Re: [syzbot] general protection fault in skb_dequeue (3)

[David Howells]

Interestingly, the problem also goes away if the ftruncate is removed from the
test program.

Are pages ever spliced from the pipe directly into the pagecache on both xfs
and ext4?  I'm not sure whether that could happen as the test seeks 2048 bytes
in before doing the sendfile().

David