)]}' { "log": [ { "commit": "52e4d5da6db788c3abc5bbb0e044e761b3540a30", "tree": "fa85034d35df564a9223c7337f2b40fe2dceb3a6", "parents": [ "0de607dc4fd80ede3b2a35e8a72f99c7a0bbc321", "00608d02ddf04b49c14801f4b0581b1b937bc766" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:50 2026 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:50 2026 +0100" }, "message": "Merge branch \u0027net-hinic3-pf-initialization\u0027\n\nFan Gong says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: hinic3: PF initialization\n\nThis is [2/3] part of hinic3 Ethernet driver second submission.\nWith this patch hinic3 becomes a complete Ethernet driver with\npf and vf.\n\nAdd cmdq detailed-response interfaces.\nAdd dump interfaces for cmdq, aeq, ceq and mailbox.\nAdd msg_send_lock for message sending concurrency.\nAdd PF device support and chip_present_flag to check cards.\nAdd rx vlan offload support.\nAdd PF FLR wait and timeout handling.\nAdd 5 ethtool ops for information of driver and link.\n\nv1: https://lore.kernel.org/netdev/cover.1771916043.git.zhuyikai1@h-partners.com/\nv2: https://lore.kernel.org/netdev/cover.1772697509.git.zhuyikai1@h-partners.com/\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/cover.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "00608d02ddf04b49c14801f4b0581b1b937bc766", "tree": "fa85034d35df564a9223c7337f2b40fe2dceb3a6", "parents": [ "330adcedd0035414b138635fd6b5f61f00cf419f" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:57 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add ethtool basic ops\n\nImplement following ethtool callback function:\n.get_link_ksettings\n.get_drvinfo\n.get_msglevel\n.set_msglevel\n.get_link\n\n These callbacks allow users to utilize ethtool for detailed\nnetwork configuration and monitoring.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/b56d490c2a06cae9541a0297d76b11d869f37161.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "330adcedd0035414b138635fd6b5f61f00cf419f", "tree": "3f5aa6372ceda552ac487a9859a859e2019f1afd", "parents": [ "33cf53672b6f386585998366c40369834f882ddb" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:56 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add PF/VF capability parsing and parameter validation\n\nAdd the ability to parse PF and VF capabilities and validate\nrelated parameters(SQ \u0026 RQ).\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/ac4733f2c0409bb778b4624ed1632dcb2ded6632.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "33cf53672b6f386585998366c40369834f882ddb", "tree": "518595ab67f729bbd54f2a2162b9446f1e2bdc2e", "parents": [ "3d36efc28078ef314445b8445b174f63bdf9579f" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:55 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add PF FLR wait and timeout handling\n\nAdd a mechanism for PF to wait for the completion of FLR, ensuring\nhardware state consistency after an FLR event.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/7a1b21426fd4274831733aca962eb209b806f4bd.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "3d36efc28078ef314445b8445b174f63bdf9579f", "tree": "4f2e54a59893602b2798d5a1c0f5cb39cb3e0d3a", "parents": [ "2a76f900d17dcb9e8322770ac9bcae34517805b3" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:54 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add PF device support and function type validation\n\nAdd PF device ID to support for PF devices in driver and enhance\nfunction type validation to ensure proper handling of both PF and\nVF.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/895cf7ac341c475e383aa8726039dc8ea3b96ffb.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "2a76f900d17dcb9e8322770ac9bcae34517805b3", "tree": "f0a66f8d8969847ba7dcf35fa27ab3ac41b14ffe", "parents": [ "0f746fc5bc77cb7421ce3f6611bd770db8c4cba8" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:53 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add msg_send_lock for message sending concurrecy\n\nAs send_mbox_msg is invoked by 3 functions: hinic3_send_mbox_to_mgmt,\nhinic3_response_mbox_to_mgmt and hinic3_send_mbox_to_mgmt_no_ack, only\nhinic3_response_mbox_to_mgmt does not has mutex and the other two has\nmbox-\u003embox_send_lock because their send actions are mutually exclusive.\n As hinic3_response_mbox_to_mgmt does not conflict with them in send\nactions but in mailbox resources, add the new mutex(msg_send_lock) in\nsend_mbox_msg to ensure message concurrency.\n\n Besdies, in mbox_send_seg change FIELD_PREP to FIELD_GET in\nMBOX_STATUS_FINISHED and MBOX_STATUS_SUCCESS to be more reasonable.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/d83f7f6eb4b5e94642a558fab75d61292c347e48.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "0f746fc5bc77cb7421ce3f6611bd770db8c4cba8", "tree": "198a41bb33c5114fda6f2ff31486bf092605a37c", "parents": [ "d69ee992fbf60dc691fed97bafcd2905c7e48832" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:52 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add RX VLAN offload support\n\nAdd vlan offload processing in RX process.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/22cf02a014c2beb7b5f92ab5e6de38c4dd928125.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "d69ee992fbf60dc691fed97bafcd2905c7e48832", "tree": "e8a803f3687344e37a0cecacb9691a952a6fb90d", "parents": [ "678c5b3b6b22f2b9851058e1624156b982891ae8" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:51 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add chip_present_flag checks to prevent errors when card is absent\n\nchip_present_flag is added for driver to prevent errors when card does\nnot exist. It has been added to multiple critical functions, including\ncommand queue, mailbox and network device operations, ensuring that the\nexistence of the network card is verified before performing operations.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/3954f22df125f5e843aaa62953d7506eb66922ac.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "678c5b3b6b22f2b9851058e1624156b982891ae8", "tree": "a80d2e644784525725798ba65e5c5b4a035459d0", "parents": [ "15b5be9389bef46884d9f970b643fedeea19105c" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:50 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add Command Queue/Async Event Queue/Complete Event Queue/Mailbox dump interfaces\n\nAdd dump interfaces for CMDQ, AEQ, CEQ and mailbox to enhance debugging\ncapabilities.\n Dump the WQE header for CMDQ.\n Dump the detailed queue information for AEQ and CEQ.\n Dump the related register status for mailbox.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/1644c5021e2059594e878812339ea025ed677f71.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "15b5be9389bef46884d9f970b643fedeea19105c", "tree": "214a5aedecc8cb019cc234a922138cbe27527235", "parents": [ "0de607dc4fd80ede3b2a35e8a72f99c7a0bbc321" ], "author": { "name": "Fan Gong", "email": "gongfan1@huawei.com", "time": "Tue Mar 10 09:04:49 2026 +0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 12:13:48 2026 +0100" }, "message": "hinic3: Add command queue detailed-response interfaces\n\nAdd new detailed response interfaces for the hinic3 command\nqueue (CMDQ), enhancing its functionality to handle commands\nrequiring input and output buffer pairs.\n\nCo-developed-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Zhu Yikai \u003czhuyikai1@h-partners.com\u003e\nSigned-off-by: Fan Gong \u003cgongfan1@huawei.com\u003e\nLink: https://patch.msgid.link/cc3cff8458aeb27b07749dc9dcee43c11c45a4c1.1773062356.git.zhuyikai1@h-partners.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "0de607dc4fd80ede3b2a35e8a72f99c7a0bbc321", "tree": "72878c52d5623288bd4e0ff528e6d6f9c73d146f", "parents": [ "17edc4e820bf8b4c7737c1de86c267e6974d543a" ], "author": { "name": "Alexander Graf", "email": "graf@amazon.com", "time": "Wed Mar 04 23:00:27 2026 +0000" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Mar 12 10:59:36 2026 +0100" }, "message": "vsock: add G2H fallback for CIDs not owned by H2G transport\n\nWhen no H2G transport is loaded, vsock currently routes all CIDs to the\nG2H transport (commit 65b422d9b61b (\"vsock: forward all packets to the\nhost when no H2G is registered\"). Extend that existing behavior: when\nan H2G transport is loaded but does not claim a given CID, the\nconnection falls back to G2H in the same way.\n\nThis matters in environments like Nitro Enclaves, where an instance may\nrun nested VMs via vhost-vsock (H2G) while also needing to reach sibling\nenclaves at higher CIDs through virtio-vsock-pci (G2H). With the old\ncode, any CID \u003e 2 was unconditionally routed to H2G when vhost was\nloaded, making those enclaves unreachable without setting\nVMADDR_FLAG_TO_HOST explicitly on every connect.\n\nRequiring every application to set VMADDR_FLAG_TO_HOST creates friction:\ntools like socat, iperf, and others would all need to learn about it.\nThe flag was introduced 6 years ago and I am still not aware of any tool\nthat supports it. Even if there was support, it would be cumbersome to\nuse. The most natural experience is a single CID address space where H2G\nonly wins for CIDs it actually owns, and everything else falls through to\nG2H, extending the behavior that already exists when H2G is absent.\n\nTo give user space at least a hint that the kernel applied this logic,\nautomatically set the VMADDR_FLAG_TO_HOST on the remote address so it\ncan determine the path taken via getpeername().\n\nAdd a per-network namespace sysctl net.vsock.g2h_fallback (default 1).\nAt 0 it forces strict routing: H2G always wins for CID \u003e VMADDR_CID_HOST,\nor ENODEV if H2G is not loaded.\n\nSigned-off-by: Alexander Graf \u003cgraf@amazon.com\u003e\nTested-by: syzbot@syzkaller.appspotmail.com\nReviewed-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nLink: https://patch.msgid.link/20260304230027.59857-1-graf@amazon.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "17edc4e820bf8b4c7737c1de86c267e6974d543a", "tree": "b1fce41669c6e65b68b07287d843aeb4256cd7b0", "parents": [ "dc9902bbd480aae510b885b67cd30cd04cfce3a8" ], "author": { "name": "Christophe Leroy (CS GROUP)", "email": "chleroy@kernel.org", "time": "Tue Mar 10 12:33:36 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 20:38:00 2026 -0700" }, "message": "net: Convert move_addr_to_user() to scoped user access\n\nmove_addr_to_user() is a critical functions that was converted to\nmasked user access by commit 1fb0e471611d (\"net: remove one stac/clac\npair from move_addr_to_user()\")\n\nConvert it to scoped user access to simplify the code.\n\nSigned-off-by: Christophe Leroy (CS GROUP) \u003cchleroy@kernel.org\u003e\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\nLink: https://patch.msgid.link/36d7f2e7f504d620c1b88526b25ebc89e3cb61d9.1773142315.git.chleroy@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "dc9902bbd480aae510b885b67cd30cd04cfce3a8", "tree": "99e19d12fa0d4ad8a3c1899bcb206382fc6ddfbc", "parents": [ "87aa0f539df0c190be7b565c1f32f9f90bf3869f" ], "author": { "name": "Wesley Atwell", "email": "atwellwea@gmail.com", "time": "Mon Mar 09 19:26:04 2026 -0600" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 20:20:03 2026 -0700" }, "message": "tcp: use WRITE_ONCE() for tsoffset in tcp_v6_connect()\n\nCommit dd23c9f1e8d5 (\"tcp: annotate data-races around tp-\u003etsoffset\")\nupdated do_tcp_getsockopt() to read tp-\u003etsoffset with READ_ONCE()\nfor TCP_TIMESTAMP because another CPU may change it concurrently.\n\ntcp_v6_connect() still stores tp-\u003etsoffset with a plain write. That\nstore runs under lock_sock() via inet_stream_connect(), but the socket\nlock does not serialize a concurrent getsockopt(TCP_TIMESTAMP) from\nanother task sharing the socket.\n\nUse WRITE_ONCE() for the tcp_v6_connect() store so the connect-time\nwriter matches the lockless TCP_TIMESTAMP reader. This also makes the\nIPv6 path consistent with tcp_v4_connect().\n\nSigned-off-by: Wesley Atwell \u003catwellwea@gmail.com\u003e\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Jiayuan Chen \u003cjiayuan.chen@shopee.com\u003e\nLink: https://patch.msgid.link/20260310012604.145661-1-atwellwea@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "87aa0f539df0c190be7b565c1f32f9f90bf3869f", "tree": "574c5809165bc5129e0aad42fdccd720a4f363ea", "parents": [ "bd4d6b955df2333511d7800a5cf9c672d3a9cad5", "f0bd19316663710de157e85afd62058312aa97e1" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:11:43 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:11:44 2026 -0700" }, "message": "Merge branch \u0027selftests-net-fix-cmd-process-timeout-handling\u0027\n\nGal Pressman says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nselftests: net: fix cmd.process() timeout handling\n\nPass the timeout argument correctly in cmd.process().\nAs Jakub noted, fixing the timeout broke the bpftrace() command\nin netpoll_basic.py, so fix it first.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260310115803.2521050-1-gal@nvidia.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "f0bd19316663710de157e85afd62058312aa97e1", "tree": "574c5809165bc5129e0aad42fdccd720a4f363ea", "parents": [ "82562972b85469e23fb787f78c1dea6ad6b16af4" ], "author": { "name": "Gal Pressman", "email": "gal@nvidia.com", "time": "Tue Mar 10 13:58:03 2026 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:11:40 2026 -0700" }, "message": "selftests: net: fix timeout passed as positional argument to communicate()\n\nThe cited commit refactored the hardcoded timeout\u003d5 into a parameter,\nbut dropped the keyword from the communicate() call.\nSince Popen.communicate()\u0027s first positional argument is \u0027input\u0027 (not\n\u0027timeout\u0027), the timeout value is silently treated as stdin input and the\ncall never enforces a timeout.\n\nPass timeout as a keyword argument to restore the intended behavior.\n\nReviewed-by: Nimrod Oren \u003cnoren@nvidia.com\u003e\nSigned-off-by: Gal Pressman \u003cgal@nvidia.com\u003e\nLink: https://patch.msgid.link/20260310115803.2521050-3-gal@nvidia.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "82562972b85469e23fb787f78c1dea6ad6b16af4", "tree": "f0522e50d1a6c1003c56797f9cbd211935b6c027", "parents": [ "bd4d6b955df2333511d7800a5cf9c672d3a9cad5" ], "author": { "name": "Gal Pressman", "email": "gal@nvidia.com", "time": "Tue Mar 10 13:58:02 2026 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:11:36 2026 -0700" }, "message": "selftests: net: pass bpftrace timeout to cmd()\n\nThe bpftrace() helper configures an interval based exit timer but does\nnot propagate the timeout to the cmd object, which defaults to 5\nseconds. Since the default BPFTRACE_TIMEOUT is 10 seconds, cmd.process()\nalways raises a TimeoutExpired exception before bpftrace has a chance to\nexit gracefully.\n\nPass timeout+5 to cmd() to allow bpftrace to complete gracefully.\n\nNote: this issue is masked by a bug in the way cmd() passes timeout,\nthis is fixed in the next commit.\n\nReviewed-by: Nimrod Oren \u003cnoren@nvidia.com\u003e\nSigned-off-by: Gal Pressman \u003cgal@nvidia.com\u003e\nLink: https://patch.msgid.link/20260310115803.2521050-2-gal@nvidia.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "bd4d6b955df2333511d7800a5cf9c672d3a9cad5", "tree": "8cfda0aa1d59ffbfe08da605af180b5c3bde82a9", "parents": [ "7e27d6202e90bc1c2ff16cd3118cb5456214ee42", "0ae998c4efd69fd5e331db7844b8cfaf07d47aea" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:00:37 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:00:38 2026 -0700" }, "message": "Merge branch \u0027net-macb-clean-up-several-member-settings-of-macb_config-instances\u0027\n\nKevin Hao says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: macb: Clean up several member settings of macb_config instances\n\nWhile debugging an issue in the macb driver, I noticed that many macb_config\ninstances have very similar member settings. This makes it difficult to\nidentify the actual differences between these instances. This patch series\naims to clean up some of these settings and clarify the specific configurations\nof each macb_config instance. No functional changes are introduced.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260310-macb-cleanup-v1-0-928c1a91a7dc@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "0ae998c4efd69fd5e331db7844b8cfaf07d47aea", "tree": "8cfda0aa1d59ffbfe08da605af180b5c3bde82a9", "parents": [ "9179711ee2f70f3ba1d56d5c2e9fce04fb754198" ], "author": { "name": "Kevin Hao", "email": "haokexin@gmail.com", "time": "Tue Mar 10 18:12:09 2026 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:00:33 2026 -0700" }, "message": "net: macb: Clean up the .usrio settings in macb_config instances\n\nAll instances of macb_config currently have the .usrio set, but most of\nthem use \u0026macb_default_usrio. In fact, there is no need to duplicate\nthis across all macb_config instances. Remove the .usrio setting from\ninstances that use \u0026macb_default_usrio, and ensure that the default is\nselected at runtime when no other value is explicitly set.\n\nSigned-off-by: Kevin Hao \u003chaokexin@gmail.com\u003e\nLink: https://patch.msgid.link/20260310-macb-cleanup-v1-3-928c1a91a7dc@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "9179711ee2f70f3ba1d56d5c2e9fce04fb754198", "tree": "8e30ef2ecb4c6ffe27788ce4895d9ee84768deaa", "parents": [ "f97977944d1587fc01736da8b138d7bb9c526d02" ], "author": { "name": "Kevin Hao", "email": "haokexin@gmail.com", "time": "Tue Mar 10 18:12:08 2026 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:00:33 2026 -0700" }, "message": "net: macb: Clean up the .init settings in macb_config instances\n\nAll instances of macb_config currently have the .init field set, but most\nof them use macb_init(). In fact, there is no need to duplicate this\nacross all macb_config instances. Introduce a new macb_init() function\nthat executes the specific .init if it is set; otherwise, it runs a\ndefault initialization function.\n\nSigned-off-by: Kevin Hao \u003chaokexin@gmail.com\u003e\nLink: https://patch.msgid.link/20260310-macb-cleanup-v1-2-928c1a91a7dc@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "f97977944d1587fc01736da8b138d7bb9c526d02", "tree": "4d996c90e7222fe599c1c9b94ee5400994a4d37a", "parents": [ "7e27d6202e90bc1c2ff16cd3118cb5456214ee42" ], "author": { "name": "Kevin Hao", "email": "haokexin@gmail.com", "time": "Tue Mar 10 18:12:07 2026 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 19:00:33 2026 -0700" }, "message": "net: macb: Clean up the .clk_init setting in the macb_config instances\n\nAll instances of macb_config currently have .clk_init set, but most of\nthem use macb_clk_init(). In fact, there is no need to duplicate this\nacross all macb_config instances. Introduce a new macb_clk_init()\nfunction that executes the specific .clk_init if it is set; otherwise,\nit runs the default clock initialization function.\n\nSigned-off-by: Kevin Hao \u003chaokexin@gmail.com\u003e\nLink: https://patch.msgid.link/20260310-macb-cleanup-v1-1-928c1a91a7dc@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7e27d6202e90bc1c2ff16cd3118cb5456214ee42", "tree": "bbe62e406a0a0dfd1d00144c8082ba04700548b5", "parents": [ "410593fec752f15c97062d2ded5e3a9dd654dcb2" ], "author": { "name": "Daniel Golle", "email": "daniel@makrotopia.org", "time": "Tue Mar 10 18:10:32 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 18:58:05 2026 -0700" }, "message": "selftests: net: local_termination: test link-local protocols\n\nAdd tests to local_termination.sh to verify that link-local frames\narrive. On some switches the DSA driver uses bridges to connect the\nuser ports to their CPU ports. More \"intelligent\" switches typically\ndon\u0027t forward link-local frames, but may trap them to an internal\nmicrocontroller. The driver may have to change trapping rules, so\nlink-local frames end up on the DSA CPU ports instead of being\nsilently dropped or trapped to the internal microcontroller of the\nswitch.\n\nAdd two tests which help to validate this has been done correctly:\n - Link-local STP BPDU should arrive at the Linux netdev when the\n bridge has STP disabled (BR_NO_STP), in which case the bridge\n forwards them rather than consuming them in the control plane\n - Link-local LLDP should arrive at standalone ports (and the test\n should be skipped on bridged ports similar to how it is done\n for the IEEE1588v2/PTP tests)\n\nSigned-off-by: Daniel Golle \u003cdaniel@makrotopia.org\u003e\nReviewed-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nTested-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nLink: https://patch.msgid.link/1a67081b2ede1e6d2d32f7dd54ae9688f3566152.1773166131.git.daniel@makrotopia.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "410593fec752f15c97062d2ded5e3a9dd654dcb2", "tree": "621eeb1eca3ba693e3ec27eb712a02fb2294d371", "parents": [ "34c0378b156f02f5fd8149f24a7aa75b255e8cda" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Mar 10 07:38:55 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 18:21:08 2026 -0700" }, "message": "tcp: add sysctl_tcp_shrink_window to netns_ipv4_sysctl.rst\n\nAdd missing entry for sysctl_tcp_shrink_window.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nLink: https://patch.msgid.link/20260310073855.564927-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "34c0378b156f02f5fd8149f24a7aa75b255e8cda", "tree": "2d22c3f2725ad0e306dfdc59af379d1e17052922", "parents": [ "482aac8b56ca21d06c588517970579474d56736e" ], "author": { "name": "Soichiro Ueda", "email": "the.latticeheart@gmail.com", "time": "Tue Mar 10 16:28:31 2026 +0900" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Mar 11 18:20:17 2026 -0700" }, "message": "selftests: af_unix: validate SO_PEEK_OFF advancement and reset\n\nExtend the so_peek_off selftest to ensure the socket peek offset is handled\ncorrectly after both MSG_PEEK and actual data consumption.\n\nVerify that the peek offset advances by the same amount as the number of\nbytes read when performing a read with MSG_PEEK.\n\nAfter exercising SO_PEEK_OFF via MSG_PEEK, drain the receive queue with a\nnon-peek recv() and verify that it can receive all the content in the\nbuffer and SO_PEEK_OFF returns back to 0.\n\nThe verification after actual data consumption was suggested by Miao Wang\nwhen the original so_peek_off selftest was introduced.\n\nLink: https://lore.kernel.org/all/7B657CC7-B5CA-46D2-8A4B-8AB5FB83C6DA@gmail.com/\nSuggested-by: Miao Wang \u003cshankerwangmiao@gmail.com\u003e\nReviewed-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nSigned-off-by: Soichiro Ueda \u003cthe.latticeheart@gmail.com\u003e\nLink: https://patch.msgid.link/20260310072832.127848-1-the.latticeheart@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "482aac8b56ca21d06c588517970579474d56736e", "tree": "2ed1beb79083c1e5e06b95ecd829906e8a88d8f4", "parents": [ "1a6ca6497a40f5c7795575a35f0da7d013b79bdd", "315bab9411f3bd3465a47a64a3e44323bfab60be" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:43 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:43 2026 -0700" }, "message": "Merge branch \u0027net-stmmac-start-to-shrink-memory-usage\u0027\n\nRussell King says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: stmmac: start to shrink memory usage\n\nStart shrinking stmmac\u0027s memory usage by avoiding using \"int\" for\nmembers that are only used for 0/1 (boolean) values, or values that\ncan\u0027t be larger than 255.\n\nIn addition, as struct stmmac_dma_cfg is approximately a cache line,\nshrinks below a cache line as a result of this patch set, and is\nrequired, there is no point separately allocating this from\nstruct plat_stmmacenet_data. Embed it into the end of this struct\nand set the existing pointer to avoid large wide-spread changes.\n\nLastly, add documentation for struct stmmac_dma_cfg, and document\nthe stmmac clocks as best we can given the driver history.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/aa6VEsmBK-S9eNYU@shell.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "315bab9411f3bd3465a47a64a3e44323bfab60be", "tree": "2ed1beb79083c1e5e06b95ecd829906e8a88d8f4", "parents": [ "9fe167ab790b10c9eb9ef82f46a03c83f9953b61" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:49 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:07 2026 -0700" }, "message": "net: stmmac: add documentation for clocks\n\nAdd documentation covering stmmac_clk, pclk, clk_ptp_ref and clk_tx_i\nin the hope that this will help understand what each of these clocks\nare for.\n\nThere is confusion around stmmac_clk and pclk which can\u0027t be easily\nresolved today as the Imagination Technologies Pistachio board that\npclk was introduced for has no public documentation and is likely now\nobsolete. So the origins of pclk are lost to the winds of time.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX5Z-0000000CVsb-1XTm@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "9fe167ab790b10c9eb9ef82f46a03c83f9953b61", "tree": "312a03832dac1e95db7d53ea58129b3bd9b5b144", "parents": [ "758ed85aadd0668c66cb359c63f384992b10938c" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:44 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:07 2026 -0700" }, "message": "net: stmmac: add documentation for stmmac_dma_cfg members\n\nAdd documentation of each of the struct stmmac_dma_cfg members. dche\nremains undocumented as I don\u0027t have documentation that covers this.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX5U-0000000CVsQ-162V@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "758ed85aadd0668c66cb359c63f384992b10938c", "tree": "0ab4dc0da1c4332fd749194c89c756f8fe44d613", "parents": [ "94808793fed71ee47741df0923d353024b6904ff" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:39 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:07 2026 -0700" }, "message": "net: stmmac: use u8 for host_dma_width and similar struct members\n\nWe aren\u0027t going to see \u003e\u003d 256-bit address busses soon, so reduce\nhost_dma_width and associated other struct members that initialise\nthis from u32 to u8.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nAcked-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e # qcom-ethqos\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX5P-0000000CVsK-0iwX@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "94808793fed71ee47741df0923d353024b6904ff", "tree": "f2feccae499cf348b8c1fb9b213365b59a9bdcb9", "parents": [ "3357642e65e9454c3da64b62c0ed987ee4010008" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:34 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:07 2026 -0700" }, "message": "net: stmmac: use u8 for ?x_queues_to_use and number_?x_queues\n\nThe maximum number of queues is a compile time constant of only eight.\nThis makes using a 32-bit quantity wastefulf. Instead, use u8 for\nthese and their associated variables.\n\nWhen reading the DT properties, saturdate at U8_MAX. Provided the core\nprovides DMA capabilities to describe the number of queues, this will\nbe capped by stmmac_hw_init() with a warning.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX5K-0000000CVsE-0J0Y@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "3357642e65e9454c3da64b62c0ed987ee4010008", "tree": "6876734357152f1f7fa58f5ecb9b4684cc5c1044", "parents": [ "c3d08424e025aaac8fb54134f76e611ef919cd08" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:28 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:06 2026 -0700" }, "message": "net: stmmac: reorder structs to reduce memory consumption\n\nReorder some of the stmmac structures to allow them to pack better,\nthereby using less memory. On aarch64, sizeof(struct stmmac_priv)\nwas 880, and with this change becomes 816, saving 64 bytes, which\nis an 8% saving.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX5E-0000000CVs8-40w4@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c3d08424e025aaac8fb54134f76e611ef919cd08", "tree": "44c876a121e19d160d43c645a1776638a240f320", "parents": [ "7a6387dec8cee5a237dc5092269e97028f5a983b" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:23 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:06 2026 -0700" }, "message": "net: stmmac: convert plat_stmmacenet_data booleans to type bool\n\nConvert members of struct plat_stmmacenet_data that are booleans to\ntype \u0027bool\u0027 and ensure their initialisers are true/false. Move the\nhas_xxx for the GMAC cores together, and move the COE members to the\nend of the list of bool to avoid unused holes in the struct.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nReviewed-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX59-0000000CVs2-3MHc@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7a6387dec8cee5a237dc5092269e97028f5a983b", "tree": "a694e7c7214b72efeb01f51ee7a9f3261c5ce0b8", "parents": [ "1a6ca6497a40f5c7795575a35f0da7d013b79bdd" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Mon Mar 09 09:39:18 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:54:06 2026 -0700" }, "message": "net: stmmac: provide plat_dat-\u003edma_cfg in stmmac_plat_dat_alloc()\n\nplat_dat-\u003edma_cfg is unconditionally required for the operation of the\ndriver, so it would make sense to allocate it along with the plat_dat.\n\nOn Arm64, sizeof(*plat_dat) has recently shrunk from 880 to 816 bytes\nand sizeof(*plat_dat-\u003edma_cfg) has shrunk from 32 to 20 bytes.\n\nGiven that dma_cfg is required, and it is now less than a cache line,\nIt doesn\u0027t make sense to allocate this separateny, so place it at the\nend of struct plat_stmmacenet_data, and set plat_dat-\u003edma_cfg to point\nat that to avoid mass changes.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nReviewed-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nTested-by: Mohd Ayaan Anwar \u003cmohd.anwar@oss.qualcomm.com\u003e\nLink: https://patch.msgid.link/E1vzX54-0000000CVrw-2jfu@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "1a6ca6497a40f5c7795575a35f0da7d013b79bdd", "tree": "cadd1c25b208db59514bc1a4c196d9233493ecb9", "parents": [ "07f56c8f54118cdbe33d566d9d4cc537c87d9202" ], "author": { "name": "Johan Hovold", "email": "johan@kernel.org", "time": "Mon Mar 09 09:26:41 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:53:19 2026 -0700" }, "message": "net: mdio: mvusb: drop redundant device reference\n\nDriver core holds a reference to the USB interface and its parent USB\ndevice while the interface is bound to a driver and there is no need to\ntake additional references unless the structures are needed after\ndisconnect.\n\nDrop the redundant device reference to reduce cargo culting, make it\neasier to spot drivers where an extra reference is needed, and reduce\nthe risk of memory leaks when drivers fail to release it.\n\nSigned-off-by: Johan Hovold \u003cjohan@kernel.org\u003e\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nLink: https://patch.msgid.link/20260309082641.15574-1-johan@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "07f56c8f54118cdbe33d566d9d4cc537c87d9202", "tree": "84640cb32f5a2b16548cdec58b5c7171582ef13a", "parents": [ "34bd3c6b0bd383a76d987c8c45c4f309b681b255", "7644e76956baa9a6bc3d208dfd92928f9ecd6a93" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:51:26 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:51:27 2026 -0700" }, "message": "Merge branch \u0027amd-xgbe-improve-power-management-for-s0i3\u0027\n\nRaju Rangoju says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\namd-xgbe: Improve power management for S0i3\n\nImprove the amd-xgbe power management handling to allow AMD platforms to\nreach the deepest suspend state (S0i3) when modern standby is used.\n\nThe first patch cleans up the xgbe_powerdown() and xgbe_powerup()\nhelpers by removing an unused caller distinction and aligning the\nordering of operations with xgbe_stop().\n\nThe second patch adds proper PCI power management operations, following\nthe standard PCI PM model, so that the device can be cleanly put into\nD3 and resumed back to D0. Without this, the amd_pmc driver reports:\n\n \"Last suspend didn\u0027t reach deepest state\"\n\nwhen the amd-xgbe driver is enabled.\n\nThese changes have been tested on AMD platforms using S0i3 modern\nstandby.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260308092851.1510214-1-Raju.Rangoju@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7644e76956baa9a6bc3d208dfd92928f9ecd6a93", "tree": "84640cb32f5a2b16548cdec58b5c7171582ef13a", "parents": [ "fe81629217e09ed8772e63a4c9cb0d864d849174" ], "author": { "name": "Raju Rangoju", "email": "Raju.Rangoju@amd.com", "time": "Sun Mar 08 14:58:51 2026 +0530" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:51:23 2026 -0700" }, "message": "amd-xgbe: add PCI power management for S0i3 support\n\nThe current suspend/resume implementation does not correctly handle PCI\ndevice power state transitions, which prevents AMD platforms from\nreaching the deepest suspend state (S0i3) when the amd-xgbe driver is\nenabled.\n\nIn particular, the amd_pmc driver reports:\n\n \"Last suspend didn\u0027t reach deepest state\"\n\nwhen this device is present.\n\nImplement proper PCI power management operations following the standard\nPCI PM model so that the device can be cleanly powered down and resumed.\n\nSuspend path:\n- Power down the network interface\n- Put the PHY into low-power mode\n- Disable bus mastering to prevent DMA activity\n- Save PCI configuration space\n- Disable the PCI device\n- Disable wake from D3 (S0i3 does not require Wake-on-LAN)\n- Set the device to D3hot\n\nResume path:\n- Restore the PCI power state to D0\n- Restore PCI configuration space\n- Enable the PCI device\n- Re-enable bus mastering\n- Re-enable device interrupts\n- Clear the PHY low-power mode\n- Power up the network interface\n\nThis allows systems using amd-xgbe to reach the deepest suspend state\nwhen entering modern standby (S0i3).\n\nSigned-off-by: Raju Rangoju \u003cRaju.Rangoju@amd.com\u003e\nLink: https://patch.msgid.link/20260308092851.1510214-3-Raju.Rangoju@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "fe81629217e09ed8772e63a4c9cb0d864d849174", "tree": "ca8639826a86bf4bc2cc00e64fa3140c2463b7aa", "parents": [ "34bd3c6b0bd383a76d987c8c45c4f309b681b255" ], "author": { "name": "Raju Rangoju", "email": "Raju.Rangoju@amd.com", "time": "Sun Mar 08 14:58:50 2026 +0530" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:51:22 2026 -0700" }, "message": "amd-xgbe: Simplify powerdown/powerup paths\n\nThe caller parameter in xgbe_powerdown() and xgbe_powerup() was intended\nto differentiate between driver and ioctl contexts, but the only\nremaining usage is from the driver suspend/resume path.\n\nSimplify this by:\n- Removing the unused XGMAC_DRIVER_CONTEXT and XGMAC_IOCTL_CONTEXT\n macros\n- Dropping the now-unused caller parameter\n- Reordering operations in xgbe_powerdown() to disable NAPI before\n stopping TX/RX, matching the order used in xgbe_stop()\n\nThis makes the powerdown/powerup paths easier to follow and keeps the\nordering consistent with the rest of the driver.\n\nSigned-off-by: Raju Rangoju \u003cRaju.Rangoju@amd.com\u003e\nLink: https://patch.msgid.link/20260308092851.1510214-2-Raju.Rangoju@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "34bd3c6b0bd383a76d987c8c45c4f309b681b255", "tree": "04a6399f04e1e115248aeace674bd128efd45333", "parents": [ "9278b888920ee8f3cea06622f04da681536b6601" ], "author": { "name": "Jiayuan Chen", "email": "jiayuan.chen@shopee.com", "time": "Mon Mar 09 20:39:16 2026 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:39:35 2026 -0700" }, "message": "net: sched: cls_u32: Avoid memcpy() false-positive warning in u32_init_knode()\n\nSyzbot reported a warning in u32_init_knode() [1].\n\nSimilar to commit 7cba18332e36 (\"net: sched: cls_u32: Avoid memcpy()\nfalse-positive warning\") which addressed the same issue in u32_change(),\nuse unsafe_memcpy() in u32_init_knode() to work around the compiler\u0027s\ninability to see into composite flexible array structs.\n\nThis silences the false-positive reported by syzbot:\n\n memcpy: detected field-spanning write (size 32) of single field\n \"\u0026new-\u003esel\" at net/sched/cls_u32.c:855 (size 16)\n\nSince the memory is correctly allocated with kzalloc_flex() using\ns-\u003enkeys, this is purely a false positive and does not need a Fixes tag.\n\n[1] https://syzkaller.appspot.com/bug?extid\u003dd5ace703ed883df56e42\n\nReported-by: syzbot+d5ace703ed883df56e42@syzkaller.appspotmail.com\nCloses: https://lore.kernel.org/all/69a811b9.a70a0220.b118c.0019.GAE@google.com/T/\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nAcked-by: Gustavo A. R. Silva \u003cgustavoars@kernel.org\u003e\nSigned-off-by: Jiayuan Chen \u003cjiayuan.chen@shopee.com\u003e\nLink: https://patch.msgid.link/20260309123917.402183-1-jiayuan.chen@linux.dev\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "9278b888920ee8f3cea06622f04da681536b6601", "tree": "eca2ddb52ee8c7b488adbbefce596b3e239a4b67", "parents": [ "1f9cab56e79eb88acec4d350c192d327244887c3" ], "author": { "name": "Yoshihiro Shimoda", "email": "yoshihiro.shimoda.uh@renesas.com", "time": "Sat Mar 07 10:55:32 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:37:21 2026 -0700" }, "message": "net: ethernet: ravb: Disable interrupts when closing device\n\nDisable E-MAC interrupts when closing the device.\n\nSigned-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n[Niklas: Rebase from BSP and reword commit message]\nSigned-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nLink: https://patch.msgid.link/20260307095532.2118495-1-niklas.soderlund+renesas@ragnatech.se\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "1f9cab56e79eb88acec4d350c192d327244887c3", "tree": "9385eee8178adaac672e3e3d63a7b59ebaed8dca", "parents": [ "7bb1970494faa6396fe4d622c4fe7edb1a9e217f" ], "author": { "name": "Eric Joyner", "email": "eric.joyner@amd.com", "time": "Fri Mar 06 13:56:34 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:34:14 2026 -0700" }, "message": "ionic: Report additional media types from firmware\n\nThe device firmware supports reporting more media types than what was\nthere in the past, so map these new media types to existing ethtool\nbits, which appears to be what other drivers do for media types that\nmatch speeds but not physical spec.\n\nAnd while here, make a very small cleanup in ionic_get_link_ksettings()\nto remove some unnecessary code duplication.\n\nReviewed-by: Brett Creeley \u003cbrett.creeley@amd.com\u003e\nSigned-off-by: Eric Joyner \u003ceric.joyner@amd.com\u003e\nLink: https://patch.msgid.link/20260306215634.64550-1-eric.joyner@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7bb1970494faa6396fe4d622c4fe7edb1a9e217f", "tree": "c42567f99ee9fe10b9c7404e651253ee90c9c888", "parents": [ "73a864352570fd30d942652f05bfe9340d7a2055", "d6df5e9b2a565be08330e46a8a615aac9ed8711b" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:33:07 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:33:07 2026 -0700" }, "message": "Merge branch \u0027tools-ynl-policy-query-support\u0027\n\nJakub Kicinski says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ntools: ynl: policy query support\n\nImprove the Netlink policy support in YNL. This series grew out of\nimprovements to policy checking, when writing selftests I realized\nthat instead of doing all the policy parsing in the test we\u0027re\nbetter off making it part of YNL itself.\n\nPatch 1 adds pad handling, apparently we never hit pad with commonly\nused families. nlctrl policy dumps use pad more frequently.\nPatch 2 is a trivial refactor.\nPatch 3 pays off some technical debt in terms of documentation.\nThe YnlFamily class is growing in size and it\u0027s quite hard to\nfind its members. So document it a little bit.\nPatch 4 is the main dish, the implementation of get_policy(op)\nin YnlFamily.\nPatch 5 plugs the new functionality into the CLI.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260310005337.3594225-1-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "d6df5e9b2a565be08330e46a8a615aac9ed8711b", "tree": "c42567f99ee9fe10b9c7404e651253ee90c9c888", "parents": [ "77a6401a8722be20ea8db98ac900c93ccc7068ff" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:53:37 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:33:00 2026 -0700" }, "message": "tools: ynl: cli: add --policy support\n\nAdd --policy flag which can be combined with --do or --dump to query\nthe kernel\u0027s netlink policy for an operation instead of executing it.\n\nExamples:\n\n $ ynl --family netdev --do dev-get --policy\n {\u0027ifindex\u0027: {\u0027max-value\u0027: 4294967295, \u0027min-value\u0027: 1, \u0027type\u0027: \u0027u32\u0027}}\n\n $ ynl --family ethtool --do channels-get --policy --output-json\n {\"header\": {\"type\": \"nested\", \"policy\": {\"dev-index\": ...}}}\n\n $ ynl --family netdev --dump dev-get --policy\n {}\n\nLink: https://patch.msgid.link/20260310005337.3594225-6-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "77a6401a8722be20ea8db98ac900c93ccc7068ff", "tree": "32364d49f98a8ce5b9900be87f28ca1c011a1838", "parents": [ "8bbcfce5db97abc6ca2066b540e88702f461128b" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:53:36 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:32:46 2026 -0700" }, "message": "tools: ynl: add Python API for easier access to policies\n\nThe format of Netlink policy dump is a bit curious with messages\nin the same dump carrying both attrs and mapping info. Plus each\nmessage carries a single piece of the puzzle the caller must then\nreassemble.\n\nI need to do this reassembly for a test, but I think it\u0027s generally\nuseful. So let\u0027s add proper support to YnlFamily to return more\nuser-friendly representation. See the various docs in the patch\nfor more details.\n\nLink: https://patch.msgid.link/20260310005337.3594225-5-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "8bbcfce5db97abc6ca2066b540e88702f461128b", "tree": "7338ead293ebdd18b1ac71e04b1507c61e64d567", "parents": [ "c26fda6212b88af1e667474728dd241ebf6ba1d1" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:53:35 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:30:03 2026 -0700" }, "message": "tools: ynl: add short doc to class YnlFamily\n\nThe class is quite long. It\u0027s getting hard to find the user-facing\nmethods. Add a short doc at the class level explaining the main API.\n\nLink: https://patch.msgid.link/20260310005337.3594225-4-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c26fda6212b88af1e667474728dd241ebf6ba1d1", "tree": "0eb7ebd02f21af3ff8e0e79c8d0f051005fb4037", "parents": [ "7b1309c33927d126d1cc47ddaf33bf2ae86f000c" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:53:34 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:30:03 2026 -0700" }, "message": "tools: ynl: move policy decoding out of NlMsg\n\nWe\u0027ll soon need to decode policies from dump so move _decode_policy()\nout of class NlMsg.\n\nLink: https://patch.msgid.link/20260310005337.3594225-3-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7b1309c33927d126d1cc47ddaf33bf2ae86f000c", "tree": "b56e8217e9dbd13ebcc704dae14c29a7894e1802", "parents": [ "73a864352570fd30d942652f05bfe9340d7a2055" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:53:33 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:30:03 2026 -0700" }, "message": "tools: ynl: handle pad type during decode\n\nApparently Python code only handled the \u0027pad\u0027 type in structs\nuntil now. Add it to attr decoding. nlctrl policy dumps need it.\n\nLink: https://patch.msgid.link/20260310005337.3594225-2-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "73a864352570fd30d942652f05bfe9340d7a2055", "tree": "4eed98c67248a9a419bd078279176fae16014f65", "parents": [ "690043b95c1804cccce8ae6a6677a6b5de33ca77" ], "author": { "name": "Rosen Penev", "email": "rosenp@gmail.com", "time": "Fri Mar 06 19:17:09 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:02:25 2026 -0700" }, "message": "net: mvneta: support EPROBE_DEFER when reading MAC address\n\nIf nvmem loads after the ethernet driver, mac address assignments will\nnot take effect. of_get_ethdev_address returns EPROBE_DEFER in such a\ncase so we need to handle that to avoid eth_hw_addr_random.\n\nAdd extra goto section to just free stats as they are allocated right\nabove.\n\nSigned-off-by: Rosen Penev \u003crosenp@gmail.com\u003e\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nLink: https://patch.msgid.link/20260307031709.640141-1-rosenp@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "690043b95c1804cccce8ae6a6677a6b5de33ca77", "tree": "6ab4a76c0927c3378a6e421b283e6bb87174c47b", "parents": [ "7da62262ec96a4b345d207b6bcd2ddf5231b7f7d" ], "author": { "name": "Dimitri Daskalakis", "email": "dimitri.daskalakis1@gmail.com", "time": "Mon Mar 09 13:42:15 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 19:01:51 2026 -0700" }, "message": "selftests: drv-net: rss: Add retries to test_rss_key_indir to reduce flakes\n\nThe test generates 16 flows, and verifies that traffic is distributed\nacross two queues via the NICs RSS indirection table. The likelihood of the\nflows skewing to a single queue is high, so we retry sending traffic up to\n3 times.\n\nAlternatively, we could increase the number of generated flows. But\ndebug kernels may struggle to ramp this many flows.\n\nDuring manual testing, the test passed for 10,000 consecutive runs.\n\nSigned-off-by: Dimitri Daskalakis \u003cdimitri.daskalakis1@gmail.com\u003e\nReviewed-by: Pavan Chebbi \u003cpavan.chebbi@broadcom.com\u003e\nLink: https://patch.msgid.link/20260309204215.2110486-1-dimitri.daskalakis1@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7da62262ec96a4b345d207b6bcd2ddf5231b7f7d", "tree": "90ceca3b452b376c2c57496778829de62be18260", "parents": [ "ae95cbaedb754633b9c0d35712a024b16a70962e" ], "author": { "name": "Fernando Fernandez Mancera", "email": "fmancera@suse.de", "time": "Mon Mar 09 03:39:45 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:59:39 2026 -0700" }, "message": "inet: add ip_local_port_step_width sysctl to improve port usage distribution\n\nWith the current port selection algorithm, ports after a reserved port\nrange or long time used port are used more often than others [1]. This\ncauses an uneven port usage distribution. This combines with cloud\nenvironments blocking connections between the application server and the\ndatabase server if there was a previous connection with the same source\nport, leading to connectivity problems between applications on cloud\nenvironments.\n\nThe real issue here is that these firewalls cannot cope with\nstandards-compliant port reuse. This is a workaround for such situations\nand an improvement on the distribution of ports selected.\n\nThe proposed solution is to implement a variant of RFC 6056 Algorithm 5.\nThe step size is selected randomly on every connect() call ensuring it\nis a coprime with respect to the size of the range of ports we want to\nscan. This way, we can ensure that all ports within the range are\nscanned before returning an error. To enable this algorithm, the user\nmust configure the new sysctl option \"net.ipv4.ip_local_port_step_width\".\n\nIn addition, on graphs generated we can observe that the distribution of\nsource ports is more even with the proposed approach. [2]\n\n[1] https://0xffsoftware.com/port_graph_current_alg.html\n\n[2] https://0xffsoftware.com/port_graph_random_step_alg.html\n\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: Fernando Fernandez Mancera \u003cfmancera@suse.de\u003e\nLink: https://patch.msgid.link/20260309023946.5473-2-fmancera@suse.de\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "ae95cbaedb754633b9c0d35712a024b16a70962e", "tree": "eaa4d874855353e410c6e8126ae34296157f5e98", "parents": [ "b8a0e5eb6a126a641ab8768e825756a66848ca29", "87fdf57ded3d3e83b90cffb160fce8e2a914142a" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:54:25 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:54:25 2026 -0700" }, "message": "Merge branch \u0027selftests-rds-ksft-cleanups\u0027\n\nAllison Henderson says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nselftests: rds: ksft cleanups\n\nThis set addresses a few rds selftests clean ups and bugs encountered\nwhen running in the ksft framework. The first patch is a clean up\npatch that addresses pylint warnings, but otherwise no functional\nchanges. The next patch moves the test time out to a ksft settings\nfile so that the time out is set appropriately. And lastly we fix a\ntcpdump segfault caused by deprecated a os.fork() call.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260308055835.1338257-1-achender@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "87fdf57ded3d3e83b90cffb160fce8e2a914142a", "tree": "eaa4d874855353e410c6e8126ae34296157f5e98", "parents": [ "b873b4e16042fac1244499dab5a72373d25ce051" ], "author": { "name": "Allison Henderson", "email": "achender@kernel.org", "time": "Sat Mar 07 22:58:35 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:54:24 2026 -0700" }, "message": "selftests: rds: Fix tcpdump segfault in rds selftests\n\nnet/rds/test.py sees a segfault in tcpdump when executed through the\nksft runner.\n\n[ 21.903713] tcpdump[1469]: segfault at 0 ip 000072100e99126d\nsp 00007ffccf740fd0 error 4\n[ 21.903721] in libc.so.6[16a26d,7798b149a000+188000]\n[ 21.905074] in libc.so.6[16a26d,72100e84f000+188000] likely on\nCPU 5 (core 5, socket 0)\n[ 21.905084] Code: 00 0f 85 a0 00 00 00 48 83 c4 38 89 d8 5b 41 5c\n41 5d 41 5e 41 5f 5d c3 0f 1f 44 00 00 48 8b 05 91 8b 09 00 8b 4d ac\n64 89 08 \u003c41\u003e 0f b6 07 83 e8 2b a8 fd 0f 84 54 ff ff ff 49 8b 36 4c 89\nff e8\n[ 21.906760] likely on CPU 9 (core 9, socket 0)\n[ 21.913469] Code: 00 0f 85 a0 00 00 00 48 83 c4 38 89 d8 5b 41 5c 41\n5d 41 5e 41 5f 5d c3 0f 1f 44 00 00 48 8b 05 91 8b 09 00 8b 4d ac 64 89\n08 \u003c41\u003e 0f b6 07 83 e8 2b a8 fd 0f 84 54 ff ff ff 49 8b 36 4c 89 ff e8\n\nThe os.fork() call creates extra complexity because it forks the entire\nprocess including the python interpreter. ip() then calls cmd() which\ncreates a subprocess.Popen. We can avoid the extra layering by simply\ncalling subprocess.Popen directly. Track the process handles directly\nand terminate them at cleanup rather than relying on killall. Further\ntcpdump\u0027s -Z flag attempts to change savefile ownership, which is not\nsupported by the 9p protocol. Fix this by writing pcap captures to\n\"/tmp\" during the test and move them to the log directory after tcpdump\nexits.\n\nSigned-off-by: Allison Henderson \u003cachender@kernel.org\u003e\nLink: https://patch.msgid.link/20260308055835.1338257-4-achender@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "b873b4e16042fac1244499dab5a72373d25ce051", "tree": "d16fe204bb66e3f5fcaa91a7d2bb55e1a37c86de", "parents": [ "5a0c5702bd0016ce761dfa2ce84c009a3a32d863" ], "author": { "name": "Allison Henderson", "email": "achender@kernel.org", "time": "Sat Mar 07 22:58:34 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:54:15 2026 -0700" }, "message": "selftests: rds: Add ksft timeout\n\nrds/run.sh sets a timer of 400s when calling test.py. However when\ntests are run through ksft, a default 45s timer is applied. Fix this\nby adding a ksft timeout in tools/testing/selftests/net/rds/settings\n\nSigned-off-by: Allison Henderson \u003cachender@kernel.org\u003e\nLink: https://patch.msgid.link/20260308055835.1338257-3-achender@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5a0c5702bd0016ce761dfa2ce84c009a3a32d863", "tree": "8945e55936497ae883781e012d677821330b904f", "parents": [ "b8a0e5eb6a126a641ab8768e825756a66848ca29" ], "author": { "name": "Allison Henderson", "email": "achender@kernel.org", "time": "Sat Mar 07 22:58:33 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 18:53:42 2026 -0700" }, "message": "selftests: rds: Fix pylint warnings\n\nTidy up all exiting pylint errors in test.py. No functional\nchanges are introduced in this patch\n\nSigned-off-by: Allison Henderson \u003cachender@kernel.org\u003e\nLink: https://patch.msgid.link/20260308055835.1338257-2-achender@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "b8a0e5eb6a126a641ab8768e825756a66848ca29", "tree": "49dc868fe4d556cd5a1d78dca4153c3a5d0ba119", "parents": [ "16767c72a40f524fbd0441386f2c3b6f5ff283b0" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Sat Mar 07 09:59:16 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 17:54:02 2026 -0700" }, "message": "tools: ynl: cli: order set-\u003elist conversion in JSON output\n\nNIPA tries to make sure that HW tests don\u0027t modify system state.\nIt dumps some well known configs before and after the test and\ncompares the outputs.\n\nMake sure that YNL json output is stable. Converting sets to lists\nwith a naive list(o) results in a random order.\n\nLink: https://patch.msgid.link/20260307175916.1652518-1-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "16767c72a40f524fbd0441386f2c3b6f5ff283b0", "tree": "bce9aba5da1c6a5ad11dcd92e6e358a689e9da24", "parents": [ "05e059510edf7b6101ff85db2a2e2d1b6f31ee6d", "aa5ec9d03b9c4459e528ecd75d84f6ef98fb2f5a" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 17:53:10 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 17:53:10 2026 -0700" }, "message": "Merge branch \u0027smc-sysctl-formatting-and-missing-entries\u0027\n\nKyoji Ogasawara says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nsmc-sysctl formatting and missing entries\n\nupdate SMC sysctl documentation in two small steps.\n\n- patch 1 fixes indentation in the smcr_buf_type section\n- patch 2 documents missing sysctl parameters limit_smc_hs and hs_ctrl,\n including values/defaults and hs_ctrl usage notes\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260309124541.22723-1-sawara04.o@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "aa5ec9d03b9c4459e528ecd75d84f6ef98fb2f5a", "tree": "bce9aba5da1c6a5ad11dcd92e6e358a689e9da24", "parents": [ "4a51ac9056c17e8216b245cd16152eefbd21abfb" ], "author": { "name": "Kyoji Ogasawara", "email": "sawara04.o@gmail.com", "time": "Mon Mar 09 21:45:40 2026 +0900" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 17:53:03 2026 -0700" }, "message": "net/smc: Add documentation for limit_smc_hs and hs_ctrl\n\nDocument missing SMC sysctl parameters limit_smc_hs and hs_ctrl\n\nSigned-off-by: Kyoji Ogasawara \u003csawara04.o@gmail.com\u003e\nReviewed-by: D. Wythe\u003calibuda@linux.alibaba.com\u003e\nLink: https://patch.msgid.link/20260309124541.22723-3-sawara04.o@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "4a51ac9056c17e8216b245cd16152eefbd21abfb", "tree": "01815cba27e6d5058d8190b0471a8497298e0d16", "parents": [ "05e059510edf7b6101ff85db2a2e2d1b6f31ee6d" ], "author": { "name": "Kyoji Ogasawara", "email": "sawara04.o@gmail.com", "time": "Mon Mar 09 21:45:39 2026 +0900" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Mar 10 17:52:36 2026 -0700" }, "message": "net/smc: fix indentation in smcr_buf_type section\n\nsmcr_buf_type section used inconsistent indentation compared\nwith the rest of this document.\n\nSigned-off-by: Kyoji Ogasawara \u003csawara04.o@gmail.com\u003e\nReviewed-by: D. Wythe\u003calibuda@linux.alibaba.com\u003e\nLink: https://patch.msgid.link/20260309124541.22723-2-sawara04.o@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "05e059510edf7b6101ff85db2a2e2d1b6f31ee6d", "tree": "409d771c2b7f3e055f4051e9aa307540f76bdb39", "parents": [ "89fe91c65992a37863241e35aec151210efc53ce", "8e5218199da48913960a0ce7e22193020dc23891" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:54 2026 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:55 2026 +0100" }, "message": "Merge branch \u0027eth-fbnic-add-fbnic-self-tests\u0027\n\nMike Marciniszyn says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\neth fbnic: Add fbnic self tests\n\nFrom: \"Mike Marciniszyn (Meta)\" \u003cmike.marciniszyn@gmail.com\u003e\n\nThis series adds self tests to test the registers, the\nmsix interrupts, the tlv, and the firmware mailbox.\n\nThis series assumes that the\n[PATCH net-next 0/2] Add debugfs hooks [1]\nis present.\n\nWhen the self tests are run the with ethtool -t:\n\n ethtool -t eth0\n The test result is PASS\n The test extra info:\n Register test (offline) 0\n MSI-X Interrupt test (offline) 0\n FW mailbox test (on/offline) 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://patch.msgid.link/20260307105847.1438-1-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "8e5218199da48913960a0ce7e22193020dc23891", "tree": "409d771c2b7f3e055f4051e9aa307540f76bdb39", "parents": [ "d522b1b004800728de5466451e4d7032a4f53de6" ], "author": { "name": "Mike Marciniszyn (Meta)", "email": "mike.marciniszyn@gmail.com", "time": "Sat Mar 07 05:58:47 2026 -0500" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:53 2026 +0100" }, "message": "eth fbnic: Add mailbox self test\n\nThe mailbox self test ensures the interface to and from\nthe firmware is healthy by sending a test message and\nfielding the response from the firmware.\n\nThis patch uses the new completion API [1][2] that allocates a\ncompletion structure, binds the completion to the TEST\nmessage, and uses a new FW parsing routine that wraps the\ncompletion processing around the TLV parser.\n\nLink: https://patch.msgid.link/20250516164804.741348-1-lee@trager.us [1]\nLink: https://patch.msgid.link/20260115003353.4150771-6-mohsin.bashr@gmail.com [2]\n\nSigned-off-by: Mike Marciniszyn (Meta) \u003cmike.marciniszyn@gmail.com\u003e\nLink: https://patch.msgid.link/20260307105847.1438-6-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "d522b1b004800728de5466451e4d7032a4f53de6", "tree": "975edff9da3ba8d06f4a00392a5ef9a79d734730", "parents": [ "99fc8d3d00c94e32c1a8b60783a0d24421053446" ], "author": { "name": "Mike Marciniszyn (Meta)", "email": "mike.marciniszyn@gmail.com", "time": "Sat Mar 07 05:58:46 2026 -0500" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:53 2026 +0100" }, "message": "eth fbnic: TLV support for use by MBX self test\n\nThe TLV (Type-Value-Length) self uses a known set of data to create a\nTLV message. These routines support the MBX self test by creating\nthe test messages and parsing the response message coming back\nfrom the firmware.\n\nSigned-off-by: Mike Marciniszyn (Meta) \u003cmike.marciniszyn@gmail.com\u003e\nLink: https://patch.msgid.link/20260307105847.1438-5-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "99fc8d3d00c94e32c1a8b60783a0d24421053446", "tree": "6e09f934f5eb24a7b72637edc807761581d6d08f", "parents": [ "b43498b7e9be0ef76440980642b2bf20dc213e19" ], "author": { "name": "Mike Marciniszyn (Meta)", "email": "mike.marciniszyn@gmail.com", "time": "Sat Mar 07 05:58:45 2026 -0500" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:53 2026 +0100" }, "message": "eth fbnic: Add msix self test\n\nThis function is meant to test the global interrupt registers and the\nPCIe IP MSI-X functionality. It essentially goes through and tests\nvarious combinations of the set, clear, and mask bits in order to\nverify the behavior is as we expect it to be from the driver.\n\nSigned-off-by: Mike Marciniszyn (Meta) \u003cmike.marciniszyn@gmail.com\u003e\nLink: https://patch.msgid.link/20260307105847.1438-4-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "b43498b7e9be0ef76440980642b2bf20dc213e19", "tree": "0fc7b677a5c9f09f46c1464a874d61111e30be71", "parents": [ "3fdd33697c2be9184668c89ba4f24a5ecbc8ec51" ], "author": { "name": "Mike Marciniszyn (Meta)", "email": "mike.marciniszyn@gmail.com", "time": "Sat Mar 07 05:58:44 2026 -0500" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:53 2026 +0100" }, "message": "eth fbnic: Add register self test\n\nThe register test will be used to verify hardware is behaving as expected.\n\nThe test itself will have us writing to registers that should have no\nside effects due to us resetting after the test has been completed.\n\nWhile the test is being run the interface should be offline.\n\nThis patch counts on the first patch of this series to export netif_open()\nand also ensures that the half close calls netif_close() to\navoid deadlock.\n\nSigned-off-by: Mike Marciniszyn (Meta) \u003cmike.marciniszyn@gmail.com\u003e\nLink: https://patch.msgid.link/20260307105847.1438-3-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "3fdd33697c2be9184668c89ba4f24a5ecbc8ec51", "tree": "8dd2190ce5b1924bc6ad3dc9b531aa01c88cc197", "parents": [ "89fe91c65992a37863241e35aec151210efc53ce" ], "author": { "name": "Mike Marciniszyn (Meta)", "email": "mike.marciniszyn@gmail.com", "time": "Sat Mar 07 05:58:43 2026 -0500" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:53:52 2026 +0100" }, "message": "net: export netif_open for self_test usage\n\ndev_open() already is exported, but drivers which use the netdev\ninstance lock need to use netif_open() instead. netif_close() is\nalso already exported [1] so this completes the pairing.\n\nThis export is required for the following fbnic self tests to\navoid calling ndo_stop() and ndo_open() in favor of the\nmore appropriate netif_open() and netif_close() that notifies\nany listeners that the interface went down to test and is now\ncoming back up.\n\nLink: https://patch.msgid.link/20250309215851.2003708-1-sdf@fomichev.me [1]\nSigned-off-by: Mike Marciniszyn (Meta) \u003cmike.marciniszyn@gmail.com\u003e\nLink: https://patch.msgid.link/20260307105847.1438-2-mike.marciniszyn@gmail.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "89fe91c65992a37863241e35aec151210efc53ce", "tree": "a9c672afb386793a033ce310bbb7a7ed643e5d10", "parents": [ "46097d011f77f5758fb47b7059b4f1f2e7403940" ], "author": { "name": "Erni Sri Satya Vennela", "email": "ernis@linux.microsoft.com", "time": "Fri Mar 06 13:12:06 2026 -0800" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:39:51 2026 +0100" }, "message": "net: mana: hardening: Validate doorbell ID from GDMA_REGISTER_DEVICE response\n\nAs a part of MANA hardening for CVM, add validation for the doorbell\nID (db_id) received from hardware in the GDMA_REGISTER_DEVICE response\nto prevent out-of-bounds memory access when calculating the doorbell\npage address.\n\nIn mana_gd_ring_doorbell(), the doorbell page address is calculated as:\n addr \u003d db_page_base + db_page_size * db_index\n \u003d (bar0_va + db_page_off) + db_page_size * db_index\n\nA hardware could return values that cause this address to fall outside\nthe BAR0 MMIO region. In Confidential VM environments, hardware responses\ncannot be fully trusted.\n\nAdd the following validations:\n- Store the BAR0 size (bar0_size) in gdma_context during probe.\n- Validate the doorbell page offset (db_page_off) read from device\n registers does not exceed bar0_size during initialization, converting\n mana_gd_init_registers() to return an error code.\n- Validate db_id from GDMA_REGISTER_DEVICE response against the\n maximum number of doorbell pages that fit within BAR0.\n\nSigned-off-by: Erni Sri Satya Vennela \u003cernis@linux.microsoft.com\u003e\nLink: https://patch.msgid.link/20260306211212.543376-1-ernis@linux.microsoft.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "46097d011f77f5758fb47b7059b4f1f2e7403940", "tree": "8e7b7b052bc2ab4728d1084bfc631e51359761ea", "parents": [ "52ede1bce557c66309f41ac29dd190be23ca9129" ], "author": { "name": "Lorenzo Bianconi", "email": "lorenzo@kernel.org", "time": "Fri Mar 06 16:09:47 2026 +0100" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Mar 10 13:23:07 2026 +0100" }, "message": "net: airoha: Move GDM forward port configuration in ndo_open/ndo_stop callbacks\n\nThis change allows to set GDM forward port configuration to\nFE_PSE_PORT_DROP stopping the network device. Hw design requires to stop\npacket forwarding putting the interface down. Moreover, PPE firmware\nrequires to use PPE1 for GDM3 or GDM4.\n\nSigned-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nLink: https://patch.msgid.link/20260306-airoha-gdm-forward-ndo-open-stop-v1-1-7b7a20dd9ef0@kernel.org\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "52ede1bce557c66309f41ac29dd190be23ca9129", "tree": "2038eebd9e9451de3c0126810b2dfddb8c032c14", "parents": [ "c127d4087930cc4706b99c9431e35ba00776a634", "687e7863f027426175791f8a23a59b7c4c816fe9" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:45:31 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:45:31 2026 -0700" }, "message": "Merge branch \u0027net-stmmac-further-ptp-cleanups\u0027\n\nRussell King says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: stmmac: further ptp cleanups\n\nThe first uses a local variable when setting n_ext_ts which is a minor\nsimplification of the code. The second removes the now unnecessary\n\"available\" flag for the PPS outputs.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/aawDiK7DjcSXSs1X@shell.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "687e7863f027426175791f8a23a59b7c4c816fe9", "tree": "2038eebd9e9451de3c0126810b2dfddb8c032c14", "parents": [ "b560d4434f98b5a8f7d04e5ec1a515a21aa8d914" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Sat Mar 07 10:53:20 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:45:27 2026 -0700" }, "message": "net: stmmac: ptp: remove redundant priv-\u003epps[].available\n\npriv-\u003epps[].available is set in stmmac_ptp_register() for all PPS\noutputs reported by hardware up to STMMAC_PPS_MAX.\n\nSince we now set priv-\u003eptp_clock_ops.n_per_out to the number of PPS\noutputs that both the hardware and driver can support to prevent\narray overflow in stmmac_enable(), this makes priv-\u003epps[].available\nredundant. Remove this struct member.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nLink: https://patch.msgid.link/E1vypHc-0000000CSbl-1X6v@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "b560d4434f98b5a8f7d04e5ec1a515a21aa8d914", "tree": "587e511c82de46068f605996cd9df83f282aa971", "parents": [ "c127d4087930cc4706b99c9431e35ba00776a634" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Sat Mar 07 10:53:15 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:45:26 2026 -0700" }, "message": "net: stmmac: ptp: rearrange n_ext_ts initialisation\n\nUse local variables for n_ext_ts rather than referencing the DMA\ncapability several times.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nLink: https://patch.msgid.link/E1vypHX-0000000CSbc-123K@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c127d4087930cc4706b99c9431e35ba00776a634", "tree": "c9e96747d51046a65875a81a3d80c507464bee72", "parents": [ "56acc7f51974c824142dd0a529b1ce05ec5fa75a" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Sat Mar 07 10:55:08 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:44:11 2026 -0700" }, "message": "net: stmmac: remove stmmac_dwmac4_get_mac_addr()\n\nstmmac_dwmac4_get_mac_addr() is identical to stmmac_get_mac_addr().\nRemove stmmac_dwmac4_get_mac_addr() to avoid this code duplication.\n\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\nLink: https://patch.msgid.link/E1vypJM-0000000CSiJ-48yO@rmk-PC.armlinux.org.uk\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "56acc7f51974c824142dd0a529b1ce05ec5fa75a", "tree": "e851648d3d965774dc02c440a228eaa9b832d771", "parents": [ "d6d4ff335db2d9242937ca474d292010acd35c38" ], "author": { "name": "Victor Nogueira", "email": "victor@mojatatu.com", "time": "Sat Mar 07 19:07:24 2026 -0300" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:42:19 2026 -0700" }, "message": "selftests/tc-testing: Adapt test\u0027s output to HFSC\u0027s iproute2 printing changes\n\nTo make the printing of HFSC\u0027s defcls consistent with HTB\u0027s,\niproute2 is now printing defcls prepended with \"0x\".\n\nThis commit adapts test a4c3 to this change.\n\nAcked-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nSigned-off-by: Victor Nogueira \u003cvictor@mojatatu.com\u003e\nLink: https://patch.msgid.link/20260307220724.2501212-1-victor@mojatatu.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "d6d4ff335db2d9242937ca474d292010acd35c38", "tree": "37e0f0abcecb0a6a48fdfbf378833353ba4f716b", "parents": [ "4b78c9cbd8f1fbb9517aee48b372646f4cf05442" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Sun Mar 08 12:35:49 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:34:00 2026 -0700" }, "message": "tcp: inline tcp_chrono_start()\n\ntcp_chrono_start() is small enough, and used in TCP sendmsg()\nfast path (from tcp_skb_entail()).\n\nNote clang is already inlining it from functions in tcp_output.c.\n\nInlining it improves performance and reduces bloat :\n\n$ scripts/bloat-o-meter -t vmlinux.old vmlinux.new\nadd/remove: 0/2 grow/shrink: 1/0 up/down: 1/-84 (-83)\nFunction old new delta\ntcp_skb_entail 280 281 +1\n__pfx_tcp_chrono_start 16 - -16\ntcp_chrono_start 68 - -68\nTotal: Before\u003d25192434, After\u003d25192351, chg -0.00%\n\nNote that tcp_chrono_stop() is too big.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Neal Cardwell \u003cncardwell@google.com\u003e\nLink: https://patch.msgid.link/20260308123549.2924460-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "4b78c9cbd8f1fbb9517aee48b372646f4cf05442", "tree": "dc44d8ee0f9fbb0feecd250f14891fb0a07efe35", "parents": [ "f2db7b80b03f268ff65fe825a7c761a8f551aa48" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Sun Mar 08 12:23:02 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:33:43 2026 -0700" }, "message": "tcp: move tp-\u003echrono_type next tp-\u003echrono_stat[]\n\nchrono_type is currently in tcp_sock_read_txrx group, which\nis supposed to hold read-mostly fields.\n\nBut chrono_type is mostly written in tx path, it should\nbe moved to tcp_sock_write_tx group, close to other\nchrono fields (chrono_stat[], chrono_start).\n\nNote this adds holes, but data locality is far more important.\n\nUse a full u8 for the time being, compiler can generate\nmore efficient code.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Neal Cardwell \u003cncardwell@google.com\u003e\nLink: https://patch.msgid.link/20260308122302.2895067-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "f2db7b80b03f268ff65fe825a7c761a8f551aa48", "tree": "af5f6bef3f8fd3f59438f90b6a20b024d4615b84", "parents": [ "e8eb33d650cd5e60b008f9d958262e489de6e7a9" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Sat Mar 07 13:36:01 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:31:41 2026 -0700" }, "message": "net/sched: refine indirect call mitigation in tc_wrapper.h\n\nSome modern cpus disable X86_FEATURE_RETPOLINE feature,\neven if a direct call can still be beneficial.\n\nEven when IBRS is present, an indirect call is more expensive\nthan a direct one:\n\nDirect Calls:\n Compilers can perform powerful optimizations like inlining,\n where the function body is directly inserted at the call site,\n eliminating call overhead entirely.\n\nIndirect Calls:\n Inlining is much harder, if not impossible, because the compiler\n doesn\u0027t know the target function at compile time.\n Techniques like Indirect Call Promotion can help by using\n profile-guided optimization to turn frequently taken indirect calls\n into conditional direct calls, but they still add complexity\n and potential overhead compared to a truly direct call.\n\nIn this patch, I split tc_skip_wrapper in two different\nstatic keys, one for tc_act() (tc_skip_wrapper_act)\nand one for tc_classify() (tc_skip_wrapper_cls).\n\nThen I enable the tc_skip_wrapper_cls only if the count\nof builtin classifiers is above one.\n\nI enable tc_skip_wrapper_act only it the count of builtin\nactions is above one.\n\nIn our production kernels, we only have CONFIG_NET_CLS_BPF\u003dy\nand CONFIG_NET_ACT_BPF\u003dy. Other are modules or are not compiled.\n\nTested on AMD Turin cpus, cls_bpf_classify() cost went\nfrom 1% down to 0.18 %, and FDO will be able to inline\nit in tcf_classify() for further gains.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nAcked-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nReviewed-by: Pedro Tammela \u003cpctammela@mojatatu.com\u003e\nReviewed-by: Victor Nogueira \u003cvictor@mojatatu.com\u003e\nLink: https://patch.msgid.link/20260307133601.3863071-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e8eb33d650cd5e60b008f9d958262e489de6e7a9", "tree": "f5815478fde4ff637cc407ed6c5af4702e5c9e82", "parents": [ "bf3471e6e6c02137dc0d26caa783ac1849f9aab8" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Sat Mar 07 09:22:14 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:31:01 2026 -0700" }, "message": "tcp: move sysctl_tcp_shrink_window to netns_ipv4_read_txrx group\n\nCommit 18fd64d25422 (\"netns-ipv4: reorganize netns_ipv4 fast path\nvariables\") missed that __tcp_select_window() is reading\nnet-\u003eipv4.sysctl_tcp_shrink_window.\n\nMove this field to netns_ipv4_read_txrx group, as __tcp_select_window()\nis used both in tx and rx paths.\n\nSaves a potential cache line miss.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nLink: https://patch.msgid.link/20260307092214.2433548-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "bf3471e6e6c02137dc0d26caa783ac1849f9aab8", "tree": "5582d6a7b16829bcb0ecfa72e74b8d51288471d6", "parents": [ "014c607f86abc903d7bf46e13373d89392e371fe" ], "author": { "name": "Lorenzo Bianconi", "email": "lorenzo@kernel.org", "time": "Fri Mar 06 09:07:27 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:18:49 2026 -0700" }, "message": "net: airoha: Make flow control source port mapping dependent on nbq parameter\n\nFlow control source port mapping for USB serdes needs to be configured\naccording to the GDM port nbq parameter. This is a preliminary patch\nsince nbq parameter is specific for the given port serdes and needs to\nbe read from the DTS (in the current codebase is assigned statically).\n\nSigned-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nLink: https://patch.msgid.link/20260306-airoha-fix-loopback-for-usb-serdes-v2-1-319de9c96826@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "014c607f86abc903d7bf46e13373d89392e371fe", "tree": "fa21841abf628dcb92aed500f0cb231f04e07e99", "parents": [ "dc9c9193c7c19a0163552da3a73a1706b7faea12" ], "author": { "name": "Ankit Garg", "email": "nktgrg@google.com", "time": "Fri Mar 06 22:48:16 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:17:52 2026 -0700" }, "message": "gve: add support for UDP GSO for DQO format\n\nEnable support for UDP GSO when using DQO format. Advertise the feature\nflag during device initialization and enable offload by default.\n\nSigned-off-by: Ankit Garg \u003cnktgrg@google.com\u003e\nReviewed-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nSigned-off-by: Harshitha Ramamurthy \u003chramamurthy@google.com\u003e\nLink: https://patch.msgid.link/20260306224816.3391551-1-hramamurthy@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "dc9c9193c7c19a0163552da3a73a1706b7faea12", "tree": "f4a32d09145a17d3787c418dfd72031bd7cde7c3", "parents": [ "51aaf65bbd21e81a0a4d6827fad76ec8873a7703" ], "author": { "name": "Alok Tiwari", "email": "alok.a.tiwari@oracle.com", "time": "Fri Mar 06 10:08:19 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:17:48 2026 -0700" }, "message": "selftests: fib_tests: fix link-local retrieval in fib6_nexthop()\n\nfib6_nexthop() retrieves the link-local address for two interfaces used\nin the test. However, both lldummy and llv1 are obtained from dummy0.\n\nllv1 is expected to be retrieved from veth1, which is the interface used\nlater in the test. The subsequent check and error message also expect\nthe address to be retrieved from veth1.\n\nFix this by retrieving llv1 from veth1.\n\nSigned-off-by: Alok Tiwari \u003calok.a.tiwari@oracle.com\u003e\nLink: https://patch.msgid.link/20260306180830.2329477-1-alok.a.tiwari@oracle.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "51aaf65bbd21e81a0a4d6827fad76ec8873a7703", "tree": "3aa52ce3aafcce8bed6b46b873441ef49e9f83a6", "parents": [ "abb0eb0b033a0a8980eb9215e02626e4801ead3f", "b6420bd5aa0c374331bad6c0fa2eb5f0f87cf5a0" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:11:21 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:11:21 2026 -0700" }, "message": "Merge tag \u0027ib-gpio-remove-of-gpio-h-for-v7.1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux into mbox\n\nBartosz Golaszewski says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nImmutable branch between GPIO and net\n\nConvert remaining users of of_gpio.h to using GPIO descriptors and\nremove the header.\n\n* tag \u0027ib-gpio-remove-of-gpio-h-for-v7.1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:\n gpio: remove of_get_named_gpio() and \u003clinux/of_gpio.h\u003e\n nfc: nfcmrvl: convert to gpio descriptors\n nfc: s3fwrn5: convert to gpio descriptors\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260309093153.10446-1-bartosz.golaszewski@oss.qualcomm.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "abb0eb0b033a0a8980eb9215e02626e4801ead3f", "tree": "61202f608353a206112e6fa8acfcd32a6c58deeb", "parents": [ "58e4d35ae7b9325ab622bbcc34312b17af425c8f" ], "author": { "name": "Qingfang Deng", "email": "dqfext@gmail.com", "time": "Fri Mar 06 17:36:49 2026 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:07:38 2026 -0700" }, "message": "ppp: simplify input error handling\n\nCurrently, ppp_input_error() indicates an error by allocating a 0-length\nskb and calling ppp_do_recv(). It takes an error code argument, which is\nstored in skb-\u003ecb, but not used by ppp_receive_frame().\n\nSimplify the error handling by removing the unused parameter and the\nunnecessary skb allocation. Instead, call ppp_receive_error() directly\nfrom ppp_input_error() under the recv lock, and the length check in\nppp_receive_frame() can be removed.\n\nSigned-off-by: Qingfang Deng \u003cdqfext@gmail.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "58e4d35ae7b9325ab622bbcc34312b17af425c8f", "tree": "ced175b3956a786214d61a1e0d835c952af2f99c", "parents": [ "a23c657e332f2feb5eb9c4a3e8371386aa7392a6" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Fri Mar 06 13:31:54 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:01:53 2026 -0700" }, "message": "net/sched: use rtnl_kfree_skbs() in pfifo_fast_reset()\n\nrtnl_kfree_skbs() reduces RTNL and qdisc spinlock hold time.\n\nskbs are freed later after RTNL has been released.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nLink: https://patch.msgid.link/20260306133154.678730-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "a23c657e332f2feb5eb9c4a3e8371386aa7392a6", "tree": "9ecf75989deca86eb846abd2d50f74c5981df4f9", "parents": [ "47e8dbb6e763e5ccfed2ab4aa55cbb163382aec1" ], "author": { "name": "Sebastian Andrzej Siewior", "email": "bigeasy@linutronix.de", "time": "Fri Mar 06 15:44:39 2026 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 19:01:08 2026 -0700" }, "message": "net: ethernet: ti: am65-cpsw: Use also port number to identify timestamps\n\nThe driver uses packet-type (RX/TX) PTP-message type and PTP-sequence\nnumber to identify a matching timestamp packet for a skb. If the same\nPTP packet arrives on both ports (as in a PRP environment) then it is\nnot obvious which event belongs to which skb.\n\nThe event contains also the port number on which it was received.\nInstead of masking it out, use it for matching.\n\nTested-by: Chintan Vankar \u003cc-vankar@ti.com\u003e\nReviewed-by: Martin Kaistra \u003cmartin.kaistra@linutronix.de\u003e\nSigned-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nLink: https://patch.msgid.link/20260306144439.cVwaaopR@linutronix.de\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "47e8dbb6e763e5ccfed2ab4aa55cbb163382aec1", "tree": "d3f680a255a0dce919442e8f1d780c508e6049c5", "parents": [ "82f36517a13e6339ddd9737d8a310949e057e596" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Sat Mar 07 16:34:30 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:55:55 2026 -0700" }, "message": "net/sched: do not reset queues in graft operations\n\nFollowing typical script is extremely disruptive,\nbecause each graft operation calls dev_deactivate()\nwhich resets all the queues of the device.\n\nQPARAM\u003d\"limit 100000 flow_limit 1000 buckets 4096\"\nTXQS\u003d64\nfor ETH in eth1\ndo\n tc qd del dev $ETH root 2\u003e/dev/null\n tc qd add dev $ETH root handle 1: mq\n for i in `seq 1 $TXQS`\n do\n slot\u003d$( printf %x $(( i )) )\n tc qd add dev $ETH parent 1:$slot fq $QPARAM\n done\ndone\n\nOne can add \"ip link set dev $ETH down/up\" to reduce the disruption time:\n\nQPARAM\u003d\"limit 100000 flow_limit 1000 buckets 4096\"\nTXQS\u003d64\nfor ETH in eth1\ndo\n ip link set dev $ETH down\n tc qd del dev $ETH root 2\u003e/dev/null\n tc qd add dev $ETH root handle 1: mq\n for i in `seq 1 $TXQS`\n do\n slot\u003d$( printf %x $(( i )) )\n tc qd add dev $ETH parent 1:$slot fq $QPARAM\n done\n ip link set dev $ETH up\ndone\n\nOr we can add a @reset_needed flag to dev_deactivate() and\ndev_deactivate_many().\n\nThis flag is set to true at device dismantle or linkwatch_do_dev(),\nand to false for graft operations.\n\nIn the future, we might only stop one queue instead of the whole\ndevice, ie call dev_deactivate_queue() instead of dev_deactivate().\n\nI think the problem (quadratic behavior) was added in commit\n2fb541c862c9 (\"net: sch_generic: aviod concurrent reset and enqueue op\nfor lockless qdisc\") but this does not look serious enough to deserve\nrisky backports.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nReviewed-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nReviewed-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nReviewed-by: Victor Nogueira \u003cvictor@mojatatu.com\u003e\nLink: https://patch.msgid.link/20260307163430.470644-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "82f36517a13e6339ddd9737d8a310949e057e596", "tree": "57a9f6164b9482a6daf4114f57f77a1da3682376", "parents": [ "6927430735802bd8bc5dbe352edbd94a04459567" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Fri Mar 06 15:43:22 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:52:56 2026 -0700" }, "message": "tcp: avoid dst-\u003eops-\u003echeck() call in tcp_v{4,6}_do_rcv()\n\nIf incoming skb dst matches the socket cached one,\nthere is no need to call again dst-\u003eops-\u003echeck().\n\nNetwork layer already validated the skb dst for us,\nusually from tcp_v{4,6}_early_demux().\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nLink: https://patch.msgid.link/20260306154322.1086539-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "6927430735802bd8bc5dbe352edbd94a04459567", "tree": "7a20652dd34882d7c90a8d08d0e8133dbe48cf35", "parents": [ "50636e5ff8861c35ebb190521ba540074ab583ad" ], "author": { "name": "Rosen Penev", "email": "rosenp@gmail.com", "time": "Thu Mar 05 18:54:49 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:51:07 2026 -0700" }, "message": "net: rocker: kzalloc + kcalloc to kzalloc_flex\n\nCombining the allocations simplifies things, especially the free path.\n\nRemove ofdpa_group_tbl_entry_free as a result. kfree is shorter.\n\nAdd __counted_by for extra runtime analysis.\n\nSigned-off-by: Rosen Penev \u003crosenp@gmail.com\u003e\nAcked-by: Gustavo A. R. Silva \u003cgustavoars@kernel.org\u003e\nReviewed-by: Jiri Pirko \u003cjiri@nvidia.com\u003e\nLink: https://patch.msgid.link/20260306025449.12333-1-rosenp@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "50636e5ff8861c35ebb190521ba540074ab583ad", "tree": "0303f6a2d2cd1fd784cc91228d17a01f8db1d57c", "parents": [ "8ca8eb05767395cd2fd0db12e052ca0bc5d1597c" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Fri Mar 06 13:11:30 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:50:24 2026 -0700" }, "message": "tcp: move tcp_v4_early_demux() to net/ipv4/ip_input.c\n\ntcp_v4_early_demux() has a single caller : ip_rcv_finish_core().\n\nMove it to net/ipv4/ip_input.c and mark it static, for possible\ncompiler/linker optimizations.\n\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: David Ahern \u003cdsahern@kernel.org\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nLink: https://patch.msgid.link/20260306131130.654991-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "8ca8eb05767395cd2fd0db12e052ca0bc5d1597c", "tree": "cdb35e6c247bb9dde45878af41bfa4014749c59d", "parents": [ "2ed4b46b4fc77749cb0f8dd31a01441b82c8dbaa" ], "author": { "name": "Keita Morisaki", "email": "kmta1236@gmail.com", "time": "Sat Mar 07 14:16:19 2026 +0900" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:49:38 2026 -0700" }, "message": "tcp: remove unused hash_size from struct tcp_out_options\n\nhash_size is declared but never read. The MD5 path always uses a\nfixed size of 16, and the TCP-AO path uses tcp_ao_maclen().\n\nThis closes a 7-byte hole and reduces the struct size from 96 to\n88 bytes.\n\nSuggested-by: Jakub Sitnicki \u003cjakub@cloudflare.com\u003e\nSigned-off-by: Keita Morisaki \u003ckmta1236@gmail.com\u003e\nReviewed-by: Kuniyuki Iwashima \u003ckuniyu@google.com\u003e\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\nLink: https://patch.msgid.link/20260307051619.51685-1-kmta1236@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "2ed4b46b4fc77749cb0f8dd31a01441b82c8dbaa", "tree": "4fb9d44406bcb0ce16693a270bcad6656c1910a2", "parents": [ "86020d7db03abd337d89b02586ce0fb0d65667e0" ], "author": { "name": "Tim Bird", "email": "tim.bird@sony.com", "time": "Wed Mar 04 17:47:22 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 18:32:45 2026 -0700" }, "message": "net: Add SPDX ids to some source files\n\nAdd SPDX-License-Identifier lines to several source\nfiles under the network sub-directory. Work on files\nin the core, dns_resolver, ipv4, ipv6 and\nnetfilter sub-dirs. Remove boilerplate\nand license reference text to avoid ambiguity.\n\nRusty Russell has expressed that his contributions\nwere intended to be GPL-2.0-or-later.\n\nSigned-off-by: Tim Bird \u003ctim.bird@sony.com\u003e\nLink: https://patch.msgid.link/20260305004724.87469-1-tim.bird@sony.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "86020d7db03abd337d89b02586ce0fb0d65667e0", "tree": "8f2584bc705f58bee10d6dbdeefd28711090fbdd", "parents": [ "0bcac7b11262557c990da1ac564d45777eb6b005", "aa234faa5a4d0f213697c4b9012c75b80b9d2c20" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:30 2026 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:30 2026 -0700" }, "message": "Merge branch \u0027tools-ynl-convert-samples-into-selftests\u0027\n\nJakub Kicinski says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ntools: ynl: convert samples into selftests\n\nThe \"samples\" were always poor man\u0027s tests, used to manually\nconfirm that C YNL works as expected. Since a proper tests/\ndirectory now exists move the samples and use the kselftest\nharness to turn them into selftests outputting KTAP.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260307033630.1396085-1-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "aa234faa5a4d0f213697c4b9012c75b80b9d2c20", "tree": "8f2584bc705f58bee10d6dbdeefd28711090fbdd", "parents": [ "e7a39b8f5fcf076b88925d0e24bdba69d988cf9e" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:30 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:27 2026 -0700" }, "message": "tools: ynl: convert rt-route sample to selftest\n\nConvert rt-route.c to use kselftest_harness.h with FIXTURE/TEST_F.\nThis is the last test to convert so clean up the Makefile.\n\nValidate that the connected routes for 192.168.1.0/24 and\n2001:db8::/64 appear in the dump.\n\nOutput:\n\n TAP version 13\n 1..1\n # Starting 1 tests from 1 test cases.\n # RUN rt_route.dump ...\n # oif: nsim0 dst: 192.168.1.0/24\n # oif: lo dst: ::1/128\n # oif: nsim0 dst: 2001:db8::1/128\n # oif: nsim0 dst: 2001:db8::/64\n # oif: nsim0 dst: fe80::/64\n # oif: nsim0 dst: ff00::/8\n # OK rt_route.dump\n ok 1 rt_route.dump\n # PASSED: 1 / 1 tests passed.\n # Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-11-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e7a39b8f5fcf076b88925d0e24bdba69d988cf9e", "tree": "09a9f556db5d3183423d411b8d4dd9e8784d9d4f", "parents": [ "1419fbf5a8179f6d5cc1e2487c01d51c97bcade0" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:29 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:27 2026 -0700" }, "message": "tools: ynl: convert rt-addr sample to selftest\n\nConvert rt-addr.c to use kselftest_harness.h with FIXTURE/TEST_F.\n\nValidate that the addresses configured by the wrapper (192.168.1.1\nand 2001:db8::1) appear in the dump.\n\nOutput:\n\n TAP version 13\n 1..1\n # Starting 1 tests from 1 test cases.\n # RUN rt_addr.dump ...\n # lo: 127.0.0.1\n # nsim0: 192.168.1.1\n # lo: ::1\n # nsim0: 2001:db8::1\n # nsim0: fe80::7c66:c9ff:fe5f:bf01\n # OK rt_addr.dump\n ok 1 rt_addr.dump\n # PASSED: 1 / 1 tests passed.\n # Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-10-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "1419fbf5a8179f6d5cc1e2487c01d51c97bcade0", "tree": "6e9fc4f37d93e6e6fb687b09f3e474b7ea2980ba", "parents": [ "db20b374e7f74d2336c8d9f8a6e4806985328980" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:28 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert ethtool sample to selftest\n\nConvert ethtool.c to use kselftest_harness.h with FIXTURE/TEST_F.\nMove ethtool from BINS to TEST_GEN_FILES and add ethtool.sh wrapper\nwhich sets up a netdevsim device before running the test binary.\n\nOutput:\n\n TAP version 13\n 1..2\n # Starting 2 tests from 1 test cases.\n # RUN ethtool.channels ...\n # nsim0: combined 1\n # OK ethtool.channels\n ok 1 ethtool.channels\n # RUN ethtool.rings ...\n # nsim0: rx 512 tx 512\n # OK ethtool.rings\n ok 2 ethtool.rings\n # PASSED: 2 / 2 tests passed.\n # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-9-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "db20b374e7f74d2336c8d9f8a6e4806985328980", "tree": "afda96c0b91d260d11ce6d5163552027f127ad0f", "parents": [ "7a95e52562936e54c77f3fb2177016d68f1f203f" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:27 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert devlink sample to selftest\n\nConvert devlink.c to use kselftest_harness.h with FIXTURE/TEST_F.\nMove devlink from BINS to TEST_GEN_FILES in the Makefile since\nit\u0027s invoked via the devlink.sh wrapper which sets up netdevsim.\n\nOutput:\n\n TAP version 13\n 1..2\n # Starting 2 tests from 1 test cases.\n # RUN devlink.dump ...\n # netdevsim/netdevsim1337\n # OK devlink.dump\n ok 1 devlink.dump\n # RUN devlink.info ...\n # netdevsim/netdevsim1337:\n # driver: netdevsim\n # running fw:\n # fw.mgmt: 10.20.30\n # OK devlink.info\n ok 2 devlink.info\n # PASSED: 2 / 2 tests passed.\n # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-8-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7a95e52562936e54c77f3fb2177016d68f1f203f", "tree": "989b25b3b6f7bcbaa745532f011fe0aa639aee8b", "parents": [ "6cf8fb4722c35270df8e087c0c589da0577377a6" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:26 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: add netdevsim wrapper library for YNL tests\n\nSome tests need netdevsim setup which is painful to do from C.\n\nAdd ynl_nsim_lib.sh, a shared library providing nsim_setup and\nnsim_cleanup functions for tests that need a netdevsim device.\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-7-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "6cf8fb4722c35270df8e087c0c589da0577377a6", "tree": "75e765f5027e5c16b47278223ca9bdcfe3b2782e", "parents": [ "5c3206786c2d2c42540ad075d73fc895a9eb7f39" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:25 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert tc and tc-filter-add samples to selftest\n\nConvert tc.c and tc-filter-add.c to produce KTAP output with\nkselftest_harness. Merge the two tests together. They both\ntest TC one is testing qdisc and the other classifiers but\nthey can easily live in a single selftest.\n\nMake the test spawn a new netns, and run the operations on\nlo to avoid onerous setup and cleanup.\n\n TAP version 13\n 1..2\n # Starting 2 tests from 1 test cases.\n # RUN tc.qdisc ...\n # lo: fq_codel limit: 10240p target: 5ms new_flow_cnt: 0\n # OK tc.qdisc\n ok 1 tc.qdisc\n # RUN tc.flower ...\n # flower pref 1 proto: 0x8100\n # flower:\n # vlan_id: 100\n # vlan_prio: 5\n # num_of_vlans: 3\n # action order: 1 vlan push id 200 protocol 0x8100 priority 0\n # action order: 2 vlan push id 300 protocol 0x8100 priority 0\n # OK tc.flower\n ok 2 tc.flower\n # PASSED: 2 / 2 tests passed.\n # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-6-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5c3206786c2d2c42540ad075d73fc895a9eb7f39", "tree": "1a97a43e513c5c4348eaccc1a137ad118bde6115", "parents": [ "7e3effbc76278bfe3c452074c301233c8e69101f" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:24 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert rt-link sample to selftest\n\nConvert rt-link.c to use kselftest_harness.h with FIXTURE/TEST_F.\nMove rt-link from BINS to TEST_GEN_PROGS.\n\nOutput:\n\n TAP version 13\n 1..3\n # Starting 3 tests from 1 test cases.\n # RUN rt_link.dump ...\n # 1: lo: mtu 65536\n # 2: sit0: mtu 1480 kind sit\n # OK rt_link.dump\n ok 1 rt_link.dump\n # RUN rt_link.netkit ...\n # 4: nk1: mtu 1500 kind netkit primary 1 policy blackhole\n # OK rt_link.netkit\n ok 2 rt_link.netkit\n # RUN rt_link.netkit_err_msg ...\n # OK rt_link.netkit_err_msg\n ok 3 rt_link.netkit_err_msg\n # PASSED: 3 / 3 tests passed.\n # Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-5-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7e3effbc76278bfe3c452074c301233c8e69101f", "tree": "d08b518113b36772259a4e03e36c8a2d1e39a160", "parents": [ "285804d63f35b333b216f90255104264c590514b" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:23 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert ovs sample to selftest\n\nConvert ovs.c to produce KTAP output with kselftest_harness.\nThe single \"crud\" test creates a new OVS datapath, fetches it back\nby name, then dumps all datapaths verifying the new one appears.\n\nIIRC I added this test because ovs is a genetlink family but\nhas a family-specific fixed header.\n\n TAP version 13\n 1..1\n # Starting 1 tests from 1 test cases.\n # RUN ovs.crud ...\n # get:\n # ynl-test(3): pid:0 cache:256\n # dump:\n # ynl-test(3): pid:0 cache:256\n # OK ovs.crud\n ok 1 ovs.crud\n # PASSED: 1 / 1 tests passed.\n # Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-4-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "285804d63f35b333b216f90255104264c590514b", "tree": "01782226068d60f3188fed01350acef2efae2c97", "parents": [ "e0aa0c61758f517ded2befa084ebcad93809b421" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:22 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:26 2026 -0700" }, "message": "tools: ynl: convert netdev sample to selftest\n\nConvert netdev.c to produce KTAP output with 3 tests:\n- dev_dump: dump all netdev devices, skip if empty\n- dev_get: query first device from dump by ifindex\n- ntf_check: subscribe to \"mgmt\", create a veth via rt-link,\n verify netdev notification is received, then delete the veth\n\nRemove stdin/scanf-based UI. Add rt-link dependency for the veth\nnotification test.\n\n TAP version 13\n 1..3\n # Starting 3 tests from 1 test cases.\n # RUN netdev.dump ...\n # lo[1]\txdp-features (0): xdp-rx-metadata-features (0): xsk-fea...\n # sit0[2]\txdp-features (0): xdp-rx-metadata-features (0): xsk-fea...\n # OK netdev.dump\n ok 1 netdev.dump\n # RUN netdev.get ...\n # lo[1]\txdp-features (0): xdp-rx-metadata-features (0): xsk-fea...\n # OK netdev.get\n ok 2 netdev.get\n # RUN netdev.ntf_check ...\n # veth0[7]\txdp-features (0): xdp-rx-metadata-features (7): timesta...\n # OK netdev.ntf_check\n ok 3 netdev.ntf_check\n # PASSED: 3 / 3 tests passed.\n # Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-3-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e0aa0c61758f517ded2befa084ebcad93809b421", "tree": "0e33c37c594526875ca917bb23835c9ef8719879", "parents": [ "0bcac7b11262557c990da1ac564d45777eb6b005" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:36:21 2026 -0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 09 17:02:25 2026 -0700" }, "message": "tools: ynl: move samples to tests\n\nThe \"samples\" were always poor man\u0027s tests (used to manually\nconfirm that C YNL works).\n\nMove all C sample programs from tools/net/ynl/samples/ to\ntools/net/ynl/tests/, \"merge\" the Makefiles. The subsequent\nchanges will convert each sample into a proper KTAP selftests.\n\nSince these are now tests rather than samples - default to\nenabling asan. After all we\u0027re testing user space code here.\n\nSort the gitignore while at it, the page-pool entry was a leftover\nso delete it.\n\nReviewed-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nTested-by: Donald Hunter \u003cdonald.hunter@gmail.com\u003e\nLink: https://patch.msgid.link/20260307033630.1396085-2-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "b6420bd5aa0c374331bad6c0fa2eb5f0f87cf5a0", "tree": "fe1185cfa2b715ab7fbc6b5d038920e05d3ba66b", "parents": [ "253350dbf3e7fbd136905c98bd9f800fddb4fead" ], "author": { "name": "Jialu Xu", "email": "xujialu@vimux.org", "time": "Sat Mar 07 11:06:26 2026 +0800" }, "committer": { "name": "Bartosz Golaszewski", "email": "bartosz.golaszewski@oss.qualcomm.com", "time": "Mon Mar 09 10:25:34 2026 +0100" }, "message": "gpio: remove of_get_named_gpio() and \u003clinux/of_gpio.h\u003e\n\nAll in-tree consumers have been converted to the descriptor-based API.\nRemove the deprecated of_get_named_gpio() helper, delete the\n\u003clinux/of_gpio.h\u003e header, and drop the corresponding entry from\nMAINTAINERS.\n\nAlso remove the completed TODO item for this cleanup.\n\nSigned-off-by: Jialu Xu \u003cxujialu@vimux.org\u003e\nReviewed-by: Linus Walleij \u003clinusw@kernel.org\u003e\nLink: https://patch.msgid.link/02ABDA1F9E3FAF1F+20260307030623.3495092-6-xujialu@vimux.org\nSigned-off-by: Bartosz Golaszewski \u003cbartosz.golaszewski@oss.qualcomm.com\u003e\n" }, { "commit": "253350dbf3e7fbd136905c98bd9f800fddb4fead", "tree": "fa77e70a5596e6c7272ee52e959c009a50048c8b", "parents": [ "b544927d75574330b0a8a33c113556b67df56f39" ], "author": { "name": "Jialu Xu", "email": "xujialu@vimux.org", "time": "Sat Mar 07 11:06:24 2026 +0800" }, "committer": { "name": "Bartosz Golaszewski", "email": "bartosz.golaszewski@oss.qualcomm.com", "time": "Mon Mar 09 10:25:34 2026 +0100" }, "message": "nfc: nfcmrvl: convert to gpio descriptors\n\nReplace the legacy of_get_named_gpio() / gpio_request_one() /\ngpio_set_value() API with the descriptor-based devm_gpiod_get_optional() /\ngpiod_set_value() API from \u003clinux/gpio/consumer.h\u003e, removing the\ndependency on \u003clinux/of_gpio.h\u003e.\n\nThe \"reset-n-io\" property rename quirk already exists in gpiolib-of.c\n(added in commit 9c2cc7171e08), so no additional quirk is needed.\n\nSigned-off-by: Jialu Xu \u003cxujialu@vimux.org\u003e\nReviewed-by: Linus Walleij \u003clinusw@kernel.org\u003e\nLink: https://patch.msgid.link/DD684946FD7EE161+20260307030623.3495092-4-xujialu@vimux.org\nSigned-off-by: Bartosz Golaszewski \u003cbartosz.golaszewski@oss.qualcomm.com\u003e\n" }, { "commit": "b544927d75574330b0a8a33c113556b67df56f39", "tree": "a448df4796623df4490d310d192fe2c0c3d7936e", "parents": [ "6de23f81a5e08be8fbf5e8d7e9febc72a5b5f27f" ], "author": { "name": "Jialu Xu", "email": "xujialu@vimux.org", "time": "Sat Mar 07 11:06:22 2026 +0800" }, "committer": { "name": "Bartosz Golaszewski", "email": "bartosz.golaszewski@oss.qualcomm.com", "time": "Mon Mar 09 10:25:34 2026 +0100" }, "message": "nfc: s3fwrn5: convert to gpio descriptors\n\nReplace the legacy of_get_named_gpio() / gpio_request_one() /\ngpio_set_value() API with the descriptor-based devm_gpiod_get() /\ngpiod_set_value() API from \u003clinux/gpio/consumer.h\u003e, removing the\ndependency on \u003clinux/of_gpio.h\u003e.\n\nThis removes the s3fwrn5_i2c_parse_dt() and s3fwrn82_uart_parse_dt()\nfunctions since devm_gpiod_get() handles both DT lookup and resource\nmanagement. The gpio_en and gpio_fw_wake fields in struct phy_common\nare changed from int to struct gpio_desc *.\n\nAdd rename quirks in gpiolib-of.c for the deprecated \"s3fwrn5,en-gpios\"\nand \"s3fwrn5,fw-gpios\" properties to maintain backward compatibility\nwith old device trees.\n\nSigned-off-by: Jialu Xu \u003cxujialu@vimux.org\u003e\nReviewed-by: Linus Walleij \u003clinusw@kernel.org\u003e\nLink: https://patch.msgid.link/94FF47746A92BD6B+20260307030623.3495092-2-xujialu@vimux.org\nSigned-off-by: Bartosz Golaszewski \u003cbartosz.golaszewski@oss.qualcomm.com\u003e\n" }, { "commit": "0bcac7b11262557c990da1ac564d45777eb6b005", "tree": "ee0217cc88ec07ebb75e60f045008c4a27c5f96d", "parents": [ "f4ac0cc88e9949d41bbdd101caa3117afe983ec9" ], "author": { "name": "Aleksei Oladko", "email": "aleksey.oladko@virtuozzo.com", "time": "Fri Mar 06 00:01:23 2026 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 06 19:26:23 2026 -0800" }, "message": "selftests: net: make ovs-dpctl.py fail when pyroute2 is unsupported\n\nThe pmtu.sh kselftest configures OVS using ovs-dpctl.py and falls back\nto ovs-vsctl only when ovs-dpctl.py fails. However, ovs-dpctl.py exits\nwith a success status when the installed pyroute2 package version is\nlower than 0.6, even though the OVS datapath is not configured.\n\nAs a result, pmtu.sh assumes that the setup was successful and\ncontinues running the test, which later fails due to the missing\nOVS configuration.\n\nFix the exit code handling in ovs-dpctl.py so that pmtu.sh can detect\nthat the setup did not complete successfully and fall back to\novs-vsctl.\n\nSigned-off-by: Aleksei Oladko \u003caleksey.oladko@virtuozzo.com\u003e\nReviewed-by: Aaron Conole \u003caconole@redhat.com\u003e\nLink: https://patch.msgid.link/20260306000127.519064-3-aleksey.oladko@virtuozzo.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" } ], "next": "f4ac0cc88e9949d41bbdd101caa3117afe983ec9" }