- 15 Mar, 2020 40 commits
-
-
Stefano Brivio authored
If the AVX2 set is available, we can exploit the repetitive characteristic of this algorithm to provide a fast, vectorised version by using 256-bit wide AVX2 operations for bucket loads and bitwise intersections. In most cases, this implementation consistently outperforms rbtree set instances despite the fact they are configured to use a given, single, ranged data type out of the ones used for performance measurements by the nft_concat_range.sh kselftest. That script, injecting packets directly on the ingoing device path with pktgen, reports, averaged over five runs on a single AMD Epyc 7402 thread (3.35GHz, 768 KiB L1D$, 12 MiB L2$), the figures below. CONFIG_RETPOLINE was not set here. Note that this is not a fair comparison over hash and rbtree set types: non-ranged entries (used to have a reference for hash types) would be matched faster than this, and matching on a single field only (which is the case for rbtree) is also significantly faster. However, it's not possible at the moment to choose this set type for non-ranged entries, and the current implementation also needs a few minor adjustments in order to match on less than two fields. ---------------.-----------------------------------.------------. AMD Epyc 7402 | baselines, Mpps | this patch | 1 thread |___________________________________|____________| 3.35GHz | | | | | | 768KiB L1D$ | netdev | hash | rbtree | | | ---------------| hook | no | single | | pipapo | type entries | drop | ranges | field | pipapo | AVX2 | ---------------|--------|--------|--------|--------|------------| net,port | | | | | | 1000 | 19.0 | 10.4 | 3.8 | 4.0 | 7.5 +87% | ---------------|--------|--------|--------|--------|------------| port,net | | | | | | 100 | 18.8 | 10.3 | 5.8 | 6.3 | 8.1 +29% | ---------------|--------|--------|--------|--------|------------| net6,port | | | | | | 1000 | 16.4 | 7.6 | 1.8 | 2.1 | 4.8 +128% | ---------------|--------|--------|--------|--------|------------| port,proto | | | | | | 30000 | 19.6 | 11.6 | 3.9 | 0.5 | 2.6 +420% | ---------------|--------|--------|--------|--------|------------| net6,port,mac | | | | | | 10 | 16.5 | 5.4 | 4.3 | 3.4 | 4.7 +38% | ---------------|--------|--------|--------|--------|------------| net6,port,mac, | | | | | | proto 1000 | 16.5 | 5.7 | 1.9 | 1.4 | 3.6 +26% | ---------------|--------|--------|--------|--------|------------| net,mac | | | | | | 1000 | 19.0 | 8.4 | 3.9 | 2.5 | 6.4 +156% | ---------------'--------'--------'--------'--------'------------' A similar strategy could be easily reused to implement specialised versions for other SIMD sets, and I plan to post at least a NEON version at a later time. Signed-off-by:Stefano Brivio <sbrivio@redhat.com> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org>
-
Stefano Brivio authored
Move most macros and helpers to a header file, so that they can be conveniently used by related implementations. No functional changes are intended here. Signed-off-by:
-
Stefano Brivio authored
SIMD vector extension sets require stricter alignment than native instruction sets to operate efficiently (AVX, NEON) or for some instructions to work at all (AltiVec). Provide facilities to define arbitrary alignment for lookup tables and scratch maps. By defining byte alignment with NFT_PIPAPO_ALIGN, lt_aligned and scratch_aligned pointers become available. Additional headroom is allocated, and pointers to the possibly unaligned, originally allocated areas are kept so that they can be freed. Signed-off-by:
-
Stefano Brivio authored
While grouping matching bits in groups of four saves memory compared to the more natural choice of 8-bit words (lookup table size is one eighth), it comes at a performance cost, as the number of lookup comparisons is doubled, and those also needs bitshifts and masking. Introduce support for 8-bit lookup groups, together with a mapping mechanism to dynamically switch, based on defined per-table size thresholds and hysteresis, between 8-bit and 4-bit groups, as tables grow and shrink. Empty sets start with 8-bit groups, and per-field tables are converted to 4-bit groups if they get too big. An alternative approach would have been to swap per-set lookup operation functions as needed, but this doesn't allow for different group sizes in the same set, which looks desirable if some fields need significantly more matching data compared to others due to heavier impact of ranges (e.g. a big number of subnets with relatively simple port specifications). Allowing different group sizes for the same lookup functions implies the need for further conditional clauses, whose cost, however, appears to be negligible in tests. The matching rate figures below were obtained for x86_64 running the nft_concat_range.sh "performance" cases, averaged over five runs, on a single thread of an AMD Epyc 7402 CPU, and for aarch64 on a single thread of a BCM2711 (Raspberry Pi 4 Model B 4GB), clocked at a stable 2147MHz frequency: ---------------.-----------------------------------.------------. AMD Epyc 7402 | baselines, Mpps | this patch | 1 thread |___________________________________|____________| 3.35GHz | | | | | | 768KiB L1D$ | netdev | hash | rbtree | | | ---------------| hook | no | single | pipapo | pipapo | type entries | drop | ranges | field | 4 bits | bit switch | ---------------|--------|--------|--------|--------|------------| net,port | | | | | | 1000 | 19.0 | 10.4 | 3.8 | 2.8 | 4.0 +43% | ---------------|--------|--------|--------|--------|------------| port,net | | | | | | 100 | 18.8 | 10.3 | 5.8 | 5.5 | 6.3 +14% | ---------------|--------|--------|--------|--------|------------| net6,port | | | | | | 1000 | 16.4 | 7.6 | 1.8 | 1.3 | 2.1 +61% | ---------------|--------|--------|--------|--------|------------| port,proto | | | | | [1] | 30000 | 19.6 | 11.6 | 3.9 | 0.3 | 0.5 +66% | ---------------|--------|--------|--------|--------|------------| net6,port,mac | | | | | | 10 | 16.5 | 5.4 | 4.3 | 2.6 | 3.4 +31% | ---------------|--------|--------|--------|--------|------------| net6,port,mac, | | | | | | proto 1000 | 16.5 | 5.7 | 1.9 | 1.0 | 1.4 +40% | ---------------|--------|--------|--------|--------|------------| net,mac | | | | | | 1000 | 19.0 | 8.4 | 3.9 | 1.7 | 2.5 +47% | ---------------'--------'--------'--------'--------'------------' [1] Causes switch of lookup table buckets for 'port', not 'proto', to 4-bit groups ---------------.-----------------------------------.------------. BCM2711 | baselines, Mpps | this patch | 1 thread |___________________________________|____________| 2147MHz | | | | | | 32KiB L1D$ | netdev | hash | rbtree | | | ---------------| hook | no | single | pipapo | pipapo | type entries | drop | ranges | field | 4 bits | bit switch | ---------------|--------|--------|--------|--------|------------| net,port | | | | | | 1000 | 1.63 | 1.37 | 0.87 | 0.61 | 0.70 +17% | ---------------|--------|--------|--------|--------|------------| port,net | | | | | | 100 | 1.64 | 1.36 | 1.02 | 0.78 | 0.81 +4% | ---------------|--------|--------|--------|--------|------------| net6,port | | | | | | 1000 | 1.56 | 1.27 | 0.65 | 0.34 | 0.50 +47% | ---------------|--------|--------|--------|--------|------------| port,proto [2] | | | | | | 10000 | 1.68 | 1.43 | 0.84 | 0.30 | 0.40 +13% | ---------------|--------|--------|--------|--------|------------| net6,port,mac | | | | | | 10 | 1.56 | 1.14 | 1.02 | 0.62 | 0.66 +6% | ---------------|--------|--------|--------|--------|------------| net6,port,mac, | | | | | | proto 1000 | 1.56 | 1.12 | 0.64 | 0.27 | 0.40 +48% | ---------------|--------|--------|--------|--------|------------| net,mac | | | | | | 1000 | 1.63 | 1.26 | 0.87 | 0.41 | 0.53 +29% | ---------------'--------'--------'--------'--------'------------' [2] Using 10000 entries instead of 30000 as it would take way too long for the test script to generate all of them Signed-off-by: -
Stefano Brivio authored
Get rid of all hardcoded assumptions that buckets in lookup tables correspond to four-bit groups, and replace them with appropriate calculations based on a variable group size, now stored in struct field. The group size could now be in principle any divisor of eight. Note, though, that lookup and get functions need an implementation intimately depending on the group size, and the only supported size there, currently, is four bits, which is also the initial and only used size at the moment. While at it, drop 'groups' from struct nft_pipapo: it was never used. Signed-off-by:
-
wenxu authored
This patch add tunnel encap decap action offload in the flowtable offload. Signed-off-by:
wenxu <wenxu@ucloud.cn> Signed-off-by:
-
wenxu authored
This patch support both ipv4 and ipv6 tunnel_id, tunnel_src and tunnel_dst match for flowtable offload Signed-off-by:
-
wenxu authored
Add etfilter flowtable support indr-block setup. It makes flowtable offload vlan and tunnel device. Signed-off-by:
-
wenxu authored
Add nf_flow_table_block_offload_init prepare for the indr block offload patch Signed-off-by:
-
Dan Carpenter authored
These lines were indented wrong so Smatch complained. net/netfilter/xt_IDLETIMER.c:81 idletimer_tg_show() warn: inconsistent indenting Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Jeremy Sowden authored
Name the mask and xor data variables, "mask" and "xor," instead of "d1" and "d2." Signed-off-by:
Jeremy Sowden <jeremy@azazel.net> Signed-off-by:
-
Gustavo A. R. Silva authored
The current codebase makes use of the zero-length array language extension to the C90 standard, but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on. Also, notice that, dynamic memory allocations won't be affected by this change: "Flexible array members have incomplete type, and so the sizeof operator may not be applied. As a quirk of the original implementation of zero-length arrays, sizeof evaluates to zero."[1] Lastly, fix checkpatch.pl warning WARNING: __aligned(size) is preferred over __attribute__((aligned(size))) in net/bridge/netfilter/ebtables.c This issue was found with the help of Coccinelle. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by:Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by:
-
Chen Wandun authored
Fix the following sparse warning: net/netfilter/nft_set_pipapo.c:739:6: warning: symbol 'nft_pipapo_get' was not declared. Should it be static? Fixes: 3c4287f6 ("nf_tables: Add set type for arbitrary concatenation of ranges") Signed-off-by:
Chen Wandun <chenwandun@huawei.com> Acked-by:
-
Li RongQing authored
TEMPLATE_NULLS_VAL is not used after commit 0838aa7f ("netfilter: fix netns dependencies with conntrack templates") PFX is not used after commit 8bee4bad ("netfilter: xt extensions: use pr_<level>") Signed-off-by:
Li RongQing <lirongqing@baidu.com> Signed-off-by:
-
Florian Westphal authored
They do not need to be writeable anymore. v2: remove left-over __read_mostly annotation in set_pipapo.c (Stefano) Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
-
Florian Westphal authored
Placing nftables set support in an extra module is pointless: 1. nf_tables needs dynamic registeration interface for sake of one module 2. nft heavily relies on sets, e.g. even simple rule like "nft ... tcp dport { 80, 443 }" will not work with _SETS=n. IOW, either nftables isn't used or both nf_tables and nf_tables_set modules are needed anyway. With extra module: 307K net/netfilter/nf_tables.ko 79K net/netfilter/nf_tables_set.ko text data bss dec filename 146416 3072 545 150033 nf_tables.ko 35496 1817 0 37313 nf_tables_set.ko This patch: 373K net/netfilter/nf_tables.ko 178563 4049 545 183157 nf_tables.ko Signed-off-by: -
Xin Long authored
Like vxlan and erspan opts, geneve opts should also be supported in nft_tunnel. The difference is geneve RFC (draft-ietf-nvo3-geneve-14) allows a geneve packet to carry multiple geneve opts. So with this patch, nftables/libnftnl would do: # nft add table ip filter # nft add chain ip filter input { type filter hook input priority 0 \; } # nft add tunnel filter geneve_02 { type geneve\; id 2\; \ ip saddr 192.168.1.1\; ip daddr 192.168.1.2\; \ sport 9000\; dport 9001\; dscp 1234\; ttl 64\; flags 1\; \ opts \"1:1:34567890,2:2:12121212,3:3:1212121234567890\"\; } # nft list tunnels table filter table ip filter { tunnel geneve_02 { id 2 ip saddr 192.168.1.1 ip daddr 192.168.1.2 sport 9000 dport 9001 tos 18 ttl 64 flags 1 geneve opts 1:1:34567890,2:2:12121212,3:3:1212121234567890 } } v1->v2: - no changes, just post it separately. Signed-off-by:Xin Long <lucien.xin@gmail.com> Signed-off-by:
-
Manoj Basapathi authored
This is a snapshot of hardidletimer netfilter target. This patch implements a hardidletimer Xtables target that can be used to identify when interfaces have been idle for a certain period of time. Timers are identified by labels and are created when a rule is set with a new label. The rules also take a timeout value (in seconds) as an option. If more than one rule uses the same timer label, the timer will be restarted whenever any of the rules get a hit. One entry for each timer is created in sysfs. This attribute contains the timer remaining for the timer to expire. The attributes are located under the xt_idletimer class: /sys/class/xt_idletimer/timers/<label> When the timer expires, the target module sends a sysfs notification to the userspace, which can then decide what to do (eg. disconnect to save power) Compared to IDLETIMER, HARDIDLETIMER can send notifications when CPU is in suspend too, to notify the timer expiry. v1->v2: Moved all functionality into IDLETIMER module to avoid code duplication per comment from Florian. Signed-off-by:
Manoj Basapathi <manojbm@codeaurora.org> Signed-off-by:
Subash Abhinov Kasiviswanathan <subashab@codeaurora.org> Signed-off-by:
-
Paul Blakey authored
This patch doesn't change any functionality. Signed-off-by:
Paul Blakey <paulb@mellanox.com> Signed-off-by:
-
Alexander Bersenev authored
The ndp32->wLength is two bytes long, so replace cpu_to_le32 with cpu_to_le16. Fixes: 0fa81b30 ("cdc_ncm: Implement the 32-bit version of NCM Transfer Block") Signed-off-by:
Alexander Bersenev <bay@hackerdom.ru> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Paolo Abeni says: ==================== mptcp: simplify mptcp_accept() Currently we allocate the MPTCP master socket at accept time. The above makes mptcp_accept() quite complex, and requires checks is several places for NULL MPTCP master socket. These series simplify the MPTCP accept implementation, moving the master socket allocation at syn-ack time, so that we drop unneeded checks with the follow-up patch. v1 -> v2: - rebased on top of 2398e399 ==================== Signed-off-by:
-
Paolo Abeni authored
After the previous patch subflow->conn is always != NULL and is never changed. We can drop a bunch of now unneeded checks. v1 -> v2: - rebased on top of commit 2398e399 ("mptcp: always include dack if possible.") Signed-off-by:
Paolo Abeni <pabeni@redhat.com> Reviewed-by:
Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by:
-
Paolo Abeni authored
This change moves the mptcp socket allocation from mptcp_accept() to subflow_syn_recv_sock(), so that subflow->conn is now always set for the non fallback scenario. It allows cleaning up a bit mptcp_accept() reducing the additional locking and will allow fourther cleanup in the next patch. Signed-off-by:
-
Dejin Zheng authored
Use devm_platform_ioremap_resource() to simplify code, which contains platform_get_resource and devm_ioremap_resource. Signed-off-by:
Dejin Zheng <zhengdejin5@gmail.com> Signed-off-by:
-
Vladimir Oltean authored
Being a non-physical port, the CPU port does not have an ocelot_port structure, so the ocelot_port_writel call inside the ocelot_port_set_maxlen() function would access data behind a NULL pointer. This is a patch for net-next only, the net tree boots fine, the bug was introduced during the net -> net-next merge. Fixes: 1d343579 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") Fixes: a8015ded ("net: mscc: ocelot: properly account for VLAN header length when setting MRU") Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
-
Hoang Le authored
Calling: tipc_node_link_down()-> - tipc_node_write_unlock()->tipc_mon_peer_down() - tipc_mon_peer_down() just after disabling bearer could be caused kernel oops. Fix this by adding a sanity check to make sure valid memory access. Acked-by:
Jon Maloy <jmaloy@redhat.com> Signed-off-by:
Hoang Le <hoang.h.le@dektech.com.au> Signed-off-by:
-
Hoang Le authored
Checking and returning 'true' boolean is useless as it will be returning at end of function Signed-off-by:
Ying Xue <ying.xue@windriver.com> Acked-by:
-
David S. Miller authored
Jakub Kicinski says: ==================== ethtool: consolidate irq coalescing - part 5 Convert more drivers following the groundwork laid in a recent patch set [1] and continued in [2], [3], [4]. The aim of the effort is to consolidate irq coalescing parameter validation in the core. This set converts further 15 drivers in drivers/net/ethernet. One more conversion sets to come. [1] https://lore.kernel.org/netdev/20200305051542.991898-1-kuba@kernel.org/ [2] https://lore.kernel.org/netdev/20200306010602.1620354-1-kuba@kernel.org/ [3] https://lore.kernel.org/netdev/20200310021512.1861626-1-kuba@kernel.org/ [4] https://lore.kernel.org/netdev/20200311223302.2171564-1-kuba@kernel.org/ ==================== Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver already correctly rejected almost all unsupported parameters (missing sample_rate_interval). As a side effect of these changes the error code for unsupported params changes from EINVAL to EOPNOTSUPP. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. As a side effect of these changes the error code for unsupported params changes from EINVAL to EOPNOTSUPP. The driver was missing a check for rate_sample_interval. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver correctly rejects all unsupported parameters, no functional changes. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver correctly rejects all unsupported parameters, no functional changes. Signed-off-by:
-
Jakub Kicinski authored
Set ethtool_ops->supported_coalesce_params to let the core reject unsupported coalescing parameters. This driver did not previously reject unsupported parameters. Signed-off-by:
Matteo Croce <mcroce@redhat.com> Signed-off-by:
-
