The scale tests are verifying behavior of mlxsw when number of instances of
some resource reaches the ASIC capacity. The number of instances is
referred to as "target" number.
No scale tests so far needed to know this target number to clean up. E.g.
the tc_flower simply removes the clsact qdisc that all the tested filters
are hooked onto, and that takes care of collecting all the filters.
However, for the RIF counter test, which is being added in a future patch,
VLAN netdevices are created. These are created as part of the test, but of
course the cleanup needs to undo them again. For that it needs to know how
many there were. To support this usage, pass the target number to the
cleanup callback.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
                else
                        log_test "'$current_test' overflow $target"
                fi
-               ${current_test}_cleanup
+               ${current_test}_cleanup $target
                devlink_reload
                RET_FIN=$(( RET_FIN || RET ))
        done
 
                        else
                                log_test "'$current_test' [$profile] overflow $target"
                        fi
-                       ${current_test}_cleanup
+                       ${current_test}_cleanup $target
                        RET_FIN=$(( RET_FIN || RET ))
                done
        done