Pull networking updates from David Miller:
"Another merge window, another pull full of stuff:
1) Support alternative names for network devices, from Jiri Pirko.
2) Introduce per-netns netdev notifiers, also from Jiri Pirko.
3) Support MSG_PEEK in vsock/virtio, from Matias Ezequiel Vara
Larsen.
4) Allow compiling out the TLS TOE code, from Jakub Kicinski.
5) Add several new tracepoints to the kTLS code, also from Jakub.
6) Support set channels ethtool callback in ena driver, from Sameeh
Jubran.
7) New SCTP events SCTP_ADDR_ADDED, SCTP_ADDR_REMOVED,
SCTP_ADDR_MADE_PRIM, and SCTP_SEND_FAILED_EVENT. From Xin Long.
8) Add XDP support to mvneta driver, from Lorenzo Bianconi.
9) Lots of netfilter hw offload fixes, cleanups and enhancements,
from Pablo Neira Ayuso.
10) PTP support for aquantia chips, from Egor Pomozov.
11) Add UDP segmentation offload support to igb, ixgbe, and i40e. From
Josh Hunt.
12) Add smart nagle to tipc, from Jon Maloy.
13) Support L2 field rewrite by TC offloads in bnxt_en, from Venkat
Duvvuru.
14) Add a flow mask cache to OVS, from Tonghao Zhang.
15) Add XDP support to ice driver, from Maciej Fijalkowski.
16) Add AF_XDP support to ice driver, from Krzysztof Kazimierczak.
17) Support UDP GSO offload in atlantic driver, from Igor Russkikh.
18) Support it in stmmac driver too, from Jose Abreu.
19) Support TIPC encryption and auth, from Tuong Lien.
20) Introduce BPF trampolines, from Alexei Starovoitov.
21) Make page_pool API more numa friendly, from Saeed Mahameed.
22) Introduce route hints to ipv4 and ipv6, from Paolo Abeni.
23) Add UDP segmentation offload to cxgb4, Rahul Lakkireddy"
* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1857 commits)
libbpf: Fix usage of u32 in userspace code
mm: Implement no-MMU variant of vmalloc_user_node_flags
slip: Fix use-after-free Read in slip_open
net: dsa: sja1105: fix sja1105_parse_rgmii_delays()
macvlan: schedule bc_work even if error
enetc: add support Credit Based Shaper(CBS) for hardware offload
net: phy: add helpers phy_(un)lock_mdio_bus
mdio_bus: don't use managed reset-controller
ax88179_178a: add ethtool_op_get_ts_info()
mlxsw: spectrum_router: Fix use of uninitialized adjacency index
mlxsw: spectrum_router: After underlay moves, demote conflicting tunnels
bpf: Simplify __bpf_arch_text_poke poke type handling
bpf: Introduce BPF_TRACE_x helper for the tracing tests
bpf: Add bpf_jit_blinding_enabled for !CONFIG_BPF_JIT
bpf, testing: Add various tail call test cases
bpf, x86: Emit patchable direct jump as tail call
bpf: Constant map key tracking for prog array pokes
bpf: Add poke dependency tracking for prog array maps
bpf: Add initial poke descriptor table for jit images
bpf: Move owner type, jited info into array auxiliary data
...
Pull cgroup updates from Tejun Heo:
"There are several notable changes here:
- Single thread migrating itself has been optimized so that it
doesn't need threadgroup rwsem anymore.
- Freezer optimization to avoid unnecessary frozen state changes.
- cgroup ID unification so that cgroup fs ino is the only unique ID
used for the cgroup and can be used to directly look up live
cgroups through filehandle interface on 64bit ino archs. On 32bit
archs, cgroup fs ino is still the only ID in use but it is only
unique when combined with gen.
- selftest and other changes"
* 'for-5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: (24 commits)
writeback: fix -Wformat compilation warnings
docs: cgroup: mm: Fix spelling of "list"
cgroup: fix incorrect WARN_ON_ONCE() in cgroup_setup_root()
cgroup: use cgrp->kn->id as the cgroup ID
kernfs: use 64bit inos if ino_t is 64bit
kernfs: implement custom exportfs ops and fid type
kernfs: combine ino/id lookup functions into kernfs_find_and_get_node_by_id()
kernfs: convert kernfs_node->id from union kernfs_node_id to u64
kernfs: kernfs_find_and_get_node_by_ino() should only look up activated nodes
kernfs: use dumber locking for kernfs_find_and_get_node_by_ino()
netprio: use css ID instead of cgroup ID
writeback: use ino_t for inodes in tracepoints
kernfs: fix ino wrap-around detection
kselftests: cgroup: Avoid the reuse of fd after it is deallocated
cgroup: freezer: don't change task and cgroups status unnecessarily
cgroup: use cgroup->last_bstat instead of cgroup->bstat_pending for consistency
cgroup: remove cgroup_enable_task_cg_lists() optimization
cgroup: pids: use atomic64_t for pids->limit
selftests: cgroup: Run test_core under interfering stress
selftests: cgroup: Add task migration tests
...
Use vlan common api to access the vlan_tag info.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Minor conflict in drivers/s390/net/qeth_l2_main.c, kept the lock
from commit c8183f5489 ("s390/qeth: fix potential deadlock on
workqueue flush"), removed the code which was removed by commit
9897d583b0 ("s390/qeth: consolidate some duplicated HW cmd code").
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Report from Dan Carpenter,
net/core/skmsg.c:792 sk_psock_write_space()
error: we previously assumed 'psock' could be null (see line 790)
net/core/skmsg.c
789 psock = sk_psock(sk);
790 if (likely(psock && sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)))
Check for NULL
791 schedule_work(&psock->work);
792 write_space = psock->saved_write_space;
^^^^^^^^^^^^^^^^^^^^^^^^
793 rcu_read_unlock();
794 write_space(sk);
Ensure psock dereference on line 792 only occurs if psock is not null.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: 604326b41a ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
kobject_put() should only be called in error path.
Fixes: b8eb718348 ("net-sysfs: Fix reference count leak in rx|netdev_queue_add_kobject")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jouni Hogander <jouni.hogander@unikie.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2019-11-20
The following pull-request contains BPF updates for your *net-next* tree.
We've added 81 non-merge commits during the last 17 day(s) which contain
a total of 120 files changed, 4958 insertions(+), 1081 deletions(-).
There are 3 trivial conflicts, resolve it by always taking the chunk from
196e8ca748:
<<<<<<< HEAD
=======
void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
>>>>>>> 196e8ca748
<<<<<<< HEAD
void *bpf_map_area_alloc(u64 size, int numa_node)
=======
static void *__bpf_map_area_alloc(u64 size, int numa_node, bool mmapable)
>>>>>>> 196e8ca748
<<<<<<< HEAD
if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) {
=======
/* kmalloc()'ed memory can't be mmap()'ed */
if (!mmapable && size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) {
>>>>>>> 196e8ca748
The main changes are:
1) Addition of BPF trampoline which works as a bridge between kernel functions,
BPF programs and other BPF programs along with two new use cases: i) fentry/fexit
BPF programs for tracing with practically zero overhead to call into BPF (as
opposed to k[ret]probes) and ii) attachment of the former to networking related
programs to see input/output of networking programs (covering xdpdump use case),
from Alexei Starovoitov.
2) BPF array map mmap support and use in libbpf for global data maps; also a big
batch of libbpf improvements, among others, support for reading bitfields in a
relocatable manner (via libbpf's CO-RE helper API), from Andrii Nakryiko.
3) Extend s390x JIT with usage of relative long jumps and loads in order to lift
the current 64/512k size limits on JITed BPF programs there, from Ilya Leoshkevich.
4) Add BPF audit support and emit messages upon successful prog load and unload in
order to have a timeline of events, from Daniel Borkmann and Jiri Olsa.
5) Extension to libbpf and xdpsock sample programs to demo the shared umem mode
(XDP_SHARED_UMEM) as well as RX-only and TX-only sockets, from Magnus Karlsson.
6) Several follow-up bug fixes for libbpf's auto-pinning code and a new API
call named bpf_get_link_xdp_info() for retrieving the full set of prog
IDs attached to XDP, from Toke Høiland-Jørgensen.
7) Add BTF support for array of int, array of struct and multidimensional arrays
and enable it for skb->cb[] access in kfree_skb test, from Martin KaFai Lau.
8) Fix AF_XDP by using the correct number of channels from ethtool, from Luigi Rizzo.
9) Two fixes for BPF selftest to get rid of a hang in test_tc_tunnel and to avoid
xdping to be run as standalone, from Jiri Benc.
10) Various BPF selftest fixes when run with latest LLVM trunk, from Yonghong Song.
11) Fix a memory leak in BPF fentry test run data, from Colin Ian King.
12) Various smaller misc cleanups and improvements mostly all over BPF selftests and
samples, from Daniel T. Lee, Andre Guedes, Anders Roxell, Mao Wenan, Yue Haibing.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce the following parameters in order to add the possibility to sync
DMA memory for device before putting allocated pages in the page_pool
caches:
- PP_FLAG_DMA_SYNC_DEV: if set in page_pool_params flags, all pages that
the driver gets from page_pool will be DMA-synced-for-device according
to the length provided by the device driver. Please note DMA-sync-for-CPU
is still device driver responsibility
- offset: DMA address offset where the DMA engine starts copying rx data
- max_len: maximum DMA memory size page_pool is allowed to flush. This
is currently used in __page_pool_alloc_pages_slow routine when pages
are allocated from page allocator
These parameters are supposed to be set by device drivers.
This optimization reduces the length of the DMA-sync-for-device.
The optimization is valid because pages are initially
DMA-synced-for-device as defined via max_len. At RX time, the driver
will perform a DMA-sync-for-CPU on the memory for the packet length.
What is important is the memory occupied by packet payload, because
this is the area CPU is allowed to read and modify. As we don't track
cache-lines written into by the CPU, simply use the packet payload length
as dma_sync_size at page_pool recycle time. This also take into account
any tail-extend.
Tested-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
kobject_init_and_add takes reference even when it fails. This has
to be given up by the caller in error handling. Otherwise memory
allocated by kobject_init_and_add is never freed. Originally found
by Syzkaller:
BUG: memory leak
unreferenced object 0xffff8880679f8b08 (size 8):
comm "netdev_register", pid 269, jiffies 4294693094 (age 12.132s)
hex dump (first 8 bytes):
72 78 2d 30 00 36 20 d4 rx-0.6 .
backtrace:
[<000000008c93818e>] __kmalloc_track_caller+0x16e/0x290
[<000000001f2e4e49>] kvasprintf+0xb1/0x140
[<000000007f313394>] kvasprintf_const+0x56/0x160
[<00000000aeca11c8>] kobject_set_name_vargs+0x5b/0x140
[<0000000073a0367c>] kobject_init_and_add+0xd8/0x170
[<0000000088838e4b>] net_rx_queue_update_kobjects+0x152/0x560
[<000000006be5f104>] netdev_register_kobject+0x210/0x380
[<00000000e31dab9d>] register_netdevice+0xa1b/0xf00
[<00000000f68b2465>] __tun_chr_ioctl+0x20d5/0x3dd0
[<000000004c50599f>] tun_chr_ioctl+0x2f/0x40
[<00000000bbd4c317>] do_vfs_ioctl+0x1c7/0x1510
[<00000000d4c59e8f>] ksys_ioctl+0x99/0xb0
[<00000000946aea81>] __x64_sys_ioctl+0x78/0xb0
[<0000000038d946e5>] do_syscall_64+0x16f/0x580
[<00000000e0aa5d8f>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[<00000000285b3d1a>] 0xffffffffffffffff
Cc: David Miller <davem@davemloft.net>
Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Jouni Hogander <jouni.hogander@unikie.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A page is NOT reusable when at least one of the following is true:
1) allocated when system was under some pressure. (page_is_pfmemalloc)
2) belongs to a different NUMA node than pool->p.nid.
To update pool->p.nid users should call page_pool_update_nid().
Holding on to such pages in the pool will hurt the consumer performance
when the pool migrates to a different numa node.
Performance testing:
XDP drop/tx rate and TCP single/multi stream, on mlx5 driver
while migrating rx ring irq from close to far numa:
mlx5 internal page cache was locally disabled to get pure page pool
results.
CPU: Intel(R) Xeon(R) CPU E5-2603 v4 @ 1.70GHz
NIC: Mellanox Technologies MT27700 Family [ConnectX-4] (100G)
XDP Drop/TX single core:
NUMA | XDP | Before | After
---------------------------------------
Close | Drop | 11 Mpps | 10.9 Mpps
Far | Drop | 4.4 Mpps | 5.8 Mpps
Close | TX | 6.5 Mpps | 6.5 Mpps
Far | TX | 3.5 Mpps | 4 Mpps
Improvement is about 30% drop packet rate, 15% tx packet rate for numa
far test.
No degradation for numa close tests.
TCP single/multi cpu/stream:
NUMA | #cpu | Before | After
--------------------------------------
Close | 1 | 18 Gbps | 18 Gbps
Far | 1 | 15 Gbps | 18 Gbps
Close | 12 | 80 Gbps | 80 Gbps
Far | 12 | 68 Gbps | 80 Gbps
In all test cases we see improvement for the far numa case, and no
impact on the close numa case.
The impact of adding a check per page is very negligible, and shows no
performance degradation whatsoever, also functionality wise it seems more
correct and more robust for page pool to verify when pages should be
recycled, since page pool can't guarantee where pages are coming from.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add page_pool_update_nid() to be called by page pool consumers when they
detect numa node changes.
It will update the page pool nid value to start allocating from the new
effective numa node.
This is to mitigate page pool allocating pages from a wrong numa node,
where the pool was originally allocated, and holding on to pages that
belong to a different numa node, which causes performance degradation.
For pages that are already being consumed and could be returned to the
pool by the consumer, in next patch we will add a check per page to avoid
recycling them back to the pool and return them to the page allocator.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When Jonathan change the page_pool to become responsible to its
own shutdown via deferred work queue, then the disconnect_cnt
counter was removed from xdp memory model tracepoint.
This patch change the page_pool_inflight tracepoint name to
page_pool_release, because it reflects the new responsability
better. And it reintroduces a counter that reflect the number of
times page_pool_release have been tried.
The counter is also used by the code, to only empty the alloc
cache once. With a stuck work queue running every second and
counter being 64-bit, it will overrun in approx 584 billion
years. For comparison, Earth lifetime expectancy is 7.5 billion
years, before the Sun will engulf, and destroy, the Earth.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When looking at the details I realised that the memory poison in
__xdp_mem_allocator_rcu_free doesn't make sense. This is because the
SLUB allocator uses the first 16 bytes (on 64 bit), for its freelist,
which overlap with members in struct xdp_mem_allocator, that were
updated. Thus, SLUB already does the "poisoning" for us.
I still believe that poisoning memory make sense in other cases.
Kernel have gained different use-after-free detection mechanism, but
enabling those is associated with a huge overhead. Experience is that
debugging facilities can change the timing so much, that that a race
condition will not be provoked when enabled. Thus, I'm still in favour
of poisoning memory where it makes sense.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
92117d8443 ("bpf: fix refcnt overflow") turned refcounting of bpf_map into
potentially failing operation, when refcount reaches BPF_MAX_REFCNT limit
(32k). Due to using 32-bit counter, it's possible in practice to overflow
refcounter and make it wrap around to 0, causing erroneous map free, while
there are still references to it, causing use-after-free problems.
But having a failing refcounting operations are problematic in some cases. One
example is mmap() interface. After establishing initial memory-mapping, user
is allowed to arbitrarily map/remap/unmap parts of mapped memory, arbitrarily
splitting it into multiple non-contiguous regions. All this happening without
any control from the users of mmap subsystem. Rather mmap subsystem sends
notifications to original creator of memory mapping through open/close
callbacks, which are optionally specified during initial memory mapping
creation. These callbacks are used to maintain accurate refcount for bpf_map
(see next patch in this series). The problem is that open() callback is not
supposed to fail, because memory-mapped resource is set up and properly
referenced. This is posing a problem for using memory-mapping with BPF maps.
One solution to this is to maintain separate refcount for just memory-mappings
and do single bpf_map_inc/bpf_map_put when it goes from/to zero, respectively.
There are similar use cases in current work on tcp-bpf, necessitating extra
counter as well. This seems like a rather unfortunate and ugly solution that
doesn't scale well to various new use cases.
Another approach to solve this is to use non-failing refcount_t type, which
uses 32-bit counter internally, but, once reaching overflow state at UINT_MAX,
stays there. This utlimately causes memory leak, but prevents use after free.
But given refcounting is not the most performance-critical operation with BPF
maps (it's not used from running BPF program code), we can also just switch to
64-bit counter that can't overflow in practice, potentially disadvantaging
32-bit platforms a tiny bit. This simplifies semantics and allows above
described scenarios to not worry about failing refcount increment operation.
In terms of struct bpf_map size, we are still good and use the same amount of
space:
BEFORE (3 cache lines, 8 bytes of padding at the end):
struct bpf_map {
const struct bpf_map_ops * ops __attribute__((__aligned__(64))); /* 0 8 */
struct bpf_map * inner_map_meta; /* 8 8 */
void * security; /* 16 8 */
enum bpf_map_type map_type; /* 24 4 */
u32 key_size; /* 28 4 */
u32 value_size; /* 32 4 */
u32 max_entries; /* 36 4 */
u32 map_flags; /* 40 4 */
int spin_lock_off; /* 44 4 */
u32 id; /* 48 4 */
int numa_node; /* 52 4 */
u32 btf_key_type_id; /* 56 4 */
u32 btf_value_type_id; /* 60 4 */
/* --- cacheline 1 boundary (64 bytes) --- */
struct btf * btf; /* 64 8 */
struct bpf_map_memory memory; /* 72 16 */
bool unpriv_array; /* 88 1 */
bool frozen; /* 89 1 */
/* XXX 38 bytes hole, try to pack */
/* --- cacheline 2 boundary (128 bytes) --- */
atomic_t refcnt __attribute__((__aligned__(64))); /* 128 4 */
atomic_t usercnt; /* 132 4 */
struct work_struct work; /* 136 32 */
char name[16]; /* 168 16 */
/* size: 192, cachelines: 3, members: 21 */
/* sum members: 146, holes: 1, sum holes: 38 */
/* padding: 8 */
/* forced alignments: 2, forced holes: 1, sum forced holes: 38 */
} __attribute__((__aligned__(64)));
AFTER (same 3 cache lines, no extra padding now):
struct bpf_map {
const struct bpf_map_ops * ops __attribute__((__aligned__(64))); /* 0 8 */
struct bpf_map * inner_map_meta; /* 8 8 */
void * security; /* 16 8 */
enum bpf_map_type map_type; /* 24 4 */
u32 key_size; /* 28 4 */
u32 value_size; /* 32 4 */
u32 max_entries; /* 36 4 */
u32 map_flags; /* 40 4 */
int spin_lock_off; /* 44 4 */
u32 id; /* 48 4 */
int numa_node; /* 52 4 */
u32 btf_key_type_id; /* 56 4 */
u32 btf_value_type_id; /* 60 4 */
/* --- cacheline 1 boundary (64 bytes) --- */
struct btf * btf; /* 64 8 */
struct bpf_map_memory memory; /* 72 16 */
bool unpriv_array; /* 88 1 */
bool frozen; /* 89 1 */
/* XXX 38 bytes hole, try to pack */
/* --- cacheline 2 boundary (128 bytes) --- */
atomic64_t refcnt __attribute__((__aligned__(64))); /* 128 8 */
atomic64_t usercnt; /* 136 8 */
struct work_struct work; /* 144 32 */
char name[16]; /* 176 16 */
/* size: 192, cachelines: 3, members: 21 */
/* sum members: 154, holes: 1, sum holes: 38 */
/* forced alignments: 2, forced holes: 1, sum forced holes: 38 */
} __attribute__((__aligned__(64)));
This patch, while modifying all users of bpf_map_inc, also cleans up its
interface to match bpf_map_put with separate operations for bpf_map_inc and
bpf_map_inc_with_uref (to match bpf_map_put and bpf_map_put_with_uref,
respectively). Also, given there are no users of bpf_map_inc_not_zero
specifying uref=true, remove uref flag and default to uref=false internally.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20191117172806.2195367-2-andriin@fb.com
Commit 78d3fd0b7d ("gro: Only use skb_gro_header for completely
non-linear packets") back in May'09 (v2.6.31-rc1) has changed the
original condition '!skb_headlen(skb)' to
'skb->mac_header == skb->tail' in gro_reset_offset() saying: "Since
the drivers that need this optimisation all provide completely
non-linear packets" (note that this condition has become the current
'skb_mac_header(skb) == skb_tail_pointer(skb)' later with commmit
ced14f6804 ("net: Correct comparisons and calculations using
skb->tail and skb-transport_header") without any functional changes).
For now, we have the following rough statistics for v5.4-rc7:
1) napi_gro_frags: 14
2) napi_gro_receive with skb->head containing (most of) payload: 83
3) napi_gro_receive with skb->head containing all the headers: 20
4) napi_gro_receive with skb->head containing only Ethernet header: 2
With the current condition, fast GRO with the usage of
NAPI_GRO_CB(skb)->frag0 is available only in the [1] case.
Packets pushed by [2] and [3] go through the 'slow' path, but
it's not a problem for them as they already contain all the needed
headers in skb->head, so pskb_may_pull() only moves skb->data.
The layout of skbs in the fourth [4] case at the moment of
dev_gro_receive() is identical to skbs that have come through [1],
as napi_frags_skb() pulls Ethernet header to skb->head. The only
difference is that the mentioned condition is always false for them,
because skb_put() and friends irreversibly alter the tail pointer.
They also go through the 'slow' path, but now every single
pskb_may_pull() in every single .gro_receive() will call the *really*
slow __pskb_pull_tail() to pull headers to head. This significantly
decreases the overall performance for no visible reasons.
The only two users of method [4] is:
* drivers/staging/qlge
* drivers/net/wireless/iwlwifi (all three variants: dvm, mvm, mvm-mq)
Note that in case with wireless drivers we can't use [1]
(napi_gro_frags()) at least for now and mac80211 stack always
performs pushes and pulls anyways, so performance hit is inavoidable.
At the moment of v2.6.31 the mentioned change was necessary (that's
why I don't add the "Fixes:" tag), but it became obsolete since
skb_gro_mac_header() has gone in commit a50e233c50 ("net-gro:
restore frag0 optimization"), so we can simply revert the condition
in gro_reset_offset() to allow skbs from [4] go through the 'fast'
path just like in case [1].
This was tested on a 600 MHz MIPS CPU and a custom driver and this
patch gave boosts up to 40 Mbps to method [4] in both directions
comparing to net-next, which made overall performance relatively
close to [1] (without it, [4] is the slowest).
v2:
- Add more references and explanations to commit message
- Fix some typos ibid
- No functional changes
Signed-off-by: Alexander Lobakin <alobakin@dlink.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
The page pool keeps track of the number of pages in flight, and
it isn't safe to remove the pool until all pages are returned.
Disallow removing the pool until all pages are back, so the pool
is always available for page producers.
Make the page pool responsible for its own delayed destruction
instead of relying on XDP, so the page pool can be used without
the xdp memory model.
When all pages are returned, free the pool and notify xdp if the
pool is registered with the xdp memory system. Have the callback
perform a table walk since some drivers (cpsw) may share the pool
among multiple xdp_rxq_info.
Note that the increment of pages_state_release_cnt may result in
inflight == 0, resulting in the pool being released.
Fixes: d956a048cd ("xdp: force mem allocator removal and periodic warning")
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Annotate BPF program context types with program-side type and kernel-side type.
This type information is used by the verifier. btf_get_prog_ctx_type() is
used in the later patches to verify that BTF type of ctx in BPF program matches to
kernel expected ctx type. For example, the XDP program type is:
BPF_PROG_TYPE(BPF_PROG_TYPE_XDP, xdp, struct xdp_md, struct xdp_buff)
That means that XDP program should be written as:
int xdp_prog(struct xdp_md *ctx) { ... }
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20191114185720.1641606-16-ast@kernel.org
btf_resolve_helper_id() caching logic is a bit racy, since under root the
verifier can verify several programs in parallel. Fix it with READ/WRITE_ONCE.
Fix the type as well, since error is also recorded.
Fixes: a7658e1a41 ("bpf: Check types of arguments passed into helpers")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191114185720.1641606-15-ast@kernel.org
1) New generic devlink param "enable_roce", for downstream devlink
reload support
2) Do vport ACL configuration on per vport basis when
enabling/disabling a vport. This enables to have vports enabled/disabled
outside of eswitch config for future
3) Split the code for legacy vs offloads mode and make it clear
4) Tide up vport locking and workqueue usage
5) Fix metadata enablement for ECPF
6) Make explicit use of VF property to publish IB_DEVICE_VIRTUAL_FUNCTION
7) E-Switch and flow steering core low level support and refactoring for
netfilter flowtables offload
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Devlink supports pair output of name and value. When the value is
binary, it must be presented in an array. If the length of the binary
value exceeds fmsg limitation, break the value into chunks internally.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cgroup ID is currently allocated using a dedicated per-hierarchy idr
and used internally and exposed through tracepoints and bpf. This is
confusing because there are tracepoints and other interfaces which use
the cgroupfs ino as IDs.
The preceding changes made kn->id exposed as ino as 64bit ino on
supported archs or ino+gen (low 32bits as ino, high gen). There's no
reason for cgroup to use different IDs. The kernfs IDs are unique and
userland can easily discover them and map them back to paths using
standard file operations.
This patch replaces cgroup IDs with kernfs IDs.
* cgroup_id() is added and all cgroup ID users are converted to use it.
* kernfs_node creation is moved to earlier during cgroup init so that
cgroup_id() is available during init.
* While at it, s/cgroup/cgrp/ in psi helpers for consistency.
* Fallback ID value is changed to 1 to be consistent with root cgroup
ID.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
kernfs_node->id is currently a union kernfs_node_id which represents
either a 32bit (ino, gen) pair or u64 value. I can't see much value
in the usage of the union - all that's needed is a 64bit ID which the
current code is already limited to. Using a union makes the code
unnecessarily complicated and prevents using 64bit ino without adding
practical benefits.
This patch drops union kernfs_node_id and makes kernfs_node->id a u64.
ino is stored in the lower 32bits and gen upper. Accessors -
kernfs[_id]_ino() and kernfs[_id]_gen() - are added to retrieve the
ino and gen. This simplifies ID handling less cumbersome and will
allow using 64bit inos on supported archs.
This patch doesn't make any functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Alexei Starovoitov <ast@kernel.org>
netprio uses cgroup ID to index the priority mapping table. This is
currently okay as cgroup IDs are allocated using idr and packed.
However, cgroup IDs will be changed to use full 64bit range and won't
be packed making this impractical. netprio doesn't care what type of
IDs it uses as long as they can identify the controller instances and
are packed. Let's switch to css IDs instead of cgroup IDs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Namhyung Kim <namhyung@kernel.org>
When setting the dump's time-stamp, use ktime_get_real in addition to
jiffies. This simplifies the user space implementation and bypasses
some inconsistent behavior with translating jiffies to current time.
The time taken is transformed into nsec, to comply with y2038 issue.
Fixes: c8e1da0bf9 ("devlink: Add health report functionality")
Signed-off-by: Aya Levin <ayal@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
New device parameter to enable/disable handling of RoCE traffic in the
device.
Signed-off-by: Michael Guralnik <michaelgur@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
There is a race between driver code that does setup/cleanup of device
and devlink reload operation that in some drivers works with the same
code. Use after free could we easily obtained by running:
while true; do
echo "0000:00:10.0" >/sys/bus/pci/drivers/mlxsw_spectrum2/bind
devlink dev reload pci/0000:00:10.0 &
echo "0000:00:10.0" >/sys/bus/pci/drivers/mlxsw_spectrum2/unbind
done
Fix this by enabling reload only after setup of device is complete and
disabling it at the beginning of the cleanup process.
Reported-by: Ido Schimmel <idosch@mellanox.com>
Fixes: 2d8dc5bbf4 ("devlink: Add support for reload")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
One conflict in the BPF samples Makefile, some fixes in 'net' whilst
we were converting over to Makefile.target rules in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a race between driver code that does setup/cleanup of device
and devlink reload operation that in some drivers works with the same
code. Use after free could we easily obtained by running:
while true; do
echo 10 > /sys/bus/netdevsim/new_device
devlink dev reload netdevsim/netdevsim10 &
echo 10 > /sys/bus/netdevsim/del_device
done
Fix this by enabling reload only after setup of device is complete and
disabling it at the beginning of the cleanup process.
Reported-by: Ido Schimmel <idosch@mellanox.com>
Fixes: 2d8dc5bbf4 ("devlink: Add support for reload")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add layer 3 generic packet exception traps that can report trapped
packets and documentation of the traps.
Unlike drop traps, these exception traps also need to inject the packet
to the kernel's receive path. For example, a packet that was trapped due
to unreachable neighbour need to be injected into the kernel so that it
will trigger an ARP request or a neighbour solicitation message.
Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add packet traps that can report packets that were dropped during layer
3 forwarding.
Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I forgot to change last_packets field in struct net_rate_estimator.
Without this fix, rate estimators would misbehave after more
than 2^32 packets have been sent.
Another solution would be to be careful and only use the
32 least significant bits of packets counters, but we have
a hole in net_rate_estimator structure and this looks
easier to read/maintain.
Fixes: d0083d98f6 ("net_sched: extend packet counter to 64bit")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A difference of two unsigned long needs long storage.
Fixes: c7fb64db00 ("[NETLINK]: Neighbour table configuration and statistics via rtnetlink")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now the kernel uses 64bit packet counters in scheduler layer,
we want to export these counters to user space.
Instead risking breaking user space by adding fields
to struct gnet_stats_basic, add a new TCA_STATS_PKT64.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After this change, qdisc packet counter is no longer
a 32bit quantity. We still export 32bit values to user.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_msg_trim() tries to only update curr pointer if it falls into
the trimmed region. The logic, however, does not take into the
account pointer wrapping that sk_msg_iter_var_prev() does nor
(as John points out) the fact that msg->sg is a ring buffer.
This means that when the message was trimmed completely, the new
curr pointer would have the value of MAX_MSG_FRAGS - 1, which is
neither smaller than any other value, nor would it actually be
correct.
Special case the trimming to 0 length a little bit and rework
the comparison between curr and end to take into account wrapping.
This bug caused the TLS code to not copy all of the message, if
zero copy filled in fewer sg entries than memcopy would need.
Big thanks to Alexander Potapenko for the non-KMSAN reproducer.
v2:
- take into account that msg->sg is a ring buffer (John).
Link: https://lore.kernel.org/netdev/20191030160542.30295-1-jakub.kicinski@netronome.com/ (v1)
Fixes: d829e9c411 ("tls: convert to generic sk_msg interface")
Reported-by: syzbot+f8495bff23a879a6d0bd@syzkaller.appspotmail.com
Reported-by: syzbot+6f50c99e8f6194bf363f@syzkaller.appspotmail.com
Co-developed-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The only slightly tricky merge conflict was the netdevsim because the
mutex locking fix overlapped a lot of driver reload reorganization.
The rest were (relatively) trivial in nature.
Signed-off-by: David S. Miller <davem@davemloft.net>
When netdevice with offloaded BPF programs is destroyed
the programs are orphaned and removed from the program
IDA - their IDs get released (the programs may remain
accessible via existing open file descriptors and pinned
files). After IDs are released they are set to 0.
This confuses dev_change_xdp_fd() because it compares
the __dev_xdp_query() result where 0 means no program
with prog->aux->id where 0 means orphaned.
dev_change_xdp_fd() would have incorrectly returned success
even though it had not installed the program.
Since drivers already catch this case via bpf_offload_dev_match()
let them handle this case. The error message drivers produce in
this case ("program loaded for a different device") is in fact
correct as the orphaned program must had to be loaded for a
different device.
Fixes: c14a9f633d ("net: Don't call XDP_SETUP_PROG when nothing is changed")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ICMP flow dissector currently parses only the Type and Code fields.
Some ICMP packets (echo, timestamp) have a 16 bit Identifier field which
is used to correlate packets.
Add such field in flow_dissector_key_icmp and replace skb_flow_get_be16()
with a more complex function which populate this field.
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
FLOW_DISSECTOR_KEY_ICMP is checked for every packet, not only ICMP ones.
Even if the test overhead is probably negligible, move the
ICMP dissector code under the big 'switch(ip_proto)' so it gets called
only for ICMP packets.
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Documents two piece of code which can't be understood at a glance.
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A simple typo fix in the nl error message (fbd -> fdb).
CC: David Ahern <dsahern@gmail.com>
Fixes: 8c6e137fbc ("rtnetlink: Update rtnl_fdb_dump for strict data checking")
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Busy polling usually runs without locks.
Let's use skb_queue_empty_lockless() instead of skb_queue_empty()
Also uses READ_ONCE() in __skb_try_recv_datagram() to address
a similar potential problem.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Many poll() handlers are lockless. Using skb_queue_empty_lockless()
instead of skb_queue_empty() is more appropriate.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
