On 03/28/2019 04:56 PM, Linus Torvalds wrote: > On Thu, Mar 28, 2019 at 1:47 PM Linus Torvalds > <[email protected]> wrote: >> On Thu, Mar 28, 2019 at 11:12 AM Waiman Long <[email protected]> wrote: >>> With the merging of owner into count for x86-64, there is only 16 bits >>> left for reader count. It is theoretically possible for an application to >>> cause more than 64k readers to acquire a rwsem leading to count overflow. >> Ahh, and here's the thing that makes 16 bits work for readers. > Hmm. Does it? > > Isn't there a race here? We're adding the READ bias, and then noticing > that it his the guard bit, and then the down_read_failed will make it > all good again. > > But this isn't actually done with preemption disabled, so things > *could* get preempted in between, and if we have a huge run of bad > luck, it can still overflow. > > Ok, so you need to have a 32k series run of bad luck (and hit > *exactly* the right small preemption point window every time), and I'm > certainly willing to say "yeah, not an issue", but maybe it's still > worth at least documenting? > > Linus
I think it is theoretically possible that this can happen, but I doubt we will ever see that. Will document that possibility in the comment. Thanks, Longman
