On Fri, Aug 31, 2018 at 03:55:58PM -0700, Sinan Kaya wrote: > On 8/31/2018 3:33 PM, Keith Busch wrote: > > Darn, you're right. The kernel allocates up to 32 vectors and the port is > > free to divvy them up however it wants amont its supported services. It > > just so happened most of the ports I tested used the same one. There's > > no way to really close this race if they are separate vectors though, > > so maybe just leave this as a 'best effort' approach and update the > > change log accodingly. > > or take the big hammer and move them into a single workqueue? That wouldn't really help if they're different interrupt vectors since they could happen in either order with a delay between the CPU seeing each of them.
