On Mon, Apr 18, 2022 at 3:20 PM Jens Axboe <axboe@xxxxxxxxx> wrote: > > Here's the thread: > > https://lore.kernel.org/all/1295659049-2688-6-git-send-email-jaxboe@xxxxxxxxxxxx/ > > I'll dig through it in a bit, but here's your reasoning for why it > should not flush on preemption: > > https://lore.kernel.org/all/BANLkTikBEJa7bJJoLFU7NoiEgOjVHVG08A@xxxxxxxxxxxxxx/ Well, that one was triggered by that whole "now it can happen anywhere" worry that people had. So yes, IO patterns are a worry, but I think the bigger worry - even back then - was that preemption points can be pretty much anywhere in the code. Linus
