On 3/3/26 14:34, Thomas Hellström wrote:
> GPU use-cases for mmu_interval_notifiers with hmm often involve
> starting a gpu operation and then waiting for it to complete.
> These operations are typically context preemption or TLB flushing.
>
> With single-pass notifiers per GPU this doesn't scale in
> multi-gpu scenarios. In those scenarios we'd want to first start
> preemption- or TLB flushing on all GPUs and as a second pass wait
> for them to complete.
>
> One can do this on per-driver basis multiplexing per-driver
> notifiers but that would mean sharing the notifier "user" lock
> across all GPUs and that doesn't scale well either, so adding support
> for multi-pass in the core appears to be the right choice.
>
> Implement two-pass capability in the mmu_interval_notifier. Use a
> linked list for the final passes to minimize the impact for
> use-cases that don't need the multi-pass functionality by avoiding
> a second interval tree walk, and to be able to easily pass data
> between the two passes.
>
> v1:
> - Restrict to two passes (Jason Gunthorpe)
> - Improve on documentation (Jason Gunthorpe)
> - Improve on function naming (Alistair Popple)
> v2:
> - Include the invalidate_finish() callback in the
> struct mmu_interval_notifier_ops.
> - Update documentation (GitHub Copilot:claude-sonnet-4.6)
> - Use lockless list for list management.
> v3:
> - Update kerneldoc for the struct mmu_interval_notifier_finish::list member
> (Matthew Brost)
> - Add a WARN_ON_ONCE() checking for NULL invalidate_finish() op if
> if invalidate_start() is non-NULL. (Matthew Brost)
>
> Cc: Jason Gunthorpe <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Simona Vetter <[email protected]>
> Cc: Dave Airlie <[email protected]>
> Cc: Alistair Popple <[email protected]>
> Cc: <[email protected]>
> Cc: <[email protected]>
> Cc: <[email protected]>
>
> Assisted-by: GitHub Copilot:claude-sonnet-4.6 # Documentation only.
> Signed-off-by: Thomas Hellström <[email protected]>
> ---
> include/linux/mmu_notifier.h | 38 +++++++++++++++++++++
> mm/mmu_notifier.c | 65 +++++++++++++++++++++++++++++++-----
> 2 files changed, 94 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
> index 07a2bbaf86e9..37b683163235 100644
> --- a/include/linux/mmu_notifier.h
> +++ b/include/linux/mmu_notifier.h
> @@ -233,16 +233,54 @@ struct mmu_notifier {
> unsigned int users;
> };
>
> +/**
> + * struct mmu_interval_notifier_finish - mmu_interval_notifier two-pass
> abstraction
> + * @link: Lockless list link for the notifiers pending pass list
> + * @notifier: The mmu_interval_notifier for which the finish pass is called.
> + *
> + * Allocate, typically using GFP_NOWAIT in the interval notifier's first
> pass.
Might want to make it clear that the fist pass is "start" and the second
pass is "finish".
Two-pass makes it sound like we'd be calling the same operation (e.g.,
invalidate() ) twice.
> + * If allocation fails (which is not unlikely under memory pressure), fall
> back
> + * to single-pass operation.
Do you mean that the core will fallback (calling invalidate() ) or that
it's the responsibility of the notifier to behave as if invalidate()
would be called to then return finish=NULL? I assume the latter.
Maybe this should be documented for @invalidate_start instead. (behave
like invalidate() if @finish is %NULL on return etc)
> Note that with a large number of notifiers
> + * implementing two passes, allocation with GFP_NOWAIT will become
> increasingly
> + * likely to fail, so consider implementing a small pool instead of using
> + * kmalloc() allocations.
> + *
> + * If the implementation needs to pass data between the two passes,
> + * the recommended way is to embed struct mmu_interval_notifier_finish into
> a larger
> + * structure that also contains the data needed to be shared. Keep in mind
> that
> + * a notifier callback can be invoked in parallel, and each invocation needs
> its
> + * own struct mmu_interval_notifier_finish.
> + */
> +struct mmu_interval_notifier_finish {
> + struct llist_node link;
> + struct mmu_interval_notifier *notifier;
> +};
> +
> /**
> * struct mmu_interval_notifier_ops
> * @invalidate: Upon return the caller must stop using any SPTEs within this
> * range. This function can sleep. Return false only if sleeping
> * was required but mmu_notifier_range_blockable(range) is
> false.
> + * @invalidate_start: Similar to @invalidate, but intended for two-pass
> notifier
> + * callbacks where the call to @invalidate_start is the
> first
> + * pass and any struct mmu_interval_notifier_finish
> pointer
> + * returned in the @finish parameter describes the final
> pass.
> + * If @finish is %NULL on return, then no final pass will
> be
> + * called.
Is @finish guaranteed to be set to %NULL before the call? The existing
code does it, but is it something notifiers can rely on?
> + * @invalidate_finish: Called as the second pass for any notifier that
> returned
> + * a non-NULL @finish from @invalidate_start. The @finish
> + * pointer passed here is the same one returned by
> + * @invalidate_start.
> */
> struct mmu_interval_notifier_ops {
> bool (*invalidate)(struct mmu_interval_notifier *interval_sub,
> const struct mmu_notifier_range *range,
> unsigned long cur_seq);
> + bool (*invalidate_start)(struct mmu_interval_notifier *interval_sub,
> + const struct mmu_notifier_range *range,
> + unsigned long cur_seq,
> + struct mmu_interval_notifier_finish **finish);
> + void (*invalidate_finish)(struct mmu_interval_notifier_finish *finish);
> };
Nothing else jumped at me, and the idea makes sense.
--
Cheers,
David
