### Configure Build Prefix with Meson

Source: https://github.com/mesa/libdrm/blob/main/README.rst

Configures the installation prefix for the project after initial Meson setup. This example sets the prefix to /usr.

```bash
meson configure builddir/ -Dprefix=/usr
```

--------------------------------

### Build and Install with Ninja

Source: https://github.com/mesa/libdrm/blob/main/README.rst

Builds and installs the project using Ninja after Meson setup. If installing to a system location, root privileges may be required for the install step.

```bash
ninja -C builddir/ install
```

--------------------------------

### Configure Build Directory with Meson

Source: https://github.com/mesa/libdrm/blob/main/README.rst

Sets up the build directory for the project using Meson. The default installation prefix is /usr/local, but can be changed.

```bash
meson builddir/
```

--------------------------------

### Example Commit Message Subject Line

Source: https://github.com/mesa/libdrm/blob/main/CONTRIBUTING.rst

Illustrates the recommended format for the first line of a commit message, including a component prefix and a concise description of the change.

```git commit
amdgpu: Use uint32_t i in amdgpu_find_bo_by_cpu_mapping
```

--------------------------------

### GEM Buffer Operations

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for specific GEM buffer operations, including opening prime buffers, getting kernel names, and setting domains.

```APIDOC
## GEM Buffer Operations

### Description
Provides specialized functions for GEM buffer operations, including inter-process communication via PRIME and setting memory domains.

### Functions
- **radeon_bo_open_prime**: Opens a GEM buffer object using PRIME file descriptor.
- **radeon_gem_bo_open_prime**: Opens a GEM buffer object using PRIME file descriptor (alternative name).
- **radeon_gem_get_kernel_name**: Retrieves the kernel name of a GEM buffer object.
- **radeon_gem_get_reloc_in_cs**: Gets relocation information for a GEM buffer in a command submission.
- **radeon_gem_name_bo**: Names a GEM buffer object.
- **radeon_gem_prime_share_bo**: Shares a GEM buffer object via PRIME.
- **radeon_gem_set_domain**: Sets the memory domain for a GEM buffer object.
```

--------------------------------

### Create and Map a Dumb-Buffer

Source: https://github.com/mesa/libdrm/blob/main/man/drm-memory.7.rst

Demonstrates the process of creating a driver-independent dumb-buffer, binding it to a DRM framebuffer, and memory-mapping it for direct access. This is useful for scanout or when a driver-specific buffer object is not required.

```c
struct drm_mode_create_dumb creq;
struct drm_mode_destroy_dumb dreq;
struct drm_mode_map_dumb mreq;
uint32_t fb;
int ret;
void *map;

/* create dumb buffer */
memset(&creq, 0, sizeof(creq));
creq.width = 1920;
creq.height = 1080;
creq.bpp = 32;
ret = drmIoctl(fd, DRM_IOCTL_MODE_CREATE_DUMB, &creq);
if (ret < 0) {
    /* buffer creation failed; see "errno" for more error codes */
    ...
}
/* creq.pitch, creq.handle and creq.size are filled by this ioctl with
 * the requested values and can be used now. */

/* create framebuffer object for the dumb-buffer */
ret = drmModeAddFB(fd, 1920, 1080, 24, 32, creq.pitch, creq.handle, &fb);
if (ret) {
    /* frame buffer creation failed; see "errno" */
    ...
}
/* the framebuffer "fb" can now used for scanout with KMS */

/* prepare buffer for memory mapping */
memset(&mreq, 0, sizeof(mreq));
mreq.handle = creq.handle;
ret = drmIoctl(fd, DRM_IOCTL_MODE_MAP_DUMB, &mreq);
if (ret) {
    /* DRM buffer preparation failed; see "errno" */
    ...
}
/* mreq.offset now contains the new offset that can be used with mmap() */

/* perform actual memory mapping */
map = mmap(0, creq.size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, mreq.offset);
if (map == MAP_FAILED) {
    /* memory-mapping failed; see "errno" */
    ...
}

/* clear the framebuffer to 0 */
memset(map, 0, creq.size);
```

--------------------------------

### Exynos Video Connection

Source: https://github.com/mesa/libdrm/blob/main/exynos/exynos-symbols.txt

Function for establishing a video connection.

```APIDOC
## Exynos Video Connection

### Description
Establishes a video connection within the Exynos system.

### Function

- **exynos_vidi_connection**: Establishes a video connection.
```

--------------------------------

### Device Management Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for initializing, deinitializing, and querying information about the AMDGPU device.

```APIDOC
## Device Management

### Functions
- `amdgpu_device_deinitialize`: Deinitializes the AMDGPU device.
- `amdgpu_device_get_fd`: Gets the file descriptor for the AMDGPU device.
- `amdgpu_device_initialize`: Initializes the AMDGPU device.
- `amdgpu_device_initialize2`: Initializes the AMDGPU device (version 2).
```

--------------------------------

### Surface Operations

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for initializing and managing Radeon surfaces.

```APIDOC
## Surface Operations

### Description
Functions related to the initialization and management of Radeon surfaces.

### Functions
- **radeon_surface_best**: Determines the best surface configuration.
- **radeon_surface_init**: Initializes a surface.
- **radeon_surface_manager_free**: Frees resources associated with the surface manager.
- **radeon_surface_manager_new**: Creates a new surface manager.
```

--------------------------------

### Client and Device Operations

Source: https://github.com/mesa/libdrm/blob/main/nouveau/nouveau-symbols.txt

Functions for managing clients and devices, including opening, closing, and creating them.

```APIDOC
## Client and Device Operations

### Description
Functions for managing clients and devices, including opening, closing, and creating them.

### Functions
- **nouveau_check_dead_channel**: Check for a dead channel.
- **nouveau_client_del**: Delete a client.
- **nouveau_client_new**: Create a new client.
- **nouveau_device_del**: Delete a device.
- **nouveau_device_new**: Create a new device.
- **nouveau_device_open**: Open a device.
- **nouveau_device_open_existing**: Open an existing device.
- **nouveau_device_wrap**: Wrap an existing device.
```

--------------------------------

### Exynos Prime Buffer Sharing

Source: https://github.com/mesa/libdrm/blob/main/exynos/exynos-symbols.txt

Functions for converting between file descriptors and handles for Exynos Prime buffers, enabling efficient buffer sharing.

```APIDOC
## Exynos Prime Buffer Sharing Functions

### Description
Functions for interoperability with PRIME buffers, allowing conversion between file descriptors and handles.

### Functions

- **exynos_prime_fd_to_handle**: Converts a PRIME buffer file descriptor to a handle.
- **exynos_prime_handle_to_fd**: Converts a handle to a PRIME buffer file descriptor.
```

--------------------------------

### Set AMDGPU ASIC ID Table Paths

Source: https://github.com/mesa/libdrm/blob/main/README.rst

Specifies alternative paths for the amdgpu.ids file at runtime by setting the AMDGPU_ASIC_ID_TABLE_PATHS environment variable. This feature requires the C library's secure_getenv() function.

```bash
export AMDGPU_ASIC_ID_TABLE_PATHS=/path/to/tables:/another/path
```

--------------------------------

### Command Submission and Synchronization Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for creating, submitting, and synchronizing command submissions, including fences and sync objects.

```APIDOC
## Command Submission and Synchronization

### Functions
- `amdgpu_cs_chunk_fence_info_to_data`: Converts fence information to data.
- `amdgpu_cs_chunk_fence_to_dep`: Converts fence to dependency.
- `amdgpu_cs_create_semaphore`: Creates a semaphore.
- `amdgpu_cs_create_syncobj`: Creates a sync object.
- `amdgpu_cs_create_syncobj2`: Creates a sync object (version 2).
- `amdgpu_cs_ctx_create`: Creates a command submission context.
- `amdgpu_cs_ctx_create2`: Creates a command submission context (version 2).
- `amdgpu_cs_ctx_free`: Frees a command submission context.
- `amdgpu_cs_ctx_override_priority`: Overrides the priority of a command submission context.
- `amdgpu_cs_ctx_stable_pstate`: Sets the stable power state for a command submission context.
- `amdgpu_cs_destroy_semaphore`: Destroys a semaphore.
- `amdgpu_cs_destroy_syncobj`: Destroys a sync object.
- `amdgpu_cs_export_syncobj`: Exports a sync object.
- `amdgpu_cs_fence_to_handle`: Converts a fence to a handle.
- `amdgpu_cs_import_syncobj`: Imports a sync object.
- `amdgpu_cs_query_fence_status`: Queries the status of a fence.
- `amdgpu_cs_query_reset_state`: Queries the reset state.
- `amdgpu_cs_query_reset_state2`: Queries the reset state (version 2).
- `amdgpu_query_sw_info`: Queries software information.
- `amdgpu_cs_signal_semaphore`: Signals a semaphore.
- `amdgpu_cs_submit`: Submits commands to the GPU.
- `amdgpu_cs_submit_raw`: Submits raw commands to the GPU.
- `amdgpu_cs_submit_raw2`: Submits raw commands to the GPU (version 2).
- `amdgpu_cs_syncobj_export_sync_file`: Exports a sync object to a file descriptor.
- `amdgpu_cs_syncobj_export_sync_file2`: Exports a sync object to a file descriptor (version 2).
- `amdgpu_cs_syncobj_import_sync_file`: Imports a sync object from a file descriptor.
- `amdgpu_cs_syncobj_import_sync_file2`: Imports a sync object from a file descriptor (version 2).
- `amdgpu_cs_syncobj_query`: Queries a sync object.
- `amdgpu_cs_syncobj_query2`: Queries a sync object (version 2).
- `amdgpu_cs_syncobj_reset`: Resets a sync object.
- `amdgpu_cs_syncobj_signal`: Signals a sync object.
- `amdgpu_cs_syncobj_timeline_signal`: Signals a timeline sync object.
- `amdgpu_cs_syncobj_timeline_wait`: Waits on a timeline sync object.
- `amdgpu_cs_syncobj_transfer`: Transfers a sync object.
- `amdgpu_cs_syncobj_wait`: Waits on a sync object.
- `amdgpu_cs_wait_fences`: Waits for multiple fences.
- `amdgpu_cs_wait_semaphore`: Waits on a semaphore.
```

--------------------------------

### Device Management Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, referencing, and deleting Etnaviv devices, as well as retrieving the device file descriptor.

```APIDOC
## Device Management Functions

### Description
Functions for managing Etnaviv devices.

### Functions
- **etna_device_new**
- **etna_device_new_dup**
- **etna_device_ref**
- **etna_device_del**
- **etna_device_fd**
```

--------------------------------

### User Queue Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for creating and managing user-space queues for command submission.

```APIDOC
## User Queue Functions

### Functions
- `amdgpu_create_userqueue`: Creates a user queue.
- `amdgpu_free_userqueue`: Frees a user queue.
- `amdgpu_userq_signal`: Signals a user queue.
- `amdgpu_userq_wait`: Waits on a user queue.
```

--------------------------------

### Buffer Object Manager (GEM)

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for initializing and destroying the GEM-based buffer object manager.

```APIDOC
## Buffer Object Manager (GEM)

### Description
Manages the creation and destruction of GEM-based buffer objects.

### Functions
- **radeon_bo_manager_gem_ctor**: Constructor for the GEM-based buffer object manager.
- **radeon_bo_manager_gem_dtor**: Destructor for the GEM-based buffer object manager.
```

--------------------------------

### Exynos Event Handling

Source: https://github.com/mesa/libdrm/blob/main/exynos/exynos-symbols.txt

Function for handling Exynos events.

```APIDOC
## Exynos Event Handling

### Description
Handles events related to the Exynos DRM system.

### Function

- **exynos_handle_event**: Processes Exynos events.
```

--------------------------------

### Virtual Address Manager (VAM) Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for managing virtual address spaces.

```APIDOC
## Virtual Address Manager (VAM)

### Functions
- `amdgpu_va_manager_alloc`: Allocates a virtual address manager.
- `amdgpu_va_manager_init`: Initializes a virtual address manager.
- `amdgpu_va_manager_init2`: Initializes a virtual address manager (version 2).
- `amdgpu_va_manager_deinit`: Deinitializes a virtual address manager.
- `amdgpu_va_manager_query_sw_info`: Queries software information for the VAM.
- `amdgpu_va_range_alloc`: Allocates a virtual address range.
- `amdgpu_va_range_alloc2`: Allocates a virtual address range (version 2).
- `amdgpu_va_range_free`: Frees a virtual address range.
- `amdgpu_va_get_start_addr`: Gets the start address of a virtual address range.
- `amdgpu_va_range_query`: Queries a virtual address range.
- `amdgpu_vm_reserve_vmid`: Reserves a Virtual Machine ID (VMID).
- `amdgpu_vm_unreserve_vmid`: Unreserves a Virtual Machine ID (VMID).
```

--------------------------------

### Buffer Object (BO) Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, managing, and accessing buffer objects, including creation from names, dmabufs, and mapping memory.

```APIDOC
## Buffer Object (BO) Functions

### Description
Functions for creating, managing, and accessing buffer objects.

### Functions
- **etna_bo_new**
- **etna_bo_from_name**
- **etna_bo_from_dmabuf**
- **etna_bo_ref**
- **etna_bo_del**
- **etna_bo_get_name**
- **etna_bo_handle**
- **etna_bo_dmabuf**
- **etna_bo_size**
- **etna_bo_map**
- **etna_bo_cpu_prep**
- **etna_bo_cpu_fini**
```

--------------------------------

### Command Submission Manager (GEM)

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for initializing and destroying the GEM-based command submission manager.

```APIDOC
## Command Submission Manager (GEM)

### Description
Manages the creation and destruction of GEM-based command submission contexts.

### Functions
- **radeon_cs_manager_gem_ctor**: Constructor for the GEM-based command submission manager.
- **radeon_cs_manager_gem_dtor**: Destructor for the GEM-based command submission manager.
```

--------------------------------

### Command Submission (CS) Operations

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for creating, managing, and submitting command streams (CSs) to the Radeon hardware.

```APIDOC
## Command Submission (CS) Operations

### Description
Provides a set of functions for managing command streams (CSs) and submitting them for execution. This includes creating, emitting commands, checking space, and managing buffer references within the command stream.

### Functions
- **radeon_cs_begin**: Begins a new command submission sequence.
- **radeon_cs_create**: Creates a new command submission context.
- **radeon_cs_destroy**: Destroys a command submission context.
- **radeon_cs_emit**: Emits a command into the command stream.
- **radeon_cs_end**: Ends the current command submission sequence.
- **radeon_cs_erase**: Erases a command submission.
- **radeon_cs_get_id**: Retrieves the ID of the command submission.
- **radeon_cs_need_flush**: Checks if a flush is needed for the command submission.
- **radeon_cs_print**: Prints the contents of a command submission.
- **radeon_cs_set_limit**: Sets a limit for the command submission.
- **radeon_cs_space_add_persistent_bo**: Adds a persistent buffer object to the command submission space.
- **radeon_cs_space_check**: Checks the available space in the command submission.
- **radeon_cs_space_check_with_bo**: Checks space with a specific buffer object.
- **radeon_cs_space_reset_bos**: Resets buffer objects in the command submission space.
- **radeon_cs_space_set_flush**: Sets the flush behavior for the command submission space.
- **radeon_cs_write_reloc**: Writes a relocation entry into the command stream.
```

--------------------------------

### Information Query Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for querying various hardware and software information related to the AMDGPU device.

```APIDOC
## Information Query

### Functions
- `amdgpu_find_bo_by_cpu_mapping`: Finds a buffer object by its CPU mapping.
- `amdgpu_get_marketing_name`: Gets the marketing name of the GPU.
- `amdgpu_query_buffer_size_alignment`: Queries buffer size and alignment.
- `amdgpu_query_crtc_from_id`: Queries the CRTC from its ID.
- `amdgpu_query_firmware_version`: Queries the firmware version.
- `amdgpu_query_gds_info`: Queries Graphics Data Store (GDS) information.
- `amdgpu_query_gpu_info`: Queries general GPU information.
- `amdgpu_query_gpuvm_fault_info`: Queries GPU VM fault information.
- `amdgpu_query_heap_info`: Queries heap information.
- `amdgpu_query_hw_ip_count`: Queries the count of hardware IP blocks.
- `amdgpu_query_hw_ip_info`: Queries information about a hardware IP block.
- `amdgpu_query_info`: Queries general information.
- `amdgpu_query_sensor_info`: Queries sensor information.
- `amdgpu_query_uq_fw_area_info`: Queries firmware update area information.
- `amdgpu_query_video_caps_info`: Queries video capabilities information.
- `amdgpu_read_mm_registers`: Reads memory-mapped registers.
```

--------------------------------

### Buffer Context Operations

Source: https://github.com/mesa/libdrm/blob/main/nouveau/nouveau-symbols.txt

Functions for managing buffer contexts, including creation, deletion, and referencing.

```APIDOC
## Buffer Context Operations

### Description
Functions for managing buffer contexts, including creation, deletion, and referencing.

### Functions
- **nouveau_bufctx_del**: Delete a buffer context.
- **nouveau_bufctx_mthd**: Execute a method within a buffer context.
- **nouveau_bufctx_new**: Create a new buffer context.
- **nouveau_bufctx_refn**: Reference a buffer context.
- **nouveau_bufctx_reset**: Reset a buffer context.
```

--------------------------------

### Exynos Device Management

Source: https://github.com/mesa/libdrm/blob/main/exynos/exynos-symbols.txt

Functions for creating and destroying Exynos DRM devices, which represent the graphics hardware.

```APIDOC
## Exynos Device Functions

### Description
Functions for managing Exynos DRM devices.

### Functions

- **exynos_device_create**: Creates a new Exynos DRM device.
- **exynos_device_destroy**: Destroys an Exynos DRM device.
```

--------------------------------

### Pipe Management Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, deleting, and waiting on Etnaviv pipes.

```APIDOC
## Pipe Management Functions

### Description
Functions for managing Etnaviv pipes and synchronization.

### Functions
- **etna_pipe_new**
- **etna_pipe_del**
- **etna_pipe_wait**
- **etna_pipe_wait_ns**
```

--------------------------------

### drmModeGetResources

Source: https://github.com/mesa/libdrm/blob/main/man/drmModeGetResources.3.rst

Allocates, populates, and returns a drmModeRes structure containing information about the current display configuration. This includes details on framebuffers, CRTCs, connectors, encoders, and supported framebuffer dimensions.

```APIDOC
## drmModeGetResources

### Description
Retrieves current display configuration information.

### Synopsis
```c
#include <xf86drm.h>
#include <xf86drmMode.h>

drmModeResPtr drmModeGetResources(int fd);
```

### Details
This function allocates and populates a `drmModeRes` structure with the following fields:

- `count_fbs`: Number of currently allocated framebuffer objects.
- `fbs`: Array of framebuffer object IDs.
- `count_crtcs`: Number of available CRTCs.
- `crtcs`: Array of CRTC IDs.
- `count_connectors`: Number of available physical connectors.
- `connectors`: Array of connector IDs.
- `count_encoders`: Number of available encoders.
- `encoders`: Array of encoder IDs.
- `min_width`, `max_width`: Minimum and maximum supported framebuffer width.
- `min_height`, `max_height`: Minimum and maximum supported framebuffer height.

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Return Value
- On success: A pointer to a `drmModeRes` structure.
- On failure: NULL.

### Notes
The returned structure must be freed using `drmModeFreeResources`.
```

--------------------------------

### Exynos Buffer Object Management

Source: https://github.com/mesa/libdrm/blob/main/exynos/exynos-symbols.txt

Functions for creating, destroying, and managing Exynos Buffer Objects (BOs). This includes operations for retrieving buffer information, handling names, and mapping buffers.

```APIDOC
## Exynos Buffer Object Functions

### Description
Provides functions to manage Exynos Buffer Objects (BOs), which are used for memory allocation and sharing.

### Functions

- **exynos_bo_create**: Creates a new Exynos Buffer Object.
- **exynos_bo_destroy**: Destroys an existing Exynos Buffer Object.
- **exynos_bo_from_name**: Retrieves an Exynos Buffer Object from its name.
- **exynos_bo_get_info**: Gets information about an Exynos Buffer Object.
- **exynos_bo_get_name**: Gets the name of an Exynos Buffer Object.
- **exynos_bo_handle**: Retrieves the handle of an Exynos Buffer Object.
- **exynos_bo_map**: Maps an Exynos Buffer Object into the application's address space.
```

--------------------------------

### Buffer Object Management Functions

Source: https://github.com/mesa/libdrm/blob/main/amdgpu/amdgpu-symbols.txt

Functions for allocating, managing, and exporting buffer objects.

```APIDOC
## Buffer Object Management

### Functions
- `amdgpu_bo_alloc`: Allocates a buffer object.
- `amdgpu_bo_cpu_map`: Maps a buffer object to CPU accessible memory.
- `amdgpu_bo_cpu_unmap`: Unmaps a buffer object from CPU memory.
- `amdgpu_bo_export`: Exports a buffer object for use by other processes or hardware.
- `amdgpu_bo_free`: Frees a buffer object.
- `amdgpu_bo_import`: Imports a buffer object from an external source.
- `amdgpu_bo_inc_ref`: Increments the reference count of a buffer object.
- `amdgpu_bo_list_create_raw`: Creates a raw buffer object list.
- `amdgpu_bo_list_destroy_raw`: Destroys a raw buffer object list.
- `amdgpu_bo_list_create`: Creates a buffer object list.
- `amdgpu_bo_list_destroy`: Destroys a buffer object list.
- `amdgpu_bo_list_update`: Updates a buffer object list.
- `amdgpu_bo_query_info`: Queries information about a buffer object.
- `amdgpu_bo_set_metadata`: Sets metadata for a buffer object.
- `amdgpu_bo_va_op`: Performs a virtual address operation on a buffer object.
- `amdgpu_bo_va_op_raw`: Performs a raw virtual address operation on a buffer object.
- `amdgpu_bo_va_op_raw2`: Performs a raw virtual address operation on a buffer object (version 2).
- `amdgpu_bo_wait_for_idle`: Waits for a buffer object to become idle.
- `amdgpu_create_bo_from_user_mem`: Creates a buffer object from user-provided memory.
```

--------------------------------

### Buffer Object Operations

Source: https://github.com/mesa/libdrm/blob/main/nouveau/nouveau-symbols.txt

Functions for managing buffer objects, including creation, mapping, naming, referencing, and prime handle operations.

```APIDOC
## Buffer Object Operations

### Description
Functions for managing buffer objects, including creation, mapping, naming, referencing, and prime handle operations.

### Functions
- **nouveau_bo_make_global**: Make a buffer object global.
- **nouveau_bo_map**: Map a buffer object into the client's address space.
- **nouveau_bo_name_get**: Get the name of a buffer object.
- **nouveau_bo_name_ref**: Reference a buffer object by its name.
- **nouveau_bo_new**: Create a new buffer object.
- **nouveau_bo_prime_handle_ref**: Reference a buffer object using a prime handle.
- **nouveau_bo_ref**: Reference a buffer object.
- **nouveau_bo_set_prime**: Set the prime capabilities of a buffer object.
- **nouveau_bo_wait**: Wait for a buffer object to become available.
- **nouveau_bo_wrap**: Wrap an existing buffer object.
```

--------------------------------

### GPU Management Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, deleting, and querying parameters of Etnaviv GPUs.

```APIDOC
## GPU Management Functions

### Description
Functions for managing Etnaviv GPU resources.

### Functions
- **etna_gpu_new**
- **etna_gpu_del**
- **etna_gpu_get_param**
```

--------------------------------

### General Tegra DRM Functions

Source: https://github.com/mesa/libdrm/blob/main/tegra/tegra-symbols.txt

Core functions for initializing, closing, and managing Tegra DRM contexts.

```APIDOC
## General Tegra DRM Functions

Core functions for interacting with the Tegra DRM device.

### `drm_tegra_close`

Closes the connection to the Tegra DRM device.

### `drm_tegra_new`

Initializes a new connection to the Tegra DRM device.
```

--------------------------------

### Command Stream Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, deleting, flushing, and finishing command streams, including timestamping and performance monitoring.

```APIDOC
## Command Stream Functions

### Description
Functions for managing and interacting with command streams.

### Functions
- **etna_cmd_stream_new**
- **etna_cmd_stream_del**
- **etna_cmd_stream_timestamp**
- **etna_cmd_stream_flush**
- **etna_cmd_stream_flush2**
- **etna_cmd_stream_finish**
- **etna_cmd_stream_perf**
- **etna_cmd_stream_reloc**
```

--------------------------------

### Performance Monitoring Functions

Source: https://github.com/mesa/libdrm/blob/main/etnaviv/etnaviv-symbols.txt

Functions for creating, deleting, and querying performance monitoring data.

```APIDOC
## Performance Monitoring Functions

### Description
Functions for creating, deleting, and querying performance monitoring resources.

### Functions
- **etna_perfmon_create**
- **etna_perfmon_del**
- **etna_perfmon_get_dom_by_name**
- **etna_perfmon_get_sig_by_name**
```

--------------------------------

### Core libdrm Functions

Source: https://github.com/mesa/libdrm/blob/main/core-symbols.txt

This section details various core functions for buffer management, context handling, device interaction, and memory operations within the libdrm library.

```APIDOC
## Core libdrm Functions

This section details various core functions for buffer management, context handling, device interaction, and memory operations within the libdrm library.

### Description
Provides fundamental operations for interacting with DRM devices, including buffer allocation, context management, device information retrieval, and memory mapping.

### Functions
- **drmAddBufs**: Add buffers to a drawable.
- **drmAddContextPrivateMapping**: Add a private mapping for a context.
- **drmAddContextTag**: Add a tag to a context.
- **drmAddMap**: Add a memory mapping.
- **drmAgpAcquire**: Acquire AGP resources.
- **drmAgpAlloc**: Allocate AGP memory.
- **drmAgpBase**: Get the AGP base address.
- **drmAgpBind**: Bind AGP memory.
- **drmAgpDeviceId**: Get the AGP device ID.
- **drmAgpEnable**: Enable AGP.
- **drmAgpFree**: Free AGP memory.
- **drmAgpGetMode**: Get the AGP mode.
- **drmAgpMemoryAvail**: Get available AGP memory.
- **drmAgpMemoryUsed**: Get used AGP memory.
- **drmAgpRelease**: Release AGP resources.
- **drmAgpSize**: Get the AGP size.
- **drmAgpUnbind**: Unbind AGP memory.
- **drmAgpVendorId**: Get the AGP vendor ID.
- **drmAgpVersionMajor**: Get the AGP major version.
- **drmAgpVersionMinor**: Get the AGP minor version.
- **drmAuthMagic**: Authenticate a magic number.
- **drmAvailable**: Check if DRM is available.
- **drmCheckModesettingSupported**: Check if modesetting is supported.
- **drmClose**: Close a DRM device.
- **drmCloseBufferHandle**: Close a buffer handle.
- **drmCloseOnce**: Close a DRM device once.
- **drmCommandNone**: No command.
- **drmCommandRead**: Read from a command buffer.
- **drmCommandWrite**: Write to a command buffer.
- **drmCommandWriteRead**: Write and read from a command buffer.
- **drmCreateContext**: Create a DRM context.
- **drmCreateDrawable**: Create a DRM drawable.
- **drmCrtcGetSequence**: Get CRTC sequence.
- **drmCrtcQueueSequence**: Queue CRTC sequence.
- **drmCtlInstHandler**: Install a control handler.
- **drmCtlUninstHandler**: Uninstall a control handler.
- **drmDelContextTag**: Delete a context tag.
- **drmDestroyContext**: Destroy a DRM context.
- **drmDestroyDrawable**: Destroy a DRM drawable.
- **drmDevicesEqual**: Check if two devices are equal.
- **drmDMA**: Perform a DMA operation.
- **drmDropMaster**: Drop master status.
- **drmError**: Get the last DRM error.
- **drmFinish**: Finish DRM operations.
- **drmFree**: Free DRM memory.
- **drmFreeBufs**: Free buffers.
- **drmFreeBusid**: Free a bus ID.
- **drmFreeDevice**: Free a DRM device.
- **drmFreeDevices**: Free DRM devices.
- **drmFreeReservedContextList**: Free reserved context list.
- **drmFreeVersion**: Free DRM version information.
- **drmGetBufInfo**: Get buffer information.
- **drmGetBusid**: Get the bus ID.
- **drmGetCap**: Get a DRM capability.
- **drmGetClient**: Get client information.
- **drmGetContextFlags**: Get context flags.
- **drmGetContextPrivateMapping**: Get private context mapping.
- **drmGetContextTag**: Get context tag.
- **drmGetDevice**: Get a DRM device.
- **drmGetDevice2**: Get a DRM device (version 2).
- **drmGetDeviceFromDevId**: Get a DRM device from device ID.
- **drmGetDeviceNameFromFd**: Get device name from file descriptor.
- **drmGetDeviceNameFromFd2**: Get device name from file descriptor (version 2).
- **drmGetDevices**: Get a list of DRM devices.
- **drmGetDevices2**: Get a list of DRM devices (version 2).
- **drmGetEntry**: Get a DRM device entry.
- **drmGetHashTable**: Get the hash table.
- **drmGetInterruptFromBusID**: Get interrupt from bus ID.
- **drmGetLibVersion**: Get the libdrm version.
- **drmGetLock**: Get a DRM lock.
- **drmGetMagic**: Get the magic number.
- **drmGetMap**: Get a memory mapping.
- **drmGetNodeTypeFromDevId**: Get node type from device ID.
- **drmGetNodeTypeFromFd**: Get node type from file descriptor.
- **drmGetPrimaryDeviceNameFromFd**: Get primary device name from file descriptor.
- **drmGetRenderDeviceNameFromFd**: Get render device name from file descriptor.
- **drmGetReservedContextList**: Get reserved context list.
- **drmGetStats**: Get DRM statistics.
- **drmGetVersion**: Get the DRM version.
- **drmHandleEvent**: Handle a DRM event.
- **drmHashCreate**: Create a hash table.
- **drmHashDelete**: Delete an entry from the hash table.
- **drmHashDestroy**: Destroy a hash table.
- **drmHashFirst**: Get the first entry in the hash table.
- **drmHashInsert**: Insert an entry into the hash table.
- **drmHashLookup**: Look up an entry in the hash table.
- **drmHashNext**: Get the next entry in the hash table.
- **drmIoctl**: Perform a generic DRM ioctl.
- **drmIsKMS**: Check if KMS is enabled.
- **drmIsMaster**: Check if the current client is master.
- **drmMalloc**: Allocate DRM memory.
- **drmMap**: Map memory.
- **drmMapBufs**: Map buffers.
- **drmMarkBufs**: Mark buffers.
- **drmMsg**: Send a DRM message.
- **drmOpen**: Open a DRM device.
- **drmOpenControl**: Open a DRM control device.
- **drmOpenOnce**: Open a DRM device once.
- **drmOpenOnceWithType**: Open a DRM device once with type.
- **drmOpenRender**: Open a DRM render device.
- **drmOpenWithType**: Open a DRM device with type.
- **drmPrimeFDToHandle**: Convert a prime FD to a handle.
- **drmPrimeHandleToFD**: Convert a handle to a prime FD.
- **drmRandom**: Generate random data.
- **drmRandomCreate**: Create a random number generator.
- **drmRandomDestroy**: Destroy a random number generator.
- **drmRandomDouble**: Generate a random double.
- **drmRmMap**: Remove a memory mapping.
- **drmScatterGatherAlloc**: Allocate scatter-gather memory.
- **drmScatterGatherFree**: Free scatter-gather memory.
- **drmSetBusid**: Set the bus ID.
- **drmSetClientCap**: Set a client capability.
- **drmSetContextFlags**: Set context flags.
- **drmSetInterfaceVersion**: Set the interface version.
- **drmSetMaster**: Set master status.
- **drmSetServerInfo**: Set server information.
- **drmSLCreate**: Create a sorted list.
- **drmSLDelete**: Delete an entry from a sorted list.
- **drmSLDestroy**: Destroy a sorted list.
- **drmSLDump**: Dump a sorted list.
- **drmSLFirst**: Get the first entry in a sorted list.
- **drmSLInsert**: Insert an entry into a sorted list.
- **drmSLLookup**: Look up an entry in a sorted list.
- **drmSLLookupNeighbors**: Look up neighbors in a sorted list.
- **drmSLNext**: Get the next entry in a sorted list.
- **drmSwitchToContext**: Switch to a DRM context.
- **drmSyncobjCreate**: Create a syncobj.
- **drmSyncobjDestroy**: Destroy a syncobj.
- **drmSyncobjEventfd**: Get an eventfd for a syncobj.
- **drmSyncobjExportSyncFile**: Export a syncobj to a sync file.
- **drmSyncobjFDToHandle**: Convert a sync file FD to a syncobj handle.
- **drmSyncobjHandleToFD**: Convert a syncobj handle to a sync file FD.
- **drmSyncobjImportSyncFile**: Import a sync file into a syncobj.
- **drmSyncobjQuery**: Query a syncobj.
- **drmSyncobjQuery2**: Query a syncobj (version 2).
- **drmSyncobjReset**: Reset a syncobj.
- **drmSyncobjSignal**: Signal a syncobj.
- **drmSyncobjTimelineSignal**: Signal a timeline syncobj.
- **drmSyncobjTimelineWait**: Wait on a timeline syncobj.
- **drmSyncobjTransfer**: Transfer a syncobj.
- **drmSyncobjWait**: Wait on a syncobj.
- **drmUnlock**: Unlock a DRM resource.
- **drmUnmap**: Unmap memory.
- **drmUnmapBufs**: Unmap buffers.
- **drmUpdateDrawableInfo**: Update drawable information.
- **drmWaitVBlank**: Wait for a vertical blank.
- **drmGetFormatModifierName**: Get the name of a format modifier.
- **drmGetFormatModifierVendor**: Get the vendor of a format modifier.
- **drmGetFormatName**: Get the name of a format.

```

--------------------------------

### Push Buffer Operations

Source: https://github.com/mesa/libdrm/blob/main/tegra/tegra-symbols.txt

Functions for managing push buffers, which are used to send commands to the hardware.

```APIDOC
## Push Buffer Operations

API for managing and interacting with push buffers.

### `drm_tegra_pushbuf_begin`

Begins a new push buffer sequence.

### `drm_tegra_pushbuf_end`

Ends the current push buffer sequence.

### `drm_tegra_pushbuf_relocate`

Relocates data within a push buffer.

### `drm_tegra_pushbuf_sync`

Synchronizes operations within a push buffer.

### `drm_tegra_pushbuf_sync_cond`

Performs conditional synchronization within a push buffer.

### `drm_tegra_pushbuf_wait`

Waits for push buffer operations to complete.
```

--------------------------------

### Find Suitable CRTC for Connector

Source: https://github.com/mesa/libdrm/blob/main/man/drm-kms.7.rst

This C function iterates through available encoders and CRTCs to find a CRTC that can drive a given connector. It requires the DRM device file descriptor, DRM resources, and connector information. Ensure the CRTC is not already in use by another connector if managing multiple displays.

```c
static int modeset_find_crtc(int fd, drmModeRes *res, drmModeConnector *conn)
{
    drmModeEncoder *enc;
    unsigned int i, j;

    /* iterate all encoders of this connector */
    for (i = 0; i < conn->count_encoders; ++i) {
        enc = drmModeGetEncoder(fd, conn->encoders[i]);
        if (!enc) {
            /* cannot retrieve encoder, ignoring... */
            continue;
        }

        /* iterate all global CRTCs */
        for (j = 0; j < res->count_crtcs; ++j) {
            /* check whether this CRTC works with the encoder */
            if (!(enc->possible_crtcs & (1 << j)))
                continue;


            /* Here you need to check that no other connector
             * currently uses the CRTC with id "crtc". If you intend
             * to drive one connector only, then you can skip this
             * step. Otherwise, simply scan your list of configured
             * connectors and CRTCs whether this CRTC is already
             * used. If it is, then simply continue the search here. */
            if (res->crtcs[j] "is unused") {
                drmModeFreeEncoder(enc);
                return res->crtcs[j];
            }
        }

        drmModeFreeEncoder(enc);
    }

    /* cannot find a suitable CRTC */
    return -ENOENT;
}
```

--------------------------------

### Buffer Object Operations

Source: https://github.com/mesa/libdrm/blob/main/radeon/radeon-symbols.txt

Functions for managing Radeon buffer objects (BOs), including debugging, retrieving information, mapping, referencing, and tiling.

```APIDOC
## Buffer Object Operations

### Description
Provides a set of functions to interact with Radeon buffer objects (BOs). These functions allow for debugging, retrieving buffer handles and properties, managing memory mapping, and controlling buffer references.

### Functions
- **radeon_bo_debug**: Debugging utility for buffer objects.
- **radeon_bo_get_handle**: Retrieves the handle for a buffer object.
- **radeon_bo_get_src_domain**: Gets the source domain of the buffer object.
- **radeon_bo_get_tiling**: Retrieves the tiling information for the buffer object.
- **radeon_bo_is_busy**: Checks if the buffer object is currently busy.
- **radeon_bo_is_referenced_by_cs**: Checks if the buffer object is referenced by a command submission.
- **radeon_bo_is_static**: Checks if the buffer object is static.
- **radeon_bo_map**: Maps the buffer object into the CPU's address space.
- **radeon_bo_open**: Opens a buffer object.
- **radeon_bo_ref**: Increments the reference count of a buffer object.
- **radeon_bo_set_tiling**: Sets the tiling information for the buffer object.
- **radeon_bo_unmap**: Unmaps the buffer object from the CPU's address space.
- **radeon_bo_unref**: Decrements the reference count of a buffer object.
- **radeon_bo_wait**: Waits for the buffer object to become idle.
```

--------------------------------

### DRM_IOCTL_MODE_CREATE_DUMB Structure

Source: https://github.com/mesa/libdrm/blob/main/man/drm-memory.7.rst

Defines the structure used to request the creation of a dumb buffer. Caller provides height, width, bpp, and flags; kernel returns handle, pitch, and size.

```c
struct drm_mode_create_dumb {
    __u32 height;
    __u32 width;
    __u32 bpp;
    __u32 flags;

    __u32 handle;
    __u32 pitch;
    __u64 size;
};

```

--------------------------------

### GEM Object OPEN Structure

Source: https://github.com/mesa/libdrm/blob/main/man/drm-memory.7.rst

Defines the structure used with the DRM_IOCTL_GEM_OPEN ioctl to import a GEM object by its name, retrieving a new handle and its size.

```c
struct drm_gem_open {
    __u32 name;

    __u32 handle;
    __u32 size;
};
```

--------------------------------

### DRM and Object Operations

Source: https://github.com/mesa/libdrm/blob/main/nouveau/nouveau-symbols.txt

Functions for interacting with Direct Rendering Manager (DRM) and managing objects.

```APIDOC
## DRM and Object Operations

### Description
Functions for interacting with Direct Rendering Manager (DRM) and managing objects.

### Functions
- **nouveau_drm_del**: Delete a DRM resource.
- **nouveau_drm_new**: Create a new DRM resource.
- **nouveau_getparam**: Get a parameter from the device.
- **nouveau_object_del**: Delete an object.
- **nouveau_object_mclass**: Get the class of an object.
- **nouveau_object_mthd**: Get the method of an object.
- **nouveau_object_new**: Create a new object.
- **nouveau_object_sclass_get**: Get the subclass of an object.
- **nouveau_object_sclass_put**: Put the subclass of an object.
- **nouveau_setparam**: Set a parameter for the device.
```

--------------------------------

### Configure Git for libdrm Patch Subject Prefix

Source: https://github.com/mesa/libdrm/blob/main/CONTRIBUTING.rst

Set a local Git configuration to automatically prefix libdrm patch subjects. This helps in easily identifying libdrm-related patches on the mailing list.

```bash
git config --local format.subjectprefix "PATCH libdrm"
```

--------------------------------

### drmHandleEvent

Source: https://github.com/mesa/libdrm/blob/main/man/drmHandleEvent.3.rst

Handles pending DRM events by reading them from the file descriptor and dispatching them to registered handlers via the drmEventContext structure.

```APIDOC
## drmHandleEvent

### Description
Processes outstanding DRM events on the DRM file-descriptor passed as `fd`. This function should be called after the DRM file-descriptor has polled readable; it will read the events and use the passed-in `evctx` structure to call function pointers.

### Function Signature
```c
int drmHandleEvent(int fd, drmEventContextPtr evctx);
```

### Parameters
#### Path Parameters
- **fd** (int) - The DRM file descriptor.
- **evctx** (drmEventContextPtr) - A pointer to a `drmEventContext` structure containing handler function pointers.

### Return Value
- **0** on success, or if there is no data to read from the file-descriptor.
- **-1** if the read on the file-descriptor fails or returns less than a full event record.

### Event Context Structure
```c
typedef struct _drmEventContext {
    int version;
    void (*vblank_handler) (int fd, unsigned int sequence, unsigned int tv_sec, unsigned int tv_usec, void *user_data);
    void (*page_flip_handler) (int fd, unsigned int sequence, unsigned int tv_sec, unsigned int tv_usec, void *user_data);
} drmEventContext, *drmEventContextPtr;
```
```

--------------------------------

### Mode Setting Functions

Source: https://github.com/mesa/libdrm/blob/main/core-symbols.txt

This section covers functions related to DRM Mode Setting, including framebuffer management, connector, CRTC, and plane operations.

```APIDOC
## Mode Setting Functions

This section covers functions related to DRM Mode Setting, including framebuffer management, connector, CRTC, and plane operations.

### Description
Provides a comprehensive set of functions for managing display modes, framebuffers, connectors, CRTCs, and planes, enabling dynamic control over display output.

### Functions
- **drmModeAddFB**: Add a framebuffer.
- **drmModeAddFB2**: Add a framebuffer (version 2).
- **drmModeAddFB2WithModifiers**: Add a framebuffer with modifiers.
- **drmModeAtomicAddProperty**: Add a property to an atomic commit.
- **drmModeAtomicAlloc**: Allocate an atomic commit.
- **drmModeAtomicCommit**: Commit an atomic configuration.
- **drmModeAtomicDuplicate**: Duplicate an atomic configuration.
- **drmModeAtomicFree**: Free an atomic configuration.
- **drmModeAtomicGetCursor**: Get cursor information from an atomic commit.
- **drmModeAtomicMerge**: Merge atomic configurations.
- **drmModeAtomicSetCursor**: Set cursor information in an atomic commit.
- **drmModeAttachMode**: Attach a mode to a connector.
- **drmModeCloseFB**: Close a framebuffer.
- **drmModeConnectorGetPossibleCrtcs**: Get possible CRTCs for a connector.
- **drmModeConnectorSetProperty**: Set a property on a connector.
- **drmModeCreateDumbBuffer**: Create a dumb buffer.
- **drmModeCreateLease**: Create a lease.
- **drmModeCreatePropertyBlob**: Create a property blob.
- **drmModeCrtcGetGamma**: Get CRTC gamma settings.
- **drmModeCrtcSetGamma**: Set CRTC gamma settings.
- **drmModeDestroyDumbBuffer**: Destroy a dumb buffer.
- **drmModeDestroyPropertyBlob**: Destroy a property blob.
- **drmModeDetachMode**: Detach a mode from a connector.
- **drmModeDirtyFB**: Mark a framebuffer as dirty.
- **drmModeFormatModifierBlobIterNext**: Iterate through format modifier blobs.
- **drmModeFreeConnector**: Free connector information.
- **drmModeFreeCrtc**: Free CRTC information.
- **drmModeFreeEncoder**: Free encoder information.
- **drmModeFreeFB**: Free framebuffer information.
- **drmModeFreeFB2**: Free framebuffer information (version 2).
- **drmModeFreeModeInfo**: Free mode information.
- **drmModeFreeObjectProperties**: Free object properties.
- **drmModeFreePlane**: Free plane information.
- **drmModeFreePlaneResources**: Free plane resources.
- **drmModeFreeProperty**: Free property information.
- **drmModeFreePropertyBlob**: Free property blob.
- **drmModeFreeResources**: Free resources.
- **drmModeGetConnector**: Get connector information.
- **drmModeGetConnectorCurrent**: Get the current connector information.
- **drmModeGetConnectorTypeName**: Get the type name of a connector.
- **drmModeGetCrtc**: Get CRTC information.
- **drmModeGetEncoder**: Get encoder information.
- **drmModeGetFB**: Get framebuffer information.
- **drmModeGetFB2**: Get framebuffer information (version 2).
- **drmModeGetLease**: Get lease information.
- **drmModeGetPlane**: Get plane information.
- **drmModeGetPlaneResources**: Get plane resources.
- **drmModeGetProperty**: Get property information.
- **drmModeGetPropertyBlob**: Get property blob.
- **drmModeGetResources**: Get DRM resources.
- **drmModeListLessees**: List lessees.
- **drmModeMapDumbBuffer**: Map a dumb buffer.
- **drmModeMoveCursor**: Move the cursor.
- **drmModeObjectGetProperties**: Get properties of an object.
- **drmModeObjectSetProperty**: Set a property on an object.
- **drmModePageFlip**: Perform a page flip.
- **drmModePageFlipTarget**: Perform a page flip with target.
- **drmModeRevokeLease**: Revoke a lease.
- **drmModeRmFB**: Remove a framebuffer.
- **drmModeSetCrtc**: Set CRTC settings.
- **drmModeSetCursor**: Set the cursor.
- **drmModeSetCursor2**: Set the cursor (version 2).
- **drmModeSetPlane**: Set plane settings.

```