 0.5131763485f, -0.7116334161f, 0.4798207128f, 0, -0.8737353606f, 0.18886992f, -0.4482350644f, 0, 0.8460043821f, -0.3725217914f, 0.3814499973f, 0, 0.8978727456f, -0.1780209141f, -0.4026575304f, 0,
+ 0.2178065647f, -0.9698322841f, -0.1094789531f, 0, -0.1518031304f, -0.7788918132f, -0.6085091231f, 0, -0.2600384876f, -0.4755398075f, -0.8403819825f, 0, 0.572313509f, -0.7474340931f, -0.3373418503f, 0, -0.7174141009f, 0.1699017182f, -0.6756111411f, 0, -0.684180784f, 0.02145707593f, -0.7289967412f, 0, -0.2007447902f, 0.06555605789f, -0.9774476623f, 0, -0.1148803697f, -0.8044887315f, 0.5827524187f, 0,
+ -0.7870349638f, 0.03447489231f, 0.6159443543f, 0, -0.2015596421f, 0.6859872284f, 0.6991389226f, 0, -0.08581082512f, -0.10920836f, -0.9903080513f, 0, 0.5532693395f, 0.7325250401f, -0.396610771f, 0, -0.1842489331f, -0.9777375055f, -0.1004076743f, 0, 0.0775473789f, -0.9111505856f, 0.4047110257f, 0, 0.1399838409f, 0.7601631212f, -0.6344734459f, 0, 0.4484419361f, -0.845289248f, 0.2904925424f, 0
+};
+
+}
+#endif // namespace fastnoiselite
diff --git a/thirdparty/noise/patches/FastNoiseLite.patch b/thirdparty/noise/patches/FastNoiseLite.patch
new file mode 100644
index 0000000000..acb1edfd73
--- /dev/null
+++ b/thirdparty/noise/patches/FastNoiseLite.patch
@@ -0,0 +1,18 @@
+--- orig/FastNoiseLite.h 1900-01-00 00:00:00 +0000
++++ noise/FastNoiseLite.h 1900-01-00 00:00:00 +0000
+@@ -52,6 +52,8 @@
+
+ #include <cmath>
+
++namespace fastnoiselite{
++
+ class FastNoiseLite
+ {
+ public:
+@@ -2583,4 +2585,5 @@
+ -0.7870349638f, 0.03447489231f, 0.6159443543f, 0, -0.2015596421f, 0.6859872284f, 0.6991389226f, 0, -0.08581082512f, -0.10920836f, -0.9903080513f, 0, 0.5532693395f, 0.7325250401f, -0.396610771f, 0, -0.1842489331f, -0.9777375055f, -0.1004076743f, 0, 0.0775473789f, -0.9111505856f, 0.4047110257f, 0, 0.1399838409f, 0.7601631212f, -0.6344734459f, 0, 0.4484419361f, -0.845289248f, 0.2904925424f, 0
+ };
+
+-#endif
++}
++#endif // namespace fastnoiselite
diff --git a/thirdparty/vulkan/patches/VMA-use-volk.patch b/thirdparty/vulkan/patches/VMA-use-volk.patch
index 1b6e0f04b8..eebe0c1bc3 100644
--- a/thirdparty/vulkan/patches/VMA-use-volk.patch
+++ b/thirdparty/vulkan/patches/VMA-use-volk.patch
@@ -1,5 +1,5 @@
diff --git a/thirdparty/vulkan/vk_mem_alloc.h b/thirdparty/vulkan/vk_mem_alloc.h
-index 52b403bede..7c450be211 100644
+index 44affc5ca4..d96f2dacc0 100644
--- a/thirdparty/vulkan/vk_mem_alloc.h
+++ b/thirdparty/vulkan/vk_mem_alloc.h
@@ -127,7 +127,11 @@ extern "C" {
diff --git a/thirdparty/vulkan/vk_mem_alloc.h b/thirdparty/vulkan/vk_mem_alloc.h
index 6618f1d1f0..d96f2dacc0 100644
--- a/thirdparty/vulkan/vk_mem_alloc.h
+++ b/thirdparty/vulkan/vk_mem_alloc.h
@@ -25,7 +25,7 @@
/** \mainpage Vulkan Memory Allocator
-<b>Version 3.0.0-development</b>
+<b>Version 3.0.1-development (2022-03-28)</b>
Copyright (c) 2017-2022 Advanced Micro Devices, Inc. All rights reserved. \n
License: MIT
@@ -84,11 +84,14 @@ License: MIT
- [Custom host memory allocator](@ref custom_memory_allocator)
- [Device memory allocation callbacks](@ref allocation_callbacks)
- [Device heap memory limit](@ref heap_memory_limit)
-- \subpage vk_khr_dedicated_allocation
-- \subpage enabling_buffer_device_address
-- \subpage vk_amd_device_coherent_memory
+- <b>Extension support</b>
+ - \subpage vk_khr_dedicated_allocation
+ - \subpage enabling_buffer_device_address
+ - \subpage vk_ext_memory_priority
+ - \subpage vk_amd_device_coherent_memory
- \subpage general_considerations
- [Thread safety](@ref general_considerations_thread_safety)
+ - [Versioning and compatibility](@ref general_considerations_versioning_and_compatibility)
- [Validation layer warnings](@ref general_considerations_validation_layer_warnings)
- [Allocation algorithm](@ref general_considerations_allocation_algorithm)
- [Features not supported](@ref general_considerations_features_not_supported)
@@ -427,6 +430,7 @@ typedef enum VmaAllocatorCreateFlagBits
VMA_ALLOCATOR_CREATE_FLAG_BITS_MAX_ENUM = 0x7FFFFFFF
} VmaAllocatorCreateFlagBits;
+/// See #VmaAllocatorCreateFlagBits.
typedef VkFlags VmaAllocatorCreateFlags;
/** @} */
@@ -534,8 +538,7 @@ typedef enum VmaAllocationCreateFlagBits
You should not use #VMA_ALLOCATION_CREATE_DEDICATED_MEMORY_BIT and
#VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT at the same time. It makes no sense.
-
- If VmaAllocationCreateInfo::pool is not null, this flag is implied and ignored. */
+ */
VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT = 0x00000002,
/** \brief Set this flag to use a memory that will be persistently mapped and retrieve pointer to it.
@@ -548,13 +551,11 @@ typedef enum VmaAllocationCreateFlagBits
support it (e.g. Intel GPU).
*/
VMA_ALLOCATION_CREATE_MAPPED_BIT = 0x00000004,
- /// \deprecated Removed. Do not use.
- VMA_ALLOCATION_CREATE_RESERVED_1_BIT = 0x00000008,
- /// \deprecated Removed. Do not use.
- VMA_ALLOCATION_CREATE_RESERVED_2_BIT = 0x00000010,
- /** Set this flag to treat VmaAllocationCreateInfo::pUserData as pointer to a
+ /** \deprecated Preserved for backward compatibility. Consider using vmaSetAllocationName() instead.
+
+ Set this flag to treat VmaAllocationCreateInfo::pUserData as pointer to a
null-terminated string. Instead of copying pointer value, a local copy of the
- string is made and stored in allocation's `pUserData`. The string is automatically
+ string is made and stored in allocation's `pName`. The string is automatically
freed together with the allocation. It is also used in vmaBuildStatsString().
*/
VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT = 0x00000020,
@@ -567,6 +568,10 @@ typedef enum VmaAllocationCreateFlagBits
It is useful when you want to bind yourself to do some more advanced binding, e.g. using some extensions.
The flag is meaningful only with functions that bind by default: vmaCreateBuffer(), vmaCreateImage().
Otherwise it is ignored.
+
+ If you want to make sure the new buffer/image is not tied to the new memory allocation
+ through `VkMemoryDedicatedAllocateInfoKHR` structure in case the allocation ends up in its own memory block,
+ use also flag #VMA_ALLOCATION_CREATE_CAN_ALIAS_BIT.
*/
VMA_ALLOCATION_CREATE_DONT_BIND_BIT = 0x00000080,
/** Create allocation only if additional device memory required for it, if any, won't exceed
@@ -591,7 +596,7 @@ typedef enum VmaAllocationCreateFlagBits
never read or accessed randomly, so a memory type can be selected that is uncached and write-combined.
\warning Violating this declaration may work correctly, but will likely be very slow.
- Watch out for implicit reads introduces by doing e.g. `pMappedData[i] += x;`
+ Watch out for implicit reads introduced by doing e.g. `pMappedData[i] += x;`
Better prepare your data in a local variable and `memcpy()` it to the mapped pointer all at once.
*/
VMA_ALLOCATION_CREATE_HOST_ACCESS_SEQUENTIAL_WRITE_BIT = 0x00000400,
@@ -604,7 +609,7 @@ typedef enum VmaAllocationCreateFlagBits
This includes allocations created in \ref custom_memory_pools.
Declares that mapped memory can be read, written, and accessed in random order,
- so a `HOST_CACHED` memory type is preferred.
+ so a `HOST_CACHED` memory type is required.
*/
VMA_ALLOCATION_CREATE_HOST_ACCESS_RANDOM_BIT = 0x00000800,
/**
@@ -648,6 +653,7 @@ typedef enum VmaAllocationCreateFlagBits
VMA_ALLOCATION_CREATE_FLAG_BITS_MAX_ENUM = 0x7FFFFFFF
} VmaAllocationCreateFlagBits;
+/// See #VmaAllocationCreateFlagBits.
typedef VkFlags VmaAllocationCreateFlags;
/// Flags to be passed as VmaPoolCreateInfo::flags.
@@ -726,16 +732,17 @@ typedef enum VmaDefragmentationFlagBits
VMA_DEFRAGMENTATION_FLAG_BITS_MAX_ENUM = 0x7FFFFFFF
} VmaDefragmentationFlagBits;
+/// See #VmaDefragmentationFlagBits.
typedef VkFlags VmaDefragmentationFlags;
-/// Operation performed on single defragmentation move.
+/// Operation performed on single defragmentation move. See structure #VmaDefragmentationMove.
typedef enum VmaDefragmentationMoveOperation
{
- /// Buffer/image has been recreated at `dstMemory` + `dstOffset`, data has been copied, old buffer/image has been destroyed. `srcAllocation` should be changed to point to the new place. This is the default value set by vmaBeginDefragmentationPass().
+ /// Buffer/image has been recreated at `dstTmpAllocation`, data has been copied, old buffer/image has been destroyed. `srcAllocation` should be changed to point to the new place. This is the default value set by vmaBeginDefragmentationPass().
VMA_DEFRAGMENTATION_MOVE_OPERATION_COPY = 0,
- /// Set this value if you cannot move the allocation. New place reserved `dstMemory` + `dstOffset` will be freed. `srcAllocation` will remain unchanged.
+ /// Set this value if you cannot move the allocation. New place reserved at `dstTmpAllocation` will be freed. `srcAllocation` will remain unchanged.
VMA_DEFRAGMENTATION_MOVE_OPERATION_IGNORE = 1,
- /// Set this value if you decide to abandon the allocation and you destroyed the buffer/image. New place reserved `dstMemory` + `dstOffset` will be freed, along with `srcAllocation`.
+ /// Set this value if you decide to abandon the allocation and you destroyed the buffer/image. New place reserved at `dstTmpAllocation` will be freed, along with `srcAllocation`, which will be destroyed.
VMA_DEFRAGMENTATION_MOVE_OPERATION_DESTROY = 2,
} VmaDefragmentationMoveOperation;
@@ -789,7 +796,7 @@ typedef enum VmaVirtualAllocationCreateFlagBits
/** Allocation strategy that chooses always the lowest offset in available space.
This is not the most efficient strategy but achieves highly packed data.
*/
- VMA_VIRTUAL_ALLOCATION_CREATE_STRATEGY_PACKED_BIT = VMA_ALLOCATION_CREATE_STRATEGY_MIN_OFFSET_BIT ,
+ VMA_VIRTUAL_ALLOCATION_CREATE_STRATEGY_MIN_OFFSET_BIT = VMA_ALLOCATION_CREATE_STRATEGY_MIN_OFFSET_BIT,
/** \brief A bit mask to extract only `STRATEGY` bits from entire set of flags.
These strategy flags are binary compatible with equivalent flags in #VmaAllocationCreateFlagBits.
@@ -1197,7 +1204,7 @@ typedef struct VmaBudget
Fetched from system using VK_EXT_memory_budget extension if enabled.
It might be different (most probably smaller) than `VkMemoryHeap::size[heapIndex]` due to factors
- external to the program, like other programs also consuming system resources.
+ external to the program, decided by the operating system.
Difference `budget - usage` is the amount of additional memory that can probably
be allocated without problems. Exceeding the budget may result in various problems.
*/
@@ -1211,6 +1218,10 @@ typedef struct VmaBudget
@{
*/
+/** \brief Parameters of new #VmaAllocation.
+
+To be used with functions like vmaCreateBuffer(), vmaCreateImage(), and many others.
+*/
typedef struct VmaAllocationCreateInfo
{
/// Use #VmaAllocationCreateFlagBits enum.
@@ -1337,7 +1348,7 @@ typedef struct VmaAllocationInfo
Same memory object can be shared by multiple allocations.
- It can change after call to vmaDefragment() if this allocation is passed to the function.
+ It can change after the allocation is moved during \ref defragmentation.
*/
VkDeviceMemory VMA_NULLABLE_NON_DISPATCHABLE deviceMemory;
/** \brief Offset in `VkDeviceMemory` object to the beginning of this allocation, in bytes. `(deviceMemory, offset)` pair is unique to this allocation.
@@ -1347,7 +1358,7 @@ typedef struct VmaAllocationInfo
not entire device memory block. Functions like vmaMapMemory(), vmaBindBufferMemory() also refer to the beginning of the allocation
and apply this offset automatically.
- It can change after call to vmaDefragment() if this allocation is passed to the function.
+ It can change after the allocation is moved during \ref defragmentation.
*/
VkDeviceSize offset;
/** \brief Size of this allocation, in bytes.
@@ -1367,7 +1378,7 @@ typedef struct VmaAllocationInfo
created with #VMA_ALLOCATION_CREATE_MAPPED_BIT flag, this value is null.
It can change after call to vmaMapMemory(), vmaUnmapMemory().
- It can also change after call to vmaDefragment() if this allocation is passed to the function.
+ It can also change after the allocation is moved during \ref defragmentation.
*/
void* VMA_NULLABLE pMappedData;
/** \brief Custom general-purpose pointer that was passed as VmaAllocationCreateInfo::pUserData or set using vmaSetAllocationUserData().
@@ -1375,6 +1386,14 @@ typedef struct VmaAllocationInfo
It can change after call to vmaSetAllocationUserData() for this allocation.
*/
void* VMA_NULLABLE pUserData;
+ /** \brief Custom allocation name that was set with vmaSetAllocationName().
+
+ It can change after call to vmaSetAllocationName() for this allocation.
+
+ Another way to set custom name is to pass it in VmaAllocationCreateInfo::pUserData with
+ additional flag #VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT set [DEPRECATED].
+ */
+ const char* VMA_NULLABLE pName;
} VmaAllocationInfo;
/** \brief Parameters for defragmentation.
@@ -1409,12 +1428,13 @@ typedef struct VmaDefragmentationMove
VmaDefragmentationMoveOperation operation;
/// Allocation that should be moved.
VmaAllocation VMA_NOT_NULL srcAllocation;
- /// Destination memory block where the allocation should be moved.
- VkDeviceMemory VMA_NOT_NULL_NON_DISPATCHABLE dstMemory;
- /// Destination offset where the allocation should be moved.
- VkDeviceSize dstOffset;
- /// Internal data used by VMA. Do not use or modify!
- void* VMA_NOT_NULL internalData;
+ /** \brief Temporary allocation pointing to destination memory that will replace `srcAllocation`.
+
+ \warning Do not store this allocation in your data structures! It exists only temporarily, for the duration of the defragmentation pass,
+ to be used for binding new buffer/image to the destination memory using e.g. vmaBindBufferMemory().
+ vmaEndDefragmentationPass() will destroy it and make `srcAllocation` point to this memory.
+ */
+ VmaAllocation VMA_NOT_NULL dstTmpAllocation;
} VmaDefragmentationMove;
/** \brief Parameters for incremental defragmentation steps.
@@ -1825,14 +1845,19 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaAllocateMemoryPages(
VmaAllocation VMA_NULLABLE* VMA_NOT_NULL VMA_LEN_IF_NOT_NULL(allocationCount) pAllocations,
VmaAllocationInfo* VMA_NULLABLE VMA_LEN_IF_NOT_NULL(allocationCount) pAllocationInfo);
-/**
+/** \brief Allocates memory suitable for given `VkBuffer`.
+
\param allocator
\param buffer
\param pCreateInfo
\param[out] pAllocation Handle to allocated memory.
\param[out] pAllocationInfo Optional. Information about allocated memory. It can be later fetched using function vmaGetAllocationInfo().
-You should free the memory using vmaFreeMemory().
+It only creates #VmaAllocation. To bind the memory to the buffer, use vmaBindBufferMemory().
+
+This is a special-purpose function. In most cases you should use vmaCreateBuffer().
+
+You must free the allocation using vmaFreeMemory() when no longer needed.
*/
VMA_CALL_PRE VkResult VMA_CALL_POST vmaAllocateMemoryForBuffer(
VmaAllocator VMA_NOT_NULL allocator,
@@ -1841,7 +1866,20 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaAllocateMemoryForBuffer(
VmaAllocation VMA_NULLABLE* VMA_NOT_NULL pAllocation,
VmaAllocationInfo* VMA_NULLABLE pAllocationInfo);
-/// Function similar to vmaAllocateMemoryForBuffer().
+/** \brief Allocates memory suitable for given `VkImage`.
+
+\param allocator
+\param image
+\param pCreateInfo
+\param[out] pAllocation Handle to allocated memory.
+\param[out] pAllocationInfo Optional. Information about allocated memory. It can be later fetched using function vmaGetAllocationInfo().
+
+It only creates #VmaAllocation. To bind the memory to the buffer, use vmaBindImageMemory().
+
+This is a special-purpose function. In most cases you should use vmaCreateImage().
+
+You must free the allocation using vmaFreeMemory() when no longer needed.
+*/
VMA_CALL_PRE VkResult VMA_CALL_POST vmaAllocateMemoryForImage(
VmaAllocator VMA_NOT_NULL allocator,
VkImage VMA_NOT_NULL_NON_DISPATCHABLE image,
@@ -1889,15 +1927,8 @@ VMA_CALL_PRE void VMA_CALL_POST vmaGetAllocationInfo(
/** \brief Sets pUserData in given allocation to new value.
-If the allocation was created with VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT,
-pUserData must be either null, or pointer to a null-terminated string. The function
-makes local copy of the string and sets it as allocation's `pUserData`. String
-passed as pUserData doesn't need to be valid for whole lifetime of the allocation -
-you can free it after this call. String previously pointed by allocation's
-pUserData is freed from memory.
-
-If the flag was not used, the value of pointer `pUserData` is just copied to
-allocation's `pUserData`. It is opaque, so you can use it however you want - e.g.
+The value of pointer `pUserData` is copied to allocation's `pUserData`.
+It is opaque, so you can use it however you want - e.g.
as a pointer, ordinal number or some handle to you own data.
*/
VMA_CALL_PRE void VMA_CALL_POST vmaSetAllocationUserData(
@@ -1905,6 +1936,19 @@ VMA_CALL_PRE void VMA_CALL_POST vmaSetAllocationUserData(
VmaAllocation VMA_NOT_NULL allocation,
void* VMA_NULLABLE pUserData);
+/** \brief Sets pName in given allocation to new value.
+
+`pName` must be either null, or pointer to a null-terminated string. The function
+makes local copy of the string and sets it as allocation's `pName`. String
+passed as pName doesn't need to be valid for whole lifetime of the allocation -
+you can free it after this call. String previously pointed by allocation's
+`pName` is freed from memory.
+*/
+VMA_CALL_PRE void VMA_CALL_POST vmaSetAllocationName(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const char* VMA_NULLABLE pName);
+
/**
\brief Given an allocation, returns Property Flags of its memory type.
@@ -2092,6 +2136,9 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaCheckCorruption(
\param allocator Allocator object.
\param pInfo Structure filled with parameters of defragmentation.
\param[out] pContext Context object that must be passed to vmaEndDefragmentation() to finish defragmentation.
+\returns
+- `VK_SUCCESS` if defragmentation can begin.
+- `VK_ERROR_FEATURE_NOT_PRESENT` if defragmentation is not supported.
For more information about defragmentation, see documentation chapter:
[Defragmentation](@ref defragmentation).
@@ -2109,7 +2156,7 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaBeginDefragmentation(
Use this function to finish defragmentation started by vmaBeginDefragmentation().
*/
-VMA_CALL_PRE VkResult VMA_CALL_POST vmaEndDefragmentation(
+VMA_CALL_PRE void VMA_CALL_POST vmaEndDefragmentation(
VmaAllocator VMA_NOT_NULL allocator,
VmaDefragmentationContext VMA_NOT_NULL context,
VmaDefragmentationStats* VMA_NULLABLE pStats);
@@ -2226,7 +2273,8 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaBindImageMemory2(
VkImage VMA_NOT_NULL_NON_DISPATCHABLE image,
const void* VMA_NULLABLE pNext);
-/**
+/** \brief Creates a new `VkBuffer`, allocates and binds memory for it.
+
\param allocator
\param pBufferCreateInfo
\param pAllocationCreateInfo
@@ -2241,7 +2289,7 @@ This function automatically:
-# Binds the buffer with the memory.
If any of these operations fail, buffer and allocation are not created,
-returned value is negative error code, *pBuffer and *pAllocation are null.
+returned value is negative error code, `*pBuffer` and `*pAllocation` are null.
If the function succeeded, you must destroy both buffer and allocation when you
no longer need them using either convenience function vmaDestroyBuffer() or
@@ -2282,6 +2330,31 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateBufferWithAlignment(
VmaAllocation VMA_NULLABLE* VMA_NOT_NULL pAllocation,
VmaAllocationInfo* VMA_NULLABLE pAllocationInfo);
+/** \brief Creates a new `VkBuffer`, binds already created memory for it.
+
+\param allocator
+\param allocation Allocation that provides memory to be used for binding new buffer to it.
+\param pBufferCreateInfo
+\param[out] pBuffer Buffer that was created.
+
+This function automatically:
+
+-# Creates buffer.
+-# Binds the buffer with the supplied memory.
+
+If any of these operations fail, buffer is not created,
+returned value is negative error code and `*pBuffer` is null.
+
+If the function succeeded, you must destroy the buffer when you
+no longer need it using `vkDestroyBuffer()`. If you want to also destroy the corresponding
+allocation you can use convenience function vmaDestroyBuffer().
+*/
+VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateAliasingBuffer(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const VkBufferCreateInfo* VMA_NOT_NULL pBufferCreateInfo,
+ VkBuffer VMA_NULLABLE_NON_DISPATCHABLE* VMA_NOT_NULL pBuffer);
+
/** \brief Destroys Vulkan buffer and frees allocated memory.
This is just a convenience function equivalent to:
@@ -2307,6 +2380,13 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateImage(
VmaAllocation VMA_NULLABLE* VMA_NOT_NULL pAllocation,
VmaAllocationInfo* VMA_NULLABLE pAllocationInfo);
+/// Function similar to vmaCreateAliasingBuffer().
+VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateAliasingImage(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const VkImageCreateInfo* VMA_NOT_NULL pImageCreateInfo,
+ VkImage VMA_NULLABLE_NON_DISPATCHABLE* VMA_NOT_NULL pImage);
+
/** \brief Destroys Vulkan image and frees allocated memory.
This is just a convenience function equivalent to:
@@ -2493,6 +2573,7 @@ VMA_CALL_PRE void VMA_CALL_POST vmaFreeStatsString(
#include <cstdlib>
#include <cstring>
#include <utility>
+#include <type_traits>
#ifdef _MSC_VER
#include <intrin.h> // For functions like __popcnt, _BitScanForward etc.
@@ -2689,6 +2770,11 @@ static void vma_aligned_free(void* VMA_NULLABLE ptr)
#endif
#endif
+#ifndef VMA_COUNT_BITS_SET
+ // Returns number of bits set to 1 in (v)
+ #define VMA_COUNT_BITS_SET(v) VmaCountBitsSet(v)
+#endif
+
#ifndef VMA_BITSCAN_LSB
// Scans integer for index of first nonzero value from the Least Significant Bit (LSB). If mask is 0 then returns UINT8_MAX
#define VMA_BITSCAN_LSB(mask) VmaBitScanLSB(mask)
@@ -3077,22 +3163,29 @@ class VmaAllocationObjectAllocator;
#endif // _VMA_FORWARD_DECLARATIONS
-#ifndef _VMA_FUNCTIONS
-// Returns number of bits set to 1 in (v).
-static inline uint32_t VmaCountBitsSet(uint32_t v)
-{
-#ifdef _MSC_VER
+#ifndef _VMA_FUNCTIONS
+
+/*
+Returns number of bits set to 1 in (v).
+
+On specific platforms and compilers you can use instrinsics like:
+
+Visual Studio:
return __popcnt(v);
-#elif defined __GNUC__ || defined __clang__
+GCC, Clang:
return static_cast<uint32_t>(__builtin_popcount(v));
-#else
+
+Define macro VMA_COUNT_BITS_SET to provide your optimized implementation.
+But you need to check in runtime whether user's CPU supports these, as some old processors don't.
+*/
+static inline uint32_t VmaCountBitsSet(uint32_t v)
+{
uint32_t c = v - ((v >> 1) & 0x55555555);
c = ((c >> 2) & 0x33333333) + (c & 0x33333333);
c = ((c >> 4) + c) & 0x0F0F0F0F;
c = ((c >> 8) + c) & 0x00FF00FF;
c = ((c >> 16) + c) & 0x0000FFFF;
return c;
-#endif
}
static inline uint8_t VmaBitScanLSB(uint64_t mask)
@@ -3631,7 +3724,11 @@ static bool FindMemoryPreferences(
// GPU access, no CPU access (e.g. a color attachment image) - prefer GPU memory
if(deviceAccess)
{
- outPreferredFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT;
+ // ...unless there is a clear preference from the user not to do so.
+ if(preferHost)
+ outNotPreferredFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT;
+ else
+ outPreferredFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT;
}
// No direct GPU access, no CPU access, just transfers.
// It may be staging copy intended for e.g. preserving image for next frame (then better GPU memory) or
@@ -5399,6 +5496,7 @@ public:
// Posts next part of an open string. The number is converted to decimal characters.
void ContinueString(uint32_t n);
void ContinueString(uint64_t n);
+ void ContinueString_Size(size_t n);
// Posts next part of an open string. Pointer value is converted to characters
// using "%p" formatting - shown as hexadecimal number, e.g.: 000000081276Ad00
void ContinueString_Pointer(const void* ptr);
@@ -5408,6 +5506,7 @@ public:
// Writes a number value.
void WriteNumber(uint32_t n);
void WriteNumber(uint64_t n);
+ void WriteSize(size_t n);
// Writes a boolean value - false or true.
void WriteBool(bool b);
// Writes a null value.
@@ -5432,6 +5531,11 @@ private:
VmaVector< StackItem, VmaStlAllocator<StackItem> > m_Stack;
bool m_InsideString;
+ // Write size_t for less than 64bits
+ void WriteSize(size_t n, std::integral_constant<bool, false>) { m_SB.AddNumber(static_cast<uint32_t>(n)); }
+ // Write size_t for 64bits
+ void WriteSize(size_t n, std::integral_constant<bool, true>) { m_SB.AddNumber(static_cast<uint64_t>(n)); }
+
void BeginValue(bool isString);
void WriteIndent(bool oneLess = false);
};
@@ -5574,6 +5678,14 @@ void VmaJsonWriter::ContinueString(uint64_t n)
m_SB.AddNumber(n);
}
+void VmaJsonWriter::ContinueString_Size(size_t n)
+{
+ VMA_ASSERT(m_InsideString);
+ // Fix for AppleClang incorrect type casting
+ // TODO: Change to if constexpr when C++17 used as minimal standard
+ WriteSize(n, std::is_same<size_t, uint64_t>{});
+}
+
void VmaJsonWriter::ContinueString_Pointer(const void* ptr)
{
VMA_ASSERT(m_InsideString);
@@ -5605,6 +5717,15 @@ void VmaJsonWriter::WriteNumber(uint64_t n)
m_SB.AddNumber(n);
}
+void VmaJsonWriter::WriteSize(size_t n)
+{
+ VMA_ASSERT(!m_InsideString);
+ BeginValue(false);
+ // Fix for AppleClang incorrect type casting
+ // TODO: Change to if constexpr when C++17 used as minimal standard
+ WriteSize(n, std::is_same<size_t, uint64_t>{});
+}
+
void VmaJsonWriter::WriteBool(bool b)
{
VMA_ASSERT(!m_InsideString);
@@ -5673,41 +5794,29 @@ static void VmaPrintDetailedStatistics(VmaJsonWriter& json, const VmaDetailedSta
json.WriteString("BlockCount");
json.WriteNumber(stat.statistics.blockCount);
-
- json.WriteString("AllocationCount");
- json.WriteNumber(stat.statistics.allocationCount);
-
- json.WriteString("UnusedRangeCount");
- json.WriteNumber(stat.unusedRangeCount);
-
json.WriteString("BlockBytes");
json.WriteNumber(stat.statistics.blockBytes);
-
+ json.WriteString("AllocationCount");
+ json.WriteNumber(stat.statistics.allocationCount);
json.WriteString("AllocationBytes");
json.WriteNumber(stat.statistics.allocationBytes);
+ json.WriteString("UnusedRangeCount");
+ json.WriteNumber(stat.unusedRangeCount);
if (stat.statistics.allocationCount > 1)
{
- json.WriteString("AllocationSize");
- json.BeginObject(true);
- json.WriteString("Min");
+ json.WriteString("AllocationSizeMin");
json.WriteNumber(stat.allocationSizeMin);
- json.WriteString("Max");
+ json.WriteString("AllocationSizeMax");
json.WriteNumber(stat.allocationSizeMax);
- json.EndObject();
}
-
if (stat.unusedRangeCount > 1)
{
- json.WriteString("UnusedRangeSize");
- json.BeginObject(true);
- json.WriteString("Min");
+ json.WriteString("UnusedRangeSizeMin");
json.WriteNumber(stat.unusedRangeSizeMin);
- json.WriteString("Max");
+ json.WriteString("UnusedRangeSizeMax");
json.WriteNumber(stat.unusedRangeSizeMax);
- json.EndObject();
}
-
json.EndObject();
}
#endif // _VMA_JSON_WRITER
@@ -5799,9 +5908,14 @@ private:
void PostMinorCounter()
{
if(m_MinorCounter < m_MajorCounter)
+ {
++m_MinorCounter;
+ }
else if(m_MajorCounter > 0)
- --m_MajorCounter, --m_MinorCounter;
+ {
+ --m_MajorCounter;
+ --m_MinorCounter;
+ }
}
};
@@ -5899,9 +6013,8 @@ struct VmaAllocation_T
enum FLAGS
{
- FLAG_USER_DATA_STRING = 0x01,
- FLAG_PERSISTENT_MAP = 0x02,
- FLAG_MAPPING_ALLOWED = 0x04,
+ FLAG_PERSISTENT_MAP = 0x01,
+ FLAG_MAPPING_ALLOWED = 0x02,
};
public:
@@ -5913,7 +6026,7 @@ public:
};
// This struct is allocated using VmaPoolAllocator.
- VmaAllocation_T(bool userDataString, bool mappingAllowed);
+ VmaAllocation_T(bool mappingAllowed);
~VmaAllocation_T();
void InitBlockAllocation(
@@ -5936,8 +6049,8 @@ public:
ALLOCATION_TYPE GetType() const { return (ALLOCATION_TYPE)m_Type; }
VkDeviceSize GetAlignment() const { return m_Alignment; }
VkDeviceSize GetSize() const { return m_Size; }
- bool IsUserDataString() const { return (m_Flags & FLAG_USER_DATA_STRING) != 0; }
void* GetUserData() const { return m_pUserData; }
+ const char* GetName() const { return m_pName; }
VmaSuballocationType GetSuballocationType() const { return (VmaSuballocationType)m_SuballocationType; }
VmaDeviceMemoryBlock* GetBlock() const { VMA_ASSERT(m_Type == ALLOCATION_TYPE_BLOCK); return m_BlockAllocation.m_Block; }
@@ -5945,8 +6058,10 @@ public:
bool IsPersistentMap() const { return (m_Flags & FLAG_PERSISTENT_MAP) != 0; }
bool IsMappingAllowed() const { return (m_Flags & FLAG_MAPPING_ALLOWED) != 0; }
- void SetUserData(VmaAllocator hAllocator, void* pUserData);
- void SwapBlockAllocation(VmaAllocation allocation);
+ void SetUserData(VmaAllocator hAllocator, void* pUserData) { m_pUserData = pUserData; }
+ void SetName(VmaAllocator hAllocator, const char* pName);
+ void FreeName(VmaAllocator hAllocator);
+ uint8_t SwapBlockAllocation(VmaAllocator hAllocator, VmaAllocation allocation);
VmaAllocHandle GetAllocHandle() const;
VkDeviceSize GetOffset() const;
VmaPool GetParentPool() const;
@@ -5992,6 +6107,7 @@ private:
VkDeviceSize m_Alignment;
VkDeviceSize m_Size;
void* m_pUserData;
+ char* m_pName;
uint32_t m_MemoryTypeIndex;
uint8_t m_Type; // ALLOCATION_TYPE
uint8_t m_SuballocationType; // VmaSuballocationType
@@ -6001,8 +6117,6 @@ private:
#if VMA_STATS_STRING_ENABLED
uint32_t m_BufferImageUsage; // 0 if unknown.
#endif
-
- void FreeUserDataString(VmaAllocator hAllocator);
};
#endif // _VMA_ALLOCATION_T
@@ -6239,6 +6353,7 @@ public:
// Validates all data structures inside this object. If not valid, returns false.
virtual bool Validate() const = 0;
virtual size_t GetAllocationCount() const = 0;
+ virtual size_t GetFreeRegionsCount() const = 0;
virtual VkDeviceSize GetSumFreeSize() const = 0;
// Returns true if this block is empty - contains only single free suballocation.
virtual bool IsEmpty() const = 0;
@@ -6248,14 +6363,14 @@ public:
virtual VmaAllocHandle GetAllocationListBegin() const = 0;
virtual VmaAllocHandle GetNextAllocation(VmaAllocHandle prevAlloc) const = 0;
+ virtual VkDeviceSize GetNextFreeRegionSize(VmaAllocHandle alloc) const = 0;
// Shouldn't modify blockCount.
virtual void AddDetailedStatistics(VmaDetailedStatistics& inoutStats) const = 0;
virtual void AddStatistics(VmaStatistics& inoutStats) const = 0;
#if VMA_STATS_STRING_ENABLED
- // mapRefCount == UINT32_MAX means unspecified.
- virtual void PrintDetailedMap(class VmaJsonWriter& json, uint32_t mapRefCount) const = 0;
+ virtual void PrintDetailedMap(class VmaJsonWriter& json) const = 0;
#endif
// Tries to find a place for suballocation with given parameters inside this block.
@@ -6299,8 +6414,7 @@ protected:
void PrintDetailedMap_Begin(class VmaJsonWriter& json,
VkDeviceSize unusedBytes,
size_t allocationCount,
- size_t unusedRangeCount,
- uint32_t mapRefCount) const;
+ size_t unusedRangeCount) const;
void PrintDetailedMap_Allocation(class VmaJsonWriter& json,
VkDeviceSize offset, VkDeviceSize size, void* userData) const;
void PrintDetailedMap_UnusedRange(class VmaJsonWriter& json,
@@ -6336,35 +6450,17 @@ void VmaBlockMetadata::DebugLogAllocation(VkDeviceSize offset, VkDeviceSize size
VmaAllocation allocation = reinterpret_cast<VmaAllocation>(userData);
userData = allocation->GetUserData();
+ const char* name = allocation->GetName();
#if VMA_STATS_STRING_ENABLED
- if (userData != VMA_NULL && allocation->IsUserDataString())
- {
- VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %s; Type: %s; Usage: %u",
- offset, size, reinterpret_cast<const char*>(userData),
- VMA_SUBALLOCATION_TYPE_NAMES[allocation->GetSuballocationType()],
- allocation->GetBufferImageUsage());
- }
- else
- {
- VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %p; Type: %s; Usage: %u",
- offset, size, userData,
- VMA_SUBALLOCATION_TYPE_NAMES[allocation->GetSuballocationType()],
- allocation->GetBufferImageUsage());
- }
+ VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %p; Name: %s; Type: %s; Usage: %u",
+ offset, size, userData, name ? name : "vma_empty",
+ VMA_SUBALLOCATION_TYPE_NAMES[allocation->GetSuballocationType()],
+ allocation->GetBufferImageUsage());
#else
- if (userData != VMA_NULL && allocation->IsUserDataString())
- {
- VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %s; Type: %u",
- offset, size, reinterpret_cast<const char*>(userData),
- (uint32_t)allocation->GetSuballocationType());
- }
- else
- {
- VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %p; Type: %u",
- offset, size, userData,
- (uint32_t)allocation->GetSuballocationType());
- }
+ VMA_DEBUG_LOG("UNFREED ALLOCATION; Offset: %llu; Size: %llu; UserData: %p; Name: %s; Type: %u",
+ offset, size, userData, name ? name : "vma_empty",
+ (uint32_t)allocation->GetSuballocationType());
#endif // VMA_STATS_STRING_ENABLED
}
@@ -6372,27 +6468,19 @@ void VmaBlockMetadata::DebugLogAllocation(VkDeviceSize offset, VkDeviceSize size
#if VMA_STATS_STRING_ENABLED
void VmaBlockMetadata::PrintDetailedMap_Begin(class VmaJsonWriter& json,
- VkDeviceSize unusedBytes, size_t allocationCount, size_t unusedRangeCount, uint32_t mapRefCount) const
+ VkDeviceSize unusedBytes, size_t allocationCount, size_t unusedRangeCount) const
{
- json.BeginObject();
-
json.WriteString("TotalBytes");
json.WriteNumber(GetSize());
json.WriteString("UnusedBytes");
- json.WriteNumber(unusedBytes);
+ json.WriteSize(unusedBytes);
json.WriteString("Allocations");
- json.WriteNumber((uint64_t)allocationCount);
+ json.WriteSize(allocationCount);
json.WriteString("UnusedRanges");
- json.WriteNumber((uint64_t)unusedRangeCount);
-
- if(mapRefCount != UINT32_MAX)
- {
- json.WriteString("MapRefCount");
- json.WriteNumber(mapRefCount);
- }
+ json.WriteSize(unusedRangeCount);
json.WriteString("Suballocations");
json.BeginArray();
@@ -6408,15 +6496,11 @@ void VmaBlockMetadata::PrintDetailedMap_Allocation(class VmaJsonWriter& json,
if (IsVirtual())
{
- json.WriteString("Type");
- json.WriteString("VirtualAllocation");
-
json.WriteString("Size");
json.WriteNumber(size);
-
- if (userData != VMA_NULL)
+ if (userData)
{
- json.WriteString("UserData");
+ json.WriteString("CustomData");
json.BeginString();
json.ContinueString_Pointer(userData);
json.EndString();
@@ -6450,7 +6534,6 @@ void VmaBlockMetadata::PrintDetailedMap_UnusedRange(class VmaJsonWriter& json,
void VmaBlockMetadata::PrintDetailedMap_End(class VmaJsonWriter& json) const
{
json.EndArray();
- json.EndObject();
}
#endif // VMA_STATS_STRING_ENABLED
#endif // _VMA_BLOCK_METADATA_FUNCTIONS
@@ -7549,12 +7632,13 @@ public:
void Init(VkDeviceSize size) override;
bool Validate() const override;
size_t GetAllocationCount() const override;
+ size_t GetFreeRegionsCount() const override;
void AddDetailedStatistics(VmaDetailedStatistics& inoutStats) const override;
void AddStatistics(VmaStatistics& inoutStats) const override;
#if VMA_STATS_STRING_ENABLED
- void PrintDetailedMap(class VmaJsonWriter& json, uint32_t mapRefCount) const override;
+ void PrintDetailedMap(class VmaJsonWriter& json) const override;
#endif
bool CreateAllocationRequest(
@@ -7577,6 +7661,7 @@ public:
void* GetAllocationUserData(VmaAllocHandle allocHandle) const override;
VmaAllocHandle GetAllocationListBegin() const override;
VmaAllocHandle GetNextAllocation(VmaAllocHandle prevAlloc) const override;
+ VkDeviceSize GetNextFreeRegionSize(VmaAllocHandle alloc) const override;
void Clear() override;
void SetAllocationUserData(VmaAllocHandle allocHandle, void* userData) override;
void DebugLogAllAllocations() const override;
@@ -7815,6 +7900,13 @@ size_t VmaBlockMetadata_Linear::GetAllocationCount() const
AccessSuballocations2nd().size() - m_2ndNullItemsCount;
}
+size_t VmaBlockMetadata_Linear::GetFreeRegionsCount() const
+{
+ // Function only used for defragmentation, which is disabled for this algorithm
+ VMA_ASSERT(0);
+ return SIZE_MAX;
+}
+
void VmaBlockMetadata_Linear::AddDetailedStatistics(VmaDetailedStatistics& inoutStats) const
{
const VkDeviceSize size = GetSize();
@@ -8133,7 +8225,7 @@ void VmaBlockMetadata_Linear::AddStatistics(VmaStatistics& inoutStats) const
}
#if VMA_STATS_STRING_ENABLED
-void VmaBlockMetadata_Linear::PrintDetailedMap(class VmaJsonWriter& json, uint32_t mapRefCount) const
+void VmaBlockMetadata_Linear::PrintDetailedMap(class VmaJsonWriter& json) const
{
const VkDeviceSize size = GetSize();
const SuballocationVectorType& suballocations1st = AccessSuballocations1st();
@@ -8295,7 +8387,7 @@ void VmaBlockMetadata_Linear::PrintDetailedMap(class VmaJsonWriter& json, uint32
}
const VkDeviceSize unusedBytes = size - usedBytes;
- PrintDetailedMap_Begin(json, unusedBytes, alloc1stCount + alloc2ndCount, unusedRangeCount, mapRefCount);
+ PrintDetailedMap_Begin(json, unusedBytes, alloc1stCount + alloc2ndCount, unusedRangeCount);
// SECOND PASS
lastOffset = 0;
@@ -8686,6 +8778,13 @@ VmaAllocHandle VmaBlockMetadata_Linear::GetNextAllocation(VmaAllocHandle prevAll
return VK_NULL_HANDLE;
}
+VkDeviceSize VmaBlockMetadata_Linear::GetNextFreeRegionSize(VmaAllocHandle alloc) const
+{
+ // Function only used for defragmentation, which is disabled for this algorithm
+ VMA_ASSERT(0);
+ return 0;
+}
+
void VmaBlockMetadata_Linear::Clear()
{
m_SumFreeSize = GetSize();
@@ -9884,6 +9983,7 @@ public:
virtual ~VmaBlockMetadata_TLSF();
size_t GetAllocationCount() const override { return m_AllocCount; }
+ size_t GetFreeRegionsCount() const override { return m_BlocksFreeCount + 1; }
VkDeviceSize GetSumFreeSize() const override { return m_BlocksFreeSize + m_NullBlock->size; }
bool IsEmpty() const override { return m_NullBlock->offset == 0; }
VkDeviceSize GetAllocationOffset(VmaAllocHandle allocHandle) const override { return ((Block*)allocHandle)->offset; };
@@ -9895,7 +9995,7 @@ public:
void AddStatistics(VmaStatistics& inoutStats) const override;
#if VMA_STATS_STRING_ENABLED
- void PrintDetailedMap(class VmaJsonWriter& json, uint32_t mapRefCount) const override;
+ void PrintDetailedMap(class VmaJsonWriter& json) const override;
#endif
bool CreateAllocationRequest(
@@ -9917,6 +10017,7 @@ public:
void* GetAllocationUserData(VmaAllocHandle allocHandle) const override;
VmaAllocHandle GetAllocationListBegin() const override;
VmaAllocHandle GetNextAllocation(VmaAllocHandle prevAlloc) const override;
+ VkDeviceSize GetNextFreeRegionSize(VmaAllocHandle alloc) const override;
void Clear() override;
void SetAllocationUserData(VmaAllocHandle allocHandle, void* userData) override;
void DebugLogAllAllocations() const override;
@@ -10167,7 +10268,7 @@ void VmaBlockMetadata_TLSF::AddStatistics(VmaStatistics& inoutStats) const
}
#if VMA_STATS_STRING_ENABLED
-void VmaBlockMetadata_TLSF::PrintDetailedMap(class VmaJsonWriter& json, uint32_t mapRefCount) const
+void VmaBlockMetadata_TLSF::PrintDetailedMap(class VmaJsonWriter& json) const
{
size_t blockCount = m_AllocCount + m_BlocksFreeCount;
VmaStlAllocator<Block*> allocator(GetAllocationCallbacks());
@@ -10184,12 +10285,10 @@ void VmaBlockMetadata_TLSF::PrintDetailedMap(class VmaJsonWriter& json, uint32_t
VmaClearDetailedStatistics(stats);
AddDetailedStatistics(stats);
- PrintDetailedMap_Begin(
- json,
+ PrintDetailedMap_Begin(json,
stats.statistics.blockBytes - stats.statistics.allocationBytes,
stats.statistics.allocationCount,
- stats.unusedRangeCount,
- mapRefCount);
+ stats.unusedRangeCount);
for (; i < blockCount; ++i)
{
@@ -10593,6 +10692,16 @@ VmaAllocHandle VmaBlockMetadata_TLSF::GetNextAllocation(VmaAllocHandle prevAlloc
return VK_NULL_HANDLE;
}
+VkDeviceSize VmaBlockMetadata_TLSF::GetNextFreeRegionSize(VmaAllocHandle alloc) const
+{
+ Block* block = (Block*)alloc;
+ VMA_ASSERT(!block->IsFree() && "Incorrect block!");
+
+ if (block->prevPhysical)
+ return block->prevPhysical->IsFree() ? block->prevPhysical->size : 0;
+ return 0;
+}
+
void VmaBlockMetadata_TLSF::Clear()
{
m_AllocCount = 0;
@@ -10837,6 +10946,7 @@ public:
size_t GetBlockCount() const { return m_Blocks.size(); }
// To be used only while the m_Mutex is locked. Used during defragmentation.
VmaDeviceMemoryBlock* GetBlock(size_t index) const { return m_Blocks[index]; }
+ VMA_RW_MUTEX &GetMutex() { return m_Mutex; }
VkResult CreateMinBlocks();
void AddStatistics(VmaStatistics& inoutStats);
@@ -10852,7 +10962,7 @@ public:
size_t allocationCount,
VmaAllocation* pAllocations);
- void Free(const VmaAllocation hAllocation, bool incrementalSort = true);
+ void Free(const VmaAllocation hAllocation);
#if VMA_STATS_STRING_ENABLED
void PrintDetailedMap(class VmaJsonWriter& json);
@@ -10878,6 +10988,9 @@ private:
// Incrementally sorted by sumFreeSize, ascending.
VmaVector<VmaDeviceMemoryBlock*, VmaStlAllocator<VmaDeviceMemoryBlock*>> m_Blocks;
uint32_t m_NextBlockId;
+ bool m_IncrementalSort = true;
+
+ void SetIncrementalSort(bool val) { m_IncrementalSort = val; }
VkDeviceSize CalcMaxBlockSize() const;
// Finds and removes given block from vector.
@@ -10928,17 +11041,26 @@ public:
const VmaDefragmentationInfo& info);
~VmaDefragmentationContext_T();
- void GetStats(VmaDefragmentationStats& outStats) { outStats = m_Stats; }
+ void GetStats(VmaDefragmentationStats& outStats) { outStats = m_GlobalStats; }
VkResult DefragmentPassBegin(VmaDefragmentationPassMoveInfo& moveInfo);
VkResult DefragmentPassEnd(VmaDefragmentationPassMoveInfo& moveInfo);
private:
- struct ImmovableBlock
+ // Max number of allocations to ignore due to size constraints before ending single pass
+ static const uint8_t MAX_ALLOCS_TO_IGNORE = 16;
+ enum class CounterStatus { Pass, Ignore, End };
+
+ struct FragmentedBlock
{
- uint32_t vectorIndex;
+ uint32_t data;
VmaDeviceMemoryBlock* block;
};
+ struct StateBalanced
+ {
+ VkDeviceSize avgFreeSize = 0;
+ VkDeviceSize avgAllocSize = UINT64_MAX;
+ };
struct StateExtensive
{
enum class Operation : uint8_t
@@ -10966,25 +11088,29 @@ private:
VmaStlAllocator<VmaDefragmentationMove> m_MoveAllocator;
VmaVector<VmaDefragmentationMove, VmaStlAllocator<VmaDefragmentationMove>> m_Moves;
+ uint8_t m_IgnoredAllocs = 0;
uint32_t m_Algorithm;
uint32_t m_BlockVectorCount;
VmaBlockVector* m_PoolBlockVector;
VmaBlockVector** m_pBlockVectors;
size_t m_ImmovableBlockCount = 0;
- VmaDefragmentationStats m_Stats = { 0 };
+ VmaDefragmentationStats m_GlobalStats = { 0 };
+ VmaDefragmentationStats m_PassStats = { 0 };
void* m_AlgorithmState = VMA_NULL;
static MoveAllocationData GetMoveData(VmaAllocHandle handle, VmaBlockMetadata* metadata);
- bool IncrementCounters(uint32_t& allocations, VkDeviceSize bytes);
+ CounterStatus CheckCounters(VkDeviceSize bytes);
+ bool IncrementCounters(VkDeviceSize bytes);
bool ReallocWithinBlock(VmaBlockVector& vector, VmaDeviceMemoryBlock* block);
bool AllocInOtherBlock(size_t start, size_t end, MoveAllocationData& data, VmaBlockVector& vector);
bool ComputeDefragmentation(VmaBlockVector& vector, size_t index);
bool ComputeDefragmentation_Fast(VmaBlockVector& vector);
- bool ComputeDefragmentation_Balanced(VmaBlockVector& vector);
+ bool ComputeDefragmentation_Balanced(VmaBlockVector& vector, size_t index, bool update);
bool ComputeDefragmentation_Full(VmaBlockVector& vector);
bool ComputeDefragmentation_Extensive(VmaBlockVector& vector, size_t index);
+ void UpdateVectorStatistics(VmaBlockVector& vector, StateBalanced& state);
bool MoveDataToFreeBlocks(VmaSuballocationType currentType,
VmaBlockVector& vector, size_t firstFreeBlock,
bool& texturePresent, bool& bufferPresent, bool& otherPresent);
@@ -11259,8 +11385,9 @@ void VmaVirtualBlock_T::BuildStatsString(bool detailedMap, VmaStringBuilder& sb)
if (detailedMap)
{
json.WriteString("Details");
- m_Metadata->PrintDetailedMap(json,
- UINT32_MAX); // mapRefCount
+ json.BeginObject();
+ m_Metadata->PrintDetailedMap(json);
+ json.EndObject();
}
json.EndObject();
@@ -11868,18 +11995,17 @@ VkResult VmaDeviceMemoryBlock::BindImageMemory(
#endif // _VMA_DEVICE_MEMORY_BLOCK_FUNCTIONS
#ifndef _VMA_ALLOCATION_T_FUNCTIONS
-VmaAllocation_T::VmaAllocation_T(bool userDataString, bool mappingAllowed)
+VmaAllocation_T::VmaAllocation_T(bool mappingAllowed)
: m_Alignment{ 1 },
m_Size{ 0 },
m_pUserData{ VMA_NULL },
+ m_pName{ VMA_NULL },
m_MemoryTypeIndex{ 0 },
m_Type{ (uint8_t)ALLOCATION_TYPE_NONE },
m_SuballocationType{ (uint8_t)VMA_SUBALLOCATION_TYPE_UNKNOWN },
m_MapCount{ 0 },
m_Flags{ 0 }
{
- if(userDataString)
- m_Flags |= (uint8_t)FLAG_USER_DATA_STRING;
if(mappingAllowed)
m_Flags |= (uint8_t)FLAG_MAPPING_ALLOWED;
@@ -11893,7 +12019,7 @@ VmaAllocation_T::~VmaAllocation_T()
VMA_ASSERT(m_MapCount == 0 && "Allocation was not unmapped before destruction.");
// Check if owned string was freed.
- VMA_ASSERT((IsUserDataString() && m_pUserData == VMA_NULL) || !IsUserDataString());
+ VMA_ASSERT(m_pName == VMA_NULL);
}
void VmaAllocation_T::InitBlockAllocation(
@@ -11948,31 +12074,25 @@ void VmaAllocation_T::InitDedicatedAllocation(
m_DedicatedAllocation.m_Next = VMA_NULL;
}
-void VmaAllocation_T::SetUserData(VmaAllocator hAllocator, void* pUserData)
+void VmaAllocation_T::SetName(VmaAllocator hAllocator, const char* pName)
{
- if (IsUserDataString())
- {
- VMA_ASSERT(pUserData == VMA_NULL || pUserData != m_pUserData);
+ VMA_ASSERT(pName == VMA_NULL || pName != m_pName);
- FreeUserDataString(hAllocator);
+ FreeName(hAllocator);
- if (pUserData != VMA_NULL)
- {
- m_pUserData = VmaCreateStringCopy(hAllocator->GetAllocationCallbacks(), (const char*)pUserData);
- }
- }
- else
- {
- m_pUserData = pUserData;
- }
+ if (pName != VMA_NULL)
+ m_pName = VmaCreateStringCopy(hAllocator->GetAllocationCallbacks(), pName);
}
-void VmaAllocation_T::SwapBlockAllocation(VmaAllocation allocation)
+uint8_t VmaAllocation_T::SwapBlockAllocation(VmaAllocator hAllocator, VmaAllocation allocation)
{
VMA_ASSERT(allocation != VMA_NULL);
VMA_ASSERT(m_Type == ALLOCATION_TYPE_BLOCK);
VMA_ASSERT(allocation->m_Type == ALLOCATION_TYPE_BLOCK);
+ if (m_MapCount != 0)
+ m_BlockAllocation.m_Block->Unmap(hAllocator, m_MapCount);
+
m_BlockAllocation.m_Block->m_pMetadata->SetAllocationUserData(m_BlockAllocation.m_AllocHandle, allocation);
VMA_SWAP(m_BlockAllocation, allocation->m_BlockAllocation);
m_BlockAllocation.m_Block->m_pMetadata->SetAllocationUserData(m_BlockAllocation.m_AllocHandle, this);
@@ -11980,6 +12100,7 @@ void VmaAllocation_T::SwapBlockAllocation(VmaAllocation allocation)
#if VMA_STATS_STRING_ENABLED
VMA_SWAP(m_BufferImageUsage, allocation->m_BufferImageUsage);
#endif
+ return m_MapCount;
}
VmaAllocHandle VmaAllocation_T::GetAllocHandle() const
@@ -12165,35 +12286,31 @@ void VmaAllocation_T::PrintParameters(class VmaJsonWriter& json) const
json.WriteString("Size");
json.WriteNumber(m_Size);
+ json.WriteString("Usage");
+ json.WriteNumber(m_BufferImageUsage);
if (m_pUserData != VMA_NULL)
{
- json.WriteString("UserData");
- if (IsUserDataString())
- {
- json.WriteString((const char*)m_pUserData);
- }
- else
- {
- json.BeginString();
- json.ContinueString_Pointer(m_pUserData);
- json.EndString();
- }
+ json.WriteString("CustomData");
+ json.BeginString();
+ json.ContinueString_Pointer(m_pUserData);
+ json.EndString();
}
-
- if (m_BufferImageUsage != 0)
+ if (m_pName != VMA_NULL)
{
- json.WriteString("Usage");
- json.WriteNumber(m_BufferImageUsage);
+ json.WriteString("Name");
+ json.WriteString(m_pName);
}
}
#endif // VMA_STATS_STRING_ENABLED
-void VmaAllocation_T::FreeUserDataString(VmaAllocator hAllocator)
+void VmaAllocation_T::FreeName(VmaAllocator hAllocator)
{
- VMA_ASSERT(IsUserDataString());
- VmaFreeString(hAllocator->GetAllocationCallbacks(), (char*)m_pUserData);
- m_pUserData = VMA_NULL;
+ if(m_pName)
+ {
+ VmaFreeString(hAllocator->GetAllocationCallbacks(), m_pName);
+ m_pName = VMA_NULL;
+ }
}
#endif // _VMA_ALLOCATION_T_FUNCTIONS
@@ -12542,9 +12659,7 @@ VkResult VmaBlockVector::AllocatePage(
return VK_ERROR_OUT_OF_DEVICE_MEMORY;
}
-void VmaBlockVector::Free(
- const VmaAllocation hAllocation,
- bool incrementalSort)
+void VmaBlockVector::Free(const VmaAllocation hAllocation)
{
VmaDeviceMemoryBlock* pBlockToDelete = VMA_NULL;
@@ -12604,8 +12719,7 @@ void VmaBlockVector::Free(
}
}
- if (incrementalSort)
- IncrementallySortBlocks();
+ IncrementallySortBlocks();
}
// Destruction of a free block. Deferred until this point, outside of mutex
@@ -12650,6 +12764,8 @@ void VmaBlockVector::Remove(VmaDeviceMemoryBlock* pBlock)
void VmaBlockVector::IncrementallySortBlocks()
{
+ if (!m_IncrementalSort)
+ return;
if (m_Algorithm != VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT)
{
// Bubble sort only until first swap.
@@ -12724,7 +12840,7 @@ VkResult VmaBlockVector::CommitAllocationRequest(
}
}
- *pAllocation = m_hAllocator->m_AllocationObjectAllocator.Allocate(isUserDataString, isMappingAllowed);
+ *pAllocation = m_hAllocator->m_AllocationObjectAllocator.Allocate(isMappingAllowed);
pBlock->m_pMetadata->Alloc(allocRequest, suballocType, *pAllocation);
(*pAllocation)->InitBlockAllocation(
pBlock,
@@ -12735,7 +12851,10 @@ VkResult VmaBlockVector::CommitAllocationRequest(
suballocType,
mapped);
VMA_HEAVY_ASSERT(pBlock->Validate());
- (*pAllocation)->SetUserData(m_hAllocator, pUserData);
+ if (isUserDataString)
+ (*pAllocation)->SetName(m_hAllocator, (const char*)pUserData);
+ else
+ (*pAllocation)->SetUserData(m_hAllocator, pUserData);
m_hAllocator->m_Budget.AddAllocation(m_hAllocator->MemoryTypeIndexToHeapIndex(m_MemoryTypeIndex), allocRequest.size);
if (VMA_DEBUG_INITIALIZE_ALLOCATIONS)
{
@@ -12770,6 +12889,7 @@ VkResult VmaBlockVector::CreateBlock(VkDeviceSize blockSize, size_t* pNewBlockIn
VkMemoryPriorityAllocateInfoEXT priorityInfo = { VK_STRUCTURE_TYPE_MEMORY_PRIORITY_ALLOCATE_INFO_EXT };
if (m_hAllocator->m_UseExtMemoryPriority)
{
+ VMA_ASSERT(m_Priority >= 0.f && m_Priority <= 1.f);
priorityInfo.priority = m_Priority;
VmaPnextChainPushFront(&allocInfo, &priorityInfo);
}
@@ -12833,50 +12953,7 @@ void VmaBlockVector::PrintDetailedMap(class VmaJsonWriter& json)
{
VmaMutexLockRead lock(m_Mutex, m_hAllocator->m_UseMutex);
- if (IsCustomPool())
- {
- const char* poolName = m_hParentPool->GetName();
- if (poolName != VMA_NULL && poolName[0] != '\0')
- {
- json.WriteString("Name");
- json.WriteString(poolName);
- }
-
- json.WriteString("MemoryTypeIndex");
- json.WriteNumber(m_MemoryTypeIndex);
-
- json.WriteString("BlockSize");
- json.WriteNumber(m_PreferredBlockSize);
-
- json.WriteString("BlockCount");
- json.BeginObject(true);
- if (m_MinBlockCount > 0)
- {
- json.WriteString("Min");
- json.WriteNumber((uint64_t)m_MinBlockCount);
- }
- if (m_MaxBlockCount < SIZE_MAX)
- {
- json.WriteString("Max");
- json.WriteNumber((uint64_t)m_MaxBlockCount);
- }
- json.WriteString("Cur");
- json.WriteNumber((uint64_t)m_Blocks.size());
- json.EndObject();
-
- if (m_Algorithm != 0)
- {
- json.WriteString("Algorithm");
- json.WriteString(VmaAlgorithmToStr(m_Algorithm));
- }
- }
- else
- {
- json.WriteString("PreferredBlockSize");
- json.WriteNumber(m_PreferredBlockSize);
- }
- json.WriteString("Blocks");
json.BeginObject();
for (size_t i = 0; i < m_Blocks.size(); ++i)
{
@@ -12884,7 +12961,12 @@ void VmaBlockVector::PrintDetailedMap(class VmaJsonWriter& json)
json.ContinueString(m_Blocks[i]->GetId());
json.EndString();
- m_Blocks[i]->m_pMetadata->PrintDetailedMap(json, m_Blocks[i]->GetMapRefCount());
+ json.BeginObject();
+ json.WriteString("MapRefCount");
+ json.WriteNumber(m_Blocks[i]->GetMapRefCount());
+
+ m_Blocks[i]->m_pMetadata->PrintDetailedMap(json);
+ json.EndObject();
}
json.EndObject();
}
@@ -12929,6 +13011,7 @@ VmaDefragmentationContext_T::VmaDefragmentationContext_T(
m_BlockVectorCount = 1;
m_PoolBlockVector = &info.pool->m_BlockVector;
m_pBlockVectors = &m_PoolBlockVector;
+ m_PoolBlockVector->SetIncrementalSort(false);
m_PoolBlockVector->SortByFreeSize();
}
else
@@ -12940,12 +13023,22 @@ VmaDefragmentationContext_T::VmaDefragmentationContext_T(
{
VmaBlockVector* vector = m_pBlockVectors[i];
if (vector != VMA_NULL)
+ {
+ vector->SetIncrementalSort(false);
vector->SortByFreeSize();
+ }
}
}
switch (m_Algorithm)
{
+ case 0: // Default algorithm
+ m_Algorithm = VMA_DEFRAGMENTATION_FLAG_ALGORITHM_BALANCED_BIT;
+ case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_BALANCED_BIT:
+ {
+ m_AlgorithmState = vma_new_array(hAllocator, StateBalanced, m_BlockVectorCount);
+ break;
+ }
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_EXTENSIVE_BIT:
{
if (hAllocator->GetBufferImageGranularity() > 1)
@@ -12959,10 +13052,27 @@ VmaDefragmentationContext_T::VmaDefragmentationContext_T(
VmaDefragmentationContext_T::~VmaDefragmentationContext_T()
{
+ if (m_PoolBlockVector != VMA_NULL)
+ {
+ m_PoolBlockVector->SetIncrementalSort(true);
+ }
+ else
+ {
+ for (uint32_t i = 0; i < m_BlockVectorCount; ++i)
+ {
+ VmaBlockVector* vector = m_pBlockVectors[i];
+ if (vector != VMA_NULL)
+ vector->SetIncrementalSort(true);
+ }
+ }
+
if (m_AlgorithmState)
{
switch (m_Algorithm)
{
+ case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_BALANCED_BIT:
+ vma_delete_array(m_MoveAllocator.m_pCallbacks, reinterpret_cast<StateBalanced*>(m_AlgorithmState), m_BlockVectorCount);
+ break;
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_EXTENSIVE_BIT:
vma_delete_array(m_MoveAllocator.m_pCallbacks, reinterpret_cast<StateExtensive*>(m_AlgorithmState), m_BlockVectorCount);
break;
@@ -12976,6 +13086,8 @@ VkResult VmaDefragmentationContext_T::DefragmentPassBegin(VmaDefragmentationPass
{
if (m_PoolBlockVector != VMA_NULL)
{
+ VmaMutexLockWrite lock(m_PoolBlockVector->GetMutex(), m_PoolBlockVector->GetAllocator()->m_UseMutex);
+
if (m_PoolBlockVector->GetBlockCount() > 1)
ComputeDefragmentation(*m_PoolBlockVector, 0);
else if (m_PoolBlockVector->GetBlockCount() == 1)
@@ -12987,6 +13099,8 @@ VkResult VmaDefragmentationContext_T::DefragmentPassBegin(VmaDefragmentationPass
{
if (m_pBlockVectors[i] != VMA_NULL)
{
+ VmaMutexLockWrite lock(m_pBlockVectors[i]->GetMutex(), m_pBlockVectors[i]->GetAllocator()->m_UseMutex);
+
if (m_pBlockVectors[i]->GetBlockCount() > 1)
{
if (ComputeDefragmentation(*m_pBlockVectors[i], i))
@@ -13017,7 +13131,11 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
VMA_ASSERT(moveInfo.moveCount > 0 ? moveInfo.pMoves != VMA_NULL : true);
VkResult result = VK_SUCCESS;
- VmaVector<ImmovableBlock, VmaStlAllocator<ImmovableBlock>> immovableBlocks(VmaStlAllocator<ImmovableBlock>(m_MoveAllocator.m_pCallbacks));
+ VmaStlAllocator<FragmentedBlock> blockAllocator(m_MoveAllocator.m_pCallbacks);
+ VmaVector<FragmentedBlock, VmaStlAllocator<FragmentedBlock>> immovableBlocks(blockAllocator);
+ VmaVector<FragmentedBlock, VmaStlAllocator<FragmentedBlock>> mappedBlocks(blockAllocator);
+
+ VmaAllocator allocator = VMA_NULL;
for (uint32_t i = 0; i < moveInfo.moveCount; ++i)
{
VmaDefragmentationMove& move = moveInfo.pMoves[i];
@@ -13037,29 +13155,54 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
vector = m_pBlockVectors[vectorIndex];
VMA_ASSERT(vector != VMA_NULL);
}
-
- VmaAllocation dst = reinterpret_cast<VmaAllocation>(move.internalData);
+
switch (move.operation)
{
case VMA_DEFRAGMENTATION_MOVE_OPERATION_COPY:
{
- move.srcAllocation->SwapBlockAllocation(dst);
- prevCount = vector->GetBlockCount();
- freedBlockSize = dst->GetBlock()->m_pMetadata->GetSize();
- vector->Free(dst, false);
- currentCount = vector->GetBlockCount();
+ uint8_t mapCount = move.srcAllocation->SwapBlockAllocation(vector->m_hAllocator, move.dstTmpAllocation);
+ if (mapCount > 0)
+ {
+ allocator = vector->m_hAllocator;
+ VmaDeviceMemoryBlock* newMapBlock = move.srcAllocation->GetBlock();
+ bool notPresent = true;
+ for (FragmentedBlock& block : mappedBlocks)
+ {
+ if (block.block == newMapBlock)
+ {
+ notPresent = false;
+ block.data += mapCount;
+ break;
+ }
+ }
+ if (notPresent)
+ mappedBlocks.push_back({ mapCount, newMapBlock });
+ }
+
+ // Scope for locks, Free have it's own lock
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ prevCount = vector->GetBlockCount();
+ freedBlockSize = move.dstTmpAllocation->GetBlock()->m_pMetadata->GetSize();
+ }
+ vector->Free(move.dstTmpAllocation);
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ currentCount = vector->GetBlockCount();
+ }
result = VK_INCOMPLETE;
break;
}
case VMA_DEFRAGMENTATION_MOVE_OPERATION_IGNORE:
{
- m_Stats.bytesMoved -= move.srcAllocation->GetSize();
- vector->Free(dst, false);
+ m_PassStats.bytesMoved -= move.srcAllocation->GetSize();
+ --m_PassStats.allocationsMoved;
+ vector->Free(move.dstTmpAllocation);
VmaDeviceMemoryBlock* newBlock = move.srcAllocation->GetBlock();
bool notPresent = true;
- for (const ImmovableBlock& block : immovableBlocks)
+ for (const FragmentedBlock& block : immovableBlocks)
{
if (block.block == newBlock)
{
@@ -13073,16 +13216,32 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
}
case VMA_DEFRAGMENTATION_MOVE_OPERATION_DESTROY:
{
- prevCount = vector->GetBlockCount();
- freedBlockSize = move.srcAllocation->GetBlock()->m_pMetadata->GetSize();
- vector->Free(move.srcAllocation, false);
- currentCount = vector->GetBlockCount();
+ m_PassStats.bytesMoved -= move.srcAllocation->GetSize();
+ --m_PassStats.allocationsMoved;
+ // Scope for locks, Free have it's own lock
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ prevCount = vector->GetBlockCount();
+ freedBlockSize = move.srcAllocation->GetBlock()->m_pMetadata->GetSize();
+ }
+ vector->Free(move.srcAllocation);
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ currentCount = vector->GetBlockCount();
+ }
freedBlockSize *= prevCount - currentCount;
- VkDeviceSize dstBlockSize = dst->GetBlock()->m_pMetadata->GetSize();
- vector->Free(dst, false);
- freedBlockSize += dstBlockSize * (currentCount - vector->GetBlockCount());
- currentCount = vector->GetBlockCount();
+ VkDeviceSize dstBlockSize;
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ dstBlockSize = move.dstTmpAllocation->GetBlock()->m_pMetadata->GetSize();
+ }
+ vector->Free(move.dstTmpAllocation);
+ {
+ VmaMutexLockRead lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+ freedBlockSize += dstBlockSize * (currentCount - vector->GetBlockCount());
+ currentCount = vector->GetBlockCount();
+ }
result = VK_INCOMPLETE;
break;
@@ -13094,8 +13253,8 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
if (prevCount > currentCount)
{
size_t freedBlocks = prevCount - currentCount;
- m_Stats.deviceMemoryBlocksFreed += static_cast<uint32_t>(freedBlocks);
- m_Stats.bytesFreed += freedBlockSize;
+ m_PassStats.deviceMemoryBlocksFreed += static_cast<uint32_t>(freedBlocks);
+ m_PassStats.bytesFreed += freedBlockSize;
}
switch (m_Algorithm)
@@ -13108,9 +13267,15 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
StateExtensive& state = reinterpret_cast<StateExtensive*>(m_AlgorithmState)[vectorIndex];
if (state.firstFreeBlock != SIZE_MAX)
{
- state.firstFreeBlock -= prevCount - currentCount;
- if (state.firstFreeBlock != 0)
- state.firstFreeBlock -= vector->GetBlock(state.firstFreeBlock - 1)->m_pMetadata->IsEmpty();
+ const size_t diff = prevCount - currentCount;
+ if (state.firstFreeBlock >= diff)
+ {
+ state.firstFreeBlock -= diff;
+ if (state.firstFreeBlock != 0)
+ state.firstFreeBlock -= vector->GetBlock(state.firstFreeBlock - 1)->m_pMetadata->IsEmpty();
+ }
+ else
+ state.firstFreeBlock = 0;
}
}
}
@@ -13120,6 +13285,13 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
moveInfo.pMoves = VMA_NULL;
m_Moves.clear();
+ // Update stats
+ m_GlobalStats.allocationsMoved += m_PassStats.allocationsMoved;
+ m_GlobalStats.bytesFreed += m_PassStats.bytesFreed;
+ m_GlobalStats.bytesMoved += m_PassStats.bytesMoved;
+ m_GlobalStats.deviceMemoryBlocksFreed += m_PassStats.deviceMemoryBlocksFreed;
+ m_PassStats = { 0 };
+
// Move blocks with immovable allocations according to algorithm
if (immovableBlocks.size() > 0)
{
@@ -13131,12 +13303,14 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
{
bool swapped = false;
// Move to the start of free blocks range
- for (const ImmovableBlock& block : immovableBlocks)
+ for (const FragmentedBlock& block : immovableBlocks)
{
- StateExtensive& state = reinterpret_cast<StateExtensive*>(m_AlgorithmState)[block.vectorIndex];
+ StateExtensive& state = reinterpret_cast<StateExtensive*>(m_AlgorithmState)[block.data];
if (state.operation != StateExtensive::Operation::Cleanup)
{
- VmaBlockVector* vector = m_pBlockVectors[block.vectorIndex];
+ VmaBlockVector* vector = m_pBlockVectors[block.data];
+ VmaMutexLockWrite lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+
for (size_t i = 0, count = vector->GetBlockCount() - m_ImmovableBlockCount; i < count; ++i)
{
if (vector->GetBlock(i) == block.block)
@@ -13144,9 +13318,12 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
VMA_SWAP(vector->m_Blocks[i], vector->m_Blocks[vector->GetBlockCount() - ++m_ImmovableBlockCount]);
if (state.firstFreeBlock != SIZE_MAX)
{
- if (i < state.firstFreeBlock - 1)
+ if (i + 1 < state.firstFreeBlock)
{
- VMA_SWAP(vector->m_Blocks[i], vector->m_Blocks[--state.firstFreeBlock]);
+ if (state.firstFreeBlock > 1)
+ VMA_SWAP(vector->m_Blocks[i], vector->m_Blocks[--state.firstFreeBlock]);
+ else
+ --state.firstFreeBlock;
}
}
swapped = true;
@@ -13163,10 +13340,12 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
default:
{
// Move to the begining
- for (const ImmovableBlock& block : immovableBlocks)
+ for (const FragmentedBlock& block : immovableBlocks)
{
- VmaBlockVector* vector = m_pBlockVectors[block.vectorIndex];
- for (size_t i = m_ImmovableBlockCount; vector->GetBlockCount(); ++i)
+ VmaBlockVector* vector = m_pBlockVectors[block.data];
+ VmaMutexLockWrite lock(vector->GetMutex(), vector->GetAllocator()->m_UseMutex);
+
+ for (size_t i = m_ImmovableBlockCount; i < vector->GetBlockCount(); ++i)
{
if (vector->GetBlock(i) == block.block)
{
@@ -13179,6 +13358,13 @@ VkResult VmaDefragmentationContext_T::DefragmentPassEnd(VmaDefragmentationPassMo
}
}
}
+
+ // Bulk-map destination blocks
+ for (const FragmentedBlock& block : mappedBlocks)
+ {
+ VkResult res = block.block->Map(allocator, block.data, VMA_NULL);
+ VMA_ASSERT(res == VK_SUCCESS);
+ }
return result;
}
@@ -13188,9 +13374,10 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation(VmaBlockVector& vector,
{
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_FAST_BIT:
return ComputeDefragmentation_Fast(vector);
- default: // Default algoritm
+ default:
+ VMA_ASSERT(0);
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_BALANCED_BIT:
- return ComputeDefragmentation_Balanced(vector);
+ return ComputeDefragmentation_Balanced(vector, index, true);
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_FULL_BIT:
return ComputeDefragmentation_Full(vector);
case VMA_DEFRAGMENTATION_FLAG_ALGORITHM_EXTENSIVE_BIT:
@@ -13216,12 +13403,27 @@ VmaDefragmentationContext_T::MoveAllocationData VmaDefragmentationContext_T::Get
return moveData;
}
-bool VmaDefragmentationContext_T::IncrementCounters(uint32_t& allocations, VkDeviceSize bytes)
+VmaDefragmentationContext_T::CounterStatus VmaDefragmentationContext_T::CheckCounters(VkDeviceSize bytes)
{
- if (++allocations >= m_MaxPassAllocations || bytes >= m_MaxPassBytes)
+ // Ignore allocation if will exceed max size for copy
+ if (m_PassStats.bytesMoved + bytes > m_MaxPassBytes)
{
- m_Stats.bytesMoved += bytes;
- m_Stats.allocationsMoved += allocations;
+ if (++m_IgnoredAllocs < MAX_ALLOCS_TO_IGNORE)
+ return CounterStatus::Ignore;
+ else
+ return CounterStatus::End;
+ }
+ return CounterStatus::Pass;
+}
+
+bool VmaDefragmentationContext_T::IncrementCounters(VkDeviceSize bytes)
+{
+ m_PassStats.bytesMoved += bytes;
+ // Early return when max found
+ if (++m_PassStats.allocationsMoved >= m_MaxPassAllocations || m_PassStats.bytesMoved >= m_MaxPassBytes)
+ {
+ VMA_ASSERT(m_PassStats.allocationsMoved == m_MaxPassAllocations ||
+ m_PassStats.bytesMoved == m_MaxPassBytes && "Exceeded maximal pass threshold!");
return true;
}
return false;
@@ -13229,8 +13431,6 @@ bool VmaDefragmentationContext_T::IncrementCounters(uint32_t& allocations, VkDev
bool VmaDefragmentationContext_T::ReallocWithinBlock(VmaBlockVector& vector, VmaDeviceMemoryBlock* block)
{
- VkDeviceSize currentBytesMoved = 0;
- uint32_t currentAllocsMoved = 0;
VmaBlockMetadata* metadata = block->m_pMetadata;
for (VmaAllocHandle handle = metadata->GetAllocationListBegin();
@@ -13241,8 +13441,18 @@ bool VmaDefragmentationContext_T::ReallocWithinBlock(VmaBlockVector& vector, Vma
// Ignore newly created allocations by defragmentation algorithm
if (moveData.move.srcAllocation->GetUserData() == this)
continue;
- VmaAllocation& dst = reinterpret_cast<VmaAllocation&>(moveData.move.internalData);
-
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
+
VkDeviceSize offset = moveData.move.srcAllocation->GetOffset();
if (offset != 0 && metadata->GetSumFreeSize() >= moveData.size)
{
@@ -13264,32 +13474,21 @@ bool VmaDefragmentationContext_T::ReallocWithinBlock(VmaBlockVector& vector, Vma
moveData.flags,
this,
moveData.type,
- &dst) == VK_SUCCESS)
+ &moveData.move.dstTmpAllocation) == VK_SUCCESS)
{
- moveData.move.dstMemory = dst->GetMemory();
- moveData.move.dstOffset = dst->GetOffset();
m_Moves.push_back(moveData.move);
- currentBytesMoved += moveData.size;
-
- if (IncrementCounters(currentAllocsMoved, currentBytesMoved))
+ if (IncrementCounters(moveData.size))
return true;
}
}
}
}
}
-
- m_Stats.bytesMoved += currentBytesMoved;
- m_Stats.allocationsMoved += currentAllocsMoved;
return false;
}
bool VmaDefragmentationContext_T::AllocInOtherBlock(size_t start, size_t end, MoveAllocationData& data, VmaBlockVector& vector)
{
- VkDeviceSize currentBytesMoved = 0;
- uint32_t currentAllocsMoved = 0;
- VmaAllocation& dst = reinterpret_cast<VmaAllocation&>(data.move.internalData);
-
for (; start < end; ++start)
{
VmaDeviceMemoryBlock* dstBlock = vector.GetBlock(start);
@@ -13302,22 +13501,15 @@ bool VmaDefragmentationContext_T::AllocInOtherBlock(size_t start, size_t end, Mo
this,
data.type,
0,
- &dst) == VK_SUCCESS)
+ &data.move.dstTmpAllocation) == VK_SUCCESS)
{
- data.move.dstMemory = dst->GetMemory();
- data.move.dstOffset = dst->GetOffset();
m_Moves.push_back(data.move);
- currentBytesMoved += data.size;
-
- if (IncrementCounters(currentAllocsMoved, currentBytesMoved))
+ if (IncrementCounters(data.size))
return true;
break;
}
}
}
-
- m_Stats.bytesMoved += currentBytesMoved;
- m_Stats.allocationsMoved += currentAllocsMoved;
return false;
}
@@ -13338,6 +13530,17 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Fast(VmaBlockVector& ve
// Ignore newly created allocations by defragmentation algorithm
if (moveData.move.srcAllocation->GetUserData() == this)
continue;
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
// Check all previous blocks for free space
if (AllocInOtherBlock(0, i, moveData, vector))
@@ -13347,19 +13550,24 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Fast(VmaBlockVector& ve
return false;
}
-bool VmaDefragmentationContext_T::ComputeDefragmentation_Balanced(VmaBlockVector& vector)
+bool VmaDefragmentationContext_T::ComputeDefragmentation_Balanced(VmaBlockVector& vector, size_t index, bool update)
{
// Go over every allocation and try to fit it in previous blocks at lowest offsets,
// if not possible: realloc within single block to minimize offset (exclude offset == 0),
// but only if there are noticable gaps between them (some heuristic, ex. average size of allocation in block)
+ VMA_ASSERT(m_AlgorithmState != VMA_NULL);
- VkDeviceSize currentBytesMoved = 0;
- uint32_t currentAllocsMoved = 0;
+ StateBalanced& vectorState = reinterpret_cast<StateBalanced*>(m_AlgorithmState)[index];
+ if (update && vectorState.avgAllocSize == UINT64_MAX)
+ UpdateVectorStatistics(vector, vectorState);
+ const size_t startMoveCount = m_Moves.size();
+ VkDeviceSize minimalFreeRegion = vectorState.avgFreeSize / 2;
for (size_t i = vector.GetBlockCount() - 1; i > m_ImmovableBlockCount; --i)
{
VmaDeviceMemoryBlock* block = vector.GetBlock(i);
VmaBlockMetadata* metadata = block->m_pMetadata;
+ VkDeviceSize prevFreeRegionSize = 0;
for (VmaAllocHandle handle = metadata->GetAllocationListBegin();
handle != VK_NULL_HANDLE;
@@ -13369,53 +13577,72 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Balanced(VmaBlockVector
// Ignore newly created allocations by defragmentation algorithm
if (moveData.move.srcAllocation->GetUserData() == this)
continue;
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
// Check all previous blocks for free space
const size_t prevMoveCount = m_Moves.size();
if (AllocInOtherBlock(0, i, moveData, vector))
return true;
+ VkDeviceSize nextFreeRegionSize = metadata->GetNextFreeRegionSize(handle);
// If no room found then realloc within block for lower offset
VkDeviceSize offset = moveData.move.srcAllocation->GetOffset();
if (prevMoveCount == m_Moves.size() && offset != 0 && metadata->GetSumFreeSize() >= moveData.size)
{
- VmaAllocationRequest request = {};
- if (metadata->CreateAllocationRequest(
- moveData.size,
- moveData.alignment,
- false,
- moveData.type,
- VMA_ALLOCATION_CREATE_STRATEGY_MIN_OFFSET_BIT,
- &request))
+ // Check if realloc will make sense
+ if (prevFreeRegionSize >= minimalFreeRegion ||
+ nextFreeRegionSize >= minimalFreeRegion ||
+ moveData.size <= vectorState.avgFreeSize ||
+ moveData.size <= vectorState.avgAllocSize)
{
- if (metadata->GetAllocationOffset(request.allocHandle) < offset)
+ VmaAllocationRequest request = {};
+ if (metadata->CreateAllocationRequest(
+ moveData.size,
+ moveData.alignment,
+ false,
+ moveData.type,
+ VMA_ALLOCATION_CREATE_STRATEGY_MIN_OFFSET_BIT,
+ &request))
{
- VmaAllocation& dst = reinterpret_cast<VmaAllocation&>(moveData.move.internalData);
- if (vector.CommitAllocationRequest(
- request,
- block,
- moveData.alignment,
- moveData.flags,
- this,
- moveData.type,
- &dst) == VK_SUCCESS)
+ if (metadata->GetAllocationOffset(request.allocHandle) < offset)
{
- moveData.move.dstMemory = dst->GetMemory();
- moveData.move.dstOffset = dst->GetOffset();
- m_Moves.push_back(moveData.move);
- currentBytesMoved += moveData.size;
-
- if (IncrementCounters(currentAllocsMoved, currentBytesMoved))
- return true;
+ if (vector.CommitAllocationRequest(
+ request,
+ block,
+ moveData.alignment,
+ moveData.flags,
+ this,
+ moveData.type,
+ &moveData.move.dstTmpAllocation) == VK_SUCCESS)
+ {
+ m_Moves.push_back(moveData.move);
+ if (IncrementCounters(moveData.size))
+ return true;
+ }
}
}
}
}
+ prevFreeRegionSize = nextFreeRegionSize;
}
}
-
- m_Stats.bytesMoved += currentBytesMoved;
- m_Stats.allocationsMoved += currentAllocsMoved;
+
+ // No moves perfomed, update statistics to current vector state
+ if (startMoveCount == m_Moves.size() && !update)
+ {
+ vectorState.avgAllocSize = UINT64_MAX;
+ return ComputeDefragmentation_Balanced(vector, index, false);
+ }
return false;
}
@@ -13424,9 +13651,6 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Full(VmaBlockVector& ve
// Go over every allocation and try to fit it in previous blocks at lowest offsets,
// if not possible: realloc within single block to minimize offset (exclude offset == 0)
- VkDeviceSize currentBytesMoved = 0;
- uint32_t currentAllocsMoved = 0;
-
for (size_t i = vector.GetBlockCount() - 1; i > m_ImmovableBlockCount; --i)
{
VmaDeviceMemoryBlock* block = vector.GetBlock(i);
@@ -13440,6 +13664,17 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Full(VmaBlockVector& ve
// Ignore newly created allocations by defragmentation algorithm
if (moveData.move.srcAllocation->GetUserData() == this)
continue;
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
// Check all previous blocks for free space
const size_t prevMoveCount = m_Moves.size();
@@ -13461,7 +13696,6 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Full(VmaBlockVector& ve
{
if (metadata->GetAllocationOffset(request.allocHandle) < offset)
{
- VmaAllocation& dst = reinterpret_cast<VmaAllocation&>(moveData.move.internalData);
if (vector.CommitAllocationRequest(
request,
block,
@@ -13469,14 +13703,10 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Full(VmaBlockVector& ve
moveData.flags,
this,
moveData.type,
- &dst) == VK_SUCCESS)
+ &moveData.move.dstTmpAllocation) == VK_SUCCESS)
{
- moveData.move.dstMemory = dst->GetMemory();
- moveData.move.dstOffset = dst->GetOffset();
m_Moves.push_back(moveData.move);
- currentBytesMoved += moveData.size;
-
- if (IncrementCounters(currentAllocsMoved, currentBytesMoved))
+ if (IncrementCounters(moveData.size))
return true;
}
}
@@ -13484,9 +13714,6 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Full(VmaBlockVector& ve
}
}
}
-
- m_Stats.bytesMoved += currentBytesMoved;
- m_Stats.allocationsMoved += currentAllocsMoved;
return false;
}
@@ -13511,6 +13738,13 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Extensive(VmaBlockVecto
case StateExtensive::Operation::FindFreeBlockTexture:
case StateExtensive::Operation::FindFreeBlockAll:
{
+ // No more blocks to free, just perform fast realloc and move to cleanup
+ if (vectorState.firstFreeBlock == 0)
+ {
+ vectorState.operation = StateExtensive::Operation::Cleanup;
+ return ComputeDefragmentation_Fast(vector);
+ }
+
// No free blocks, have to clear last one
size_t last = (vectorState.firstFreeBlock == SIZE_MAX ? vector.GetBlockCount() : vectorState.firstFreeBlock) - 1;
VmaBlockMetadata* freeMetadata = vector.GetBlock(last)->m_pMetadata;
@@ -13521,6 +13755,17 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Extensive(VmaBlockVecto
handle = freeMetadata->GetNextAllocation(handle))
{
MoveAllocationData moveData = GetMoveData(handle, freeMetadata);
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
// Check all previous blocks for free space
if (AllocInOtherBlock(0, last, moveData, vector))
@@ -13568,8 +13813,7 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Extensive(VmaBlockVecto
}
vectorState.firstFreeBlock = last;
// Nothing done, block found without reallocations, can perform another reallocs in same pass
- if (prevMoveCount == m_Moves.size())
- return ComputeDefragmentation_Extensive(vector, index);
+ return ComputeDefragmentation_Extensive(vector, index);
}
break;
}
@@ -13637,6 +13881,9 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Extensive(VmaBlockVecto
}
break;
}
+ case StateExtensive::Operation::Cleanup:
+ // Cleanup is handled below so that other operations may reuse the cleanup code. This case is here to prevent the unhandled enum value warning (C4062).
+ break;
}
if (vectorState.operation == StateExtensive::Operation::Cleanup)
@@ -13655,6 +13902,27 @@ bool VmaDefragmentationContext_T::ComputeDefragmentation_Extensive(VmaBlockVecto
return false;
}
+void VmaDefragmentationContext_T::UpdateVectorStatistics(VmaBlockVector& vector, StateBalanced& state)
+{
+ size_t allocCount = 0;
+ size_t freeCount = 0;
+ state.avgFreeSize = 0;
+ state.avgAllocSize = 0;
+
+ for (size_t i = 0; i < vector.GetBlockCount(); ++i)
+ {
+ VmaBlockMetadata* metadata = vector.GetBlock(i)->m_pMetadata;
+
+ allocCount += metadata->GetAllocationCount();
+ freeCount += metadata->GetFreeRegionsCount();
+ state.avgFreeSize += metadata->GetSumFreeSize();
+ state.avgAllocSize += metadata->GetSize();
+ }
+
+ state.avgAllocSize = (state.avgAllocSize - state.avgFreeSize) / allocCount;
+ state.avgFreeSize /= freeCount;
+}
+
bool VmaDefragmentationContext_T::MoveDataToFreeBlocks(VmaSuballocationType currentType,
VmaBlockVector& vector, size_t firstFreeBlock,
bool& texturePresent, bool& bufferPresent, bool& otherPresent)
@@ -13673,6 +13941,17 @@ bool VmaDefragmentationContext_T::MoveDataToFreeBlocks(VmaSuballocationType curr
// Ignore newly created allocations by defragmentation algorithm
if (moveData.move.srcAllocation->GetUserData() == this)
continue;
+ switch (CheckCounters(moveData.move.srcAllocation->GetSize()))
+ {
+ case CounterStatus::Ignore:
+ continue;
+ case CounterStatus::End:
+ return true;
+ default:
+ VMA_ASSERT(0);
+ case CounterStatus::Pass:
+ break;
+ }
// Move only single type of resources at once
if (!VmaIsBufferImageGranularityConflict(moveData.type, currentType))
@@ -14404,6 +14683,7 @@ VkResult VmaAllocator_T::AllocateDedicatedMemory(
VkMemoryPriorityAllocateInfoEXT priorityInfo = { VK_STRUCTURE_TYPE_MEMORY_PRIORITY_ALLOCATE_INFO_EXT };
if(m_UseExtMemoryPriority)
{
+ VMA_ASSERT(priority >= 0.f && priority <= 1.f);
priorityInfo.priority = priority;
VmaPnextChainPushFront(&allocInfo, &priorityInfo);
}
@@ -14515,9 +14795,12 @@ VkResult VmaAllocator_T::AllocateDedicatedMemoryPage(
}
}
- *pAllocation = m_AllocationObjectAllocator.Allocate(isUserDataString, isMappingAllowed);
+ *pAllocation = m_AllocationObjectAllocator.Allocate(isMappingAllowed);
(*pAllocation)->InitDedicatedAllocation(pool, memTypeIndex, hMemory, suballocType, pMappedData, size);
- (*pAllocation)->SetUserData(this, pUserData);
+ if (isUserDataString)
+ (*pAllocation)->SetName(this, (const char*)pUserData);
+ else
+ (*pAllocation)->SetUserData(this, pUserData);
m_Budget.AddAllocation(MemoryTypeIndexToHeapIndex(memTypeIndex), size);
if(VMA_DEBUG_INITIALIZE_ALLOCATIONS)
{
@@ -14629,8 +14912,8 @@ VkResult VmaAllocator_T::FindMemoryTypeIndex(
if((requiredFlags & ~currFlags) == 0)
{
// Calculate cost as number of bits from preferredFlags not present in this memory type.
- uint32_t currCost = VmaCountBitsSet(preferredFlags & ~currFlags) +
- VmaCountBitsSet(currFlags & notPreferredFlags);
+ uint32_t currCost = VMA_COUNT_BITS_SET(preferredFlags & ~currFlags) +
+ VMA_COUNT_BITS_SET(currFlags & notPreferredFlags);
// Remember memory type with lowest cost.
if(currCost < minCost)
{
@@ -14849,6 +15132,8 @@ void VmaAllocator_T::FreeMemory(
FillAllocation(allocation, VMA_ALLOCATION_FILL_PATTERN_DESTROYED);
}
+ allocation->FreeName(this);
+
switch(allocation->GetType())
{
case VmaAllocation_T::ALLOCATION_TYPE_BLOCK:
@@ -14994,6 +15279,7 @@ void VmaAllocator_T::GetAllocationInfo(VmaAllocation hAllocation, VmaAllocationI
pAllocationInfo->size = hAllocation->GetSize();
pAllocationInfo->pMappedData = hAllocation->GetMappedData();
pAllocationInfo->pUserData = hAllocation->GetUserData();
+ pAllocationInfo->pName = hAllocation->GetName();
}
VkResult VmaAllocator_T::CreatePool(const VmaPoolCreateInfo* pCreateInfo, VmaPool* pPool)
@@ -15664,89 +15950,90 @@ uint32_t VmaAllocator_T::GetGpuDefragmentationMemoryTypeBits()
#if VMA_STATS_STRING_ENABLED
void VmaAllocator_T::PrintDetailedMap(VmaJsonWriter& json)
{
- bool dedicatedAllocationsStarted = false;
- for(uint32_t memTypeIndex = 0; memTypeIndex < GetMemoryTypeCount(); ++memTypeIndex)
- {
- VmaDedicatedAllocationList& dedicatedAllocList = m_DedicatedAllocations[memTypeIndex];
- if(!dedicatedAllocList.IsEmpty())
- {
- if(dedicatedAllocationsStarted == false)
- {
- dedicatedAllocationsStarted = true;
- json.WriteString("DedicatedAllocations");
- json.BeginObject();
- }
-
- json.BeginString("Type ");
- json.ContinueString(memTypeIndex);
- json.EndString();
-
- dedicatedAllocList.BuildStatsString(json);
- }
- }
- if(dedicatedAllocationsStarted)
- {
- json.EndObject();
- }
-
+ json.WriteString("DefaultPools");
+ json.BeginObject();
{
- bool allocationsStarted = false;
- for(uint32_t memTypeIndex = 0; memTypeIndex < GetMemoryTypeCount(); ++memTypeIndex)
+ for (uint32_t memTypeIndex = 0; memTypeIndex < GetMemoryTypeCount(); ++memTypeIndex)
{
VmaBlockVector* pBlockVector = m_pBlockVectors[memTypeIndex];
- if(pBlockVector != VMA_NULL)
+ VmaDedicatedAllocationList& dedicatedAllocList = m_DedicatedAllocations[memTypeIndex];
+ if (pBlockVector != VMA_NULL)
{
- if (pBlockVector->IsEmpty() == false)
+ json.BeginString("Type ");
+ json.ContinueString(memTypeIndex);
+ json.EndString();
+ json.BeginObject();
{
- if (allocationsStarted == false)
- {
- allocationsStarted = true;
- json.WriteString("DefaultPools");
- json.BeginObject();
- }
-
- json.BeginString("Type ");
- json.ContinueString(memTypeIndex);
- json.EndString();
+ json.WriteString("PreferredBlockSize");
+ json.WriteNumber(pBlockVector->GetPreferredBlockSize());
- json.BeginObject();
+ json.WriteString("Blocks");
pBlockVector->PrintDetailedMap(json);
- json.EndObject();
+
+ json.WriteString("DedicatedAllocations");
+ dedicatedAllocList.BuildStatsString(json);
}
+ json.EndObject();
}
}
- if(allocationsStarted)
- {
- json.EndObject();
- }
}
+ json.EndObject();
- // Custom pools
+ json.WriteString("CustomPools");
+ json.BeginObject();
{
VmaMutexLockRead lock(m_PoolsMutex, m_UseMutex);
- if(!m_Pools.IsEmpty())
+ if (!m_Pools.IsEmpty())
{
- json.WriteString("Pools");
- json.BeginObject();
- for(VmaPool pool = m_Pools.Front(); pool != VMA_NULL; pool = m_Pools.GetNext(pool))
+ for (uint32_t memTypeIndex = 0; memTypeIndex < GetMemoryTypeCount(); ++memTypeIndex)
{
- json.BeginString();
- json.ContinueString(pool->GetId());
- json.EndString();
+ bool displayType = true;
+ size_t index = 0;
+ for (VmaPool pool = m_Pools.Front(); pool != VMA_NULL; pool = m_Pools.GetNext(pool))
+ {
+ VmaBlockVector& blockVector = pool->m_BlockVector;
+ if (blockVector.GetMemoryTypeIndex() == memTypeIndex)
+ {
+ if (displayType)
+ {
+ json.BeginString("Type ");
+ json.ContinueString(memTypeIndex);
+ json.EndString();
+ json.BeginArray();
+ displayType = false;
+ }
- json.BeginObject();
- pool->m_BlockVector.PrintDetailedMap(json);
+ json.BeginObject();
+ {
+ json.WriteString("Name");
+ json.BeginString();
+ json.ContinueString_Size(index++);
+ if (pool->GetName())
+ {
+ json.WriteString(" - ");
+ json.WriteString(pool->GetName());
+ }
+ json.EndString();
- if (!pool->m_DedicatedAllocations.IsEmpty())
- {
- json.WriteString("DedicatedAllocations");
- pool->m_DedicatedAllocations.BuildStatsString(json);
+ json.WriteString("PreferredBlockSize");
+ json.WriteNumber(blockVector.GetPreferredBlockSize());
+
+ json.WriteString("Blocks");
+ blockVector.PrintDetailedMap(json);
+
+ json.WriteString("DedicatedAllocations");
+ pool->m_DedicatedAllocations.BuildStatsString(json);
+ }
+ json.EndObject();
+ }
}
- json.EndObject();
+
+ if (!displayType)
+ json.EndArray();
}
- json.EndObject();
}
}
+ json.EndObject();
}
#endif // VMA_STATS_STRING_ENABLED
#endif // _VMA_ALLOCATOR_T_FUNCTIONS
@@ -15857,127 +16144,176 @@ VMA_CALL_PRE void VMA_CALL_POST vmaBuildStatsString(
VmaStringBuilder sb(allocator->GetAllocationCallbacks());
{
- VmaJsonWriter json(allocator->GetAllocationCallbacks(), sb);
- json.BeginObject();
-
VmaBudget budgets[VK_MAX_MEMORY_HEAPS];
allocator->GetHeapBudgets(budgets, 0, allocator->GetMemoryHeapCount());
VmaTotalStatistics stats;
allocator->CalculateStatistics(&stats);
- json.WriteString("Total");
- VmaPrintDetailedStatistics(json, stats.total);
-
- for(uint32_t heapIndex = 0; heapIndex < allocator->GetMemoryHeapCount(); ++heapIndex)
+ VmaJsonWriter json(allocator->GetAllocationCallbacks(), sb);
+ json.BeginObject();
{
- json.BeginString("Heap ");
- json.ContinueString(heapIndex);
- json.EndString();
+ json.WriteString("General");
json.BeginObject();
+ {
+ const VkPhysicalDeviceProperties& deviceProperties = allocator->m_PhysicalDeviceProperties;
+ const VkPhysicalDeviceMemoryProperties& memoryProperties = allocator->m_MemProps;
- json.WriteString("Size");
- json.WriteNumber(allocator->m_MemProps.memoryHeaps[heapIndex].size);
+ json.WriteString("API");
+ json.WriteString("Vulkan");
- json.WriteString("Flags");
- json.BeginArray(true);
- if((allocator->m_MemProps.memoryHeaps[heapIndex].flags & VK_MEMORY_HEAP_DEVICE_LOCAL_BIT) != 0)
- {
- json.WriteString("DEVICE_LOCAL");
- }
- json.EndArray();
+ json.WriteString("apiVersion");
+ json.BeginString();
+ json.ContinueString(VK_API_VERSION_MAJOR(deviceProperties.apiVersion));
+ json.ContinueString(".");
+ json.ContinueString(VK_API_VERSION_MINOR(deviceProperties.apiVersion));
+ json.ContinueString(".");
+ json.ContinueString(VK_API_VERSION_PATCH(deviceProperties.apiVersion));
+ json.EndString();
- json.WriteString("Budget");
- json.BeginObject();
- {
- json.WriteString("BlockBytes");
- json.WriteNumber(budgets[heapIndex].statistics.blockBytes);
- json.WriteString("AllocationBytes");
- json.WriteNumber(budgets[heapIndex].statistics.allocationBytes);
- json.WriteString("BlockCount");
- json.WriteNumber(budgets[heapIndex].statistics.blockCount);
- json.WriteString("AllocationCount");
- json.WriteNumber(budgets[heapIndex].statistics.allocationCount);
- json.WriteString("Usage");
- json.WriteNumber(budgets[heapIndex].usage);
- json.WriteString("Budget");
- json.WriteNumber(budgets[heapIndex].budget);
+ json.WriteString("GPU");
+ json.WriteString(deviceProperties.deviceName);
+ json.WriteString("deviceType");
+ json.WriteNumber(static_cast<uint32_t>(deviceProperties.deviceType));
+
+ json.WriteString("maxMemoryAllocationCount");
+ json.WriteNumber(deviceProperties.limits.maxMemoryAllocationCount);
+ json.WriteString("bufferImageGranularity");
+ json.WriteNumber(deviceProperties.limits.bufferImageGranularity);
+ json.WriteString("nonCoherentAtomSize");
+ json.WriteNumber(deviceProperties.limits.nonCoherentAtomSize);
+
+ json.WriteString("memoryHeapCount");
+ json.WriteNumber(memoryProperties.memoryHeapCount);
+ json.WriteString("memoryTypeCount");
+ json.WriteNumber(memoryProperties.memoryTypeCount);
}
json.EndObject();
-
- if(stats.memoryHeap[heapIndex].statistics.blockCount > 0)
- {
- json.WriteString("Stats");
- VmaPrintDetailedStatistics(json, stats.memoryHeap[heapIndex]);
- }
-
- for(uint32_t typeIndex = 0; typeIndex < allocator->GetMemoryTypeCount(); ++typeIndex)
+ }
+ {
+ json.WriteString("Total");
+ VmaPrintDetailedStatistics(json, stats.total);
+ }
+ {
+ json.WriteString("MemoryInfo");
+ json.BeginObject();
{
- if(allocator->MemoryTypeIndexToHeapIndex(typeIndex) == heapIndex)
+ for (uint32_t heapIndex = 0; heapIndex < allocator->GetMemoryHeapCount(); ++heapIndex)
{
- json.BeginString("Type ");
- json.ContinueString(typeIndex);
+ json.BeginString("Heap ");
+ json.ContinueString(heapIndex);
json.EndString();
-
json.BeginObject();
-
- json.WriteString("Flags");
- json.BeginArray(true);
- VkMemoryPropertyFlags flags = allocator->m_MemProps.memoryTypes[typeIndex].propertyFlags;
- if((flags & VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT) != 0)
- {
- json.WriteString("DEVICE_LOCAL");
- }
- if((flags & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT) != 0)
- {
- json.WriteString("HOST_VISIBLE");
- }
- if((flags & VK_MEMORY_PROPERTY_HOST_COHERENT_BIT) != 0)
- {
- json.WriteString("HOST_COHERENT");
- }
- if((flags & VK_MEMORY_PROPERTY_HOST_CACHED_BIT) != 0)
- {
- json.WriteString("HOST_CACHED");
- }
- if((flags & VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT) != 0)
- {
- json.WriteString("LAZILY_ALLOCATED");
- }
-#if VMA_VULKAN_VERSION >= 1001000
- if((flags & VK_MEMORY_PROPERTY_PROTECTED_BIT) != 0)
- {
- json.WriteString("PROTECTED");
- }
-#endif // #if VMA_VULKAN_VERSION >= 1001000
-#if VK_AMD_device_coherent_memory
- if((flags & VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD_COPY) != 0)
- {
- json.WriteString("DEVICE_COHERENT");
- }
- if((flags & VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD_COPY) != 0)
{
- json.WriteString("DEVICE_UNCACHED");
- }
-#endif // #if VK_AMD_device_coherent_memory
- json.EndArray();
+ const VkMemoryHeap& heapInfo = allocator->m_MemProps.memoryHeaps[heapIndex];
+ json.WriteString("Flags");
+ json.BeginArray(true);
+ {
+ if (heapInfo.flags & VK_MEMORY_HEAP_DEVICE_LOCAL_BIT)
+ json.WriteString("DEVICE_LOCAL");
+ #if VMA_VULKAN_VERSION >= 1001000
+ if (heapInfo.flags & VK_MEMORY_HEAP_MULTI_INSTANCE_BIT)
+ json.WriteString("MULTI_INSTANCE");
+ #endif
+
+ VkMemoryHeapFlags flags = heapInfo.flags &
+ ~(VK_MEMORY_HEAP_DEVICE_LOCAL_BIT
+ #if VMA_VULKAN_VERSION >= 1001000
+ | VK_MEMORY_HEAP_MULTI_INSTANCE_BIT
+ #endif
+ );
+ if (flags != 0)
+ json.WriteNumber(flags);
+ }
+ json.EndArray();
+
+ json.WriteString("Size");
+ json.WriteNumber(heapInfo.size);
+
+ json.WriteString("Budget");
+ json.BeginObject();
+ {
+ json.WriteString("BudgetBytes");
+ json.WriteNumber(budgets[heapIndex].budget);
+ json.WriteString("UsageBytes");
+ json.WriteNumber(budgets[heapIndex].usage);
+ }
+ json.EndObject();
- if(stats.memoryType[typeIndex].statistics.blockCount > 0)
- {
json.WriteString("Stats");
- VmaPrintDetailedStatistics(json, stats.memoryType[typeIndex]);
- }
+ VmaPrintDetailedStatistics(json, stats.memoryHeap[heapIndex]);
+ json.WriteString("MemoryPools");
+ json.BeginObject();
+ {
+ for (uint32_t typeIndex = 0; typeIndex < allocator->GetMemoryTypeCount(); ++typeIndex)
+ {
+ if (allocator->MemoryTypeIndexToHeapIndex(typeIndex) == heapIndex)
+ {
+ json.BeginString("Type ");
+ json.ContinueString(typeIndex);
+ json.EndString();
+ json.BeginObject();
+ {
+ json.WriteString("Flags");
+ json.BeginArray(true);
+ {
+ VkMemoryPropertyFlags flags = allocator->m_MemProps.memoryTypes[typeIndex].propertyFlags;
+ if (flags & VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT)
+ json.WriteString("DEVICE_LOCAL");
+ if (flags & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT)
+ json.WriteString("HOST_VISIBLE");
+ if (flags & VK_MEMORY_PROPERTY_HOST_COHERENT_BIT)
+ json.WriteString("HOST_COHERENT");
+ if (flags & VK_MEMORY_PROPERTY_HOST_CACHED_BIT)
+ json.WriteString("HOST_CACHED");
+ if (flags & VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT)
+ json.WriteString("LAZILY_ALLOCATED");
+ #if VMA_VULKAN_VERSION >= 1001000
+ if (flags & VK_MEMORY_PROPERTY_PROTECTED_BIT)
+ json.WriteString("PROTECTED");
+ #endif
+ #if VK_AMD_device_coherent_memory
+ if (flags & VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD_COPY)
+ json.WriteString("DEVICE_COHERENT_AMD");
+ if (flags & VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD_COPY)
+ json.WriteString("DEVICE_UNCACHED_AMD");
+ #endif
+
+ flags &= ~(VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT
+ #if VMA_VULKAN_VERSION >= 1001000
+ | VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT
+ #endif
+ #if VK_AMD_device_coherent_memory
+ | VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD_COPY
+ | VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD_COPY
+ #endif
+ | VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT
+ | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT
+ | VK_MEMORY_PROPERTY_HOST_CACHED_BIT);
+ if (flags != 0)
+ json.WriteNumber(flags);
+ }
+ json.EndArray();
+
+ json.WriteString("Stats");
+ VmaPrintDetailedStatistics(json, stats.memoryType[typeIndex]);
+ }
+ json.EndObject();
+ }
+ }
+
+ }
+ json.EndObject();
+ }
json.EndObject();
}
}
-
json.EndObject();
}
- if(detailedMap == VK_TRUE)
- {
+
+ if (detailedMap == VK_TRUE)
allocator->PrintDetailedMap(json);
- }
json.EndObject();
}
@@ -16428,6 +16764,14 @@ VMA_CALL_PRE void VMA_CALL_POST vmaSetAllocationUserData(
allocation->SetUserData(allocator, pUserData);
}
+VMA_CALL_PRE void VMA_CALL_POST vmaSetAllocationName(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const char* VMA_NULLABLE pName)
+{
+ allocation->SetName(allocator, pName);
+}
+
VMA_CALL_PRE void VMA_CALL_POST vmaGetAllocationMemoryProperties(
VmaAllocator VMA_NOT_NULL allocator,
VmaAllocation VMA_NOT_NULL allocation,
@@ -16567,13 +16911,20 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaBeginDefragmentation(
VMA_DEBUG_LOG("vmaBeginDefragmentation");
+ if (pInfo->pool != VMA_NULL)
+ {
+ // Check if run on supported algorithms
+ if (pInfo->pool->m_BlockVector.GetAlgorithm() & VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT)
+ return VK_ERROR_FEATURE_NOT_PRESENT;
+ }
+
VMA_DEBUG_GLOBAL_MUTEX_LOCK
*pContext = vma_new(allocator, VmaDefragmentationContext_T)(allocator, *pInfo);
return VK_SUCCESS;
}
-VMA_CALL_PRE VkResult VMA_CALL_POST vmaEndDefragmentation(
+VMA_CALL_PRE void VMA_CALL_POST vmaEndDefragmentation(
VmaAllocator allocator,
VmaDefragmentationContext context,
VmaDefragmentationStats* pStats)
@@ -16587,7 +16938,6 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaEndDefragmentation(
if (pStats)
context->GetStats(*pStats);
vma_delete(allocator, context);
- return VK_SUCCESS;
}
VMA_CALL_PRE VkResult VMA_CALL_POST vmaBeginDefragmentationPass(
@@ -16864,6 +17214,50 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateBufferWithAlignment(
return res;
}
+VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateAliasingBuffer(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const VkBufferCreateInfo* VMA_NOT_NULL pBufferCreateInfo,
+ VkBuffer VMA_NULLABLE_NON_DISPATCHABLE* VMA_NOT_NULL pBuffer)
+{
+ VMA_ASSERT(allocator && pBufferCreateInfo && pBuffer && allocation);
+
+ VMA_DEBUG_LOG("vmaCreateAliasingBuffer");
+
+ *pBuffer = VK_NULL_HANDLE;
+
+ if (pBufferCreateInfo->size == 0)
+ {
+ return VK_ERROR_INITIALIZATION_FAILED;
+ }
+ if ((pBufferCreateInfo->usage & VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT_COPY) != 0 &&
+ !allocator->m_UseKhrBufferDeviceAddress)
+ {
+ VMA_ASSERT(0 && "Creating a buffer with VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT is not valid if VMA_ALLOCATOR_CREATE_BUFFER_DEVICE_ADDRESS_BIT was not used.");
+ return VK_ERROR_INITIALIZATION_FAILED;
+ }
+
+ VMA_DEBUG_GLOBAL_MUTEX_LOCK
+
+ // 1. Create VkBuffer.
+ VkResult res = (*allocator->GetVulkanFunctions().vkCreateBuffer)(
+ allocator->m_hDevice,
+ pBufferCreateInfo,
+ allocator->GetAllocationCallbacks(),
+ pBuffer);
+ if (res >= 0)
+ {
+ // 2. Bind buffer with memory.
+ res = allocator->BindBufferMemory(allocation, 0, *pBuffer, VMA_NULL);
+ if (res >= 0)
+ {
+ return VK_SUCCESS;
+ }
+ (*allocator->GetVulkanFunctions().vkDestroyBuffer)(allocator->m_hDevice, *pBuffer, allocator->GetAllocationCallbacks());
+ }
+ return res;
+}
+
VMA_CALL_PRE void VMA_CALL_POST vmaDestroyBuffer(
VmaAllocator allocator,
VkBuffer buffer,
@@ -16985,10 +17379,52 @@ VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateImage(
return res;
}
+VMA_CALL_PRE VkResult VMA_CALL_POST vmaCreateAliasingImage(
+ VmaAllocator VMA_NOT_NULL allocator,
+ VmaAllocation VMA_NOT_NULL allocation,
+ const VkImageCreateInfo* VMA_NOT_NULL pImageCreateInfo,
+ VkImage VMA_NULLABLE_NON_DISPATCHABLE* VMA_NOT_NULL pImage)
+{
+ VMA_ASSERT(allocator && pImageCreateInfo && pImage && allocation);
+
+ *pImage = VK_NULL_HANDLE;
+
+ VMA_DEBUG_LOG("vmaCreateImage");
+
+ if (pImageCreateInfo->extent.width == 0 ||
+ pImageCreateInfo->extent.height == 0 ||
+ pImageCreateInfo->extent.depth == 0 ||
+ pImageCreateInfo->mipLevels == 0 ||
+ pImageCreateInfo->arrayLayers == 0)
+ {
+ return VK_ERROR_INITIALIZATION_FAILED;
+ }
+
+ VMA_DEBUG_GLOBAL_MUTEX_LOCK
+
+ // 1. Create VkImage.
+ VkResult res = (*allocator->GetVulkanFunctions().vkCreateImage)(
+ allocator->m_hDevice,
+ pImageCreateInfo,
+ allocator->GetAllocationCallbacks(),
+ pImage);
+ if (res >= 0)
+ {
+ // 2. Bind image with memory.
+ res = allocator->BindImageMemory(allocation, 0, *pImage, VMA_NULL);
+ if (res >= 0)
+ {
+ return VK_SUCCESS;
+ }
+ (*allocator->GetVulkanFunctions().vkDestroyImage)(allocator->m_hDevice, *pImage, allocator->GetAllocationCallbacks());
+ }
+ return res;
+}
+
VMA_CALL_PRE void VMA_CALL_POST vmaDestroyImage(
- VmaAllocator allocator,
- VkImage image,
- VmaAllocation allocation)
+ VmaAllocator VMA_NOT_NULL allocator,
+ VkImage VMA_NULLABLE_NON_DISPATCHABLE image,
+ VmaAllocation VMA_NULLABLE allocation)
{
VMA_ASSERT(allocator);
@@ -17176,9 +17612,10 @@ before including these headers (like `WIN32_LEAN_AND_MEAN` or
`WINVER` for Windows, `VK_USE_PLATFORM_WIN32_KHR` for Vulkan), you must define
them before every `#include` of this library.
-\note This library is written in C++, but has C-compatible interface.
+This library is written in C++, but has C-compatible interface.
Thus you can include and use vk_mem_alloc.h in C or C++ code, but full
implementation with `VMA_IMPLEMENTATION` macro must be compiled as C++, NOT as C.
+Some features of C++14 used. STL containers, RTTI, or C++ exceptions are not used.
\section quick_start_initialization Initialization
@@ -17775,14 +18212,28 @@ To use custom memory pools:
Example:
\code
+// Find memoryTypeIndex for the pool.
+VkBufferCreateInfo sampleBufCreateInfo = { VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO };
+sampleBufCreateInfo.size = 0x10000; // Doesn't matter.
+sampleBufCreateInfo.usage = VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT | VK_BUFFER_USAGE_TRANSFER_DST_BIT;
+
+VmaAllocationCreateInfo sampleAllocCreateInfo = {};
+sampleAllocCreateInfo.usage = VMA_MEMORY_USAGE_AUTO;
+
+uint32_t memTypeIndex;
+VkResult res = vmaFindMemoryTypeIndexForBufferInfo(allocator,
+ &sampleBufCreateInfo, &sampleAllocCreateInfo, &memTypeIndex);
+// Check res...
+
// Create a pool that can have at most 2 blocks, 128 MiB each.
VmaPoolCreateInfo poolCreateInfo = {};
-poolCreateInfo.memoryTypeIndex = ...
+poolCreateInfo.memoryTypeIndex = memTypeIndex;
poolCreateInfo.blockSize = 128ull * 1024 * 1024;
poolCreateInfo.maxBlockCount = 2;
VmaPool pool;
-vmaCreatePool(allocator, &poolCreateInfo, &pool);
+res = vmaCreatePool(allocator, &poolCreateInfo, &pool);
+// Check res...
// Allocate a buffer out of it.
VkBufferCreateInfo bufCreateInfo = { VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO };
@@ -17794,8 +18245,8 @@ allocCreateInfo.pool = pool;
VkBuffer buf;
VmaAllocation alloc;
-VmaAllocationInfo allocInfo;
-vmaCreateBuffer(allocator, &bufCreateInfo, &allocCreateInfo, &buf, &alloc, &allocInfo);
+res = vmaCreateBuffer(allocator, &bufCreateInfo, &allocCreateInfo, &buf, &alloc, nullptr);
+// Check res...
\endcode
You have to free all allocations made from this pool before destroying it.
@@ -17933,6 +18384,8 @@ you can achieve behavior of a ring buffer / queue.
Ring buffer is available only in pools with one memory block -
VmaPoolCreateInfo::maxBlockCount must be 1. Otherwise behavior is undefined.
+\note \ref defragmentation is not supported in custom pools created with #VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT.
+
\page defragmentation Defragmentation
@@ -17965,41 +18418,105 @@ for(;;)
res = vmaBeginDefragmentationPass(allocator, defragCtx, &pass);
if(res == VK_SUCCESS)
break;
- else if(res == VK_INCOMPLETE)
+ else if(res != VK_INCOMPLETE)
+ // Handle error...
+
+ for(uint32_t i = 0; i < pass.moveCount; ++i)
{
- for(uint32_t i = 0; i < pass.moveCount; ++i)
- {
- //- Inspect pass.pMoves[i].srcAllocation, identify what buffer or image it represents.
- //- Recreate this buffer or image at pass.pMoves[i].dstMemory, pass.pMoves[i].dstOffset.
- //- Issue a vkCmdCopyBuffer/vkCmdCopyImage to copy its content to the new place.
- }
- //- Make sure the copy commands finished executing.
- //- Update appropriate descriptors to point to the new places.
- res = vmaEndDefragmentationPass(allocator, defragCtx, &pass);
- if(res == VK_SUCCESS)
- break;
- else if(res != VK_INCOMPLETE)
- // Handle error...
+ // Inspect pass.pMoves[i].srcAllocation, identify what buffer/image it represents.
+ VmaAllocationInfo allocInfo;
+ vmaGetAllocationInfo(allocator, pMoves[i].srcAllocation, &allocInfo);
+ MyEngineResourceData* resData = (MyEngineResourceData*)allocInfo.pUserData;
+
+ // Recreate and bind this buffer/image at: pass.pMoves[i].dstMemory, pass.pMoves[i].dstOffset.
+ VkImageCreateInfo imgCreateInfo = ...
+ VkImage newImg;
+ res = vkCreateImage(device, &imgCreateInfo, nullptr, &newImg);
+ // Check res...
+ res = vmaBindImageMemory(allocator, pMoves[i].dstTmpAllocation, newImg);
+ // Check res...
+
+ // Issue a vkCmdCopyBuffer/vkCmdCopyImage to copy its content to the new place.
+ vkCmdCopyImage(cmdBuf, resData->img, ..., newImg, ...);
}
- else
+
+ // Make sure the copy commands finished executing.
+ vkWaitForFences(...);
+
+ // Destroy old buffers/images bound with pass.pMoves[i].srcAllocation.
+ for(uint32_t i = 0; i < pass.moveCount; ++i)
+ {
+ // ...
+ vkDestroyImage(device, resData->img, nullptr);
+ }
+
+ // Update appropriate descriptors to point to the new places...
+
+ res = vmaEndDefragmentationPass(allocator, defragCtx, &pass);
+ if(res == VK_SUCCESS)
+ break;
+ else if(res != VK_INCOMPLETE)
// Handle error...
}
vmaEndDefragmentation(allocator, defragCtx, nullptr);
\endcode
-You can defragment a specific custom pool by setting VmaDefragmentationInfo::pool
-(like in the example above) or all the default pools by setting this member to null.
+Although functions like vmaCreateBuffer(), vmaCreateImage(), vmaDestroyBuffer(), vmaDestroyImage()
+create/destroy an allocation and a buffer/image at once, these are just a shortcut for
+creating the resource, allocating memory, and binding them together.
+Defragmentation works on memory allocations only. You must handle the rest manually.
+Defragmentation is an iterative process that should repreat "passes" as long as related functions
+return `VK_INCOMPLETE` not `VK_SUCCESS`.
+In each pass:
+
+1. vmaBeginDefragmentationPass() function call:
+ - Calculates and returns the list of allocations to be moved in this pass.
+ Note this can be a time-consuming process.
+ - Reserves destination memory for them by creating temporary destination allocations
+ that you can query for their `VkDeviceMemory` + offset using vmaGetAllocationInfo().
+2. Inside the pass, **you should**:
+ - Inspect the returned list of allocations to be moved.
+ - Create new buffers/images and bind them at the returned destination temporary allocations.
+ - Copy data from source to destination resources if necessary.
+ - Destroy the source buffers/images, but NOT their allocations.
+3. vmaEndDefragmentationPass() function call:
+ - Frees the source memory reserved for the allocations that are moved.
+ - Modifies source #VmaAllocation objects that are moved to point to the destination reserved memory.
+ - Frees `VkDeviceMemory` blocks that became empty.
Unlike in previous iterations of the defragmentation API, there is no list of "movable" allocations passed as a parameter.
Defragmentation algorithm tries to move all suitable allocations.
You can, however, refuse to move some of them inside a defragmentation pass, by setting
`pass.pMoves[i].operation` to #VMA_DEFRAGMENTATION_MOVE_OPERATION_IGNORE.
-However, this is not recommended and may result in suboptimal packing of the allocations after defragmentation.
+This is not recommended and may result in suboptimal packing of the allocations after defragmentation.
If you cannot ensure any allocation can be moved, it is better to keep movable allocations separate in a custom pool.
-You can also decide to destroy an allocation instead of moving it.
-You should then set `pass.pMoves[i].operation` to #VMA_DEFRAGMENTATION_MOVE_OPERATION_DESTROY.
+Inside a pass, for each allocation that should be moved:
+
+- You should copy its data from the source to the destination place by calling e.g. `vkCmdCopyBuffer()`, `vkCmdCopyImage()`.
+ - You need to make sure these commands finished executing before destroying the source buffers/images and before calling vmaEndDefragmentationPass().
+- If a resource doesn't contain any meaningful data, e.g. it is a transient color attachment image to be cleared,
+ filled, and used temporarily in each rendering frame, you can just recreate this image
+ without copying its data.
+- If the resource is in `HOST_VISIBLE` and `HOST_CACHED` memory, you can copy its data on the CPU
+ using `memcpy()`.
+- If you cannot move the allocation, you can set `pass.pMoves[i].operation` to #VMA_DEFRAGMENTATION_MOVE_OPERATION_IGNORE.
+ This will cancel the move.
+ - vmaEndDefragmentationPass() will then free the destination memory
+ not the source memory of the allocation, leaving it unchanged.
+- If you decide the allocation is unimportant and can be destroyed instead of moved (e.g. it wasn't used for long time),
+ you can set `pass.pMoves[i].operation` to #VMA_DEFRAGMENTATION_MOVE_OPERATION_DESTROY.
+ - vmaEndDefragmentationPass() will then free both source and destination memory, and will destroy the source #VmaAllocation object.
+
+You can defragment a specific custom pool by setting VmaDefragmentationInfo::pool
+(like in the example above) or all the default pools by setting this member to null.
+
+Defragmentation is always performed in each pool separately.
+Allocations are never moved between different Vulkan memory types.
+The size of the destination memory reserved for a moved allocation is the same as the original one.
+Alignment of an allocation as it was determined using `vkGetBufferMemoryRequirements()` etc. is also respected after defragmentation.
+Buffers/images should be recreated with the same `VkBufferCreateInfo` / `VkImageCreateInfo` parameters as the original ones.
You can perform the defragmentation incrementally to limit the number of allocations and bytes to be moved
in each pass, e.g. to call it in sync with render frames and not to experience too big hitches.
@@ -18009,6 +18526,13 @@ It is also safe to perform the defragmentation asynchronously to render frames a
usage, possibly from multiple threads, with the exception that allocations
returned in VmaDefragmentationPassMoveInfo::pMoves shouldn't be destroyed until the defragmentation pass is ended.
+<b>Mapping</b> is preserved on allocations that are moved during defragmentation.
+Whether through #VMA_ALLOCATION_CREATE_MAPPED_BIT or vmaMapMemory(), the allocations
+are mapped at their new place. Of course, pointer to the mapped data changes, so it needs to be queried
+using VmaAllocationInfo::pMappedData.
+
+\note Defragmentation is not supported in custom pools created with #VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT.
+
\page statistics Statistics
@@ -18081,6 +18605,8 @@ To do that, fill VmaAllocationCreateInfo::pUserData field when creating
an allocation. It is an opaque `void*` pointer. You can use it e.g. as a pointer,
some handle, index, key, ordinal number or any other value that would associate
the allocation with your custom metadata.
+It it useful to identify appropriate data structures in your engine given #VmaAllocation,
+e.g. when doing \ref defragmentation.
\code
VkBufferCreateInfo bufCreateInfo = ...
@@ -18111,44 +18637,21 @@ vmaBuildStatsString() in hexadecimal form.
\section allocation_names Allocation names
-There is alternative mode available where `pUserData` pointer is used to point to
-a null-terminated string, giving a name to the allocation. To use this mode,
-set #VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT flag in VmaAllocationCreateInfo::flags.
-Then `pUserData` passed as VmaAllocationCreateInfo::pUserData or argument to
-vmaSetAllocationUserData() must be either null or a pointer to a null-terminated string.
+An allocation can also carry a null-terminated string, giving a name to the allocation.
+To set it, call vmaSetAllocationName().
The library creates internal copy of the string, so the pointer you pass doesn't need
to be valid for whole lifetime of the allocation. You can free it after the call.
\code
-VkImageCreateInfo imageInfo = ...
-
std::string imageName = "Texture: ";
imageName += fileName;
-
-VmaAllocationCreateInfo allocCreateInfo = {};
-allocCreateInfo.usage = VMA_MEMORY_USAGE_AUTO;
-allocCreateInfo.flags = VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT;
-allocCreateInfo.pUserData = imageName.c_str();
-
-VkImage image;
-VmaAllocation allocation;
-vmaCreateImage(allocator, &imageInfo, &allocCreateInfo, &image, &allocation, nullptr);
+vmaSetAllocationName(allocator, allocation, imageName.c_str());
\endcode
-The value of `pUserData` pointer of the allocation will be different than the one
-you passed when setting allocation's name - pointing to a buffer managed
-internally that holds copy of the string.
-
-\code
-VmaAllocationInfo allocInfo;
-vmaGetAllocationInfo(allocator, allocation, &allocInfo);
-const char* imageName = (const char*)allocInfo.pUserData;
-printf("Image name: %s\n", imageName);
-\endcode
+The string can be later retrieved by inspecting VmaAllocationInfo::pName.
+It is also printed in JSON report created by vmaBuildStatsString().
-That string is also printed in JSON report created by vmaBuildStatsString().
-
-\note Passing string name to VMA allocation doesn't automatically set it to the Vulkan buffer or image created with it.
+\note Setting string name to VMA allocation doesn't automatically set it to the Vulkan buffer or image created with it.
You must do it manually using an extension like VK_EXT_debug_utils, which is independent of this library.
@@ -18491,6 +18994,7 @@ imgCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
VmaAllocationCreateInfo allocCreateInfo = {};
allocCreateInfo.usage = VMA_MEMORY_USAGE_AUTO;
allocCreateInfo.flags = VMA_ALLOCATION_CREATE_DEDICATED_MEMORY_BIT;
+allocCreateInfo.priority = 1.0f;
VkImage img;
VmaAllocation alloc;
@@ -18502,7 +19006,8 @@ Consider creating them as dedicated allocations using #VMA_ALLOCATION_CREATE_DED
especially if they are large or if you plan to destroy and recreate them with different sizes
e.g. when display resolution changes.
Prefer to create such resources first and all other GPU resources (like textures and vertex buffers) later.
-
+When VK_EXT_memory_priority extension is enabled, it is also worth setting high priority to such allocation
+to decrease chances to be evicted to system memory by the operating system.
\section usage_patterns_staging_copy_upload Staging copy for upload
@@ -18653,6 +19158,7 @@ else
// [Executed in runtime]:
memcpy(stagingAllocInfo.pMappedData, myData, myDataSize);
+ //vkCmdPipelineBarrier: VK_ACCESS_HOST_WRITE_BIT --> VK_ACCESS_TRANSFER_READ_BIT
VkBufferCopy bufCopy = {
0, // srcOffset
0, // dstOffset,
@@ -18815,6 +19321,86 @@ To learn more about this extension, see:
+\page vk_ext_memory_priority VK_EXT_memory_priority
+
+VK_EXT_memory_priority is a device extension that allows to pass additional "priority"
+value to Vulkan memory allocations that the implementation may use prefer certain
+buffers and images that are critical for performance to stay in device-local memory
+in cases when the memory is over-subscribed, while some others may be moved to the system memory.
+
+VMA offers convenient usage of this extension.
+If you enable it, you can pass "priority" parameter when creating allocations or custom pools
+and the library automatically passes the value to Vulkan using this extension.
+
+If you want to use this extension in connection with VMA, follow these steps:
+
+\section vk_ext_memory_priority_initialization Initialization
+
+1) Call `vkEnumerateDeviceExtensionProperties` for the physical device.
+Check if the extension is supported - if returned array of `VkExtensionProperties` contains "VK_EXT_memory_priority".
+
+2) Call `vkGetPhysicalDeviceFeatures2` for the physical device instead of old `vkGetPhysicalDeviceFeatures`.
+Attach additional structure `VkPhysicalDeviceMemoryPriorityFeaturesEXT` to `VkPhysicalDeviceFeatures2::pNext` to be returned.
+Check if the device feature is really supported - check if `VkPhysicalDeviceMemoryPriorityFeaturesEXT::memoryPriority` is true.
+
+3) While creating device with `vkCreateDevice`, enable this extension - add "VK_EXT_memory_priority"
+to the list passed as `VkDeviceCreateInfo::ppEnabledExtensionNames`.
+
+4) While creating the device, also don't set `VkDeviceCreateInfo::pEnabledFeatures`.
+Fill in `VkPhysicalDeviceFeatures2` structure instead and pass it as `VkDeviceCreateInfo::pNext`.
+Enable this device feature - attach additional structure `VkPhysicalDeviceMemoryPriorityFeaturesEXT` to
+`VkPhysicalDeviceFeatures2::pNext` chain and set its member `memoryPriority` to `VK_TRUE`.
+
+5) While creating #VmaAllocator with vmaCreateAllocator() inform VMA that you
+have enabled this extension and feature - add #VMA_ALLOCATOR_CREATE_EXT_MEMORY_PRIORITY_BIT
+to VmaAllocatorCreateInfo::flags.
+
+\section vk_ext_memory_priority_usage Usage
+
+When using this extension, you should initialize following member:
+
+- VmaAllocationCreateInfo::priority when creating a dedicated allocation with #VMA_ALLOCATION_CREATE_DEDICATED_MEMORY_BIT.
+- VmaPoolCreateInfo::priority when creating a custom pool.
+
+It should be a floating-point value between `0.0f` and `1.0f`, where recommended default is `0.5f`.
+Memory allocated with higher value can be treated by the Vulkan implementation as higher priority
+and so it can have lower chances of being pushed out to system memory, experiencing degraded performance.
+
+It might be a good idea to create performance-critical resources like color-attachment or depth-stencil images
+as dedicated and set high priority to them. For example:
+
+\code
+VkImageCreateInfo imgCreateInfo = { VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO };
+imgCreateInfo.imageType = VK_IMAGE_TYPE_2D;
+imgCreateInfo.extent.width = 3840;
+imgCreateInfo.extent.height = 2160;
+imgCreateInfo.extent.depth = 1;
+imgCreateInfo.mipLevels = 1;
+imgCreateInfo.arrayLayers = 1;
+imgCreateInfo.format = VK_FORMAT_R8G8B8A8_UNORM;
+imgCreateInfo.tiling = VK_IMAGE_TILING_OPTIMAL;
+imgCreateInfo.initialLayout = VK_IMAGE_LAYOUT_UNDEFINED;
+imgCreateInfo.usage = VK_IMAGE_USAGE_SAMPLED_BIT | VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT;
+imgCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
+
+VmaAllocationCreateInfo allocCreateInfo = {};
+allocCreateInfo.usage = VMA_MEMORY_USAGE_AUTO;
+allocCreateInfo.flags = VMA_ALLOCATION_CREATE_DEDICATED_MEMORY_BIT;
+allocCreateInfo.priority = 1.0f;
+
+VkImage img;
+VmaAllocation alloc;
+vmaCreateImage(allocator, &imgCreateInfo, &allocCreateInfo, &img, &alloc, nullptr);
+\endcode
+
+`priority` member is ignored in the following situations:
+
+- Allocations created in custom pools: They inherit the priority, along with all other allocation parameters
+ from the parametrs passed in #VmaPoolCreateInfo when the pool was created.
+- Allocations created in default pools: They inherit the priority from the parameters
+ VMA used when creating default pools, which means `priority == 0.5f`.
+
+
\page vk_amd_device_coherent_memory VK_AMD_device_coherent_memory
VK_AMD_device_coherent_memory is a device extension that enables access to
@@ -18939,6 +19525,28 @@ accompanying this library.
functions.
- #VmaVirtualBlock is not safe to be used from multiple threads simultaneously.
+\section general_considerations_versioning_and_compatibility Versioning and compatibility
+
+The library uses [**Semantic Versioning**](https://semver.org/),
+which means version numbers follow convention: Major.Minor.Patch (e.g. 2.3.0), where:
+
+- Incremented Patch version means a release is backward- and forward-compatible,
+ introducing only some internal improvements, bug fixes, optimizations etc.
+ or changes that are out of scope of the official API described in this documentation.
+- Incremented Minor version means a release is backward-compatible,
+ so existing code that uses the library should continue to work, while some new
+ symbols could have been added: new structures, functions, new values in existing
+ enums and bit flags, new structure members, but not new function parameters.
+- Incrementing Major version means a release could break some backward compatibility.
+
+All changes between official releases are documented in file "CHANGELOG.md".
+
+\warning Backward compatiblity is considered on the level of C++ source code, not binary linkage.
+Adding new members to existing structures is treated as backward compatible if initializing
+the new members to binary zero results in the old behavior.
+You should always fully initialize all library structures to zeros and not rely on their
+exact binary size.
+
\section general_considerations_validation_layer_warnings Validation layer warnings
When using this library, you can meet following types of warnings issued by