virtio-scsi: Cleanup of I/Os that never started
commite9c0f0f58ad0a41c3c4b19e1911cfe095afc09ca
authorEric Farman <farman@linux.vnet.ibm.com>
Tue, 14 Jan 2014 19:16:25 +0000 (14 14:16 -0500)
committerPaolo Bonzini <pbonzini@redhat.com>
Wed, 15 Jan 2014 09:36:18 +0000 (15 10:36 +0100)
treed2b82107f015daa573986a22d14bd9b5c485784c
parent33325a53f15ab5370e1917b2a11cadffc77c5a52
virtio-scsi: Cleanup of I/Os that never started

There is still a small window that occurs when a cancel I/O affects
an asynchronous I/O operation that hasn't started.  In other words,
when the residual data length equals the expected data length.

Today, the routine virtio_scsi_command_complete fails because the
VirtIOSCSIReq pointer (from the hba_private field in SCSIRequest)
was cleared earlier when virtio_scsi_complete_req was called by
the virtio_scsi_request_cancelled routine.  As a result, the
virtio_scsi_command_complete routine needs to simply return when
it is processing a SCSIRequest block that was marked canceled.

Signed-off-by: Eric Farman <farman@linux.vnet.ibm.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
hw/scsi/virtio-scsi.c