Re: [Regression x2, 3.13-git] virtio block mq hang, iostat bustedon virtio devices

[Jens Axboe]

On 11/20/2013 01:07 AM, Christoph Hellwig wrote:
>> Just stumbled on that too. You need one more, btw, for the sg failure
>> case:
>>
>>
>> diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
>> index 588479d58f52..6a680d4de7f1 100644
>> --- a/drivers/block/virtio_blk.c
>> +++ b/drivers/block/virtio_blk.c
>> @@ -199,15 +199,16 @@ static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req)
>>  
>>  	spin_lock_irqsave(&vblk->vq_lock, flags);
>>  	if (__virtblk_add_req(vblk->vq, vbr, vbr->sg, num) < 0) {
>> +		virtqueue_kick(vblk->vq);
>>  		spin_unlock_irqrestore(&vblk->vq_lock, flags);
>>  		blk_mq_stop_hw_queue(hctx);
>> -		virtqueue_kick(vblk->vq);
>>  		return BLK_MQ_RQ_QUEUE_BUSY;
>>  	}
>> -	spin_unlock_irqrestore(&vblk->vq_lock, flags);
>>  
>>  	if (last)
>>  		virtqueue_kick(vblk->vq);
>> +
>> +	spin_unlock_irqrestore(&vblk->vq_lock, flags);
> 
> Note that virtqueue_kick can be split into a virtqueue_prepare_split
> that needs the lock, and a virtqueue_notify that doesn't, but it might
> not be nessecary to do that optimization right now.

Right, that would be easily doable. Seems it's in the realm of micro
optimizations, but worth doing next time it's touched.

> Btw, how much do I need to bribe you to send mails where I don't have 
> step over 500 lines of dead quotes to find the actual patch?

Sorry, the next one was trimmed :-)

-- 
Jens Axboe

--
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/