mirror-linux/drivers/accel/habanalabs/Kconfig

53 lines
1.4 KiB
Plaintext

# SPDX-License-Identifier: GPL-2.0-only
#
# HabanaLabs AI accelerators driver
#
config DRM_ACCEL_HABANALABS
tristate "HabanaLabs AI accelerators"
depends on DRM_ACCEL
depends on X86 && X86_64
depends on PCI && HAS_IOMEM
select GENERIC_ALLOCATOR
select HWMON
select DMA_SHARED_BUFFER
select CRC32
select FW_LOADER
help
Enables PCIe card driver for Habana's AI Processors (AIP) that are
designed to accelerate Deep Learning inference and training workloads.
The driver manages the PCIe devices and provides IOCTL interface for
the user to submit workloads to the devices.
The user-space interface is described in
include/uapi/drm/habanalabs_accel.h
If unsure, say N.
To compile this driver as a module, choose M here: the
module will be called habanalabs.
if DRM_ACCEL_HABANALABS
config HL_HLDIO
bool "Habanalabs NVMe Direct I/O (HLDIO)"
depends on PCI_P2PDMA
depends on BLOCK
help
Enable NVMe peer-to-peer direct I/O support for Habanalabs AI
accelerators.
This allows direct data transfers between NVMe storage devices
and Habanalabs accelerators without involving system memory,
using PCI peer-to-peer DMA capabilities.
Requirements:
- CONFIG_PCI_P2PDMA=y
- NVMe device and Habanalabs accelerator under same PCI root complex
- IOMMU disabled or in passthrough mode
- Hardware supporting PCI P2P DMA
If unsure, say N
endif # DRM_ACCEL_HABANALABS