Re: 2.6.13-rc7-rt1

[Daniel Walker <dwalker@mvista.com>]

On Thu, 2005-08-25 at 23:54 +0200, Ingo Molnar wrote:
> * Daniel Walker <dwalker@mvista.com> wrote:
> 
> > @@ -257,6 +257,7 @@ void check_preempt_wakeup(struct task_st
> >  	 * hangs and race conditions.
> >  	 */
> >  	if (!preempt_count() &&
> > +		!__raw_irqs_disabled() &&
> >  		p->prio < current->prio &&
> >  		rt_task(p) &&
> >  		(current->rcu_read_lock_nesting != 0 ||
> 
> did you get a false positive? If yes, in what code/driver?


Yes, one was in trigger_softirqs() , the other in init_sd . Both traces
below.

BUG: swapper/1, possible wake_up race on softirq-scsi/3/31

Call Trace:<ffffffff80121ade>{wake_up_process+17} <ffffffff8012ddec>{trigger_softirqs+76}
       <ffffffff80379fed>{__scsi_done+101} <ffffffff803e55b4>{ata_scsi_rbuf_fill+114}
       <ffffffff80379fff>{scsi_done+0} <ffffffff803e5e0e>{ata_scsi_simulate+320}
       <ffffffff80379fff>{scsi_done+0} <ffffffff803e5f08>{ata_scsi_queuecmd+228}
       <ffffffff8037a1ff>{scsi_dispatch_cmd+488} <ffffffff8037f7d3>{scsi_request_fn+1046}
       <ffffffff8031c9b7>{blk_insert_request+156} <ffffffff8037e23f>{scsi_insert_special_req+51}
       <ffffffff8037e513>{scsi_wait_req+339} <ffffffff8037fae7>{__scsi_mode_sense+228}
       <ffffffff803e8f63>{sd_revalidate_disk+3037} <ffffffff80143953>{add_preempt_count_ti+35}
       <ffffffff80143442>{atomic_dec_and_spin_lock+52} <ffffffff801a9fc2>{rescan_partitions+133}
       <ffffffff8017729b>{do_open+672} <ffffffff80176fc2>{blkdev_get+161}

BUG: swapper/1, possible wake_up race on softirq-scsi/1/15

Call Trace:<ffffffff80121ade>{wake_up_p   <ffffffff8031581d>{driver_register+91} <ffffffff80834e5b>{init_sd+31}
       <ffffffff8010b24f>{init+503} <ffffffff8010e93e>{child_rip+8}
       <ffffffff8010b058>{init+0} <ffffffff8010e936>{child_rip+0}