Currently we may merge incorrectly a received GSO packet
or a packet with frag_list into a packet sitting in the
gro_hash list. skb_segment() may crash case because
the assumptions on the skb layout are not met.
The correct behaviour would be to flush the packet in the
gro_hash list and send the received GSO packet directly
afterwards. Commit d61d072e87 ("net-gro: avoid reorders")
sets NAPI_GRO_CB(skb)->flush in this case, but this is not
checked before merging. This patch makes sure to check this
flag and to not merge in that case.
Fixes: d61d072e87 ("net-gro: avoid reorders")
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of declaring a function in a .c file, declare it in a header
file and include that header file from the source files that define
and that use the function. That allows the compiler to verify
consistency of declaration and definition. See also commit
52267790ef ("sock: add MSG_ZEROCOPY") # v4.14.
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch addresses the fact that there are drivers, specifically tun,
that will call into the network page fragment allocators with buffer sizes
that are not cache aligned. Doing this could result in data alignment
and DMA performance issues as these fragment pools are also shared with the
skb allocator and any other devices that will use napi_alloc_frags or
netdev_alloc_frags.
Fixes: ffde7328a3 ("net: Split netdev_alloc_frag into __alloc_page_frag and add __napi_alloc_frag")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit dcda9b0471 ("mm, tree wide: replace __GFP_REPEAT by
__GFP_RETRY_MAYFAIL with more useful semantic") replaced __GFP_REPEAT in
alloc_skb_with_frags() with __GFP_RETRY_MAYFAIL when the allocation may
directly reclaim.
The previous behavior would require reclaim up to 1 << order pages for
skb aligned header_len of order > PAGE_ALLOC_COSTLY_ORDER before failing,
otherwise the allocations in alloc_skb() would loop in the page allocator
looking for memory. __GFP_RETRY_MAYFAIL makes both allocations failable
under memory pressure, including for the HEAD allocation.
This can cause, among many other things, write() to fail with ENOTCONN
during RPC when under memory pressure.
These allocations should succeed as they did previous to dcda9b0471
even if it requires calling the oom killer and additional looping in the
page allocator to find memory. There is no way to specify the previous
behavior of __GFP_REPEAT, but it's unlikely to be necessary since the
previous behavior only guaranteed that 1 << order pages would be reclaimed
before failing for order > PAGE_ALLOC_COSTLY_ORDER. That reclaim is not
guaranteed to be contiguous memory, so repeating for such large orders is
usually not beneficial.
Removing the setting of __GFP_RETRY_MAYFAIL to restore the previous
behavior, specifically not allowing alloc_skb() to fail for small orders
and oom kill if necessary rather than allowing RPCs to fail.
Fixes: dcda9b0471 ("mm, tree wide: replace __GFP_REPEAT by __GFP_RETRY_MAYFAIL with more useful semantic")
Signed-off-by: David Rientjes <rientjes@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the extension to be added is already present, the only
skb field we may need to update is 'extensions': we can reorder
the code and avoid a branch.
v1 -> v2:
- be sure to flag the newly added extension as active
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
On cow we can free the old extension: we must avoid dereferencing
such extension after skb_ext_maybe_cow(). Since 'new' contents
are always equal to 'old' after the copy, we can fix the above
accessing the relevant data using 'new'.
Fixes: df5042f4c5 ("sk_buff: add skb extension infrastructure")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove skb->sp and allocate secpath storage via extension
infrastructure. This also reduces sk_buff by 8 bytes on x86_64.
Total size of allyesconfig kernel is reduced slightly, as there is
less inlined code (one conditional atomic op instead of two on
skb_clone).
No differences in throughput in following ipsec performance tests:
- transport mode with aes on 10GB link
- tunnel mode between two network namespaces with aes and null cipher
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This converts the bridge netfilter (calling iptables hooks from bridge)
facility to use the extension infrastructure.
The bridge_nf specific hooks in skb clone and free paths are removed, they
have been replaced by the skb_ext hooks that do the same as the bridge nf
allocations hooks did.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds an optional extension infrastructure, with ispec (xfrm) and
bridge netfilter as first users.
objdiff shows no changes if kernel is built without xfrm and br_netfilter
support.
The third (planned future) user is Multipath TCP which is still
out-of-tree.
MPTCP needs to map logical mptcp sequence numbers to the tcp sequence
numbers used by individual subflows.
This DSS mapping is read/written from tcp option space on receive and
written to tcp option space on transmitted tcp packets that are part of
and MPTCP connection.
Extending skb_shared_info or adding a private data field to skb fclones
doesn't work for incoming skb, so a different DSS propagation method would
be required for the receive side.
mptcp has same requirements as secpath/bridge netfilter:
1. extension memory is released when the sk_buff is free'd.
2. data is shared after cloning an skb (clone inherits extension)
3. adding extension to an skb will COW the extension buffer if needed.
The "MPTCP upstreaming" effort adds SKB_EXT_MPTCP extension to store the
mapping for tx and rx processing.
Two new members are added to sk_buff:
1. 'active_extensions' byte (filling a hole), telling which extensions
are available for this skb.
This has two purposes.
a) avoids the need to initialize the pointer.
b) allows to "delete" an extension by clearing its bit
value in ->active_extensions.
While it would be possible to store the active_extensions byte
in the extension struct instead of sk_buff, there is one problem
with this:
When an extension has to be disabled, we can always clear the
bit in skb->active_extensions. But in case it would be stored in the
extension buffer itself, we might have to COW it first, if
we are dealing with a cloned skb. On kmalloc failure we would
be unable to turn an extension off.
2. extension pointer, located at the end of the sk_buff.
If the active_extensions byte is 0, the pointer is undefined,
it is not initialized on skb allocation.
This adds extra code to skb clone and free paths (to deal with
refcount/free of extension area) but this replaces similar code that
manages skb->nf_bridge and skb->sp structs in the followup patches of
the series.
It is possible to add support for extensions that are not preseved on
clones/copies.
To do this, it would be needed to define a bitmask of all extensions that
need copy/cow semantics, and change __skb_ext_copy() to check
->active_extensions & SKB_EXT_PRESERVE_ON_CLONE, then just set
->active_extensions to 0 on the new clone.
This isn't done here because all extensions that get added here
need the copy/cow semantics.
v2:
Allocate entire extension space using kmem_cache.
Upside is that this allows better tracking of used memory,
downside is that we will allocate more space than strictly needed in
most cases (its unlikely that all extensions are active/needed at same
time for same skb).
The allocated memory (except the small extension header) is not cleared,
so no additonal overhead aside from memory usage.
Avoid atomic_dec_and_test operation on skb_ext_put()
by using similar trick as kfree_skbmem() does with fclone_ref:
If recount is 1, there is no concurrent user and we can free right away.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit abf4bb6b63 ("skbuff: Add the offload_mr_fwd_mark field") added
the 'offload_mr_fwd_mark' field to indicate that a packet has already
undergone L3 multicast routing by a capable device. The field is used to
prevent the kernel from forwarding a packet through a netdev through
which the device has already forwarded the packet.
Currently, no unicast packet is routed by both the device and the
kernel, but this is about to change by subsequent patches and we need to
be able to mark such packets, so that they will no be forwarded twice.
Instead of adding yet another field to 'struct sk_buff', we can just
rename 'offload_mr_fwd_mark' to 'offload_l3_fwd_mark', as a packet
either has a multicast or a unicast destination IP.
While at it, add a comment about both 'offload_fwd_mark' and
'offload_l3_fwd_mark'.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With MSG_ZEROCOPY, each skb holds a reference to a struct ubuf_info.
Release of its last reference triggers a completion notification.
The TCP stack in tcp_sendmsg_locked holds an extra ref independent of
the skbs, because it can build, send and free skbs within its loop,
possibly reaching refcount zero and freeing the ubuf_info too soon.
The UDP stack currently also takes this extra ref, but does not need
it as all skbs are sent after return from __ip(6)_append_data.
Avoid the extra refcount_inc and refcount_dec_and_test, and generally
the sock_zerocopy_put in the common path, by passing the initial
reference to the first skb.
This approach is taken instead of initializing the refcount to 0, as
that would generate error "refcount_t: increment on 0" on the
next skb_zcopy_set.
Changes
v3 -> v4
- Move skb_zcopy_set below the only kfree_skb that might cause
a premature uarg destroy before skb_zerocopy_put_abort
- Move the entire skb_shinfo assignment block, to keep that
cacheline access in one place
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend zerocopy to udp sockets. Allow setting sockopt SO_ZEROCOPY and
interpret flag MSG_ZEROCOPY.
This patch was previously part of the zerocopy RFC patchsets. Zerocopy
is not effective at small MTU. With segmentation offload building
larger datagrams, the benefit of page flipping outweights the cost of
generating a completion notification.
tools/testing/selftests/net/msg_zerocopy.sh after applying follow-on
test patch and making skb_orphan_frags_rx same as skb_orphan_frags:
ipv4 udp -t 1
tx=191312 (11938 MB) txc=0 zc=n
rx=191312 (11938 MB)
ipv4 udp -z -t 1
tx=304507 (19002 MB) txc=304507 zc=y
rx=304507 (19002 MB)
ok
ipv6 udp -t 1
tx=174485 (10888 MB) txc=0 zc=n
rx=174485 (10888 MB)
ipv6 udp -z -t 1
tx=294801 (18396 MB) txc=294801 zc=y
rx=294801 (18396 MB)
ok
Changes
v1 -> v2
- Fixup reverse christmas tree violation
v2 -> v3
- Split refcount avoidance optimization into separate patch
- Fix refcount leak on error in fragmented case
(thanks to Paolo Abeni for pointing this one out!)
- Fix refcount inc on zero
- Test sock_flag SOCK_ZEROCOPY directly in __ip_append_data.
This is needed since commit 5cf4a8532c ("tcp: really ignore
MSG_ZEROCOPY if no SO_ZEROCOPY") did the same for tcp.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I do not see how one can effectively use skb_insert() without holding
some kind of lock. Otherwise other cpus could have changed the list
right before we have a chance of acquiring list->lock.
Only existing user is in drivers/infiniband/hw/nes/nes_mgt.c and this
one probably meant to use __skb_insert() since it appears nesqp->pau_list
is protected by nesqp->pau_lock. This looks like nesqp->pau_lock
could be removed, since nesqp->pau_list.lock could be used instead.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Faisal Latif <faisal.latif@intel.com>
Cc: Doug Ledford <dledford@redhat.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: linux-rdma <linux-rdma@vger.kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a packet is trapped and the corresponding SKB marked as
already-forwarded, it retains this marking even after it is forwarded
across veth links into another bridge. There, since it ingresses the
bridge over veth, which doesn't have offload_fwd_mark, it triggers a
warning in nbp_switchdev_frame_mark().
Then nbp_switchdev_allowed_egress() decides not to allow egress from
this bridge through another veth, because the SKB is already marked, and
the mark (of 0) of course matches. Thus the packet is incorrectly
blocked.
Solve by resetting offload_fwd_mark() in skb_scrub_packet(). That
function is called from tunnels and also from veth, and thus catches the
cases where traffic is forwarded between bridges and transformed in a
way that invalidates the marking.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Suggested-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
if list is NULL pointer, and the following access of list
will trigger panic, which is same as BUG_ON
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently netdev_rx_csum_fault() only shows a device name,
we need more information about the skb for debugging csum
failures.
Sample output:
ens3: hw csum failure
dev features: 0x0000000000014b89
skb len=84 data_len=0 pkt_type=0 gso_size=0 gso_type=0 nr_frags=0 ip_summed=0 csum=0 csum_complete_sw=0 csum_valid=0 csum_level=0
Note, I use pr_err() just to be consistent with the existing one.
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This removes assumptions about VLAN_TAG_PRESENT bit.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
__skb_checksum_complete_head() and __skb_checksum_complete()
are both declared in skbuff.h, they fit better in skbuff.c
than datagram.c.
Cc: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove kernel-doc warning:
net/core/skbuff.c:4953: warning: Function parameter or member 'skb' not described in 'skb_gso_size_check'
Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern's dump indexing bug fix in 'net' overlapped the
change of the function signature of inet6_fill_ifaddr() in
'net-next'. Trivially resolved.
Signed-off-by: David S. Miller <davem@davemloft.net>
We've been getting checksum errors involving small UDP packets, usually
59B packets with 1 extra non-zero padding byte. netdev_rx_csum_fault()
has been complaining that HW is providing bad checksums. Turns out the
problem is in pskb_trim_rcsum_slow(), introduced in commit 88078d98d1
("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends").
The source of the problem is that when the bytes we are trimming start
at an odd address, as in the case of the 1 padding byte above,
skb_checksum() returns a byte-swapped value. We cannot just combine this
with skb->csum using csum_sub(). We need to use csum_block_sub() here
that takes into account the parity of the start address and handles the
swapping.
Matches existing code in __skb_postpull_rcsum() and esp_remove_trailer().
Fixes: 88078d98d1 ("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends")
Signed-off-by: Dimitris Michailidis <dmichail@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts were easy to resolve using immediate context mostly,
except the cls_u32.c one where I simply too the entire HEAD
chunk.
Signed-off-by: David S. Miller <davem@davemloft.net>
This helper is unused since commit 988cf74deb ("inet:
Stop generating UFO packets.")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
According to the documentation in msg_zerocopy.rst, the SO_ZEROCOPY
flag was introduced because send(2) ignores unknown message flags and
any legacy application which was accidentally passing the equivalent of
MSG_ZEROCOPY earlier should not see any new behaviour.
Before commit f214f915e7 ("tcp: enable MSG_ZEROCOPY"), a send(2) call
which passed the equivalent of MSG_ZEROCOPY without setting SO_ZEROCOPY
would succeed. However, after that commit, it fails with -ENOBUFS. So
it appears that the SO_ZEROCOPY flag fails to fulfill its intended
purpose. Fix it.
Fixes: f214f915e7 ("tcp: enable MSG_ZEROCOPY")
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is needed for veth XDP which does skb_copy_expand()-like operation.
v2:
- Drop skb_copy_header part because it has already been exported now.
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested: see the next patch is the series.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 784abe24c9 ("net: Add decrypted field to skb")
introduced a 'decrypted' field that is explicitly copied on skb
copy and clone.
Move it between headers_start[0] and headers_end[0], so that we
don't need to copy it explicitly as it's copied by the memcpy()
in __copy_skb_header().
While at it, drop the assignment in __skb_clone(), it was
already redundant.
This doesn't change the size of sk_buff or cacheline boundaries.
The 15-bits hole before tc_index becomes a 14-bits hole, and
will be again a 15-bits hole when this change is merged with
commit 8b7008620b ("net: Don't copy pfmemalloc flag in
__copy_skb_header()").
v2: as reported by kbuild test robot (oops, I forgot to build
with CONFIG_TLS_DEVICE it seems), we can't use
CHECK_SKB_FIELD() on a bit-field member. Just drop the
check for the moment being, perhaps we could think of some
magic to also check bit-field members one day.
Fixes: 784abe24c9 ("net: Add decrypted field to skb")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The decrypted bit is propogated to cloned/copied skbs.
This will be used later by the inline crypto receive side offload
of tls.
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 8b7008620b ("net: Don't copy pfmemalloc flag in
__copy_skb_header()") introduced a different handling for the
pfmemalloc flag in copy and clone paths.
In __skb_clone(), now, the flag is set only if it was set in the
original skb, but not cleared if it wasn't. This is wrong and
might lead to socket buffers being flagged with pfmemalloc even
if the skb data wasn't allocated from pfmemalloc reserves. Copy
the flag instead of ORing it.
Reported-by: Sabrina Dubroca <sd@queasysnail.net>
Fixes: 8b7008620b ("net: Don't copy pfmemalloc flag in __copy_skb_header()")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Tested-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The pfmemalloc flag indicates that the skb was allocated from
the PFMEMALLOC reserves, and the flag is currently copied on skb
copy and clone.
However, an skb copied from an skb flagged with pfmemalloc
wasn't necessarily allocated from PFMEMALLOC reserves, and on
the other hand an skb allocated that way might be copied from an
skb that wasn't.
So we should not copy the flag on skb copy, and rather decide
whether to allow an skb to be associated with sockets unrelated
to page reclaim depending only on how it was allocated.
Move the pfmemalloc flag before headers_start[0] using an
existing 1-bit hole, so that __copy_skb_header() doesn't copy
it.
When cloning, we'll now take care of this flag explicitly,
contravening to the warning comment of __skb_clone().
While at it, restore the newline usage introduced by commit
b193722731 ("net: reorganize sk_buff for faster
__copy_skb_header()") to visually separate bytes used in
bitfields after headers_start[0], that was gone after commit
a9e419dc7b ("netfilter: merge ctinfo into nfct pointer storage
area"), and describe the pfmemalloc flag in the kernel-doc
structure comment.
This doesn't change the size of sk_buff or cacheline boundaries,
but consolidates the 15 bits hole before tc_index into a 2 bytes
hole before csum, that could now be filled more easily.
Reported-by: Patrick Talbert <ptalbert@redhat.com>
Fixes: c93bdd0e03 ("netvm: allow skb allocation to use PFMEMALLOC reserves")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is done in preparation for the upcoming time based transmission
patchset. Now that skb->tstamp will be used to hold packet's txtime,
we must ensure that it is being cleared when traversing namespaces.
Also, doing that from skb_scrub_packet() before the early return would
break our feature when tunnels are used.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simple overlapping changes in stmmac driver.
Adjust skb_gro_flush_final_remcsum function signature to make GRO list
changes in net-next, as per Stephen Rothwell's example merge
resolution.
Signed-off-by: David S. Miller <davem@davemloft.net>
alloc_skb_with_frags uses __GFP_NORETRY for non-sleeping allocations
which is just a noop and a little bit confusing.
__GFP_NORETRY was added by ed98df3361 ("net: use __GFP_NORETRY for
high order allocations") to prevent from the OOM killer. Yet this was
not enough because fb05e7a89f ("net: don't wait for order-3 page
allocation") didn't want an excessive reclaim for non-costly orders
so it made it completely NOWAIT while it preserved __GFP_NORETRY in
place which is now redundant.
Drop the pointless __GFP_NORETRY because this function is used as
copy&paste source for other places.
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sock reference is lost when scrubbing the packet and that breaks
TSQ (TCP Small Queues) and XPS (Transmit Packet Steering) causing
performance impacts of about 50% in a single TCP stream when crossing
network namespaces.
XPS breaks because the queue mapping stored in the socket is not
available, so another random queue might be selected when the stack
needs to transmit something like a TCP ACK, or TCP Retransmissions.
That causes packet re-ordering and/or performance issues.
TSQ breaks because it orphans the packet while it is still in the
host, so packets are queued contributing to the buffer bloat problem.
Preserving the sock reference fixes both issues. The socket is
orphaned anyways in the receiving path before any relevant action
and on TX side the netfilter checks if the reference is local before
use it.
Signed-off-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Manage pending per-NAPI GRO packets via list_head.
Return an SKB pointer from the GRO receive handlers. When GRO receive
handlers return non-NULL, it means that this SKB needs to be completed
at this time and removed from the NAPI queue.
Several operations are greatly simplified by this transformation,
especially timing out the oldest SKB in the list when gro_count
exceeds MAX_GRO_SKBS, and napi_gro_flush() which walks the queue
in reverse order.
Signed-off-by: David S. Miller <davem@davemloft.net>
copy_skb_header is renamed to skb_copy_header and
exported. Exposing this function give more flexibility
in copying SKBs.
skb_copy and skb_copy_expand do not give enough control
over which parts are copied.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement generic segmentation offload support for udp datagrams. A
follow-up patch adds support to the protocol stack to generate such
packets.
UDP GSO is not UFO. UFO fragments a single large datagram. GSO splits
a large payload into a number of discrete UDP datagrams.
The implementation adds a GSO type SKB_UDP_GSO_L4 to differentiate it
from UFO (SKB_UDP_GSO).
IPPROTO_UDPLITE is excluded, as that protocol has no gso handler
registered.
[ Export __udp_gso_segment for ipv6. -DaveM ]
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After working on IP defragmentation lately, I found that some large
packets defeat CHECKSUM_COMPLETE optimization because of NIC adding
zero paddings on the last (small) fragment.
While removing the padding with pskb_trim_rcsum(), we set skb->ip_summed
to CHECKSUM_NONE, forcing a full csum validation, even if all prior
fragments had CHECKSUM_COMPLETE set.
We can instead compute the checksum of the part we are trimming,
usually smaller than the part we keep.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzbot reported __skb_try_recv_from_queue() was using skb->peeked
while it was potentially unitialized.
We need to clear it in __skb_clone()
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Minor conflicts in drivers/net/ethernet/mellanox/mlx5/core/en_rep.c,
we had some overlapping changes:
1) In 'net' MLX5E_PARAMS_LOG_{SQ,RQ}_SIZE -->
MLX5E_REP_PARAMS_LOG_{SQ,RQ}_SIZE
2) In 'net-next' params->log_rq_size is renamed to be
params->log_rq_mtu_frames.
3) In 'net-next' params->hard_mtu is added.
Signed-off-by: David S. Miller <davem@davemloft.net>
Fun set of conflict resolutions here...
For the mac80211 stuff, these were fortunately just parallel
adds. Trivially resolved.
In drivers/net/phy/phy.c we had a bug fix in 'net' that moved the
function phy_disable_interrupts() earlier in the file, whilst in
'net-next' the phy_error() call from this function was removed.
In net/ipv4/xfrm4_policy.c, David Ahern's changes to remove the
'rt_table_id' member of rtable collided with a bug fix in 'net' that
added a new struct member "rt_mtu_locked" which needs to be copied
over here.
The mlxsw driver conflict consisted of net-next separating
the span code and definitions into separate files, whilst
a 'net' bug fix made some changes to that moved code.
The mlx5 infiniband conflict resolution was quite non-trivial,
the RDMA tree's merge commit was used as a guide here, and
here are their notes:
====================
Due to bug fixes found by the syzkaller bot and taken into the for-rc
branch after development for the 4.17 merge window had already started
being taken into the for-next branch, there were fairly non-trivial
merge issues that would need to be resolved between the for-rc branch
and the for-next branch. This merge resolves those conflicts and
provides a unified base upon which ongoing development for 4.17 can
be based.
Conflicts:
drivers/infiniband/hw/mlx5/main.c - Commit 42cea83f95
(IB/mlx5: Fix cleanup order on unload) added to for-rc and
commit b5ca15ad7e (IB/mlx5: Add proper representors support)
add as part of the devel cycle both needed to modify the
init/de-init functions used by mlx5. To support the new
representors, the new functions added by the cleanup patch
needed to be made non-static, and the init/de-init list
added by the representors patch needed to be modified to
match the init/de-init list changes made by the cleanup
patch.
Updates:
drivers/infiniband/hw/mlx5/mlx5_ib.h - Update function
prototypes added by representors patch to reflect new function
names as changed by cleanup patch
drivers/infiniband/hw/mlx5/ib_rep.c - Update init/de-init
stage list to match new order from cleanup patch
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When errors are enqueued to the error queue via sock_queue_err_skb()
function, it is possible that the waiting application is not notified.
Calling 'sk->sk_data_ready()' would not notify applications that
selected only POLLERR events in poll() (for example).
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Randy E. Witt <randy.e.witt@intel.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we have a bridge with vlan_filtering on and a vlan device on top of
it, packets would be corrupted in skb_vlan_untag() called from
br_dev_xmit().
The problem sits in skb_reorder_vlan_header() used in skb_vlan_untag(),
which makes use of skb->mac_len. In this function mac_len is meant for
handling rx path with vlan devices with reorder_header disabled, but in
tx path mac_len is typically 0 and cannot be used, which is the problem
in this case.
The current code even does not properly handle rx path (skb_vlan_untag()
called from __netif_receive_skb_core()) with reorder_header off actually.
In rx path single tag case, it works as follows:
- Before skb_reorder_vlan_header()
mac_header data
v v
+-------------------+-------------+------+----
| ETH | VLAN | ETH |
| ADDRS | TPID | TCI | TYPE |
+-------------------+-------------+------+----
<-------- mac_len --------->
<------------->
to be removed
- After skb_reorder_vlan_header()
mac_header data
v v
+-------------------+------+----
| ETH | ETH |
| ADDRS | TYPE |
+-------------------+------+----
<-------- mac_len --------->
This is ok, but in rx double tag case, it corrupts packets:
- Before skb_reorder_vlan_header()
mac_header data
v v
+-------------------+-------------+-------------+------+----
| ETH | VLAN | VLAN | ETH |
| ADDRS | TPID | TCI | TPID | TCI | TYPE |
+-------------------+-------------+-------------+------+----
<--------------- mac_len ---------------->
<------------->
should be removed
<--------------------------->
actually will be removed
- After skb_reorder_vlan_header()
mac_header data
v v
+-------------------+------+----
| ETH | ETH |
| ADDRS | TYPE |
+-------------------+------+----
<--------------- mac_len ---------------->
So, two of vlan tags are both removed while only inner one should be
removed and mac_header (and mac_len) is broken.
skb_vlan_untag() is meant for removing the vlan header at (skb->data - 2),
so use skb->data and skb->mac_header to calculate the right offset.
Reported-by: Brandon Carpenter <brandon.carpenter@cypherpath.com>
Fixes: a6e18ff111 ("vlan: Fix untag operations of stacked vlans with REORDER_HEADER off")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
As well as the basic conversion, I noticed that a lot of the
SCTP code checks gso_type without first checking skb_is_gso()
so I have added that where appropriate.
Also, document the helper.
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
All of the conflicts were cases of overlapping changes.
In net/core/devlink.c, we have to make care that the
resouce size_params have become a struct member rather
than a pointer to such an object.
Signed-off-by: David S. Miller <davem@davemloft.net>
They're very hard to use properly as they do not consider the
GSO_BY_FRAGS case. Code should use skb_gso_validate_network_len
and skb_gso_validate_mac_len as they do consider this case.
Make the seglen functions static, which stops people using them
outside of skbuff.c
Signed-off-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If you take a GSO skb, and split it into packets, will the network
length (L3 headers + L4 headers + payload) of those packets be small
enough to fit within a given MTU?
skb_gso_validate_mtu gives you the answer to that question. However,
we recently added to add a way to validate the MAC length of a split GSO
skb (L2+L3+L4+payload), and the names get confusing, so rename
skb_gso_validate_mtu to skb_gso_validate_network_len
Signed-off-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All kmem caches aren't reallocated once set up.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
RDS would like to use the helper functions for managing pinned pages
added by Commit a91dbff551 ("sock: ulimit on MSG_ZEROCOPY pages")
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Most callers of put_cmsg() use a "sizeof(foo)" for the length argument.
Within put_cmsg(), a copy_to_user() call is made with a dynamic size, as a
result of the cmsg header calculations. This means that hardened usercopy
will examine the copy, even though it was technically a fixed size and
should be implicitly whitelisted. All the put_cmsg() calls being built
from values in skbuff_head_cache are coming out of the protocol-defined
"cb" field, so whitelist this field entirely instead of creating per-use
bounce buffers, for which there are concerns about performance.
Original report was:
Bad or missing usercopy whitelist? Kernel memory exposure attempt detected from SLAB object 'skbuff_head_cache' (offset 64, size 16)!
WARNING: CPU: 0 PID: 3663 at mm/usercopy.c:81 usercopy_warn+0xdb/0x100 mm/usercopy.c:76
...
__check_heap_object+0x89/0xc0 mm/slab.c:4426
check_heap_object mm/usercopy.c:236 [inline]
__check_object_size+0x272/0x530 mm/usercopy.c:259
check_object_size include/linux/thread_info.h:112 [inline]
check_copy_size include/linux/thread_info.h:143 [inline]
copy_to_user include/linux/uaccess.h:154 [inline]
put_cmsg+0x233/0x3f0 net/core/scm.c:242
sock_recv_errqueue+0x200/0x3e0 net/core/sock.c:2913
packet_recvmsg+0xb2e/0x17a0 net/packet/af_packet.c:3296
sock_recvmsg_nosec net/socket.c:803 [inline]
sock_recvmsg+0xc9/0x110 net/socket.c:810
___sys_recvmsg+0x2a4/0x640 net/socket.c:2179
__sys_recvmmsg+0x2a9/0xaf0 net/socket.c:2287
SYSC_recvmmsg net/socket.c:2368 [inline]
SyS_recvmmsg+0xc4/0x160 net/socket.c:2352
entry_SYSCALL_64_fastpath+0x29/0xa0
Reported-by: syzbot+e2d6cfb305e9f3911dea@syzkaller.appspotmail.com
Fixes: 6d07d1cd30 ("usercopy: Restrict non-usercopy caches to size 0")
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If you take a GSO skb, and split it into packets, will the MAC
length (L2 + L3 + L4 headers + payload) of those packets be small
enough to fit within a given length?
Move skb_gso_mac_seglen() to skbuff.h with other related functions
like skb_gso_network_seglen() so we can use it, and then create
skb_gso_validate_mac_len to do the full calculation.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/ip6_gre.c is a case of parallel adds.
include/trace/events/tcp.h is a little bit more tricky. The removal
of in-trace-macro ifdefs in 'net' paralleled with moving
show_tcp_state_name and friends over to include/trace/events/sock.h
in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_copy_ubufs must unclone before it is safe to modify its
skb_shared_info with skb_zcopy_clear.
Commit b90ddd5687 ("skbuff: skb_copy_ubufs must release uarg even
without user frags") ensures that all skbs release their zerocopy
state, even those without frags.
But I forgot an edge case where such an skb arrives that is cloned.
The stack does not build such packets. Vhost/tun skbs have their
frags orphaned before cloning. TCP skbs only attach zerocopy state
when a frag is added.
But if TCP packets can be trimmed or linearized, this might occur.
Tracing the code I found no instance so far (e.g., skb_linearize
ends up calling skb_zcopy_clear if !skb->data_len).
Still, it is non-obvious that no path exists. And it is fragile to
rely on this.
Fixes: b90ddd5687 ("skbuff: skb_copy_ubufs must release uarg even without user frags")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a net-next follow-up to commit 268b790679 ("skbuff: orphan
frags before zerocopy clone"), which fixed a bug in net, but added a
call to skb_zerocopy_clone at each frag to do so.
When segmenting skbs with user frags, either the user frags must be
replaced with private copies and uarg released, or the uarg must have
its refcount increased for each new skb.
skb_orphan_frags does the first, except for cases that can handle
reference counting. skb_zerocopy_clone then does the second.
Call these once per nskb, instead of once per frag.
That is, in the common case. With a frag list, also refresh when the
origin skb (frag_skb) changes.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_copy_ubufs creates a private copy of frags[] to release its hold
on user frags, then calls uarg->callback to notify the owner.
Call uarg->callback even when no frags exist. This edge case can
happen when zerocopy_sg_from_iter finds enough room in skb_headlen
to copy all the data.
Fixes: 3ece782693 ("sock: skb_copy_ubufs support for compound pages")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Call skb_zerocopy_clone after skb_orphan_frags, to avoid duplicate
calls to skb_uarg(skb)->callback for the same data.
skb_zerocopy_clone associates skb_shinfo(skb)->uarg from frag_skb
with each segment. This is only safe for uargs that do refcounting,
which is those that pass skb_orphan_frags without dropping their
shared frags. For others, skb_orphan_frags drops the user frags and
sets the uarg to NULL, after which sock_zerocopy_clone has no effect.
Qemu hangs were reported due to duplicate vhost_net_zerocopy_callback
calls for the same data causing the vhost_net_ubuf_ref_>refcount to
drop below zero.
Link: http://lkml.kernel.org/r/<CAF=yD-LWyCD4Y0aJ9O0e_CHLR+3JOeKicRRTEVCPxgw4XOcqGQ@mail.gmail.com>
Fixes: 1f8b977ab3 ("sock: enable MSG_ZEROCOPY")
Reported-by: Andreas Hartmann <andihartmann@01019freenet.de>
Reported-by: David Hill <dhill@redhat.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_complete_tx_timestamp must ingest the skb it is passed. Call
kfree_skb if the skb cannot be enqueued.
Fixes: b245be1f4d ("net-timestamp: no-payload only sysctl")
Fixes: 9ac25fc063 ("net: fix socket refcounting in skb_complete_tx_timestamp()")
Reported-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Merge updates from Andrew Morton:
- a few misc bits
- ocfs2 updates
- almost all of MM
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (131 commits)
memory hotplug: fix comments when adding section
mm: make alloc_node_mem_map a void call if we don't have CONFIG_FLAT_NODE_MEM_MAP
mm: simplify nodemask printing
mm,oom_reaper: remove pointless kthread_run() error check
mm/page_ext.c: check if page_ext is not prepared
writeback: remove unused function parameter
mm: do not rely on preempt_count in print_vma_addr
mm, sparse: do not swamp log with huge vmemmap allocation failures
mm/hmm: remove redundant variable align_end
mm/list_lru.c: mark expected switch fall-through
mm/shmem.c: mark expected switch fall-through
mm/page_alloc.c: broken deferred calculation
mm: don't warn about allocations which stall for too long
fs: fuse: account fuse_inode slab memory as reclaimable
mm, page_alloc: fix potential false positive in __zone_watermark_ok
mm: mlock: remove lru_add_drain_all()
mm, sysctl: make NUMA stats configurable
shmem: convert shmem_init_inodecache() to void
Unify migrate_pages and move_pages access checks
mm, pagevec: rename pagevec drained field
...
As the page free path makes no distinction between cache hot and cold
pages, there is no real useful ordering of pages in the free list that
allocation requests can take advantage of. Juding from the users of
__GFP_COLD, it is likely that a number of them are the result of copying
other sites instead of actually measuring the impact. Remove the
__GFP_COLD parameter which simplifies a number of paths in the page
allocator.
This is potentially controversial but bear in mind that the size of the
per-cpu pagelists versus modern cache sizes means that the whole per-cpu
list can often fit in the L3 cache. Hence, there is only a potential
benefit for microbenchmarks that alloc/free pages in a tight loop. It's
even worse when THP is taken into account which has little or no chance
of getting a cache-hot page as the per-cpu list is bypassed and the
zeroing of multiple pages will thrash the cache anyway.
The truncate microbenchmarks are not shown as this patch affects the
allocation path and not the free path. A page fault microbenchmark was
tested but it showed no sigificant difference which is not surprising
given that the __GFP_COLD branches are a miniscule percentage of the
fault path.
Link: http://lkml.kernel.org/r/20171018075952.10627-9-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "kmemcheck: kill kmemcheck", v2.
As discussed at LSF/MM, kill kmemcheck.
KASan is a replacement that is able to work without the limitation of
kmemcheck (single CPU, slow). KASan is already upstream.
We are also not aware of any users of kmemcheck (or users who don't
consider KASan as a suitable replacement).
The only objection was that since KASAN wasn't supported by all GCC
versions provided by distros at that time we should hold off for 2
years, and try again.
Now that 2 years have passed, and all distros provide gcc that supports
KASAN, kill kmemcheck again for the very same reasons.
This patch (of 4):
Remove kmemcheck annotations, and calls to kmemcheck from the kernel.
[alexander.levin@verizon.com: correctly remove kmemcheck call from dma_map_sg_attrs]
Link: http://lkml.kernel.org/r/20171012192151.26531-1-alexander.levin@verizon.com
Link: http://lkml.kernel.org/r/20171007030159.22241-2-alexander.levin@verizon.com
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tim Hansen <devtimhansen@gmail.com>
Cc: Vegard Nossum <vegardno@ifi.uio.no>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Simple cases of overlapping changes in the packet scheduler.
Must easier to resolve this time.
Which probably means that I screwed it up somehow.
Signed-off-by: David S. Miller <davem@davemloft.net>
When run ipvs in two different network namespace at the same host, and one
ipvs transport network traffic to the other network namespace ipvs.
'ipvs_property' flag will make the second ipvs take no effect. So we should
clear 'ipvs_property' when SKB network namespace changed.
Fixes: 621e84d6f3 ("dev: introduce skb_scrub_packet()")
Signed-off-by: Ye Yin <hustcat@gmail.com>
Signed-off-by: Wei Zhou <chouryzhou@gmail.com>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
There were quite a few overlapping sets of changes here.
Daniel's bug fix for off-by-ones in the new BPF branch instructions,
along with the added allowances for "data_end > ptr + x" forms
collided with the metadata additions.
Along with those three changes came veritifer test cases, which in
their final form I tried to group together properly. If I had just
trimmed GIT's conflict tags as-is, this would have split up the
meta tests unnecessarily.
In the socketmap code, a set of preemption disabling changes
overlapped with the rename of bpf_compute_data_end() to
bpf_compute_data_pointers().
Changes were made to the mv88e6060.c driver set addr method
which got removed in net-next.
The hyperv transport socket layer had a locking change in 'net'
which overlapped with a change of socket state macro usage
in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
Syzkaller hits WARN_ON(sk->sk_wmem_queued) in sk_stream_kill_queues
after triggering an EFAULT in __zerocopy_sg_from_iter.
On this error, skb_zerocopy_stream_iter resets the skb to its state
before the operation with __pskb_trim. It cannot kfree_skb like
datagram callers, as the skb may have data from a previous send call.
__pskb_trim calls skb_condense for unowned skbs, which adjusts their
truesize. These tcp skbuffs are owned and their truesize must add up
to sk_wmem_queued. But they match because their skb->sk is NULL until
tcp_transmit_skb.
Temporarily set skb->sk when calling __pskb_trim to signal that the
skbuffs are owned and avoid the skb_condense path.
Fixes: 52267790ef ("sock: add MSG_ZEROCOPY")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix BUG() calls to use BUG_ON(conditional) macros.
This was found using make coccicheck M=net/core on linux next
tag next-2017092
Signed-off-by: Tim Hansen <devtimhansen@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Compiler does not really know that skb_shinfo(to|from) are constants
in skb_try_coalesce(), lets cache their values to shrink code.
We might even take care of skb_zcopy() calls later.
$ size net/core/skbuff.o.before net/core/skbuff.o
text data bss dec hex filename
40727 1298 0 42025 a429 net/core/skbuff.o.before
40631 1298 0 41929 a3c9 net/core/skbuff.o
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This work enables generic transfer of metadata from XDP into skb. The
basic idea is that we can make use of the fact that the resulting skb
must be linear and already comes with a larger headroom for supporting
bpf_xdp_adjust_head(), which mangles xdp->data. Here, we base our work
on a similar principle and introduce a small helper bpf_xdp_adjust_meta()
for adjusting a new pointer called xdp->data_meta. Thus, the packet has
a flexible and programmable room for meta data, followed by the actual
packet data. struct xdp_buff is therefore laid out that we first point
to data_hard_start, then data_meta directly prepended to data followed
by data_end marking the end of packet. bpf_xdp_adjust_head() takes into
account whether we have meta data already prepended and if so, memmove()s
this along with the given offset provided there's enough room.
xdp->data_meta is optional and programs are not required to use it. The
rationale is that when we process the packet in XDP (e.g. as DoS filter),
we can push further meta data along with it for the XDP_PASS case, and
give the guarantee that a clsact ingress BPF program on the same device
can pick this up for further post-processing. Since we work with skb
there, we can also set skb->mark, skb->priority or other skb meta data
out of BPF, thus having this scratch space generic and programmable
allows for more flexibility than defining a direct 1:1 transfer of
potentially new XDP members into skb (it's also more efficient as we
don't need to initialize/handle each of such new members). The facility
also works together with GRO aggregation. The scratch space at the head
of the packet can be multiple of 4 byte up to 32 byte large. Drivers not
yet supporting xdp->data_meta can simply be set up with xdp->data_meta
as xdp->data + 1 as bpf_xdp_adjust_meta() will detect this and bail out,
such that the subsequent match against xdp->data for later access is
guaranteed to fail.
The verifier treats xdp->data_meta/xdp->data the same way as we treat
xdp->data/xdp->data_end pointer comparisons. The requirement for doing
the compare against xdp->data is that it hasn't been modified from it's
original address we got from ctx access. It may have a range marking
already from prior successful xdp->data/xdp->data_end pointer comparisons
though.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As measured in my prior patch ("sch_netem: faster rb tree removal"),
rbtree_postorder_for_each_entry_safe() is nice looking but much slower
than using rb_next() directly, except when tree is small enough
to fit in CPU caches (then the cost is the same)
Also note that there is not even an increase of text size :
$ size net/core/skbuff.o.before net/core/skbuff.o
text data bss dec hex filename
40711 1298 0 42009 a419 net/core/skbuff.o.before
40711 1298 0 42009 a419 net/core/skbuff.o
From: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit 0ddf3fb2c4 ("udp: preserve skb->dst if required
for IP options processing") we clear the skb head state as soon
as the skb carrying them is first processed.
Since the same skb can be processed several times when MSG_PEEK
is used, we can end up lacking the required head states, and
eventually oopsing.
Fix this clearing the skb head state only when processing the
last skb reference.
Reported-by: Eric Dumazet <edumazet@google.com>
Fixes: 0ddf3fb2c4 ("udp: preserve skb->dst if required for IP options processing")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
v2: added the change in drivers/vhost/net.c as spotted
by Willem.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to convert this atomic_t refcnt to refcount_t,
we need to init the refcount to one to not trigger
a 0 -> 1 transition.
This also removes one atomic operation in fast path.
v2: removed dead code in sock_zerocopy_put_abort()
as suggested by Willem.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename skb_pad() into __skb_pad() and make it take a third argument:
free_on_error which controls whether kfree_skb() should be called or
not, skb_pad() directly makes use of it and passes true to preserve its
existing behavior. Do exactly the same thing with __skb_put_padto() and
skb_put_padto().
Suggested-by: David Miller <davem@davemloft.net>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Woojung Huh <Woojung.Huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A couple fixes to new skb_send_sock infrastructure. However, no users
currently exist for this code (adding user in next handful of patches)
so it should not be possible to trigger a panic with existing in-kernel
code.
Fixes: 306b13eb3c ("proto_ops: Add locked held versions of sendmsg and sendpage")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only call mm_unaccount_pinned_pages when releasing a struct ubuf_info
that has initialized its field uarg->mmp.
Before this patch, a vhost-net with experimental_zcopytx can crash in
mm_unaccount_pinned_pages
sock_zerocopy_put
skb_zcopy_clear
skb_release_data
Only sock_zerocopy_alloc initializes this field. Move the unaccount
call from generic sock_zerocopy_put to its specific callback
sock_zerocopy_callback.
Fixes: a91dbff551 ("sock: ulimit on MSG_ZEROCOPY pages")
Reported-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bound the number of pages that a user may pin.
Follow the lead of perf tools to maintain a per-user bound on memory
locked pages commit 789f90fcf6 ("perf_counter: per user mlock gift")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the simple case, each sendmsg() call generates data and eventually
a zerocopy ready notification N, where N indicates the Nth successful
invocation of sendmsg() with the MSG_ZEROCOPY flag on this socket.
TCP and corked sockets can cause send() calls to append new data to an
existing sk_buff and, thus, ubuf_info. In that case the notification
must hold a range. odify ubuf_info to store a inclusive range [N..N+m]
and add skb_zerocopy_realloc() to optionally extend an existing range.
Also coalesce notifications in this common case: if a notification
[1, 1] is about to be queued while [0, 0] is the queue tail, just modify
the head of the queue to read [0, 1].
Coalescing is limited to a few TSO frames worth of data to bound
notification latency.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prepare the datapath for refcounted ubuf_info. Clone ubuf_info with
skb_zerocopy_clone() wherever needed due to skb split, merge, resize
or clone.
Split skb_orphan_frags into two variants. The split, merge, .. paths
support reference counted zerocopy buffers, so do not do a deep copy.
Add skb_orphan_frags_rx for paths that may loop packets to receive
sockets. That is not allowed, as it may cause unbounded latency.
Deep copy all zerocopy copy buffers, ref-counted or not, in this path.
The exact locations to modify were chosen by exhaustively searching
through all code that might modify skb_frag references and/or the
the SKBTX_DEV_ZEROCOPY tx_flags bit.
The changes err on the safe side, in two ways.
(1) legacy ubuf_info paths virtio and tap are not modified. They keep
a 1:1 ubuf_info to sk_buff relationship. Calls to skb_orphan_frags
still call skb_copy_ubufs and thus copy frags in this case.
(2) not all copies deep in the stack are addressed yet. skb_shift,
skb_split and skb_try_coalesce can be refined to avoid copying.
These are not in the hot path and this patch is hairy enough as
is, so that is left for future refinement.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The send call ignores unknown flags. Legacy applications may already
unwittingly pass MSG_ZEROCOPY. Continue to ignore this flag unless a
socket opts in to zerocopy.
Introduce socket option SO_ZEROCOPY to enable MSG_ZEROCOPY processing.
Processes can also query this socket option to detect kernel support
for the feature. Older kernels will return ENOPROTOOPT.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kernel supports zerocopy sendmsg in virtio and tap. Expand the
infrastructure to support other socket types. Introduce a completion
notification channel over the socket error queue. Notifications are
returned with ee_origin SO_EE_ORIGIN_ZEROCOPY. ee_errno is 0 to avoid
blocking the send/recv path on receiving notifications.
Add reference counting, to support the skb split, merge, resize and
clone operations possible with SOCK_STREAM and other socket types.
The patch does not yet modify any datapaths.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refine skb_copy_ubufs to support compound pages. With upcoming TCP
zerocopy sendmsg, such fragments may appear.
The existing code replaces each page one for one. Splitting each
compound page into an independent number of regular pages can result
in exceeding limit MAX_SKB_FRAGS if data is not exactly page aligned.
Instead, fill all destination pages but the last to PAGE_SIZE.
Split the existing alloc + copy loop into separate stages:
1. compute bytelength and minimum number of pages to store this.
2. allocate
3. copy, filling each page except the last to PAGE_SIZE bytes
4. update skb frag array
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Skb frags may contain compound pages. Various operations map frags
temporarily using kmap_atomic, but this function works on single
pages, not whole compound pages. The distinction is only relevant
for high mem pages that require temporary mappings.
Introduce a looping mechanism that for compound highmem pages maps
one page at a time, does not change behavior on other pages.
Use the loop in the kmap_atomic callers in net/core/skbuff.c.
Verified by triggering skb_copy_bits with
tcpdump -n -c 100 -i ${DEV} -w /dev/null &
netperf -t TCP_STREAM -H ${HOST}
and by triggering __skb_checksum with
ethtool -K ${DEV} tx off
repeated the tests with looping on a non-highmem platform
(x86_64) by making skb_frag_must_loop always return true.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add skb_send_sock to send an skbuff on a socket within the kernel.
Arguments include an offset so that an skbuf might be sent in mulitple
calls (e.g. send buffer limit is hit).
Signed-off-by: Tom Herbert <tom@quantonium.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
A null check is needed after all. netlink skbs can have skb->head be
backed by vmalloc. The netlink destructor vfree()s head, then sets it to
NULL. We then panic in skb_release_data with a NULL dereference.
Re-add such a test.
Alternative would be to switch to kvfree to free skb->head memory
and remove the special handling in netlink destructor.
Reported-by: kernel test robot <fengguang.wu@intel.com>
Fixes: 06dc75ab06 ("net: Revert "net: add function to allocate sk_buff head without data area")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
It was added for netlink mmap tx, there are no callers in the tree.
The commit also added a check for skb->head != NULL in kfree_skb path,
remove that too -- all skbs ought to have skb->head set.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the pull_pages code block, if the first frag size > eat,
we can end the loop in advance to avoid extra copy.
Signed-off-by: Lin Zhang <xiaolou4617@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__GFP_REPEAT was designed to allow retry-but-eventually-fail semantic to
the page allocator. This has been true but only for allocations
requests larger than PAGE_ALLOC_COSTLY_ORDER. It has been always
ignored for smaller sizes. This is a bit unfortunate because there is
no way to express the same semantic for those requests and they are
considered too important to fail so they might end up looping in the
page allocator for ever, similarly to GFP_NOFAIL requests.
Now that the whole tree has been cleaned up and accidental or misled
usage of __GFP_REPEAT flag has been removed for !costly requests we can
give the original flag a better name and more importantly a more useful
semantic. Let's rename it to __GFP_RETRY_MAYFAIL which tells the user
that the allocator would try really hard but there is no promise of a
success. This will work independent of the order and overrides the
default allocator behavior. Page allocator users have several levels of
guarantee vs. cost options (take GFP_KERNEL as an example)
- GFP_KERNEL & ~__GFP_RECLAIM - optimistic allocation without _any_
attempt to free memory at all. The most light weight mode which even
doesn't kick the background reclaim. Should be used carefully because
it might deplete the memory and the next user might hit the more
aggressive reclaim
- GFP_KERNEL & ~__GFP_DIRECT_RECLAIM (or GFP_NOWAIT)- optimistic
allocation without any attempt to free memory from the current
context but can wake kswapd to reclaim memory if the zone is below
the low watermark. Can be used from either atomic contexts or when
the request is a performance optimization and there is another
fallback for a slow path.
- (GFP_KERNEL|__GFP_HIGH) & ~__GFP_DIRECT_RECLAIM (aka GFP_ATOMIC) -
non sleeping allocation with an expensive fallback so it can access
some portion of memory reserves. Usually used from interrupt/bh
context with an expensive slow path fallback.
- GFP_KERNEL - both background and direct reclaim are allowed and the
_default_ page allocator behavior is used. That means that !costly
allocation requests are basically nofail but there is no guarantee of
that behavior so failures have to be checked properly by callers
(e.g. OOM killer victim is allowed to fail currently).
- GFP_KERNEL | __GFP_NORETRY - overrides the default allocator behavior
and all allocation requests fail early rather than cause disruptive
reclaim (one round of reclaim in this implementation). The OOM killer
is not invoked.
- GFP_KERNEL | __GFP_RETRY_MAYFAIL - overrides the default allocator
behavior and all allocation requests try really hard. The request
will fail if the reclaim cannot make any progress. The OOM killer
won't be triggered.
- GFP_KERNEL | __GFP_NOFAIL - overrides the default allocator behavior
and all allocation requests will loop endlessly until they succeed.
This might be really dangerous especially for larger orders.
Existing users of __GFP_REPEAT are changed to __GFP_RETRY_MAYFAIL
because they already had their semantic. No new users are added.
__alloc_pages_slowpath is changed to bail out for __GFP_RETRY_MAYFAIL if
there is no progress and we have already passed the OOM point.
This means that all the reclaim opportunities have been exhausted except
the most disruptive one (the OOM killer) and a user defined fallback
behavior is more sensible than keep retrying in the page allocator.
[akpm@linux-foundation.org: fix arch/sparc/kernel/mdesc.c]
[mhocko@suse.com: semantic fix]
Link: http://lkml.kernel.org/r/20170626123847.GM11534@dhcp22.suse.cz
[mhocko@kernel.org: address other thing spotted by Vlastimil]
Link: http://lkml.kernel.org/r/20170626124233.GN11534@dhcp22.suse.cz
Link: http://lkml.kernel.org/r/20170623085345.11304-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Alex Belits <alex.belits@cavium.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David Daney <david.daney@cavium.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: NeilBrown <neilb@suse.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
This patch uses refcount_inc_not_zero() instead of
atomic_inc_not_zero_hint() due to absense of a _hint()
version of refcount API. If the hint() version must
be used, we might need to revisit API.
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Steffen Klassert