Re: [Qemu-devel] [PATCH for-4.0 v8 3/7] migration: fix the multifd code when receiving less channels

2018-12-19 Thread Fei Li
On 12/19/2018 10:11 PM, Markus Armbruster wrote: Fei Li writes: On 12/13/2018 02:17 PM, Markus Armbruster wrote: Fei Li writes: In our current code, when multifd is used during migration, if there is an error before the destination receives all new channels, the source keeps running,

Re: [Qemu-devel] [PATCH for-4.0 v8 3/7] migration: fix the multifd code when receiving less channels

2018-12-19 Thread Markus Armbruster
Fei Li writes: > On 12/13/2018 02:17 PM, Markus Armbruster wrote: >> Fei Li writes: >> >>> In our current code, when multifd is used during migration, if there >>> is an error before the destination receives all new channels, the >>> source keeps running, however the destination does not exit

Re: [Qemu-devel] [PATCH for-4.0 v8 3/7] migration: fix the multifd code when receiving less channels

2018-12-17 Thread Fei Li
On 12/13/2018 02:17 PM, Markus Armbruster wrote: Fei Li writes: In our current code, when multifd is used during migration, if there is an error before the destination receives all new channels, the source keeps running, however the destination does not exit but keeps waiting until the

Re: [Qemu-devel] [PATCH for-4.0 v8 3/7] migration: fix the multifd code when receiving less channels

2018-12-12 Thread Markus Armbruster
Fei Li writes: > In our current code, when multifd is used during migration, if there > is an error before the destination receives all new channels, the > source keeps running, however the destination does not exit but keeps > waiting until the source is killed deliberately. > > Fix this by

[Qemu-devel] [PATCH for-4.0 v8 3/7] migration: fix the multifd code when receiving less channels

2018-12-11 Thread Fei Li
In our current code, when multifd is used during migration, if there is an error before the destination receives all new channels, the source keeps running, however the destination does not exit but keeps waiting until the source is killed deliberately. Fix this by dumping the specific error and