lkml.org 
[lkml]   [2017]   [Dec]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    From
    SubjectRe: [PATCH 4.14 108/159] kvm, mm: account kvm related kmem slabs to kmemcg
    Date
    On Fri, Dec 22, 2017 at 02:06:07PM +0100, Michal Hocko wrote:
    >On Fri 22-12-17 13:41:22, Greg KH wrote:
    >> On Fri, Dec 22, 2017 at 10:34:07AM +0100, Michal Hocko wrote:
    >> > On Fri 22-12-17 09:46:33, Greg KH wrote:
    >> > > 4.14-stable review patch. If anyone has any objections, please let me know.
    >> > >
    >> > > ------------------
    >> > >
    >> > > From: Shakeel Butt <shakeelb@google.com>
    >> > >
    >> > >
    >> > > [ Upstream commit 46bea48ac241fe0b413805952dda74dd0c09ba8b ]
    >> > >
    >> > > The kvm slabs can consume a significant amount of system memory
    >> > > and indeed in our production environment we have observed that
    >> > > a lot of machines are spending significant amount of memory that
    >> > > can not be left as system memory overhead. Also the allocations
    >> > > from these slabs can be triggered directly by user space applications
    >> > > which has access to kvm and thus a buggy application can leak
    >> > > such memory. So, these caches should be accounted to kmemcg.
    >> > >
    >> > > Signed-off-by: Shakeel Butt <shakeelb@google.com>
    >> > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
    >> > > Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
    >> > > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
    >> >
    >> > The patch is not marked for stable, neither it fixes an existing bug.
    >> > It is a nice to have thing for sure but I am wondering how this got
    >> > through stable-filter.
    >>
    >> Sasha picked it out, and it seemed like a sane thing to backport. If
    >> you think it's not worthy, I'll gladly drop it, but it seemed like such
    >> a simple bugfix to include.
    >
    >It is not that I would have some specific concerns about this particular
    >patch. It is more of a worry about the overal process. I thought that
    >_any_ patch backported to the stable tree would require a specific bug
    >to be fixed or in exceptional cases a performance issue. I have
    >experienced this pushback myself when trying to push "no real bug report
    >but better to have this plugged" patches.
    >
    >So something has apparently changed in the process, I just haven't
    >noticed it. I am worried this might lead to more regression in future.
    >Not that my worry counts all that much as I am not a stable kernel user
    >though. So this is just my 2c worth of worry.

    The way I see it is that stable commits are supposed to fix a bug that
    a user can hit/exploit, it doesn't have to have an actual user
    complaining about it.

    For this particular commit, the way I read it is that a user can avoid
    his kmemcg limits (maybe maliciously), which would qualify as an
    actual bug we want to get fixed.

    --

    Thanks,
    Sasha
    \
     
     \ /
      Last update: 2017-12-22 18:42    [W:4.529 / U:0.352 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site