Re: [RFC] bdrv_flush: only use fast path when in owned AioContext

2020-05-12 Thread Stefan Reiter
On 5/12/20 1:32 PM, Kevin Wolf wrote: Am 12.05.2020 um 12:57 hat Kevin Wolf geschrieben: Am 11.05.2020 um 18:50 hat Stefan Reiter geschrieben: Just because we're in a coroutine doesn't imply ownership of the context of the flushed drive. In such a case use the slow path which explicitly enters

Re: [RFC] bdrv_flush: only use fast path when in owned AioContext

2020-05-12 Thread Kevin Wolf
Am 12.05.2020 um 12:57 hat Kevin Wolf geschrieben: > Am 11.05.2020 um 18:50 hat Stefan Reiter geschrieben: > > Just because we're in a coroutine doesn't imply ownership of the context > > of the flushed drive. In such a case use the slow path which explicitly > > enters bdrv_flush_co_entry in the c

Re: [RFC] bdrv_flush: only use fast path when in owned AioContext

2020-05-12 Thread Kevin Wolf
Am 11.05.2020 um 18:50 hat Stefan Reiter geschrieben: > Just because we're in a coroutine doesn't imply ownership of the context > of the flushed drive. In such a case use the slow path which explicitly > enters bdrv_flush_co_entry in the correct AioContext. > > Signed-off-by: Stefan Reiter > ---

[RFC] bdrv_flush: only use fast path when in owned AioContext

2020-05-11 Thread Stefan Reiter
Just because we're in a coroutine doesn't imply ownership of the context of the flushed drive. In such a case use the slow path which explicitly enters bdrv_flush_co_entry in the correct AioContext. Signed-off-by: Stefan Reiter --- We've experienced some lockups in this codepath when taking snap