gug prohibits users from calling get_user_pages() with FOLL_PIN. But it
allows users to call get_user_pages() with FOLL_LONGTERM only. It seems
insensible.
since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit
users from calling get_user_pages() with FOLL_LONGTERM while not with
FOLL_PIN.
mm/gup_benchmark.c used to be the only user who did this improperly.
But it has been fixed by moving to use pin_user_pages().
Cc: John Hubbard <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Jérôme Glisse <[email protected]>
Cc: "Matthew Wilcox (Oracle)" <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Mike Kravetz <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Signed-off-by: Barry Song <[email protected]>
---
mm/gup.c | 37 ++++++++++++++++++++++---------------
1 file changed, 22 insertions(+), 15 deletions(-)
diff --git a/mm/gup.c b/mm/gup.c
index ae096ea7583f..4da669f79566 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1789,6 +1789,25 @@ static long __get_user_pages_remote(struct mm_struct *mm,
gup_flags | FOLL_TOUCH | FOLL_REMOTE);
}
+static bool is_valid_gup_flags(unsigned int gup_flags)
+{
+ /*
+ * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
+ * never directly by the caller, so enforce that with an assertion:
+ */
+ if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
+ return false;
+ /*
+ * FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying
+ * that is, FOLL_LONGTERM is a specific case, more restrictive case of
+ * FOLL_PIN.
+ */
+ if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
+ return false;
+
+ return true;
+}
+
/**
* get_user_pages_remote() - pin user pages in memory
* @mm: mm_struct of target mm
@@ -1854,11 +1873,7 @@ long get_user_pages_remote(struct mm_struct *mm,
unsigned int gup_flags, struct page **pages,
struct vm_area_struct **vmas, int *locked)
{
- /*
- * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
- * never directly by the caller, so enforce that with an assertion:
- */
- if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
+ if (!is_valid_gup_flags(gup_flags))
return -EINVAL;
return __get_user_pages_remote(mm, start, nr_pages, gup_flags,
@@ -1904,11 +1919,7 @@ long get_user_pages(unsigned long start, unsigned long nr_pages,
unsigned int gup_flags, struct page **pages,
struct vm_area_struct **vmas)
{
- /*
- * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
- * never directly by the caller, so enforce that with an assertion:
- */
- if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
+ if (!is_valid_gup_flags(gup_flags))
return -EINVAL;
return __gup_longterm_locked(current->mm, start, nr_pages,
@@ -2810,11 +2821,7 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast_only);
int get_user_pages_fast(unsigned long start, int nr_pages,
unsigned int gup_flags, struct page **pages)
{
- /*
- * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
- * never directly by the caller, so enforce that:
- */
- if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
+ if (!is_valid_gup_flags(gup_flags))
return -EINVAL;
/*
--
2.27.0
On Wed, Aug 19, 2020 at 11:01:00PM +1200, Barry Song wrote:
> gug prohibits users from calling get_user_pages() with FOLL_PIN. But it
> allows users to call get_user_pages() with FOLL_LONGTERM only. It seems
> insensible.
>
> since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit
> users from calling get_user_pages() with FOLL_LONGTERM while not with
> FOLL_PIN.
>
> mm/gup_benchmark.c used to be the only user who did this improperly.
> But it has been fixed by moving to use pin_user_pages().
>
> Cc: John Hubbard <[email protected]>
> Cc: Jan Kara <[email protected]>
> Cc: J?r?me Glisse <[email protected]>
> Cc: "Matthew Wilcox (Oracle)" <[email protected]>
> Cc: Al Viro <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Cc: Dan Williams <[email protected]>
> Cc: Dave Chinner <[email protected]>
> Cc: Jason Gunthorpe <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Cc: Michal Hocko <[email protected]>
> Cc: Mike Kravetz <[email protected]>
> Cc: Shuah Khan <[email protected]>
> Cc: Vlastimil Babka <[email protected]>
> Signed-off-by: Barry Song <[email protected]>
Seems reasonable to me.
Reviewed-by: Ira Weiny <[email protected]>
> ---
> mm/gup.c | 37 ++++++++++++++++++++++---------------
> 1 file changed, 22 insertions(+), 15 deletions(-)
>
> diff --git a/mm/gup.c b/mm/gup.c
> index ae096ea7583f..4da669f79566 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1789,6 +1789,25 @@ static long __get_user_pages_remote(struct mm_struct *mm,
> gup_flags | FOLL_TOUCH | FOLL_REMOTE);
> }
>
> +static bool is_valid_gup_flags(unsigned int gup_flags)
> +{
> + /*
> + * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> + * never directly by the caller, so enforce that with an assertion:
> + */
> + if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + return false;
> + /*
> + * FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying
> + * that is, FOLL_LONGTERM is a specific case, more restrictive case of
> + * FOLL_PIN.
> + */
> + if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
> + return false;
> +
> + return true;
> +}
> +
> /**
> * get_user_pages_remote() - pin user pages in memory
> * @mm: mm_struct of target mm
> @@ -1854,11 +1873,7 @@ long get_user_pages_remote(struct mm_struct *mm,
> unsigned int gup_flags, struct page **pages,
> struct vm_area_struct **vmas, int *locked)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that with an assertion:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> return __get_user_pages_remote(mm, start, nr_pages, gup_flags,
> @@ -1904,11 +1919,7 @@ long get_user_pages(unsigned long start, unsigned long nr_pages,
> unsigned int gup_flags, struct page **pages,
> struct vm_area_struct **vmas)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that with an assertion:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> return __gup_longterm_locked(current->mm, start, nr_pages,
> @@ -2810,11 +2821,7 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast_only);
> int get_user_pages_fast(unsigned long start, int nr_pages,
> unsigned int gup_flags, struct page **pages)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> /*
> --
> 2.27.0
>
>
>
On 8/19/20 4:01 AM, Barry Song wrote:
> gug prohibits users from calling get_user_pages() with FOLL_PIN. But it
Maybe Andrew can fix the typo above: gug --> gup.
> allows users to call get_user_pages() with FOLL_LONGTERM only. It seems
> insensible.
>
> since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit
> users from calling get_user_pages() with FOLL_LONGTERM while not with
> FOLL_PIN.
>
> mm/gup_benchmark.c used to be the only user who did this improperly.
> But it has been fixed by moving to use pin_user_pages().
For future patches, you don't have to write everything in the
commit log. Some things are better placed in a cover letter or after
the "---" line, because they don't need to be recorded forever.
Anyway, the diffs seem fine, assuming that you've audited the call sites.
thanks,
--
John Hubbard
NVIDIA
>
> Cc: John Hubbard <[email protected]>
> Cc: Jan Kara <[email protected]>
> Cc: Jérôme Glisse <[email protected]>
> Cc: "Matthew Wilcox (Oracle)" <[email protected]>
> Cc: Al Viro <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Cc: Dan Williams <[email protected]>
> Cc: Dave Chinner <[email protected]>
> Cc: Jason Gunthorpe <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Cc: Michal Hocko <[email protected]>
> Cc: Mike Kravetz <[email protected]>
> Cc: Shuah Khan <[email protected]>
> Cc: Vlastimil Babka <[email protected]>
> Signed-off-by: Barry Song <[email protected]>
> ---
> mm/gup.c | 37 ++++++++++++++++++++++---------------
> 1 file changed, 22 insertions(+), 15 deletions(-)
>
> diff --git a/mm/gup.c b/mm/gup.c
> index ae096ea7583f..4da669f79566 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1789,6 +1789,25 @@ static long __get_user_pages_remote(struct mm_struct *mm,
> gup_flags | FOLL_TOUCH | FOLL_REMOTE);
> }
>
> +static bool is_valid_gup_flags(unsigned int gup_flags)
> +{
> + /*
> + * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> + * never directly by the caller, so enforce that with an assertion:
> + */
> + if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + return false;
> + /*
> + * FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying
> + * that is, FOLL_LONGTERM is a specific case, more restrictive case of
> + * FOLL_PIN.
> + */
> + if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
> + return false;
> +
> + return true;
> +}
> +
> /**
> * get_user_pages_remote() - pin user pages in memory
> * @mm: mm_struct of target mm
> @@ -1854,11 +1873,7 @@ long get_user_pages_remote(struct mm_struct *mm,
> unsigned int gup_flags, struct page **pages,
> struct vm_area_struct **vmas, int *locked)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that with an assertion:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> return __get_user_pages_remote(mm, start, nr_pages, gup_flags,
> @@ -1904,11 +1919,7 @@ long get_user_pages(unsigned long start, unsigned long nr_pages,
> unsigned int gup_flags, struct page **pages,
> struct vm_area_struct **vmas)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that with an assertion:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> return __gup_longterm_locked(current->mm, start, nr_pages,
> @@ -2810,11 +2821,7 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast_only);
> int get_user_pages_fast(unsigned long start, int nr_pages,
> unsigned int gup_flags, struct page **pages)
> {
> - /*
> - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> - * never directly by the caller, so enforce that:
> - */
> - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> + if (!is_valid_gup_flags(gup_flags))
> return -EINVAL;
>
> /*
>
On Wed, Aug 19, 2020 at 11:45 PM John Hubbard <[email protected]> wrote:
>
> On 8/19/20 4:01 AM, Barry Song wrote:
> > gug prohibits users from calling get_user_pages() with FOLL_PIN. But it
>
> Maybe Andrew can fix the typo above: gug --> gup.
>
>
> > allows users to call get_user_pages() with FOLL_LONGTERM only. It seems
> > insensible.
> >
> > since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit
> > users from calling get_user_pages() with FOLL_LONGTERM while not with
> > FOLL_PIN.
> >
> > mm/gup_benchmark.c used to be the only user who did this improperly.
> > But it has been fixed by moving to use pin_user_pages().
>
> For future patches, you don't have to write everything in the
> commit log. Some things are better placed in a cover letter or after
> the "---" line, because they don't need to be recorded forever.
>
> Anyway, the diffs seem fine, assuming that you've audited the call sites.
We can use is_valid_gup_flags() inside ->
get_user_pages_locked(),
get_user_pages_unlocked(),
pin_user_pages_locked() as well.
Are you planning to add it in future patches ?
>
> thanks,
> --
> John Hubbard
> NVIDIA
>
> >
> > Cc: John Hubbard <[email protected]>
> > Cc: Jan Kara <[email protected]>
> > Cc: Jérôme Glisse <[email protected]>
> > Cc: "Matthew Wilcox (Oracle)" <[email protected]>
> > Cc: Al Viro <[email protected]>
> > Cc: Christoph Hellwig <[email protected]>
> > Cc: Dan Williams <[email protected]>
> > Cc: Dave Chinner <[email protected]>
> > Cc: Jason Gunthorpe <[email protected]>
> > Cc: Jonathan Corbet <[email protected]>
> > Cc: Michal Hocko <[email protected]>
> > Cc: Mike Kravetz <[email protected]>
> > Cc: Shuah Khan <[email protected]>
> > Cc: Vlastimil Babka <[email protected]>
> > Signed-off-by: Barry Song <[email protected]>
> > ---
> > mm/gup.c | 37 ++++++++++++++++++++++---------------
> > 1 file changed, 22 insertions(+), 15 deletions(-)
> >
> > diff --git a/mm/gup.c b/mm/gup.c
> > index ae096ea7583f..4da669f79566 100644
> > --- a/mm/gup.c
> > +++ b/mm/gup.c
> > @@ -1789,6 +1789,25 @@ static long __get_user_pages_remote(struct mm_struct *mm,
> > gup_flags | FOLL_TOUCH | FOLL_REMOTE);
> > }
> >
> > +static bool is_valid_gup_flags(unsigned int gup_flags)
> > +{
> > + /*
> > + * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> > + * never directly by the caller, so enforce that with an assertion:
> > + */
> > + if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> > + return false;
> > + /*
> > + * FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying
> > + * that is, FOLL_LONGTERM is a specific case, more restrictive case of
> > + * FOLL_PIN.
> > + */
> > + if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
> > + return false;
> > +
> > + return true;
> > +}
> > +
> > /**
> > * get_user_pages_remote() - pin user pages in memory
> > * @mm: mm_struct of target mm
> > @@ -1854,11 +1873,7 @@ long get_user_pages_remote(struct mm_struct *mm,
> > unsigned int gup_flags, struct page **pages,
> > struct vm_area_struct **vmas, int *locked)
> > {
> > - /*
> > - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> > - * never directly by the caller, so enforce that with an assertion:
> > - */
> > - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> > + if (!is_valid_gup_flags(gup_flags))
> > return -EINVAL;
> >
> > return __get_user_pages_remote(mm, start, nr_pages, gup_flags,
> > @@ -1904,11 +1919,7 @@ long get_user_pages(unsigned long start, unsigned long nr_pages,
> > unsigned int gup_flags, struct page **pages,
> > struct vm_area_struct **vmas)
> > {
> > - /*
> > - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> > - * never directly by the caller, so enforce that with an assertion:
> > - */
> > - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> > + if (!is_valid_gup_flags(gup_flags))
> > return -EINVAL;
> >
> > return __gup_longterm_locked(current->mm, start, nr_pages,
> > @@ -2810,11 +2821,7 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast_only);
> > int get_user_pages_fast(unsigned long start, int nr_pages,
> > unsigned int gup_flags, struct page **pages)
> > {
> > - /*
> > - * FOLL_PIN must only be set internally by the pin_user_pages*() APIs,
> > - * never directly by the caller, so enforce that:
> > - */
> > - if (WARN_ON_ONCE(gup_flags & FOLL_PIN))
> > + if (!is_valid_gup_flags(gup_flags))
> > return -EINVAL;
> >
> > /*
> >
>
>
On 9/3/20 12:12 AM, Souptick Joarder wrote:
> On Wed, Aug 19, 2020 at 11:45 PM John Hubbard <[email protected]> wrote:
>>
>> On 8/19/20 4:01 AM, Barry Song wrote:
>>> gug prohibits users from calling get_user_pages() with FOLL_PIN. But it
>>
>> Maybe Andrew can fix the typo above: gug --> gup.
>>
>>
>>> allows users to call get_user_pages() with FOLL_LONGTERM only. It seems
>>> insensible.
>>>
>>> since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit
>>> users from calling get_user_pages() with FOLL_LONGTERM while not with
>>> FOLL_PIN.
>>>
>>> mm/gup_benchmark.c used to be the only user who did this improperly.
>>> But it has been fixed by moving to use pin_user_pages().
>>
>> For future patches, you don't have to write everything in the
>> commit log. Some things are better placed in a cover letter or after
>> the "---" line, because they don't need to be recorded forever.
>>
>> Anyway, the diffs seem fine, assuming that you've audited the call sites.
>
> We can use is_valid_gup_flags() inside ->
> get_user_pages_locked(),
> get_user_pages_unlocked(),
> pin_user_pages_locked() as well.
Probably it's best to discern between valid pup flags, and valid gup flags.
As in: separate functions for those. Maybe one is a subset of the other, but
still.
>
> Are you planning to add it in future patches ?
>
It's not on my list. I don't see anything wrong with doing so, other
than avoiding the minor pitfall I called out above. So if you want to
do that, then feel free...
thanks,
--
John Hubbard
NVIDIA
On Thu, Sep 03, 2020 at 12:42:44PM +0530, Souptick Joarder wrote:
> We can use is_valid_gup_flags() inside ->
> get_user_pages_locked(),
> get_user_pages_unlocked(),
> pin_user_pages_locked() as well.
>
> Are you planning to add it in future patches ?
If you're looking for a new project, adding a foll_t or gup_t or
something for the FOLL flags (like we have for gfp_t or vm_fault_t)
would be helpful. We're inconsistent with our naming here.
On Fri, Sep 4, 2020 at 12:09 AM Matthew Wilcox <[email protected]> wrote:
>
> On Thu, Sep 03, 2020 at 12:42:44PM +0530, Souptick Joarder wrote:
> > We can use is_valid_gup_flags() inside ->
> > get_user_pages_locked(),
> > get_user_pages_unlocked(),
> > pin_user_pages_locked() as well.
> >
> > Are you planning to add it in future patches ?
>
> If you're looking for a new project, adding a foll_t or gup_t or
> something for the FOLL flags (like we have for gfp_t or vm_fault_t)
> would be helpful. We're inconsistent with our naming here.
Sure. I will start looking into this and come up with a RFC version.