On 3/20/21 1:18 PM, Linus Torvalds wrote: > On Sat, Mar 20, 2021 at 10:51 AM Linus Torvalds > <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote: >> >> Alternatively, make it not use >> CLONE_SIGHAND|CLONE_THREAD at all, but that would make it >> unnecessarily allocate its own signal state, so that's "cleaner" but >> not great either. > > Thinking some more about that, it would be problematic for things like > the resource counters too. They'd be much better shared. > > Not adding it to the thread list etc might be clever, but feels a bit too scary. That would be my immediate concern - it might very well be the right long term solution, but I'd be wary of doing it upfront and having weird fallout due to it. > So on the whole I think Jens' minor patches to just not have IO helper > threads accept signals are probably the right thing to do. I do think we should just go with those two - they are simple and straight forward. I'm also totally fine replacing 2/2 with Eric's variant if he prefers that, I've confirmed that it works fine for me as well. -- Jens Axboe
