[PATCH net] tcp: call sk_wmem_schedule before sk_mem_charge in zerocopy path

Subsystems: networking [general], networking [tcp], the rest

STALE1873d REVIEWED: 3 (1M)

3 review trailers.

2 messages, 2 authors, 2021-07-09 · open the first message on its own page

[PATCH net] tcp: call sk_wmem_schedule before sk_mem_charge in zerocopy path

From: Talal Ahmad <hidden>
Date: 2021-07-09 15:43:53

From: Talal Ahmad <redacted>

sk_wmem_schedule makes sure that sk_forward_alloc has enough
bytes for charging that is going to be done by sk_mem_charge.

In the transmit zerocopy path, there is sk_mem_charge but there was
no call to sk_wmem_schedule. This change adds that call.

Without this call to sk_wmem_schedule, sk_forward_alloc can go
negetive which is a bug because sk_forward_alloc is a per-socket
space that has been forward charged so this can't be negative.

Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY")
Signed-off-by: Talal Ahmad <redacted>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Wei Wang <redacted>
Reviewed-by: Soheil Hassas Yeganeh <redacted>
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/ipv4/tcp.c | 3 +++
 1 file changed, 3 insertions(+)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index d5ab5f243640..8cb44040ec68 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1375,6 +1375,9 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
 			}
 			pfrag->offset += copy;
 		} else {
+			if (!sk_wmem_schedule(sk, copy))
+				goto wait_for_space;
+
 			err = skb_zerocopy_iter_stream(sk, skb, msg, copy, uarg);
 			if (err == -EMSGSIZE || err == -EEXIST) {
 				tcp_mark_push(tp, skb);
-- 
2.32.0.93.g670b81a890-goog

Re: [PATCH net] tcp: call sk_wmem_schedule before sk_mem_charge in zerocopy path

From: patchwork-bot+netdevbpf@kernel.org
Date: 2021-07-09 18:30:09

Hello:

This patch was applied to netdev/net.git (refs/heads/master):

On Fri,  9 Jul 2021 11:43:06 -0400 you wrote:
From: Talal Ahmad <redacted>

sk_wmem_schedule makes sure that sk_forward_alloc has enough
bytes for charging that is going to be done by sk_mem_charge.

In the transmit zerocopy path, there is sk_mem_charge but there was
no call to sk_wmem_schedule. This change adds that call.

[...]
Here is the summary with links:
  - [net] tcp: call sk_wmem_schedule before sk_mem_charge in zerocopy path
    https://git.kernel.org/netdev/net/c/358ed6242070

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help