Bug#1079170: linux [amd64]: Enable DRM_ACCEL_HABANALABS and DRM_ACCEL_IVPU
Source: linux
Version: 6.10.4-1
Severity: wishlist
Tags: patch
X-Debbugs-Cc: jair.gonzalez@linux.intel.com,
miguel.bernal.marin@linux.intel.com
Dear Maintainer,
Please enable the HabanaLabs AI accelerators (DRM_ACCEL_HABANALABS)
and the Intel NPU (Neural Processing Unit) (DRM_ACCEL_IVPU) drivers.
They both depend on the Compute Acceleration Framework (DRM_ACCEL).
Habana's AI Processors (AIP) are designed to accelerate Deep Learning
inference and training workloads. The driver manages the PCIe devices
and provides the IOCTL interface for the user to submit workloads to
the devices. The module is defined in drivers/accel/habanalabs/Kconfig,
while the user-space interface is described in
include/uapi/drm/habanalabs_accel.h
Intel NPU (formerly called Intel VPU) is a CPU-integrated inference
accelerator for Computer Vision and Deep Learning applications.
This option is useful for systems with a 14th generation Intel CPU
(Meteor Lake) or newer. The module is defined in
drivers/accel/ivpu/Kconfig.
The Compute Accelerator Framework targets device drivers of compute
acceleration devices, such as, but not limited to, Machine-Learning
and Deep-Learning acceleration devices. This framework is integrated
with the DRM subsystem as compute accelerators and GPUs share a lot in
common and can use almost the same infrastructure code. Having said
that, acceleration devices will have a different major number than
GPUs, and will be exposed to user-space using different device files,
called accel/accel* (in /dev, sysfs and debugfs).
A MR was created with this proposal at:
https://salsa.debian.org/kernel-team/linux/-/merge_requests/1165
Thanks,
Jair Gonzalez
Reply to: