This serie of patches aims to reduce the usage of the AioContexlock in block-copy, by introducing smaller granularity locks thus on making the block layer thread safe.
This serie depends on my previous serie that brings thread safety to the smaller API used by block-copy, like ratelimit, progressmeter abd co-shared-resource. What's missing for block-copy to be fully thread-safe is fixing the CoSleep API to allow cross-thread sleep and wakeup. Paolo is working on it. Patch 1 introduces the .method field in BlockCopyState, to be used instead of .use_copy_range, .copy_size and .zeros. Patch 2-3 provide comments and refactoring in preparation to the lock added in patch 4 on BlockCopyTask, BlockCopyCallState and BlockCopyState. Patch 5 uses load_acquire/store_release to make sure BlockCopyCallState OUT fields are updated before finished is set to true. Based-on: <20210518094058.25952-1-eespo...@redhat.com> Signed-off-by: Emanuele Giuseppe Esposito <eespo...@redhat.com> --- v3: * Use a single lock instead of two [Paolo, Vladimir] * Extend lock to protect also BdrvDirtyBitmap API [Vladimir] * Drop patch 6 (set .method as atomic) since with current refactoring it can be simply included in the near critical sections protected by the lock Emanuele Giuseppe Esposito (4): block-copy: improve comments of BlockCopyTask and BlockCopyState types and functions block-copy: move progress_set_remaining in block_copy_task_end block-copy: add a CoMutex block-copy: atomic .cancelled and .finished fields in BlockCopyCallState Paolo Bonzini (1): block-copy: streamline choice of copy_range vs. read/write block/block-copy.c | 319 +++++++++++++++++++++++++-------------------- 1 file changed, 181 insertions(+), 138 deletions(-) -- 2.30.2