| Message ID | 20220216050320.3222-1-kerneljasonxing@gmail.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [v2,net-next] net: introduce SO_RCVBUFAUTO to let the rcv_buf tune automatically | expand |
On Tue, Feb 15, 2022 at 9:03 PM <kerneljasonxing@gmail.com> wrote: > > From: Jason Xing <xingwanli@kuaishou.com> > > Normally, user doesn't care the logic behind the kernel if they're > trying to set receive buffer via setsockopt. However, once the new > value of the receive buffer is set even though it's not smaller than > the initial value which is sysctl_tcp_rmem[1] implemented in > tcp_rcv_space_adjust(),, the server's wscale will shrink and then > lead to the bad bandwidth as intended. Quite confusing changelog, honestly. Users of SO_RCVBUF specifically told the kernel : I want to use _this_ buffer size, I do not want the kernel to decide for me. Also, I think your changelog does not really explain that _if_ you set SO_RCVBUF to a small value before connect() or in general the 3WHS, the chosen wscale will be small, and this won't allow future 10x increase of the effective RWIN. > > For now, introducing a new socket option to let the receive buffer > grow automatically no matter what the new value is can solve > the bad bandwidth issue meanwhile it's not breaking the application > with SO_RCVBUF option set. > > Here are some numbers: > $ sysctl -a | grep rmem > net.core.rmem_default = 212992 > net.core.rmem_max = 40880000 > net.ipv4.tcp_rmem = 4096 425984 40880000 > > Case 1 > on the server side > # iperf -s -p 5201 > on the client side > # iperf -c [client ip] -p 5201 > It turns out that the bandwidth is 9.34 Gbits/sec while the wscale of > server side is 10. It's good. > > Case 2 > on the server side > #iperf -s -p 5201 -w 425984 > on the client side > # iperf -c [client ip] -p 5201 > It turns out that the bandwidth is reduced to 2.73 Gbits/sec while the > wcale is 2, even though the receive buffer is not changed at all at the > very beginning. > > After this patch is applied, the bandwidth of case 2 is recovered to > 9.34 Gbits/sec as expected at the cost of consuming more memory per > socket. How does your patch allow wscale to increase after flow is established ? I would remove from the changelog these experimental numbers that look quite wrong, maybe copy/pasted from your prior version. Instead I would describe why an application might want to clear the 'receive buffer size is locked' socket attribute. > > Signed-off-by: Jason Xing <xingwanli@kuaishou.com> > -- > v2: suggested by Eric > - introduce new socket option instead of breaking the logic in SO_RCVBUF > - Adjust the title and description of this patch > link: https://lore.kernel.org/lkml/CANn89iL8vOUOH9bZaiA-cKcms+PotuKCxv7LpVx3RF0dDDSnmg@mail.gmail.com/ > --- > I think adding another parallel SO_RCVBUF option is not good. It is adding confusion (and net/core/filter.c has been unchanged) Also we want CRIU to work correctly. So if you have a SO_XXXX setsockopt() call, you also need to provide getsockopt() implementation. I would suggest an option to clear or set SOCK_RCVBUF_LOCK, and getsockopt() would return if the bit is currently set or not. Something clearly describing the intent, like SO_RCVBUF_LOCK maybe.
On Wed, Feb 16, 2022 at 2:25 PM Eric Dumazet <edumazet@google.com> wrote: > > On Tue, Feb 15, 2022 at 9:03 PM <kerneljasonxing@gmail.com> wrote: > > > > From: Jason Xing <xingwanli@kuaishou.com> > > > > Normally, user doesn't care the logic behind the kernel if they're > > trying to set receive buffer via setsockopt. However, once the new > > value of the receive buffer is set even though it's not smaller than > > the initial value which is sysctl_tcp_rmem[1] implemented in > > tcp_rcv_space_adjust(),, the server's wscale will shrink and then > > lead to the bad bandwidth as intended. > > Quite confusing changelog, honestly. > > Users of SO_RCVBUF specifically told the kernel : I want to use _this_ > buffer size, I do not want the kernel to decide for me. > > Also, I think your changelog does not really explain that _if_ you set > SO_RCVBUF to a small value before > connect() or in general the 3WHS, the chosen wscale will be small, and > this won't allow future 10x increase > of the effective RWIN. > Yes, you hit the point really. > > > > > For now, introducing a new socket option to let the receive buffer > > grow automatically no matter what the new value is can solve > > the bad bandwidth issue meanwhile it's not breaking the application > > with SO_RCVBUF option set. > > > > Here are some numbers: > > $ sysctl -a | grep rmem > > net.core.rmem_default = 212992 > > net.core.rmem_max = 40880000 > > net.ipv4.tcp_rmem = 4096 425984 40880000 > > > > Case 1 > > on the server side > > # iperf -s -p 5201 > > on the client side > > # iperf -c [client ip] -p 5201 > > It turns out that the bandwidth is 9.34 Gbits/sec while the wscale of > > server side is 10. It's good. > > > > Case 2 > > on the server side > > #iperf -s -p 5201 -w 425984 > > on the client side > > # iperf -c [client ip] -p 5201 > > It turns out that the bandwidth is reduced to 2.73 Gbits/sec while the > > wcale is 2, even though the receive buffer is not changed at all at the > > very beginning. > > > > After this patch is applied, the bandwidth of case 2 is recovered to > > 9.34 Gbits/sec as expected at the cost of consuming more memory per > > socket. > > How does your patch allow wscale to increase after flow is established ? > > I would remove from the changelog these experimental numbers that look > quite wrong, > maybe copy/pasted from your prior version. > My fault. I should have removed this part. > Instead I would describe why an application might want to clear the > 'receive buffer size is locked' socket attribute. > > > > > Signed-off-by: Jason Xing <xingwanli@kuaishou.com> > > -- > > v2: suggested by Eric > > - introduce new socket option instead of breaking the logic in SO_RCVBUF > > - Adjust the title and description of this patch > > link: https://lore.kernel.org/lkml/CANn89iL8vOUOH9bZaiA-cKcms+PotuKCxv7LpVx3RF0dDDSnmg@mail.gmail.com/ > > --- > > > > I think adding another parallel SO_RCVBUF option is not good. It is > adding confusion (and net/core/filter.c has been unchanged) I'll change the filter.c altogether in the next submission. > > Also we want CRIU to work correctly. > > So if you have a SO_XXXX setsockopt() call, you also need to provide > getsockopt() implementation. > > I would suggest an option to clear or set SOCK_RCVBUF_LOCK, and > getsockopt() would return if the bit is currently set or not. > > Something clearly describing the intent, like SO_RCVBUF_LOCK maybe. Just now, I found out that the latest kernel has merged a similar patch (commit 04190bf89) about three months ago. Is it still necessary to add another separate option to clear the SOCK_RCVBUF_LOCK explicitly? Thanks, Jason
On Tue, Feb 15, 2022 at 10:58 PM Jason Xing <kerneljasonxing@gmail.com> wrote: > Just now, I found out that the latest kernel has merged a similar > patch (commit 04190bf89) about three months ago. There you go :) > > Is it still necessary to add another separate option to clear the > SOCK_RCVBUF_LOCK explicitly? What do you mean, SO_BUF_LOCK is all that is needed.
On Thu, Feb 17, 2022 at 12:56 AM Eric Dumazet <edumazet@google.com> wrote: > > On Tue, Feb 15, 2022 at 10:58 PM Jason Xing <kerneljasonxing@gmail.com> wrote: > > Just now, I found out that the latest kernel has merged a similar > > patch (commit 04190bf89) about three months ago. > > There you go :) > > > > > Is it still necessary to add another separate option to clear the > > SOCK_RCVBUF_LOCK explicitly? > > What do you mean, SO_BUF_LOCK is all that is needed. Yeah, I think SO_BUF_LOCK is enough and we don't have to add a new option like SOCK_RCVBUF_LOCK as we've talked about before. Thanks, Eric.
diff --git a/include/uapi/asm-generic/socket.h b/include/uapi/asm-generic/socket.h index c77a131..f4ce79b 100644 --- a/include/uapi/asm-generic/socket.h +++ b/include/uapi/asm-generic/socket.h @@ -18,6 +18,7 @@ #define SO_RCVBUF 8 #define SO_SNDBUFFORCE 32 #define SO_RCVBUFFORCE 33 +#define SO_RCVBUFAUTO 74 #define SO_KEEPALIVE 9 #define SO_OOBINLINE 10 #define SO_NO_CHECK 11 diff --git a/net/core/sock.c b/net/core/sock.c index 4ff806d..8565684 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -917,13 +917,14 @@ void sock_set_keepalive(struct sock *sk) } EXPORT_SYMBOL(sock_set_keepalive); -static void __sock_set_rcvbuf(struct sock *sk, int val) +static void __sock_set_rcvbuf(struct sock *sk, int val, bool is_set) { /* Ensure val * 2 fits into an int, to prevent max_t() from treating it * as a negative value. */ val = min_t(int, val, INT_MAX / 2); - sk->sk_userlocks |= SOCK_RCVBUF_LOCK; + if (is_set) + sk->sk_userlocks |= SOCK_RCVBUF_LOCK; /* We double it on the way in to account for "struct sk_buff" etc. * overhead. Applications assume that the SO_RCVBUF setting they make @@ -941,7 +942,7 @@ static void __sock_set_rcvbuf(struct sock *sk, int val) void sock_set_rcvbuf(struct sock *sk, int val) { lock_sock(sk); - __sock_set_rcvbuf(sk, val); + __sock_set_rcvbuf(sk, val, true); release_sock(sk); } EXPORT_SYMBOL(sock_set_rcvbuf); @@ -1106,7 +1107,7 @@ int sock_setsockopt(struct socket *sock, int level, int optname, * play 'guess the biggest size' games. RCVBUF/SNDBUF * are treated in BSD as hints */ - __sock_set_rcvbuf(sk, min_t(u32, val, sysctl_rmem_max)); + __sock_set_rcvbuf(sk, min_t(u32, val, sysctl_rmem_max), true); break; case SO_RCVBUFFORCE: @@ -1118,7 +1119,14 @@ int sock_setsockopt(struct socket *sock, int level, int optname, /* No negative values (to prevent underflow, as val will be * multiplied by 2). */ - __sock_set_rcvbuf(sk, max(val, 0)); + __sock_set_rcvbuf(sk, max(val, 0), true); + break; + + case SO_RCVBUFAUTO: + /* Though similar to SO_RCVBUF, we do not use userlocks in + * order to let the receive buffer tune automatically. + */ + __sock_set_rcvbuf(sk, min_t(u32, val, sysctl_rmem_max), false); break; case SO_KEEPALIVE: