Reproduce environment:
network with 3 VM linuxs is connected as below:
VM1<---->VM2(latest kernel 6.5.0-rc7)<---->VM3
VM1: eth0 ip: 192.168.122.207 MTU 1500
VM2: eth0 ip: 192.168.122.208, eth1 ip: 192.168.123.224 MTU 1500
VM3: eth0 ip: 192.168.123.240 MTU 1500
Reproduce:
VM1 send 1400 bytes UDP data to VM3 using tools scapy with flags=0.
scapy command:
send(IP(dst="192.168.123.240",flags=0)/UDP()/str('0'*1400),count=1,
inter=1.000000)
Result:
Before IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails
FragOKs FragFails FragCreates
Ip: 1 64 11 0 3 4 0 0 4 7 0 0 0 0 0 0 0 0 0
......
----------------------------------------------------------------------
After IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails
FragOKs FragFails FragCreates
Ip: 1 64 12 0 3 5 0 0 4 8 0 0 0 0 0 0 0 0 0
......
----------------------------------------------------------------------
"ForwDatagrams" increase from 4 to 5 and "OutRequests" also increase
from 7 to 8.
Issue description and patch:
IPSTATS_MIB_OUTPKTS("OutRequests") is counted with IPSTATS_MIB_OUTOCTETS
("OutOctets") in ip_finish_output2().
According to RFC 4293, it is "OutOctets" counted with "OutTransmits" but
not "OutRequests". "OutRequests" does not include any datagrams counted
in "ForwDatagrams".
ipSystemStatsOutOctets OBJECT-TYPE
DESCRIPTION
"The total number of octets in IP datagrams delivered to the
lower layers for transmission. Octets from datagrams
counted in ipIfStatsOutTransmits MUST be counted here.
ipSystemStatsOutRequests OBJECT-TYPE
DESCRIPTION
"The total number of IP datagrams that local IP user-
protocols (including ICMP) supplied to IP in requests for
transmission. Note that this counter does not include any
datagrams counted in ipSystemStatsOutForwDatagrams.
So do patch to define IPSTATS_MIB_OUTPKTS to "OutTransmits" and add
IPSTATS_MIB_OUTREQUESTS for "OutRequests".
Add IPSTATS_MIB_OUTREQUESTS counter in __ip_local_out() for ipv4 and add
IPSTATS_MIB_OUT counter in ip6_finish_output2() for ipv6.
Test result with patch:
Before IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails
FragOKs FragFails FragCreates OutTransmits
Ip: 1 64 9 0 5 1 0 0 3 3 0 0 0 0 0 0 0 0 0 4
......
root@qemux86-64:~# cat /proc/net/netstat
......
IpExt: InNoRoutes InTruncatedPkts InMcastPkts OutMcastPkts InBcastPkts
OutBcastPkts InOctets OutOctets InMcastOctets OutMcastOctets
InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts
InECT0Pkts InCEPkts ReasmOverlaps
IpExt: 0 0 0 0 0 0 2976 1896 0 0 0 0 0 9 0 0 0 0
----------------------------------------------------------------------
After IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails
FragOKs FragFails FragCreates OutTransmits
Ip: 1 64 10 0 5 2 0 0 3 3 0 0 0 0 0 0 0 0 0 5
......
root@qemux86-64:~# cat /proc/net/netstat
......
IpExt: InNoRoutes InTruncatedPkts InMcastPkts OutMcastPkts InBcastPkts
OutBcastPkts InOctets OutOctets InMcastOctets OutMcastOctets
InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts
InECT0Pkts InCEPkts ReasmOverlaps
IpExt: 0 0 0 0 0 0 4404 3324 0 0 0 0 0 10 0 0 0 0
----------------------------------------------------------------------
"ForwDatagrams" increase from 1 to 2 and "OutRequests" is keeping 3.
"OutTransmits" increase from 4 to 5 and "OutOctets" increase 1428.
Signed-off-by: Heng Guo <heng.guo@windriver.com>
Reviewed-by: Kun Song <Kun.Song@windriver.com>
Reviewed-by: Filip Pudak <filip.pudak@windriver.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some reads of inet->tos are racy.
Add needed READ_ONCE() annotations and convert IP_TOS option lockless.
v2: missing changes in include/net/route.h (David Ahern)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet->pmtudisc can be read locklessly.
Implement proper lockless reads and writes to inet->pmtudisc
ip_sock_set_mtu_discover() can now be called from arbitrary
contexts.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet->mc_ttl can be read locklessly.
Implement proper lockless reads and writes to inet->mc_ttl
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a followup of 8bf43be799 ("net: annotate data-races
around sk->sk_priority").
sk->sk_priority can be read and written without holding the socket lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk->sk_tsflags can be read locklessly, add corresponding annotations.
Fixes: b9f40e21ef ("net-timestamp: move timestamp flags out of sk_flags")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit edf391ff17 ("snmp: add missing counters for RFC 4293") had
already added OutOctets for RFC 4293. In commit 2d8dbb04c6 ("snmp: fix
OutOctets counter to include forwarded datagrams"), OutOctets was
counted again, but not removed from ip_output().
According to RFC 4293 "3.2.3. IP Statistics Tables",
ipipIfStatsOutTransmits is not equal to ipIfStatsOutForwDatagrams. So
"IPSTATS_MIB_OUTOCTETS must be incremented when incrementing" is not
accurate. And IPSTATS_MIB_OUTOCTETS should be counted after fragment.
This patch reverts commit 2d8dbb04c6 ("snmp: fix OutOctets counter to
include forwarded datagrams") and move IPSTATS_MIB_OUTOCTETS to
ip_finish_output2 for ipv4.
Reviewed-by: Filip Pudak <filip.pudak@windriver.com>
Signed-off-by: Heng Guo <heng.guo@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
LWTUNNEL_XMIT_CONTINUE is implicitly assumed in ip(6)_finish_output2,
such that any positive return value from a xmit hook could cause
unexpected continue behavior, despite that related skb may have been
freed. This could be error-prone for future xmit hook ops. One of the
possible errors is to return statuses of dst_output directly.
To make the code safer, redefine LWTUNNEL_XMIT_CONTINUE value to
distinguish from dst_output statuses and check the continue
condition explicitly.
Fixes: 3a0af8fd61 ("bpf: BPF for lightweight tunnel infrastructure")
Suggested-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/96b939b85eda00e8df4f7c080f770970a4c5f698.1692326837.git.yan@cloudflare.com
ip_select_ttl() is racy, because it reads inet->uc_ttl
without proper locking.
Add READ_ONCE()/WRITE_ONCE() annotations while
allowing IP_TTL socket option to be set/read without
holding the socket lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
IP_HDRINCL socket option can now be set/read
without locking the socket.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
For now, No matter what error pointer ip_neigh_for_gw() returns,
ip_finish_output2() always return -EINVAL, which may mislead the upper
users.
For exemple, an application uses sendto to send an UDP packet, but when the
neighbor table overflows, sendto() will get a value of -EINVAL, and it will
cause users to waste a lot of time checking parameters for errors.
Return the real errno instead of -EINVAL.
Signed-off-by: xu xin <xu.xin16@zte.com.cn>
Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
Cc: Si Hao <si.hao@zte.com.cn>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Link: https://lore.kernel.org/r/20230807015408.248237-1-xu.xin16@zte.com.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
__ip_append_data() can get into an infinite loop when asked to splice into
a partially-built UDP message that has more than the frag-limit data and up
to the MTU limit. Something like:
pipe(pfd);
sfd = socket(AF_INET, SOCK_DGRAM, 0);
connect(sfd, ...);
send(sfd, buffer, 8161, MSG_CONFIRM|MSG_MORE);
write(pfd[1], buffer, 8);
splice(pfd[0], 0, sfd, 0, 0x4ffe0ul, 0);
where the amount of data given to send() is dependent on the MTU size (in
this instance an interface with an MTU of 8192).
The problem is that the calculation of the amount to copy in
__ip_append_data() goes negative in two places, and, in the second place,
this gets subtracted from the length remaining, thereby increasing it.
This happens when pagedlen > 0 (which happens for MSG_ZEROCOPY and
MSG_SPLICE_PAGES), because the terms in:
copy = datalen - transhdrlen - fraggap - pagedlen;
then mostly cancel when pagedlen is substituted for, leaving just -fraggap.
This causes:
length -= copy + transhdrlen;
to increase the length to more than the amount of data in msg->msg_iter,
which causes skb_splice_from_iter() to be unable to fill the request and it
returns less than 'copied' - which means that length never gets to 0 and we
never exit the loop.
Fix this by:
(1) Insert a note about the dodgy calculation of 'copy'.
(2) If MSG_SPLICE_PAGES, clear copy if it is negative from the above
equation, so that 'offset' isn't regressed and 'length' isn't
increased, which will mean that length and thus copy should match the
amount left in the iterator.
(3) When handling MSG_SPLICE_PAGES, give a warning and return -EIO if
we're asked to splice more than is in the iterator. It might be
better to not give the warning or even just give a 'short' write.
[!] Note that this ought to also affect MSG_ZEROCOPY, but MSG_ZEROCOPY
avoids the problem by simply assuming that everything asked for got copied,
not just the amount that was in the iterator. This is a potential bug for
the future.
Fixes: 7ac7c98785 ("udp: Convert udp_sendpage() to use MSG_SPLICE_PAGES")
Reported-by: syzbot+f527b971b4bdc8e79f9e@syzkaller.appspotmail.com
Link: https://lore.kernel.org/r/000000000000881d0606004541d1@google.com/
Signed-off-by: David Howells <dhowells@redhat.com>
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/1420063.1690904933@warthog.procyon.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
sk_getsockopt() runs locklessly. This means sk->sk_priority
can be read while other threads are changing its value.
Other reads also happen without socket lock being held.
Add missing annotations where needed.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk->sk_mark is often read while another thread could change the value.
Fixes: 4a19ec5800 ("[NET]: Introducing socket mark socket option.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Splicing to SOCK_RAW sockets may set MSG_SPLICE_PAGES, but in such a case,
__ip_append_data() will call skb_splice_from_iter() to access the 'from'
data, assuming it to point to a msghdr struct with an iter, instead of
using the provided getfrag function to access it.
In the case of raw_sendmsg(), however, this is not the case and 'from' will
point to a raw_frag_vec struct and raw_getfrag() will be the frag-getting
function. A similar issue may occur with rawv6_sendmsg().
Fix this by ignoring MSG_SPLICE_PAGES if getfrag != ip_generic_getfrag as
ip_generic_getfrag() expects "from" to be a msghdr*, but the other getfrags
don't. Note that this will prevent MSG_SPLICE_PAGES from being effective
for udplite.
This likely affects ping sockets too. udplite looks like it should be okay
as it expects "from" to be a msghdr.
Signed-off-by: David Howells <dhowells@redhat.com>
Reported-by: syzbot+d8486855ef44506fd675@syzkaller.appspotmail.com
Link: https://lore.kernel.org/r/000000000000ae4cbf05fdeb8349@google.com/
Fixes: 2dc334f1a6 ("splice, net: Use sendmsg(MSG_SPLICE_PAGES) rather than ->sendpage()")
Tested-by: syzbot+d8486855ef44506fd675@syzkaller.appspotmail.com
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/1410156.1686729856@warthog.procyon.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Move declarations into include/net/gso.h and code into net/core/gso.c
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230608191738.3947077-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When using IPv4/TCP, skb->hash comes from sk->sk_txhash except in
TIME_WAIT and SYN_RECV where it's not set in the reply skb from
ip_send_unicast_reply. Those packets will have a mismatched hash with
others from the same flow as their hashes will be 0. IPv6 does not have
the same issue as the hash is set from the socket txhash in those cases.
This commits sets the hash in the reply skb from ip_send_unicast_reply,
which makes the IPv4 code behaving like IPv6.
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
ip_append_page() is no longer used with the removal of udp_sendpage(), so
remove it.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Make IP/UDP sendmsg() support MSG_SPLICE_PAGES. This causes pages to be
spliced from the source iterator.
This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pass the maximum number of fragments into skb_append_pagefrags() rather
than using MAX_SKB_FRAGS so that it can be used from code that wants to
specify sysctl_max_skb_frags.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Like commit ea30388bae ("ipv6: Fix an uninit variable access bug in
__ip6_make_skb()"). icmphdr does not in skb linear region under the
scenario of SOCK_RAW socket. Access icmp_hdr(skb)->type directly will
trigger the uninit variable access bug.
Use a local variable icmp_type to carry the correct value in different
scenarios.
Fixes: 96793b4825 ("[IPV4]: Add ICMPMsgStats MIB (RFC 4293)")
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ziyang Xuan <william.xuanziyang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rcu_bh is no longer a win, especially for objects freed
with standard call_rcu().
Switch neighbour code to no longer disable BH when not necessary.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When the feature was added it was enabled for SW timestamps only but
with current hardware the same out-of-order timestamps can be seen.
Let's expand the area for the feature to all types of timestamps.
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to Eric's IPv6 BIG TCP, this patch is to enable IPv4 BIG TCP.
Firstly, allow sk->sk_gso_max_size to be set to a value greater than
GSO_LEGACY_MAX_SIZE by not trimming gso_max_size in sk_trim_gso_size()
for IPv4 TCP sockets.
Then on TX path, set IP header tot_len to 0 when skb->len > IP_MAX_MTU
in __ip_local_out() to allow to send BIG TCP packets, and this implies
that skb->len is the length of a IPv4 packet; On RX path, use skb->len
as the length of the IPv4 packet when the IP header tot_len is 0 and
skb->len > IP_MAX_MTU in ip_rcv_core(). As the API iph_set_totlen() and
skb_ip_totlen() are used in __ip_local_out() and ip_rcv_core(), we only
need to update these APIs.
Also in GRO receive, add the check for ETH_P_IP/IPPROTO_TCP, and allows
the merged packet size >= GRO_LEGACY_MAX_SIZE in skb_gro_receive(). In
GRO complete, set IP header tot_len to 0 when the merged packet size
greater than IP_MAX_MTU in iph_set_totlen() so that it can be processed
on RX path.
Note that by checking skb_is_gso_tcp() in API iph_totlen(), it makes
this implementation safe to use iph->len == 0 indicates IPv4 BIG TCP
packets.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We can benefit from a smaller struct ubuf_info, so leave only mandatory
fields and let users to decide how they want to extend it. Convert
MSG_ZEROCOPY to struct ubuf_info_msgzc and remove duplicated fields.
This reduces the size from 48 bytes to just 16.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
While reading sysctl_[rw]mem_(max|default), they can be changed
concurrently. Thus, we need to add READ_ONCE() to its readers.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pavel Begunkov says:
====================
io_uring zerocopy send
The patchset implements io_uring zerocopy send. It works with both registered
and normal buffers, mixing is allowed but not recommended. Apart from usual
request completions, just as with MSG_ZEROCOPY, io_uring separately notifies
the userspace when buffers are freed and can be reused (see API design below),
which is delivered into io_uring's Completion Queue. Those "buffer-free"
notifications are not necessarily per request, but the userspace has control
over it and should explicitly attaching a number of requests to a single
notification. The series also adds some internal optimisations when used with
registered buffers like removing page referencing.
From the kernel networking perspective there are two main changes. The first
one is passing ubuf_info into the network layer from io_uring (inside of an
in kernel struct msghdr). This allows extra optimisations, e.g. ubuf_info
caching on the io_uring side, but also helps to avoid cross-referencing
and synchronisation problems. The second part is an optional optimisation
removing page referencing for requests with registered buffers.
Benchmarking UDP with an optimised version of the selftest (see [1]), which
sends a bunch of requests, waits for completions and repeats. "+ flush" column
posts one additional "buffer-free" notification per request, and just "zc"
doesn't post buffer notifications at all.
NIC (requests / second):
IO size | non-zc | zc | zc + flush
4000 | 495134 | 606420 (+22%) | 558971 (+12%)
1500 | 551808 | 577116 (+4.5%) | 565803 (+2.5%)
1000 | 584677 | 592088 (+1.2%) | 560885 (-4%)
600 | 596292 | 598550 (+0.4%) | 555366 (-6.7%)
dummy (requests / second):
IO size | non-zc | zc | zc + flush
8000 | 1299916 | 2396600 (+84%) | 2224219 (+71%)
4000 | 1869230 | 2344146 (+25%) | 2170069 (+16%)
1200 | 2071617 | 2361960 (+14%) | 2203052 (+6%)
600 | 2106794 | 2381527 (+13%) | 2195295 (+4%)
Previously it also brought a massive performance speedup compared to the
msg_zerocopy tool (see [3]), which is probably not super interesting. There
is also an additional bunch of refcounting optimisations that was omitted from
the series for simplicity and as they don't change the picture drastically,
they will be sent as follow up, as well as flushing optimisations closing the
performance gap b/w two last columns.
For TCP on localhost (with hacks enabling localhost zerocopy) and including
additional overhead for receive:
IO size | non-zc | zc
1200 | 4174 | 4148
4096 | 7597 | 11228
Using a real NIC 1200 bytes, zc is worse than non-zc ~5-10%, maybe the
omitted optimisations will somewhat help, should look better for 4000,
but couldn't test properly because of setup problems.
Links:
liburing (benchmark + tests):
[1] https://github.com/isilence/liburing/tree/zc_v4
kernel repo:
[2] https://github.com/isilence/linux/tree/zc_v4
RFC v1:
[3] https://lore.kernel.org/io-uring/cover.1638282789.git.asml.silence@gmail.com/
RFC v2:
https://lore.kernel.org/io-uring/cover.1640029579.git.asml.silence@gmail.com/
Net patches based:
git@github.com:isilence/linux.git zc_v4-net-base
or
https://github.com/isilence/linux/tree/zc_v4-net-base
API design overview:
The series introduces an io_uring concept of notifactors. From the userspace
perspective it's an entity to which it can bind one or more requests and then
requesting to flush it. Flushing a notifier makes it impossible to attach new
requests to it, and instructs the notifier to post a completion once all
requests attached to it are completed and the kernel doesn't need the buffers
anymore.
Notifications are stored in notification slots, which should be registered as
an array in io_uring. Each slot stores only one notifier at any particular
moment. Flushing removes it from the slot and the slot automatically replaces
it with a new notifier. All operations with notifiers are done by specifying
an index of a slot it's currently in.
When registering a notification the userspace specifies a u64 tag for each
slot, which will be copied in notification completion entries as
cqe::user_data. cqe::res is 0 and cqe::flags is equal to wrap around u32
sequence number counting notifiers of a slot.
====================
Link: https://lore.kernel.org/r/cover.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Teach ipv4/udp how to use external ubuf_info provided in msghdr and
also prepare it for managed frags by sprinkling
skb_zcopy_downgrade_managed() when it could mix managed and not managed
frags.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Even when zerocopy transmission is requested and possible,
__ip_append_data() will still copy a small chunk of data just because it
allocated some extra linear space (e.g. 148 bytes). It wastes CPU cycles
on copy and iter manipulations and also misalignes potentially aligned
data. Avoid such copies. And as a bonus we can allocate smaller skb.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If we set XFRM security policy by calling setsockopt with option
IPV6_XFRM_POLICY, the policy will be stored in 'sock_policy' in 'sock'
struct. However tcp_v6_send_response doesn't look up dst_entry with the
actual socket but looks up with tcp control socket. This may cause a
problem that a RST packet is sent without ESP encryption & peer's TCP
socket can't receive it.
This patch will make the function look up dest_entry with actual socket,
if the socket has XFRM policy(sock_policy), so that the TCP response
packet via this function can be encrypted, & aligned on the encrypted
TCP socket.
Tested: We encountered this problem when a TCP socket which is encrypted
in ESP transport mode encryption, receives challenge ACK at SYN_SENT
state. After receiving challenge ACK, TCP needs to send RST to
establish the socket at next SYN try. But the RST was not encrypted &
peer TCP socket still remains on ESTABLISHED state.
So we verified this with test step as below.
[Test step]
1. Making a TCP state mismatch between client(IDLE) & server(ESTABLISHED).
2. Client tries a new connection on the same TCP ports(src & dst).
3. Server will return challenge ACK instead of SYN,ACK.
4. Client will send RST to server to clear the SOCKET.
5. Client will retransmit SYN to server on the same TCP ports.
[Expected result]
The TCP connection should be established.
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Cc: Sehee Lee <seheele@google.com>
Signed-off-by: Sewook Seo <sewookseo@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the len fields manipulation in the skbs to a helper function.
There is a comment specifically requesting this and there are several
other areas in the code displaying the same pattern which can be
refactored.
This improves code readability.
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
Link: https://lore.kernel.org/r/20220622160853.GA6478@debian
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The previous patches handled the delivery_time before sch_handle_ingress().
This patch can now set the skb->mono_delivery_time to flag the skb->tstamp
is used as the mono delivery_time (EDT) instead of the (rcv) timestamp
and also clear it with skb_clear_delivery_time() after
sch_handle_ingress(). This will make the bpf_redirect_*()
to keep the mono delivery_time and used by a qdisc (fq) of
the egress-ing interface.
A latter patch will postpone the skb_clear_delivery_time() until the
stack learns that the skb is being delivered locally and that will
make other kernel forwarding paths (ip[6]_forward) able to keep
the delivery_time also. Thus, like the previous patches on using
the skb->mono_delivery_time bit, calling skb_clear_delivery_time()
is not limited within the CONFIG_NET_INGRESS to avoid too many code
churns among this set.
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb->tstamp was first used as the (rcv) timestamp.
The major usage is to report it to the user (e.g. SO_TIMESTAMP).
Later, skb->tstamp is also set as the (future) delivery_time (e.g. EDT in TCP)
during egress and used by the qdisc (e.g. sch_fq) to make decision on when
the skb can be passed to the dev.
Currently, there is no way to tell skb->tstamp having the (rcv) timestamp
or the delivery_time, so it is always reset to 0 whenever forwarded
between egress and ingress.
While it makes sense to always clear the (rcv) timestamp in skb->tstamp
to avoid confusing sch_fq that expects the delivery_time, it is a
performance issue [0] to clear the delivery_time if the skb finally
egress to a fq@phy-dev. For example, when forwarding from egress to
ingress and then finally back to egress:
tcp-sender => veth@netns => veth@hostns => fq@eth0@hostns
^ ^
reset rest
This patch adds one bit skb->mono_delivery_time to flag the skb->tstamp
is storing the mono delivery_time (EDT) instead of the (rcv) timestamp.
The current use case is to keep the TCP mono delivery_time (EDT) and
to be used with sch_fq. A latter patch will also allow tc-bpf@ingress
to read and change the mono delivery_time.
In the future, another bit (e.g. skb->user_delivery_time) can be added
for the SCM_TXTIME where the clock base is tracked by sk->sk_clockid.
[ This patch is a prep work. The following patches will
get the other parts of the stack ready first. Then another patch
after that will finally set the skb->mono_delivery_time. ]
skb_set_delivery_time() function is added. It is used by the tcp_output.c
and during ip[6] fragmentation to assign the delivery_time to
the skb->tstamp and also set the skb->mono_delivery_time.
A note on the change in ip_send_unicast_reply() in ip_output.c.
It is only used by TCP to send reset/ack out of a ctl_sk.
Like the new skb_set_delivery_time(), this patch sets
the skb->mono_delivery_time to 0 for now as a place
holder. It will be enabled in a latter patch.
A similar case in tcp_ipv6 can be done with
skb_set_delivery_time() in tcp_v6_send_response().
[0] (slide 22): https://linuxplumbersconf.org/event/11/contributions/953/attachments/867/1658/LPC_2021_BPF_Datapath_Extensions.pdf
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace kfree_skb() which is used in the packet egress path of IP layer
with kfree_skb_reason(). Functions that are involved include:
__ip_queue_xmit()
ip_finish_output()
ip_mc_finish_output()
ip6_output()
ip6_finish_output()
ip6_finish_output2()
Following new drop reasons are introduced:
SKB_DROP_REASON_IP_OUTNOROUTES
SKB_DROP_REASON_BPF_CGROUP_EGRESS
SKB_DROP_REASON_IPV6DISABLED
SKB_DROP_REASON_NEIGH_CREATEFAIL
Reviewed-by: Mengen Sun <mengensun@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP sendmsg() can be lockless, this is causing all kinds
of data races.
This patch converts sk->sk_tskey to remove one of these races.
BUG: KCSAN: data-race in __ip_append_data / __ip_append_data
read to 0xffff8881035d4b6c of 4 bytes by task 8877 on cpu 1:
__ip_append_data+0x1c1/0x1de0 net/ipv4/ip_output.c:994
ip_make_skb+0x13f/0x2d0 net/ipv4/ip_output.c:1636
udp_sendmsg+0x12bd/0x14c0 net/ipv4/udp.c:1249
inet_sendmsg+0x5f/0x80 net/ipv4/af_inet.c:819
sock_sendmsg_nosec net/socket.c:705 [inline]
sock_sendmsg net/socket.c:725 [inline]
____sys_sendmsg+0x39a/0x510 net/socket.c:2413
___sys_sendmsg net/socket.c:2467 [inline]
__sys_sendmmsg+0x267/0x4c0 net/socket.c:2553
__do_sys_sendmmsg net/socket.c:2582 [inline]
__se_sys_sendmmsg net/socket.c:2579 [inline]
__x64_sys_sendmmsg+0x53/0x60 net/socket.c:2579
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
write to 0xffff8881035d4b6c of 4 bytes by task 8880 on cpu 0:
__ip_append_data+0x1d8/0x1de0 net/ipv4/ip_output.c:994
ip_make_skb+0x13f/0x2d0 net/ipv4/ip_output.c:1636
udp_sendmsg+0x12bd/0x14c0 net/ipv4/udp.c:1249
inet_sendmsg+0x5f/0x80 net/ipv4/af_inet.c:819
sock_sendmsg_nosec net/socket.c:705 [inline]
sock_sendmsg net/socket.c:725 [inline]
____sys_sendmsg+0x39a/0x510 net/socket.c:2413
___sys_sendmsg net/socket.c:2467 [inline]
__sys_sendmmsg+0x267/0x4c0 net/socket.c:2553
__do_sys_sendmmsg net/socket.c:2582 [inline]
__se_sys_sendmmsg net/socket.c:2579 [inline]
__x64_sys_sendmmsg+0x53/0x60 net/socket.c:2579
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
value changed: 0x0000054d -> 0x0000054e
Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 8880 Comm: syz-executor.5 Not tainted 5.17.0-rc2-syzkaller-00167-gdcb85f85fa6f-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Fixes: 09c2d251b7 ("net-timestamp: add key to disambiguate concurrent datagrams")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since v2.5.44 and addition of ip_options_fragment()
ip_options_build() does not render headers for fragments
directly. @is_frag is always 0.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 431280eebe ("ipv4: tcp: send zero IPID for RST and
ACK sent in SYN-RECV and TIME-WAIT state") we took care of some
ctl packets sent by TCP.
It turns out we need to use a similar strategy for SYNACK packets.
By default, they carry IP_DF and IPID==0, but there are ways
to ask them to use the hashed IP ident generator and thus
be used to build off-path attacks.
(Ref: Off-Path TCP Exploits of the Mixed IPID Assignment)
One of this way is to force (before listener is started)
echo 1 >/proc/sys/net/ipv4/ip_no_pmtu_disc
Another way is using forged ICMP ICMP_FRAG_NEEDED
with a very small MTU (like 68) to force a false return from
ip_dont_fragment()
In this patch, ip_build_and_send_pkt() uses the following
heuristics.
1) Most SYNACK packets are smaller than IPV4_MIN_MTU and therefore
can use IP_DF regardless of the listener or route pmtu setting.
2) In case the SYNACK packet is bigger than IPV4_MIN_MTU,
we use prandom_u32() generator instead of the IPv4 hashed ident one.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Ray Che <xijiache@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Cc: Geoff Alexander <alexandg@cs.unm.edu>
Cc: Willy Tarreau <w@1wt.eu>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
During IP fragmentation we sanitize IP options. This means overwriting
options which should not be copied with NOPs. Only the first fragment
has the original, full options.
ip_fraglist_prepare() copies the IP header and options from previous
fragment to the next one. Commit 19c3401a91 ("net: ipv4: place control
buffer handling away from fragmentation iterators") moved sanitizing
options before ip_fraglist_prepare() which means options are sanitized
and then overwritten again with the old values.
Fixing this is not enough, however, nor did the sanitization work
prior to aforementioned commit.
ip_options_fragment() (which does the sanitization) uses ipcb->opt.optlen
for the length of the options. ipcb->opt of fragments is not populated
(it's 0), only the head skb has the state properly built. So even when
called at the right time ip_options_fragment() does nothing. This seems
to date back all the way to v2.5.44 when the fast path for pre-fragmented
skbs had been introduced. Prior to that ip_options_build() would have been
called for every fragment (in fact ever since v2.5.44 the fragmentation
handing in ip_options_build() has been dead code, I'll clean it up in
-next).
In the original patch (see Link) caixf mentions fixing the handling
for fragments other than the second one, but I'm not sure how _any_
fragment could have had their options sanitized with the code
as it stood.
Tested with python (MTU on lo lowered to 1000 to force fragmentation):
import socket
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
s.setsockopt(socket.IPPROTO_IP, socket.IP_OPTIONS,
bytearray([7,4,5,192, 20|0x80,4,1,0]))
s.sendto(b'1'*2000, ('127.0.0.1', 1234))
Before:
IP (tos 0x0, ttl 64, id 1053, offset 0, flags [+], proto UDP (17), length 996, options (RR [bad length 4] [bad ptr 5] 192.148.4.1,,RA value 256))
localhost.36500 > localhost.search-agent: UDP, length 2000
IP (tos 0x0, ttl 64, id 1053, offset 968, flags [+], proto UDP (17), length 996, options (RR [bad length 4] [bad ptr 5] 192.148.4.1,,RA value 256))
localhost > localhost: udp
IP (tos 0x0, ttl 64, id 1053, offset 1936, flags [none], proto UDP (17), length 100, options (RR [bad length 4] [bad ptr 5] 192.148.4.1,,RA value 256))
localhost > localhost: udp
After:
IP (tos 0x0, ttl 96, id 42549, offset 0, flags [+], proto UDP (17), length 996, options (RR [bad length 4] [bad ptr 5] 192.148.4.1,,RA value 256))
localhost.51607 > localhost.search-agent: UDP, bad length 2000 > 960
IP (tos 0x0, ttl 96, id 42549, offset 968, flags [+], proto UDP (17), length 996, options (NOP,NOP,NOP,NOP,RA value 256))
localhost > localhost: udp
IP (tos 0x0, ttl 96, id 42549, offset 1936, flags [none], proto UDP (17), length 100, options (NOP,NOP,NOP,NOP,RA value 256))
localhost > localhost: udp
RA (20 | 0x80) is now copied as expected, RR (7) is "NOPed out".
Link: https://lore.kernel.org/netdev/20220107080559.122713-1-ooppublic@163.com/
Fixes: 19c3401a91 ("net: ipv4: place control buffer handling away from fragmentation iterators")
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: caixf <ooppublic@163.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
there is a same action when the variable is initialized
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: luo penghao <luo.penghao@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a if statements to avoid the warning.
Dan Carpenter report:
The patch faf482ca19: "net: ipv4: Move ip_options_fragment() out of
loop" from Aug 23, 2021, leads to the following Smatch complaint:
net/ipv4/ip_output.c:833 ip_do_fragment()
warn: variable dereferenced before check 'iter.frag' (see line 828)
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: faf482ca19 ("net: ipv4: Move ip_options_fragment() out of loop")
Link: https://lore.kernel.org/netdev/20210830073802.GR7722@kadam/T/#t
Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ip_options_fragment() only called when iter->offset is equal to zero,
so move it out of loop, and inline 'Copy the flags to each fragment.'
As also, remove the unused parameter in ip_frag_ipcb().
Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
Unlike skb_realloc_headroom, new helper skb_expand_head
does not allocate a new skb if possible.
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix the following out-of-bounds warning:
In function 'ip_copy_addrs',
inlined from '__ip_queue_xmit' at net/ipv4/ip_output.c:517:2:
net/ipv4/ip_output.c:449:2: warning: 'memcpy' offset [40, 43] from the object at 'fl' is out of the bounds of referenced subobject 'saddr' with type 'unsigned int' at offset 36 [-Warray-bounds]
449 | memcpy(&iph->saddr, &fl4->saddr,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
450 | sizeof(fl4->saddr) + sizeof(fl4->daddr));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The problem is that the original code is trying to copy data into a
couple of struct members adjacent to each other in a single call to
memcpy(). This causes a legitimate compiler warning because memcpy()
overruns the length of &iph->saddr and &fl4->saddr. As these are just
a couple of struct members, fix this by using direct assignments,
instead of memcpy().
This helps with the ongoing efforts to globally enable -Warray-bounds
and get us closer to being able to tighten the FORTIFY_SOURCE routines
on memcpy().
Link: https://github.com/KSPP/linux/issues/109
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/lkml/d5ae2e65-1f18-2577-246f-bada7eee6ccd@intel.com/
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dave observed number of machines hitting OOM on the UDP send
path. The workload seems to be sending large UDP packets over
loopback. Since loopback has MTU of 64k kernel will try to
allocate an skb with up to 64k of head space. This has a good
chance of failing under memory pressure. What's worse if
the message length is <32k the allocation may trigger an
OOM killer.
This is entirely avoidable, we can use an skb with page frags.
af_unix solves a similar problem by limiting the head
length to SKB_MAX_ALLOC. This seems like a good and simple
approach. It means that UDP messages > 16kB will now
use fragments if underlying device supports SG, if extra
allocator pressure causes regressions in real workloads
we can switch to trying the large allocation first and
falling back.
v4: pre-calculate all the additions to alloclen so
we can be sure it won't go over order-2
Reported-by: Dave Jones <dsj@fb.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
s/readibility/readability/
s/insufficent/insufficient/
Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>