The reference counts for ZONE_DEVICE private pages should be
initialised by the driver when the page is actually allocated by the
driver allocator, not when they are first created. This is currently
the case for MEMORY_DEVICE_PRIVATE and MEMORY_DEVICE_COHERENT pages
but not MEMORY_DEVICE_PCI_P2PDMA pages so fix that up.
Signed-off-by: Alistair Popple <apopple@nvidia.com>
---
drivers/pci/p2pdma.c | 6 ++++++
mm/memremap.c | 17 +++++++++++++----
mm/mm_init.c | 22 ++++++++++++++++++----
3 files changed, 37 insertions(+), 8 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 4f47a13..210b9f4 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj,
}
/*
+ * Initialise the refcount for the freshly allocated page. As we have
+ * just allocated the page no one else should be using it.
+ */
+ set_page_count(virt_to_page(kaddr), 1);
+
+ /*
* vm_insert_page() can sleep, so a reference is taken to mapping
* such that rcu_read_unlock() can be done before inserting the
* pages
diff --git a/mm/memremap.c b/mm/memremap.c
index 40d4547..07bbe0e 100644
--- a/mm/memremap.c
+++ b/mm/memremap.c
@@ -488,15 +488,24 @@ void free_zone_device_folio(struct folio *folio)
folio->mapping = NULL;
folio->page.pgmap->ops->page_free(folio_page(folio, 0));
- if (folio->page.pgmap->type != MEMORY_DEVICE_PRIVATE &&
- folio->page.pgmap->type != MEMORY_DEVICE_COHERENT)
+ switch (folio->page.pgmap->type) {
+ case MEMORY_DEVICE_PRIVATE:
+ case MEMORY_DEVICE_COHERENT:
+ put_dev_pagemap(folio->page.pgmap);
+ break;
+
+ case MEMORY_DEVICE_FS_DAX:
+ case MEMORY_DEVICE_GENERIC:
/*
* Reset the refcount to 1 to prepare for handing out the page
* again.
*/
folio_set_count(folio, 1);
- else
- put_dev_pagemap(folio->page.pgmap);
+ break;
+
+ case MEMORY_DEVICE_PCI_P2PDMA:
+ break;
+ }
}
void zone_device_page_init(struct page *page)
diff --git a/mm/mm_init.c b/mm/mm_init.c
index 4ba5607..0489820 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1015,12 +1015,26 @@ static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
}
/*
- * ZONE_DEVICE pages are released directly to the driver page allocator
- * which will set the page count to 1 when allocating the page.
+ * ZONE_DEVICE pages other than MEMORY_TYPE_GENERIC and
+ * MEMORY_TYPE_FS_DAX pages are released directly to the driver page
+ * allocator which will set the page count to 1 when allocating the
+ * page.
+ *
+ * MEMORY_TYPE_GENERIC and MEMORY_TYPE_FS_DAX pages automatically have
+ * their refcount reset to one whenever they are freed (ie. after
+ * their refcount drops to 0).
*/
- if (pgmap->type == MEMORY_DEVICE_PRIVATE ||
- pgmap->type == MEMORY_DEVICE_COHERENT)
+ switch (pgmap->type) {
+ case MEMORY_DEVICE_PRIVATE:
+ case MEMORY_DEVICE_COHERENT:
+ case MEMORY_DEVICE_PCI_P2PDMA:
set_page_count(page, 0);
+ break;
+
+ case MEMORY_DEVICE_FS_DAX:
+ case MEMORY_DEVICE_GENERIC:
+ break;
+ }
}
/*
--
git-series 0.9.1Alistair Popple wrote:
> The reference counts for ZONE_DEVICE private pages should be
> initialised by the driver when the page is actually allocated by the
> driver allocator, not when they are first created. This is currently
> the case for MEMORY_DEVICE_PRIVATE and MEMORY_DEVICE_COHERENT pages
> but not MEMORY_DEVICE_PCI_P2PDMA pages so fix that up.
>
> Signed-off-by: Alistair Popple <apopple@nvidia.com>
> ---
> drivers/pci/p2pdma.c | 6 ++++++
> mm/memremap.c | 17 +++++++++++++----
> mm/mm_init.c | 22 ++++++++++++++++++----
> 3 files changed, 37 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
> index 4f47a13..210b9f4 100644
> --- a/drivers/pci/p2pdma.c
> +++ b/drivers/pci/p2pdma.c
> @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj,
> }
>
> /*
> + * Initialise the refcount for the freshly allocated page. As we have
> + * just allocated the page no one else should be using it.
> + */
> + set_page_count(virt_to_page(kaddr), 1);
Perhaps VM_WARN_ONCE to back up that assumption?
I also notice that there are multiple virt_to_page() lookups in this
routine, so maybe time for a local @page variable.
> +
> + /*
> * vm_insert_page() can sleep, so a reference is taken to mapping
> * such that rcu_read_unlock() can be done before inserting the
> * pages
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 40d4547..07bbe0e 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -488,15 +488,24 @@ void free_zone_device_folio(struct folio *folio)
> folio->mapping = NULL;
> folio->page.pgmap->ops->page_free(folio_page(folio, 0));
>
> - if (folio->page.pgmap->type != MEMORY_DEVICE_PRIVATE &&
> - folio->page.pgmap->type != MEMORY_DEVICE_COHERENT)
> + switch (folio->page.pgmap->type) {
> + case MEMORY_DEVICE_PRIVATE:
> + case MEMORY_DEVICE_COHERENT:
> + put_dev_pagemap(folio->page.pgmap);
> + break;
> +
> + case MEMORY_DEVICE_FS_DAX:
> + case MEMORY_DEVICE_GENERIC:
> /*
> * Reset the refcount to 1 to prepare for handing out the page
> * again.
> */
> folio_set_count(folio, 1);
> - else
> - put_dev_pagemap(folio->page.pgmap);
> + break;
> +
> + case MEMORY_DEVICE_PCI_P2PDMA:
> + break;
A follow on cleanup is that either all implementations should be
put_dev_pagemap(), or none of them. Put the onus on the implementation
to track how many pages it has handed out in the implementation
allocator.
> + }
> }
>
> void zone_device_page_init(struct page *page)
> diff --git a/mm/mm_init.c b/mm/mm_init.c
> index 4ba5607..0489820 100644
> --- a/mm/mm_init.c
> +++ b/mm/mm_init.c
> @@ -1015,12 +1015,26 @@ static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
> }
>
> /*
> - * ZONE_DEVICE pages are released directly to the driver page allocator
> - * which will set the page count to 1 when allocating the page.
> + * ZONE_DEVICE pages other than MEMORY_TYPE_GENERIC and
> + * MEMORY_TYPE_FS_DAX pages are released directly to the driver page
> + * allocator which will set the page count to 1 when allocating the
> + * page.
> + *
> + * MEMORY_TYPE_GENERIC and MEMORY_TYPE_FS_DAX pages automatically have
> + * their refcount reset to one whenever they are freed (ie. after
> + * their refcount drops to 0).
I'll send some follow on patches to clean up device-dax.
For this one:
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Dan Williams <dan.j.williams@intel.com> writes:
> Alistair Popple wrote:
[...]
>> diff --git a/mm/memremap.c b/mm/memremap.c
>> index 40d4547..07bbe0e 100644
>> --- a/mm/memremap.c
>> +++ b/mm/memremap.c
>> @@ -488,15 +488,24 @@ void free_zone_device_folio(struct folio *folio)
>> folio->mapping = NULL;
>> folio->page.pgmap->ops->page_free(folio_page(folio, 0));
>>
>> - if (folio->page.pgmap->type != MEMORY_DEVICE_PRIVATE &&
>> - folio->page.pgmap->type != MEMORY_DEVICE_COHERENT)
>> + switch (folio->page.pgmap->type) {
>> + case MEMORY_DEVICE_PRIVATE:
>> + case MEMORY_DEVICE_COHERENT:
>> + put_dev_pagemap(folio->page.pgmap);
>> + break;
>> +
>> + case MEMORY_DEVICE_FS_DAX:
>> + case MEMORY_DEVICE_GENERIC:
>> /*
>> * Reset the refcount to 1 to prepare for handing out the page
>> * again.
>> */
>> folio_set_count(folio, 1);
>> - else
>> - put_dev_pagemap(folio->page.pgmap);
>> + break;
>> +
>> + case MEMORY_DEVICE_PCI_P2PDMA:
>> + break;
>
> A follow on cleanup is that either all implementations should be
> put_dev_pagemap(), or none of them. Put the onus on the implementation
> to track how many pages it has handed out in the implementation
> allocator.
Agreed. I've ignored the get/put_dev_pagemap() calls for this clean up
but am planning to do a follow up to clean those up too, probably by
removing them entirely as you suggest.
[...]
> For this one:
>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Thanks.
On 2024-09-09 22:14, Alistair Popple wrote: > The reference counts for ZONE_DEVICE private pages should be > initialised by the driver when the page is actually allocated by the > driver allocator, not when they are first created. This is currently > the case for MEMORY_DEVICE_PRIVATE and MEMORY_DEVICE_COHERENT pages > but not MEMORY_DEVICE_PCI_P2PDMA pages so fix that up. > > Signed-off-by: Alistair Popple <apopple@nvidia.com> > --- > drivers/pci/p2pdma.c | 6 ++++++ > mm/memremap.c | 17 +++++++++++++---- > mm/mm_init.c | 22 ++++++++++++++++++---- > 3 files changed, 37 insertions(+), 8 deletions(-) > > diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c > index 4f47a13..210b9f4 100644 > --- a/drivers/pci/p2pdma.c > +++ b/drivers/pci/p2pdma.c > @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj, > } > > /* > + * Initialise the refcount for the freshly allocated page. As we have > + * just allocated the page no one else should be using it. > + */ > + set_page_count(virt_to_page(kaddr), 1); > + > + /* > * vm_insert_page() can sleep, so a reference is taken to mapping > * such that rcu_read_unlock() can be done before inserting the > * pages This seems to only set reference count to the first page, when there can be more than one page referenced by kaddr. I suspect the page count adjustment should be done in the for loop that's a few lines lower than this. I think a similar mistake was made by other recent changes. Thanks, Logan
Logan Gunthorpe <logang@deltatee.com> writes: > On 2024-09-09 22:14, Alistair Popple wrote: >> The reference counts for ZONE_DEVICE private pages should be >> initialised by the driver when the page is actually allocated by the >> driver allocator, not when they are first created. This is currently >> the case for MEMORY_DEVICE_PRIVATE and MEMORY_DEVICE_COHERENT pages >> but not MEMORY_DEVICE_PCI_P2PDMA pages so fix that up. >> >> Signed-off-by: Alistair Popple <apopple@nvidia.com> >> --- >> drivers/pci/p2pdma.c | 6 ++++++ >> mm/memremap.c | 17 +++++++++++++---- >> mm/mm_init.c | 22 ++++++++++++++++++---- >> 3 files changed, 37 insertions(+), 8 deletions(-) >> >> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c >> index 4f47a13..210b9f4 100644 >> --- a/drivers/pci/p2pdma.c >> +++ b/drivers/pci/p2pdma.c >> @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj, >> } >> >> /* >> + * Initialise the refcount for the freshly allocated page. As we have >> + * just allocated the page no one else should be using it. >> + */ >> + set_page_count(virt_to_page(kaddr), 1); >> + >> + /* >> * vm_insert_page() can sleep, so a reference is taken to mapping >> * such that rcu_read_unlock() can be done before inserting the >> * pages > This seems to only set reference count to the first page, when there can > be more than one page referenced by kaddr. Good point. > I suspect the page count adjustment should be done in the for loop > that's a few lines lower than this. Have moved it there for the next version, thanks! > I think a similar mistake was made by other recent changes. > > Thanks, > > Logan
In subject: PCI/P2PDMA: ... would match previous history. On Tue, Sep 10, 2024 at 02:14:27PM +1000, Alistair Popple wrote: > The reference counts for ZONE_DEVICE private pages should be > initialised by the driver when the page is actually allocated by the > driver allocator, not when they are first created. This is currently > the case for MEMORY_DEVICE_PRIVATE and MEMORY_DEVICE_COHERENT pages > but not MEMORY_DEVICE_PCI_P2PDMA pages so fix that up. > > Signed-off-by: Alistair Popple <apopple@nvidia.com> > --- > drivers/pci/p2pdma.c | 6 ++++++ > mm/memremap.c | 17 +++++++++++++---- > mm/mm_init.c | 22 ++++++++++++++++++---- > 3 files changed, 37 insertions(+), 8 deletions(-) > > diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c > index 4f47a13..210b9f4 100644 > --- a/drivers/pci/p2pdma.c > +++ b/drivers/pci/p2pdma.c > @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj, > } > > /* > + * Initialise the refcount for the freshly allocated page. As we have > + * just allocated the page no one else should be using it. > + */ > + set_page_count(virt_to_page(kaddr), 1); No doubt the subject line is true in some overall context, but it does seem to say the opposite of what happens here. Bjorn
>> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c >> index 4f47a13..210b9f4 100644 >> --- a/drivers/pci/p2pdma.c >> +++ b/drivers/pci/p2pdma.c >> @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj, >> } >> >> /* >> + * Initialise the refcount for the freshly allocated page. As we have >> + * just allocated the page no one else should be using it. >> + */ >> + set_page_count(virt_to_page(kaddr), 1); > > No doubt the subject line is true in some overall context, but it does > seem to say the opposite of what happens here. Fair. It made sense to me from the mm context I was coming from (it was being initialised to 1 there) but not overall. Something like "move page refcount initialisation to p2pdma driver" would make more sense? - Alistair > Bjorn
On Wed, Sep 11, 2024 at 11:07:51AM +1000, Alistair Popple wrote: > > >> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c > >> index 4f47a13..210b9f4 100644 > >> --- a/drivers/pci/p2pdma.c > >> +++ b/drivers/pci/p2pdma.c > >> @@ -129,6 +129,12 @@ static int p2pmem_alloc_mmap(struct file *filp, struct kobject *kobj, > >> } > >> > >> /* > >> + * Initialise the refcount for the freshly allocated page. As we have > >> + * just allocated the page no one else should be using it. > >> + */ > >> + set_page_count(virt_to_page(kaddr), 1); > > > > No doubt the subject line is true in some overall context, but it does > > seem to say the opposite of what happens here. > > Fair. It made sense to me from the mm context I was coming from (it was > being initialised to 1 there) but not overall. Something like "move page > refcount initialisation to p2pdma driver" would make more sense? Definitely would, thanks.
© 2016 - 2024 Red Hat, Inc.