> > qla2300 0000:05:0d.0: LOOP DOWN detected. > > Debug: sleeping function called from invalid context at include/linux/rwsem.h:43 > > in_atomic():1, irqs_disabled():1 > > [<c0120a74>] __might_sleep+0xa4/0xc0 > > [<c026a466>] device_for_each_child+0x26/0x80 > > [<c02b3180>] target_block+0x0/0x30 > > [<c02bbdae>] fc_remote_port_block+0x2e/0x60 > > [<c02bdbf5>] qla2x00_mark_all_devices_lost+0x55/0x60 > > [<c02c597e>] qla2x00_async_event+0x83e/0xd60 > > [<c011dd2b>] find_busiest_group+0xbb/0x310 > > [<c02cdce4>] sd_rw_intr+0x164/0x320 > > [<c02c4e37>] qla2300_intr_handler+0x77/0x240 > > [<c0144882>] handle_IRQ_event+0x32/0x70 > > Without wanting to making a number of large changes to the qla2xxx > internals to deal with these pre-qualifications, could you try the > following patch (lightly tested with latest linus git tree). > > We'll need to update the fc_remote_port docs in order to account for > this semantic change in device_for_each_child(). > Indeed, it fixed this bug. Thanks for the prompt fix. This leaves us with DM related bugs only :/ Regards, -- christophe varoqui <christophe.varoqui@xxxxxxx>
