Change Details

When we get topological removal events, fail the devices faster rather than retrying events. When we've made the decision to remove the device, return CAM_DEV_NOT_THERE instead of a milder error to signal to cam_periph_error that it should return ENXIO up the stack.

Before issing the REMOVE_DEVICE command to the firmware, make sure that all commands have completed. When we're removing a target that's not a volume, advertise up the stack that it's actually gone, as opposed to having a transient selection error we should retry. Do this both in the action routine, and when we get a notification of an aborted command. Add assert that all commands are complete to REMOVE_DEVICE completion routine. We shouldn't have and commands in the queue because we've waited for them all. Any commands that make it into our action routine after we mark the target in removal will complete immediately with an error. It's not OK to complete commands before we send the REMOVE_DEVICE. Instead, make sure that all pending commands are complete before sending that. By trying to second guess the firmware here, we run the risk of completing commands twice, which leads to corruption. This removes the forced completion of commands introduced in r218811. So it's a partial backout of that commit, but replaces it with a more robust mechanism.

When we get topological removal eventsBefore issing the REMOVE_DEVICE command to the firmware, make sure that all commands have completed. When we're removing a target that's not a volume, advertise up the stack that it's actually gone, as opposed to having a transient selection error we should retry. Do this both in the action routine, and when we get a notification of an aborted command. Add assert that all commands are complete to REMOVE_DEVICE completion routine. We shouldn't have and commands in the queue because we've waited for them all. Any commands that make it into our action routine after we mark the target in removal will complete immediately with an error. It's not OK to complete commands before we send the REMOVE_DEVICE. Instead, fail the devices faster rathermake sure that all pending commands are complete before sending than ret. By trying events.to second guess the firmware here, When we've madewe run the decision to remove the devicerisk of completing commands twice, return CAM_DEV_NOT_THERE instead of a milder error to signal to cam_periph_error that it should return ENXIO up the stackwhich leads to corruption. This removes the forced completion of commands introduced in r218811. So it's a partial backout of that commit, but replaces it with a more robust mechanism.