mbox series

[for-4.9,0/5] l2tp session lookup fixes

Message ID 1498492039-26905-1-git-send-email-amit.pundir@linaro.org
Headers show
Series l2tp session lookup fixes | expand

Message

Amit Pundir June 26, 2017, 3:47 p.m. UTC 
Hi,

Kindly consider/review following l2tp fixes for stable 4.9.y.

This patchset fix race in l2tp sessions, avoid duplicate sessions
and probable "use after free" bugs. Here is the original submission
https://www.mail-archive.com/netdev@vger.kernel.org/msg161040.html

Cherry-picked and build tested on Linux 4.9.34 for
ARCH=x86/x86_64 + allmodconfig.

Regards,
Amit Pundir


Guillaume Nault (5):
  l2tp: fix race in l2tp_recv_common()
  l2tp: ensure session can't get removed during pppol2tp_session_ioctl()
  l2tp: fix duplicate session creation
  l2tp: hold session while sending creation notifications
  l2tp: take a reference on sessions used in genetlink handlers

 net/l2tp/l2tp_core.c    | 152 ++++++++++++++++++++++++++++++++++++++----------
 net/l2tp/l2tp_core.h    |   6 +-
 net/l2tp/l2tp_eth.c     |  10 +---
 net/l2tp/l2tp_ip.c      |  17 ++++--
 net/l2tp/l2tp_ip6.c     |  18 ++++--
 net/l2tp/l2tp_netlink.c |  45 +++++++++-----
 net/l2tp/l2tp_ppp.c     |  75 +++++++++++++-----------
 7 files changed, 222 insertions(+), 101 deletions(-)

-- 
2.7.4

Comments

Guillaume Nault June 27, 2017, 7:57 a.m. UTC | #1 
On Mon, Jun 26, 2017 at 09:17:14PM +0530, Amit Pundir wrote:
> Hi,

> 

> Kindly consider/review following l2tp fixes for stable 4.9.y.

> 

> This patchset fix race in l2tp sessions, avoid duplicate sessions

> and probable "use after free" bugs. Here is the original submission

> https://www.mail-archive.com/netdev@vger.kernel.org/msg161040.html

> 

> Cherry-picked and build tested on Linux 4.9.34 for

> ARCH=x86/x86_64 + allmodconfig.

> 

Looks good to me. Thanks.
Greg Kroah-Hartman July 3, 2017, 8:48 a.m. UTC | #2 
On Mon, Jun 26, 2017 at 09:17:14PM +0530, Amit Pundir wrote:
> Hi,

> 

> Kindly consider/review following l2tp fixes for stable 4.9.y.

> 

> This patchset fix race in l2tp sessions, avoid duplicate sessions

> and probable "use after free" bugs. Here is the original submission

> https://www.mail-archive.com/netdev@vger.kernel.org/msg161040.html

> 

> Cherry-picked and build tested on Linux 4.9.34 for

> ARCH=x86/x86_64 + allmodconfig.


Thanks for these, should they also go into 4.4 and older kernels?  If
so, care to make up patches for them?

thanks,

greg k-h
Amit Pundir July 3, 2017, 8:56 a.m. UTC | #3 
On 3 July 2017 at 14:18, Greg KH <gregkh@linuxfoundation.org> wrote:
> On Mon, Jun 26, 2017 at 09:17:14PM +0530, Amit Pundir wrote:

>> Hi,

>>

>> Kindly consider/review following l2tp fixes for stable 4.9.y.

>>

>> This patchset fix race in l2tp sessions, avoid duplicate sessions

>> and probable "use after free" bugs. Here is the original submission

>> https://www.mail-archive.com/netdev@vger.kernel.org/msg161040.html

>>

>> Cherry-picked and build tested on Linux 4.9.34 for

>> ARCH=x86/x86_64 + allmodconfig.

>

> Thanks for these, should they also go into 4.4 and older kernels?  If

> so, care to make up patches for them?


Tried on 4.4 but one or more net-namespace patches are missing in
between. So I was not sure if I should pull them and other dependent
patches (if any) into this patchset.

Regards,
Amit Pundir

>

> thanks,

>

> greg k-h
Greg Kroah-Hartman July 3, 2017, 9:03 a.m. UTC | #4 
On Mon, Jul 03, 2017 at 02:26:11PM +0530, Amit Pundir wrote:
> On 3 July 2017 at 14:18, Greg KH <gregkh@linuxfoundation.org> wrote:

> > On Mon, Jun 26, 2017 at 09:17:14PM +0530, Amit Pundir wrote:

> >> Hi,

> >>

> >> Kindly consider/review following l2tp fixes for stable 4.9.y.

> >>

> >> This patchset fix race in l2tp sessions, avoid duplicate sessions

> >> and probable "use after free" bugs. Here is the original submission

> >> https://www.mail-archive.com/netdev@vger.kernel.org/msg161040.html

> >>

> >> Cherry-picked and build tested on Linux 4.9.34 for

> >> ARCH=x86/x86_64 + allmodconfig.

> >

> > Thanks for these, should they also go into 4.4 and older kernels?  If

> > so, care to make up patches for them?

> 

> Tried on 4.4 but one or more net-namespace patches are missing in

> between. So I was not sure if I should pull them and other dependent

> patches (if any) into this patchset.


Ok, if you don't think so, that's fine with me, it's just that the
patches said they fixed something that old.  If no one has complained
yet about 4.4 being broken for these issues, don't worry about it :)

thanks,

greg k-h