Use sendmsg() with MSG_SPLICE_PAGES rather than sendpage in
skb_send_sock(). This causes pages to be spliced from the source iterator
if possible.
This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.
Note that this could perhaps be improved to fill out a bvec array with all
the frags and then make a single sendmsg call, possibly sticking the header
on the front also.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Link: https://lore.kernel.org/r/20230623225513.2732256-3-dhowells@redhat.com
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>
The "len" variable needs to be signed for the error handling to work
correctly.
Fixes: 2e910b9532 ("net: Add a function to splice pages into an skbuff for MSG_SPLICE_PAGES")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/366861a7-87c8-4bbf-9101-69dd41021d07@kili.mountain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 50749f2dd6 ("tcp/udp: Fix memleaks of sk and zerocopy skbs with
TX timestamp.") added a call to skb_orphan_frags_rx() to fix leaks with
zerocopy skbs. But it ended up adding a leak of its own. When
skb_orphan_frags_rx() fails, the function just returns, leaking the skb
it just cloned. Free it before returning.
This bug was discovered and resolved using Coverity Static Analysis
Security Testing (SAST) by Synopsys, Inc.
Fixes: 50749f2dd6 ("tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp.")
Signed-off-by: Pratyush Yadav <ptyadav@amazon.de>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20230522153020.32422-1-ptyadav@amazon.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a function to handle MSG_SPLICE_PAGES being passed internally to
sendmsg(). Pages are spliced into the given socket buffer if possible and
copied in if not (e.g. they're slab pages or have a zero refcount).
Signed-off-by: David Howells <dhowells@redhat.com>
cc: David Ahern <dsahern@kernel.org>
cc: Al Viro <viro@zeniv.linux.org.uk>
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>
Most users use __skb_frag_set_page()/skb_frag_off_set()/
skb_frag_size_set() to fill the page desc for a skb frag.
Introduce skb_frag_fill_page_desc() to do that.
net/bpf/test_run.c does not call skb_frag_off_set() to
set the offset, "copy_from_user(page_address(page), ...)"
and 'shinfo' being part of the 'data' kzalloced in
bpf_test_init() suggest that it is assuming offset to be
initialized as zero, so call skb_frag_fill_page_desc()
with offset being zero for this case.
Also, skb_frag_set_page() is not used anymore, so remove
it.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit c9929f0e34 ("mm/slob: remove CONFIG_SLOB") removes CONFIG_SLOB.
Now, we can also remove special handling for socket buffers with the SLOB
allocator. The code with HAVE_SKB_SMALL_HEAD_CACHE=1 is now the default
behavior for all allocators.
Remove an unnecessary distinction between SLOB and SLAB/SLUB allocator
after the SLOB allocator is gone.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230509071207.28942-1-lukas.bulwahn@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
David Ahern reported crashes in skb_copy_ubufs() caused by TCP tx zerocopy
using hugepages, and skb length bigger than ~68 KB.
skb_copy_ubufs() assumed it could copy all payload using up to
MAX_SKB_FRAGS order-0 pages.
This assumption broke when BIG TCP was able to put up to 512 KB per skb.
We did not hit this bug at Google because we use CONFIG_MAX_SKB_FRAGS=45
and limit gso_max_size to 180000.
A solution is to use higher order pages if needed.
v2: add missing __GFP_COMP, or we leak memory.
Fixes: 7c4e983c4f ("net: allow gso_max_size to exceed 65536")
Reported-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/netdev/c70000f6-baa4-4a05-46d0-4b3e0dc1ccc8@gmail.com/T/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Xin Long <lucien.xin@gmail.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Coco Li <lixiaoyan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzkaller reported [0] memory leaks of an UDP socket and ZEROCOPY
skbs. We can reproduce the problem with these sequences:
sk = socket(AF_INET, SOCK_DGRAM, 0)
sk.setsockopt(SOL_SOCKET, SO_TIMESTAMPING, SOF_TIMESTAMPING_TX_SOFTWARE)
sk.setsockopt(SOL_SOCKET, SO_ZEROCOPY, 1)
sk.sendto(b'', MSG_ZEROCOPY, ('127.0.0.1', 53))
sk.close()
sendmsg() calls msg_zerocopy_alloc(), which allocates a skb, sets
skb->cb->ubuf.refcnt to 1, and calls sock_hold(). Here, struct
ubuf_info_msgzc indirectly holds a refcnt of the socket. When the
skb is sent, __skb_tstamp_tx() clones it and puts the clone into
the socket's error queue with the TX timestamp.
When the original skb is received locally, skb_copy_ubufs() calls
skb_unclone(), and pskb_expand_head() increments skb->cb->ubuf.refcnt.
This additional count is decremented while freeing the skb, but struct
ubuf_info_msgzc still has a refcnt, so __msg_zerocopy_callback() is
not called.
The last refcnt is not released unless we retrieve the TX timestamped
skb by recvmsg(). Since we clear the error queue in inet_sock_destruct()
after the socket's refcnt reaches 0, there is a circular dependency.
If we close() the socket holding such skbs, we never call sock_put()
and leak the count, sk, and skb.
TCP has the same problem, and commit e0c8bccd40 ("net: stream:
purge sk_error_queue in sk_stream_kill_queues()") tried to fix it
by calling skb_queue_purge() during close(). However, there is a
small chance that skb queued in a qdisc or device could be put
into the error queue after the skb_queue_purge() call.
In __skb_tstamp_tx(), the cloned skb should not have a reference
to the ubuf to remove the circular dependency, but skb_clone() does
not call skb_copy_ubufs() for zerocopy skb. So, we need to call
skb_orphan_frags_rx() for the cloned skb to call skb_copy_ubufs().
[0]:
BUG: memory leak
unreferenced object 0xffff88800c6d2d00 (size 1152):
comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 cd af e8 81 00 00 00 00 ................
02 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00 ...@............
backtrace:
[<0000000055636812>] sk_prot_alloc+0x64/0x2a0 net/core/sock.c:2024
[<0000000054d77b7a>] sk_alloc+0x3b/0x800 net/core/sock.c:2083
[<0000000066f3c7e0>] inet_create net/ipv4/af_inet.c:319 [inline]
[<0000000066f3c7e0>] inet_create+0x31e/0xe40 net/ipv4/af_inet.c:245
[<000000009b83af97>] __sock_create+0x2ab/0x550 net/socket.c:1515
[<00000000b9b11231>] sock_create net/socket.c:1566 [inline]
[<00000000b9b11231>] __sys_socket_create net/socket.c:1603 [inline]
[<00000000b9b11231>] __sys_socket_create net/socket.c:1588 [inline]
[<00000000b9b11231>] __sys_socket+0x138/0x250 net/socket.c:1636
[<000000004fb45142>] __do_sys_socket net/socket.c:1649 [inline]
[<000000004fb45142>] __se_sys_socket net/socket.c:1647 [inline]
[<000000004fb45142>] __x64_sys_socket+0x73/0xb0 net/socket.c:1647
[<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
[<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
BUG: memory leak
unreferenced object 0xffff888017633a00 (size 240):
comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 2d 6d 0c 80 88 ff ff .........-m.....
backtrace:
[<000000002b1c4368>] __alloc_skb+0x229/0x320 net/core/skbuff.c:497
[<00000000143579a6>] alloc_skb include/linux/skbuff.h:1265 [inline]
[<00000000143579a6>] sock_omalloc+0xaa/0x190 net/core/sock.c:2596
[<00000000be626478>] msg_zerocopy_alloc net/core/skbuff.c:1294 [inline]
[<00000000be626478>] msg_zerocopy_realloc+0x1ce/0x7f0 net/core/skbuff.c:1370
[<00000000cbfc9870>] __ip_append_data+0x2adf/0x3b30 net/ipv4/ip_output.c:1037
[<0000000089869146>] ip_make_skb+0x26c/0x2e0 net/ipv4/ip_output.c:1652
[<00000000098015c2>] udp_sendmsg+0x1bac/0x2390 net/ipv4/udp.c:1253
[<0000000045e0e95e>] inet_sendmsg+0x10a/0x150 net/ipv4/af_inet.c:819
[<000000008d31bfde>] sock_sendmsg_nosec net/socket.c:714 [inline]
[<000000008d31bfde>] sock_sendmsg+0x141/0x190 net/socket.c:734
[<0000000021e21aa4>] __sys_sendto+0x243/0x360 net/socket.c:2117
[<00000000ac0af00c>] __do_sys_sendto net/socket.c:2129 [inline]
[<00000000ac0af00c>] __se_sys_sendto net/socket.c:2125 [inline]
[<00000000ac0af00c>] __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2125
[<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
[<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
Fixes: f214f915e7 ("tcp: enable MSG_ZEROCOPY")
Fixes: b5947e5d1e ("udp: msg_zerocopy")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ocelot_xmit_get_vlan_info() calls __skb_vlan_pop() as the most
appropriate helper I could find which strips away a VLAN header.
That's all I need it to do, but __skb_vlan_pop() has more logic, which
will become incompatible with the future revert of commit 6d1ccff627
("net: reset mac header in dev_start_xmit()").
Namely, it performs a sanity check on skb_mac_header(), which will stop
being set after the above revert, so it will return an error instead of
removing the VLAN tag.
ocelot_xmit_get_vlan_info() gets called in 2 circumstances:
(1) the port is under a VLAN-aware bridge and the bridge sends
VLAN-tagged packets
(2) the port is under a VLAN-aware bridge and somebody else (an 8021q
upper) sends VLAN-tagged packets (using a VID that isn't in the
bridge vlan tables)
In case (1), there is actually no bug to defend against, because
br_dev_xmit() calls skb_reset_mac_header() and things continue to work.
However, in case (2), illustrated using the commands below, it can be
seen that our intervention is needed, since __skb_vlan_pop() complains:
$ ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up
$ ip link set $eth master br0 && ip link set $eth up
$ ip link add link $eth name $eth.100 type vlan id 100 && ip link set $eth.100 up
$ ip addr add 192.168.100.1/24 dev $eth.100
I could fend off the checks in __skb_vlan_pop() with some
skb_mac_header_was_set() calls, but seeing how few callers of
__skb_vlan_pop() there are from TX paths, that seems rather
unproductive.
As an alternative solution, extract the bare minimum logic to strip a
VLAN header, and move it to a new helper named vlan_remove_tag(), close
to the definition of vlan_insert_tag(). Document it appropriately and
make ocelot_xmit_get_vlan_info() call this smaller helper instead.
Seeing that it doesn't appear illegal to test skb->protocol in the TX
path, I guess it would be a good for vlan_remove_tag() to also absorb
the vlan_set_encap_proto() function call.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
kfree_skb() can be called from hard irq handlers,
but skb_attempt_defer_free() is meant to be used
from process or BH contexts, and skb_defer_free_flush()
is meant to be called from BH contexts.
Not having to mask hard irq can save some cycles.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend drop reasons to make them usable by subsystems
other than core by reserving the high 16 bits for a
new subsystem ID, of which 0 of course is used for the
existing reasons immediately.
To still be able to have string reasons, restructure
that code a bit to make the loopup under RCU, the only
user of this (right now) is drop_monitor.
Link: https://lore.kernel.org/netdev/00659771ed54353f92027702c5bbb84702da62ce.camel@sipsolutions.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
__kfree_skb_defer() uses the old naming where "defer" meant
slab bulk free/alloc APIs. In the meantime we also made
__kfree_skb_defer() feed the per-NAPI skb cache, which
implies bulk APIs. So take away the 'defer' and add 'napi'.
While at it add a drop reason. This only matters on the
tx_action path, if the skb has a frag_list. But getting
rid of a SKB_DROP_REASON_NOT_SPECIFIED seems like a net
benefit so why not.
Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230420020005.815854-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Datacenter kernel builds will very likely not include WIRELESS,
so let them shave 2 bits off the skb by hiding the wifi fields.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Recent patches to mlx5 mentioned a regression when moving from
driver local page pool to only using the generic page pool code.
Page pool has two recycling paths (1) direct one, which runs in
safe NAPI context (basically consumer context, so producing
can be lockless); and (2) via a ptr_ring, which takes a spin
lock because the freeing can happen from any CPU; producer
and consumer may run concurrently.
Since the page pool code was added, Eric introduced a revised version
of deferred skb freeing. TCP skbs are now usually returned to the CPU
which allocated them, and freed in softirq context. This places the
freeing (producing of pages back to the pool) enticingly close to
the allocation (consumer).
If we can prove that we're freeing in the same softirq context in which
the consumer NAPI will run - lockless use of the cache is perfectly fine,
no need for the lock.
Let drivers link the page pool to a NAPI instance. If the NAPI instance
is scheduled on the same CPU on which we're freeing - place the pages
in the direct cache.
With that and patched bnxt (XDP enabled to engage the page pool, sigh,
bnxt really needs page pool work :() I see a 2.6% perf boost with
a TCP stream test (app on a different physical core than softirq).
The CPU use of relevant functions decreases as expected:
page_pool_refill_alloc_cache 1.17% -> 0%
_raw_spin_lock 2.41% -> 0.98%
Only consider lockless path to be safe when NAPI is scheduled
- in practice this should cover majority if not all of steady state
workloads. It's usually the NAPI kicking in that causes the skb flush.
The main case we'll miss out on is when application runs on the same
CPU as NAPI. In that case we don't use the deferred skb free path.
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We maintain a NAPI-local cache of skbs which is fed by napi_consume_skb().
Going forward we will also try to cache head and data pages.
Plumb the "are we in a normal NAPI context" information thru
deeper into the freeing path, up to skb_release_data() and
skb_free_head()/skb_pp_recycle(). The "not normal NAPI context"
comes from netpoll which passes budget of 0 to try to reap
the Tx completions but not perform any Rx.
Use "bool napi_safe" rather than bare "int budget",
the further we get from NAPI the more confusing the budget
argument may seem (particularly whether 0 or MAX is the
correct value to pass in when not in NAPI).
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 1effe8ca4e ("skbuff: fix coalescing for page_pool fragment
recycling") allowed coalescing to proceed with non page pool page and page
pool page when @from is cloned, i.e.
to->pp_recycle --> false
from->pp_recycle --> true
skb_cloned(from) --> true
However, it actually requires skb_cloned(@from) to hold true until
coalescing finishes in this situation. If the other cloned SKB is
released while the merging is in process, from_shinfo->nr_frags will be
set to 0 toward the end of the function, causing the increment of frag
page _refcount to be unexpectedly skipped resulting in inconsistent
reference counts. Later when SKB(@to) is released, it frees the page
directly even though the page pool page is still in use, leading to
use-after-free or double-free errors. So it should be prohibited.
The double-free error message below prompted us to investigate:
BUG: Bad page state in process swapper/1 pfn:0e0d1
page:00000000c6548b28 refcount:-1 mapcount:0 mapping:0000000000000000
index:0x2 pfn:0xe0d1
flags: 0xfffffc0000000(node=0|zone=1|lastcpupid=0x1fffff)
raw: 000fffffc0000000 0000000000000000 ffffffff00000101 0000000000000000
raw: 0000000000000002 0000000000000000 ffffffffffffffff 0000000000000000
page dumped because: nonzero _refcount
CPU: 1 PID: 0 Comm: swapper/1 Tainted: G E 6.2.0+
Call Trace:
<IRQ>
dump_stack_lvl+0x32/0x50
bad_page+0x69/0xf0
free_pcp_prepare+0x260/0x2f0
free_unref_page+0x20/0x1c0
skb_release_data+0x10b/0x1a0
napi_consume_skb+0x56/0x150
net_rx_action+0xf0/0x350
? __napi_schedule+0x79/0x90
__do_softirq+0xc8/0x2b1
__irq_exit_rcu+0xb9/0xf0
common_interrupt+0x82/0xa0
</IRQ>
<TASK>
asm_common_interrupt+0x22/0x40
RIP: 0010:default_idle+0xb/0x20
Fixes: 53e0961da1 ("page_pool: add frag page recycling support in page pool")
Signed-off-by: Liang Chen <liangchen.linux@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230413090353.14448-1-liangchen.linux@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Similarly to napi_build_skb(), it is likely the skb allocation in
build_skb() succeeded. frag_size != 0 is also likely, as stated in
__build_skb_around().
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use skb_propagate_pfmemalloc() in build_skb()/build_skb_around() instead
of open-coding it.
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
MSG_ZEROCOPY ensures that pinned user pages do not exceed the limit.
If no limit is set, skip this accounting as otherwise expensive
atomic_long operations are called for no reason.
This accounting is already skipped for privileged (CAP_IPC_LOCK)
users. Rely on the same mechanism: if no mmp->user is set,
mm_unaccount_pinned_pages does not decrement either.
Tested by running tools/testing/selftests/net/msg_zerocopy.sh with
an unprivileged user for the TXMODE binary:
ip netns exec "${NS1}" sudo -u "{$USER}" "${BIN}" "-${IP}" ...
Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230214155740.3448763-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
skbuff_head_cache is misnamed (perhaps for historical reasons?)
because it does not hold heads. Head is the buffer which skb->data
points to, and also where shinfo lives. struct sk_buff is a metadata
structure, not the head.
Eric recently added skb_small_head_cache (which allocates actual
head buffers), let that serve as an excuse to finally clean this up :)
Leave the user-space visible name intact, it could possibly be uAPI.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Recent removal of ksize() in alloc_skb() increased
performance because we no longer read
the associated struct page.
We have an equivalent cost at kfree_skb() time.
kfree(skb->head) has to access a struct page,
often cold in cpu caches to get the owning
struct kmem_cache.
Considering that many allocations are small (at least for TCP ones)
we can have our own kmem_cache to avoid the cache line miss.
This also saves memory because these small heads
are no longer padded to 1024 bytes.
CONFIG_SLUB=y
$ grep skbuff_small_head /proc/slabinfo
skbuff_small_head 2907 2907 640 51 8 : tunables 0 0 0 : slabdata 57 57 0
CONFIG_SLAB=y
$ grep skbuff_small_head /proc/slabinfo
skbuff_small_head 607 624 640 6 1 : tunables 54 27 8 : slabdata 104 104 5
Notes:
- After Kees Cook patches and this one, we might
be able to revert commit
dbae2b0628 ("net: skb: introduce and use a single page frag cache")
because GRO_MAX_HEAD is also small.
- This patch is a NOP for CONFIG_SLOB=y builds.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All kmalloc_reserve() callers have to make the same computation,
we can factorize them, to prepare following patch in the series.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This is a cleanup patch, to prepare following change.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We have many places using this expression:
SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
Use of SKB_HEAD_ALIGN() will allow to clean them.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
First user of skb_poison_list is in kfree_skb_list_reason, to catch bugs
earlier like introduced in commit eedade12f4 ("net: kfree_skb_list use
kmem_cache_free_bulk"). For completeness mentioned bug have been fixed in
commit f72ff8b81e ("net: fix kfree_skb_list use of skb_mark_not_on_list").
In case of a bug like mentioned commit we would have seen OOPS with:
general protection fault, probably for non-canonical address 0xdead000000000870
And content of one the registers e.g. R13: dead000000000800
In this case skb->len is at offset 112 bytes (0x70) why fault happens at
0x800+0x70 = 0x870
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This include was added for skb_find_text() but all we need there
is a forward declaration of struct ts_config.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
A bug was introduced by commit eedade12f4 ("net: kfree_skb_list use
kmem_cache_free_bulk"). It unconditionally unlinked the SKB list via
invoking skb_mark_not_on_list().
In this patch we choose to remove the skb_mark_not_on_list() call as it
isn't necessary. It would be possible and correct to call
skb_mark_not_on_list() only when __kfree_skb_reason() returns true,
meaning the SKB is ready to be free'ed, as it calls/check skb_unref().
This fix is needed as kfree_skb_list() is also invoked on skb_shared_info
frag_list (skb_drop_fraglist() calling kfree_skb_list()). A frag_list can
have SKBs with elevated refcnt due to cloning via skb_clone_fraglist(),
which takes a reference on all SKBs in the list. This implies the
invariant that all SKBs in the list must have the same refcnt, when using
kfree_skb_list().
Reported-by: syzbot+c8a2e66e37eee553c4fd@syzkaller.appspotmail.com
Reported-and-tested-by: syzbot+c8a2e66e37eee553c4fd@syzkaller.appspotmail.com
Fixes: eedade12f4 ("net: kfree_skb_list use kmem_cache_free_bulk")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/167421088417.1125894.9761158218878962159.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The kfree_skb_list function walks SKB (via skb->next) and frees them
individually to the SLUB/SLAB allocator (kmem_cache). It is more
efficient to bulk free them via the kmem_cache_free_bulk API.
This patches create a stack local array with SKBs to bulk free while
walking the list. Bulk array size is limited to 16 SKBs to trade off
stack usage and efficiency. The SLUB kmem_cache "skbuff_head_cache"
uses objsize 256 bytes usually in an order-1 page 8192 bytes that is
32 objects per slab (can vary on archs and due to SLUB sharing). Thus,
for SLUB the optimal bulk free case is 32 objects belonging to same
slab, but runtime this isn't likely to occur.
The expected gain from using kmem_cache bulk alloc and free API
have been assessed via a microbencmark kernel module[1].
The module 'slab_bulk_test01' results at bulk 16 element:
kmem-in-loop Per elem: 109 cycles(tsc) 30.532 ns (step:16)
kmem-bulk Per elem: 64 cycles(tsc) 17.905 ns (step:16)
More detailed description of benchmarks avail in [2].
[1] https://github.com/netoptimizer/prototype-kernel/tree/master/kernel/mm
[2] https://github.com/xdp-project/xdp-project/blob/master/areas/mem/kfree_skb_list01.org
V2: rename function to kfree_skb_add_bulk.
Reviewed-by: Saeed Mahameed <saeed@kernel.org>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The SKB drop reason uses __builtin_return_address(0) to give the call
"location" to trace_kfree_skb() tracepoint skb:kfree_skb.
To keep this stable for compilers kfree_skb_reason() is annotated with
__fix_address (noinline __noclone) as fixed in commit c205cc7534
("net: skb: prevent the split of kfree_skb_reason() by gcc").
The function kfree_skb_list_reason() invoke kfree_skb_reason(), which
cause the __builtin_return_address(0) "location" to report the
unexpected address of kfree_skb_list_reason.
Example output from 'perf script':
kpktgend_0 1337 [000] 81.002597: skb:kfree_skb: skbaddr=0xffff888144824700 protocol=2048 location=kfree_skb_list_reason+0x1e reason: QDISC_DROP
Patch creates an __always_inline __kfree_skb_reason() helper call that
is called from both kfree_skb_list() and kfree_skb_list_reason().
Suggestions for solutions that shares code better are welcome.
As preparation for next patch move __kfree_skb() invocation out of
this helper function.
Reviewed-by: Saeed Mahameed <saeed@kernel.org>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Since commit ce098da149 ("skbuff: Introduce slab_build_skb()")
drivers trying to build skb around slab-backed buffers should
go via slab_build_skb() rather than passing frag_size = 0 to
the main build_skb().
Remove the copy'n'pasted comments about 0 meaning slab.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extending the tail can have some unexpected side effects if a program uses
a helper like BPF_FUNC_skb_pull_data to read partial content beyond the
head skb headlen when all the skbs in the gso frag_list are linear with no
head_frag -
kernel BUG at net/core/skbuff.c:4219!
pc : skb_segment+0xcf4/0xd2c
lr : skb_segment+0x63c/0xd2c
Call trace:
skb_segment+0xcf4/0xd2c
__udp_gso_segment+0xa4/0x544
udp4_ufo_fragment+0x184/0x1c0
inet_gso_segment+0x16c/0x3a4
skb_mac_gso_segment+0xd4/0x1b0
__skb_gso_segment+0xcc/0x12c
udp_rcv_segment+0x54/0x16c
udp_queue_rcv_skb+0x78/0x144
udp_unicast_rcv_skb+0x8c/0xa4
__udp4_lib_rcv+0x490/0x68c
udp_rcv+0x20/0x30
ip_protocol_deliver_rcu+0x1b0/0x33c
ip_local_deliver+0xd8/0x1f0
ip_rcv+0x98/0x1a4
deliver_ptype_list_skb+0x98/0x1ec
__netif_receive_skb_core+0x978/0xc60
Fix this by marking these skbs as GSO_DODGY so segmentation can handle
the tail updates accordingly.
Fixes: 3dcbdb134f ("net: gso: Fix skb_segment splat when splitting gso_size mangled skb having linear-headed frag_list")
Signed-off-by: Sean Tranchetti <quic_stranche@quicinc.com>
Signed-off-by: Subash Abhinov Kasiviswanathan <quic_subashab@quicinc.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://lore.kernel.org/r/1671084718-24796-1-git-send-email-quic_subashab@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
syzkaller reported:
BUG: KASAN: slab-out-of-bounds in __build_skb_around+0x235/0x340 net/core/skbuff.c:294
Write of size 32 at addr ffff88802aa172c0 by task syz-executor413/5295
For bpf_prog_test_run_skb(), which uses a kmalloc()ed buffer passed to
build_skb().
When build_skb() is passed a frag_size of 0, it means the buffer came
from kmalloc. In these cases, ksize() is used to find its actual size,
but since the allocation may not have been made to that size, actually
perform the krealloc() call so that all the associated buffer size
checking will be correctly notified (and use the "new" pointer so that
compiler hinting works correctly). Split this logic out into a new
interface, slab_build_skb(), but leave the original 0 checking for now
to catch any stragglers.
Reported-by: syzbot+fda18eaa8c12534ccb3b@syzkaller.appspotmail.com
Link: https://groups.google.com/g/syzkaller-bugs/c/UnIKxTtU5-0/m/-wbXinkgAQAJ
Fixes: 38931d8989 ("mm: Make ksize() a reporting-only function")
Cc: Pavel Begunkov <asml.silence@gmail.com>
Cc: pepsipu <soopthegoop@gmail.com>
Cc: syzbot+fda18eaa8c12534ccb3b@syzkaller.appspotmail.com
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: kasan-dev <kasan-dev@googlegroups.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: ast@kernel.org
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Hao Luo <haoluo@google.com>
Cc: Jesper Dangaard Brouer <hawk@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: jolsa@kernel.org
Cc: KP Singh <kpsingh@kernel.org>
Cc: martin.lau@linux.dev
Cc: Stanislav Fomichev <sdf@google.com>
Cc: song@kernel.org
Cc: Yonghong Song <yhs@fb.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20221208060256.give.994-kees@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Adopt atomic_long_try_cmpxchg() in mm_account_pinned_pages()
as it is slightly more efficient.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
-----BEGIN PGP SIGNATURE-----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=q04p
-----END PGP SIGNATURE-----
Merge tag 'rxrpc-next-20221108' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
rxrpc changes
David Howells says:
====================
rxrpc: Increasing SACK size and moving away from softirq, part 1
AF_RXRPC has some issues that need addressing:
(1) The SACK table has a maximum capacity of 255, but for modern networks
that isn't sufficient. This is hard to increase in the upstream code
because of the way the application thread is coupled to the softirq
and retransmission side through a ring buffer. Adjustments to the rx
protocol allows a capacity of up to 8192, and having a ring
sufficiently large to accommodate that would use an excessive amount
of memory as this is per-call.
(2) Processing ACKs in softirq mode causes the ACKs get conflated, with
only the most recent being considered. Whilst this has the upside
that the retransmission algorithm only needs to deal with the most
recent ACK, it causes DATA transmission for a call to be very bursty
because DATA packets cannot be transmitted in softirq mode. Rather
transmission must be delegated to either the application thread or a
workqueue, so there tend to be sudden bursts of traffic for any
particular call due to scheduling delays.
(3) All crypto in a single call is done in series; however, each DATA
packet is individually encrypted so encryption and decryption of large
calls could be parallelised if spare CPU resources are available.
This is the first of a number of sets of patches that try and address them.
The overall aims of these changes include:
(1) To get rid of the TxRx ring and instead pass the packets round in
queues (eg. sk_buff_head). On the Tx side, each ACK packet comes with
a SACK table that can be parsed as-is, so there's no particular need
to maintain our own; we just have to refer to the ACK.
On the Rx side, we do need to maintain a SACK table with one bit per
entry - but only if packets go missing - and we don't want to have to
perform a complex transformation to get the information into an ACK
packet.
(2) To try and move almost all processing of received packets out of the
softirq handler and into a high-priority kernel I/O thread. Only the
transferral of packets would be left there. I would still use the
encap_rcv hook to receive packets as there's a noticeable performance
drop from letting the UDP socket put the packets into its own queue
and then getting them out of there.
(3) To make the I/O thread also do all the transmission. The app thread
would be responsible for packaging the data into packets and then
buffering them for the I/O thread to transmit. This would make it
easier for the app thread to run ahead of the I/O thread, and would
mean the I/O thread is less likely to have to wait around for a new
packet to come available for transmission.
(4) To logically partition the socket/UAPI/KAPI side of things from the
I/O side of things. The local endpoint, connection, peer and call
objects would belong to the I/O side. The socket side would not then
touch the private internals of calls and suchlike and would not change
their states. It would only look at the send queue, receive queue and
a way to pass a message to cause an abort.
(5) To remove as much locking, synchronisation, barriering and atomic ops
as possible from the I/O side. Exclusion would be achieved by
limiting modification of state to the I/O thread only. Locks would
still need to be used in communication with the UDP socket and the
AF_RXRPC socket API.
(6) To provide crypto offload kernel threads that, when there's slack in
the system, can see packets that need crypting and provide
parallelisation in dealing with them.
(7) To remove the use of system timers. Since each timer would then send
a poke to the I/O thread, which would then deal with it when it had
the opportunity, there seems no point in using system timers if,
instead, a list of timeouts can be sensibly consulted. An I/O thread
only then needs to schedule with a timeout when it is idle.
(8) To use zero-copy sendmsg to send packets. This would make use of the
I/O thread being the sole transmitter on the socket to manage the
dead-reckoning sequencing of the completion notifications. There is a
problem with zero-copy, though: the UDP socket doesn't handle running
out of option memory very gracefully.
With regard to this first patchset, the changes made include:
(1) Some fixes, including a fallback for proc_create_net_single_write(),
setting ack.bufferSize to 0 in ACK packets and a fix for rxrpc
congestion management, which shouldn't be saving the cwnd value
between calls.
(2) Improvements in rxrpc tracepoints, including splitting the timer
tracepoint into a set-timer and a timer-expired trace.
(3) Addition of a new proc file to display some stats.
(4) Some code cleanups, including removing some unused bits and
unnecessary header inclusions.
(5) A change to the recently added UDP encap_err_rcv hook so that it has
the same signature as {ip,ipv6}_icmp_error(), and then just have rxrpc
point its UDP socket's hook directly at those.
(6) Definition of a new struct, rxrpc_txbuf, that is used to hold
transmissible packets of DATA and ACK type in a single 2KiB block
rather than using an sk_buff. This allows the buffer to be on a
number of queues simultaneously more easily, and also guarantees that
the entire block is in a single unit for zerocopy purposes and that
the data payload is aligned for in-place crypto purposes.
(7) ACK txbufs are allocated at proposal and queued for later transmission
rather than being stored in a single place in the rxrpc_call struct,
which means only a single ACK can be pending transmission at a time.
The queue is then drained at various points. This allows the ACK
generation code to be simplified.
(8) The Rx ring buffer is removed. When a jumbo packet is received (which
comprises a number of ordinary DATA packets glued together), it used
to be pointed to by the ring multiple times, with an annotation in a
side ring indicating which subpacket was in that slot - but this is no
longer possible. Instead, the packet is cloned once for each
subpacket, barring the last, and the range of data is set in the skb
private area. This makes it easier for the subpackets in a jumbo
packet to be decrypted in parallel.
(9) The Tx ring buffer is removed. The side annotation ring that held the
SACK information is also removed. Instead, in the event of packet
loss, the SACK data attached an ACK packet is parsed.
(10) Allocate an skcipher request when needed in the rxkad security class
rather than caching one in the rxrpc_call struct. This deals with a
race between externally-driven call disconnection getting rid of the
skcipher request and sendmsg/recvmsg trying to use it because they
haven't seen the completion yet. This is also needed to support
parallelisation as the skcipher request cannot be used by two or more
threads simultaneously.
(11) Call udp_sendmsg() and udpv6_sendmsg() directly rather than going
through kernel_sendmsg() so that we can provide our own iterator
(zerocopy explicitly doesn't work with a KVEC iterator). This also
lets us avoid the overhead of the security hook.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Improve the tracking of which packets need to be transmitted by saving the
last ACK packet that we receive that has a populated soft-ACK table rather
than marking packets. Then we can step through the soft-ACK table and look
at the packets we've transmitted beyond that to determine which packets we
might want to retransmit.
We also look at the highest serial number that has been acked to try and
guess which packets we've transmitted the peer is likely to have seen. If
necessary, we send a ping to retrieve that number.
One downside that might be a problem is that we can't then compare the
previous acked/unacked state so easily in rxrpc_input_soft_acks() - which
is a potential problem for the slow-start algorithm.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
