From: Joel Fernandes <joelagnelf@nvidia.com>
To: linux-kernel@vger.kernel.org
Cc: Miguel Ojeda <ojeda@kernel.org>, Boqun Feng <boqun@kernel.org>,
Gary Guo <gary@garyguo.net>,
Bjorn Roy Baron <bjorn3_gh@protonmail.com>,
Benno Lossin <lossin@kernel.org>,
Andreas Hindborg <a.hindborg@kernel.org>,
Alice Ryhl <aliceryhl@google.com>,
Trevor Gross <tmgross@umich.edu>,
Danilo Krummrich <dakr@kernel.org>,
Dave Airlie <airlied@redhat.com>,
Daniel Almeida <daniel.almeida@collabora.com>,
Koen Koning <koen.koning@linux.intel.com>,
dri-devel@lists.freedesktop.org, rust-for-linux@vger.kernel.org,
Nikola Djukic <ndjukic@nvidia.com>,
Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
Maxime Ripard <mripard@kernel.org>,
Thomas Zimmermann <tzimmermann@suse.de>,
David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
Jonathan Corbet <corbet@lwn.net>,
Alex Deucher <alexander.deucher@amd.com>,
Christian Koenig <christian.koenig@amd.com>,
Jani Nikula <jani.nikula@linux.intel.com>,
Joonas Lahtinen <joonas.lahtinen@linux.intel.com>,
Rodrigo Vivi <rodrigo.vivi@intel.com>,
Tvrtko Ursulin <tursulin@ursulin.net>,
Huang Rui <ray.huang@amd.com>,
Matthew Auld <matthew.auld@intel.com>,
Lucas De Marchi <lucas.demarchi@intel.com>,
Thomas Hellstrom <thomas.hellstrom@linux.intel.com>,
Helge Deller <deller@gmx.de>, Alex Gaynor <alex.gaynor@gmail.com>,
Boqun Feng <boqun.feng@gmail.com>,
John Hubbard <jhubbard@nvidia.com>,
Alistair Popple <apopple@nvidia.com>,
Timur Tabi <ttabi@nvidia.com>, Edwin Peer <epeer@nvidia.com>,
Alexandre Courbot <acourbot@nvidia.com>,
Andrea Righi <arighi@nvidia.com>,
Andy Ritger <aritger@nvidia.com>, Zhi Wang <zhiw@nvidia.com>,
Balbir Singh <balbirs@nvidia.com>,
Philipp Stanner <phasta@kernel.org>,
Elle Rhumsaa <elle@weathered-steel.dev>,
alexeyi@nvidia.com, Eliot Courtney <ecourtney@nvidia.com>,
joel@joelfernandes.org, linux-doc@vger.kernel.org,
amd-gfx@lists.freedesktop.org, intel-gfx@lists.freedesktop.org,
intel-xe@lists.freedesktop.org, linux-fbdev@vger.kernel.org
Subject: Re: [PATCH v11 07/20] gpu: nova-core: mm: Add TLB flush support
Date: Thu, 16 Apr 2026 17:23:12 -0400 [thread overview]
Message-ID: <20260416212312.GA667928@joelbox2> (raw)
In-Reply-To: <20260415210548.3776595-7-joelagnelf@nvidia.com>
On Wed, Apr 15, 2026 at 05:05:34PM -0400, Joel Fernandes wrote:
> Add TLB (Translation Lookaside Buffer) flush support for GPU MMU.
>
> After modifying page table entries, the GPU's TLB must be invalidated
> to ensure the new mappings take effect. The Tlb struct provides flush
> functionality through BAR0 registers.
>
> The flush operation writes the page directory base address and triggers
> an invalidation, polling for completion with a 2 second timeout matching
> the Nouveau driver.
>
> Cc: Nikola Djukic <ndjukic@nvidia.com>
> Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
> ---
> drivers/gpu/nova-core/mm.rs | 1 +
> drivers/gpu/nova-core/mm/tlb.rs | 97 +++++++++++++++++++++++++++++++++
> drivers/gpu/nova-core/regs.rs | 44 +++++++++++++++
> 3 files changed, 142 insertions(+)
> create mode 100644 drivers/gpu/nova-core/mm/tlb.rs
>
> diff --git a/drivers/gpu/nova-core/mm.rs b/drivers/gpu/nova-core/mm.rs
> index fa29f525f282..314d660d898b 100644
> --- a/drivers/gpu/nova-core/mm.rs
> +++ b/drivers/gpu/nova-core/mm.rs
> @@ -25,6 +25,7 @@ fn from(pfn: Pfn) -> Self {
> }
>
> pub(crate) mod pramin;
> +pub(super) mod tlb;
>
> use kernel::{
> bitfield,
> diff --git a/drivers/gpu/nova-core/mm/tlb.rs b/drivers/gpu/nova-core/mm/tlb.rs
> new file mode 100644
> index 000000000000..6d384f447635
> --- /dev/null
> +++ b/drivers/gpu/nova-core/mm/tlb.rs
> @@ -0,0 +1,97 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +//! TLB (Translation Lookaside Buffer) flush support for GPU MMU.
> +//!
> +//! After modifying page table entries, the GPU's TLB must be flushed to
> +//! ensure the new mappings take effect. This module provides TLB flush
> +//! functionality for virtual memory managers.
> +//!
> +//! # Examples
> +//!
> +//! ```ignore
> +//! use crate::mm::tlb::Tlb;
> +//!
> +//! fn page_table_update(tlb: &Tlb, pdb_addr: VramAddress) -> Result<()> {
> +//! // ... modify page tables ...
> +//!
> +//! // Flush TLB to make changes visible (polls for completion).
> +//! tlb.flush(pdb_addr)?;
> +//!
> +//! Ok(())
> +//! }
> +//! ```
> +
> +use kernel::{
> + devres::Devres,
> + io::poll::read_poll_timeout,
> + io::Io,
> + new_mutex,
> + prelude::*,
> + sync::{
> + Arc,
> + Mutex, //
> + },
> + time::Delta, //
> +};
> +
> +use crate::{
> + driver::Bar0,
> + mm::VramAddress,
> + regs, //
> +};
> +
> +/// TLB manager for GPU translation buffer operations.
> +#[pin_data]
> +pub(crate) struct Tlb {
> + bar: Arc<Devres<Bar0>>,
> + /// TLB flush serialization lock: This lock is designed to be acquired during
> + /// the DMA fence signalling critical path. It should NEVER be held across any
> + /// reclaimable CPU memory allocations because the memory reclaim path can
> + /// call `dma_fence_wait()` (when implemented), which would deadlock if lock held.
> + #[pin]
> + lock: Mutex<()>,
> +}
> +
> +impl Tlb {
> + /// Create a new TLB manager.
> + pub(super) fn new(bar: Arc<Devres<Bar0>>) -> impl PinInit<Self> {
> + pin_init!(Self {
> + bar,
> + lock <- new_mutex!((), "tlb_flush"),
> + })
> + }
> +
> + /// Flush the GPU TLB for a specific page directory base.
> + ///
> + /// This invalidates all TLB entries associated with the given PDB address.
> + /// Must be called after modifying page table entries to ensure the GPU sees
> + /// the updated mappings.
> + pub(super) fn flush(&self, pdb_addr: VramAddress) -> Result {
> + let _guard = self.lock.lock();
> +
> + let bar = self.bar.try_access().ok_or(ENODEV)?;
> +
> + // Write PDB address.
> + bar.write_reg(regs::NV_TLB_FLUSH_PDB_LO::from_pdb_addr(pdb_addr.raw_u64()));
> + bar.write_reg(regs::NV_TLB_FLUSH_PDB_HI::from_pdb_addr(pdb_addr.raw_u64()));
> +
> + // Trigger flush: invalidate all pages, require global acknowledgment
> + // from all engines before completion.
> + bar.write_reg(
> + regs::NV_TLB_FLUSH_CTRL::zeroed()
> + .with_page_all(true)
> + .with_ack_globally(true)
> + .with_enable(true),
> + );
> +
> + // Poll for completion - enable bit clears when flush is done.
> + read_poll_timeout(
> + || Ok(bar.read(regs::NV_TLB_FLUSH_CTRL)),
> + |ctrl: ®s::NV_TLB_FLUSH_CTRL| !ctrl.enable(),
> + Delta::ZERO,
> + Delta::from_secs(2),
> + )?;
> +
> + Ok(())
> + }
> +}
Btw, I changed this to doing it in 2 phases to avoid holding the RCU readlock
across read_poll_timeout, which can sleep. Will squash it in for v12.
---8<-----------------------
diff --git a/drivers/gpu/nova-core/mm/tlb.rs b/drivers/gpu/nova-core/mm/tlb.rs
index 6d384f447635..3a65db7d9a1a 100644
--- a/drivers/gpu/nova-core/mm/tlb.rs
+++ b/drivers/gpu/nova-core/mm/tlb.rs
@@ -69,24 +69,35 @@ pub(super) fn new(bar: Arc<Devres<Bar0>>) -> impl PinInit<Self> {
pub(super) fn flush(&self, pdb_addr: VramAddress) -> Result {
let _guard = self.lock.lock();
- let bar = self.bar.try_access().ok_or(ENODEV)?;
+ // Broken into 2 phases with scopes (Write and Poll) to avoid holding
+ // RevecablableGuard (and hence RCU read-side critical section) across
+ // the read_poll_timeout() call that can sleep.
- // Write PDB address.
- bar.write_reg(regs::NV_TLB_FLUSH_PDB_LO::from_pdb_addr(pdb_addr.raw_u64()));
- bar.write_reg(regs::NV_TLB_FLUSH_PDB_HI::from_pdb_addr(pdb_addr.raw_u64()));
+ // Write phase — hold bar access briefly for register writes only.
+ {
+ let bar = self.bar.try_access().ok_or(ENODEV)?;
- // Trigger flush: invalidate all pages, require global acknowledgment
- // from all engines before completion.
- bar.write_reg(
- regs::NV_TLB_FLUSH_CTRL::zeroed()
- .with_page_all(true)
- .with_ack_globally(true)
- .with_enable(true),
- );
+ // Write PDB address.
+ bar.write_reg(regs::NV_TLB_FLUSH_PDB_LO::from_pdb_addr(pdb_addr.raw_u64()));
+ bar.write_reg(regs::NV_TLB_FLUSH_PDB_HI::from_pdb_addr(pdb_addr.raw_u64()));
- // Poll for completion - enable bit clears when flush is done.
+ // Trigger flush: invalidate all pages, require global acknowledgment
+ // from all engines before completion.
+ bar.write_reg(
+ regs::NV_TLB_FLUSH_CTRL::zeroed()
+ .with_page_all(true)
+ .with_ack_globally(true)
+ .with_enable(true),
+ );
+ }
+
+ // Poll for completion — re-acquire bar access each iteration to avoid
+ // holding the RCU read-side lock (via RevocableGuard) across sleep.
read_poll_timeout(
- || Ok(bar.read(regs::NV_TLB_FLUSH_CTRL)),
+ || {
+ let bar = self.bar.try_access().ok_or(ENODEV)?;
+ Ok(bar.read(regs::NV_TLB_FLUSH_CTRL))
+ },
|ctrl: ®s::NV_TLB_FLUSH_CTRL| !ctrl.enable(),
Delta::ZERO,
Delta::from_secs(2),
--
2.34.1
next prev parent reply other threads:[~2026-04-16 21:23 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-15 21:05 [PATCH v11 01/20] gpu: nova-core: gsp: Return GspStaticInfo from boot() Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 02/20] gpu: nova-core: gsp: Extract usable FB region from GSP Joel Fernandes
2026-04-16 23:04 ` John Hubbard
2026-04-16 23:26 ` John Hubbard
2026-04-15 21:05 ` [PATCH v11 03/20] gpu: nova-core: gsp: Expose total physical VRAM end from FB region info Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 04/20] gpu: nova-core: mm: Add support to use PRAMIN windows to write to VRAM Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 05/20] docs: gpu: nova-core: Document the PRAMIN aperture mechanism Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 06/20] gpu: nova-core: mm: Add common memory management types Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 07/20] gpu: nova-core: mm: Add TLB flush support Joel Fernandes
2026-04-16 21:23 ` Joel Fernandes [this message]
2026-04-16 21:45 ` Danilo Krummrich
2026-04-16 22:18 ` Joel Fernandes
2026-04-16 22:53 ` Danilo Krummrich
2026-04-15 21:05 ` [PATCH v11 08/20] gpu: nova-core: mm: Add GpuMm centralized memory manager Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 09/20] gpu: nova-core: mm: Add common types for all page table formats Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 10/20] gpu: nova-core: mm: Add MMU v2 page table types Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 11/20] gpu: nova-core: mm: Add MMU v3 " Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 12/20] gpu: nova-core: mm: Add unified page table entry wrapper enums Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 13/20] gpu: nova-core: mm: Add page table walker for MMU v2/v3 Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 14/20] gpu: nova-core: mm: Add Virtual Memory Manager Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 15/20] gpu: nova-core: mm: Add virtual address range tracking to VMM Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 16/20] gpu: nova-core: mm: Add multi-page mapping API " Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 17/20] gpu: nova-core: Add BAR1 aperture type and size constant Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 18/20] gpu: nova-core: mm: Add BAR1 user interface Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 19/20] gpu: nova-core: mm: Add BAR1 memory management self-tests Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 20/20] gpu: nova-core: mm: Add PRAMIN aperture self-tests Joel Fernandes
2026-04-15 21:05 ` [PATCH v11 00/21] gpu: nova-core: Add memory management support Joel Fernandes
2026-04-16 22:57 ` [PATCH v11 01/20] gpu: nova-core: gsp: Return GspStaticInfo from boot() John Hubbard
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260416212312.GA667928@joelbox2 \
--to=joelagnelf@nvidia.com \
--cc=a.hindborg@kernel.org \
--cc=acourbot@nvidia.com \
--cc=airlied@gmail.com \
--cc=airlied@redhat.com \
--cc=alex.gaynor@gmail.com \
--cc=alexander.deucher@amd.com \
--cc=alexeyi@nvidia.com \
--cc=aliceryhl@google.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=apopple@nvidia.com \
--cc=arighi@nvidia.com \
--cc=aritger@nvidia.com \
--cc=balbirs@nvidia.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun.feng@gmail.com \
--cc=boqun@kernel.org \
--cc=christian.koenig@amd.com \
--cc=corbet@lwn.net \
--cc=dakr@kernel.org \
--cc=daniel.almeida@collabora.com \
--cc=deller@gmx.de \
--cc=dri-devel@lists.freedesktop.org \
--cc=ecourtney@nvidia.com \
--cc=elle@weathered-steel.dev \
--cc=epeer@nvidia.com \
--cc=gary@garyguo.net \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=jani.nikula@linux.intel.com \
--cc=jhubbard@nvidia.com \
--cc=joel@joelfernandes.org \
--cc=joonas.lahtinen@linux.intel.com \
--cc=koen.koning@linux.intel.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-fbdev@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lossin@kernel.org \
--cc=lucas.demarchi@intel.com \
--cc=maarten.lankhorst@linux.intel.com \
--cc=matthew.auld@intel.com \
--cc=mripard@kernel.org \
--cc=ndjukic@nvidia.com \
--cc=ojeda@kernel.org \
--cc=phasta@kernel.org \
--cc=ray.huang@amd.com \
--cc=rodrigo.vivi@intel.com \
--cc=rust-for-linux@vger.kernel.org \
--cc=simona@ffwll.ch \
--cc=thomas.hellstrom@linux.intel.com \
--cc=tmgross@umich.edu \
--cc=ttabi@nvidia.com \
--cc=tursulin@ursulin.net \
--cc=tzimmermann@suse.de \
--cc=zhiw@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox