Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-08 Thread Heiko Carstens
On Thu, Apr 07, 2016 at 05:14:00PM +0200, Sebastian Andrzej Siewior wrote: > On 04/06/2016 09:51 PM, Heiko Carstens wrote: > > This fixes the issue that a second cpu_down() will take forever, if > > __cpu_disable() fails. > > Yes. But even without the second take down your CPU isn't complete up.

Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-08 Thread Heiko Carstens
On Thu, Apr 07, 2016 at 05:14:00PM +0200, Sebastian Andrzej Siewior wrote: > On 04/06/2016 09:51 PM, Heiko Carstens wrote: > > This fixes the issue that a second cpu_down() will take forever, if > > __cpu_disable() fails. > > Yes. But even without the second take down your CPU isn't complete up.

Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-07 Thread Sebastian Andrzej Siewior
On 04/06/2016 09:51 PM, Heiko Carstens wrote: > This fixes the issue that a second cpu_down() will take forever, if > __cpu_disable() fails. Yes. But even without the second take down your CPU isn't complete up. > However it does not fix the issue that CPU_DOWN_FAILED will be seen on a >

Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-07 Thread Sebastian Andrzej Siewior
On 04/06/2016 09:51 PM, Heiko Carstens wrote: > This fixes the issue that a second cpu_down() will take forever, if > __cpu_disable() fails. Yes. But even without the second take down your CPU isn't complete up. > However it does not fix the issue that CPU_DOWN_FAILED will be seen on a >

Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-06 Thread Heiko Carstens
On Tue, Apr 05, 2016 at 05:59:04PM +0200, Sebastian Andrzej Siewior wrote: > If we error out in __cpu_disable() (via takedown_cpu() which is > currently the last one that can fail) we don't rollback entirely to > CPUHP_ONLINE (where we started) but to CPUHP_AP_ONLINE_IDLE. This > happens because

Re: [PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-06 Thread Heiko Carstens
On Tue, Apr 05, 2016 at 05:59:04PM +0200, Sebastian Andrzej Siewior wrote: > If we error out in __cpu_disable() (via takedown_cpu() which is > currently the last one that can fail) we don't rollback entirely to > CPUHP_ONLINE (where we started) but to CPUHP_AP_ONLINE_IDLE. This > happens because

[PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-05 Thread Sebastian Andrzej Siewior
If we error out in __cpu_disable() (via takedown_cpu() which is currently the last one that can fail) we don't rollback entirely to CPUHP_ONLINE (where we started) but to CPUHP_AP_ONLINE_IDLE. This happens because the former states were on the target CPU (the AP states) and during the rollback we

[PATCH] cpu/hotplug: fix rollback during error-out in __cpu_disable()

2016-04-05 Thread Sebastian Andrzej Siewior
If we error out in __cpu_disable() (via takedown_cpu() which is currently the last one that can fail) we don't rollback entirely to CPUHP_ONLINE (where we started) but to CPUHP_AP_ONLINE_IDLE. This happens because the former states were on the target CPU (the AP states) and during the rollback we