From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=91.218.175.145 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1795169361; b=zGiqGGY46OtgV5k6PhVuWNi/HJIZURm+F8cAlK8nzrjvDXOqF/0S6CkrUm4UUYgHui3i Th5PbfWNMi6ylm5TZ4XofpRLo0gXoZNVlETAofORKl5WQig7t/BiDEULDqbMKzF0zTRxe MGVvL88h6J+n6pF8gOyrxLPvfqZBf6vU6XFjYD/FIk2+By38aCZDh0qQIfbHbcb/d/31N Va3mv9UrdpFlDjmea5DgVYeoFAVcojA41XQW/m1yJGVqJHDQEeLZEnFgf94pwbf86c5i1 52ateI7N0EvnSN6YOwUYLJ0Ui1tngOb4I7gEk9HDSiCnEWgEF5ObujjD3y9wPOXlcog== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1795169361; h=DMARC-Filter:X-Envelope-To:DKIM-Signature:X-Envelope-To:Received: X-Mizu-Trace-ID:X-Migadu-Flow:Message-ID:Date:MIME-Version:User-Agent: Subject:To:Cc:References:From:In-Reply-To:Content-Type: Content-Transfer-Encoding; bh=8l1rxUVHALrLazVSWALlznjCgIYdHTWqcToTvFjFBKg=; b=yGPIW1wP0O0mVWq3la99oFmNugVFUVeFAmeLIDelBRd53eyjhPeyU/Xp0RDY4jvYsM/K StMMgXIZiQK6mAUpTsh9/Lbw+tXlJpRUUBAEyIiejeD2dGI/fKn5htBJ8bCl6rejEb0LK Cm4tbri/2GDI3HPqWzTF6afTkffygU7tx/Kr6nkXfkr/sqculjZviNhUusaryFJd41pvN 4E+PML6NqWd521g+U1xlaAgvN5T0hu8sXPnIA8bJFxd9DyUm/pn5brZaU5SItmoe72atu lhTYG1Jw4zUyHYduybnUmKL4Y1R/y6tSBLodOtwOAfLCOFV03a+LGRrDVV3UFG+JYIQ== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass header.d=linux.dev header.i=@linux.dev header.a=rsa-sha256 header.s=key1 header.b="KK4/wbl6"; arc=none smtp.remote-ip=91.218.175.145 Received: from mta0.migadu.com (out-145.mta0.migadu.com [91.218.175.145]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 2E1DF606A334 for ; Fri, 20 Nov 2026 10:09:20 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org 2E1DF606A334 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key, unprotected) header.d=linux.dev header.i=@linux.dev header.a=rsa-sha256 header.s=key1 header.b=KK4/wbl6 X-Envelope-To: netdev@vger.kernel.org DKIM-Signature: a=rsa-sha256; bh=lmupTUCv65fg5pP75anIseOQtkdlOKkZnDml59Hx+vg=; c=simple/simple; d=linux.dev; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1795169359; v=1; x=1795774159; b=KK4/wbl619CRE9joqmaw8LIvKdOKg2jk4pyiVIF4hVznoWuyV3N6otfaoXWUnWgbBhC6H2rj QAV0ijBAhkGuwRWbSwbaVe4Et5hMFScAWdZATXbMESpD6ibOrOzRRufWfOb8Gbmq+aOep6XNsbd QDMcI1kaejZZPt53EQ3B6JTM= X-Envelope-To: netdev@vger.kernel.org Received: by smtp.migadu.com with ESMTPS id 3039bbd6bb4976e6; Fri, 20 Nov 2026 10:09:19 +0000 X-Mizu-Trace-ID: 3039bbd6bb4976e6 X-Migadu-Flow: FLOW_OUT Message-ID: <4198a993-1c7c-4ca5-8d31-8f8e36778702@linux.dev> Date: Fri, 20 Nov 2026 18:09:09 +0800 X-Mailing-List: netdev@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH net v2] tcp: fix negative sk_forward_alloc after MSG_ZEROCOPY append To: netdev@vger.kernel.org Cc: Eric Dumazet , Neal Cardwell , Kuniyuki Iwashima , "David S. Miller" , Jakub Kicinski , Paolo Abeni , Simon Horman , Soheil Hassas Yeganeh , Shakeel Butt , Wei Wang , linux-kernel@vger.kernel.org References: <20261120094605.46063-1-jiayuan.chen@linux.dev> From: Jiayuan Chen In-Reply-To: <20261120094605.46063-1-jiayuan.chen@linux.dev> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit On 11/20/26 5:46 PM, Jiayuan Chen wrote: > Our local syzbot reported the following warning: > > WARNING: ./include/linux/ucopysize.h:57 at tcp_sendmsg_locked+0x1e13/0x1f20 > Modules linked in: > CPU: 3 UID: 0 PID: 377 Comm: syz.2.2248 Not tainted 7.4.0-rc2+ #398 PREEMPT > RIP: 0010:tcp_sendmsg_locked+0x1e13/0x1f20 > Call Trace: > > tcp_sendmsg+0x2c/0x50 > inet_sendmsg+0x62/0xb0 > __sys_sendto+0x2fe/0x350 > __x64_sys_sendto+0x7f/0xb0 > x64_sys_call+0x1375/0x27e0 > do_syscall_64+0xe4/0x470 > entry_SYSCALL_64_after_hwframe+0x76/0x7e > > > The problem is here: > > if (!skb_zcopy_pure(skb)) { > copy = tcp_wmem_schedule(sk, copy); > if (!copy) > goto wait_for_space; > } > > When MSG_ZEROCOPY data is appended to a skb that already holds copied > data, only the user length is scheduled, but the pinned pages are then > charged by their truesize: > > skb_zerocopy_iter_stream > __zerocopy_sg_from_iter > zerocopy_fill_skb_from_iter //truesize += PAGE_ALIGN(copied + start) > sk_mem_charge(sk, truesize) //sk_forward_alloc goes negative > > For example, sending 8 iovecs of 1 byte each schedules 8 bytes and > charges 32768 with 4K pages. > > On a later sendmsg() without MSG_ZEROCOPY, if sk_wmem_schedule() fails > (e.g. tcp_mem hard limit is reached), tcp_wmem_schedule() returns > min(copy, sk->sk_forward_alloc), which is negative. The negative copy > is passed to skb_copy_to_page_nocache() and triggers the warning, and > sendmsg() fails with -EFAULT. > > A few options were considered: > > 1) Schedule the truesize up front. It is not known before the pages > are pinned, it depends on the iovec layout. > 2) Do not append zerocopy data to a non pure skb. This splits a small > header and its zerocopy payload into two skbs. > 3) Schedule the missing part right after the pages are charged. They > are already in the skb at this point, so this can not fail. > > Go with 3). > > Fixes: 849b425cd091 ("tcp: fix possible freeze in tx path under memory pressure") > Signed-off-by: Jiayuan Chen > --- > v1: https://lore.kernel.org/netdev/20261119090828.343748-1-jiayuan.chen@linux.dev/ > > Changes in v2: > - Only call sk_forced_mem_schedule() for non pure zerocopy skbs (Eric) > --- > net/ipv4/tcp.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c > index d510e982dcea..343f6d61b408 100644 > --- a/net/ipv4/tcp.c > +++ b/net/ipv4/tcp.c > @@ -1359,6 +1359,8 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) > if (err < 0) > goto do_error; > copy = err; > + if (!skb_zcopy_pure(skb)) > + sk_forced_mem_schedule(sk, 0); I think I should move such code before the 'if (err < 0)', like the sashiko said: https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20261119090828.343748-1-jiayuan.chen%40linux.dev > } else if (zc == MSG_SPLICE_PAGES) { > /* Splice in data if we can; copy if we can't. */ > if (tcp_downgrade_zcopy_pure(sk, skb))