On Tue, Jul 30, 2019 at 11:35:39AM +0200, Stefano Garzarella wrote: > On Mon, Jul 29, 2019 at 03:10:15PM -0400, Michael S. Tsirkin wrote: > > On Mon, Jul 29, 2019 at 06:50:56PM +0200, Stefano Garzarella wrote: > > > On Mon, Jul 29, 2019 at 06:19:03PM +0200, Stefano Garzarella wrote: > > > > On Mon, Jul 29, 2019 at 11:49:02AM -0400, Michael S. Tsirkin wrote: > > > > > On Mon, Jul 29, 2019 at 05:36:56PM +0200, Stefano Garzarella wrote: > > > > > > On Mon, Jul 29, 2019 at 10:04:29AM -0400, Michael S. Tsirkin wrote: > > > > > > > On Wed, Jul 17, 2019 at 01:30:26PM +0200, Stefano Garzarella wrote: > > > > > > > > Since virtio-vsock was introduced, the buffers filled by the host > > > > > > > > and pushed to the guest using the vring, are directly queued in > > > > > > > > a per-socket list. These buffers are preallocated by the guest > > > > > > > > with a fixed size (4 KB). > > > > > > > > > > > > > > > > The maximum amount of memory used by each socket should be > > > > > > > > controlled by the credit mechanism. > > > > > > > > The default credit available per-socket is 256 KB, but if we use > > > > > > > > only 1 byte per packet, the guest can queue up to 262144 of 4 KB > > > > > > > > buffers, using up to 1 GB of memory per-socket. In addition, the > > > > > > > > guest will continue to fill the vring with new 4 KB free buffers > > > > > > > > to avoid starvation of other sockets. > > > > > > > > > > > > > > > > This patch mitigates this issue copying the payload of small > > > > > > > > packets (< 128 bytes) into the buffer of last packet queued, in > > > > > > > > order to avoid wasting memory. > > > > > > > > > > > > > > > > Reviewed-by: Stefan Hajnoczi <stefanha@xxxxxxxxxx> > > > > > > > > Signed-off-by: Stefano Garzarella <sgarzare@xxxxxxxxxx> > > > > > > > > > > > > > > This is good enough for net-next, but for net I think we > > > > > > > should figure out how to address the issue completely. > > > > > > > Can we make the accounting precise? What happens to > > > > > > > performance if we do? > > > > > > > > > > > > > > > > > > > In order to do more precise accounting maybe we can use the buffer size, > > > > > > instead of payload size when we update the credit available. > > > > > > In this way, the credit available for each socket will reflect the memory > > > > > > actually used. > > > > > > > > > > > > I should check better, because I'm not sure what happen if the peer sees > > > > > > 1KB of space available, then it sends 1KB of payload (using a 4KB > > > > > > buffer). > > > > > > > > > > > > The other option is to copy each packet in a new buffer like I did in > > > > > > the v2 [2], but this forces us to make a copy for each packet that does > > > > > > not fill the entire buffer, perhaps too expensive. > > > > > > > > > > > > [2] https://patchwork.kernel.org/patch/10938741/ > > > > > > > > > > > > > > > > > > Thanks, > > > > > > Stefano > > > > > > > > > > Interesting. You are right, and at some level the protocol forces copies. > > > > > > > > > > We could try to detect that the actual memory is getting close to > > > > > admin limits and force copies on queued packets after the fact. > > > > > Is that practical? > > > > > > > > Yes, I think it is doable! > > > > We can decrease the credit available with the buffer size queued, and > > > > when the buffer size of packet to queue is bigger than the credit > > > > available, we can copy it. > > > > > > > > > > > > > > And yes we can extend the credit accounting to include buffer size. > > > > > That's a protocol change but maybe it makes sense. > > > > > > > > Since we send to the other peer the credit available, maybe this > > > > change can be backwards compatible (I'll check better this). > > > > > > What I said was wrong. > > > > > > We send a counter (increased when the user consumes the packets) and the > > > "buf_alloc" (the max memory allowed) to the other peer. > > > It makes a difference between a local counter (increased when the > > > packets are sent) and the remote counter to calculate the credit available: > > > > > > u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit) > > > { > > > u32 ret; > > > > > > spin_lock_bh(&vvs->tx_lock); > > > ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt); > > > if (ret > credit) > > > ret = credit; > > > vvs->tx_cnt += ret; > > > spin_unlock_bh(&vvs->tx_lock); > > > > > > return ret; > > > } > > > > > > Maybe I can play with "buf_alloc" to take care of bytes queued but not > > > used. > > > > > > Thanks, > > > Stefano > > > > Right. And the idea behind it all was that if we send a credit > > to remote then we have space for it. > > Yes. > > > I think the basic idea was that if we have actual allocated > > memory and can copy data there, then we send the credit to > > remote. > > > > Of course that means an extra copy every packet. > > So as an optimization, it seems that we just assume > > that we will be able to allocate a new buffer. > > Yes, we refill the virtqueue when half of the buffers were used. > > > > > First this is not the best we can do. We can actually do > > allocate memory in the socket before sending credit. > > In this case, IIUC we should allocate an entire buffer (4KB), > so we can reuse it if the packet is big. > > > If packet is small then we copy it there. > > If packet is big then we queue the packet, > > take the buffer out of socket and add it to the virtqueue. > > > > Second question is what to do about medium sized packets. > > Packet is 1K but buffer is 4K, what do we do? > > And here I wonder - why don't we add the 3K buffer > > to the vq? > > This would allow us to have an accurate credit account. > > The problem here is the compatibility. Before this series virtio-vsock > and vhost-vsock modules had the RX buffer size hard-coded > (VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE = 4K). So, if we send a buffer smaller > of 4K, there might be issues. Shouldn't be if they are following the spec. If not let's fix the broken parts. > > Maybe it is the time to add add 'features' to virtio-vsock device. > > Thanks, > Stefano Why would a remote care about buffer sizes? Let's first see what the issues are. If they exist we can either fix the bugs, or code the bug as a feature in spec. -- MST _______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization
