Messages in this thread | | | Date | Fri, 03 Oct 2008 12:56:32 +0200 | From | Heiko Carstens <> | Subject | [PATCH/RFC 0/4] Add stop_machine_get/put_threads to stop_machine infrastructrue. |
| |
This patch series would allow to convert s390 to the generic IPI interface. We can't to that currently since our etr/stp code relies on the old semantics of smp_call_function that guarantee that the function only returns after all receiving cpus have acknowledged the IPI. That way it is known that all other cpus are running in an interrupt handler with interrupts disabled. This is not true anymore with the generic IPI infrastructure.
So one idea was to use stop_machine in order to synchronize all cpus. Rusty was kind enough to extend it so that it is now possible to run a function on several cpus, instead of just one. However we need to be able to do that without allocating any memory. That's what this patch set is about. It introduces two new interfaces which allow to create the kstop threads in advance so that a subsequent stop_machine call will be able to run without the need to create any threads and therefore avoids the need to allocate any memory.
The two interfaces (see patch 2) are:
stop_machine_get_threads() - create all needed kstop threads in advance. If it is called multiple times it will just increase an internal usecount.
stop_machine_put_threads() - kill all previously created kstop threads, if the internal usecount drops to zero.
Patch 1 is a stop_machine bugfix and is independent of the rest Patch 2 introduces the new proposed interface Patch 3 converts the s390 etr and stp code to the new API Patch 4 converts s390 to the generic IPI interface
Thanks, Heiko
| |