On Mon, Oct 21, 2019 at 09:31:13AM -0700, Tom Rix wrote: > When preempt rt is full, softirq and interrupts run in kthreads. So it > is possible for the tasklet to sleep and for its queue to get modified > while it sleeps.
This is ridiculous. The network stack is full of assumptions like this. So I think we need to fix preempt rt instead because you can't make a major change like this without auditing the entire kernel first rather than relying on a whack-a-mole approach. Cheers, -- Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
