kernel/git/next/linux-next.git/Documentation/gpu/nova, branch master

docs: gpu: nova-core: Document the PRAMIN aperture mechanism

2026-09-07T02:55:43+00:00

Add documentation for the PRAMIN aperture mechanism used by nova-core for direct VRAM access. Nova only uses TARGET=VRAM for VRAM access. The SYS_MEM target values are documented for completeness but not used by the driver. Signed-off-by: Joel Fernandes [ecourtney: fix the TARGET encodings, KiB/MiB units, wording nits] [ecourtney: wrap at 80 columns] Signed-off-by: Eliot Courtney Acked-by: Danilo Krummrich Link: https://patch.msgid.link/20260827-pramin-split-v3-7-24b24d7afc52@nvidia.com Signed-off-by: Alexandre Courbot

gpu: nova-core: add TLV parser for firmware files

2026-08-05T23:54:12+00:00

TLV (type, length, value) files are the new image format used by Nova to encapsulate firmware images and their metadata. Unlike the firmware files for previous versions of the firmware, TLV filenames are not versioned, and they have a .tlv suffix. Add function request_tlv() to load TLV firmware images. Add the Tlv struct and supporting types for parsing TLV firmware images. TLV files begin with a 4-byte magic header, which must be "NVFW" for Nvidia firmware files. This is followed by a sequence of blocks each containing a 4-byte ASCII tag, a 4-byte little-endian length, and a payload padded to a 4-byte boundary. Tlv::new() validates the entire image up front, so that the iterator can subsequently yield blocks without fallible parsing. Also add accessor methods for the various encoded types that will be used by the driver. Signed-off-by: Timur Tabi Reviewed-by: Alexandre Courbot Tested-by: Alexandre Courbot Link: https://patch.msgid.link/20260731201017.2580713-4-ttabi@nvidia.com [ Drop unnecessary payload.is_empty() check in Tlv::new(), use EINVAL instead of ENODATA in Tlv::get_bytes() and add a corresponding TODO comment. - Danilo ] Signed-off-by: Danilo Krummrich

gpu: nova-core: add FSP and PRC protocol documentation

2026-06-17T12:30:17+00:00

Add documentation for the Foundation Security Processor (FSP) interface covering the simplified Hopper/Blackwell boot flow, the Chain of Trust (COT) message protocol, the MCTP/NVDM message format, and the Product Reconfiguration Control (PRC) protocol used to query device configuration knobs such as vGPU mode. Signed-off-by: Zhi Wang Link: https://patch.msgid.link/20260604114339.1565660-6-zhiw@nvidia.com Signed-off-by: Alexandre Courbot

gpu: nova-core: vbios: move constants and functions to be associated

2026-05-25T18:18:08+00:00

Move constants and functions to be inside the impls of the types they are related to. This makes it more obvious what each type and value is for. Signed-off-by: Eliot Courtney Link: https://patch.msgid.link/20260525-fix-vbios-v5-21-e5e455251537@nvidia.com Signed-off-by: Danilo Krummrich

Documentation: gpu: nova: document the IFR header layout

2026-04-28T23:13:55+00:00

Init-from-ROM (IFR) is a special GPU feature used for power management on some Nvidia GPUs. It references data in the VBIOS for its operation, but for drivers the important piece is the header that precedes the VBIOS PCI Expansion ROM. Parsing VBIOS is necessary to boot GSP-RM on Turing, Ampere, and Ada GPUs. Most such GPUs do not need to parse the IFR header in order to find the VBIOS, but the Nvidia GA100 is the exception. GA100 lacks a display engine, so the PRAMIN method (which reads the VBIOS from VRAM via display hardware) is unavailable, forcing the driver to read the ROM directly via PROM. On other similar GPUs, either PRAMIN succeeds before PROM is tried, or the IFR hardware has already applied the ROM offset so that PROM reads transparently skip the IFR header. Note that GH100 also does not have a display engine, but it uses a completely different method to boot GSP-RM. This information is derived from NVIDIA's open-source GPU kernel module driver (aka OpenRM), specifically the NV_PBUS_IFR_FMT_FIXEDx definitions in dev_bus.h and the parsing logic in s_romImgFindPciHeader_TU102(). Signed-off-by: Timur Tabi Reviewed-by: Joel Fernandes Link: https://patch.msgid.link/20260408142132.3911466-1-ttabi@nvidia.com [acourbot: fix TOTAL_DATA_SIZE definition.] Signed-off-by: Alexandre Courbot

Documentation: nova: remove register abstraction task

2026-03-26T06:10:29+00:00

The `register!` macro has been implemented and all nova-core code converted to use it. Remove the corresponding task in todo.rst. Reviewed-by: Eliot Courtney Reviewed-by: Gary Guo Acked-by: Danilo Krummrich Link: https://patch.msgid.link/20260325-b4-nova-register-v4-10-bdf172f0f6ca@nvidia.com Signed-off-by: Alexandre Courbot

Documentation: nova: update pending tasks

2026-01-26T17:16:14+00:00

Some tasks, while already having some progress, are still pending, hence update them. Link: https://patch.msgid.link/20260119153129.59876-2-dakr@kernel.org [ Remove PCI config space, as it is completed. - Danilo ] Signed-off-by: Danilo Krummrich

Documentation: nova: remove completed tasks

2026-01-26T17:06:00+00:00

Remove completed tasks from Nova's task list. Link: https://patch.msgid.link/20260119153129.59876-1-dakr@kernel.org Signed-off-by: Danilo Krummrich

Documentation: nova: Update the todo list

2025-11-07T14:10:44+00:00

This small patch updates the nova todo list to remove some tasks that have been solved lately: * COHA is solved in this patch series * TRSM was solved recently [1] [1] https://lore.kernel.org/rust-for-linux/DCEJ9SV4LBJL.11EUZVXX6EB9H@nvidia.com/ Signed-off-by: Daniel del Castillo [acourbot@nvidia.com: set prefix to "Documentation: nova:".] Signed-off-by: Alexandre Courbot Message-ID: <20251104193756.57726-4-delcastillodelarosadaniel@gmail.com>

gpu: nova-core: replace wait_on with kernel equivalents

2025-10-25T04:14:21+00:00

wait_on was a temporary helper function waiting for a kernel crate equivalent. Now that read_poll_timeout and fsleep are available, use them and remove wait_on. Acked-by: Danilo Krummrich Signed-off-by: Alexandre Courbot Message-ID: <20251020-nova_wait_on-v1-1-2eb87fb38d14@nvidia.com>