Messages in this thread | | | Subject | Re: 2.6.8.1-mm4 - more cpu hotplug breakage | From | Rusty Russell <> | Date | Thu, 26 Aug 2004 12:54:47 +1000 |
| |
On Thu, 2004-08-26 at 09:09, Nathan Lynch wrote: > My apologies if this is getting annoying, but it occurred to me that any > user of stop_machine_run is broken similarly... which means that > unloading a module will also hang your machine. I have verified this on > 2.6.8.1-mm4 on ppc64.
My apologies for not responding earlier.
I've been chasing this, and I realized that both my previous patch and the one in -mm4 are broken. They fix the case where we release_task ourselves, then a CPU goes down, but not the case where our parent release_tasks us and the CPU goes down.
The correct fix is for the hotplug CPU code to scan the dead CPU's runqueue as well, and migrate those.
I'll patch soon, Rusty. -- Anyone who quotes me in their signature is an idiot -- Rusty Russell
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |