Re: KASAN: out-of-bounds Read in rds_cong_queue_updates (2)

[Dmitry Vyukov <dvyukov@google.com>]

On Wed, Jun 13, 2018 at 12:19 PM, Sowmini Varadhan
<sowmini.varadhan@oracle.com> wrote:
> On (06/13/18 09:52), Dmitry Vyukov wrote:
>> I think this is:
>>
>> #syz dup: KASAN: use-after-free Read in rds_cong_queue_updates
>
> Indeed. We'd had a discussion about getting a dump of threads
> using sysrq (or similar), given the challenges around actually
> getting a crash dump, is that now possible? That will certainly help.

Still no automation around it.

But you can add thread dump on panic locally.

This is a common pattern recently that kernel does not provide enough
information for debugging on bugs. +Testuo
Since we panic on all kernel bugs, perhaps it's panic's work to dump
as much info as possible.


> another missing bit is that we still need the sychronize_net()
> in rds_release(). I realize synchronize_net() is sub-optimal for perf,
> but leaving this existing hole where races can occur in unexpected
> manifestations is not ideal either.
> (See https://www.spinics.net/lists/netdev/msg475074.html for earlier
> discussion thread)