- 22 Mar, 2022 10 commits
-
-
Jakub Kicinski authored
Building htmldocs complains: include/linux/netdevice.h:2295: warning: Function parameter or member 'dm_private' not described in 'net_device' Fixes: b26ef81c ("drop_monitor: remove quadratic behavior") Signed-off-by:
Jakub Kicinski <kuba@kernel.org> Reviewed-by:
Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20220322051053.1883186-1-kuba@kernel.orgSigned-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Eric Dumazet authored
Call br_mst_info_size() only if vg pointer is not NULL. general protection fault, probably for non-canonical address 0xdffffc0000000058: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x00000000000002c0-0x00000000000002c7] CPU: 0 PID: 975 Comm: syz-executor.0 Tainted: G W 5.17.0-next-20220321-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:br_mst_info_size+0x97/0x270 net/bridge/br_mst.c:242 Code: 00 00 31 c0 e8 ba 10 53 f9 31 c0 b9 40 00 00 00 4c 8d 6c 24 30 4c 89 ef f3 48 ab 48 8d 83 c0 02 00 00 48 89 04 24 48 c1 e8 03 <80> 3c 28 00 0f 85 ae 01 00 00 48 8b 83 c0 02 00 00 41 bf 04 00 00 RSP: 0018:ffffc900153770a8 EFLAGS: 00010202 RAX: 0000000000000058 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000040000 RSI: ffffffff88259876 RDI: ffffc900153772d8 RBP: dffffc0000000000 R08: 0000000000000000 R09: ffffffff8db68957 R10: ffffffff881f737b R11: 0000000000000000 R12: 0000000000000000 R13: ffffc900153770d8 R14: 00000000000002a0 R15: 00000000ffffffff FS: 00007f18bbb6f700(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020001a80 CR3: 000000001a7d9000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 00000000000000d8 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> br_get_link_af_size_filtered+0x6e9/0xc00 net/bridge/br_netlink.c:123 rtnl_link_get_af_size net/core/rtnetlink.c:598 [inline] if_nlmsg_size+0x40c/0xa50 net/core/rtnetlink.c:1040 rtnl_calcit.isra.0+0x25f/0x460 net/core/rtnetlink.c:3780 rtnetlink_rcv_msg+0xa65/0xb80 net/core/rtnetlink.c:5937 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2496 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline] netlink_unicast+0x543/0x7f0 net/netlink/af_netlink.c:1345 netlink_sendmsg+0x904/0xe00 net/netlink/af_netlink.c:1921 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:725 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2413 ___sys_sendmsg+0xf3/0x170 net/socket.c:2467 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2496 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0x80 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f18baa89049 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f18bbb6f168 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00007f18bab9bf60 RCX: 00007f18baa89049 RDX: 0000000000000000 RSI: 0000000020001a80 RDI: 0000000000000004 RBP: 00007f18baae308d R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00007ffeedb2be2f R14: 00007f18bbb6f300 R15: 0000000000022000 </TASK> Modules linked in: ---[ end trace 0000000000000000 ]--- RIP: 0010:br_mst_info_size+0x97/0x270 net/bridge/br_mst.c:242 Code: 00 00 31 c0 e8 ba 10 53 f9 31 c0 b9 40 00 00 00 4c 8d 6c 24 30 4c 89 ef f3 48 ab 48 8d 83 c0 02 00 00 48 89 04 24 48 c1 e8 03 <80> 3c 28 00 0f 85 ae 01 00 00 48 8b 83 c0 02 00 00 41 bf 04 00 00 RSP: 0018:ffffc900153770a8 EFLAGS: 00010202 RAX: 0000000000000058 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000040000 RSI: ffffffff88259876 RDI: ffffc900153772d8 RBP: dffffc0000000000 R08: 0000000000000000 R09: ffffffff8db68957 R10: ffffffff881f737b R11: 0000000000000000 R12: 0000000000000000 R13: ffffc900153770d8 R14: 00000000000002a0 R15: 00000000ffffffff FS: 00007f18bbb6f700(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2ca22000 CR3: 000000001a7d9000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 00000000000000d8 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Fixes: 122c2948 ("net: bridge: mst: Support setting and reporting MST port states") Reported-by:
syzbot <syzkaller@googlegroups.com> Signed-off-by:
Tobias Waldekranz <tobias@waldekranz.com> Acked-by:
Nikolay Aleksandrov <razor@blackwall.org> Link: https://lore.kernel.org/r/20220322012314.795187-1-eric.dumazet@gmail.comSigned-off-by:
-
Paolo Abeni authored
Ido Schimmel says: ==================== selftests: forwarding: Locked bridge port fixes Two fixes for the locked bridge port selftest. ==================== Link: https://lore.kernel.org/r/20220321175102.978020-1-idosch@nvidia.comSigned-off-by:
-
Ido Schimmel authored
The test creates a separate VRF for the VLAN upper, but does not destroy it during cleanup, resulting in "RTNETLINK answers: File exists" errors. Fix by using the same VRF for the port and its VLAN upper. This is OK since their IP addresses do not overlap. Before: # ./bridge_locked_port.sh TEST: Locked port ipv4 [ OK ] TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] # ./bridge_locked_port.sh RTNETLINK answers: File exists RTNETLINK answers: File exists RTNETLINK answers: File exists RTNETLINK answers: File exists RTNETLINK answers: File exists RTNETLINK answers: File exists TEST: Locked port ipv4 [ OK ] TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] After: # ./bridge_locked_port.sh TEST: Locked port ipv4 [ OK ] TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] # ./bridge_locked_port.sh TEST: Locked port ipv4 [ OK ] TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] Fixes: b2b681a4 ("selftests: forwarding: tests of locked port feature") Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Signed-off-by:
-
Ido Schimmel authored
Disable learning before bringing the bridge port up in order to avoid the FDB being populated and the test failing. Before: # ./bridge_locked_port.sh RTNETLINK answers: File exists TEST: Locked port ipv4 [FAIL] Ping worked after locking port, but before adding FDB entry TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] After: # ./bridge_locked_port.sh TEST: Locked port ipv4 [ OK ] TEST: Locked port ipv6 [ OK ] TEST: Locked port vlan [ OK ] Fixes: b2b681a4 ("selftests: forwarding: tests of locked port feature") Signed-off-by: -
Bill Wendling authored
The original behavior was to print out unsigned short or unsigned char values. The change in commit d65aea8e ("bnx2x: use correct format characters") prints out the whole value if not truncated. So truncate the value to an unsigned {short|char} to retain the original behavior. Fixes: d65aea8e ("bnx2x: use correct format characters") Link: https://github.com/ClangBuiltLinux/linux/issues/378Signed-off-by:
Bill Wendling <morbo@google.com> Link: https://lore.kernel.org/r/20220321023155.106066-1-morbo@google.comSigned-off-by:
-
Jakub Kicinski authored
Michael Walle says: ==================== net: mscc-miim: add integrated PHY reset support The MDIO driver has support to release the integrated PHYs from reset. This was implemented for the SparX-5 for now. Now add support for the LAN966x, too. ==================== Link: https://lore.kernel.org/r/20220318201324.1647416-1-michael@walle.ccSigned-off-by:
-
Michael Walle authored
The LAN966x has two internal PHYs which are in reset by default. The driver already supported the internal PHYs of the SparX-5. Now add support for the LAN966x, too. Add a new compatible to distinguish them. The LAN966x has additional control bits in this register, thus convert the regmap_write() to regmap_update_bits() to leave the remaining bits untouched. This doesn't change anything for the SparX-5 SoC, because there, the register consists only of reset bits. Signed-off-by:
Michael Walle <michael@walle.cc> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
Michael Walle authored
Replace the magic numbers by macros which are already defined. It seems the original commit missed to use them. Signed-off-by:
-
Michael Walle authored
The MDIO controller has support to release the internal PHYs from reset by specifying a second memory resource. This is different between the currently supported SparX-5 and the LAN966x. Add a new compatible to distinguish between these two. Signed-off-by:
Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by:
-
- 21 Mar, 2022 30 commits
-
-
Tobias Waldekranz authored
Some chips using the split VTU/STU design will not accept VTU entries who's SID points to an invalid STU entry. Therefore, mark all those chips with either the mv88e6352_g1_stu_* or mv88e6390_g1_stu_* ops as appropriate. Notably, chips for the Opal Plus (6085/6097) era seem to use a different implementation than those from Agate (6352) and onwards, even though their external interface is the same. The former happily accepts VTU entries referencing invalid STU entries, while the latter does not. This fixes an issue where the driver would fail to probe switch trees that contained chips of the Agate/Topaz generation which did not declare STU support, as loaded VTU entries would be read back as invalid. Fixes: 49c98c1d ("net: dsa: mv88e6xxx: Disentangle STU from VTU") Reported-by:
Marek Behún <kabel@kernel.org> Signed-off-by:
Vladimir Oltean <olteanv@gmail.com> Tested-by:
-
Guo Zhengkui authored
`cur64`, `start64` and `ts_delta` are int64_t. Change format specifiers in fprintf from `"%lu"` to `"%" PRId64` to adapt to 32-bit and 64-bit systems. Signed-off-by:
Guo Zhengkui <guozhengkui@vivo.com> Link: https://lore.kernel.org/r/20220319073730.5235-1-guozhengkui@vivo.comSigned-off-by:
-
Vladimir Oltean authored
The Felix switch has 6 ports, 2 of which are internal. Due to some misunderstanding, my initial suggestion for vsc9959_port_modes[]: https://patchwork.kernel.org/project/netdevbpf/patch/20220129220221.2823127-10-colin.foster@in-advantage.com/#24718277 got translated by Colin into a 5-port array, leading to an all-zero port mode mask for port 5. Fixes: acf242fc ("net: dsa: felix: remove prevalidate_phy_mode interface") Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Link: https://lore.kernel.org/r/20220318195812.276276-1-vladimir.oltean@nxp.comSigned-off-by:
-
Jakub Kicinski authored
Tobias Waldekranz says: ==================== net: dsa: mv88e6xxx: MST Fixes 1/2 fixes the issue reported by Marek here: https://lore.kernel.org/netdev/20220318182817.5ade8ecd@dellmb/ 2/2 adds a missing capability check to the new .vlan_msti_set callback. ==================== Link: https://lore.kernel.org/r/20220318201321.4010543-1-tobias@waldekranz.comSigned-off-by:
-
Tobias Waldekranz authored
In the same way that we check for STU support in the MST state callback, we should also verify it before trying to change a VLANs MSTI membership. Fixes: acaf4d2e ("net: dsa: mv88e6xxx: MST Offloading") Signed-off-by:
-
Tobias Waldekranz authored
Simply having a physical STU table in the device doesn't do us any good if there's no implementation of the relevant ops to access that table. So ensure that chips that claim STU support can also talk to the hardware. This fixes an issue where chips that had a their ->info->max_sid set (due to their family membership), but no implementation (due to their chip-specific ops struct) would fail to probe. Fixes: 49c98c1d ("net: dsa: mv88e6xxx: Disentangle STU from VTU") Reported-by:
-
Jakub Kicinski authored
Ziyang Xuan says: ==================== net/tls: some optimizations for tls Do some small optimizations for tls, including jump instructions optimization, and judgement processes optimization. ==================== Link: https://lore.kernel.org/r/cover.1647658604.git.william.xuanziyang@huawei.comSigned-off-by:
-
Ziyang Xuan authored
It is known that priority setting HW offload when set tls TX/RX offload by setsockopt(). Check netdevice whether support NETIF_F_HW_TLS_TX or not at the later stages in the whole tls_set_device_offload() process, some memory allocations have been done before that. We must release those memory and return error if we judge the netdevice not support NETIF_F_HW_TLS_TX. It is redundant. Move NETIF_F_HW_TLS_TX judgement forward, and move start_marker_record and offload_ctx memory allocation back slightly. Thus, we can get simpler exception handling process. Signed-off-by:
Ziyang Xuan <william.xuanziyang@huawei.com> Signed-off-by:
-
Ziyang Xuan authored
Avoid using "goto" jump instruction unconditionally when we can return directly. Remove unnecessary jump instructions in do_tls_setsockopt_conf(). Signed-off-by:
-
Sebastian Andrzej Siewior authored
The lockdep annotation lockdep_assert_softirq_will_run() expects that either hard or soft interrupts are disabled because both guaranty that the "raised" soft-interrupts will be processed once the context is left. This triggers in flush_smp_call_function_from_idle() but it this case it explicitly calls do_softirq() in case of pending softirqs. Revert the "softirq will run" annotation in ____napi_schedule() and move the check back to __netif_rx() as it was. Keep the IRQ-off assert in ____napi_schedule() because this is always required. Fixes: fbd9a2ce ("net: Add lockdep asserts to ____napi_schedule().") Signed-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Reviewed-by:
Jason A. Donenfeld <Jason@zx2c4.com> Link: https://lore.kernel.org/r/YjhD3ZKWysyw8rc6@linutronix.deSigned-off-by:
-
David S. Miller authored
Jakub Kicinski says: ==================== devlink: hold the instance lock in eswitch callbacks Series number 2 in the effort to hold the devlink instance lock in call driver callbacks. We have the following drivers using this API: - bnxt, nfp, netdevsim - their own locking is removed / simplified by this series; all of them needed a lock to protect from changes to the number of VFs while switching modes, now the VF config bus callback takes the devlink instance lock via devl_lock(); - ice - appears not to allow changing modes while SR-IOV enabled, so nothing to do there; - liquidio - does not contain any locking; - octeontx2/af - is very special but at least doesn't have locking so doesn't get in the way either; - mlx5 has a wealth of locks - I chickened out and dropped the lock in the callbacks so that I can leave the driver be, for now. The last one is obviously not ideal, but I would prefer to transition the API already as it make take longer. v2: use a wrapper in mlx5 and extend the comment ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
Make the devlink core hold the instance lock during eswitch_mode callbacks. Cheat in case of mlx5 (see the cover letter). Reviewed-by:
Leon Romanovsky <leonro@nvidia.com> Reviewed-by:
Jiri Pirko <jiri@nvidia.com> Signed-off-by:
-
Jakub Kicinski authored
Similarly to the previous commit, use the devlink instance lock and let it replace the vfs_lock. nsim_esw_legacy_enable() was locked by both port lock and vfs lock so one set of lock/unlocks goes away. netdevsim's .eswitch_mode_set callback is now ready for the callback to take the instance lock. Signed-off-by:
-
Jakub Kicinski authored
Take advantage of the devlink instance lock for protecting the port list. This will simplify locking even more once all devlink callbacks hold the instance lock. We need to add locking in nsim_dev_port_add_all() which used to assume higher layer protection when accessing the list. Signed-off-by:
-
Jakub Kicinski authored
We'll need an explicitly locked rate node API for netdevsim to switch eswitch mode setting to locked. Reviewed-by:
-
Jakub Kicinski authored
In prep for .eswitch_mode_set being called with the devlink instance lock held use that lock explicitly instead of creating a local mutex just for the sriov reconfig. Signed-off-by:
-
David S. Miller authored
Tong Zhang says: ==================== fix typos: "to short" -> "too short" doing some code review and I found out there are a couple of places where "too short" is misspelled as "to short". ==================== Signed-off-by: -
Tong Zhang authored
"frame to short" -> "frame too short" Signed-off-by:
Tong Zhang <ztong0001@gmail.com> Signed-off-by:
-
Tong Zhang authored
"Frame to short" -> "Frame too short" Signed-off-by:
-
Tong Zhang authored
"packet length to short" -> "packet length too short" Signed-off-by:
-
Tong Zhang authored
"RX USB to short" -> "RX USB too short" Signed-off-by:
-
David S. Miller authored
Casper Andersson says: ==================== net: sparx5: Add multicast support Add multicast support to Sparx5. ==================== Signed-off-by: -
Casper Andersson authored
Adds mdb handlers. Uses the PGID arbiter to find a free entry in the PGID table for the multicast group port mask. Signed-off-by:
Casper Andersson <casper.casan@gmail.com> Signed-off-by:
-
Casper Andersson authored
The PGID (Port Group ID) table holds port masks for different purposes. The first 72 are reserved for port destination masks, flood masks, and CPU forwarding. The rest are shared between multicast, link aggregation, and virtualization profiles. The GLAG area is reserved to not be used by anything else, since it is a subset of the MCAST area. The arbiter keeps track of which entries are in use. You can ask for a free ID or give back one you are done using. Signed-off-by:
-
David S. Miller authored
Simon Horman says: ==================== nfp: support for NFP-3800 Yinjun Zhan says: This is the second of a two part series to support the NFP-3800 device. To utilize the new hardware features of the NFP-3800, driver adds support of a new data path NFDK. This series mainly does some refactor work to the data path related implementations. The data path specific implementations are now separated into nfd3 and nfdk directories respectively, and the common part is also moved into a new file. * The series starts with a small refinement in Patch 1/10. Patches 2/10 and 3/10 are the main refactoring of data path implementation, which prepares for the adding the NFDK data path. * Before the introduction of NFDK, there's some more preparation work for NFP-3800 features, such as multi-descriptor per-packet and write-back mechanism of TX pointer, which is done in patches 4/10, 5/10, 6/10, 7/10. * Patch 8/10 allows the driver to select data path according to firmware version. Finally, patches 9/10 and 10/10 introduce the new NFDK data path. Changes between v1 and v2 * Correct kdoc for nfp_nfdk_tx() * Correct build warnings on 32-bit Thanks to everyone who contributed to this work. ==================== Signed-off-by: -
Yinjun Zhang authored
Due to the different definition of txbuf in NFDK comparing to NFD3, there're no pre-allocated txbufs for xdp use in NFDK's implementation, we just use the existed rxbuf and recycle it when xdp tx is completed. For each packet to transmit in xdp path, we cannot use more than `NFDK_TX_DESC_PER_SIMPLE_PKT` txbufs, one is to stash virtual address, and another is for dma address, so currently the amount of transmitted bytes is not accumulated. Also we borrow the last bit of virtual addr to indicate a new transmitted packet due to address's alignment attribution. Signed-off-by:
Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by:
Fei Qin <fei.qin@corigine.com> Signed-off-by:
Simon Horman <simon.horman@corigine.com> Signed-off-by:
-
Jakub Kicinski authored
Add new data path. The TX is completely different, each packet has multiple descriptor entries (between 2 and 32). TX ring is divided into blocks 32 descriptor, and descritors of one packet can't cross block bounds. The RX side is the same for now. ABI version 5 or later is required. There is no support for VLAN insertion on TX. XDP_TX action and AF_XDP zero-copy is not implemented in NFDK path. Changes to Jakub's work: * Move statistics of hw_csum_tx after jumbo packet's segmentation. * Set L3_CSUM flag to enable recaculating of L3 header checksum in ipv4 case. * Mark the case of TSO a packet with metadata prepended as unsupported. Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by:
Xingfeng Hu <xingfeng.hu@corigine.com> Signed-off-by:
Dianchao Wang <dianchao.wang@corigine.com> Signed-off-by:
-
Jakub Kicinski authored
Prepare for choosing data path based on the firmware version field. Exploit one bit from the reserved byte in the firmware version field as the data path type. We need the firmware version right after vNIC is allocated, so it has to be read inside nfp_net_alloc(), callers don't have to set it afterwards. Following patches will bring the implementation of the second data path. Signed-off-by:
-
Jakub Kicinski authored
Make sure that features supported only by some of the data paths are not enabled for all. Add a mask of supported features into the data path op structure. Signed-off-by:
-
Jakub Kicinski authored
Newer versions of the PCIe microcode support writing back the position of the TX pointer back into host memory. This speeds up TX completions, because we avoid a read from device memory (replacing PCIe read with DMA coherent read). Signed-off-by:
-
