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 <jgg@ziepe.ca>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Cc: Dave Airlie <airlied@gmail.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: <dri-devel@lists.freedesktop.org>
Cc: <linux-mm@kvack.org>
Cc: <linux-kernel@vger.kernel.org>
Assisted-by: GitHub Copilot:claude-sonnet-4.6 # Documentation only.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
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.
+ * If allocation fails (which is not unlikely under memory pressure), fall back
+ * to single-pass operation. 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.
+ * @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);
};
struct mmu_interval_notifier {
diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
index a6cdf3674bdc..4d8a64ce8eda 100644
--- a/mm/mmu_notifier.c
+++ b/mm/mmu_notifier.c
@@ -260,6 +260,15 @@ mmu_interval_read_begin(struct mmu_interval_notifier *interval_sub)
}
EXPORT_SYMBOL_GPL(mmu_interval_read_begin);
+static void mn_itree_finish_pass(struct llist_head *finish_passes)
+{
+ struct llist_node *first = llist_reverse_order(__llist_del_all(finish_passes));
+ struct mmu_interval_notifier_finish *f, *next;
+
+ llist_for_each_entry_safe(f, next, first, link)
+ f->notifier->ops->invalidate_finish(f);
+}
+
static void mn_itree_release(struct mmu_notifier_subscriptions *subscriptions,
struct mm_struct *mm)
{
@@ -271,6 +280,7 @@ static void mn_itree_release(struct mmu_notifier_subscriptions *subscriptions,
.end = ULONG_MAX,
};
struct mmu_interval_notifier *interval_sub;
+ LLIST_HEAD(finish_passes);
unsigned long cur_seq;
bool ret;
@@ -278,11 +288,27 @@ static void mn_itree_release(struct mmu_notifier_subscriptions *subscriptions,
mn_itree_inv_start_range(subscriptions, &range, &cur_seq);
interval_sub;
interval_sub = mn_itree_inv_next(interval_sub, &range)) {
- ret = interval_sub->ops->invalidate(interval_sub, &range,
- cur_seq);
+ if (interval_sub->ops->invalidate_start) {
+ struct mmu_interval_notifier_finish *finish = NULL;
+
+ ret = interval_sub->ops->invalidate_start(interval_sub,
+ &range,
+ cur_seq,
+ &finish);
+ if (ret && finish) {
+ finish->notifier = interval_sub;
+ __llist_add(&finish->link, &finish_passes);
+ }
+
+ } else {
+ ret = interval_sub->ops->invalidate(interval_sub,
+ &range,
+ cur_seq);
+ }
WARN_ON(!ret);
}
+ mn_itree_finish_pass(&finish_passes);
mn_itree_inv_end(subscriptions);
}
@@ -430,7 +456,9 @@ static int mn_itree_invalidate(struct mmu_notifier_subscriptions *subscriptions,
const struct mmu_notifier_range *range)
{
struct mmu_interval_notifier *interval_sub;
+ LLIST_HEAD(finish_passes);
unsigned long cur_seq;
+ int err = 0;
for (interval_sub =
mn_itree_inv_start_range(subscriptions, range, &cur_seq);
@@ -438,23 +466,41 @@ static int mn_itree_invalidate(struct mmu_notifier_subscriptions *subscriptions,
interval_sub = mn_itree_inv_next(interval_sub, range)) {
bool ret;
- ret = interval_sub->ops->invalidate(interval_sub, range,
- cur_seq);
+ if (interval_sub->ops->invalidate_start) {
+ struct mmu_interval_notifier_finish *finish = NULL;
+
+ ret = interval_sub->ops->invalidate_start(interval_sub,
+ range,
+ cur_seq,
+ &finish);
+ if (ret && finish) {
+ finish->notifier = interval_sub;
+ __llist_add(&finish->link, &finish_passes);
+ }
+
+ } else {
+ ret = interval_sub->ops->invalidate(interval_sub,
+ range,
+ cur_seq);
+ }
if (!ret) {
if (WARN_ON(mmu_notifier_range_blockable(range)))
continue;
- goto out_would_block;
+ err = -EAGAIN;
+ break;
}
}
- return 0;
-out_would_block:
+ mn_itree_finish_pass(&finish_passes);
+
/*
* On -EAGAIN the non-blocking caller is not allowed to call
* invalidate_range_end()
*/
- mn_itree_inv_end(subscriptions);
- return -EAGAIN;
+ if (err)
+ mn_itree_inv_end(subscriptions);
+
+ return err;
}
static int mn_hlist_invalidate_range_start(
@@ -976,6 +1022,7 @@ int mmu_interval_notifier_insert(struct mmu_interval_notifier *interval_sub,
struct mmu_notifier_subscriptions *subscriptions;
int ret;
+ WARN_ON_ONCE(ops->invalidate_start && !ops->invalidate_finish);
might_lock(&mm->mmap_lock);
subscriptions = smp_load_acquire(&mm->notifier_subscriptions);
--
2.53.0
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 <jgg@ziepe.ca>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Simona Vetter <simona.vetter@ffwll.ch>
> Cc: Dave Airlie <airlied@gmail.com>
> Cc: Alistair Popple <apopple@nvidia.com>
> Cc: <dri-devel@lists.freedesktop.org>
> Cc: <linux-mm@kvack.org>
> Cc: <linux-kernel@vger.kernel.org>
>
> Assisted-by: GitHub Copilot:claude-sonnet-4.6 # Documentation only.
> Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> 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
On Wed, 2026-03-04 at 20:45 +0100, David Hildenbrand (Arm) wrote:
> 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 <jgg@ziepe.ca>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Simona Vetter <simona.vetter@ffwll.ch>
> > Cc: Dave Airlie <airlied@gmail.com>
> > Cc: Alistair Popple <apopple@nvidia.com>
> > Cc: <dri-devel@lists.freedesktop.org>
> > Cc: <linux-mm@kvack.org>
> > Cc: <linux-kernel@vger.kernel.org>
> >
> > Assisted-by: GitHub Copilot:claude-sonnet-4.6 # Documentation only.
> > Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> > ---
> > 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.
Thanks. I sent out a v4 addressing the above and to a wider audience.
Thanks,
Thomas
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. Please CC all maintainers+reviewers that MAINTAINERS recommends you to cc. -- Cheers, David
On Wed, 2026-03-04 at 20:32 +0100, David Hildenbrand (Arm) wrote: > 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. > > Please CC all maintainers+reviewers that MAINTAINERS recommends you > to cc. Hmm. Good point. I missed a fair number of those. Will Resend Thanks, Thomas
© 2016 - 2026 Red Hat, Inc.