Re: [PATCH v1.2 17/16] arcmsr: fixed calling scsi_scan_host after all initialization done

[ching <ching2048@areca.com.tw>]

Hi Tomas,

Although in http://www.makelinux.net/ldd3/chp-7-sect-6, there said to flush the shared workqueue requires calling function
flush_scheduled_work.

But in kernel workqueue.c, there are comment of flush_scheduled_work said that 
"In most situations flushing the entire workqueue is overkill; you merely need to know that a particular work item isn't
queued and isn't running."

In such cases use flush_work is OK.
This can be proved in so many kernel device driver calling flush_work instead of flush_scheduled_work.


On Thu, 2014-05-15 at 13:52 +0200, Tomas Henzl wrote:
> On 05/15/2014 12:49 PM, ching wrote:
> > Hi Tomas,
> >
> > Then I will have a patch 18/16 to replace flush_scheduled_work with flush_work.
> > Thanks for your advice.
> 
> Here http://www.makelinux.net/ldd3/chp-7-sect-6 is a short text about workqueue,
> the use of schedule_work in your sources means that you use the global queue.
> If that is true, flush_scheduled_work is the right choice.
> 
> >
> > On Wed, 2014-05-14 at 14:53 +0200, Tomas Henzl wrote:
> >> On 05/14/2014 01:55 PM, ching wrote:
> >>> From: Ching<ching2048@areca.com.tw>
> >>>
> >>> This is a new patch after patch v1.1 series.
> >>>
> >>> Fixed calling scsi_scan_host until all initialization are done.
> >>> And fixed error path free allocated resource.
> >>>
> >>> Signed-off-by: Ching<ching2048@areca.com.tw>
> >>> ---
> >>>
> >>> diff -uprN a/drivers/scsi/arcmsr/arcmsr_hba.c b/drivers/scsi/arcmsr/arcmsr_hba.c
> >>> --- a/drivers/scsi/arcmsr/arcmsr_hba.c	2014-05-08 17:45:34.000000000 +0800
> >>> +++ b/drivers/scsi/arcmsr/arcmsr_hba.c	2014-05-14 18:29:30.000000000 +0800
> >>> @@ -112,6 +112,7 @@ static void arcmsr_hbaD_message_isr(stru
> >>>  static void arcmsr_hardware_reset(struct AdapterControlBlock *acb);
> >>>  static const char *arcmsr_info(struct Scsi_Host *);
> >>>  static irqreturn_t arcmsr_interrupt(struct AdapterControlBlock *acb);
> >>> +static void arcmsr_free_irq(struct pci_dev *, struct AdapterControlBlock *);
> >>>  static int arcmsr_adjust_disk_queue_depth(struct scsi_device *sdev,
> >>>  					  int queue_depth, int reason)
> >>>  {
> >>> @@ -778,12 +779,11 @@ static int arcmsr_probe(struct pci_dev *
> >>>  	}
> >>>  	error = scsi_add_host(host, &pdev->dev);
> >>>  	if(error){
> >>> -		goto RAID_controller_stop;
> >>> +		goto free_ccb_pool;
> >>>  	}
> >>>  	if (arcmsr_request_irq(pdev, acb) == ARC_FAILURE)
> >>>  		goto scsi_host_remove;
> >>>  	arcmsr_iop_init(acb);
> >>> -    	scsi_scan_host(host);
> >>>  	INIT_WORK(&acb->arcmsr_do_message_isr_bh, arcmsr_message_isr_bh_fn);
> >>>  	atomic_set(&acb->rq_map_token, 16);
> >>>  	atomic_set(&acb->ante_token_value, 16);
> >>> @@ -795,13 +795,17 @@ static int arcmsr_probe(struct pci_dev *
> >>>  	add_timer(&acb->eternal_timer);
> >>>  	if(arcmsr_alloc_sysfs_attr(acb))
> >>>  		goto out_free_sysfs;
> >>> +    	scsi_scan_host(host);
> >>>  	return 0;
> >>>  out_free_sysfs:
> >>> -scsi_host_remove:
> >>> -	scsi_remove_host(host);
> >>> -RAID_controller_stop:
> >>> +	del_timer_sync(&acb->eternal_timer);
> >>> + 	flush_work(&acb->arcmsr_do_message_isr_bh);
> >> In your driver you use both flush_work(struct work_struct *) and flush_scheduled_work(),
> >> the second form is used used to flush the shared workqueue. 
> >> That said (and I think you use just one queue) you should decide which one is correct.
> >> In my previous mail I have asked about flush_work because i have it seen elsewhere in
> >> your code, now I've noticed the use of schedule_work and that means that you are using the global wq.
> >>  
> >>
> >>>  	arcmsr_stop_adapter_bgrb(acb);
> >>>  	arcmsr_flush_adapter_cache(acb);
> >>> +	arcmsr_free_irq(pdev, acb);
> >>> +scsi_host_remove:
> >>> +	scsi_remove_host(host);
> >>> +free_ccb_pool:
> >>>  	arcmsr_free_ccb_pool(acb);
> >>>  free_hbb_mu:
> >>>  	arcmsr_free_mu(acb);
> >>>
> >>>
> >>> --
> >>> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> >>> the body of a message to majordomo@vger.kernel.org
> >>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >
>