linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RFC][PATCH 0/7] kref improvements
@ 2016-11-14 17:39 Peter Zijlstra
  2016-11-14 17:39 ` [RFC][PATCH 1/7] kref: Add KREF_INIT() Peter Zijlstra
                   ` (7 more replies)
  0 siblings, 8 replies; 110+ messages in thread
From: Peter Zijlstra @ 2016-11-14 17:39 UTC (permalink / raw)
  To: gregkh, keescook, will.deacon, elena.reshetova, arnd, tglx,
	mingo, hpa, dave
  Cc: linux-kernel

This series unfscks kref and then implements it in terms of refcount_t.

x86_64-allyesconfig compile tested and boot tested with my regular config.

refcount_t is as per the previous thread, it BUGs on over-/underflow and
saturates at UINT_MAX, such that if we ever overflow, we'll never free again.

^ permalink raw reply	[flat|nested] 110+ messages in thread
* Re: [RFC][PATCH 2/7] kref: Add kref_read()
@ 2016-11-16 20:08 Alexei Starovoitov
  2016-11-17  8:53 ` Peter Zijlstra
  0 siblings, 1 reply; 110+ messages in thread
From: Alexei Starovoitov @ 2016-11-16 20:08 UTC (permalink / raw)
  To: Kees Cook
  Cc: Peter Zijlstra, Greg KH, Will Deacon, Reshetova, Elena,
	Arnd Bergmann, Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
	David Windsor, LKML, Daniel Borkmann

On Wed, Nov 16, 2016 at 10:58 AM, Kees Cook <keescook@chromium.org> wrote:
> On Wed, Nov 16, 2016 at 2:09 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>> On Tue, Nov 15, 2016 at 12:53:35PM -0800, Kees Cook wrote:
>>>
>>> What should we do about things like this (bpf_prog_put() and callbacks
>>> from kernel/bpf/syscall.c):
>>>
>>>
>>> static void bpf_prog_uncharge_memlock(struct bpf_prog *prog)
>>> {
>>>         struct user_struct *user = prog->aux->user;
>>>
>>>         atomic_long_sub(prog->pages, &user->locked_vm);
>>>         free_uid(user);
>>> }
>>>
>>> static void __bpf_prog_put_rcu(struct rcu_head *rcu)
>>> {
>>>         struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu);
>>>
>>>         free_used_maps(aux);
>>>         bpf_prog_uncharge_memlock(aux->prog);
>>>         bpf_prog_free(aux->prog);
>>> }
>>>
>>> void bpf_prog_put(struct bpf_prog *prog)
>>> {
>>>         if (atomic_dec_and_test(&prog->aux->refcnt))
>>>                 call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu);
>>> }
>>>
>>>
>>> Not only do we want to protect prog->aux->refcnt, but I think we want
>>> to protect user->locked_vm too ... I don't think it's sane for
>>> user->locked_vm to be a stats_t ?
>>
>> Why would you want to mess with locked_vm? You seem of the opinion that
>> everything atomic_t is broken, this isn't the case.
>
> What I mean to say is that while the refcnt here should clearly be
> converted to kref or refcount_t, it looks like locked_vm should become
> a new stats_t. However, it seems weird for locked_vm to ever wrap
> either...

I prefer to avoid 'fixing' things that are not broken.
Note, prog->aux->refcnt already has explicit checks for overflow.
locked_vm is used for resource accounting and not refcnt,
so I don't see issues there either.

^ permalink raw reply	[flat|nested] 110+ messages in thread

end of thread, other threads:[~2017-01-30 13:41 UTC | newest]

Thread overview: 110+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-11-14 17:39 [RFC][PATCH 0/7] kref improvements Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 1/7] kref: Add KREF_INIT() Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 2/7] kref: Add kref_read() Peter Zijlstra
2016-11-14 18:16   ` Christoph Hellwig
2016-11-15  7:28     ` Greg KH
2016-11-15  7:47       ` Peter Zijlstra
2016-11-15  8:37       ` [PATCH] printk, locking/atomics, kref: Introduce new %pAr and %pAk format string options for atomic_t and 'struct kref' Ingo Molnar
2016-11-15  8:43         ` [PATCH v2] " Ingo Molnar
2016-11-15  9:21           ` Peter Zijlstra
2016-11-15  9:41             ` [PATCH v3] printk, locking/atomics, kref: Introduce new %pAa " Ingo Molnar
2016-11-15 10:10           ` [PATCH v2] printk, locking/atomics, kref: Introduce new %pAr " kbuild test robot
2016-11-15 16:42         ` [PATCH] " Linus Torvalds
2016-11-16  8:13           ` Ingo Molnar
2016-11-15  7:33   ` [RFC][PATCH 2/7] kref: Add kref_read() Greg KH
2016-11-15  8:03     ` Peter Zijlstra
2016-11-15 20:53       ` Kees Cook
2016-11-16  8:21         ` Greg KH
2016-11-16 10:10           ` Peter Zijlstra
2016-11-16 10:18             ` Greg KH
2016-11-16 10:11           ` Daniel Borkmann
2016-11-16 10:19             ` Greg KH
2016-11-16 10:09         ` Peter Zijlstra
2016-11-16 18:58           ` Kees Cook
2016-11-17  8:34             ` Peter Zijlstra
2016-11-17 12:30               ` David Windsor
2016-11-17 12:43                 ` Peter Zijlstra
2016-11-17 13:01                   ` Reshetova, Elena
2016-11-17 13:22                     ` Peter Zijlstra
2016-11-17 15:42                       ` Reshetova, Elena
2016-11-17 18:02                       ` Reshetova, Elena
2016-11-17 19:10                         ` Peter Zijlstra
2016-11-17 19:29                         ` Peter Zijlstra
2016-11-17 19:34               ` Kees Cook
2016-11-14 17:39 ` [RFC][PATCH 3/7] kref: Kill kref_sub() Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 4/7] kref: Use kref_get_unless_zero() more Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 5/7] kref: Implement kref_put_lock() Peter Zijlstra
2016-11-14 20:35   ` Kees Cook
2016-11-15  7:50     ` Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 6/7] kref: Avoid more abuse Peter Zijlstra
2016-11-14 17:39 ` [RFC][PATCH 7/7] kref: Implement using refcount_t Peter Zijlstra
2016-11-15  8:40   ` Ingo Molnar
2016-11-15  9:47     ` Peter Zijlstra
2016-11-15 10:03       ` Ingo Molnar
2016-11-15 10:46         ` Peter Zijlstra
2016-11-15 13:03           ` Ingo Molnar
2016-11-15 18:06             ` Kees Cook
2016-11-15 19:16               ` Peter Zijlstra
2016-11-15 19:23                 ` Kees Cook
2016-11-16  8:31                   ` Ingo Molnar
2016-11-16  8:51                     ` Greg KH
2016-11-16  9:07                       ` Ingo Molnar
2016-11-16  9:24                         ` Greg KH
2016-11-16 10:15                     ` Peter Zijlstra
2016-11-16 18:55                       ` Kees Cook
2016-11-17  8:33                         ` Peter Zijlstra
2016-11-17 19:50                           ` Kees Cook
2016-11-16 18:41                     ` Kees Cook
2016-11-15 12:33   ` Boqun Feng
2016-11-15 13:01     ` Peter Zijlstra
2016-11-15 14:19       ` Boqun Feng
2016-11-17  9:28         ` Peter Zijlstra
2016-11-17  9:48           ` Boqun Feng
2016-11-17 10:29             ` Peter Zijlstra
2016-11-17 10:39               ` Peter Zijlstra
2016-11-17 11:03                 ` Greg KH
2016-11-17 12:48                   ` Peter Zijlstra
     [not found]               ` <CAL0jBu-GnREUPSX4kUDp-Cc8ZGp6+Cb2q0HVandswcLzPRnChQ@mail.gmail.com>
2016-11-17 12:08                 ` Peter Zijlstra
2016-11-17 12:08           ` Will Deacon
2016-11-17 16:11             ` Peter Zijlstra
2016-11-17 16:36               ` Will Deacon
2016-11-18  8:26                 ` Boqun Feng
2016-11-18 10:16                   ` Will Deacon
2016-11-18 10:07   ` Reshetova, Elena
2016-11-18 11:37     ` Peter Zijlstra
2016-11-18 17:06       ` Will Deacon
2016-11-18 18:57         ` Peter Zijlstra
2016-11-21  4:06         ` Boqun Feng
2016-11-21  7:48           ` Ingo Molnar
2016-11-21  8:38             ` Boqun Feng
2016-11-21  8:44       ` Boqun Feng
2016-11-21  9:02         ` Peter Zijlstra
2016-11-21  9:37           ` Boqun Feng
2016-11-18 10:47   ` Reshetova, Elena
2016-11-18 10:52     ` Peter Zijlstra
2016-11-18 16:58       ` Reshetova, Elena
2016-11-18 18:53         ` Peter Zijlstra
2016-11-19  7:14           ` Reshetova, Elena
2016-11-19 11:45             ` Peter Zijlstra
2017-01-26 23:14   ` Kees Cook
2017-01-27  9:58     ` Peter Zijlstra
2017-01-27 21:07       ` Kees Cook
2017-01-30 13:40         ` Peter Zijlstra
2016-11-15  7:27 ` [RFC][PATCH 0/7] kref improvements Greg KH
2016-11-15  7:42   ` Ingo Molnar
2016-11-15 15:05     ` Greg KH
2016-11-15  7:48   ` Peter Zijlstra
2016-11-16 20:08 [RFC][PATCH 2/7] kref: Add kref_read() Alexei Starovoitov
2016-11-17  8:53 ` Peter Zijlstra
2016-11-17 16:19   ` Alexei Starovoitov
2016-11-17 16:34     ` Thomas Gleixner
2016-11-18 17:33     ` Reshetova, Elena
2016-11-19  3:47       ` Alexei Starovoitov
2016-11-21  8:18         ` Reshetova, Elena
2016-11-21 12:47       ` David Windsor
2016-11-21 15:39         ` Reshetova, Elena
2016-11-21 15:49           ` Peter Zijlstra
2016-11-21 16:00             ` Peter Zijlstra
2016-11-21 19:27               ` Reshetova, Elena
2016-11-21 20:12                 ` David Windsor
2016-11-22 10:37                   ` Peter Zijlstra

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).