FreeBSD source tree https://cgit-dev.freebsd.org/src-test/atom?h=main 2020-11-16T10:15:03Z 2020-11-16T10:15:03Z Hans Petter Selasky hselasky@FreeBSD.org 2020-11-16T10:15:03Z urn:sha1:7eefcb5eead518d9db75411beece922618587297 APIs that have deferred callbacks should have some kind of cleanup function that callers can use to fence the callbacks. Otherwise things like module unloading can lead to dangling function pointers, or worse. The IB MR code is the only place that calls this function and had a really poor attempt at creating this fence. Provide a good version in the core code as future patches will add more places that need this fence. Linux commit: e355477ed9e4f401e3931043df97325d38552d54 MFC after: 1 week Sponsored by: Mellanox Technologies // NVIDIA Networking 2020-11-16T10:10:53Z Hans Petter Selasky hselasky@FreeBSD.org 2020-11-16T10:10:53Z urn:sha1:f34f0a65b2223db5f307a0e6d376e1cf89623da6 Report EQE data upon CQ completion to let upper layers use this data. Linux commit: 4e0e2ea1886afe8c001971ff767f6670312a9b04 MFC after: 1 week Sponsored by: Mellanox Technologies // NVIDIA Networking 2020-11-16T10:06:10Z Hans Petter Selasky hselasky@FreeBSD.org 2020-11-16T10:06:10Z urn:sha1:ffdb195f318ac3ceeacbae6cf19cd2949660a55c Enhance mlx5_core_create_cq() to get the command out buffer from the callers to let them use the output. Linux commit: 38164b771947be9baf06e78ffdfb650f8f3e908e MFC after: 1 week Sponsored by: Mellanox Technologies // NVIDIA Networking 2020-11-16T10:03:18Z Hans Petter Selasky hselasky@FreeBSD.org 2020-11-16T10:03:18Z urn:sha1:4a64b690f130b7500493869e41456ce4f3d9c50b To prevent a hardware memory leak when a DEVX DCT object is destroyed without calling drain DCT before, (e.g. under cleanup flow), need to manage its creation and destruction via mlx5 core. Linux commit: c5ae1954c47d3fd8815bd5a592aba18702c93f33 MFC after: 1 week Sponsored by: Mellanox Technologies // NVIDIA Networking 2020-10-07T17:46:49Z Hans Petter Selasky hselasky@FreeBSD.org 2020-10-07T17:46:49Z urn:sha1:194ddc011ad10900a405fb81cf48a022badf77b8 Cleanup all host resources, SYSCTLs, MSIX vectors and memory used by the host and only leave the device allocated memory behind, if any, because it may still be in use, when the PCI remove function is called. Else future probe calls may fail due to SYSCTLs already existing. MFC after: 1 week Sponsored by: Mellanox Technologies // NVIDIA Networking 2020-08-31T16:32:17Z Konstantin Belousov kib@FreeBSD.org 2020-08-31T16:32:17Z urn:sha1:596b98ba160e9a3e9aff06664e968afc534ecfcb Setting GUIDs make RoCE offloads functional on VFs. Reported and tested by: chuck Sponsored by: Mellanox Technologies - Nvidia MFC after: 1 week 2020-08-31T16:30:52Z Konstantin Belousov kib@FreeBSD.org 2020-08-31T16:30:52Z urn:sha1:cca1f7a12fab1942b3bbff3dccddb02de14ceac4 Sponsored by: Mellanox Technologies - Nvidia MFC after: 1 week 2020-08-31T16:27:03Z Konstantin Belousov kib@FreeBSD.org 2020-08-31T16:27:03Z urn:sha1:2ea114b34e98862ffb75c3285c8f66f22bd4fb99 Sponsored by: Mellanox Technologies - Nvidia MFC after: 1 week 2020-08-31T16:25:55Z Konstantin Belousov kib@FreeBSD.org 2020-08-31T16:25:55Z urn:sha1:62daa4b6e897d772e338cb921fc194dda1a65761 And use it in mlx5_query_pddr_range_info() instead of direct register access. Sponsored by: Mellanox Technologies - Nvidia MFC after: 1 week 2020-08-31T16:23:51Z Konstantin Belousov kib@FreeBSD.org 2020-08-31T16:23:51Z urn:sha1:e088db5eaec5b43e1effc7edb82b37c4a0ff0282 PDDR (Port Diagnostics Database Register) is used to read the physical layer debug database, which contains helpful troubleshooting information regarding the state of the link. PDDR register can only be queried when PCAM register reports it as supported in its register mask. A new helper macro was added to the MLX5_CAP_* infrastructure in order to access this mask. Sponsored by: Mellanox Technologies - Nvidia MFC after: 1 week