[PATCH 1/2] xsk: do not use mmap_sem

[PATCH 1/2] xsk: do not use mmap_sem

[Davidlohr Bueso]

Holding mmap_sem exclusively for a gup() is an overkill.
Lets replace the call for gup_fast() and let the mm take
it if necessary.

Cc: David S. Miller <davem@davemloft.net>
Cc: Bjorn Topel <bjorn.topel@intel.com>
Cc: Magnus Karlsson <magnus.karlsson@intel.com>
CC: netdev@vger.kernel.org
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
---
 net/xdp/xdp_umem.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 5ab236c5c9a5..25e1e76654a8 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -265,10 +265,8 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
 	if (!umem->pgs)
 		return -ENOMEM;
 
-	down_write(&current->mm->mmap_sem);
-	npgs = get_user_pages(umem->address, umem->npgs,
-			      gup_flags, &umem->pgs[0], NULL);
-	up_write(&current->mm->mmap_sem);
+	npgs = get_user_pages_fast(umem->address, umem->npgs,
+				   gup_flags, &umem->pgs[0]);
 
 	if (npgs != umem->npgs) {
 		if (npgs >= 0) {
-- 
2.16.4

Re: [PATCH 1/2] xsk: do not use mmap_sem

[Björn Töpel]

Den tors 7 feb. 2019 kl 06:38 skrev Davidlohr Bueso <dave@stgolabs.net>:
>
> Holding mmap_sem exclusively for a gup() is an overkill.
> Lets replace the call for gup_fast() and let the mm take
> it if necessary.
>
> Cc: David S. Miller <davem@davemloft.net>
> Cc: Bjorn Topel <bjorn.topel@intel.com>
> Cc: Magnus Karlsson <magnus.karlsson@intel.com>
> CC: netdev@vger.kernel.org
> Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
> ---
>  net/xdp/xdp_umem.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
>
> diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
> index 5ab236c5c9a5..25e1e76654a8 100644
> --- a/net/xdp/xdp_umem.c
> +++ b/net/xdp/xdp_umem.c
> @@ -265,10 +265,8 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
>         if (!umem->pgs)
>                 return -ENOMEM;
>
> -       down_write(&current->mm->mmap_sem);
> -       npgs = get_user_pages(umem->address, umem->npgs,
> -                             gup_flags, &umem->pgs[0], NULL);
> -       up_write(&current->mm->mmap_sem);
> +       npgs = get_user_pages_fast(umem->address, umem->npgs,
> +                                  gup_flags, &umem->pgs[0]);
>

Thanks for the patch!

The lifetime of the pinning is similar to RDMA umem mapping, so isn't
gup_longterm preferred?


Björn

>         if (npgs != umem->npgs) {
>                 if (npgs >= 0) {
> --
> 2.16.4
>

Re: [PATCH 1/2] xsk: do not use mmap_sem

[Daniel Borkmann]

[ +Dan ]

On 02/07/2019 08:43 AM, Björn Töpel wrote:
> Den tors 7 feb. 2019 kl 06:38 skrev Davidlohr Bueso <dave@stgolabs.net>:
>>
>> Holding mmap_sem exclusively for a gup() is an overkill.
>> Lets replace the call for gup_fast() and let the mm take
>> it if necessary.
>>
>> Cc: David S. Miller <davem@davemloft.net>
>> Cc: Bjorn Topel <bjorn.topel@intel.com>
>> Cc: Magnus Karlsson <magnus.karlsson@intel.com>
>> CC: netdev@vger.kernel.org
>> Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
>> ---
>>  net/xdp/xdp_umem.c | 6 ++----
>>  1 file changed, 2 insertions(+), 4 deletions(-)
>>
>> diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
>> index 5ab236c5c9a5..25e1e76654a8 100644
>> --- a/net/xdp/xdp_umem.c
>> +++ b/net/xdp/xdp_umem.c
>> @@ -265,10 +265,8 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
>>         if (!umem->pgs)
>>                 return -ENOMEM;
>>
>> -       down_write(&current->mm->mmap_sem);
>> -       npgs = get_user_pages(umem->address, umem->npgs,
>> -                             gup_flags, &umem->pgs[0], NULL);
>> -       up_write(&current->mm->mmap_sem);
>> +       npgs = get_user_pages_fast(umem->address, umem->npgs,
>> +                                  gup_flags, &umem->pgs[0]);
>>
> 
> Thanks for the patch!
> 
> The lifetime of the pinning is similar to RDMA umem mapping, so isn't
> gup_longterm preferred?

Seems reasonable from reading what gup_longterm seems to do. Davidlohr
or Dan, any thoughts on the above?

Thanks,
Daniel

[PATCH v2] xsk: share the mmap_sem for page pinning

[Davidlohr Bueso]

Holding mmap_sem exclusively for a gup() is an overkill.
Lets share the lock and replace the gup call for gup_longterm(),
as it is better suited for the lifetime of the pinning.

Cc: David S. Miller <davem@davemloft.net>
Cc: Bjorn Topel <bjorn.topel@intel.com>
Cc: Magnus Karlsson <magnus.karlsson@intel.com>
CC: netdev@vger.kernel.org
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
---
 net/xdp/xdp_umem.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 5ab236c5c9a5..e7fa8d0d7090 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -265,10 +265,10 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
 	if (!umem->pgs)
 		return -ENOMEM;
 
-	down_write(&current->mm->mmap_sem);
-	npgs = get_user_pages(umem->address, umem->npgs,
+	down_read(&current->mm->mmap_sem);
+	npgs = get_user_pages_longterm(umem->address, umem->npgs,
 			      gup_flags, &umem->pgs[0], NULL);
-	up_write(&current->mm->mmap_sem);
+	up_read(&current->mm->mmap_sem);
 
 	if (npgs != umem->npgs) {
 		if (npgs >= 0) {
-- 
2.16.4

Re: [PATCH v2] xsk: share the mmap_sem for page pinning

[Björn Töpel]

Den mån 11 feb. 2019 kl 17:15 skrev Davidlohr Bueso <dave@stgolabs.net>:
>
> Holding mmap_sem exclusively for a gup() is an overkill.
> Lets share the lock and replace the gup call for gup_longterm(),
> as it is better suited for the lifetime of the pinning.
>

Thanks for the cleanup!

Acked-by: Björn Töpel <bjorn.topel@intel.com>

> Cc: David S. Miller <davem@davemloft.net>
> Cc: Bjorn Topel <bjorn.topel@intel.com>
> Cc: Magnus Karlsson <magnus.karlsson@intel.com>
> CC: netdev@vger.kernel.org
> Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
> ---
>  net/xdp/xdp_umem.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
> index 5ab236c5c9a5..e7fa8d0d7090 100644
> --- a/net/xdp/xdp_umem.c
> +++ b/net/xdp/xdp_umem.c
> @@ -265,10 +265,10 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
>         if (!umem->pgs)
>                 return -ENOMEM;
>
> -       down_write(&current->mm->mmap_sem);
> -       npgs = get_user_pages(umem->address, umem->npgs,
> +       down_read(&current->mm->mmap_sem);
> +       npgs = get_user_pages_longterm(umem->address, umem->npgs,
>                               gup_flags, &umem->pgs[0], NULL);
> -       up_write(&current->mm->mmap_sem);
> +       up_read(&current->mm->mmap_sem);
>
>         if (npgs != umem->npgs) {
>                 if (npgs >= 0) {
> --
> 2.16.4
>

Re: [PATCH 1/2] xsk: do not use mmap_sem

[Dan Williams]

[ add Ira ]

On Mon, Feb 11, 2019 at 7:33 AM Daniel Borkmann <daniel@iogearbox.net> wrote:
>
> [ +Dan ]
>
> On 02/07/2019 08:43 AM, Björn Töpel wrote:
> > Den tors 7 feb. 2019 kl 06:38 skrev Davidlohr Bueso <dave@stgolabs.net>:
> >>
> >> Holding mmap_sem exclusively for a gup() is an overkill.
> >> Lets replace the call for gup_fast() and let the mm take
> >> it if necessary.
> >>
> >> Cc: David S. Miller <davem@davemloft.net>
> >> Cc: Bjorn Topel <bjorn.topel@intel.com>
> >> Cc: Magnus Karlsson <magnus.karlsson@intel.com>
> >> CC: netdev@vger.kernel.org
> >> Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
> >> ---
> >>  net/xdp/xdp_umem.c | 6 ++----
> >>  1 file changed, 2 insertions(+), 4 deletions(-)
> >>
> >> diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
> >> index 5ab236c5c9a5..25e1e76654a8 100644
> >> --- a/net/xdp/xdp_umem.c
> >> +++ b/net/xdp/xdp_umem.c
> >> @@ -265,10 +265,8 @@ static int xdp_umem_pin_pages(struct xdp_umem *umem)
> >>         if (!umem->pgs)
> >>                 return -ENOMEM;
> >>
> >> -       down_write(&current->mm->mmap_sem);
> >> -       npgs = get_user_pages(umem->address, umem->npgs,
> >> -                             gup_flags, &umem->pgs[0], NULL);
> >> -       up_write(&current->mm->mmap_sem);
> >> +       npgs = get_user_pages_fast(umem->address, umem->npgs,
> >> +                                  gup_flags, &umem->pgs[0]);
> >>
> >
> > Thanks for the patch!
> >
> > The lifetime of the pinning is similar to RDMA umem mapping, so isn't
> > gup_longterm preferred?
>
> Seems reasonable from reading what gup_longterm seems to do. Davidlohr
> or Dan, any thoughts on the above?

Yes, any gup where the lifetime of the corresponding put_page() is
under direct control of userspace should be using the _longterm flavor
to coordinate be careful in the presence of dax. In the dax case there
is no page cache indirection to coordinate truncate vs page pins. Ira
is looking to add a "fast + longterm" flavor for cases like this.

RE: [PATCH 1/2] xsk: do not use mmap_sem

[Weiny, Ira]

> > >> ---
> > >>  net/xdp/xdp_umem.c | 6 ++----
> > >>  1 file changed, 2 insertions(+), 4 deletions(-)
> > >>
> > >> diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c index
> > >> 5ab236c5c9a5..25e1e76654a8 100644
> > >> --- a/net/xdp/xdp_umem.c
> > >> +++ b/net/xdp/xdp_umem.c
> > >> @@ -265,10 +265,8 @@ static int xdp_umem_pin_pages(struct
> xdp_umem *umem)
> > >>         if (!umem->pgs)
> > >>                 return -ENOMEM;
> > >>
> > >> -       down_write(&current->mm->mmap_sem);
> > >> -       npgs = get_user_pages(umem->address, umem->npgs,
> > >> -                             gup_flags, &umem->pgs[0], NULL);
> > >> -       up_write(&current->mm->mmap_sem);
> > >> +       npgs = get_user_pages_fast(umem->address, umem->npgs,
> > >> +                                  gup_flags, &umem->pgs[0]);
> > >>
> > >
> > > Thanks for the patch!
> > >
> > > The lifetime of the pinning is similar to RDMA umem mapping, so
> > > isn't gup_longterm preferred?
> >
> > Seems reasonable from reading what gup_longterm seems to do. Davidlohr
> > or Dan, any thoughts on the above?
> 
> Yes, any gup where the lifetime of the corresponding put_page() is under
> direct control of userspace should be using the _longterm flavor to
> coordinate be careful in the presence of dax. In the dax case there is no page
> cache indirection to coordinate truncate vs page pins. Ira is looking to add a
> "fast + longterm" flavor for cases like this.

Yes I'm just about ready with a small patch set to add a GUP fast longterm.

I think this should wait for that series.

Ira
 

Re: [PATCH v2] xsk: share the mmap_sem for page pinning

[Daniel Borkmann]

On 02/11/2019 05:15 PM, Davidlohr Bueso wrote:
> Holding mmap_sem exclusively for a gup() is an overkill.
> Lets share the lock and replace the gup call for gup_longterm(),
> as it is better suited for the lifetime of the pinning.
> 
> Cc: David S. Miller <davem@davemloft.net>
> Cc: Bjorn Topel <bjorn.topel@intel.com>
> Cc: Magnus Karlsson <magnus.karlsson@intel.com>
> CC: netdev@vger.kernel.org
> Signed-off-by: Davidlohr Bueso <dbueso@suse.de>

Applied, thanks!