Re: [PATCH] block: mq flush: fix race between IPI handler and mq flush worker

[Ming Lei]

On Mon, May 19, 2014 at 11:18 PM, Christoph Hellwig <hch@xxxxxx> wrote:
> On Mon, May 19, 2014 at 11:05:50PM +0800, Ming Lei wrote:
>> Another simple fix is to disable ipi for flush request, but looks
>> this one should be better.
>
> I think the first thing is to bite the bullet and sort out and document
> the various unions in struct request for real.

I agree, unions should be documented in detail.

>
> For example the first union has the call_single_data for the blk-mq

blk-softirq need rq->csd too in raise_blk_irq().

> completions, while the second one has the ipi_list that is used by
> the old blk-softirq code.

Also we can put some mq specific fields and legacy fields into
one union too.

>
> If we get this right with a single union that contains a struct for
> each phase of the request we might find enough space to keep using
> the current way.

If we can figure it out, that should be better solution, but changing/
merging fields may affect performance too, and need careful
verification.

Thanks,
-- 
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/