- Sep 21, 2022
-
-
Edward Cree authored
There's no clear explanation of what VF Representors are for, their semantics, etc., outside of vendor docs and random conference slides. Add a document explaining Representors and defining what drivers that implement them are expected to do. Signed-off-by:
Edward Cree <ecree.xilinx@gmail.com> Reviewed-by:
Bagas Sanjaya <bagasdotme@gmail.com> Link: https://lore.kernel.org/r/20220905135557.39233-1-ecree@xilinx.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
- Sep 20, 2022
-
-
Kuniyuki Iwashima authored
The more sockets we have in the hash table, the longer we spend looking up the socket. While running a number of small workloads on the same host, they penalise each other and cause performance degradation. The root cause might be a single workload that consumes much more resources than the others. It often happens on a cloud service where different workloads share the same computing resource. On EC2 c5.24xlarge instance (196 GiB memory and 524288 (1Mi / 2) ehash entries), after running iperf3 in different netns, creating 24Mi sockets without data transfer in the root netns causes about 10% performance regression for the iperf3's connection. thash_entries sockets length Gbps 524288 1 1 50.7 24Mi 48 45.1 It is basically related to the length of the list of each hash bucket. For testing purposes to see how performance drops along the length, I set 131072 (1Mi / 8) to thash_entries, and here's the result. thash_entries sockets length Gbps 131072 1 1 50.7 1Mi 8 49.9 2Mi 16 48.9 4Mi 32 47.3 8Mi 64 44.6 16Mi 128 40.6 24Mi 192 36.3 32Mi 256 32.5 40Mi 320 27.0 48Mi 384 25.0 To resolve the socket lookup degradation, we introduce an optional per-netns hash table for TCP, but it's just ehash, and we still share the global bhash, bhash2 and lhash2. With a smaller ehash, we can look up non-listener sockets faster and isolate such noisy neighbours. In addition, we can reduce lock contention. We can control the ehash size by a new sysctl knob. However, depending on workloads, it will require very sensitive tuning, so we disable the feature by default (net.ipv4.tcp_child_ehash_entries == 0). Moreover, we can fall back to using the global ehash in case we fail to allocate enough memory for a new ehash. The maximum size is 16Mi, which is large enough that even if we have 48Mi sockets, the average list length is 3, and regression would be less than 1%. We can check the current ehash size by another read-only sysctl knob, net.ipv4.tcp_ehash_entries. A negative value means the netns shares the global ehash (per-netns ehash is disabled or failed to allocate memory). # dmesg | cut -d ' ' -f 5- | grep "established hash" TCP established hash table entries: 524288 (order: 10, 4194304 bytes, vmalloc hugepage) # sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = 524288 # can be changed by thash_entries # sysctl net.ipv4.tcp_child_ehash_entries net.ipv4.tcp_child_ehash_entries = 0 # disabled by default # ip netns add test1 # ip netns exec test1 sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = -524288 # share the global ehash # sysctl -w net.ipv4.tcp_child_ehash_entries=100 net.ipv4.tcp_child_ehash_entries = 100 # ip netns add test2 # ip netns exec test2 sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = 128 # own a per-netns ehash with 2^n buckets When more than two processes in the same netns create per-netns ehash concurrently with different sizes, we need to guarantee the size in one of the following ways: 1) Share the global ehash and create per-netns ehash First, unshare() with tcp_child_ehash_entries==0. It creates dedicated netns sysctl knobs where we can safely change tcp_child_ehash_entries and clone()/unshare() to create a per-netns ehash. 2) Control write on sysctl by BPF We can use BPF_PROG_TYPE_CGROUP_SYSCTL to allow/deny read/write on sysctl knobs. Note that the global ehash allocated at the boot time is spread over available NUMA nodes, but inet_pernet_hashinfo_alloc() will allocate pages for each per-netns ehash depending on the current process's NUMA policy. By default, the allocation is done in the local node only, so the per-netns hash table could fully reside on a random node. Thus, depending on the NUMA policy the netns is created with and the CPU the current thread is running on, we could see some performance differences for highly optimised networking applications. Note also that the default values of two sysctl knobs depend on the ehash size and should be tuned carefully: tcp_max_tw_buckets : tcp_child_ehash_entries / 2 tcp_max_syn_backlog : max(128, tcp_child_ehash_entries / 128) As a bonus, we can dismantle netns faster. Currently, while destroying netns, we call inet_twsk_purge(), which walks through the global ehash. It can be potentially big because it can have many sockets other than TIME_WAIT in all netns. Splitting ehash changes that situation, where it's only necessary for inet_twsk_purge() to clean up TIME_WAIT sockets in each netns. With regard to this, we do not free the per-netns ehash in inet_twsk_kill() to avoid UAF while iterating the per-netns ehash in inet_twsk_purge(). Instead, we do it in tcp_sk_exit_batch() after calling tcp_twsk_purge() to keep it protocol-family-independent. In the future, we could optimise ehash lookup/iteration further by removing netns comparison for the per-netns ehash. Signed-off-by:Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
Vladimir Oltean authored
Replace the free-form description of device tree bindings for VSC9959 and VSC9953 with a YAML formatted dt-schema description. This contains more or less the same information, but reworded to be a bit more succint. Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by:
Maxim Kochetkov <fido_max@inbox.ru> Reviewed-by:
Rob Herring <robh@kernel.org> Link: https://lore.kernel.org/r/20220913125806.524314-1-vladimir.oltean@nxp.com Signed-off-by:
-
Alexandru Tachici authored
Add bindings for the ADIN1110/2111 MAC-PHY/SWITCH. Reviewed-by:
Alexandru Tachici <alexandru.tachici@analog.com> Signed-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Vladimir Oltean authored
DSA now supports multiple CPU ports, explain the use cases that are covered, the new UAPI, the permitted degrees of freedom, the driver API, and remove some old "hanging fruits". Signed-off-by:
-
Vladimir Oltean authored
This is not used by the DSA dt-binding, so remove it from all examples. Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Acked-by:
Arınç ÜNAL <arinc.unal@arinc9.com> Acked-by:
-
Vladimir Oltean authored
The common dsa-port.yaml does this (and more) since commit 2ec2fb83 ("dt-bindings: net: dsa: make phylink bindings required for CPU/DSA ports"). Signed-off-by:
-
Vladimir Oltean authored
The fact that some DSA device trees use 'label = "cpu"' for the CPU port is nothing but blind cargo cult copying. The 'label' property was never part of the DSA DT bindings for anything except the user ports, where it provided a hint as to what name the created netdevs should use. DSA does use the "cpu" port label to identify a CPU port in dsa_port_parse(), but this is only for non-OF code paths (platform data). The proper way to identify a CPU port is to look at whether the 'ethernet' phandle is present. Signed-off-by:
-
Geert Uytterhoeven authored
Document support for the Renesas Ethernet AVB (EtherAVB-IF) block in the Renesas R-Car V4H (R8A779G0) SoC. Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Acked-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by:
Wolfram Sang <wsa+renesas@sang-engineering.com> Reviewed-by:
Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by:
-
Geert Uytterhoeven authored
Despite the name, R-Car V3U is the first member of the R-Car Gen4 family. Hence move its compatible value to the R-Car Gen4 section. Signed-off-by:
-
Wei Fang authored
Add fsl,s32v234-fec to compatible property to support s32v234 platform. Signed-off-by:
Wei Fang <wei.fang@nxp.com> Acked-by:
-
- Sep 09, 2022
-
-
Colin Foster authored
Add devicetree bindings for SPI-controlled Ocelot chips, specifically the VSC7512. Signed-off-by:
Colin Foster <colin.foster@in-advantage.com> Reviewed-by:
Lee Jones <lee@kernel.org> Link: https://lore.kernel.org/r/20220905162132.2943088-8-colin.foster@in-advantage.com
-
- Sep 06, 2022
-
-
Dario Binacchi authored
The Amarula contact info email address is wrong, so fix it up to use the correct one. Signed-off-by:
Dario Binacchi <dario.binacchi@amarulasolutions.com> Link: https://lore.kernel.org/all/20220828134442.794990-1-dario.binacchi@amarulasolutions.com Signed-off-by:
Marc Kleine-Budde <mkl@pengutronix.de>
-
Krzysztof Kozlowski authored
reg-io-width is a standard property, so no need for defining its type Signed-off-by:
-
- Sep 05, 2022
-
-
Biju Das authored
Document RZ/N1 power-domains support. Also update the example with power-domains property. Signed-off-by:
Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by:
-
Sean Anderson authored
Add 1000BASE-KX interface mode. This 1G backplane ethernet as described in clause 70. Clause 73 autonegotiation is mandatory, and only full duplex operation is supported. Although at the PMA level this interface mode is identical to 1000BASE-X, it uses a different form of in-band autonegation. This justifies a separate interface mode, since the interface mode (along with the MLO_AN_* autonegotiation mode) sets the type of autonegotiation which will be used on a link. This results in more than just electrical differences between the link modes. With regard to 1000BASE-X, 1000BASE-KX holds a similar position to SGMII: same signaling, but different autonegotiation. PCS drivers (which typically handle in-band autonegotiation) may only support 1000BASE-X, and not 1000BASE-KX. Similarly, the phy mode is used to configure serdes phys with phy_set_mode_ext. Due to the different electrical standards (SFI or XFI vs Clause 70), they will likely want to use different configuration. Adding a phy interface mode for 1000BASE-KX helps simplify configuration in these areas. Signed-off-by:
Sean Anderson <sean.anderson@seco.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Maxime Chevallier authored
Some implementations of the TSE have their PCS as an external bloc, exposed at its own register range. Document this, and add a new example showing a case using the pcs and the new phylink conversion to connect an sfp port to a TSE mac. Signed-off-by:
Maxime Chevallier <maxime.chevallier@bootlin.com> Signed-off-by:
-
Maxime Chevallier authored
Convert the bindings for the Altera Triple-Speed Ethernet to yaml. Signed-off-by:
-
- Sep 02, 2022
-
-
Ming Lei authored
Add documentation for ublk subsystem. It was supposed to be documented when merging the driver, but missing at that time. Cc: Bagas Sanjaya <bagasdotme@gmail.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Richard W.M. Jones <rjones@redhat.com> Cc: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com> Reviewed-by:
Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by:
ZiyangZhang <ZiyangZhang@linux.alibaba.com> Signed-off-by:
Ming Lei <ming.lei@redhat.com> [axboe: correct MAINTAINERS addition] Signed-off-by:
Jens Axboe <axboe@kernel.dk>
-
Michael Walle authored
Make the reset line optional. It turns out, there is no dedicated reset for the switch. Instead, the reset which was used up until now, was kind of a global reset. This is now handled elsewhere, thus don't require a reset. Signed-off-by:
Michael Walle <michael@walle.cc> Acked-by:
-
- Sep 01, 2022
-
-
David Howells authored
Remove rxrpc_get_reply_time() as that is no longer used now that the call issue time is used instead of the reply time. Signed-off-by:David Howells <dhowells@redhat.com>
-
Sebastian Reichel authored
Add compatible string for RK3588 gmac, which is similar to the RK3568 one, but needs another syscon device for clock selection. Acked-by:
Sebastian Reichel <sebastian.reichel@collabora.com> Signed-off-by:
-
Eric Dumazet authored
Because per host rate limiting has been proven problematic (side channel attacks can be based on it), per host rate limiting of challenge acks ideally should be per netns and turned off by default. This is a long due followup of following commits: 083ae308 ("tcp: enable per-socket rate limiting of all 'challenge acks'") f2b2c582 ("tcp: mitigate ACK loops for connections as tcp_sock") 75ff39cc ("tcp: make challenge acks less predictable") Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
-
- Aug 31, 2022
-
-
Randy Dunlap authored
Change occurrences of "it's" that are possessive to "its" so that they don't read as "it is". Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Eric Dumazet <edumazet@google.com> Cc: Paolo Abeni <pabeni@redhat.com> Cc: Jiri Pirko <jiri@nvidia.com> Link: https://lore.kernel.org/r/20220829235414.17110-1-rdunlap@infradead.org Signed-off-by:
-
Fernando Fernandez Mancera authored
tlb_dynamic_lb bonding option is compatible with balance-tlb and balance-alb modes. In order to be consistent with other option documentation, it should mention both modes not only balance-tlb. Signed-off-by:
Fernando Fernandez Mancera <ffmancera@riseup.net> Acked-by:
Jay Vosburgh <jay.vosburgh@canonical.com> Link: https://lore.kernel.org/r/20220826154738.4039-1-ffmancera@riseup.net Signed-off-by:
-
Arınç ÜNAL authored
Update the description of the binding. - Describe the switches, which SoCs they are in, or if they are standalone. - Explain the various ways of configuring MT7530's port 5. - Remove phy-mode = "rgmii-txid" from description. Same code path is followed for delayed rgmii and rgmii phy-mode on mtk_eth_soc.c. Signed-off-by:
-
Arınç ÜNAL authored
Define acceptable phy-mode values for the CPU ports of mt7530 and mt7531 switches. Remove relevant information from the description of the binding. Signed-off-by:
-
Arınç ÜNAL authored
Update the examples on the binding. - Add examples which include a wide variation of configurations. - Make example comments YAML comment instead of DT binding comment. - Add interrupt controller to the examples. Include header file for interrupt. - Change reset line for MT7621 examples. - Pretty formatting for the examples. - Change switch reg to 0. - Change port labels to fit the example, change port 4 label to wan. Signed-off-by:
-
Arınç ÜNAL authored
- Add description for reset-gpios. - Invalidate reset-gpios if mediatek,mcm is used. We cannot use multiple reset lines at the same time. - Invalidate mediatek,mcm if the compatible device is mediatek,mt7531. There is no multi-chip module version of mediatek,mt7531. - Require mediatek,mcm for mediatek,mt7621 as the compatible string is only used for the multi-chip module version of MT7530. Signed-off-by:
-
Arınç ÜNAL authored
Fix the description of mediatek,mcm. mediatek,mcm is not used on MT7623NI. Signed-off-by:
-
Arınç ÜNAL authored
Make trivial changes on the binding. - Update title to include MT7531 switch. - Add me as a maintainer. List maintainers in alphabetical order by first name. - Add description to compatible strings. - Stretch descriptions up to the 80 character limit. - Remove lists for single items. - Remove requiring reg as it's already required by dsa-port.yaml. - Define acceptable reg values for the CPU ports. - Remove quotes from $ref: "dsa.yaml#". Signed-off-by:
-
- Aug 30, 2022
-
-
Mengyuan Lou authored
Add build options and guidance doc. Initialize pci device access for Wangxun Gigabit Ethernet devices. Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
Mengyuan Lou <mengyuanlou@net-swift.com> Link: https://lore.kernel.org/r/20220826034609.51854-1-mengyuanlou@net-swift.com Signed-off-by:
-
Jakub Kicinski authored
There is currently no way to report via extack in a structured way that an attribute is missing. This leads to families resorting to string messages. Add a pair of attributes - @offset and @type for machine-readable way of reporting missing attributes. The @offset points to the nest which should have contained the attribute, @type is the expected nla_type. The offset will be skipped if the attribute is missing at the message level rather than inside a nest. User space should be able to figure out which attribute enum (AKA attribute space AKA attribute set) the nest pointed to by @offset is using. Reviewed-by:
Johannes Berg <johannes@sipsolutions.net> Signed-off-by:
-
- Aug 29, 2022
-
-
Greg Tulli authored
Add a new iforce_device entry to support the Boeder Force Feedback Wheel device. Signed-off-by:
Greg Tulli <greg.iforce@gmail.com> Link: https://lore.kernel.org/r/3256420-c8ac-31b-8499-3c488a9880fd@gmail.com Signed-off-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com>
-
- Aug 26, 2022
-
-
Menglong Dong authored
Stephen Rothwell reported htmldocs warning when merging net-next: Documentation/networking/kapi:26: net/core/skbuff.c:780: WARNING: Error in declarator or parameters Invalid C declaration: Expecting "(" in parameters. [error at 19] void __fix_address kfree_skb_reason (struct sk_buff *skb, enum skb_drop_reason reason) -------------------^ Add __fix_address keyword to c_id_attributes array in conf.py to fix the warning. Link: https://lore.kernel.org/linux-next/20220825154105.534d78ab@canb.auug.org.au/ Reported-by:Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
Menglong Dong <imagedong@tencent.com> Tested-by:
Jonathan Corbet <corbet@lwn.net>
-
Mikulas Patocka authored
There are several places in the kernel where wait_on_bit is not followed by a memory barrier (for example, in drivers/md/dm-bufio.c:new_read). On architectures with weak memory ordering, it may happen that memory accesses that follow wait_on_bit are reordered before wait_on_bit and they may return invalid data. Fix this class of bugs by introducing a new function "test_bit_acquire" that works like test_bit, but has acquire memory ordering semantics. Signed-off-by:
Mikulas Patocka <mpatocka@redhat.com> Acked-by:
Will Deacon <will@kernel.org> Cc: stable@vger.kernel.org Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Rob Herring authored
In order to ensure only documented properties are present, node schemas must have unevaluatedProperties or additionalProperties set to false (typically). Add missing properties/$refs as exposed by this addition. Signed-off-by:
-
- Aug 25, 2022
-
-
Kairui Song authored
The workflow example code is not working since it got the file names wrong. So fix this. Fixes: b1840272 ("Docs/admin-guide/mm/damon/usage: document DAMON sysfs interface") Reviewed-by:
SeongJae Park <sj@kernel.org> Signed-off-by:
Kairui Song <kasong@tencent.com> Link: https://lore.kernel.org/r/20220823114053.53305-1-ryncsn@gmail.com Signed-off-by:
-
Lukas Bulwahn authored
A quick 'grep "5\.x" . -R' on Documentation shows that README.rst, 2.Process.rst and applying-patches.rst all mention the version number "5.x" for kernel releases. As the next release will be version 6.0, updating the version number to 6.x in README.rst seems reasonable. The description in 2.Process.rst is just a description of recent kernel releases, it was last updated in the beginning of 2020, and can be revisited at any time on a regular basis, independent of changing the version number from 5 to 6. So, there is no need to update this document now when transitioning from 5.x to 6.x numbering. The document applying-patches.rst is probably obsolete for most users anyway, a reader will sufficiently well understand the steps, even it mentions version 5 rather than version 6. So, do not update that to a version 6.x numbering scheme. Update version number from 5.x to 6.x in README.rst only. Signed-off-by:
Lukas Bulwahn <lukas.bulwahn@gmail.com> Link: https://lore.kernel.org/r/20220824080836.23087-1-lukas.bulwahn@gmail.com Signed-off-by:
-
Salvatore Bonaccorso authored
While reporting for the AMD retbleed vulnerability was added in 6b80b59b ("x86/bugs: Report AMD retbleed vulnerability") the new sysfs file was not mentioned so far in the ABI documentation for sysfs-devices-system-cpu. Fix that. Fixes: 6b80b59b ("x86/bugs: Report AMD retbleed vulnerability") Signed-off-by:
Salvatore Bonaccorso <carnil@debian.org> Signed-off-by:
Borislav Petkov <bp@suse.de> Link: https://lore.kernel.org/r/20220801091529.325327-1-carnil@debian.org
-
