On Wed, Aug 22, 2018 at 09:54:48PM -0700, Linus Torvalds wrote: > It honored it for the *normal* case, which is why it took so long to > notice that the TLB shootdown had been broken on x86 when it moved to > the "generic" code. The *normal* case does this all right, and batches > things up, and then when the batch fills up it does a > tlb_table_flush() which does the TLB flush and schedules the actual > freeing. > > But there were two cases that *didn't* do that. The special "I'm the > only thread" fast case, and the "oops I ran out of memory, so now I'll > fake it, and just synchronize with page twalkers manually, and then do > that special direct remove without flushing the tlb". The actual RCU batching case was also busted; there was no guarantee that by the time we run the RCU callbacks the invalidate would've happened. Exceedingly unlikely, but no guarantee. So really, all 3 cases in tlb_remove_table() were busted in this respect.
