ESP-IDF applications use the common computer architecture patterns of *stack* (dynamic memory allocated by program control flow) and *heap* (dynamic memory allocated by function calls), as well as statically allocated memory (allocated at compile time).
Because ESP-IDF is a multi-threaded RTOS environment, each RTOS task has its own stack. By default, each of these stacks is allocated from the heap when the task is created. (See :cpp:func:`xTaskCreateStatic` for the alternative where stacks are statically allocated.)
Because {IDF_TARGET_NAME} uses multiple types of RAM, it also contains multiple heaps with different capabilities. A capabilities-based memory allocator allows apps to make heap allocations for different purposes.
For most purposes, the standard libc ``malloc()`` and ``free()`` functions can be used for heap allocation without any special consideration.
However, in order to fully make use of all of the memory types and their characteristics, ESP-IDF also has a
capabilities-based heap memory allocator. If you want to have memory with certain properties (for example, :ref:`dma-capable-memory` or executable-memory), you can create an OR-mask of the required capabilities and pass that to :cpp:func:`heap_caps_malloc`.
- DRAM (Data RAM) is memory used to hold data. This is the most common kind of memory accessed as heap.
- IRAM (Instruction RAM) usually holds executable data only. If accessed as generic memory, all accesses must be :ref:`32-bit aligned<32-Bit Accessible Memory>`.
- D/IRAM is RAM which can be used as either Instruction or Data RAM.
For more details on these internal memory types, see :ref:`memory-layout`.
It's also possible to connect external SPI RAM to the {IDF_TARGET_NAME} - :doc:`external RAM </api-guides/external-ram>` can be integrated into the {IDF_TARGET_NAME}'s memory map using the flash cache, and accessed similarly to DRAM.
DRAM uses capability ``MALLOC_CAP_8BIT`` (accessible in single byte reads and writes). To test the free DRAM heap size at runtime, call cpp:func:`heap_caps_get_free_size(MALLOC_CAP_8BIT)`.
When calling ``malloc()``, the ESP-IDF ``malloc()`` implementation internally calls cpp:func:`heap_caps_malloc_default(size)`. This will allocate memory with capability ``MALLOC_CAP_DEFAULT``, which is byte-addressable.
At startup, the DRAM heap contains all data memory which is not statically allocated by the app. Reducing statically allocated buffers will increase the amount of available free heap.
..note:: At runtime, the available heap DRAM may be less than calculated at compile time, because at startup some memory is allocated from the heap before the FreeRTOS scheduler is started (including memory for the stacks of initial FreeRTOS tasks).
IRAM
^^^^
At startup, the IRAM heap contains all instruction memory which is not used by the app executable code.
Some memory in the {IDF_TARGET_NAME} is available as either DRAM or IRAM. If memory is allocated from a D/IRAM region, the free heap size for both types of memory will decrease.
Use the ``MALLOC_CAP_DMA`` flag to allocate memory which is suitable for use with hardware DMA engines (for example SPI and I2S). This capability flag excludes any external PSRAM.
The EDMA hardware feature allows DMA buffers to be placed in external PSRAM, but there may be additional alignment constraints. Consult the {IDF_TARGET_NAME} Technical Reference Manual for details. To allocate a DMA-capable external memory buffer, use the ``MALLOC_CAP_SPIRAM`` capabilities flag together with :cpp:func:`heap_caps_aligned_alloc` with the necessary alignment specified.
If a certain memory structure is only addressed in 32-bit units, for example an array of ints or pointers, it can be useful to allocate it with the ``MALLOC_CAP_32BIT`` flag. This also allows the allocator to give out IRAM memory; something which it can't do for a normal malloc() call. This can help to use all the available memory in the {IDF_TARGET_NAME}.
..only:: CONFIG_IDF_TARGET_ARCH_XTENSA and SOC_CPU_HAS_FPU
Please note that on {IDF_TARGET_NAME} series chips, ``MALLOC_CAP_32BIT`` cannot be used for storing floating-point variables. This is because ``MALLOC_CAP_32BIT`` may return instruction RAM, and the floating-point assembly instructions on {IDF_TARGET_NAME} cannot access instruction RAM.
Memory allocated with ``MALLOC_CAP_32BIT`` can *only* be accessed via 32-bit reads and writes, any other type of access will generate a fatal LoadStoreError exception.
When :doc:`external RAM </api-guides/external-ram>` is enabled, external SPI RAM under 4MiB in size can be allocated using standard ``malloc`` calls, or via ``heap_caps_malloc(MALLOC_CAP_SPIRAM)``, depending on configuration. See :ref:`external_ram_config` for more details.
Heap functions are thread safe, meaning they can be called from different tasks simultaneously without any limitations.
It is technically possible to call ``malloc``, ``free``, and related functions from interrupt handler (ISR) context. However this is not recommended, as heap function calls may delay other interrupts. It is strongly recommended to refactor applications so that any buffers used by an ISR are pre-allocated outside of the ISR. Support for calling heap functions from ISRs may be removed in a future update.
Knowledge about the regions of memory in the chip comes from the "soc" component, which contains memory layout information for the chip, and the different capabilities of each region. Each region's capabilities are prioritised, so that (for example) dedicated DRAM and IRAM regions will be used for allocations ahead of the more versatile D/IRAM regions.
Each contiguous region of memory contains its own memory heap. The heaps are created using the :ref:`multi_heap <multi-heap>` functionality. multi_heap allows any contiguous region of memory to be used as a heap.
The heap capabilities allocator uses knowledge of the memory regions to initialize each individual heap. Allocation functions in the heap capabilities API will find the most appropriate heap for the allocation (based on desired capabilities, available space, and preferences for each region's use) and then calling :cpp:func:`multi_heap_malloc` for the heap situated in that particular region.
Calling ``free()`` involves finding the particular heap corresponding to the freed address, and then calling :cpp:func:`multi_heap_free` on that particular multi_heap instance.
