- 22 Apr, 2024 19 commits
-
-
Eric Dumazet authored
While investigating TCP performance, I found that TCP would sometimes send big skbs followed by a single MSS skb, in a 'locked' pattern. For instance, BIG TCP is enabled, MSS is set to have 4096 bytes of payload per segment. gso_max_size is set to 181000. This means that an optimal TCP packet size should contain 44 * 4096 = 180224 bytes of payload, However, I was seeing packets sizes interleaved in this pattern: 172032, 8192, 172032, 8192, 172032, 8192, <repeat> tcp_tso_should_defer() heuristic is defeated, because after a split of a packet in write queue for whatever reason (this might be a too small CWND or a small enough pacing_rate), the leftover packet in the queue is smaller than the optimal size. It is time to try to make 'leftover packets' bigger so that tcp_tso_should_defer() can give its full potential. After this patch, we can see the following output: 14:13:34.009273 IP6 sender > receiver: Flags [P.], seq 4048380:4098360, ack 1, win 256, options [nop,nop,TS val 3425678144 ecr 1561784500], length 49980 14:13:34.010272 IP6 sender > receiver: Flags [P.], seq 4098360:4148340, ack 1, win 256, options [nop,nop,TS val 3425678145 ecr 1561784501], length 49980 14:13:34.011271 IP6 sender > receiver: Flags [P.], seq 4148340:4198320, ack 1, win 256, options [nop,nop,TS val 3425678146 ecr 1561784502], length 49980 14:13:34.012271 IP6 sender > receiver: Flags [P.], seq 4198320:4248300, ack 1, win 256, options [nop,nop,TS val 3425678147 ecr 1561784503], length 49980 14:13:34.013272 IP6 sender > receiver: Flags [P.], seq 4248300:4298280, ack 1, win 256, options [nop,nop,TS val 3425678148 ecr 1561784504], length 49980 14:13:34.014271 IP6 sender > receiver: Flags [P.], seq 4298280:4348260, ack 1, win 256, options [nop,nop,TS val 3425678149 ecr 1561784505], length 49980 14:13:34.015272 IP6 sender > receiver: Flags [P.], seq 4348260:4398240, ack 1, win 256, options [nop,nop,TS val 3425678150 ecr 1561784506], length 49980 14:13:34.016270 IP6 sender > receiver: Flags [P.], seq 4398240:4448220, ack 1, win 256, options [nop,nop,TS val 3425678151 ecr 1561784507], length 49980 14:13:34.017269 IP6 sender > receiver: Flags [P.], seq 4448220:4498200, ack 1, win 256, options [nop,nop,TS val 3425678152 ecr 1561784508], length 49980 14:13:34.018276 IP6 sender > receiver: Flags [P.], seq 4498200:4548180, ack 1, win 256, options [nop,nop,TS val 3425678153 ecr 1561784509], length 49980 14:13:34.019259 IP6 sender > receiver: Flags [P.], seq 4548180:4598160, ack 1, win 256, options [nop,nop,TS val 3425678154 ecr 1561784510], length 49980 With 200 concurrent flows on a 100Gbit NIC, we can see a reduction of TSO packets (and ACK packets) of about 30 %. Signed-off-by:
Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20240418214600.1291486-4-edumazet@google.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Eric Dumazet authored
tcp_write_xmit() calls tcp_init_tso_segs() to set gso_size and gso_segs on the packet. tcp_init_tso_segs() requires the stack to maintain an up to date tcp_skb_pcount(), and this makes sense for packets in rtx queue. Not so much for packets still in the write queue. In the following patch, we don't want to deal with tcp_skb_pcount() when moving payload from 2nd skb to 1st skb in the write queue. Signed-off-by:
-
Eric Dumazet authored
tcp_cwnd_test() has a special handing for the last packet in the write queue if it is smaller than one MSS and has the FIN flag. This is in violation of TCP RFC, and seems quite dubious. This packet can be sent only if the current CWND is bigger than the number of packets in flight. Making tcp_cwnd_test() result independent of the first skb in the write queue is needed for the last patch of the series. Signed-off-by:
-
Jakub Kicinski authored
Tariq Toukan says: ==================== mlx5e per-queue coalescing This patchset adds ethtool per-queue coalescing support for the mlx5e driver. The series introduce some changes needed as preparations for the final patch which adds the support and implements the callbacks. Main changes: - DIM code movements into its own header file. - Switch to dynamic allocation of the DIM struct in the RQs/SQs. - Allow coalescing config change without channels reset when possible. ==================== Link: https://lore.kernel.org/r/20240419080445.417574-1-tariqt@nvidia.comSigned-off-by:
-
Rahul Rameshbabu authored
Use mlx5 on-the-fly coalescing configuration support to enable individual channel configuration. Co-developed-by:
Nabil S. Alramli <dev@nalramli.com> Signed-off-by:
Joe Damato <jdamato@fastly.com> Signed-off-by:
Rahul Rameshbabu <rrameshbabu@nvidia.com> Reviewed-by:
Saeed Mahameed <saeedm@nvidia.com> Signed-off-by:
Tariq Toukan <tariqt@nvidia.com> Link: https://lore.kernel.org/r/20240419080445.417574-6-tariqt@nvidia.comSigned-off-by:
-
Rahul Rameshbabu authored
When CQE mode or DIM state is changed, gracefully reconfigure channels to handle new configuration. Previously, would create new channels that would reflect the changes rather than update the original channels. Co-developed-by:
-
Rahul Rameshbabu authored
Make it possible for the DIM structure to be torn down while an SQ or RQ is still active. Changing the CQ period mode is an example where the previous sampling done with the DIM structure would need to be invalidated. Co-developed-by:
-
Rahul Rameshbabu authored
Use core DIM CQ period mode enum values for the CQ parameter for the period mode. Translate the value to the specific mlx5 device constant for the selected period mode when creating a CQ. Avoid needing to translate mlx5 device constants to DIM constants for core DIM functionality. Co-developed-by:
-
Rahul Rameshbabu authored
Create a header specifically for DIM-related declarations. Move existing DIM-specific functionality from en.h. Future DIM-related functionality will be declared in en/dim.h in subsequent patches. Co-developed-by:
-
Jakub Kicinski authored
Pawel Dembicki says: ==================== net: dsa: vsc73xx: convert to PHYLINK and do some cleanup This patch series is a result of splitting a larger patch series [0], where some parts needed to be refactored. The first patch switches from a poll loop to read_poll_timeout. The second patch is a simple conversion to phylink because adjust_link won't work anymore. The third patch is preparation for future use. Using the "phy_interface_mode_is_rgmii" macro allows for the proper recognition of all RGMII modes. Patches 4-5 involve some cleanup: The fourth patch introduces a definition with the maximum number of ports to avoid using magic numbers. The next one fills in documentation. [0] https://patchwork.kernel.org/project/netdevbpf/list/?series=841034&state=%2A&archive=both ==================== Link: https://lore.kernel.org/r/20240417205048.3542839-1-paweldembicki@gmail.comSigned-off-by:
-
Pawel Dembicki authored
This commit adds updates to the documentation describing the structures used in vsc73xx. This will help prevent kdoc-related issues in the future. Reviewed-by:
Linus Walleij <linus.walleij@linaro.org> Reviewed-by:
Florian Fainelli <florian.fainelli@broadcom.com> Reviewed-by:
Vladimir Oltean <olteanv@gmail.com> Signed-off-by:
Pawel Dembicki <paweldembicki@gmail.com> Link: https://lore.kernel.org/r/20240417205048.3542839-6-paweldembicki@gmail.comSigned-off-by:
-
Pawel Dembicki authored
This patch introduces a new define: VSC73XX_MAX_NUM_PORTS, which can be used in the future instead of a hardcoded value. Currently, the only hardcoded value is vsc->ds->num_ports. It is being replaced with the new define. Suggested-by:
-
Pawel Dembicki authored
It's preparation for future use. At this moment, the RGMII port is used only for a connection to the MAC interface, but in the future, someone could connect a PHY to it. Using the "phy_interface_mode_is_rgmii" macro allows for the proper recognition of all RGMII modes. Suggested-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by:
-
Pawel Dembicki authored
This patch replaces the adjust_link api with the phylink apis that provide equivalent functionality. The remaining functionality from the adjust_link is now covered in the mac_link_* and mac_config from phylink_mac_ops structure. Removes: .adjust_link Adds phylink_mac_ops structure: .mac_config .mac_link_up .mac_link_down Signed-off-by:
-
Pawel Dembicki authored
Switch the delay loop during the Arbiter empty check from vsc73xx_adjust_link() to use read_poll_timeout(). Functionally, one msleep() call is eliminated at the end of the loop in the timeout case. As Russell King suggested: "This [change] avoids the issue that on the last iteration, the code reads the register, tests it, finds the condition that's being waiting for is false, _then_ waits and end up printing the error message - that last wait is rather useless, and as the arbiter state isn't checked after waiting, it could be that we had success during the last wait." Suggested-by:
Russell King <linux@armlinux.org.uk> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Reviewed-by:
-
Eric Dumazet authored
After commit 1eeb5043 ("tcp/dccp: do not care about families in inet_twsk_purge()") tcp_twsk_purge() is no longer potentially called from a module. Signed-off-by:
Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Geetha sowjanya authored
Support offloading of skbedit mark action. For example, to mark with 0x0008, with dest ip 60.60.60.2 on eth2 interface: # tc qdisc add dev eth2 ingress # tc filter add dev eth2 ingress protocol ip flower \ dst_ip 60.60.60.2 action skbedit mark 0x0008 skip_sw Signed-off-by:Geetha sowjanya <gakula@marvell.com> Reviewed-by:
Jacob Keller <jacob.e.keller@intel.com> Reviewed-by:
Simon Horman <horms@kernel.org> Signed-off-by:
-
Julien Panis authored
When an ethX port is disabled in the device tree, an error is returned by xdp_rxq_info_reg() function while transitioning the CPSW device to the up state. The message 'Missing net_device from driver' is output. This patch fixes the issue by registering xdp_rxq info only if ethX port is enabled (i.e. ndev pointer is not NULL). Fixes: 8acacc40 ("net: ethernet: ti: am65-cpsw: Add minimal XDP support") Link: https://lore.kernel.org/all/260d258f-87a1-4aac-8883-aab4746b32d8@ti.com/Reported-by:
Siddharth Vadapalli <s-vadapalli@ti.com> Closes: https://gist.github.com/Siddharth-Vadapalli-at-TI/5ed0e436606001c247a7da664f75edeeSigned-off-by:
Julien Panis <jpanis@baylibre.com> Signed-off-by:
-
Thomas Weißschuh authored
To be able to constify instances of struct ctl_tables it is necessary to remove ways through which non-const versions are exposed from the sysctl core. One of these is the ctl_table_arg member of struct ctl_table_header. Constify this reference as a prerequisite for the full constification of struct ctl_table instances. No functional change. Signed-off-by:
Thomas Weißschuh <linux@weissschuh.net> Reviewed-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
- 20 Apr, 2024 13 commits
-
-
Jakub Kicinski authored
Florian Westphal says: ==================== testing: make netfilter selftests functional in vng environment This is the second batch of the netfilter selftest move. Changes since v1: - makefile and kernel config are updated to have all required features - fix makefile with missing bits to make kselftest-install work - test it via vng as per https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style (Thanks Jakub!) - squash a few fixes, e.g. nft_queue.sh v1 had a race w. NFNETLINK_QUEUE=m - add a settings file with 8m timeout, for nft_concat_range.sh sake. That script can be sped up a bit, I think, but its not contained in this batch yet. - toss the first two bogus rebase artifacts (Matthieu Baerts) scripts are moved to lib.sh infra. This allows to use busywait helper and ditch various 'sleep 2' all over the place. Tested on Fedora 39: vng --build --config tools/testing/selftests/net/netfilter/config make -C tools/testing/selftests/ TARGETS=net/netfilter vng -v --run . --user root --cpus 2 -- \ make -C tools/testing/selftests TARGETS=net/netfilter run_tests ... all tests pass except nft_audit.sh which SKIPs due to nft version mismatch (Fedora is on nft 1.0.7 which lacks reset keyword support). Missing/WIP bits: - speed up nf_concat_range.sh test - extend flowtable selftest - shellcheck fixups for remaining scripts ==================== Link: https://lore.kernel.org/r/20240418152744.15105-1-fw@strlen.deSigned-off-by:
-
Florian Westphal authored
Jakub reports the Makefile missed a few updates to make kselftest-install work for the netfilter tests and points out that config file lacks many dependencies such as VETH support. The settings file (timeout 8m) is added for nft_concat_range.sh script which can take several minutes to complete. Fixes: 3f189349 ("selftests: netfilter: move to net subdir") Reported-by:
Florian Westphal <fw@strlen.de> Link: https://lore.kernel.org/r/20240418152744.15105-13-fw@strlen.deSigned-off-by:
-
Florian Westphal authored
This testcase doesn't work if auditd is running, audit_logread will not receive any data in that case. Add a nftables feature test for the reset keyword and skip this test if that fails. While at it, do a few minor shellcheck cleanups. Signed-off-by:
-
Florian Westphal authored
shellcheck complains about missing "", so add those. Signed-off-by:
-
Florian Westphal authored
no functional change intended. Signed-off-by:
-
Florian Westphal authored
No functional change intended. Signed-off-by:
-
Florian Westphal authored
While at it: No need for iperf here, use socat. This also reduces the script runtime. Signed-off-by:
-
Florian Westphal authored
no functional change intended. Signed-off-by:
-
Florian Westphal authored
Intentional changes: - Use socat instead of netcat - Use a temporary file instead of pipe, else packets do not match "-m string" rules, multiple writes to the pipe cause multiple packets, but this needs only one to work. Signed-off-by:
-
Florian Westphal authored
Also do shellcheck cleanups here, no functional changes intended. When running tests via vng tool, the packetpath insertion test fails: dd: failed to open '/dev/stdout': Device or resource busy Just omit 'of=' and this will work as intended. Signed-off-by:
-
Florian Westphal authored
use checktool helper where applicable. Signed-off-by:
-
Florian Westphal authored
No functional change intended. Disable frequent shellcheck warnings wrt. "unreachable" code, those helpers get called indirectly from busywait helper. Signed-off-by:
-
Florian Westphal authored
- switch to socat, like other tests - use buswait helper to test once listener netns is ready - do not generate multiple input test files, only generate one and use cleanup hook to remove it, like other temporary files. Signed-off-by:
-
- 19 Apr, 2024 8 commits
-
-
David S. Miller authored
Eric Dumazet says: ==================== neighbour: convert neigh_dump_info() to RCU Remove RTNL requirement for "ip neighbour show" command. ==================== Signed-off-by: -
Eric Dumazet authored
neigh_dump_table() is already relying on RCU protection. pneigh_dump_table() is using its own protection (tbl->lock) Signed-off-by:
-
Eric Dumazet authored
Change neigh_dump_table() and pneigh_dump_table() to either return 0 or -EMSGSIZE if not enough space was available in the skb. Then neigh_dump_info() can do the same. This allows NLMSG_DONE to be appended to the current skb at the end of a dump, saving a couple of recvmsg() system calls. Signed-off-by:
-
Eric Dumazet authored
In order to remove RTNL protection from neightbl_dump_info() and neigh_dump_info() later, we need to add RCU protection to neigh_tables[]. Signed-off-by:
-
Russell King (Oracle) authored
The kernel build bot identified the following mistake in the recently merged 860a9bed ("net: dsa: xrs700x: provide own phylink MAC operations") patch: drivers/net/dsa/xrs700x/xrs700x.c:714:37: warning: 'xrs700x_phylink_mac_ops' defined but not used [-Wunused-const-variable=] 714 | static const struct phylink_mac_ops xrs700x_phylink_mac_ops = { | ^~~~~~~~~~~~~~~~~~~~~~~ Fix the omitted assignment of ds->phylink_mac_ops. Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
-
David S. Miller authored
Jason Xing says: ==================== locklessly protect left members in struct rps_dev_flow From: Jason Xing <kernelxing@tencent.com> Since Eric did a more complicated locklessly change to last_qtail member[1] in struct rps_dev_flow, the left members are easier to change as the same. One thing important I would like to share by qooting Eric: "rflow is located in rxqueue->rps_flow_table, it is thus private to current thread. Only one cpu can service an RX queue at a time." So we only pay attention to the reader in the rps_may_expire_flow() and writer in the set_rps_cpu(). They are in the two different contexts. [1]: https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/?id=3b4cf29bdab v3 Link: https://lore.kernel.org/all/20240417062721.45652-1-kerneljasonxing@gmail.com/ 1. adjust the protection in a right way (Eric) v2 1. fix passing wrong type qtail. ==================== Signed-off-by:
-
Jason Xing authored
This is the last member in struct rps_dev_flow which should be protected locklessly. So finish it. Signed-off-by:
Jason Xing <kernelxing@tencent.com> Reviewed-by:
-
Jason Xing authored
As we can see, rflow->filter can be written/read concurrently, so lockless access is needed. Signed-off-by:
-
