)]}' { "log": [ { "commit": "f2b6cfda76d2119871e10fa01ecdc7178401ef22", "tree": "14ef50253c092134b568af026e777f536f5a09f8", "parents": [ "05b953a550625dc05652e15aeb9af34771c96325" ], "author": { "name": "Leon Romanovsky", "email": "leonro@nvidia.com", "time": "Sun Feb 19 11:09:10 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:52:56 2023 -0800" }, "message": "net/mlx5e: Align IPsec ASO result memory to be as required by hardware\n\nHardware requires an alignment to 64 bytes to return ASO data. Missing\nthis alignment caused to unpredictable results while ASO events were\ngenerated.\n\nFixes: 8518d05b8f9a (\"net/mlx5e: Create Advanced Steering Operation object for IPsec\")\nReported-by: Emeel Hakim \u003cehakim@nvidia.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleonro@nvidia.com\u003e\nLink: https://lore.kernel.org/r/de0302c572b90c9224a72868d4e0d657b6313c4b.1676797613.git.leon@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "05b953a550625dc05652e15aeb9af34771c96325", "tree": "208967e73cc62f25f0fe6d4bb433d53e0871c8c1", "parents": [ "981f40458e7a6ffbdff1a09ece6099b3b49d08a5", "993fd9bd656a082c9b713171622c70f72f0af59f" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:52:50 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:52:50 2023 -0800" }, "message": "Merge tag \u0027mlx5-updates-2023-02-15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux\n\nSaeed Mahameed says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nmlx5-updates-2023-02-15\n\n1) From Gal Tariq and Parav, Few cleanups for mlx5 driver.\n\n2) From Vlad: Allow offloading of ct \u0027new\u0027 match based on [1]\n\n[1] https://lore.kernel.org/netdev/20230201163100.1001180-1-vladbu@nvidia.com/\n\n* tag \u0027mlx5-updates-2023-02-15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:\n net/mlx5e: RX, Remove doubtful unlikely call\n net/mlx5e: Fix outdated TLS comment\n net/mlx5e: Remove unused function mlx5e_sq_xmit_simple\n net/mlx5e: Allow offloading of ct \u0027new\u0027 match\n net/mlx5e: Implement CT entry update\n net/mlx5: Simplify eq list traversal\n net/mlx5e: Remove redundant page argument in mlx5e_xdp_handle()\n net/mlx5e: Remove redundant page argument in mlx5e_xmit_xdp_buff()\n net/mlx5e: Switch to using napi_build_skb()\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230218090513.284718-1-saeed@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "981f40458e7a6ffbdff1a09ece6099b3b49d08a5", "tree": "35d3a3a7992301b0ef4bbe9baaf1f9e1f203db60", "parents": [ "871489dd01b67483248edc8873c389a66e469f30", "6702782845a5bf381a19b204c369e63420041665" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:11 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:12 2023 -0800" }, "message": "Merge branch \u0027net-sched-cls_api-support-hardware-miss-to-tc-action\u0027\n\nPaul Blakey says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet/sched: cls_api: Support hardware miss to tc action\n\nThis series adds support for hardware miss to instruct tc to continue execution\nin a specific tc action instance on a filter\u0027s action list. The mlx5 driver patch\n(besides the refactors) shows its usage instead of using just chain restore.\n\nCurrently a filter\u0027s action list must be executed all together or\nnot at all as driver are only able to tell tc to continue executing from a\nspecific tc chain, and not a specific filter/action.\n\nThis is troublesome with regards to action CT, where new connections should\nbe sent to software (via tc chain restore), and established connections can\nbe handled in hardware.\n\nChecking for new connections is done when executing the ct action in hardware\n(by checking the packet\u0027s tuple against known established tuples).\nBut if there is a packet modification (pedit) action before action CT and the\nchecked tuple is a new connection, hardware will need to revert the previous\npacket modifications before sending it back to software so it can\nre-match the same tc filter in software and re-execute its CT action.\n\nThe following is an example configuration of stateless nat\non mlx5 driver that isn\u0027t supported before this patchet:\n\n #Setup corrosponding mlx5 VFs in namespaces\n $ ip netns add ns0\n $ ip netns add ns1\n $ ip link set dev enp8s0f0v0 netns ns0\n $ ip netns exec ns0 ifconfig enp8s0f0v0 1.1.1.1/24 up\n $ ip link set dev enp8s0f0v1 netns ns1\n $ ip netns exec ns1 ifconfig enp8s0f0v1 1.1.1.2/24 up\n\n #Setup tc arp and ct rules on mxl5 VF representors\n $ tc qdisc add dev enp8s0f0_0 ingress\n $ tc qdisc add dev enp8s0f0_1 ingress\n $ ifconfig enp8s0f0_0 up\n $ ifconfig enp8s0f0_1 up\n\n #Original side\n $ tc filter add dev enp8s0f0_0 ingress chain 0 proto ip flower \\\n ct_state -trk ip_proto tcp dst_port 8888 \\\n action pedit ex munge tcp dport set 5001 pipe \\\n action csum ip tcp pipe \\\n action ct pipe \\\n action goto chain 1\n $ tc filter add dev enp8s0f0_0 ingress chain 1 proto ip flower \\\n ct_state +trk+est \\\n action mirred egress redirect dev enp8s0f0_1\n $ tc filter add dev enp8s0f0_0 ingress chain 1 proto ip flower \\\n ct_state +trk+new \\\n action ct commit pipe \\\n action mirred egress redirect dev enp8s0f0_1\n $ tc filter add dev enp8s0f0_0 ingress chain 0 proto arp flower \\\n action mirred egress redirect dev enp8s0f0_1\n\n #Reply side\n $ tc filter add dev enp8s0f0_1 ingress chain 0 proto arp flower \\\n action mirred egress redirect dev enp8s0f0_0\n $ tc filter add dev enp8s0f0_1 ingress chain 0 proto ip flower \\\n ct_state -trk ip_proto tcp \\\n action ct pipe \\\n action pedit ex munge tcp sport set 8888 pipe \\\n action csum ip tcp pipe \\\n action mirred egress redirect dev enp8s0f0_0\n\n #Run traffic\n $ ip netns exec ns1 iperf -s -p 5001\u0026\n $ sleep 2 #wait for iperf to fully open\n $ ip netns exec ns0 iperf -c 1.1.1.2 -p 8888\n\n #dump tc filter stats on enp8s0f0_0 chain 0 rule and see hardware packets:\n $ tc -s filter show dev enp8s0f0_0 ingress chain 0 proto ip | grep \"hardware.*pkt\"\n Sent hardware 9310116832 bytes 6149672 pkt\n Sent hardware 9310116832 bytes 6149672 pkt\n Sent hardware 9310116832 bytes 6149672 pkt\n\nA new connection executing the first filter in hardware will first rewrite\nthe dst port to the new port, and then the ct action is executed,\nbecause this is a new connection, hardware will need to be send this back\nto software, on chain 0, to execute the first filter again in software.\nThe dst port needs to be reverted otherwise it won\u0027t re-match the old\ndst port in the first filter. Because of that, currently mlx5 driver will\nreject offloading the above action ct rule.\n\nThis series adds support for hardware partially executing a filter\u0027s action list,\nand letting tc software continue processing in the specific action instance\nwhere hardware left off (in the above case after the \"action pedit ex munge tcp\ndport... of the first rule\") allowing support for scenarios such as the above.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230217223620.28508-1-paulb@nvidia.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "6702782845a5bf381a19b204c369e63420041665", "tree": "35d3a3a7992301b0ef4bbe9baaf1f9e1f203db60", "parents": [ "235ff07da7ec2e5addcee68a366ac987571b4d7c" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:20 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/mlx5e: TC, Set CT miss to the specific ct action instance\n\nCurrently, CT misses restore the missed chain on the tc skb extension so\ntc will continue from the relevant chain. Instead, restore the CT action\u0027s\nmiss cookie on the extension, which will instruct tc to continue from the\nthis specific CT action instance on the relevant filter\u0027s action list.\n\nMap the CT action\u0027s miss_cookie to a new miss object (ACT_MISS), and use\nthis miss mapping instead of the current chain miss object (CHAIN_MISS)\nfor CT action misses.\n\nTo restore this new miss mapping value, add a RX restore rule for each\nsuch mapping value.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Roi Dayan \u003croid@nvidia.com\u003e\nReviewed-by: Oz Sholmo \u003cozsh@nvidia.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "235ff07da7ec2e5addcee68a366ac987571b4d7c", "tree": "a8b43bd61787e67f1826f54e8424ff1024d2b74b", "parents": [ "93a1ab2c545b88fa70f8c91b60b54e4175008742" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:19 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG\n\nThis reg usage is always a mapped object, not necessarily\ncontaining chain info.\n\nRename to properly convey what it stores.\nThis patch doesn\u0027t change any functionality.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Roi Dayan \u003croid@nvidia.com\u003e\nReviewed-by: Leon Romanovsky \u003cleonro@nvidia.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "93a1ab2c545b88fa70f8c91b60b54e4175008742", "tree": "6acb7cab8420564d4b5fc46447090c7da3bd9a16", "parents": [ "03a283cdc8c80fd6002f8ab898ff5ccf78e33f95" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:18 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/mlx5: Refactor tc miss handling to a single function\n\nMove tc miss handling code to en_tc.c, and remove\nduplicate code.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Roi Dayan \u003croid@nvidia.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "03a283cdc8c80fd6002f8ab898ff5ccf78e33f95", "tree": "9a1c1a24afc59ccadc53a59c65189b2238b1d9bd", "parents": [ "606c7c43d08c4daf954ec8d58ae6dd39292fb457" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:17 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/mlx5: Kconfig: Make tc offload depend on tc skb extension\n\nTc skb extension is a basic requirement for using tc\noffload to support correct restoration on action miss.\n\nDepend on it.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "606c7c43d08c4daf954ec8d58ae6dd39292fb457", "tree": "81a89d33194b6ce79b5d31898995f7ec20488951", "parents": [ "08a0063df3aed8d76a4034279117db12dbc1050f" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:16 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/sched: flower: Support hardware miss to tc action\n\nTo support hardware miss to tc action in actions on the flower\nclassifier, implement the required getting of filter actions,\nand setup filter exts (actions) miss by giving it the filter\u0027s\nhandle and actions.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@nvidia.com\u003e\nReviewed-by: Simon Horman \u003csimon.horman@corigine.com\u003e\nReviewed-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "08a0063df3aed8d76a4034279117db12dbc1050f", "tree": "669bec635c692e22d0e0c11af84d33098fea9e38", "parents": [ "80cd22c35c9001fe72bf614d29439de41933deca" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:15 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/sched: flower: Move filter handle initialization earlier\n\nTo support miss to action during hardware offload the filter\u0027s\nhandle is needed when setting up the actions (tcf_exts_init()),\nand before offloading.\n\nMove filter handle initialization earlier.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@nvidia.com\u003e\nReviewed-by: Simon Horman \u003csimon.horman@corigine.com\u003e\nReviewed-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "80cd22c35c9001fe72bf614d29439de41933deca", "tree": "0a4953069a9b1876740444178bc5091f4ff6d089", "parents": [ "db4b49025c0c7116f1d2dfe8d5bbfc983ac054de" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:14 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/sched: cls_api: Support hardware miss to tc action\n\nFor drivers to support partial offload of a filter\u0027s action list,\nadd support for action miss to specify an action instance to\ncontinue from in sw.\n\nCT action in particular can\u0027t be fully offloaded, as new connections\nneed to be handled in software. This imposes other limitations on\nthe actions that can be offloaded together with the CT action, such\nas packet modifications.\n\nAssign each action on a filter\u0027s action list a unique miss_cookie\nwhich drivers can then use to fill action_miss part of the tc skb\nextension. On getting back this miss_cookie, find the action\ninstance with relevant cookie and continue classifying from there.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@nvidia.com\u003e\nReviewed-by: Simon Horman \u003csimon.horman@corigine.com\u003e\nReviewed-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nAcked-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "db4b49025c0c7116f1d2dfe8d5bbfc983ac054de", "tree": "80a16930c1e9f34eda2c728f571d3e61d78dbbaa", "parents": [ "871489dd01b67483248edc8873c389a66e469f30" ], "author": { "name": "Paul Blakey", "email": "paulb@nvidia.com", "time": "Sat Feb 18 00:36:13 2023 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:46:10 2023 -0800" }, "message": "net/sched: Rename user cookie and act cookie\n\nstruct tc_action-\u003eact_cookie is a user defined cookie,\nand the related struct flow_action_entry-\u003eact_cookie is\nused as an handle similar to struct flow_cls_offload-\u003ecookie.\n\nRename tc_action-\u003eact_cookie to user_cookie, and\nflow_action_entry-\u003eact_cookie to cookie so their names\nwould better fit their usage.\n\nSigned-off-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nReviewed-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "871489dd01b67483248edc8873c389a66e469f30", "tree": "6829f042291a2aa824294d3e808ab7ff39367855", "parents": [ "5f22c3b6215d34b8af45c390c8d5ea2cef2ef5d2", "ed9a8ad7d8a1a0eb7d4e1414d0a04ece7c2265df" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:40:52 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:40:52 2023 -0800" }, "message": "Merge tag \u0027ieee802154-for-net-next-2023-02-20\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next\n\nStefan Schmidt says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: ieee802154-next 2023-02-20\n\nMiquel Raynal build upon his earlier work and introduced two new\nfeatures into the ieee802154 stack. Beaconing to announce existing\nPAN\u0027s and passive scanning to discover the beacons and associated\nPAN\u0027s. The matching changes to the userspace configuration tool\nhave been posted as well and will be released together with the\nkernel release.\n\nArnd Bergmann and Dmitry Torokhov worked on converting the\nat86rf230 and cc2520 drivers away from the unused platform_data\nusage and towards the new gpiod API. (I had to add a revert as\nDmitry found a regression on an already pushed tree on my side).\n\nChanges since v1 (pull request 2023-02-02)\n- Netlink API extack and NLA_POLICY* usage as suggested by Jakub\n- Removed always true condition found by kernel test robot\n- Simplify device removal with running background job for scanning\n- Fix problems with beacon sending in some cases by using the MLME\n tx path\n\n* tag \u0027ieee802154-for-net-next-2023-02-20\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next:\n ieee802154: Drop device trackers\n mac802154: Fix an always true condition\n mac802154: Send beacons using the MLME Tx path\n ieee802154: Change error code on monitor scan netlink request\n ieee802154: Convert scan error messages to extack\n ieee802154: Use netlink policies when relevant on scan parameters\n ieee802154: at86rf230: switch to using gpiod API\n ieee802154: at86rf230: drop support for platform data\n Revert \"at86rf230: convert to gpio descriptors\"\n cc2520: move to gpio descriptors\n mac802154: Avoid superfluous endianness handling\n at86rf230: convert to gpio descriptors\n mac802154: Handle basic beaconing\n ieee802154: Add support for user beaconing requests\n mac802154: Handle passive scanning\n mac802154: Add MLME Tx locked helpers\n mac802154: Prepare forcing specific symbol duration\n ieee802154: Introduce a helper to validate a channel\n ieee802154: Define a beacon frame header\n ieee802154: Add support for user scanning requests\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230220213749.386451-1-stefan@datenfreihafen.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5f22c3b6215d34b8af45c390c8d5ea2cef2ef5d2", "tree": "6c06272ceddeb2ed8b922d2776e3332aa81be6a8", "parents": [ "5feeaba1063102baf789f4df263f655333924ccb" ], "author": { "name": "Alejandro Lucero", "email": "alejandro.lucero-palau@amd.com", "time": "Mon Feb 20 11:01:33 2023 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:39:50 2023 -0800" }, "message": "sfc: fix builds without CONFIG_RTC_LIB\n\nAdd an embarrassingly missed semicolon plus and embarrassingly missed\nparenthesis breaking kernel building when CONFIG_RTC_LIB is not set\nlike the one reported with ia64 config.\n\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nLink: https://lore.kernel.org/oe-kbuild-all/202302170047.EjCPizu3-lkp@intel.com/\nFixes: 14743ddd2495 (\"sfc: add devlink info support for ef100\")\nSigned-off-by: Alejandro Lucero \u003calejandro.lucero-palau@amd.com\u003e\nLink: https://lore.kernel.org/r/20230220110133.29645-1-alejandro.lucero-palau@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5feeaba1063102baf789f4df263f655333924ccb", "tree": "5d83985e809c4cab990532b498b7059a8a346c59", "parents": [ "f8f185e39b4de91bc5235e5be0d829bea69d9b06" ], "author": { "name": "Yang Li", "email": "yang.lee@linux.alibaba.com", "time": "Mon Feb 20 14:59:58 2023 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:39:00 2023 -0800" }, "message": "sfc: clean up some inconsistent indentings\n\nFix some indentngs and remove the warning below:\ndrivers/net/ethernet/sfc/mae.c:657 efx_mae_enumerate_mports() warn: inconsistent indenting\n\nReported-by: Abaci Robot \u003cabaci@linux.alibaba.com\u003e\nLink: https://bugzilla.openanolis.cn/show_bug.cgi?id\u003d4117\nSigned-off-by: Yang Li \u003cyang.lee@linux.alibaba.com\u003e\nAcked-by: Martin Habets \u003chabetsm.xilinx@gmail.com\u003e\nLink: https://lore.kernel.org/r/20230220065958.52941-1-yang.lee@linux.alibaba.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "f8f185e39b4de91bc5235e5be0d829bea69d9b06", "tree": "5c759a791acf23ad13047546a7af50eacc968508", "parents": [ "be9832c2e9cc4c15906a77baddcd906fb4bb864b" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Sat Feb 18 10:38:50 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:38:00 2023 -0800" }, "message": "net/mlx4_en: Introduce flexible array to silence overflow warning\n\nThe call \"skb_copy_from_linear_data(skb, inl + 1, spc)\" triggers a FORTIFY\nmemcpy() warning on ppc64 platform:\n\nIn function ‘fortify_memcpy_chk’,\n inlined from ‘skb_copy_from_linear_data’ at ./include/linux/skbuff.h:4029:2,\n inlined from ‘build_inline_wqe’ at drivers/net/ethernet/mellanox/mlx4/en_tx.c:722:4,\n inlined from ‘mlx4_en_xmit’ at drivers/net/ethernet/mellanox/mlx4/en_tx.c:1066:3:\n./include/linux/fortify-string.h:513:25: error: call to ‘__write_overflow_field’ declared with\nattribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()?\n[-Werror\u003dattribute-warning]\n 513 | __write_overflow_field(p_size_field, size);\n | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nSame behaviour on x86 you can get if you use \"__always_inline\" instead of\n\"inline\" for skb_copy_from_linear_data() in skbuff.h\n\nThe call here copies data into inlined tx destricptor, which has 104\nbytes (MAX_INLINE) space for data payload. In this case \"spc\" is known\nin compile-time but the destination is used with hidden knowledge\n(real structure of destination is different from that the compiler\ncan see). That cause the fortify warning because compiler can check\nbounds, but the real bounds are different. \"spc\" can\u0027t be bigger than\n64 bytes (MLX4_INLINE_ALIGN), so the data can always fit into inlined\ntx descriptor. The fact that \"inl\" points into inlined tx descriptor is\ndetermined earlier in mlx4_en_xmit().\n\nAvoid confusing the compiler with \"inl + 1\" constructions to get to past\nthe inl header by introducing a flexible array \"data\" to the struct so\nthat the compiler can see that we are not dealing with an array of inl\nstructs, but rather, arbitrary data following the structure. There are\nno changes to the structure layout reported by pahole, and the resulting\nmachine code is actually smaller.\n\nReported-by: Josef Oskera \u003cjoskera@redhat.com\u003e\nLink: https://lore.kernel.org/lkml/20230217094541.2362873-1-joskera@redhat.com\nFixes: f68f2ff91512 (\"fortify: Detect struct member overflows in memcpy() at compile-time\")\nCc: Yishai Hadas \u003cyishaih@nvidia.com\u003e\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nLink: https://lore.kernel.org/r/20230218183842.never.954-kees@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "be9832c2e9cc4c15906a77baddcd906fb4bb864b", "tree": "d1b9df10dad7e0e3be7409743dbe0a0f481235c6", "parents": [ "ee8d72a157ebb4b8c4b8b664f5a78a341fede2ef" ], "author": { "name": "Kuniyuki Iwashima", "email": "kuniyu@amazon.com", "time": "Fri Feb 17 12:09:20 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:31:49 2023 -0800" }, "message": "net/ulp: Remove redundant -\u003eclone() test in inet_clone_ulp().\n\nCommit 2c02d41d71f9 (\"net/ulp: prevent ULP without clone op from entering\nthe LISTEN status\") guarantees that all ULP listeners have clone() op, so\nwe no longer need to test it in inet_clone_ulp().\n\nSigned-off-by: Kuniyuki Iwashima \u003ckuniyu@amazon.com\u003e\nLink: https://lore.kernel.org/r/20230217200920.85306-1-kuniyu@amazon.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "ee8d72a157ebb4b8c4b8b664f5a78a341fede2ef", "tree": "4c6abf1db03b7bffaf8d9a41ecda74b9041312fb", "parents": [ "01bb11ad828b320749764fa93ad078db20d08a9e", "168de0233586fb06c5c5c56304aa9a928a09b0ba" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 15:38:41 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 16:31:14 2023 -0800" }, "message": "Merge tag \u0027for-netdev\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next\n\nDaniel Borkmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: bpf-next 2023-02-17\n\nWe\u0027ve added 64 non-merge commits during the last 7 day(s) which contain\na total of 158 files changed, 4190 insertions(+), 988 deletions(-).\n\nThe main changes are:\n\n1) Add a rbtree data structure following the \"next-gen data structure\"\n precedent set by recently-added linked-list, that is, by using\n kfunc + kptr instead of adding a new BPF map type, from Dave Marchevsky.\n\n2) Add a new benchmark for hashmap lookups to BPF selftests,\n from Anton Protopopov.\n\n3) Fix bpf_fib_lookup to only return valid neighbors and add an option\n to skip the neigh table lookup, from Martin KaFai Lau.\n\n4) Add cgroup.memory\u003dnobpf kernel parameter option to disable BPF memory\n accouting for container environments, from Yafang Shao.\n\n5) Batch of ice multi-buffer and driver performance fixes,\n from Alexander Lobakin.\n\n6) Fix a bug in determining whether global subprog\u0027s argument is\n PTR_TO_CTX, which is based on type names which breaks kprobe progs,\n from Andrii Nakryiko.\n\n7) Prep work for future -mcpu\u003dv4 LLVM option which includes usage of\n BPF_ST insn. Thus improve BPF_ST-related value tracking in verifier,\n from Eduard Zingerman.\n\n8) More prep work for later building selftests with Memory Sanitizer\n in order to detect usages of undefined memory, from Ilya Leoshkevich.\n\n9) Fix xsk sockets to check IFF_UP earlier to avoid a NULL pointer\n dereference via sendmsg(), from Maciej Fijalkowski.\n\n10) Implement BPF trampoline for RV64 JIT compiler, from Pu Lehui.\n\n11) Fix BPF memory allocator in combination with BPF hashtab where it could\n corrupt special fields e.g. used in bpf_spin_lock, from Hou Tao.\n\n12) Fix LoongArch BPF JIT to always use 4 instructions for function\n address so that instruction sequences don\u0027t change between passes,\n from Hengqi Chen.\n\n* tag \u0027for-netdev\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (64 commits)\n selftests/bpf: Add bpf_fib_lookup test\n bpf: Add BPF_FIB_LOOKUP_SKIP_NEIGH for bpf_fib_lookup\n riscv, bpf: Add bpf trampoline support for RV64\n riscv, bpf: Add bpf_arch_text_poke support for RV64\n riscv, bpf: Factor out emit_call for kernel and bpf context\n riscv: Extend patch_text for multiple instructions\n Revert \"bpf, test_run: fix \u0026xdp_frame misplacement for LIVE_FRAMES\"\n selftests/bpf: Add global subprog context passing tests\n selftests/bpf: Convert test_global_funcs test to test_loader framework\n bpf: Fix global subprog context argument resolution logic\n LoongArch, bpf: Use 4 instructions for function address in JIT\n bpf: bpf_fib_lookup should not return neigh in NUD_FAILED state\n bpf: Disable bh in bpf_test_run for xdp and tc prog\n xsk: check IFF_UP earlier in Tx path\n Fix typos in selftest/bpf files\n selftests/bpf: Use bpf_{btf,link,map,prog}_get_info_by_fd()\n samples/bpf: Use bpf_{btf,link,map,prog}_get_info_by_fd()\n bpftool: Use bpf_{btf,link,map,prog}_get_info_by_fd()\n libbpf: Use bpf_{btf,link,map,prog}_get_info_by_fd()\n libbpf: Introduce bpf_{btf,link,map,prog}_get_info_by_fd()\n ...\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230217221737.31122-1-daniel@iogearbox.net\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "01bb11ad828b320749764fa93ad078db20d08a9e", "tree": "7c85bb1f6473ba62607a6a37b1d75f1660d6620a", "parents": [ "3fcdf2dfefb6313ea0395519d1784808c0b6559b" ], "author": { "name": "Yury Norov", "email": "yury.norov@gmail.com", "time": "Thu Feb 16 17:39:08 2023 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 20 11:45:33 2023 -0800" }, "message": "sched/topology: fix KASAN warning in hop_cmp()\n\nDespite that prev_hop is used conditionally on cur_hop\nis not the first hop, it\u0027s initialized unconditionally.\n\nBecause initialization implies dereferencing, it might happen\nthat the code dereferences uninitialized memory, which has been\nspotted by KASAN. Fix it by reorganizing hop_cmp() logic.\n\nReported-by: Bruno Goncalves \u003cbgoncalv@redhat.com\u003e\nFixes: cd7f55359c90 (\"sched: add sched_numa_find_nth_cpu()\")\nSigned-off-by: Yury Norov \u003cyury.norov@gmail.com\u003e\nLink: https://lore.kernel.org/r/Y+7avK6V9SyAWsXi@yury-laptop/\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "3fcdf2dfefb6313ea0395519d1784808c0b6559b", "tree": "865d92753eb13d18f249ae1d316ce7e6158c5a20", "parents": [ "5f1eb1ff58ea122e24adf0bc940f268ed2227462" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Fri Feb 17 10:34:14 2023 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:40:53 2023 +0000" }, "message": "net: bcmgenet: Support wake-up from s2idle\n\nWhen we suspend into s2idle we also need to enable the interrupt line\nthat generates the MPD and HFB interrupts towards the host CPU interrupt\ncontroller (typically the ARM GIC or MIPS L1) to make it exit s2idle.\n\nWhen we suspend into other modes such as \"standby\" or \"mem\" we engage a\npower management state machine which will gate off the CPU L1 controller\n(priv-\u003eirq0) and ungate the side band wake-up interrupt (priv-\u003ewol_irq).\nIt is safe to have both enabled as wake-up sources because they are\nmutually exclusive given any suspend mode.\n\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5f1eb1ff58ea122e24adf0bc940f268ed2227462", "tree": "9e570cd6d4c78167faf5aa6a2b5f618ff093046b", "parents": [ "fce10282a03db59bdb1cba6333d0564461d47bd6" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Fri Feb 17 18:24:54 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:39:59 2023 +0000" }, "message": "scm: add user copy checks to put_cmsg()\n\nThis is a followup of commit 2558b8039d05 (\"net: use a bounce\nbuffer for copying skb-\u003emark\")\n\nx86 and powerpc define user_access_begin, meaning\nthat they are not able to perform user copy checks\nwhen using user_write_access_begin() / unsafe_copy_to_user()\nand friends [1]\n\nInstead of waiting bugs to trigger on other arches,\nadd a check_object_size() in put_cmsg() to make sure\nthat new code tested on x86 with CONFIG_HARDENED_USERCOPY\u003dy\nwill perform more security checks.\n\n[1] We can not generically call check_object_size() from\nunsafe_copy_to_user() because UACCESS is enabled at this point.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fce10282a03db59bdb1cba6333d0564461d47bd6", "tree": "ccee6e3ad8db16127ccd47bef97c3a6aa7c23323", "parents": [ "f6aa90a7a94a64f8c7ad28e09c1339624a824506" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Fri Feb 17 19:09:20 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:38:35 2023 +0000" }, "message": "devlink: drop leftover duplicate/unused code\n\nThe recent merge from net left-over some unused code in\nleftover.c - nomen omen.\n\nJust drop the unused bits.\n\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@nvidia.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f6aa90a7a94a64f8c7ad28e09c1339624a824506", "tree": "3f18a147571b2d7dbe65beb77eaa640f2b3575cc", "parents": [ "4d3e050b548878a14edfa2aeaca5d583b8b980ef", "6ad172748db49deef0da9038d29019aedf991a7e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:31:22 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:31:22 2023 +0000" }, "message": "Merge tag \u0027linux-can-next-for-6.3-20230217\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next\nMarc Kleine-Budde says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: can-next 2023-02-17 - fixed\n\nthis is a pull request of 4 patches for net-next/master.\n\nThe first patch is by Yang Li and converts the ctucanfd driver to\ndevm_platform_ioremap_resource().\n\nThe last 3 patches are by Frank Jungclaus, target the esd_usb driver\nand contains preparations for the upcoming support of the esd\nCAN-USB/3 hardware.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4d3e050b548878a14edfa2aeaca5d583b8b980ef", "tree": "85e8d36d490ad4ce5fab8896690261bacaba65cd", "parents": [ "38d711aacc3d6df13bec75577f966c43b83576fb" ], "author": { "name": "Horatiu Vultur", "email": "horatiu.vultur@microchip.com", "time": "Fri Feb 17 14:28:31 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:30:08 2023 +0000" }, "message": "net: lan966x: Use automatic selection of VCAP rule actionset\n\nSince commit 81e164c4aec5 (\"net: microchip: sparx5: Add automatic\nselection of VCAP rule actionset\") the VCAP API has the capability to\nselect automatically the actionset based on the actions that are attached\nto the rule. So it is not needed anymore to hardcode the actionset in the\ndriver, therefore it is OK to remove this.\n\nSigned-off-by: Horatiu Vultur \u003choratiu.vultur@microchip.com\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "38d711aacc3d6df13bec75577f966c43b83576fb", "tree": "74deeba1123d2df5b4700532693b73e246bafa94", "parents": [ "e469b6268d8cb7c6cf6b9cd5ce7f403de695627f", "3a7d84eae03bef4c02c39822b2ea6be5ac73de7b" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:22:54 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:22:54 2023 +0000" }, "message": "Merge branch \u0027default_rps_mask-follow-up\u0027\n\nPaolo Abeni says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: default_rps_mask follow-up\n\nThe first patch namespacify the setting. In the common case, once\nproper isolation is in place in the main namespace, forwarding\nto/from each child netns will allways happen on the desidered CPUs.\n\nAny additional RPS stage inside the child namespace will not provide\nadditional isolation and could hurt performance badly if picking a\nCPU on a remote node.\n\nThe 2nd patch adds more self-tests coverage.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3a7d84eae03bef4c02c39822b2ea6be5ac73de7b", "tree": "74deeba1123d2df5b4700532693b73e246bafa94", "parents": [ "50bcfe8df7c73ce51762f65d218b4ef0cc5da3ee" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Fri Feb 17 13:28:50 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:22:54 2023 +0000" }, "message": "self-tests: more rps self tests\n\nExplicitly check for child netns and main ns independency\n\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "50bcfe8df7c73ce51762f65d218b4ef0cc5da3ee", "tree": "095c3d2d38e056301ab0d59842e38fa85e331d0f", "parents": [ "e469b6268d8cb7c6cf6b9cd5ce7f403de695627f" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Fri Feb 17 13:28:49 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:22:54 2023 +0000" }, "message": "net: make default_rps_mask a per netns attribute\n\nThat really was meant to be a per netns attribute from the beginning.\n\nThe idea is that once proper isolation is in place in the main\nnamespace, additional demux in the child namespaces will be redundant.\nLet\u0027s make child netns default rps mask empty by default.\n\nTo avoid bloating the netns with a possibly large cpumask, allocate\nit on-demand during the first write operation.\n\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e469b6268d8cb7c6cf6b9cd5ce7f403de695627f", "tree": "3878f0e54a82132e20a2ee13f2be749d8f4b9a09", "parents": [ "1155a2281de9e7c08c5c6e265b32b28d1fe9ea07", "38ae3192296924181537544e7cfc43ca78eadcda" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:18:13 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 11:18:13 2023 +0000" }, "message": "Merge tag \u0027wireless-next-2023-02-17\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next\n\nKalle Valo says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nwireless-next patches for v6.3\n\nThird set of patches for v6.3. This time only a set of small fixes\nsubmitted during the last day or two.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1155a2281de9e7c08c5c6e265b32b28d1fe9ea07", "tree": "275c9e476ba3ec43ed530e9ef88438e5ac430460", "parents": [ "129ff4de58ff0ca981b833834d8b9bddb71d4bfe", "2954fe60e33da0f4de4d81a4c95c7dddb517d00c" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:53:56 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:53:56 2023 +0000" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next\n\nPablo Neira Ayuso says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nNetfilter/IPVS updates for net-next\n\nThe following patchset contains Netfilter updates for net-next:\n\n1) Add safeguard to check for NULL tupe in objects updates via\n NFT_MSG_NEWOBJ, this should not ever happen. From Alok Tiwari.\n\n2) Incorrect pointer check in the new destroy rule command,\n from Yang Yingliang.\n\n3) Incorrect status bitcheck in nf_conntrack_udp_packet(),\n from Florian Westphal.\n\n4) Simplify seq_print_acct(), from Ilia Gavrilov.\n\n5) Use 2-arg optimal variant of kfree_rcu() in IPVS,\n from Julian Anastasov.\n\n6) TCP connection enters CLOSE state in conntrack for locally\n originated TCP reset packet from the reject target,\n from Florian Westphal.\n\nThe fixes #2 and #3 in this series address issues from the previous pull\nnf-next request in this net-next cycle.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "129ff4de58ff0ca981b833834d8b9bddb71d4bfe", "tree": "e3d968a1018152c2ef8e07af5ad921faa2d29070", "parents": [ "a59f832a71c938779b8abdb022d6d39903e635c3" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Feb 17 10:58:06 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:50:47 2023 +0000" }, "message": "net: microchip: sparx5: reduce stack usage\n\nThe vcap_admin structures in vcap_api_next_lookup_advanced_test()\ntake several hundred bytes of stack frame, but when CONFIG_KASAN_STACK\nis enabled, each one of them also has extra padding before and after\nit, which ends up blowing the warning limit:\n\nIn file included from drivers/net/ethernet/microchip/vcap/vcap_api.c:3521:\ndrivers/net/ethernet/microchip/vcap/vcap_api_kunit.c: In function \u0027vcap_api_next_lookup_advanced_test\u0027:\ndrivers/net/ethernet/microchip/vcap/vcap_api_kunit.c:1954:1: error: the frame size of 1448 bytes is larger than 1400 bytes [-Werror\u003dframe-larger-than\u003d]\n 1954 | }\n\nReduce the total stack usage by replacing the five structures with\nan array that only needs one pair of padding areas.\n\nFixes: 1f741f001160 (\"net: microchip: sparx5: Add KUNIT tests for enabling/disabling chains\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a59f832a71c938779b8abdb022d6d39903e635c3", "tree": "8c2bb62c8cf3091bfa4897119c88493a01ba2cde", "parents": [ "8173c2f9a1a4a5afcf465a62bc8b616c1fef252d" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Feb 17 10:56:39 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:49:30 2023 +0000" }, "message": "sfc: use IS_ENABLED() checks for CONFIG_SFC_SRIOV\n\nOne local variable has become unused after a recent change:\n\ndrivers/net/ethernet/sfc/ef100_nic.c: In function \u0027ef100_probe_netdev_pf\u0027:\ndrivers/net/ethernet/sfc/ef100_nic.c:1155:21: error: unused variable \u0027net_dev\u0027 [-Werror\u003dunused-variable]\n struct net_device *net_dev \u003d efx-\u003enet_dev;\n ^~~~~~~\n\nThe variable is still used in an #ifdef. Replace the #ifdef with\nan if(IS_ENABLED()) check that lets the compiler see where it is\nused, rather than adding another #ifdef.\n\nThis also fixes an uninitialized return value in ef100_probe_netdev_pf()\nthat gcc did not spot.\n\nFixes: 7e056e2360d9 (\"sfc: obtain device mac address based on firmware handle for ef100\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nReviewed-by: Simon Horman \u003csimon.horman@corigine.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8173c2f9a1a4a5afcf465a62bc8b616c1fef252d", "tree": "d9d648cd7fefb2b458bba164104bcf1ff5592cdf", "parents": [ "0d39ad3e1b045ca1600169ca8de1267ab4ae1a81" ], "author": { "name": "Michal Swiatkowski", "email": "michal.swiatkowski@linux.intel.com", "time": "Fri Feb 17 11:50:17 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:47:34 2023 +0000" }, "message": "ice: properly alloc ICE_VSI_LB\n\nDevlink reload patchset introduced regression. ICE_VSI_LB wasn\u0027t\ntaken into account when doing default allocation. Fix it by adding a\ncase for ICE_VSI_LB in ice_vsi_alloc_def().\n\nFixes: 6624e780a577 (\"ice: split ice_vsi_setup into smaller functions\")\nReported-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nAcked-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nSigned-off-by: Michal Swiatkowski \u003cmichal.swiatkowski@linux.intel.com\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0d39ad3e1b045ca1600169ca8de1267ab4ae1a81", "tree": "12ddc6299659dd6c35c3ac14924efb4e8680cc8c", "parents": [ "933a01ad599766cf9bcda788f956f425a8b0b0a2" ], "author": { "name": "Colin Ian King", "email": "colin.i.king@gmail.com", "time": "Fri Feb 17 09:25:28 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:44:33 2023 +0000" }, "message": "sfc: Fix spelling mistake \"creationg\" -\u003e \"creating\"\n\nThere is a spelling mistake in a pci_warn message. Fix it.\n\nSigned-off-by: Colin Ian King \u003ccolin.i.king@gmail.com\u003e\nReviewed-by:  Alejandro Lucero \u003calejandro.lucero-palau@amd.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "933a01ad599766cf9bcda788f956f425a8b0b0a2", "tree": "ab85573d13b2761177ada86bfa410aa94f4057cc", "parents": [ "c2a978c171a6d44d4d9710e7e4455f75d34aecee" ], "author": { "name": "Geetha sowjanya", "email": "gakula@marvell.com", "time": "Fri Feb 17 11:21:12 2023 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:42:37 2023 +0000" }, "message": "octeontx2-af: Add NIX Errata workaround on CN10K silicon\n\nThis patch adds workaround for below 2 HW erratas\n\n1. Due to improper clock gating, NIXRX may free the same\nNPA buffer multiple times.. to avoid this, always enable\nNIX RX conditional clock.\n\n2. NIX FIFO does not get initialized on reset, if the SMQ\nflush is triggered before the first packet is processed, it\nwill lead to undefined state. The workaround to perform SMQ\nflush only if packet count is non-zero in MDQ.\n\nSigned-off-by: Geetha sowjanya \u003cgakula@marvell.com\u003e\nSigned-off-by: Sunil Kovvuri Goutham \u003csgoutham@marvell.com\u003e\nSigned-off-by: Sai Krishna \u003csaikrishnag@marvell.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c2a978c171a6d44d4d9710e7e4455f75d34aecee", "tree": "9382002b9365b5956e40da64447930309626ea8d", "parents": [ "0b1dbf889d0bebfa533a6220b0379300d4007204" ], "author": { "name": "Andrew Lunn", "email": "andrew@lunn.ch", "time": "Fri Feb 17 04:15:20 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:06:49 2023 +0000" }, "message": "net: phy: Read EEE abilities when using .features\n\nA PHY driver can use a static integer value to indicate what link mode\nfeatures it supports, i.e, its abilities.. This is the old way, but\nuseful when dynamically determining the devices features does not\nwork, e.g. support of fibre.\n\nEEE support has been moved into phydev-\u003esupported_eee. This needs to\nbe set otherwise the code assumes EEE is not supported. It is normally\nset as part of reading the devices abilities. However if a static\ninteger value was used, the dynamic reading of the abilities is not\nperformed. Add a call to genphy_c45_read_eee_abilities() to read the\nEEE abilities.\n\nFixes: 8b68710a3121 (\"net: phy: start using genphy_c45_ethtool_get/set_eee()\")\nSigned-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nReviewed-by: Oleksij Rempel \u003co.rempel@pengutronix.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0b1dbf889d0bebfa533a6220b0379300d4007204", "tree": "7d5796d139f3fcbc4fe1e0997ee5ce0d900c7ae3", "parents": [ "cf06eef0c86bd2088bafecb211e06a2855b4c327", "2f987d486610752b364fd238fa3bb1a142d80b44" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:04:22 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:04:22 2023 +0000" }, "message": "Merge branch \u0027phydev-locks\u0027\n\nAndrew Lunn says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nAdd additional phydev locks\n\nThe phydev lock should be held when accessing members of phydev, or\ncalling into the driver. Some of the phy_ethtool_ functions are\nmissing locks. Add them. To avoid deadlock the marvell driver is\nmodified since it calls one of the functions which gain locks, which\nwould result in a deadlock.\n\nThe missing locks have not caused noticeable issues, so these patches\nare for net-next.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2f987d486610752b364fd238fa3bb1a142d80b44", "tree": "7d5796d139f3fcbc4fe1e0997ee5ce0d900c7ae3", "parents": [ "3365777a6a2243f1cca5a441f2c89002d16fc580" ], "author": { "name": "Andrew Lunn", "email": "andrew@lunn.ch", "time": "Fri Feb 17 04:07:14 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:04:22 2023 +0000" }, "message": "net: phy: Add locks to ethtool functions\n\nThe phydev lock should be held while accessing members of phydev,\nor calling into the driver.\n\nSigned-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3365777a6a2243f1cca5a441f2c89002d16fc580", "tree": "e2b663d44b92d86cdfbccb34d868c2c71762adbd", "parents": [ "cf06eef0c86bd2088bafecb211e06a2855b4c327" ], "author": { "name": "Andrew Lunn", "email": "andrew@lunn.ch", "time": "Fri Feb 17 04:07:13 2023 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 10:04:22 2023 +0000" }, "message": "net: phy: marvell: Use the unlocked genphy_c45_ethtool_get_eee()\n\nphy_ethtool_get_eee() is about to gain locking of the phydev lock.\nThis means it cannot be used within a PHY driver without causing a\ndeadlock. Swap to using genphy_c45_ethtool_get_eee() which assumes the\nlock has already been taken.\n\nSigned-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cf06eef0c86bd2088bafecb211e06a2855b4c327", "tree": "5ad08fa6db860df81a5d89216b704328765f5675", "parents": [ "dd1b527831a3ed659afa01b672d8e1f7e6ca95a5", "ac03694bc009703022cf45a9c90675d5505c584c" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:24 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:24 2023 +0000" }, "message": "Merge branch \u0027icmp6-drop-reason\u0027\n\nEric Dumazet says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nipv6: icmp6: better drop reason support\n\nThis series aims to have more precise drop reason reports for icmp6.\n\nThis should reduce false positives on most usual cases.\n\nThis can be extended as needed later.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ac03694bc009703022cf45a9c90675d5505c584c", "tree": "5ad08fa6db860df81a5d89216b704328765f5675", "parents": [ "c34b8bb11ebc135e970653bd6fc8e3f863fb6a81" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:42 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to icmpv6_echo_reply()\n\nChange icmpv6_echo_reply() to return a drop reason.\n\nFor the moment, return NOT_SPECIFIED or SKB_CONSUMED.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c34b8bb11ebc135e970653bd6fc8e3f863fb6a81", "tree": "ef1b9c3f83a0573e50cc31d5dcaf7687267ef837", "parents": [ "784d4477f07b930df73bc77e842e03f1dacb83aa" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:41 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST\n\nHosts can often receive neighbour discovery messages\nthat are not for them.\n\nUse a dedicated drop reason to make clear the packet is dropped\nfor this normal case.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "784d4477f07b930df73bc77e842e03f1dacb83aa", "tree": "1e27c24768d6f98d776565fcbb4b9a00d97738e4", "parents": [ "ec993edf05ca4eee5878edf51fdb5ffa2b1decc3" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:40 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add SKB_DROP_REASON_IPV6_NDISC_BAD_OPTIONS\n\nThis is a generic drop reason for any error detected\nin ndisc_parse_options().\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ec993edf05ca4eee5878edf51fdb5ffa2b1decc3", "tree": "25e3035e0b313523c0e8de0272b973c5af115dae", "parents": [ "2f326d9d9ff46fb2e45fb3b6ae77eff04332dde6" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:39 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to ndisc_redirect_rcv()\n\nChange ndisc_redirect_rcv() to return a drop reason.\n\nFor the moment, return PKT_TOO_SMALL, NOT_SPECIFIED\nand values from icmpv6_notify().\n\nMore reasons are added later.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2f326d9d9ff46fb2e45fb3b6ae77eff04332dde6", "tree": "dd050826f585e57ad99cd5e798c9133c1663b38c", "parents": [ "243e37c642ac1d52858bc5f3559e380babf21169" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:38 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to ndisc_router_discovery()\n\nChange ndisc_router_discovery() to return a drop reason.\n\nFor the moment, return PKT_TOO_SMALL, NOT_SPECIFIED\nand SKB_CONSUMED.\n\nMore reasons are added later.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "243e37c642ac1d52858bc5f3559e380babf21169", "tree": "fb46c8500cb2d08c213345d46ee154997e83f451", "parents": [ "3009f9ae21ec539985977dcaa0cebb628afeb181" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:37 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to ndisc_recv_rs()\n\nChange ndisc_recv_rs() to return a drop reason.\n\nFor the moment, return PKT_TOO_SMALL, NOT_SPECIFIED\nor SKB_CONSUMED. More reasons are added later.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3009f9ae21ec539985977dcaa0cebb628afeb181", "tree": "40b688a81d7f8f509d87b8e5024c546d2444757a", "parents": [ "7c9c8913f4523cda30a710736844d74bfee060a4" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:36 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to ndisc_recv_na()\n\nChange ndisc_recv_na() to return a drop reason.\n\nFor the moment, return PKT_TOO_SMALL, NOT_SPECIFIED\nor SKB_CONSUMED. More reasons are added later.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c9c8913f4523cda30a710736844d74bfee060a4", "tree": "9861b2ccdbc33c181977d0a48013d95fa923a698", "parents": [ "dd1b527831a3ed659afa01b672d8e1f7e6ca95a5" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 16:28:35 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:54:23 2023 +0000" }, "message": "ipv6: icmp6: add drop reason support to ndisc_recv_ns()\n\nChange ndisc_recv_ns() to return a drop reason.\n\nFor the moment, return PKT_TOO_SMALL, NOT_SPECIFIED\nor SKB_CONSUMED.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "dd1b527831a3ed659afa01b672d8e1f7e6ca95a5", "tree": "49d25216963d33e8acc4e8a9cf85bd9fd1ce464f", "parents": [ "9f78bf330a66cd400b3e00f370f597e9fa939207" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Feb 16 15:47:18 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:28:49 2023 +0000" }, "message": "net: add location to trace_consume_skb()\n\nkfree_skb() includes the location, it makes sense\nto add it to consume_skb() as well.\n\nAfter patch:\n\n taskd_EventMana 8602 [004] 420.406239: skb:consume_skb: skbaddr\u003d0xffff893a4a6d0500 location\u003dunix_stream_read_generic\n swapper 0 [011] 422.732607: skb:consume_skb: skbaddr\u003d0xffff89597f68cee0 location\u003dmlx4_en_free_tx_desc\n discipline 9141 [043] 423.065653: skb:consume_skb: skbaddr\u003d0xffff893a487e9c00 location\u003dskb_consume_udp\n swapper 0 [010] 423.073166: skb:consume_skb: skbaddr\u003d0xffff8949ce9cdb00 location\u003dicmpv6_rcv\n borglet 8672 [014] 425.628256: skb:consume_skb: skbaddr\u003d0xffff8949c42e9400 location\u003dnetlink_dump\n swapper 0 [028] 426.263317: skb:consume_skb: skbaddr\u003d0xffff893b1589dce0 location\u003dnet_rx_action\n wget 14339 [009] 426.686380: skb:consume_skb: skbaddr\u003d0xffff893a51b552e0 location\u003dtcp_rcv_state_process\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9f78bf330a66cd400b3e00f370f597e9fa939207", "tree": "2d260b0cc0471c30641dadb8ae18cef51bf85b2e", "parents": [ "b148d400f820637bcc95f6aca64c8763a2db858f" ], "author": { "name": "Xuan Zhuo", "email": "xuanzhuo@linux.alibaba.com", "time": "Thu Feb 16 16:30:47 2023 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 20 08:22:12 2023 +0000" }, "message": "xsk: support use vaddr as ring\n\nWhen we try to start AF_XDP on some machines with long running time, due\nto the machine\u0027s memory fragmentation problem, there is no sufficient\ncontiguous physical memory that will cause the start failure.\n\nIf the size of the queue is 8 * 1024, then the size of the desc[] is\n8 * 1024 * 8 \u003d 16 * PAGE, but we also add struct xdp_ring size, so it is\n16page+. This is necessary to apply for a 4-order memory. If there are a\nlot of queues, it is difficult to these machine with long running time.\n\nHere, that we actually waste 15 pages. 4-Order memory is 32 pages, but\nwe only use 17 pages.\n\nThis patch replaces __get_free_pages() by vmalloc() to allocate memory\nto solve these problems.\n\nSigned-off-by: Xuan Zhuo \u003cxuanzhuo@linux.alibaba.com\u003e\nAcked-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b148d400f820637bcc95f6aca64c8763a2db858f", "tree": "25a035e2be7cb279ca8ac7d20c00ed95b9218baf", "parents": [ "c078381856230f1e8e13738661d83c2b4b433819", "64cb6aad12328015202af5b2a9623c6bcc021855" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:46:59 2023 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:46:59 2023 +0100" }, "message": "Merge branch \u0027taprio-queuemaxsdu-fixes\u0027\n\nVladimir Oltean says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ntaprio queueMaxSDU fixes\n\nThis fixes 3 issues noticed while attempting to reoffload the\ndynamically calculated queueMaxSDU values. These are:\n- Dynamic queueMaxSDU is not calculated correctly due to a lost patch\n- Dynamically calculated queueMaxSDU needs to be clamped on the low end\n- Dynamically calculated queueMaxSDU needs to be clamped on the high end\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230215224632.2532685-1-vladimir.oltean@nxp.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "64cb6aad12328015202af5b2a9623c6bcc021855", "tree": "25a035e2be7cb279ca8ac7d20c00ed95b9218baf", "parents": [ "bdf366bd867c4565b535a5825df7ddcb4773fc28" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Thu Feb 16 00:46:32 2023 +0200" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:46:57 2023 +0100" }, "message": "net/sched: taprio: dynamic max_sdu larger than the max_mtu is unlimited\n\nIt makes no sense to keep randomly large max_sdu values, especially if\nlarger than the device\u0027s max_mtu. These are visible in \"tc qdisc show\".\nSuch a max_sdu is practically unlimited and will cause no packets for\nthat traffic class to be dropped on enqueue.\n\nJust set max_sdu_dynamic to U32_MAX, which in the logic below causes\ntaprio to save a max_frm_len of U32_MAX and a max_sdu presented to user\nspace of 0 (unlimited).\n\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nReviewed-by: Kurt Kanzenbach \u003ckurt@linutronix.de\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "bdf366bd867c4565b535a5825df7ddcb4773fc28", "tree": "b8340300f9126da0ee4725605da45cab81965cb2", "parents": [ "09dbdf28f9f9fa27e16895c67fbd94ff36124766" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Thu Feb 16 00:46:31 2023 +0200" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:46:57 2023 +0100" }, "message": "net/sched: taprio: don\u0027t allow dynamic max_sdu to go negative after stab adjustment\n\nThe overhead specified in the size table comes from the user. With small\ntime intervals (or gates always closed), the overhead can be larger than\nthe max interval for that traffic class, and their difference is\nnegative.\n\nWhat we want to happen is for max_sdu_dynamic to have the smallest\nnon-zero value possible (1) which means that all packets on that traffic\nclass are dropped on enqueue. However, since max_sdu_dynamic is u32, a\nnegative is represented as a large value and oversized dropping never\nhappens.\n\nUse max_t with int to force a truncation of max_frm_len to no smaller\nthan dev-\u003ehard_header_len + 1, which in turn makes max_sdu_dynamic no\nsmaller than 1.\n\nFixes: fed87cc6718a (\"net/sched: taprio: automatically calculate queueMaxSDU based on TC gate durations\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nReviewed-by: Kurt Kanzenbach \u003ckurt@linutronix.de\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "09dbdf28f9f9fa27e16895c67fbd94ff36124766", "tree": "e374b9affdba50025e5d08013433bd5f816e7ae8", "parents": [ "c078381856230f1e8e13738661d83c2b4b433819" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Thu Feb 16 00:46:30 2023 +0200" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:46:57 2023 +0100" }, "message": "net/sched: taprio: fix calculation of maximum gate durations\n\ntaprio_calculate_gate_durations() depends on netdev_get_num_tc() and\nthis returns 0. So it calculates the maximum gate durations for no\ntraffic class.\n\nI had tested the blamed commit only with another patch in my tree, one\nwhich in the end I decided isn\u0027t valuable enough to submit (\"net/sched:\ntaprio: mask off bits in gate mask that exceed number of TCs\").\n\nThe problem is that having this patch threw off my testing. By moving\nthe netdev_set_num_tc() call earlier, we implicitly gave to\ntaprio_calculate_gate_durations() the information it needed.\n\nExtract only the portion from the unsubmitted change which applies the\nmqprio configuration to the netdev earlier.\n\nLink: https://patchwork.kernel.org/project/netdevbpf/patch/20230130173145.475943-15-vladimir.oltean@nxp.com/\nFixes: a306a90c8ffe (\"net/sched: taprio: calculate tc gate durations\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nReviewed-by: Kurt Kanzenbach \u003ckurt@linutronix.de\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "c078381856230f1e8e13738661d83c2b4b433819", "tree": "a4c3287e98c4aca17b78ea5b0c47656511b57883", "parents": [ "d269ac136ede573c90c18a66961143294ef564e5" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Feb 15 21:48:05 2023 +0000" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:33:25 2023 +0100" }, "message": "rxrpc: Fix overproduction of wakeups to recvmsg()\n\nFix three cases of overproduction of wakeups:\n\n (1) rxrpc_input_split_jumbo() conditionally notifies the app that there\u0027s\n data for recvmsg() to collect if it queues some data - and then its\n only caller, rxrpc_input_data(), goes and wakes up recvmsg() anyway.\n\n Fix the rxrpc_input_data() to only do the wakeup in failure cases.\n\n (2) If a DATA packet is received for a call by the I/O thread whilst\n recvmsg() is busy draining the call\u0027s rx queue in the app thread, the\n call will left on the recvmsg() queue for recvmsg() to pick up, even\n though there isn\u0027t any data on it.\n\n This can cause an unexpected recvmsg() with a 0 return and no MSG_EOR\n set after the reply has been posted to a service call.\n\n Fix this by discarding pending calls from the recvmsg() queue that\n don\u0027t need servicing yet.\n\n (3) Not-yet-completed calls get requeued after having data read from them,\n even if they have no data to read.\n\n Fix this by only requeuing them if they have data waiting on them; if\n they don\u0027t, the I/O thread will requeue them when data arrives or they\n fail.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: Marc Dionne \u003cmarc.dionne@auristor.com\u003e\ncc: linux-afs@lists.infradead.org\nLink: https://lore.kernel.org/r/3386149.1676497685@warthog.procyon.org.uk\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "d269ac136ede573c90c18a66961143294ef564e5", "tree": "8b0e94aeda862c5c9dc74afe30a7bd49ba43f787", "parents": [ "1c93e48cc39147723abdcadc251611f6f0c18dee", "f651334e1ef5d1c8fba16a27dff2f3629e7e86e9" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:22 2023 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:22 2023 +0100" }, "message": "Merge branch \u0027net-final-gsi-register-updates\u0027\n\nAlex Elder says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: final GSI register updates\n\nI believe this is the last set of changes required to allow IPA v5.0\nto be supported. There is a little cleanup work remaining, but that\ncan happen in the next Linux release cycle. Otherwise we just need\nconfig data and register definitions for IPA v5.0 (and DTS updates).\nThese are ready but won\u0027t be posted without further testing.\n\nThe first patch in this series fixes a minor bug in a patch just\nposted, which I found too late. The second eliminates the GSI\nmemory \"adjustment\"; this was done previously to avoid/delay the\nneed to implement a more general way to define GSI register offsets.\nNote that this patch causes \"checkpatch\" warnings due to indentation\nthat aligns with an open parenthesis.\n\nThe third patch makes use of the newly-defined register offsets, to\neliminate the need for a function that hid a few details. The next\nmodifies a different helper function to work properly for IPA v5.0+.\nThe fifth patch changes the way the event ring size is specified\nbased on how it\u0027s now done for IPA v5.0+. And the last defines a\nnew register required for IPA v5.0+.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20230215195352.755744-1-elder@linaro.org\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "f651334e1ef5d1c8fba16a27dff2f3629e7e86e9", "tree": "8b0e94aeda862c5c9dc74afe30a7bd49ba43f787", "parents": [ "37cd29ec84016297e5b834345dd272a7d3bd8fae" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:52 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: add HW_PARAM_4 GSI register\n\nStarting at IPA v5.0, the number of event rings per EE is defined\nin a field in a new HW_PARAM_4 GSI register rather than HW_PARAM_2.\nDefine this new register and its fields, and update the code that\nchecks the number of rings supported by hardware to use the proper\nfield based on IPA version.\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "37cd29ec84016297e5b834345dd272a7d3bd8fae", "tree": "a8c39c123765e360c2ea263202f0be191b20edc7", "parents": [ "62747512ebe6cd292e280aa7dc1dd37d36dcb637" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:51 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: support different event ring encoding\n\nStarting with IPA v5.0, a channel\u0027s event ring index is encoded in\na field in the CH_C_CNTXT_1 GSI register rather than CH_C_CNTXT_0.\nDefine a new field ID for the former register and encode the event\nring in the appropriate register.\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "62747512ebe6cd292e280aa7dc1dd37d36dcb637", "tree": "0ef19ec5f31472b0ebb372eab5815349e3e41d5b", "parents": [ "f75f44ddd4cb4bb0ede9c6c7f29679e9794552df" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:50 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: avoid setting an undefined field\n\nThe GSI channel protocol field in the CH_C_CNTXT_0 GSI register is\nwidened starting IPA v5.0, making the CHTYPE_PROTOCOL_MSB field\nadded in IPA v4.5 unnecessary. Update the code to reflect this.\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "f75f44ddd4cb4bb0ede9c6c7f29679e9794552df", "tree": "066973f2e981a5a9cc007cb6d374ba4398cd5480", "parents": [ "59b12b1d27f3f82e56462f0da6413e1849a06d3a" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:49 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: kill ev_ch_e_cntxt_1_length_encode()\n\nNow that we explicitly define each register field width there is no\nneed to have a special encoding function for the event ring length.\nAdd a field for this to the EV_CH_E_CNTXT_1 GSI register, and use it\nin place of ev_ch_e_cntxt_1_length_encode() (which can be removed).\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "59b12b1d27f3f82e56462f0da6413e1849a06d3a", "tree": "8364fb7dfcfba693c0dc9911bb04346c93134228", "parents": [ "ecfa80ce3b8740d0760f44cbd9e60f446682b3fa" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:48 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: kill gsi-\u003evirt_raw\n\nStarting at IPA v4.5, almost all GSI registers had their offsets\nchanged by a fixed amount (shifted downward by 0xd000). Rather than\ndefining offsets for all those registers dependent on version, an\nadjustment was applied for most register accesses. This was\nimplemented in commit cdeee49f3ef7f (\"net: ipa: adjust GSI register\naddresses\"). It was later modified to be a bit more obvious about\nthe adjusment, in commit 571b1e7e58ad3 (\"net: ipa: use a separate\npointer for adjusted GSI memory\").\n\nWe now are able to define every GSI register with its own offset, so\nthere\u0027s no need to implement this special adjustment.\n\nSo get rid of the \"virt_raw\" pointer, and just maintain \"virt\" as\nthe (non-adjusted) base address of I/O mapped GSI register memory.\n\nRedefine the offsets of all GSI registers (other than the INTER_EE\nones, which were not subject to the adjustment) for IPA v4.5+,\nsubtracting 0xd000 from their defined offsets instead.\n\nMove the ERROR_LOG and ERROR_LOG_CLR definitions further down in the\nregister definition files so all registers are defined in order of\ntheir offset.\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "ecfa80ce3b8740d0760f44cbd9e60f446682b3fa", "tree": "b0d23f6b9234118294a1911ada90a2cf19c1eb98", "parents": [ "1c93e48cc39147723abdcadc251611f6f0c18dee" ], "author": { "name": "Alex Elder", "email": "elder@linaro.org", "time": "Wed Feb 15 13:53:47 2023 -0600" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 08:14:20 2023 +0100" }, "message": "net: ipa: fix an incorrect assignment\n\nI spotted an error in a patch posted this week, unfortunately just\nafter it got accepted. The effect of the bug is that time-based\ninterrupt moderation is disabled. This is not technically a bug,\nbut it is not what is intended. The problem is that a |\u003d assignment\ngot implemented as a simple assignment, so the previously assigned\nvalue was ignored.\n\nFixes: edc6158b18af (\"net: ipa: define fields for event-ring related registers\")\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "1c93e48cc39147723abdcadc251611f6f0c18dee", "tree": "ea8030a9dec49dc17e09a0ed80f2c74ba2f7afce", "parents": [ "675f176b4dcc2b75adbcea7ba0e9a649527f53bd" ], "author": { "name": "Lorenzo Bianconi", "email": "lorenzo@kernel.org", "time": "Wed Feb 15 15:32:57 2023 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Mon Feb 20 07:49:44 2023 +0100" }, "message": "net: dpaa2-eth: do not always set xsk support in xdp_features flag\n\nDo not always add NETDEV_XDP_ACT_XSK_ZEROCOPY bit in xdp_features flag\nbut check if the NIC really supports it.\n\nSigned-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nReviewed-by: Larysa Zaremba \u003clarysa.zaremba@intel.com\u003e\nLink: https://lore.kernel.org/r/3dba6ea42dc343a9f2d7d1a6a6a6c173235e1ebf.1676471386.git.lorenzo@kernel.org\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "ed9a8ad7d8a1a0eb7d4e1414d0a04ece7c2265df", "tree": "280ce63657fdfa9790fc0779860905553a49c190", "parents": [ "61d7dddf46caa1c6dd869385a275e4d2931e7090" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:35 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:49:53 2023 +0100" }, "message": "ieee802154: Drop device trackers\n\nIn order to prevent a device from disappearing when a background job was\nstarted, dev_hold() and dev_put() calls were made. During the\nstabilization phase of the scan/beacon features, it was later decided\nthat removing the device while a background job was ongoing was a valid use\ncase, and we should instead stop the background job and then remove the\ndevice, rather than prevent the device from being removed. This is what\nis currently done, which means manually reference counting the device\nduring background jobs is no longer needed.\n\nFixes: ed3557c947e1 (\"ieee802154: Add support for user scanning requests\")\nFixes: 9bc114504b07 (\"ieee802154: Add support for user beaconing requests\")\nReported-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-7-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "61d7dddf46caa1c6dd869385a275e4d2931e7090", "tree": "ebefb095c6ef91e7033024bdee9ea248e969bef1", "parents": [ "1375e3ba9d773f2dbac96ebddfdd0d160276ca40" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:34 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:47:26 2023 +0100" }, "message": "mac802154: Fix an always true condition\n\nAt this stage we simply do not care about the delayed work value,\nbecause active scan is not yet supported, so we can blindly queue\nanother work once a beacon has been sent.\n\nIt fixes a smatch warning:\n mac802154_beacon_worker() warn: always true condition\n \u0027(local-\u003ebeacon_interval \u003e\u003d 0) \u003d\u003e (0-u32max \u003e\u003d 0)\u0027\n\nFixes: 3accf4762734 (\"mac802154: Handle basic beaconing\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-6-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "1375e3ba9d773f2dbac96ebddfdd0d160276ca40", "tree": "127339b391bab4a0e043e9e44975206fc051c23c", "parents": [ "1edecbd0bd45c9c899e0f82b123342f28423468c" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:33 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:44:53 2023 +0100" }, "message": "mac802154: Send beacons using the MLME Tx path\n\nUsing ieee802154_subif_start_xmit() to bypass the net queue when\nsending beacons is broken because it does not acquire the\nHARD_TX_LOCK(), hence not preventing datagram buffers to be smashed by\nbeacons upon contention situation. Using the mlme_tx helper is not the\nbest fit either but at least it is not buggy and has little-to-no\nperformance hit. More details are given in the comment explaining this\nchoice in the code.\n\nFixes: 3accf4762734 (\"mac802154: Handle basic beaconing\")\nReported-by: Alexander Aring \u003caahringo@redhat.com\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-5-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "1edecbd0bd45c9c899e0f82b123342f28423468c", "tree": "b4e5d1fcc32bf75b28216074bd3c981bb5b872a3", "parents": [ "a0b6106672b53158bf141d1a713aef2c891d74b1" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:32 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:41:23 2023 +0100" }, "message": "ieee802154: Change error code on monitor scan netlink request\n\nReturning EPERM gives the impression that \"right now\" it is not\npossible, but \"later\" it could be, while what we want to express is the\nfact that this is not currently supported at all (might change in the\nfuture). So let\u0027s return EOPNOTSUPP instead.\n\nFixes: ed3557c947e1 (\"ieee802154: Add support for user scanning requests\")\nSuggested-by: Alexander Aring \u003caahringo@redhat.com\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-4-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "a0b6106672b53158bf141d1a713aef2c891d74b1", "tree": "8e73948236f82a48296aee9c36943e58a7901061", "parents": [ "648324c9b690bcda0ac6f6499370effc0336ee27" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:31 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:38:41 2023 +0100" }, "message": "ieee802154: Convert scan error messages to extack\n\nInstead of printing error messages in the kernel log, let\u0027s use extack.\nWhen there is a netlink error returned that could be further specified\nwith a string, use extack as well.\n\nApply this logic to the very recent scan/beacon infrastructure.\n\nFixes: ed3557c947e1 (\"ieee802154: Add support for user scanning requests\")\nFixes: 9bc114504b07 (\"ieee802154: Add support for user beaconing requests\")\nSuggested-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-3-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "648324c9b690bcda0ac6f6499370effc0336ee27", "tree": "82785cd19f719b8ff87be712485d3a26bd8cd10f", "parents": [ "6130543654e0e5a79485ed8538c0bf2259fe7431" ], "author": { "name": "Miquel Raynal", "email": "miquel.raynal@bootlin.com", "time": "Tue Feb 14 14:50:30 2023 +0100" }, "committer": { "name": "Stefan Schmidt", "email": "stefan@datenfreihafen.org", "time": "Sat Feb 18 16:35:10 2023 +0100" }, "message": "ieee802154: Use netlink policies when relevant on scan parameters\n\nInstead of open-coding scan parameters (page, channels, duration, etc),\nlet\u0027s use the existing NLA_POLICY* macros. This help greatly reducing\nthe error handling and clarifying the overall logic.\n\nFixes: ed3557c947e1 (\"ieee802154: Add support for user scanning requests\")\nSuggested-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nLink: https://lore.kernel.org/r/20230214135035.1202471-2-miquel.raynal@bootlin.com\nSigned-off-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n" }, { "commit": "993fd9bd656a082c9b713171622c70f72f0af59f", "tree": "f15b062092911d2ef4db8a8b7cf572b6d931b2a8", "parents": [ "afce9271facb4cd41ffdf063cc6dd3d67341e723" ], "author": { "name": "Gal Pressman", "email": "gal@nvidia.com", "time": "Wed Feb 15 16:09:18 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:34 2023 -0800" }, "message": "net/mlx5e: RX, Remove doubtful unlikely call\n\nWhen building an skb in non-linear mode, it is not likely nor unlikely\nthat the xdp buff has fragments, it depends on the size of the packet\nreceived.\n\nSigned-off-by: Gal Pressman \u003cgal@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\nReviewed-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\n" }, { "commit": "afce9271facb4cd41ffdf063cc6dd3d67341e723", "tree": "ffdda0fc6320dfd2151777c9dfd0b4c0bfefa28a", "parents": [ "b5618a6b19c9dfead5133b72c40112f343b55d76" ], "author": { "name": "Tariq Toukan", "email": "tariqt@nvidia.com", "time": "Wed Feb 15 16:09:17 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:34 2023 -0800" }, "message": "net/mlx5e: Fix outdated TLS comment\n\nComment is outdated since\ncommit 40379a0084c2 (\"net/mlx5_fpga: Drop INNOVA TLS support\").\n\nSigned-off-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nReviewed-by: Gal Pressman \u003cgal@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "b5618a6b19c9dfead5133b72c40112f343b55d76", "tree": "6189368c842641322bad7f1af820b0f4146f7b4f", "parents": [ "f869bcb0d28ec981dd5f6d1958c607c457c1bb26" ], "author": { "name": "Tariq Toukan", "email": "tariqt@nvidia.com", "time": "Wed Feb 15 16:09:16 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:34 2023 -0800" }, "message": "net/mlx5e: Remove unused function mlx5e_sq_xmit_simple\n\nThe last usage was removed as part of\ncommit 40379a0084c2 (\"net/mlx5_fpga: Drop INNOVA TLS support\").\n\nSigned-off-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nReviewed-by: Gal Pressman \u003cgal@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "f869bcb0d28ec981dd5f6d1958c607c457c1bb26", "tree": "7d52bc556045fac8fc7f6653dde84d0e690d82ab", "parents": [ "94ceffb48eac7692677d8093dcde6965b70c4b35" ], "author": { "name": "Vlad Buslov", "email": "vladbu@nvidia.com", "time": "Sun Nov 06 17:22:21 2022 +0100" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:34 2023 -0800" }, "message": "net/mlx5e: Allow offloading of ct \u0027new\u0027 match\n\nAllow offloading filters that match on conntrack \u0027new\u0027 state in order to\nenable UDP NEW offload in the following patch.\n\nUnhardcode ct \u0027established\u0027 from ct modify header infrastructure code and\ndetermine correct ct state bit according to the metadata action \u0027cookie\u0027\nfield.\n\nSigned-off-by: Vlad Buslov \u003cvladbu@nvidia.com\u003e\nReviewed-by: Oz Shlomo \u003cozsh@nvidia.com\u003e\nReviewed-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "94ceffb48eac7692677d8093dcde6965b70c4b35", "tree": "43830bc4a39c97eb9166142b13896e145b5238d0", "parents": [ "3ac0b6aa892a1961d3bc62c87e3e3c9646e9d309" ], "author": { "name": "Vlad Buslov", "email": "vladbu@nvidia.com", "time": "Thu Dec 01 16:56:10 2022 +0100" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:33 2023 -0800" }, "message": "net/mlx5e: Implement CT entry update\n\nWith support for UDP NEW offload the flow_table may now send updates for\nexisting flows. Support properly replacing existing entries by updating\nflow restore_cookie and replacing the rule with new one with the same match\nbut new mod_hdr action that sets updated ctinfo.\n\nSigned-off-by: Vlad Buslov \u003cvladbu@nvidia.com\u003e\nReviewed-by: Oz Shlomo \u003cozsh@nvidia.com\u003e\nReviewed-by: Paul Blakey \u003cpaulb@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "3ac0b6aa892a1961d3bc62c87e3e3c9646e9d309", "tree": "1c968110b5a38a40bb8521405853e9baffdba382", "parents": [ "9da5294e2c6adc7169ccdfa5bef87f9e893d5a12" ], "author": { "name": "Parav Pandit", "email": "parav@nvidia.com", "time": "Wed Feb 15 16:09:13 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:33 2023 -0800" }, "message": "net/mlx5: Simplify eq list traversal\n\nEQ list is read only while finding the matching EQ.\nHence, avoid *_safe() version.\n\nSigned-off-by: Parav Pandit \u003cparav@nvidia.com\u003e\nReviewed-by: Shay Drory \u003cshayd@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\nReviewed-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\n" }, { "commit": "9da5294e2c6adc7169ccdfa5bef87f9e893d5a12", "tree": "e4352fa95756cde1afe77ad0cfe7a6dd1a2a4248", "parents": [ "bfc63c9796900b4f72a0700ac68e4c6306017833" ], "author": { "name": "Tariq Toukan", "email": "tariqt@nvidia.com", "time": "Wed Feb 15 16:09:12 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:33 2023 -0800" }, "message": "net/mlx5e: Remove redundant page argument in mlx5e_xdp_handle()\n\nRemove the page parameter, it can be derived from the xdp_buff member\nof mlx5e_xdp_buff.\n\nSigned-off-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nReviewed-by: Dragos Tatulea \u003cdtatulea@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "bfc63c9796900b4f72a0700ac68e4c6306017833", "tree": "c1fa58c76bcec1aebd6ad37db335f3f24f6edf6a", "parents": [ "53ee9142717714881fe0b4292b237e1707246f2c" ], "author": { "name": "Tariq Toukan", "email": "tariqt@nvidia.com", "time": "Wed Feb 15 16:09:11 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:33 2023 -0800" }, "message": "net/mlx5e: Remove redundant page argument in mlx5e_xmit_xdp_buff()\n\nRemove the page parameter, it can be derived from the xdp_buff.\n\nSigned-off-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nReviewed-by: Dragos Tatulea \u003cdtatulea@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\n" }, { "commit": "53ee9142717714881fe0b4292b237e1707246f2c", "tree": "fe73d164e0a73c277542bc63974f6d1e8a20ec9f", "parents": [ "675f176b4dcc2b75adbcea7ba0e9a649527f53bd" ], "author": { "name": "Tariq Toukan", "email": "tariqt@nvidia.com", "time": "Wed Feb 15 16:09:10 2023 -0800" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@nvidia.com", "time": "Sat Feb 18 01:01:33 2023 -0800" }, "message": "net/mlx5e: Switch to using napi_build_skb()\n\nUse napi_build_skb() which uses NAPI percpu caches to obtain\nskbuff_head instead of inplace allocation.\n\nnapi_build_skb() calls napi_skb_cache_get(), which returns a cached\nskb, or allocates a bulk of NAPI_SKB_CACHE_BULK (16) if cache is empty.\n\nPerformance test:\nTCP single stream, single ring, single core, default MTU (1500B).\n\nBefore: 26.5 Gbits/sec\nAfter: 30.1 Gbits/sec (+13.6%)\n\nSigned-off-by: Tariq Toukan \u003ctariqt@nvidia.com\u003e\nReviewed-by: Gal Pressman \u003cgal@nvidia.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@nvidia.com\u003e\nReviewed-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\n" }, { "commit": "168de0233586fb06c5c5c56304aa9a928a09b0ba", "tree": "5704b8b4453664107faf69ab5f6129df8572041c", "parents": [ "31de4105f00d64570139bc5494a201b0bd57349f" ], "author": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Fri Feb 17 12:55:15 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 22:12:04 2023 +0100" }, "message": "selftests/bpf: Add bpf_fib_lookup test\n\nThis patch tests the bpf_fib_lookup helper when looking up\na neigh in NUD_FAILED and NUD_STALE state. It also adds test\nfor the new BPF_FIB_LOOKUP_SKIP_NEIGH flag.\n\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nLink: https://lore.kernel.org/bpf/20230217205515.3583372-2-martin.lau@linux.dev\n" }, { "commit": "31de4105f00d64570139bc5494a201b0bd57349f", "tree": "16b85d933eb7572af8530eb3a2d1dcd04f8f816d", "parents": [ "49b5e77ae3e214acff4728595b4ac7bf776693ca" ], "author": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Fri Feb 17 12:55:14 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 22:12:04 2023 +0100" }, "message": "bpf: Add BPF_FIB_LOOKUP_SKIP_NEIGH for bpf_fib_lookup\n\nThe bpf_fib_lookup() also looks up the neigh table.\nThis was done before bpf_redirect_neigh() was added.\n\nIn the use case that does not manage the neigh table\nand requires bpf_fib_lookup() to lookup a fib to\ndecide if it needs to redirect or not, the bpf prog can\ndepend only on using bpf_redirect_neigh() to lookup the\nneigh. It also keeps the neigh entries fresh and connected.\n\nThis patch adds a bpf_fib_lookup flag, SKIP_NEIGH, to avoid\nthe double neigh lookup when the bpf prog always call\nbpf_redirect_neigh() to do the neigh lookup. The params-\u003esmac\noutput is skipped together when SKIP_NEIGH is set because\nbpf_redirect_neigh() will figure out the smac also.\n\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nLink: https://lore.kernel.org/bpf/20230217205515.3583372-1-martin.lau@linux.dev\n" }, { "commit": "49b5e77ae3e214acff4728595b4ac7bf776693ca", "tree": "8d8439f133914a38aa7aa4b7436726e9ca76ecba", "parents": [ "596f2e6f9cf41436a5512a3f278c86da5c5598fb" ], "author": { "name": "Pu Lehui", "email": "pulehui@huawei.com", "time": "Wed Feb 15 21:52:05 2023 +0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:45:30 2023 +0100" }, "message": "riscv, bpf: Add bpf trampoline support for RV64\n\nBPF trampoline is the critical infrastructure of the BPF subsystem, acting\nas a mediator between kernel functions and BPF programs. Numerous important\nfeatures, such as using BPF program for zero overhead kernel introspection,\nrely on this key component. We can\u0027t wait to support bpf trampoline on RV64.\nThe related tests have passed, as well as the test_verifier with no new\nfailure ceses.\n\nSigned-off-by: Pu Lehui \u003cpulehui@huawei.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nTested-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nAcked-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nLink: https://lore.kernel.org/bpf/20230215135205.1411105-5-pulehui@huaweicloud.com\n" }, { "commit": "596f2e6f9cf41436a5512a3f278c86da5c5598fb", "tree": "379ec18fcd887274967b6b0d31c1ab3941b412c5", "parents": [ "0fd1fd0104954380477353aea29c347e85dff16d" ], "author": { "name": "Pu Lehui", "email": "pulehui@huawei.com", "time": "Wed Feb 15 21:52:04 2023 +0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:45:30 2023 +0100" }, "message": "riscv, bpf: Add bpf_arch_text_poke support for RV64\n\nImplement bpf_arch_text_poke for RV64. For call scenario, to make BPF\ntrampoline compatible with the kernel and BPF context, we follow the\nframework of RV64 ftrace to reserve 4 nops for BPF programs as function\nentry, and use auipc+jalr instructions for function call. However, since\nauipc+jalr call instruction is non-atomic operation, we need to use\nstop-machine to make sure instructions patching in atomic context. Also,\nwe use auipc+jalr pair and need to patch in stop-machine context for\njump scenario.\n\nSigned-off-by: Pu Lehui \u003cpulehui@huawei.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nTested-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nAcked-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nLink: https://lore.kernel.org/bpf/20230215135205.1411105-4-pulehui@huaweicloud.com\n" }, { "commit": "0fd1fd0104954380477353aea29c347e85dff16d", "tree": "347bd2795977125923c003b13953166a0989786c", "parents": [ "5e57fb7b0bd3ea7e994ef1c0ab3562d1fe0676b2" ], "author": { "name": "Pu Lehui", "email": "pulehui@huawei.com", "time": "Wed Feb 15 21:52:03 2023 +0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:45:30 2023 +0100" }, "message": "riscv, bpf: Factor out emit_call for kernel and bpf context\n\nThe current emit_call function is not suitable for kernel function call as\nit store return value to bpf R0 register. We can separate it out for common\nuse. Meanwhile, simplify judgment logic, that is, fixed function address\ncan use jal or auipc+jalr, while the unfixed can use only auipc+jalr.\n\nSigned-off-by: Pu Lehui \u003cpulehui@huawei.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nTested-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nAcked-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nLink: https://lore.kernel.org/bpf/20230215135205.1411105-3-pulehui@huaweicloud.com\n" }, { "commit": "5e57fb7b0bd3ea7e994ef1c0ab3562d1fe0676b2", "tree": "150dfe4d1f95aec03c11f35144a27cb39bf117e5", "parents": [ "181127fb76e62d06ab17a75fd610129688612343" ], "author": { "name": "Pu Lehui", "email": "pulehui@huawei.com", "time": "Wed Feb 15 21:52:02 2023 +0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:45:30 2023 +0100" }, "message": "riscv: Extend patch_text for multiple instructions\n\nExtend patch_text for multiple instructions. This is the preparaiton for\nmultiple instructions text patching in riscv BPF trampoline, and may be\nuseful for other scenario.\n\nSigned-off-by: Pu Lehui \u003cpulehui@huawei.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nTested-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nReviewed-by: Conor Dooley \u003cconor.dooley@microchip.com\u003e\nAcked-by: Björn Töpel \u003cbjorn@rivosinc.com\u003e\nLink: https://lore.kernel.org/bpf/20230215135205.1411105-2-pulehui@huaweicloud.com\n" }, { "commit": "181127fb76e62d06ab17a75fd610129688612343", "tree": "1d3bc497410981d7d026f52e328a62d2753ea421", "parents": [ "e2b5cfc978f871996d1f8667515c0e06b33e620e" ], "author": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Fri Feb 17 12:13:09 2023 -0800" }, "committer": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Fri Feb 17 12:24:33 2023 -0800" }, "message": "Revert \"bpf, test_run: fix \u0026xdp_frame misplacement for LIVE_FRAMES\"\n\nThis reverts commit 6c20822fada1b8adb77fa450d03a0d449686a4a9.\n\nbuild bot failed on arch with different cache line size:\nhttps://lore.kernel.org/bpf/50c35055-afa9-d01e-9a05-ea5351280e4f@intel.com/\n\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\n" }, { "commit": "e2b5cfc978f871996d1f8667515c0e06b33e620e", "tree": "980f38a1b4e043ec3167cf63ce4d19c9eba42f6f", "parents": [ "95ebb376176c52382293e05e63f142114a5e40ef" ], "author": { "name": "Andrii Nakryiko", "email": "andrii@kernel.org", "time": "Wed Feb 15 20:59:54 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:21:50 2023 +0100" }, "message": "selftests/bpf: Add global subprog context passing tests\n\nAdd tests validating that it\u0027s possible to pass context arguments into\nglobal subprogs for various types of programs, including a particularly\ntricky KPROBE programs (which cover kprobes, uprobes, USDTs, a vast and\nimportant class of programs).\n\nSigned-off-by: Andrii Nakryiko \u003candrii@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Stanislav Fomichev \u003csdf@google.com\u003e\nLink: https://lore.kernel.org/bpf/20230216045954.3002473-4-andrii@kernel.org\n" }, { "commit": "95ebb376176c52382293e05e63f142114a5e40ef", "tree": "17b8bedca1f8aa7ab08816ef2f6fede6108a38fe", "parents": [ "d384dce281ed1b504fae2e279507827638d56fa3" ], "author": { "name": "Andrii Nakryiko", "email": "andrii@kernel.org", "time": "Wed Feb 15 20:59:53 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:20:44 2023 +0100" }, "message": "selftests/bpf: Convert test_global_funcs test to test_loader framework\n\nConvert 17 test_global_funcs subtests into test_loader framework for\neasier maintenance and more declarative way to define expected\nfailures/successes.\n\nSigned-off-by: Andrii Nakryiko \u003candrii@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Stanislav Fomichev \u003csdf@google.com\u003e\nLink: https://lore.kernel.org/bpf/20230216045954.3002473-3-andrii@kernel.org\n" }, { "commit": "d384dce281ed1b504fae2e279507827638d56fa3", "tree": "18dd9f650dca039d06f89a28977c950a06a4404e", "parents": [ "64f50f6575721ef03d001e907455cbe3baa2a5b1" ], "author": { "name": "Andrii Nakryiko", "email": "andrii@kernel.org", "time": "Wed Feb 15 20:59:52 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 21:20:44 2023 +0100" }, "message": "bpf: Fix global subprog context argument resolution logic\n\nKPROBE program\u0027s user-facing context type is defined as typedef\nbpf_user_pt_regs_t. This leads to a problem when trying to passing\nkprobe/uprobe/usdt context argument into global subprog, as kernel\nalways strip away mods and typedefs of user-supplied type, but takes\nexpected type from bpf_ctx_convert as is, which causes mismatch.\n\nCurrent way to work around this is to define a fake struct with the same\nname as expected typedef:\n\n struct bpf_user_pt_regs_t {};\n\n __noinline my_global_subprog(struct bpf_user_pt_regs_t *ctx) { ... }\n\nThis patch fixes the issue by resolving expected type, if it\u0027s not\na struct. It still leaves the above work-around working for backwards\ncompatibility.\n\nFixes: 91cc1a99740e (\"bpf: Annotate context types\")\nSigned-off-by: Andrii Nakryiko \u003candrii@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Stanislav Fomichev \u003csdf@google.com\u003e\nLink: https://lore.kernel.org/bpf/20230216045954.3002473-2-andrii@kernel.org\n" }, { "commit": "64f50f6575721ef03d001e907455cbe3baa2a5b1", "tree": "7c487fd3cc1be21498883f3e7c2290da88c13dcd", "parents": [ "1fe4850b34ab512ff911e2c035c75fb6438f7307" ], "author": { "name": "Hengqi Chen", "email": "hengqi.chen@gmail.com", "time": "Tue Feb 14 15:26:33 2023 +0000" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 17:43:07 2023 +0100" }, "message": "LoongArch, bpf: Use 4 instructions for function address in JIT\n\nThis patch fixes the following issue of function calls in JIT, like:\n\n [ 29.346981] multi-func JIT bug 105 !\u003d 103\n\nThe issus can be reproduced by running the \"inline simple bpf_loop call\"\nverifier test.\n\nThis is because we are emiting 2-4 instructions for 64-bit immediate moves.\nDuring the first pass of JIT, the placeholder address is zero, emiting two\ninstructions for it. In the extra pass, the function address is in XKVRANGE,\nemiting four instructions for it. This change the instruction index in\nJIT context. Let\u0027s always use 4 instructions for function address in JIT.\nSo that the instruction sequences don\u0027t change between the first pass and\nthe extra pass for function calls.\n\nFixes: 5dc615520c4d (\"LoongArch: Add BPF JIT support\")\nSigned-off-by: Hengqi Chen \u003chengqi.chen@gmail.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nTested-by: Tiezhu Yang \u003cyangtiezhu@loongson.cn\u003e\nLink: https://lore.kernel.org/bpf/20230214152633.2265699-1-hengqi.chen@gmail.com\n" }, { "commit": "38ae3192296924181537544e7cfc43ca78eadcda", "tree": "49982a0e541d191186e75857a55a36cd14aa3ddf", "parents": [ "3a156b52c73c9491441799bbbac18075bf96480e" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Feb 17 10:59:04 2023 +0100" }, "committer": { "name": "Kalle Valo", "email": "kvalo@kernel.org", "time": "Fri Feb 17 18:30:44 2023 +0200" }, "message": "wifi: rtl8xxxu: add LEDS_CLASS dependency\n\nrtl8xxxu now unconditionally uses LEDS_CLASS, so a Kconfig dependency\nis required to avoid link errors:\n\naarch64-linux-ld: drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.o: in function `rtl8xxxu_disconnect\u0027:\nrtl8xxxu_core.c:(.text+0x730): undefined reference to `led_classdev_unregister\u0027\n\nERROR: modpost: \"led_classdev_unregister\" [drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.ko] undefined!\nERROR: modpost: \"led_classdev_register_ext\" [drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.ko] undefined!\n\nFixes: 3be01622995b (\"wifi: rtl8xxxu: Register the LED and make it blink\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/20230217095910.2480356-1-arnd@kernel.org\n" }, { "commit": "1fe4850b34ab512ff911e2c035c75fb6438f7307", "tree": "dd92d0ddc7a8ab638b7381afb79a8681d407c3af", "parents": [ "af2d0d09eabe98b01bf02b236e381edae4209778" ], "author": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Thu Feb 16 16:41:48 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 16:19:42 2023 +0100" }, "message": "bpf: bpf_fib_lookup should not return neigh in NUD_FAILED state\n\nThe bpf_fib_lookup() helper does not only look up the fib (ie. route)\nbut it also looks up the neigh. Before returning the neigh, the helper\ndoes not check for NUD_VALID. When a neigh state (neigh-\u003enud_state)\nis in NUD_FAILED, its dmac (neigh-\u003eha) could be all zeros. The helper\nstill returns SUCCESS instead of NO_NEIGH in this case. Because of the\nSUCCESS return value, the bpf prog directly uses the returned dmac\nand ends up filling all zero in the eth header.\n\nThis patch checks for NUD_VALID and returns NO_NEIGH if the neigh is\nnot valid.\n\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nLink: https://lore.kernel.org/bpf/20230217004150.2980689-3-martin.lau@linux.dev\n" }, { "commit": "af2d0d09eabe98b01bf02b236e381edae4209778", "tree": "3fd4af59d6648e671783a6f4430719bffaa9766a", "parents": [ "1596dae2f17ec5c6e8c8f0e3fec78c5ae55c1e0b" ], "author": { "name": "Martin KaFai Lau", "email": "martin.lau@kernel.org", "time": "Thu Feb 16 16:41:47 2023 -0800" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 16:19:23 2023 +0100" }, "message": "bpf: Disable bh in bpf_test_run for xdp and tc prog\n\nSome of the bpf helpers require bh disabled. eg. The bpf_fib_lookup\nhelper that will be used in a latter selftest. In particular, it\ncalls ___neigh_lookup_noref that expects the bh disabled.\n\nThis patch disables bh before calling bpf_prog_run[_xdp], so\nthe testing prog can also use those helpers.\n\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nLink: https://lore.kernel.org/bpf/20230217004150.2980689-2-martin.lau@linux.dev\n" }, { "commit": "1596dae2f17ec5c6e8c8f0e3fec78c5ae55c1e0b", "tree": "34b5d564a90cea0a7599ece2564277faee8a45f5", "parents": [ "df71a42cc37a44cdc7682f57aecf14ff44391eed" ], "author": { "name": "Maciej Fijalkowski", "email": "maciej.fijalkowski@intel.com", "time": "Wed Feb 15 15:33:09 2023 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 17 16:05:28 2023 +0100" }, "message": "xsk: check IFF_UP earlier in Tx path\n\nXsk Tx can be triggered via either sendmsg() or poll() syscalls. These\ntwo paths share a call to common function xsk_xmit() which has two\nsanity checks within. A pseudo code example to show the two paths:\n\n__xsk_sendmsg() : xsk_poll():\nif (unlikely(!xsk_is_bound(xs))) if (unlikely(!xsk_is_bound(xs)))\n return -ENXIO; return mask;\nif (unlikely(need_wait)) (...)\n return -EOPNOTSUPP; xsk_xmit()\nmark napi id\n(...)\nxsk_xmit()\n\nxsk_xmit():\nif (unlikely(!(xs-\u003edev-\u003eflags \u0026 IFF_UP)))\n\treturn -ENETDOWN;\nif (unlikely(!xs-\u003etx))\n\treturn -ENOBUFS;\n\nAs it can be observed above, in sendmsg() napi id can be marked on\ninterface that was not brought up and this causes a NULL ptr\ndereference:\n\n[31757.505631] BUG: kernel NULL pointer dereference, address: 0000000000000018\n[31757.512710] #PF: supervisor read access in kernel mode\n[31757.517936] #PF: error_code(0x0000) - not-present page\n[31757.523149] PGD 0 P4D 0\n[31757.525726] Oops: 0000 [#1] PREEMPT SMP NOPTI\n[31757.530154] CPU: 26 PID: 95641 Comm: xdpsock Not tainted 6.2.0-rc5+ #40\n[31757.536871] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019\n[31757.547457] RIP: 0010:xsk_sendmsg+0xde/0x180\n[31757.551799] Code: 00 75 a2 48 8b 00 a8 04 75 9b 84 d2 74 69 8b 85 14 01 00 00 85 c0 75 1b 48 8b 85 28 03 00 00 48 8b 80 98 00 00 00 48 8b 40 20 \u003c8b\u003e 40 18 89 85 14 01 00 00 8b bd 14 01 00 00 81 ff 00 01 00 00 0f\n[31757.570840] RSP: 0018:ffffc90034f27dc0 EFLAGS: 00010246\n[31757.576143] RAX: 0000000000000000 RBX: ffffc90034f27e18 RCX: 0000000000000000\n[31757.583389] RDX: 0000000000000001 RSI: ffffc90034f27e18 RDI: ffff88984cf3c100\n[31757.590631] RBP: ffff88984714a800 R08: ffff88984714a800 R09: 0000000000000000\n[31757.597877] R10: 0000000000000001 R11: 0000000000000000 R12: 00000000fffffffa\n[31757.605123] R13: 0000000000000000 R14: 0000000000000003 R15: 0000000000000000\n[31757.612364] FS: 00007fb4c5931180(0000) GS:ffff88afdfa00000(0000) knlGS:0000000000000000\n[31757.620571] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[31757.626406] CR2: 0000000000000018 CR3: 000000184b41c003 CR4: 00000000007706e0\n[31757.633648] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n[31757.640894] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n[31757.648139] PKRU: 55555554\n[31757.650894] Call Trace:\n[31757.653385] \u003cTASK\u003e\n[31757.655524] sock_sendmsg+0x8f/0xa0\n[31757.659077] ? sockfd_lookup_light+0x12/0x70\n[31757.663416] __sys_sendto+0xfc/0x170\n[31757.667051] ? do_sched_setscheduler+0xdb/0x1b0\n[31757.671658] __x64_sys_sendto+0x20/0x30\n[31757.675557] do_syscall_64+0x38/0x90\n[31757.679197] entry_SYSCALL_64_after_hwframe+0x72/0xdc\n[31757.687969] Code: 8e f6 ff 44 8b 4c 24 2c 4c 8b 44 24 20 41 89 c4 44 8b 54 24 28 48 8b 54 24 18 b8 2c 00 00 00 48 8b 74 24 10 8b 7c 24 08 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 3a 44 89 e7 48 89 44 24 08 e8 b5 8e f6 ff 48\n[31757.707007] RSP: 002b:00007ffd49c73c70 EFLAGS: 00000293 ORIG_RAX: 000000000000002c\n[31757.714694] RAX: ffffffffffffffda RBX: 000055a996565380 RCX: 00007fb4c5727c16\n[31757.721939] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003\n[31757.729184] RBP: 0000000000000040 R08: 0000000000000000 R09: 0000000000000000\n[31757.736429] R10: 0000000000000040 R11: 0000000000000293 R12: 0000000000000000\n[31757.743673] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000\n[31757.754940] \u003c/TASK\u003e\n\nTo fix this, let\u0027s make xsk_xmit a function that will be responsible for\ngeneric Tx, where RCU is handled accordingly and pull out sanity checks\nand xs-\u003ezc handling. Populate sanity checks to __xsk_sendmsg() and\nxsk_poll().\n\nFixes: ca2e1a627035 (\"xsk: Mark napi_id on sendmsg()\")\nFixes: 18b1ab7aa76b (\"xsk: Fix race at socket teardown\")\nSigned-off-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nReviewed-by: Alexander Lobakin \u003caleksander.lobakin@intel.com\u003e\nLink: https://lore.kernel.org/r/20230215143309.13145-1-maciej.fijalkowski@intel.com\nSigned-off-by: Martin KaFai Lau \u003cmartin.lau@kernel.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "2954fe60e33da0f4de4d81a4c95c7dddb517d00c", "tree": "dc62dd7ec9f19796605a78780e4ca0075f12446a", "parents": [ "e4d0fe71f59dc5137a2793ff7560730d80d1e1f4" ], "author": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Wed Feb 01 14:45:22 2023 +0100" }, "committer": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Fri Feb 17 13:04:56 2023 +0100" }, "message": "netfilter: let reset rules clean out conntrack entries\n\niptables/nftables support responding to tcp packets with tcp resets.\n\nThe generated tcp reset packet passes through both output and postrouting\nnetfilter hooks, but conntrack will never see them because the generated\nskb has its -\u003enfct pointer copied over from the packet that triggered the\nreset rule.\n\nIf the reset rule is used for established connections, this\nmay result in the conntrack entry to be around for a very long\ntime (default timeout is 5 days).\n\nOne way to avoid this would be to not copy the nf_conn pointer\nso that the rest packet passes through conntrack too.\n\nProblem is that output rules might not have the same conntrack\nzone setup as the prerouting ones, so its possible that the\nreset skb won\u0027t find the correct entry. Generating a template\nentry for the skb seems error prone as well.\n\nAdd an explicit \"closing\" function that switches a confirmed\nconntrack entry to closed state and wire this up for tcp.\n\nIf the entry isn\u0027t confirmed, no action is needed because\nthe conntrack entry will never be committed to the table.\n\nReported-by: Russel King \u003clinux@armlinux.org.uk\u003e\nSigned-off-by: Florian Westphal \u003cfw@strlen.de\u003e\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "675f176b4dcc2b75adbcea7ba0e9a649527f53bd", "tree": "dc5eea870c19b405bb6558964e5367bb6c9205c5", "parents": [ "84cb1b53cdbad52642571e31a8aee301206d2043", "ec35307e18ba8174e2a3f701956059f6a36f22fb" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 17 11:06:39 2023 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 17 11:06:39 2023 +0000" }, "message": "Merge ra.kernel.org:/pub/scm/linux/kernel/git/netdev/net\n\nSome of the devlink bits were tricky, but I think I got it right.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3a156b52c73c9491441799bbbac18075bf96480e", "tree": "8553b47562045b4b05194fe805dc0ecef7f88c94", "parents": [ "d9fcf94b8f68f4a0b440d26a1143932525c2f3dd" ], "author": { "name": "Johannes Berg", "email": "johannes.berg@intel.com", "time": "Thu Feb 16 21:34:44 2023 +0100" }, "committer": { "name": "Kalle Valo", "email": "kvalo@kernel.org", "time": "Fri Feb 17 11:35:54 2023 +0200" }, "message": "wifi: iwlegacy: avoid fortify warning\n\nThere are two different alive messages, the \"init\" one is\nbigger than the other one, so we have a fortify read warn\nhere. Avoid it by copying from the variable-sized \u0027raw\u0027\ninstead.\n\nSigned-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/20230216203444.134310-1-johannes@sipsolutions.net\n" }, { "commit": "d9fcf94b8f68f4a0b440d26a1143932525c2f3dd", "tree": "30e3bc40fc59f3baf4f68c45bdd5f01bbe72e722", "parents": [ "0731d0b664f26f54b6293421af54da15b9eb1c8c" ], "author": { "name": "Johannes Berg", "email": "johannes.berg@intel.com", "time": "Thu Feb 16 20:57:54 2023 +0100" }, "committer": { "name": "Kalle Valo", "email": "kvalo@kernel.org", "time": "Fri Feb 17 11:34:56 2023 +0200" }, "message": "wifi: iwlwifi: mvm: remove unused iwl_dbgfs_is_match()\n\nThis inline function is unused, remove it.\n\nSigned-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/20230216205754.d500dcc2e90c.Id87df297263f86b5bba002f7cbb387abc13adf53@changeid\n" }, { "commit": "0731d0b664f26f54b6293421af54da15b9eb1c8c", "tree": "8dc9a8a60db29025656cdeed4468d15201715161", "parents": [ "4a267bc5ea8f159b614d0549030216d0434eccca" ], "author": { "name": "Po-Hao Huang", "email": "phhuang@realtek.com", "time": "Thu Feb 16 16:28:07 2023 +0800" }, "committer": { "name": "Kalle Valo", "email": "kvalo@kernel.org", "time": "Fri Feb 17 11:34:08 2023 +0200" }, "message": "wifi: rtw89: fix AP mode authentication transmission failed\n\nFor some ICs, packets can\u0027t be sent correctly without initializing\nCMAC table first. Previous flow do this initialization after\nassociated, results in authentication response fails to transmit.\nMove the initialization up front to a proper place to solve this.\n\nSigned-off-by: Po-Hao Huang \u003cphhuang@realtek.com\u003e\nSigned-off-by: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/20230216082807.22285-1-pkshih@realtek.com\n" }, { "commit": "4a267bc5ea8f159b614d0549030216d0434eccca", "tree": "82236bd9426aebff73d486fb6d5d8bb7ab8a8f32", "parents": [ "84cb1b53cdbad52642571e31a8aee301206d2043" ], "author": { "name": "Ping-Ke Shih", "email": "pkshih@realtek.com", "time": "Thu Feb 16 13:36:33 2023 +0800" }, "committer": { "name": "Kalle Valo", "email": "kvalo@kernel.org", "time": "Fri Feb 17 11:33:13 2023 +0200" }, "message": "wifi: rtw88: use RTW_FLAG_POWERON flag to prevent to power on/off twice\n\nUse power state to decide whether we can enter or leave IPS accurately,\nand then prevent to power on/off twice.\n\nThe commit 6bf3a083407b (\"wifi: rtw88: add flag check before enter or leave IPS\")\nwould like to prevent this as well, but it still can\u0027t entirely handle all\ncases. The exception is that WiFi gets connected and does suspend/resume,\nit will power on twice and cause it failed to power on after resuming,\nlike:\n\n rtw_8723de 0000:03:00.0: failed to poll offset\u003d0x6 mask\u003d0x2 value\u003d0x2\n rtw_8723de 0000:03:00.0: mac power on failed\n rtw_8723de 0000:03:00.0: failed to power on mac\n rtw_8723de 0000:03:00.0: leave idle state failed\n rtw_8723de 0000:03:00.0: failed to leave ips state\n rtw_8723de 0000:03:00.0: failed to leave idle state\n rtw_8723de 0000:03:00.0: failed to send h2c command\n\nTo fix this, introduce new flag RTW_FLAG_POWERON to reflect power state,\nand call rtw_mac_pre_system_cfg() to configure registers properly between\npower-off/-on.\n\nReported-by: Paul Gover \u003cpmw.gover@yahoo.co.uk\u003e\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d217016\nFixes: 6bf3a083407b (\"wifi: rtw88: add flag check before enter or leave IPS\")\nCc: \u003cStable@vger.kernel.org\u003e\nSigned-off-by: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/20230216053633.20366-1-pkshih@realtek.com\n" }, { "commit": "ec35307e18ba8174e2a3f701956059f6a36f22fb", "tree": "0c213e562624728f43a94163fa103ea7f9273dda", "parents": [ "3ac88fa4605ec98e545fb3ad0154f575fda2de5f", "f7597e3c58eeb9ce534993f53c982f2e91e6dd4d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 20:23:32 2023 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 20:23:32 2023 -0800" }, "message": "Merge tag \u0027drm-fixes-2023-02-17\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Just a final collection of misc fixes, the biggest disables the\n recently added dynamic debugging support, it has a regression that\n needs some bigger fixes.\n\n Otherwise a bunch of fixes across the board, vc4, amdgpu and vmwgfx\n mostly, with some smaller i915 and ast fixes.\n\n drm:\n - dynamic debug disable for now\n\n fbdev:\n - deferred i/o device close fix\n\n amdgpu:\n - Fix GC11.x suspend warning\n - Fix display warning\n\n vc4:\n - YUV planes fix\n - hdmi display fix\n - crtc reduced blanking fix\n\n ast:\n - fix start address computation\n\n vmwgfx:\n - fix bo/handle races\n\n i915:\n - gen11 WA fix\"\n\n* tag \u0027drm-fixes-2023-02-17\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/amd/display: Fail atomic_check early on normalize_zpos error\n drm/amd/amdgpu: fix warning during suspend\n drm/vmwgfx: Do not drop the reference to the handle too soon\n drm/vmwgfx: Stop accessing buffer objects which failed init\n drm/i915/gen11: Wa_1408615072/Wa_1407596294 should be on GT list\n drm: Disable dynamic debug as broken\n drm/ast: Fix start address computation\n fbdev: Fix invalid page access after closing deferred I/O devices\n drm/vc4: crtc: Increase setup cost in core clock calculation to handle extreme reduced blanking\n drm/vc4: hdmi: Always enable GCP with AVMUTE cleared\n drm/vc4: Fix YUV plane handling when planes are in different buffers\n" }, { "commit": "df71a42cc37a44cdc7682f57aecf14ff44391eed", "tree": "08187f8129d2193cf6e8922ef9d8528f83d49047", "parents": [ "c5a237a4db21ca7a28518c994def39d7bd62a0d1" ], "author": { "name": "Taichi Nishimura", "email": "awkrail01@gmail.com", "time": "Thu Feb 16 17:55:37 2023 +0900" }, "committer": { "name": "Andrii Nakryiko", "email": "andrii@kernel.org", "time": "Thu Feb 16 16:56:17 2023 -0800" }, "message": "Fix typos in selftest/bpf files\n\nRun spell checker on files in selftest/bpf and fixed typos.\n\nSigned-off-by: Taichi Nishimura \u003cawkrail01@gmail.com\u003e\nSigned-off-by: Andrii Nakryiko \u003candrii@kernel.org\u003e\nReviewed-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nLink: https://lore.kernel.org/bpf/20230216085537.519062-1-awkrail01@gmail.com\n" }, { "commit": "f7597e3c58eeb9ce534993f53c982f2e91e6dd4d", "tree": "7b5ad32e6c32c2d940d89967777600d697fa3243", "parents": [ "a2a04b515562bbb040e8ae3decae8c1e9ef05327", "d5a1224aa68c8b124a4c5c390186e571815ed390" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 17 09:49:12 2023 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 17 09:50:06 2023 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2023-02-16\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\n- Moving gen11 hw wa to the right place. (Matt)\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/Y+47eUvwbafER35/@intel.com\n" } ], "next": "c5a237a4db21ca7a28518c994def39d7bd62a0d1" }