697  * queue a task_work based retry of the operation, attempting to copy the data
699  * do a thread based blocking retry of the operation. That's the unexpected
723  * based retry. If we return false here, the request is handed to the async
724  * worker threads for retry. If we're doing buffered reads on a regular file,
725  * we prepare a private wait_page_queue entry and retry the operation. This
728  * that callback, io_uring uses task_work to setup a retry of the operation.
729  * That retry will attempt the buffered read again. The retry will generally
731  * async worker threads for a blocking retry.
740 	/* never retry for NOWAIT, we just complete with -EAGAIN */  in io_rw_should_retry()
805 	 * If the file is marked O_NONBLOCK, still allow retry for it if it  in io_rw_init_file()
807 	 * reliably. If not, or it IOCB_NOWAIT is set, don't retry.  in io_rw_init_file()
870 	 * retry from blocking context for either.  in __io_read()
880 		/* IOPOLL retry should happen for io-wq threads */  in __io_read()
883 		/* no retry on NONBLOCK nor RWF_NOWAIT */  in __io_read()
891 		/* read all, failed, already did sync or don't want to retry */  in __io_read()
914 		/* if we can retry, do so with the callbacks armed */  in __io_read()
922 		 * Now retry read with the IOCB_WAITQ parts set in the iocb. If  in __io_read()
925 		 * here, and if we do, then just retry at the new offset.  in __io_read()
930 		/* we got some bytes, but not all. retry. */  in __io_read()
995 				 * Force retry, as we might have more data to  in io_read_mshot()
1091 	 * retry them without IOCB_NOWAIT.  in io_write()
1095 	/* no retry on NONBLOCK nor RWF_NOWAIT */  in io_write()
1099 		/* IOPOLL retry should happen for io-wq threads */  in io_write()