Re: [PATCH] virtio-net: parameterize min ring num_free for virtio receive

On Fri, Jul 19, 2019 at 03:31:29PM +0000, 冉 jiang wrote:

On 2019/7/19 22:29, Jiang wrote:

quoted

On 2019/7/19 10:36, Jason Wang wrote:

quoted

On 2019/7/18 下午10:43, Michael S. Tsirkin wrote:

quoted

On Thu, Jul 18, 2019 at 10:42:47AM -0400, Michael S. Tsirkin wrote:

quoted

On Thu, Jul 18, 2019 at 10:01:05PM +0800, Jason Wang wrote:

quoted

On 2019/7/18 下午9:04, Michael S. Tsirkin wrote:

quoted

On Thu, Jul 18, 2019 at 12:55:50PM +0000, ? jiang wrote:

quoted

This change makes ring buffer reclaim threshold num_free 
configurable
for better performance, while it's hard coded as 1/2 * queue now.
According to our test with qemu + dpdk, packet dropping happens 
when
the guest is not able to provide free buffer in avail ring timely.
Smaller value of num_free does decrease the number of packet 
dropping
during our test as it makes virtio_net reclaim buffer earlier.

At least, we should leave the value changeable to user while the
default value as 1/2 * queue is kept.

Signed-off-by: jiangkidd<redacted>

That would be one reason, but I suspect it's not the
true one. If you need more buffer due to jitter
then just increase the queue size. Would be cleaner.


However are you sure this is the reason for
packet drops? Do you see them dropped by dpdk
due to lack of space in the ring? As opposed to
by guest?

Besides those, this patch depends on the user to choose a suitable 
threshold
which is not good. You need either a good value with demonstrated 
numbers or
something smarter.

Thanks

I do however think that we have a problem right now: try_fill_recv can
take up a long time during which net stack does not run at all. 
Imagine
a 1K queue - we are talking 512 packets. That's exceessive.

Yes, we will starve a fast host in this case.

quoted

quoted

   napi poll
weight solves a similar problem, so it might make sense to cap this at
napi_poll_weight.

Which will allow tweaking it through a module parameter as a
side effect :) Maybe just do NAPI_POLL_WEIGHT.

Or maybe NAPI_POLL_WEIGHT/2 like we do at half the queue ;). Please
experiment, measure performance and let the list know

quoted

Need to be careful though: queues can also be small and I don't 
think we
want to exceed queue size / 2, or maybe queue size - napi_poll_weight.
Definitely must not exceed the full queue size.

Looking at intel, it uses 16 and i40e uses 32.  It looks to me 
NAPI_POLL_WEIGHT/2 is better.

Jiang, want to try that and post a new patch?

Thanks

quoted

quoted

-- 
MST

We did have completed several rounds of test with setting the value to 
budget (64 as the default value). It does improve a lot with pps is 
below 400pps for a single stream. Let me consolidate the data and will 
send it soon. Actually, we are confident that it runs out of free 
buffer in avail ring when packet dropping happens with below systemtap:

Just a snippet:

probe module("virtio_ring").function("virtqueue_get_buf")
{
    x = (@cast($_vq, "vring_virtqueue")->vring->used->idx)- 
(@cast($_vq, "vring_virtqueue")->last_used_idx) ---> we use this one 
to verify if the queue is full, which means guest is not able to take 
buffer from the queue timely

    if (x<0 && (x+65535)<4096)
        x = x+65535

    if((x==1024) && @cast($_vq, "vring_virtqueue")->vq->callback == 
callback_addr)
        netrxcount[x] <<< gettimeofday_s()
}


probe module("virtio_ring").function("virtqueue_add_inbuf")
{
    y = (@cast($vq, "vring_virtqueue")->vring->avail->idx)- 
(@cast($vq, "vring_virtqueue")->vring->used->idx) ---> we use this one 
to verify if we run out of free buffer in avail ring
    if (y<0 && (y+65535)<4096)
        y = y+65535

    if(@2=="debugon")
    {
        if(y==0 && @cast($vq, "vring_virtqueue")->vq->callback == 
callback_addr)
        {
            netrxfreecount[y] <<< gettimeofday_s()

            printf("no avail ring left seen, printing most recent 5 
num free, vq: %lx, current index: %d\n", $vq, recentfreecount)
            for(i=recentfreecount; i!=((recentfreecount+4) % 5); 
i=((i+1) % 5))
            {
                printf("index: %d, num free: %d\n", i, recentfree[$vq, 
i])
            }

            printf("index: %d, num free: %d\n", i, recentfree[$vq, i])
            //exit()
        }
    }
}


probe 
module("virtio_net").statement("virtnet_receive@drivers/net/virtio_net.c:732")
{
    recentfreecount++
    recentfreecount = recentfreecount % 5
    recentfree[$rq->vq, recentfreecount] = $rq->vq->num_free ---> 
record the num_free for the last 5 calls to virtnet_receive, so we can 
see if lowering the bar helps.
}


Here is the result:

no avail ring left seen, printing most recent 5 num free, vq: 
ffff9c13c1200000, current index: 1
index: 1, num free: 561
index: 2, num free: 305
index: 3, num free: 369
index: 4, num free: 433
index: 0, num free: 497
no avail ring left seen, printing most recent 5 num free, vq: 
ffff9c13c1200000, current index: 1
index: 1, num free: 543
index: 2, num free: 463
index: 3, num free: 469
index: 4, num free: 476
index: 0, num free: 479
no avail ring left seen, printing most recent 5 num free, vq: 
ffff9c13c1200000, current index: 2
index: 2, num free: 555
index: 3, num free: 414
index: 4, num free: 420
index: 0, num free: 427
index: 1, num free: 491

You can see in the last 4 calls to virtnet_receive before we run out 
of free buffer and start to relaim, num_free is quite high. So if we 
can do the reclaim earlier, it will certainly help.

Meanwhile, the patch I proposed actually keeps the default value as 
1/2 * queue. So the default behavior remains and only leave the 
interface to advanced users, who really understands what they are 
doing. Also, the best value may vary in different environment. Do you 
still think hardcoding this is better option?


Jiang

Here is the snippet from our test result. Test1 was done with default 
driver with the value of 1/2 * queue, while test2 is with my patch and 
min_numfree set to 64 (the default budget value). We can see average 
drop packets do decrease a lot in test2. Let me know if you need the 
full testing data.

test1Time    avgDropPackets    test2Time    avgDropPackets    pps

quoted

16:21.0    12.295    56:50.4    0    300k
17:19.1    15.244    56:50.4    0    300k
18:17.5    18.789    56:50.4    0    300k
19:15.1    14.208    56:50.4    0    300k
20:13.2    20.818    56:50.4    0.267    300k
21:11.2    12.397    56:50.4    0    300k
22:09.3    12.599    56:50.4    0    300k
23:07.3    15.531    57:48.4    0    300k
24:05.5    13.664    58:46.5    0    300k
25:03.7    13.158    59:44.5    4.73    300k
26:01.1    2.486    00:42.6    0    300k
26:59.1    11.241    01:40.6    0    300k
27:57.2    20.521    02:38.6    0    300k
28:55.2    30.094    03:36.7    0    300k
29:53.3    16.828    04:34.7    0.963    300k
30:51.3    46.916    05:32.8    0    400k
31:49.3    56.214    05:32.8    0    400k
32:47.3    58.69    05:32.8    0    400k
33:45.3    61.486    05:32.8    0    400k
34:43.3    72.175    05:32.8    0.598    400k
35:41.3    56.699    05:32.8    0    400k
36:39.3    61.071    05:32.8    0    400k
37:37.3    43.355    06:30.8    0    400k
38:35.4    44.644    06:30.8    0    400k
39:33.4    72.336    06:30.8    0    400k
40:31.4    70.676    06:30.8    0    400k
41:29.4    108.009    06:30.8    0    400k
42:27.4    65.216    06:30.8    0    400k

Jiang

OK I find this surprising but I accept what you see.
I'm inclined not to add a tunable and just select
a value ourselves.
I'm also fine with using the napi poll module parameter
which will give you a bit of tunability.

-- 
MST