Re: [PATCH] block: Fix bug in runtime-resume handling

[Alan Stern <stern@rowland.harvard.edu>]

On Mon, Aug 24, 2020 at 10:48:32AM -0700, Bart Van Assche wrote:
> On 2020-08-23 07:57, Alan Stern wrote:
> > The problem is fixed by checking that the queue's runtime-PM status
> > isn't RPM_SUSPENDED before allowing a request to be issued, [ ... ]
> 
> Reviewed-by: Bart Van Assche <bvanassche@acm.org>
> 
> Can, can you help with testing this patch?
> 
> Thanks,
> 
> Bart.

Martin:

(I forgot to ask this question several weeks ago, while you were running 
your tests.  Better ask it now before I forget again...)

I suspect the old runtime-PM code in the block layer would have worked 
okay in your SD cardreader test if the BLK_MQ_REQ_PREEMPT flag had not 
been set.  Do you know why the flag was set, or what line of code caused 
it to be set?

As I recall, the request that failed was a read-ahead.  It's not clear 
to me why such a request would need to have BLK_MQ_REQ_PREEMPT set.  
Although there may be other reasons for having that flag, at this point 
we're concerned with requests that are part of the runtime-resume 
procedure.  A read-ahead does not fall into that category.

Do you think you can find the answer?  Perhaps we should add a separate 
BLK_MQ_REQ_PM flag.

Alan Stern