diff --git a/docs/gfx/Linear_allocator_1_algo_default.png b/docs/gfx/Linear_allocator_1_algo_default.png new file mode 100644 index 0000000..e3e8f47 Binary files /dev/null and b/docs/gfx/Linear_allocator_1_algo_default.png differ diff --git a/docs/gfx/Linear_allocator_2_algo_linear.png b/docs/gfx/Linear_allocator_2_algo_linear.png new file mode 100644 index 0000000..c2a38b2 Binary files /dev/null and b/docs/gfx/Linear_allocator_2_algo_linear.png differ diff --git a/docs/gfx/Linear_allocator_3_free_at_once.png b/docs/gfx/Linear_allocator_3_free_at_once.png new file mode 100644 index 0000000..a000ad5 Binary files /dev/null and b/docs/gfx/Linear_allocator_3_free_at_once.png differ diff --git a/docs/gfx/Linear_allocator_4_stack.png b/docs/gfx/Linear_allocator_4_stack.png new file mode 100644 index 0000000..4dba064 Binary files /dev/null and b/docs/gfx/Linear_allocator_4_stack.png differ diff --git a/docs/gfx/Linear_allocator_5_ring_buffer.png b/docs/gfx/Linear_allocator_5_ring_buffer.png new file mode 100644 index 0000000..0686702 Binary files /dev/null and b/docs/gfx/Linear_allocator_5_ring_buffer.png differ diff --git a/docs/gfx/Linear_allocator_6_ring_buffer_lost.png b/docs/gfx/Linear_allocator_6_ring_buffer_lost.png new file mode 100644 index 0000000..9a2338c Binary files /dev/null and b/docs/gfx/Linear_allocator_6_ring_buffer_lost.png differ diff --git a/docs/gfx/Linear_allocator_7_double_stack.png b/docs/gfx/Linear_allocator_7_double_stack.png new file mode 100644 index 0000000..77d1468 Binary files /dev/null and b/docs/gfx/Linear_allocator_7_double_stack.png differ diff --git a/docs/html/custom_memory_pools.html b/docs/html/custom_memory_pools.html index fed21c8..9f72b3f 100644 --- a/docs/html/custom_memory_pools.html +++ b/docs/html/custom_memory_pools.html @@ -78,7 +78,7 @@ $(function() {
  1. Fill VmaPoolCreateInfo structure.
  2. Call vmaCreatePool() to obtain VmaPool handle.
    3. -
  3. When making an allocation, set VmaAllocationCreateInfo::pool to this handle. You don't need to specify any other parameters of this structure, like usage.
    4. +
  4. When making an allocation, set VmaAllocationCreateInfo::pool to this handle. You don't need to specify any other parameters of this structure, like usage.

Example:

// Create a pool that can have at most 2 blocks, 128 MiB each.
VmaPoolCreateInfo poolCreateInfo = {};
poolCreateInfo.memoryTypeIndex = ...
poolCreateInfo.blockSize = 128ull * 1024 * 1024;
poolCreateInfo.maxBlockCount = 2;
VmaPool pool;
vmaCreatePool(allocator, &poolCreateInfo, &pool);
// Allocate a buffer out of it.
VkBufferCreateInfo bufCreateInfo = { VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO };
bufCreateInfo.size = 1024;
bufCreateInfo.usage = VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT | VK_BUFFER_USAGE_TRANSFER_DST_BIT;
VmaAllocationCreateInfo allocCreateInfo = {};
allocCreateInfo.pool = pool;
VkBuffer buf;
VmaAllocation alloc;
VmaAllocationInfo allocInfo;
vmaCreateBuffer(allocator, &bufCreateInfo, &allocCreateInfo, &buf, &alloc, &allocInfo);

You have to free all allocations made from this pool before destroying it.

@@ -88,9 +88,55 @@ Choosing memory type index
VkBufferCreateInfo exampleBufCreateInfo = { VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO };
exampleBufCreateInfo.size = 1024; // Whatever.
exampleBufCreateInfo.usage = VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT | VK_BUFFER_USAGE_TRANSFER_DST_BIT; // Change if needed.
VmaAllocationCreateInfo allocCreateInfo = {};
allocCreateInfo.usage = VMA_MEMORY_USAGE_GPU_ONLY; // Change if needed.
uint32_t memTypeIndex;
vmaFindMemoryTypeIndexForBufferInfo(allocator, &exampleBufCreateInfo, &allocCreateInfo, &memTypeIndex);
VmaPoolCreateInfo poolCreateInfo = {};
poolCreateInfo.memoryTypeIndex = memTypeIndex;
// ...

When creating buffers/images allocated in that pool, provide following parameters:

- +

+Linear allocation algorithm

+

Each Vulkan memory block managed by this library has accompanying metadata that keeps track of used and unused regions. By default, the metadata structure and algorithm tries to find best place for new allocations among free regions to optimize memory usage. This way you can allocate and free objects in any order.

+
+Default allocation algorithm +
+

Sometimes there is a need to use simpler, linear allocation algorithm. You can create custom pool that uses such algorithm by adding flag VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT to VmaPoolCreateInfo::flags while creating VmaPool object. Then an alternative metadata management is used. It always creates new allocations after last one and doesn't reuse free regions after allocations freed in the middle. It results in better allocation performance and less memory consumed by metadata.

+
+Linear allocation algorithm +
+

With this one flag, you can create a custom pool that can be used in many ways: free-at-once, stack, double stack, and ring buffer. See below for details.

+

Pools with linear algorithm must have only one memory block - VmaPoolCreateInfo::maxBlockCount must be 1.

+

+Free-at-once

+

In a pool that uses linear algorithm, you still need to free all the allocations individually, e.g. by using vmaFreeMemory() or vmaDestroyBuffer(). You can free them in any order. New allocations are always made after last one - free space in the middle is not reused. However, when you release all the allocation and the pool becomes empty, allocation starts from the beginning again. This way you can use linear algorithm to speed up creation of allocations that you are going to release all at once.

+
+Free-at-once +
+

+Stack

+

When you free an allocation that was created last, its space can be reused. Thanks to this, if you always release allocations in the order opposite to their creation (LIFO - Last In First Out), you can achieve behavior of a stack.

+
+Stack +
+

+Double stack

+

The space reserved by a custom pool with linear algorithm may be used by two stacks:

+ +

To make allocation from upper stack, add flag VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT to VmaAllocationCreateInfo::flags.

+

When the two stacks' ends meet so there is not enough space between them for a new allocation, such allocation fails with usual VK_ERROR_OUT_OF_DEVICE_MEMORY error.

+
+Double stack +
+

+Ring buffer

+

When you free some allocations from the beginning and there is not enough free space for a new one at the end of a pool, allocator's "cursor" wraps around to the beginning and starts allocation there. Thanks to this, if you always release allocations in the same order as you created them (FIFO - First In First Out), you can achieve behavior of a ring buffer / queue.

+
+Ring buffer +
+

Pools with linear algorithm support lost allocations when used as ring buffer. If there is not enough free space for a new allocation, but existing allocations from the front of the queue can become lost, they become lost and the allocation succeeds.

+
+Ring buffer with lost allocations +
+
Generated by   diff --git a/docs/html/debugging_memory_usage.html b/docs/html/debugging_memory_usage.html index dd84301..83e8d35 100644 --- a/docs/html/debugging_memory_usage.html +++ b/docs/html/debugging_memory_usage.html @@ -69,10 +69,10 @@ $(function() {

If you suspect a bug with memory usage, like usage of uninitialized memory or memory being overwritten out of bounds of an allocation, you can use debug features of this library to verify this.

Memory initialization

-

If you experience a bug with incorrect data in your program and you suspect uninitialized memory to be used, you can enable automatic memory initialization to verify this. To do it, define macro VMA_DEBUG_INITIALIZE_ALLOCATIONS to 1.

+

If you experience a bug with incorrect and nondeterministic data in your program and you suspect uninitialized memory to be used, you can enable automatic memory initialization to verify this. To do it, define macro VMA_DEBUG_INITIALIZE_ALLOCATIONS to 1.

#define VMA_DEBUG_INITIALIZE_ALLOCATIONS 1
#include "vk_mem_alloc.h"

It makes memory of all new allocations initialized to bit pattern 0xDCDCDCDC. Before an allocation is destroyed, its memory is filled with bit pattern 0xEFEFEFEF. Memory is automatically mapped and unmapped if necessary.

If you find these values while debugging your program, good chances are that you incorrectly read Vulkan memory that is allocated but not initialized, or already freed, respectively.

-

Memory initialization works only with memory types that are HOST_VISIBLE. It works also with dedicated allocations. It doesn't work with allocations created with VMA_ALLOCATION_CREATE_CAN_BECOME_LOST_BIT flag, as these they cannot be mapped.

+

Memory initialization works only with memory types that are HOST_VISIBLE. It works also with dedicated allocations. It doesn't work with allocations created with VMA_ALLOCATION_CREATE_CAN_BECOME_LOST_BIT flag, as they cannot be mapped.

Margins

By default, allocations are laid out in memory blocks next to each other if possible (considering required alignment, bufferImageGranularity, and nonCoherentAtomSize).

diff --git a/docs/html/globals.html b/docs/html/globals.html index 2f1f2fb..73bdc76 100644 --- a/docs/html/globals.html +++ b/docs/html/globals.html @@ -89,6 +89,9 @@ $(function() {
  • VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT : vk_mem_alloc.h
    • +
  • VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT +: vk_mem_alloc.h +
  • VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT : vk_mem_alloc.h
    • @@ -128,6 +131,9 @@ $(function() {
  • VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT : vk_mem_alloc.h
    • +
  • VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT +: vk_mem_alloc.h +
  • VMA_RECORD_FLAG_BITS_MAX_ENUM : vk_mem_alloc.h
    • diff --git a/docs/html/globals_eval.html b/docs/html/globals_eval.html index 091f76b..38d7955 100644 --- a/docs/html/globals_eval.html +++ b/docs/html/globals_eval.html @@ -77,6 +77,9 @@ $(function() {
  • VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT : vk_mem_alloc.h
    • +
  • VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT +: vk_mem_alloc.h +
  • VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT : vk_mem_alloc.h
    • @@ -113,6 +116,9 @@ $(function() {
  • VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT : vk_mem_alloc.h
    • +
  • VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT +: vk_mem_alloc.h +
  • VMA_RECORD_FLAG_BITS_MAX_ENUM : vk_mem_alloc.h
    • diff --git a/docs/html/index.html b/docs/html/index.html index dc62c22..e231206 100644 --- a/docs/html/index.html +++ b/docs/html/index.html @@ -62,7 +62,7 @@ $(function() {
    Vulkan Memory Allocator
    -

    Version 2.1.0-alpha.3 (2018-06-14)

    +

    Version 2.1.0-alpha.4 (2018-08-22)

    Copyright (c) 2017-2018 Advanced Micro Devices, Inc. All rights reserved.
    License: MIT

    Documentation of all members: vk_mem_alloc.h

    @@ -92,6 +92,13 @@ Table of contents
  • Custom memory pools
  • Defragmentation
    • diff --git a/docs/html/search/all_f.js b/docs/html/search/all_f.js index 60dbc10..cc8c748 100644 --- a/docs/html/search/all_f.js +++ b/docs/html/search/all_f.js @@ -25,6 +25,7 @@ var searchData= ['vma_5fallocation_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_ALLOCATION_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597ae5633ec569f4899cf8f29e7385b2f882',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fmapped_5fbit',['VMA_ALLOCATION_CREATE_MAPPED_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a11da372cc3a82931c5e5d6146cd9dd1f',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fnever_5fallocate_5fbit',['VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a89759603401014eb325eb22a3839f2ff',1,'vk_mem_alloc.h']]], + ['vma_5fallocation_5fcreate_5fupper_5faddress_5fbit',['VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a42ba3a2d2c7117953210b7c3ef8da0df',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fuser_5fdata_5fcopy_5fstring_5fbit',['VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597aa6f24f821cd6a7c5e4a443f7bf59c520',1,'vk_mem_alloc.h']]], ['vma_5fallocator_5fcreate_5fexternally_5fsynchronized_5fbit',['VMA_ALLOCATOR_CREATE_EXTERNALLY_SYNCHRONIZED_BIT',['../vk__mem__alloc_8h.html#a4f87c9100d154a65a4ad495f7763cf7ca4816ddaed324ba110172ca608a20f29d',1,'vk_mem_alloc.h']]], ['vma_5fallocator_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_ALLOCATOR_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a4f87c9100d154a65a4ad495f7763cf7cae4d5ad929caba5f23eb502b13bd5286c',1,'vk_mem_alloc.h']]], @@ -38,6 +39,7 @@ var searchData= ['vma_5fmemory_5fusage_5funknown',['VMA_MEMORY_USAGE_UNKNOWN',['../vk__mem__alloc_8h.html#aa5846affa1e9da3800e3e78fae2305ccaf50d27e34e0925cf3a63db8c839121dd',1,'vk_mem_alloc.h']]], ['vma_5fpool_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_POOL_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a1c7312bea9ea246846b9054fd6bd6aec',1,'vk_mem_alloc.h']]], ['vma_5fpool_5fcreate_5fignore_5fbuffer_5fimage_5fgranularity_5fbit',['VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a9f1a499508a8edb4e8ba40aa0290a3d2',1,'vk_mem_alloc.h']]], + ['vma_5fpool_5fcreate_5flinear_5falgorithm_5fbit',['VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a13c8a444197c67866be9cb05599fc726',1,'vk_mem_alloc.h']]], ['vma_5frecord_5fflag_5fbits_5fmax_5fenum',['VMA_RECORD_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a4dd2c44642312a147a4e93373a6e64d2a20dd17d69966dbffa054739d6090b85e',1,'vk_mem_alloc.h']]], ['vma_5frecord_5fflush_5fafter_5fcall_5fbit',['VMA_RECORD_FLUSH_AFTER_CALL_BIT',['../vk__mem__alloc_8h.html#a4dd2c44642312a147a4e93373a6e64d2a8e7ab322e8732654be627c4ea8f36cc7',1,'vk_mem_alloc.h']]], ['vma_5frecording_5fenabled',['VMA_RECORDING_ENABLED',['../vk__mem__alloc_8h.html#a1f0c126759fc96ccb6e2d23c101d770c',1,'vk_mem_alloc.h']]], diff --git a/docs/html/search/enumvalues_0.js b/docs/html/search/enumvalues_0.js index 9f462fc..ee262d0 100644 --- a/docs/html/search/enumvalues_0.js +++ b/docs/html/search/enumvalues_0.js @@ -6,6 +6,7 @@ var searchData= ['vma_5fallocation_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_ALLOCATION_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597ae5633ec569f4899cf8f29e7385b2f882',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fmapped_5fbit',['VMA_ALLOCATION_CREATE_MAPPED_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a11da372cc3a82931c5e5d6146cd9dd1f',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fnever_5fallocate_5fbit',['VMA_ALLOCATION_CREATE_NEVER_ALLOCATE_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a89759603401014eb325eb22a3839f2ff',1,'vk_mem_alloc.h']]], + ['vma_5fallocation_5fcreate_5fupper_5faddress_5fbit',['VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597a42ba3a2d2c7117953210b7c3ef8da0df',1,'vk_mem_alloc.h']]], ['vma_5fallocation_5fcreate_5fuser_5fdata_5fcopy_5fstring_5fbit',['VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT',['../vk__mem__alloc_8h.html#ad9889c10c798b040d59c92f257cae597aa6f24f821cd6a7c5e4a443f7bf59c520',1,'vk_mem_alloc.h']]], ['vma_5fallocator_5fcreate_5fexternally_5fsynchronized_5fbit',['VMA_ALLOCATOR_CREATE_EXTERNALLY_SYNCHRONIZED_BIT',['../vk__mem__alloc_8h.html#a4f87c9100d154a65a4ad495f7763cf7ca4816ddaed324ba110172ca608a20f29d',1,'vk_mem_alloc.h']]], ['vma_5fallocator_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_ALLOCATOR_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a4f87c9100d154a65a4ad495f7763cf7cae4d5ad929caba5f23eb502b13bd5286c',1,'vk_mem_alloc.h']]], @@ -18,6 +19,7 @@ var searchData= ['vma_5fmemory_5fusage_5funknown',['VMA_MEMORY_USAGE_UNKNOWN',['../vk__mem__alloc_8h.html#aa5846affa1e9da3800e3e78fae2305ccaf50d27e34e0925cf3a63db8c839121dd',1,'vk_mem_alloc.h']]], ['vma_5fpool_5fcreate_5fflag_5fbits_5fmax_5fenum',['VMA_POOL_CREATE_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a1c7312bea9ea246846b9054fd6bd6aec',1,'vk_mem_alloc.h']]], ['vma_5fpool_5fcreate_5fignore_5fbuffer_5fimage_5fgranularity_5fbit',['VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a9f1a499508a8edb4e8ba40aa0290a3d2',1,'vk_mem_alloc.h']]], + ['vma_5fpool_5fcreate_5flinear_5falgorithm_5fbit',['VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT',['../vk__mem__alloc_8h.html#a9a7c45f9c863695d98c83fa5ac940fe7a13c8a444197c67866be9cb05599fc726',1,'vk_mem_alloc.h']]], ['vma_5frecord_5fflag_5fbits_5fmax_5fenum',['VMA_RECORD_FLAG_BITS_MAX_ENUM',['../vk__mem__alloc_8h.html#a4dd2c44642312a147a4e93373a6e64d2a20dd17d69966dbffa054739d6090b85e',1,'vk_mem_alloc.h']]], ['vma_5frecord_5fflush_5fafter_5fcall_5fbit',['VMA_RECORD_FLUSH_AFTER_CALL_BIT',['../vk__mem__alloc_8h.html#a4dd2c44642312a147a4e93373a6e64d2a8e7ab322e8732654be627c4ea8f36cc7',1,'vk_mem_alloc.h']]] ]; diff --git a/docs/html/struct_vma_pool_create_info.html b/docs/html/struct_vma_pool_create_info.html index 153ff87..aa5ba2a 100644 --- a/docs/html/struct_vma_pool_create_info.html +++ b/docs/html/struct_vma_pool_create_info.html @@ -160,8 +160,8 @@ Public Attributes

    Maximum number of blocks that can be allocated in this pool. Optional.

    -

    Optional. Set to 0 to use SIZE_MAX, which means no limit.

    -

    Set to same value as minBlockCount to have fixed amount of memory allocated throuout whole lifetime of this pool.

    +

    Set to 0 to use default, which is SIZE_MAX, which means no limit. When VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT is used, default is 1.

    +

    Set to same value as VmaPoolCreateInfo::minBlockCount to have fixed amount of memory allocated throughout whole lifetime of this pool.

    @@ -194,7 +194,7 @@ Public Attributes

    Minimum number of blocks to be always allocated in this pool, even if they stay empty.

    -

    Set to 0 to have no preallocated blocks and let the pool be completely empty.

    +

    Set to 0 to have no preallocated blocks and allow the pool be completely empty.

    diff --git a/docs/html/usage_patterns.html b/docs/html/usage_patterns.html index 0495f56..897dca1 100644 --- a/docs/html/usage_patterns.html +++ b/docs/html/usage_patterns.html @@ -84,7 +84,7 @@ Immutable resources Dynamic resources

    When: Any resources that change frequently (aka "dynamic"), e.g. every frame or every draw call, written on CPU, read on GPU.

    What to do: Create them using VMA_MEMORY_USAGE_CPU_TO_GPU. You can map it and write to it directly on CPU, as well as read from it on GPU.

    -

    This is a more complex situation. Different solutions are possible, and the best one depends on specific GPU type, but you can use this simple approach for the start. Prefer to write to such resource sequentially (e.g. using memcpy). Don't perform random access or any reads from it, as it may be very slow.

    +

    This is a more complex situation. Different solutions are possible, and the best one depends on specific GPU type, but you can use this simple approach for the start. Prefer to write to such resource sequentially (e.g. using memcpy). Don't perform random access or any reads from it on CPU, as it may be very slow.

    Readback

    When: Resources that contain data written by GPU that you want to read back on CPU, e.g. results of some computations.

    @@ -93,7 +93,7 @@ Readback Advanced patterns

    Detecting integrated graphics

    -

    You can support integrated graphics (like Intel HD Graphics, AMD APU) better by detecting it in Vulkan. To do it, call vkGetPhysicalDeviceProperties(), inspect VkPhysicalDeviceProperties::deviceType and look for VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU. When you find it, you can assume that memory is unified and all memory types are equally fast to access from GPU, regardless of VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT.

    +

    You can support integrated graphics (like Intel HD Graphics, AMD APU) better by detecting it in Vulkan. To do it, call vkGetPhysicalDeviceProperties(), inspect VkPhysicalDeviceProperties::deviceType and look for VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU. When you find it, you can assume that memory is unified and all memory types are comparably fast to access from GPU, regardless of VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT.

    You can then sum up sizes of all available memory heaps and treat them as useful for your GPU resources, instead of only DEVICE_LOCAL ones. You can also prefer to create your resources in memory types that are HOST_VISIBLE to map them directly instead of submitting explicit transfer (see below).

    Direct access versus transfer

    @@ -105,7 +105,7 @@ Direct access versus transfer

    Which solution is the most efficient depends on your resource and especially on the GPU. It is best to measure it and then make the decision. Some general recommendations:

    diff --git a/docs/html/vk__mem__alloc_8h.html b/docs/html/vk__mem__alloc_8h.html index 4c2a48f..49c5731 100644 --- a/docs/html/vk__mem__alloc_8h.html +++ b/docs/html/vk__mem__alloc_8h.html @@ -217,12 +217,14 @@ Enumerations
      VMA_ALLOCATION_CREATE_CAN_MAKE_OTHER_LOST_BIT = 0x00000010, VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT = 0x00000020, +VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT = 0x00000040, VMA_ALLOCATION_CREATE_FLAG_BITS_MAX_ENUM = 0x7FFFFFFF
    } Flags to be passed as VmaAllocationCreateInfo::flags. More...
      enum  VmaPoolCreateFlagBits { VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT = 0x00000002, +VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT = 0x00000004, VMA_POOL_CREATE_FLAG_BITS_MAX_ENUM = 0x7FFFFFFF } Flags to be passed as VmaPoolCreateInfo::flags. More...
    @@ -780,6 +782,9 @@ Functions VMA_ALLOCATION_CREATE_USER_DATA_COPY_STRING_BIT 

    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 freed together with the allocation. It is also used in vmaBuildStatsString().

    +VMA_ALLOCATION_CREATE_UPPER_ADDRESS_BIT 

    Allocation will be created from upper stack in a double stack pool.

    +

    This flag is only allowed for custom pools created with VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT flag.

    + VMA_ALLOCATION_CREATE_FLAG_BITS_MAX_ENUM  @@ -874,10 +879,15 @@ Functions

    Flags to be passed as VmaPoolCreateInfo::flags.

    +
    Enumerator
    VMA_POOL_CREATE_IGNORE_BUFFER_IMAGE_GRANULARITY_BIT 

    Use this flag if you always allocate only buffers and linear images or only optimal images out of this pool and so Buffer-Image Granularity can be ignored.

    -

    This is na optional optimization flag.

    +

    This is an optional optimization flag.

    If you always allocate using vmaCreateBuffer(), vmaCreateImage(), vmaAllocateMemoryForBuffer(), then you don't need to use it because allocator knows exact type of your allocations so it can handle Buffer-Image Granularity in the optimal way.

    If you also allocate using vmaAllocateMemoryForImage() or vmaAllocateMemory(), exact type of such allocations is not known, so allocator must be conservative in handling Buffer-Image Granularity, which can lead to suboptimal allocation (wasted memory). In that case, if you can make sure you always allocate only buffers and linear images or only optimal images out of this pool, use this flag to make allocator disregard Buffer-Image Granularity and so make allocations more optimal.

    VMA_POOL_CREATE_LINEAR_ALGORITHM_BIT 

    Enables alternative, linear allocation algorithm in this pool.

    +

    Specify this flag to enable linear allocation algorithm, which always creates new allocations after last one and doesn't reuse space from allocations freed in between. It trades memory consumption for simplified algorithm and data structure, which has better performance and uses less memory for metadata.

    +

    By using this flag, you can achieve behavior of free-at-once, stack, ring buffer, and double stack. For details, see documentation chapter Linear allocation algorithm.

    +

    When using this flag, you must specify VmaPoolCreateInfo::maxBlockCount == 1 (or 0 for default).

    +
    VMA_POOL_CREATE_FLAG_BITS_MAX_ENUM 
    @@ -1246,7 +1256,7 @@ Functions -

    Corruption detection is enabled only when VMA_DEBUG_DETECT_CORRUPTION macro is defined to nonzero, VMA_DEBUG_MARGIN is defined to nonzero and only for memory types that are HOST_VISIBLE and HOST_COHERENT. For more information, see Corruption detection.

    +

    Corruption detection is enabled only when VMA_DEBUG_DETECT_CORRUPTION macro is defined to nonzero, VMA_DEBUG_MARGIN is defined to nonzero and only for memory types that are HOST_VISIBLE and HOST_COHERENT. For more information, see Corruption detection.

    Possible return values: