Re: [LSFMM] RDMA data corruption potential during FS writeback

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 05/18/2018 08:51 PM, Dan Williams wrote:
> On Fri, May 18, 2018 at 8:24 PM, Jason Gunthorpe <jgg@xxxxxxxx> wrote:
>> On Fri, May 18, 2018 at 07:33:41PM -0700, John Hubbard wrote:
>>> On 05/18/2018 01:23 PM, Dan Williams wrote:
>>>> On Fri, May 18, 2018 at 10:36 AM, Jason Gunthorpe <jgg@xxxxxxxx> wrote:
>>>>> On Fri, May 18, 2018 at 04:47:48PM +0000, Christopher Lameter wrote:
>>>>>> On Fri, 18 May 2018, Jason Gunthorpe wrote:
>>>>>>
>>>>>>
>>>>>> The newcomer here is RDMA. The FS side is the mainstream use case and has
>>>>>> been there since Unix learned to do paging.
>>>>>
>>>>> Well, it has been this way for 12 years, so it isn't that new.
>>>>>
>>>>> Honestly it sounds like get_user_pages is just a broken Linux
>>>>> API??
>>>>>
>>>>> Nothing can use it to write to pages because the FS could explode -
>>>>> RDMA makes it particularly easy to trigger this due to the longer time
>>>>> windows, but presumably any get_user_pages could generate a race and
>>>>> hit this? Is that right?
>>>
>>> +1, and I am now super-interested in this conversation, because
>>> after tracking down a kernel BUG to this classic mistaken pattern:
>>>
>>>     get_user_pages (on file-backed memory from ext4)
>>>     ...do some DMA
>>>     set_pages_dirty
>>>     put_page(s)
>>
>> Ummm, RDMA has done essentially that since 2005, since when did it
>> become wrong? Do you have some references? Is there some alternative?
>>
>> See __ib_umem_release
>>
>>> ...there is (rarely!) a backtrace from ext4, that disavows ownership of
>>> any such pages.
>>
>> Yes, I've seen that oops with RDMA, apparently isn't actually that
>> rare if you tweak things just right.
>>
>> I thought it was an obscure ext4 bug :(
>>
>>> Because the obvious "fix" in device driver land is to use a dedicated
>>> buffer for DMA, and copy to the filesystem buffer, and of course I will
>>> get *killed* if I propose such a performance-killing approach. But a
>>> core kernel fix really is starting to sound attractive.
>>
>> Yeah, killed is right. That idea totally cripples RDMA.
>>
>> What is the point of get_user_pages FOLL_WRITE if you can't write to
>> and dirty the pages!?!
>>
> 
> You're oversimplifying the problem, here are the details:
> 
> https://www.spinics.net/lists/linux-mm/msg142700.html
> 

Hi Dan,

The thing is, the above still leads to a fairly simple conclusion, which
is: unless something is changed, then no, you cannot do the standard,
simple RDMA thing. Because someone can hit the case that Jan Kara describes
in the link above.

So I think we're all saying the same thing, right? We need to fix something so
that this pattern actually works in all cases? 

I just want to be sure that this doesn't get characterized as "oh this is 
just a special case that you might need to avoid". From my view (which at
the moment is sort of RDMA-centric, for this bug), I don't see a way to 
avoid the problem, other than avoiding calling get_user_pages on file-backed
memory--and even *that* is darned difficult.

Please tell me I'm wildly wrong, though--I'd really love to be wildly wrong
here. :)

thanks,
-- 
John Hubbard
NVIDIA




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux