>>>>> "Nick" == Nick Piggin <npiggin@suse.de> writes:

Nick,

quoted text
>> Filesystems will inevitably have to be integrity-aware for that to
>> work.  And it will be their job to keep the data pages stable during
>> DMA.

Nick> Closing the while it is dirty, while it is being written back
Nick> window still leaves a pretty big window. Also, how do you handle
Nick> mmap writes?  Write protect and checksum the destination page
Nick> after every store? Or leave some window between when the pagecache
Nick> is dirtied and when it is written back? So I don't know whether
Nick> it's worth putting a lot of effort into this case.

I'm mostly interested in the cases where the filesystem acts as a
conduit for integrity metadata from user space. 

I agree the corruption windows inside the kernel are only of moderate
interest.  No filesystems have added support for explicitly protecting a
bio because the block layer's function to do so automatically is just a
few function calls away.

-- 
Martin K. Petersen	Oracle Linux Engineering
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/