Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 8 Apr 2021 16:51:27 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/4fa56ad0d12e24df768c98bffe9039f915d1bc02 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Nothing very exciting here, just a few small bug fixes. No red flags for this release have shown up. Thanks, Jason The following changes since commit a5e13c6df0e41702d2b2c77c8ad41677ebb065b3: Linux 5.12-rc5 (2021-03-28 15:48:16 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to d1c803a9ccd7bd3aff5e989ccfb39ed3b799b975: RDMA/addr: Be strict with gid size (2021-04-08 16:14:56 -0300) RDMA 5.12 third rc pull request Several bug fixes: - Regression from the last pull request in cxgb4 related to the ipv6 fixes - KASAN crasher in rtrs - oops in hfi1 related to a buggy BIOS - Userspace could oops qedr's XRC support - Uninitialized memory when parsing a LS_NLA_TYPE_DGID netlink message Kamal Heib (1): RDMA/qedr: Fix kernel panic when trying to access recv_cq Leon Romanovsky (1): RDMA/addr: Be strict with gid size Md Haris Iqbal (1): RDMA/rtrs-clt: Close rtrs client conn before destroying rtrs clt session files Mike Marciniszyn (1): IB/hfi1: Fix probe time panic when AIP is enabled with a buggy BIOS Potnuri Bharat Teja (1): RDMA/cxgb4: check for ipv6 address properly while destroying listener drivers/infiniband/core/addr.c | 4 +++- drivers/infiniband/hw/cxgb4/cm.c | 3 ++- drivers/infiniband/hw/hfi1/affinity.c | 21 + drivers/infiniband/hw/hfi1/hfi.h | 1 + drivers/infiniband/hw/hfi1/init.c | 10 +- drivers/infiniband/hw/hfi1/netdev_rx.c | 3 +-- drivers/infiniband/hw/qedr/verbs.c | 3 ++- drivers/infiniband/ulp/rtrs/rtrs-clt.c | 2 +- 8 files changed, 24 insertions(+), 23 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 25 Mar 2021 15:04:21 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/2ba9bea2d3682361f0f22f68a400bcee4248c205 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Not much going on, just some small bug fixes Thanks, Jason The following changes since commit a38fd8748464831584a19438cbb3082b5a2dab15: Linux 5.12-rc2 (2021-03-05 17:33:41 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 3408be145a5d6418ff955fe5badde652be90e700: RDMA/cxgb4: Fix adapter LE hash errors while destroying ipv6 listening server (2021-03-25 10:25:58 -0300) RDMA 5.12 second rc pull request - Typo causing a regression in mlx5 devx - Regression in the recent hns rework causing the HW to get out of sync - Longstanding cxgb4 adaptor crash when destroying cm ids Lang Cheng (1): RDMA/hns: Fix bug during CMDQ initialization Mark Zhang (1): RDMA/mlx5: Fix typo in destroy_mkey inbox Potnuri Bharat Teja (1): RDMA/cxgb4: Fix adapter LE hash errors while destroying ipv6 listening server drivers/infiniband/hw/cxgb4/cm.c | 4 ++-- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 4 +++- drivers/infiniband/hw/mlx5/devx.c | 2 +- 3 files changed, 6 insertions(+), 4 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 5 Mar 2021 19:35:41 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/f3ed4de6cc8327e4ef79e6c7892b2b5cbbc02405 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Nothing special here, though Bob's regression fixes for rxe would have made it before the rc cycle had there not been such strong winter weather! Thanks, Jason The following changes since commit fe07bfda2fb9cdef8a4d4008a409bb02f35f1bd8: Linux 5.12-rc1 (2021-02-28 16:05:19 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 545c4ab463c2224557e56b2609f88ed5be265405: RDMA/rxe: Fix errant WARN_ONCE in rxe_completer() (2021-03-05 14:15:22 -0400) RDMA 5.12 first rc pull request - Fix corner cases in the rxe reference counting cleanup that are causing regressions in blktests for SRP - Two kdoc fixes so W=1 is clean - Missing error return in error unwind for mlx5 - Wrong lock type nesting in IB CM Bob Pearson (3): RDMA/rxe: Fix missed IB reference counting in loopback RDMA/rxe: Fix extra deref in rxe_rcv_mcast_pkt() RDMA/rxe: Fix errant WARN_ONCE in rxe_completer() Julian Braha (1): RDMA/rxe: Fix missing kconfig dependency on CRYPTO Leon Romanovsky (2): RDMA/mlx5: Set correct kernel-doc identifier RDMA/uverbs: Fix kernel-doc warning of _uverbs_alloc Saeed Mahameed (1): RDMA/cm: Fix IRQ restore in ib_send_cm_sidr_rep YueHaibing (1): IB/mlx5: Add missing error code drivers/infiniband/core/cm.c | 5 +-- drivers/infiniband/core/uverbs_ioctl.c | 2 +- drivers/infiniband/hw/mlx5/devx.c | 4 ++- drivers/infiniband/hw/mlx5/odp.c | 2 +- drivers/infiniband/sw/rxe/Kconfig | 1 + drivers/infiniband/sw/rxe/rxe_comp.c | 55 +-- drivers/infiniband/sw/rxe/rxe_net.c| 10 +- drivers/infiniband/sw/rxe/rxe_recv.c | 59 -- 8 files changed, 76 insertions(+), 62 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Mon, 22 Feb 2021 10:59:21 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/3672ac8ac0d8bece188f82c48770bbe40f234f1e Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.12.
This is quite a small cycle, if not for Lee's 70 patches cleaning the kdocs it
would be well below typical for patch count.
Most of the interesting work here was in the HNS and rxe drivers which got
fairly major internal changes.
As before there was a slightly tricky conflict with the rdma for-rc
tree, which I've resolved by merging in v5.11 at the top.
The following changes since commit a6a217dddcd544f6b75f0e2a60b6e84c1d494b7e:
net/mlx5: Add new timestamp mode bits (2021-02-16 16:10:39 +0200)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 7289e26f395b583f68b676d4d12a0971e4f6f65c:
Merge tag 'v5.11' into rdma.git for-next (2021-02-18 11:19:29 -0400)
RDMA 5.12 merge window pull request
- Driver updates and bug fixes: siw, hns, bnxt_re, mlx5, efa
- Significant rework in rxe to get it ready to have XRC support added
- Several rts bug fixes
- Big series to get to 'make W=1' cleanness, primarily updating kdocs
- Support for creating a RDMA MR from a DMABUF fd to allow PCI peer to
peer transfers to GPU VRAM
- Device disassociation now works properly with umad
- Work to support more than 255 ports on a RDMA device
- Further support for the new HNS HIP09 hardware
- Coding style cleanups: comma to semicolon, unneded semicolon/blank
lines, remove 'h' printk format, don't check for NULL before kfree,
use true/false for bool.
Aharon Landau (1):
RDMA/mlx5: Fail QP creation if the device can not support the CQE TS
Avihai Horon (1):
RDMA/ucma: Fix use-after-free bug in ucma_create_uevent
Bernard Metzler (1):
RDMA/siw: Fix handling of zero-sized Read and Receive Queues.
Bob Pearson (21):
RDMA/rxe: Remove unneeded RXE_POOL_ATOMIC flag
RDMA/rxe: Let pools support both keys and indices
RDMA/rxe: Add elem_offset field to rxe_type_info
RDMA/rxe: Make pool lookup and alloc APIs type safe
RDMA/rxe: Make add/drop key/index APIs type safe
RDMA/rxe: Add unlocked versions of pool APIs
RDMA/rxe: Fix race in rxe_mcast.c
RDMA/rxe: Fix bug in rxe_alloc()
RDMA/rxe: Fix misleading comments and names
RDMA/rxe: Remove RXE_POOL_ATOMIC
RDMA/rxe: Remove references to ib_device and pool
RDMA/rxe: Remove unneeded pool->state
RDMA/rxe: Replace missing rxe_pool_get_index_locked
RDMA/rxe: Fix coding error in rxe_recv.c
RDMA/rxe: Remove useless code in rxe_recv.c
RDMA/rxe: Fix coding error in rxe_rcv_mcast_pkt
RDMA/rxe: Correct skb on loopback path
RDMA/rxe: Fix FIXME in rxe_udp_encap_recv()
RDMA/rxe: Fix minor coding style issues
RDMA/rxe: Cleanup init_send_wqe
RDMA/rxe: Remove unused pkt->offset
Christoph Lameter (1):
RDMA/ipoib: Remove racy Subnet Manager sendonly join checks
Gal Pressman (5):
RDMA/efa: Remove redundant NULL pointer check of CQE
RDMA/efa: Remove duplication of upper/lower_32_bits
RDMA/efa: Remove unnecessary indentation in defs comments
RDMA/efa: Remove unused 'select' field from get/set feature command
descriptor
RDMA/efa: Remove unused syndrome enum values
Gioh Kim (2):
RDMA/rtrs-srv: fix memory leak by missing kobject free
RDMA/rtrs-srv-sysfs: fix missing put_device
Guoqing Jiang (8):
RDMA/rtrs-srv: Jump to dereg_mr label if allocate iu fails
RDMA/rtrs: Call kobject_put in the failure path
RDMA/rtrs-clt: Consolidate rtrs_clt_destroy_sysfs_root_{folder,files}
RDMA/rtrs-clt: Kill wait_for_inflight_permits
RDMA/rtrs-clt: Remove unnecessary 'goto out'
RDMA/rtrs-clt: Kill rtrs_clt_change_state
RDMA/rtrs-clt: Rename __rtrs_clt_change_state to rtrs_clt_change_state
RDMA/rtrs-clt: Refactor the failure cases in alloc_clt
Jack Wang (12):
RDMA/rtrs: Extend ibtrs_cq_qp_create
RDMA/rtrs-srv: Release lock before call into close_sess
RDMA/rtrs-srv: Use sysfs_remove_file_self for disconnect
RDMA/rtrs-clt: Set mininum limit when create QP
RDMA/rtrs-srv: Fix missing wr_cqe
RDMA/rtrs: Do not signal for heatbeat
RDMA/rtrs-clt: Use bitmask to check sess->flags
RDMA/rtrs-srv: Do not signal REG_MR
RDMA/rtrs-srv: Init wr_cnt as 1
RDMA/rtrs: Fix KASAN: stack-out-of-bounds bug
RDMA/rtrs-srv: Fix stack-out-of-bounds
RDMA/rtrs-srv: Do not pass a valid pointer to PTR_ERR()
Jason Gunthorpe (3):
Merge branch 'devx_set_get' into rdma.git for-next
Merge branch 'mlx5_timestamp' into rdma.git for-next
Merge tag 'v5.11' into rdma.git for-next
Jianxin Xiong (4):
RDMA/umem: Support importing dma-buf as user memory region
RDMA/core: Add device method for registering dma-buf based memory region
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Nothing very exciting here, more evidence that people are infrequently testing areas, at least. Thanks, Jason The following changes since commit 19c329f6808995b142b3966301f217c831e7cf31: Linux 5.11-rc4 (2021-01-17 16:37:05 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to f1b0a8ea9f12b8ade0dbe40dd57e4ffa9a30ed93: Revert "RDMA/rxe: Remove VLAN code leftovers from RXE" (2021-01-20 13:29:28 -0400) RDMA 5.11 second RC pull request Several recent regressions and some bug fixes - Typo corrupting the max_recv_sge for cxgb4 - Regression from re-using kernel enums as a HW AbI in vmw_pvrdma - Sleeping inside a spinlock in hns - Revert the attempt to fix devlink deadlocks as the fix is more buggy - Typo in sysfs_emit_at conversions - Revert the removal of VLAN support in rxe Bryan Tan (1): RDMA/vmw_pvrdma: Fix network_hdr_type reported in WC Joe Perches (1): RDMA/usnic: Fix misuse of sysfs_emit_at Kamal Heib (1): RDMA/cxgb4: Fix the reported max_recv_sge value Martin Wilck (1): Revert "RDMA/rxe: Remove VLAN code leftovers from RXE" Parav Pandit (1): Revert "RDMA/mlx5: Fix devlink deadlock on net namespace deletion" Yangyang Li (1): RDMA/hns: Use mutex instead of spinlock for ida allocation drivers/infiniband/hw/cxgb4/qp.c | 2 +- drivers/infiniband/hw/hns/hns_roce_device.h| 2 +- drivers/infiniband/hw/hns/hns_roce_qp.c| 11 ++- drivers/infiniband/hw/mlx5/main.c | 6 ++ drivers/infiniband/hw/usnic/usnic_ib_sysfs.c | 7 +++ drivers/infiniband/hw/vmw_pvrdma/pvrdma.h | 14 ++ drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c | 2 +- drivers/infiniband/sw/rxe/rxe_net.c| 6 ++ drivers/infiniband/sw/rxe/rxe_resp.c | 5 + drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h | 5 + include/linux/mlx5/driver.h| 18 -- include/uapi/rdma/vmw_pvrdma-abi.h | 7 +++ 12 files changed, 51 insertions(+), 34 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 15 Jan 2021 15:21:40 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/8cbe71e7e01a9e45a390b204403880c90a226039 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Bit later for the first rc pull request than normal due to the holidays, but a fairly modest set of bug fixes, nothing abnormal from the merge window The ucma patch is a bit on the larger side, but given the regression was recently added I've opted to forward it to the rc stream. Thanks, Jason The following changes since commit e71ba9452f0b5b2e8dc8aa5445198cd9214a6a62: Linux 5.11-rc2 (2021-01-03 15:55:30 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 7c7b3e5d9aeed31d35c5dab0bf9c0fd4c8923206: RDMA/cma: Fix error flow in default_roce_mode_store (2021-01-14 12:53:13 -0400) RDMA 5.11 first RC pull request Several bug fixes. - Fix a ucma memory leak introduced in v5.9 while fixing the Syzkaller bugs - Don't fail when the xarray wraps for user verbs objects - User triggerable oops regression from the umem page size rework - Error unwind bugs in usnic, ocrdma, mlx5 and cma Aharon Landau (1): RDMA/umem: Avoid undefined behavior of rounddown_pow_of_two() Dinghao Liu (1): RDMA/usnic: Fix memleak in find_free_vf_and_create_qp_grp Jason Gunthorpe (1): RDMA/ucma: Do not miss ctx destruction steps in some cases Leon Romanovsky (1): RDMA/restrack: Don't treat as an error allocation ID wrapping Mark Bloch (1): RDMA/mlx5: Fix wrong free of blue flame register on error Neta Ostrovsky (1): RDMA/cma: Fix error flow in default_roce_mode_store Parav Pandit (1): IB/mlx5: Fix error unwinding when set_has_smi_cap fails Tom Rix (1): RDMA/ocrdma: Fix use after free in ocrdma_dealloc_ucontext_pd() drivers/infiniband/core/cma_configfs.c | 4 +- drivers/infiniband/core/restrack.c | 1 + drivers/infiniband/core/ucma.c | 135 ++- drivers/infiniband/core/umem.c | 2 +- drivers/infiniband/hw/mlx5/main.c| 4 +- drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 2 +- drivers/infiniband/hw/usnic/usnic_ib_verbs.c | 3 + 7 files changed, 83 insertions(+), 68 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Wed, 16 Dec 2020 13:57:30 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/009bd55dfcc857d8b00a5bbb17a8db060317af6f Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.12.
The biggest item in this PR would be the new HIP09 HW support from
HNS, otherwise it was pretty quiet for new work here.
Thanks,
Jason
The following changes since commit b65054597872ce3aefbc6a666385eabdf9e288da:
Linux 5.10-rc6 (2020-11-29 15:50:50 -0800)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to e246b7c035d74abfb3507fa10082d0c42cc016c3:
RDMA/cma: Don't overwrite sgid_attr after device is released (2020-12-14
15:23:06 -0400)
RDMA 5.11 pull request
A smaller set of patches, nothing stands out as being particularly major
this cycle:
- Driver bug fixes and updates: bnxt_re, cxgb4, rxe, hns, i40iw, cxgb4,
mlx4 and mlx5
- Bug fixes and polishing for the new rts ULP
- Cleanup of uverbs checking for allowed driver operations
- Use sysfs_emit all over the place
- Lots of bug fixes and clarity improvements for hns
- hip09 support for hns
- NDR and 50/100Gb signaling rates
- Remove dma_virt_ops and go back to using the IB DMA wrappers
- mlx5 optimizations for contiguous DMA regions
Arnd Bergmann (2):
IB/verbs: avoid nested container_of()
RDMa/mthca: Work around -Wenum-conversion warning
Avihai Horon (3):
RDMA/mlx5: Enable querying AH for XRC QP types
RDMA/mlx4: Enable querying AH for XRC QP types
RDMA/uverbs: Fix incorrect variable type
Bob Pearson (3):
RDMA/rxe: Remove unused RXE_MR_TYPE_FMR
RDMA/rxe: Compute PSN windows correctly
RDMA/rxe: Use acquire/release for memory ordering
Christoph Hellwig (9):
RDMA/umem: Use ib_dma_max_seg_size instead of dma_get_max_seg_size
RDMA/core: Remove ib_dma_{alloc,free}_coherent
RDMA: Lift ibdev_to_node from rds to common code
nvme-rdma: Use ibdev_to_node instead of dereferencing ->dma_device
rds: stop using dmapool
RDMA/core: remove use of dma_virt_ops
PCI/P2PDMA: Remove the DMA_VIRT_OPS hacks
PCI/P2PDMA: Cleanup __pci_p2pdma_map_sg a bit
dma-mapping: remove dma_virt_ops
Christophe JAILLET (1):
IB/qib: Use dma_set_mask_and_coherent to simplify code
Danil Kipnis (1):
RDMA/rtrs-clt: Remove destroy_con_cq_qp in case route resolving failed
Gal Pressman (2):
RDMA/efa: Remove .create_ah callback assignment
RDMA/efa: Use dma_set_mask_and_coherent() to simplify code
Gioh Kim (4):
RDMA/rtrs-clt: Missing error from rtrs_rdma_conn_established
RDMA/rtrs: Remove unnecessary argument dir of rtrs_iu_free
RDMA/rtrs-clt: Remove duplicated switch-case handling for CM error events
RDMA/rtrs-clt: Remove duplicated code
Guoqing Jiang (5):
RDMA/rtrs-srv: Don't guard the whole __alloc_srv with srv_mutex
RDMA/rtrs-srv: Fix typo
RDMA/rtrs-srv: Kill rtrs_srv_change_state_get_old
RDMA/rtrs: Introduce rtrs_post_send
RDMA/rtrs-clt: Remove 'addr' from rtrs_clt_add_path_to_arr
Gustavo A. R. Silva (4):
IB/hfi1: Fix fall-through warnings for Clang
IB/mlx4: Fix fall-through warnings for Clang
IB/qedr: Fix fall-through warnings for Clang
IB/mlx5: Fix fall-through warnings for Clang
Jack Morgenstein (2):
RDMA/core: Clean up cq pool mechanism
RDMA/core: Do not indicate device ready when device enablement fails
Jack Wang (3):
RDMA/rtrs-clt: Remove outdated comment in create_con_cq_qp
RDMA/rtrs-clt: Avoid run destroy_con_cq_qp/create_con_cq_qp in parallel
RDMA/ipoib: Distribute cq completion vector better
Jason Gunthorpe (43):
RDMA/cxgb4: Remove MW support
RDMA: Remove uverbs_ex_cmd_mask values that are linked to functions
RDMA: Remove elements in uverbs_cmd_mask that all drivers set
RDMA: Move more uverbs_cmd_mask settings to the core
RDMA: Check srq_type during create_srq
RDMA: Check attr_mask during modify_qp
RDMA: Check flags during create_cq
RDMA: Check create_flags during create_qp
RDMA/core Remove uverbs_ex_cmd_mask
RDMA: Remove uverbs cmds from drivers that don't use them
RDMA: Remove AH from uverbs_cmd_mask
RDMA/mlx5: Remove mlx5_ib_mr->order
RDMA/mlx5: Fix corruption of reg_pages in mlx5_ib_rereg_user_mr()
RDMA/mlx5: Remove mlx5_ib_mr->npages
RDMA/mlx5: Move mlx5_ib_cont_pages() to the creation of the mlx5_ib_mr
RDMA/mlx5: Remove order from mlx5_ib_cont_pages()
RDMA/mlx5: Remove ncont from mlx5_ib_cont_pages()
RDMA/mlx5: Remove npages from mlx5_ib_cont_pages()
RDMA/mlx5: Change mlx5_ib_populate_pas() to use rdma_for_each_block()
RDMA/mlx5: Move xlt_emergency_page_mutex into mr.c
RDMA/mlx5: Split the WR setup out of mlx5_ib_update_xlt()
RDMA/mlx5: Split mlx5_ib_update_xlt() into ODP and non-ODP cases
RDMA/mlx5: Use
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 10 Dec 2020 11:50:05 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/9fca90cf28920c6d0723d7efd1eae0b0fb90309c Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, A few fairly normal bug fixes late in the cycle, not seeing anything worrysome for release here. The following changes since commit b65054597872ce3aefbc6a666385eabdf9e288da: Linux 5.10-rc6 (2020-11-29 15:50:50 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 340b940ea0ed12d9adbb8f72dea17d516b2019e8: RDMA/cm: Fix an attempt to use non-valid pointer when cleaning timewait (2020-12-09 15:51:35 -0400) RDMA 5.10 fifth rc pull request Two user triggerable crashers and a some EFA related regressions: - Syzkaller found bug in CM - Restore access to the GID table and fix modify_qp for EFA - Crasher in qedr Alok Prasad (1): RDMA/qedr: iWARP invalid(zero) doorbell address fix Gal Pressman (2): RDMA/efa: Use the correct current and new states in modify QP RDMA/core: Fix empty gid table for non IB/RoCE devices Leon Romanovsky (1): RDMA/cm: Fix an attempt to use non-valid pointer when cleaning timewait drivers/infiniband/core/cache.c | 3 --- drivers/infiniband/core/cm.c | 2 ++ drivers/infiniband/hw/efa/efa_verbs.c | 4 ++-- drivers/infiniband/hw/qedr/verbs.c| 9 + 4 files changed, 13 insertions(+), 5 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 27 Nov 2020 10:00:52 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/d41e9b22eb871a7a7060964db9ce1ceb1c6e5b57 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Two security issues and several small bug fixes. Things seem to have stabilized for this release here. The following changes since commit 418baf2c28f3473039f2f7377760bd8f6897ae18: Linux 5.10-rc5 (2020-11-22 15:36:08 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 17475e104dcb74217c282781817f8f52b46130d3: RDMA/hns: Bugfix for memory window mtpt configuration (2020-11-26 10:57:32 -0400) RDMA 5.10 fourth rc pull request Two notable security issues and a collection of minor fixes: - Significant out of bounds access security issue in i40iw - Fix misuse of mmu notifiers in hfi1 - Several errors in the register map/usage in hns - Missing error returns in mthca Dennis Dalessandro (1): IB/hfi1: Ensure correct mm is used at all times Shiraz Saleem (1): RDMA/i40iw: Address an mmap handler exploit in i40iw Wenpeng Liang (2): RDMA/hns: Fix wrong field of SRQ number the device supports RDMA/hns: Fix retry_cnt and rnr_cnt when querying QP Xiongfeng Wang (1): IB/mthca: fix return value of error branch in mthca_init_cq() Yixian Liu (1): RDMA/hns: Bugfix for memory window mtpt configuration drivers/infiniband/hw/hfi1/file_ops.c | 4 +- drivers/infiniband/hw/hfi1/hfi.h | 2 +- drivers/infiniband/hw/hfi1/mmu_rb.c| 68 +++--- drivers/infiniband/hw/hfi1/mmu_rb.h| 16 ++- drivers/infiniband/hw/hfi1/user_exp_rcv.c | 12 -- drivers/infiniband/hw/hfi1/user_exp_rcv.h | 6 +++ drivers/infiniband/hw/hfi1/user_sdma.c | 13 +++--- drivers/infiniband/hw/hfi1/user_sdma.h | 7 ++- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 9 ++-- drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 2 +- drivers/infiniband/hw/i40iw/i40iw_main.c | 5 --- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 37 +++- drivers/infiniband/hw/mthca/mthca_cq.c | 10 +++-- 13 files changed, 98 insertions(+), 93 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 19 Nov 2020 15:29:25 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/3be28e93cd88fbcbe97cabcbe92b1ccc9f830450 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, The last two weeks have been quiet here, just the usual smattering of long standing bug fixes. The following changes since commit f8394f232b1eab649ce2df5c5f15b0e528c92091: Linux 5.10-rc3 (2020-11-08 16:10:16 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to ee415d73dcc24caef7f6bbf292dcc365613d2188: tools/testing/scatterlist: Fix test to compile and run (2020-11-17 20:02:20 -0400) RDMA 5.10 third rc pull request A collection of error case bug fixes - Improper nesting of spinlock types in cm - Missing error codes and kfree() - Ensure dma_virt_ops users have the right kconfig symbols to work properly - Compilation failure of tools/testing Christoph Hellwig (1): RMDA/sw: Don't allow drivers using dma_virt_ops on highmem configs Jason Gunthorpe (1): RDMA/cm: Make the local_id_table xarray non-irq Maor Gottlieb (1): tools/testing/scatterlist: Fix test to compile and run Qinglang Miao (1): RDMA/pvrdma: Fix missing kfree() in pvrdma_register_device() Zhang Changzhong (1): IB/hfi1: Fix error return code in hfi1_init_dd() drivers/infiniband/Kconfig | 3 +++ drivers/infiniband/core/cm.c | 12 ++-- drivers/infiniband/hw/hfi1/chip.c | 3 ++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c | 2 +- drivers/infiniband/sw/rdmavt/Kconfig | 3 ++- drivers/infiniband/sw/rxe/Kconfig | 2 +- drivers/infiniband/sw/siw/Kconfig | 1 + tools/testing/scatterlist/linux/mm.h | 1 + tools/testing/scatterlist/main.c | 4 ++-- 9 files changed, 19 insertions(+), 12 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 5 Nov 2020 14:16:36 -0400: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/6f3f374ac05d05cfa63d04f4479ead7e3cb6d087 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, This rc cycle continues to be challenging with an abnormal number of regressions. The following changes since commit 3cea11cd5e3b00d91caf0b4730194039b45c5891: Linux 5.10-rc2 (2020-11-01 14:43:51 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 21fcdeec09ff461b2f9a9ef4fcc3a136249e58a1: RDMA/srpt: Fix typo in srpt_unregister_mad_agent docstring (2020-11-05 11:38:29 -0400) RDMA 5.10 second rc pull request A few more merge window regressions that didn't make rc1: - New validation in the DMA layer triggers wrong use of the DMA layer in rxe, siw and rdmavt - Accidental change of a hypervisor facing ABI when widening the port speed u8 to u16 in vmw_pvrdma - Memory leak on error unwind in SRP target Adit Ranadive (1): RDMA/vmw_pvrdma: Fix the active_speed and phys_state value Jason Gunthorpe (1): RDMA/srpt: Fix typo in srpt_unregister_mad_agent docstring Maor Gottlieb (1): IB/srpt: Fix memory leak in srpt_add_one Parav Pandit (1): RDMA: Fix software RDMA drivers for dma mapping error drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h | 2 +- drivers/infiniband/sw/rdmavt/vt.c | 7 +-- drivers/infiniband/sw/rxe/rxe_verbs.c | 6 +- drivers/infiniband/sw/siw/siw_main.c| 7 +-- drivers/infiniband/ulp/srpt/ib_srpt.c | 13 - 5 files changed, 24 insertions(+), 11 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 29 Oct 2020 15:41:23 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/b9c0f4bd5b8114ee1773734e07cda921b6e8248b Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
Re: [GIT PULL] Please pull RDMA subsystem changes
On Thu, Oct 29, 2020 at 11:41 AM Jason Gunthorpe wrote: > > Three notable merge window regressions that didn't get caught/fixed in > time for rc1: Three .. and then you list five things ;) Linus
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request The good news is people are testing rc1 in the RDMA world - the bad news is testing of the for-next area is not as good as I had hoped, as we really should have caught at least the rdma_connect_locked() issue before now. Here are several regression fixes for issues found by people testing rc1. The following changes since commit 3650b228f83adda7e5ee532e2b90429c03f7b9ec: Linux 5.10-rc1 (2020-10-25 15:14:11 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to a2267f8a52eea9096861affd463f691be0f0e8c9: RDMA/qedr: Fix memory leak in iWARP CM (2020-10-28 09:45:25 -0300) RDMA 5.10 first rc pull request Three notable merge window regressions that didn't get caught/fixed in time for rc1: - Fix in kernel users of rxe, they were broken by the rapid fix to undo the uABI breakage in rxe from another patch - EFA userspace needs to read the GID table but was broken with the new GID table logic - Fix user triggerable deadlock in mlx5 using devlink reload - Fix deadlock in several ULPs using rdma_connect from the CM handler callbacks - Memory leak in qedr Alok Prasad (1): RDMA/qedr: Fix memory leak in iWARP CM Bob Pearson (1): RDMA/rxe: Fix small problem in network_type patch Gal Pressman (1): RDMA/uverbs: Fix false error in query gid IOCTL Jason Gunthorpe (1): RDMA: Add rdma_connect_locked() Parav Pandit (1): RDMA/mlx5: Fix devlink deadlock on net namespace deletion drivers/infiniband/core/cma.c | 48 +- drivers/infiniband/core/uverbs_std_types_device.c | 3 -- drivers/infiniband/hw/mlx5/main.c | 6 ++- drivers/infiniband/hw/qedr/qedr_iw_cm.c| 1 + drivers/infiniband/sw/rxe/rxe_av.c | 35 ++-- drivers/infiniband/sw/rxe/rxe_net.c| 2 +- drivers/infiniband/ulp/iser/iser_verbs.c | 2 +- drivers/infiniband/ulp/rtrs/rtrs-clt.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h | 5 --- drivers/nvme/host/rdma.c | 4 +- include/linux/mlx5/driver.h| 18 include/rdma/rdma_cm.h | 14 +-- net/rds/ib_cm.c| 5 ++- 13 files changed, 103 insertions(+), 44 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sat, Oct 17, 2020 at 11:21:51AM -0700, Linus Torvalds wrote: > On Fri, Oct 16, 2020 at 11:52 AM Jason Gunthorpe wrote: > > > > You'll need to apply this fixup to the merge commit (it is in the tag > > for-linus-merged for reference): > > Ugh. That's unbelievably and unnecessarily ugly. > > There's no point in that unnecessary "ret" variable and the "goto out" > etc, when all the error cases can be handled much more directly. Yep > So I resolved that merge issue somewhat differently. I can't test the > end result, but it looks TriviallyCorrect(tm). > > Famous last words. Feel free to make fun of me and call me names if that > breaks. Not familiar with DRM land, but it looks trivially fine to me too. Thanks, Jason
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 16 Oct 2020 15:51:55 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/a1e16bc7d5f7ca3599d8a7f061841c93a563665e Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
Re: [GIT PULL] Please pull RDMA subsystem changes
On Fri, Oct 16, 2020 at 11:52 AM Jason Gunthorpe wrote: > > You'll need to apply this fixup to the merge commit (it is in the tag > for-linus-merged for reference): Ugh. That's unbelievably and unnecessarily ugly. There's no point in that unnecessary "ret" variable and the "goto out" etc, when all the error cases can be handled much more directly. So I resolved that merge issue somewhat differently. I can't test the end result, but it looks TriviallyCorrect(tm). Famous last words. Feel free to make fun of me and call me names if that breaks. Linus I did it v
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.10.
A usual cycle for RDMA with a typical mix of driver and core subsystem
updates. There was a lot of activity in the last week on fixing up some small
patches that were merged during the rc8 period, so the top couple commits are
new things that haven't been in linux-next, but they are all rc-ish unbreaking
stuff.
There is a compilation break after merging this, the signature of
__sg_alloc_table_from_pages() was changed in this PR. The series that did this
has been reviewed and been in linux-next for a while.
You'll need to apply this fixup to the merge commit (it is in the tag
for-linus-merged for reference):
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 11fe9ff76fd572..8ee53839825338 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -807,6 +807,7 @@ struct sg_table *drm_prime_pages_to_sg(struct drm_device
*dev,
struct page **pages, unsigned int
nr_pages)
{
struct sg_table *sg = NULL;
+ struct scatterlist *sge;
size_t max_segment = 0;
int ret;
@@ -820,11 +821,13 @@ struct sg_table *drm_prime_pages_to_sg(struct drm_device
*dev,
max_segment = dma_max_mapping_size(dev->dev);
if (max_segment == 0 || max_segment > SCATTERLIST_MAX_SEGMENT)
max_segment = SCATTERLIST_MAX_SEGMENT;
- ret = __sg_alloc_table_from_pages(sg, pages, nr_pages, 0,
- nr_pages << PAGE_SHIFT,
- max_segment, GFP_KERNEL);
- if (ret)
+ sge = __sg_alloc_table_from_pages(sg, pages, nr_pages, 0,
+ nr_pages << PAGE_SHIFT, max_segment,
+ NULL, 0, GFP_KERNEL);
+ if (IS_ERR(sge)) {
+ ret = PTR_ERR(sge);
goto out;
+ }
return sg;
out:
Thanks,
Jason
The following changes since commit 549738f15da0e5a00275977623be199fbbf7df50:
Linux 5.9-rc8 (2020-10-04 16:04:34 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to c7a198c700763ac89abbb166378f546aeb9afb33:
RDMA/ucma: Fix use after free in destroy id flow (2020-10-16 14:07:08 -0300)
RDMA 5.10 pull request
The typical set of driver updates across the subsystem:
- Driver minor changes and bug fixes for mlx5, efa, rxe, vmw_pvrdma, hns,
usnic, qib, qedr, cxgb4, hns, bnxt_re
- Various rtrs fixes and updates
- Bug fix for mlx4 CM emulation for virtualization scenarios where MRA
wasn't working right
- Use tracepoints instead of pr_debug in the CM code
- Scrub the locking in ucma and cma to close more syzkaller bugs
- Use tasklet_setup in the subsystem
- Revert the idea that 'destroy' operations are not allowed to fail at
the driver level. This proved unworkable from a HW perspective.
- Revise how the umem API works so drivers make fewer mistakes using it
- XRC support for qedr
- Convert uverbs objects RWQ and MW to new the allocation scheme
- Large queue entry sizes for hns
- Use hmm_range_fault() for mlx5 On Demand Paging
- uverbs APIs to inspect the GID table instead of sysfs
- Move some of the RDMA code for building large page SGLs into
lib/scatterlist
Aharon Landau (3):
net/mlx5: Refactor query port speed functions
RDMA/mlx5: Delete duplicated mlx5_ptys_width enum
RDMA: Fix link active_speed size
Alex Dewar (3):
RDMA/qib: Remove superfluous fallthrough statements
RDMA/qib: Tidy up process_cc()
RDMA/ucma: Fix resource leak on error path
Alex Vesker (3):
RDMA/mlx5: Add sw_owner_v2 bit capability
RDMA/mlx5: Allow DM allocation for sw_owner_v2 enabled devices
RDMA/mlx5: Expose TIR and QP ICM address for sw_owner_v2 devices
Allen Pais (5):
RDMA/bnxt_re: Convert tasklets to use new tasklet_setup() API
RDMA/hfi1: Convert tasklets to use new tasklet_setup() API
RDMA/i40iw: Convert tasklets to use new tasklet_setup() API
RDMA/qib: Convert tasklets to use new tasklet_setup() API
RDMA/rxe: Convert tasklets to use new tasklet_setup() API
Alok Prasad (1):
RDMA/qedr: Endianness warnings cleanup
Avihai Horon (4):
RDMA/core: Change rdma_get_gid_attr returned error code
RDMA/core: Modify enum ib_gid_type and enum rdma_network_type
RDMA/core: Introduce new GID table query API
RDMA/uverbs: Expose the new GID query API to user space
Bob Pearson (8):
RDMA/rxe: Fix style warnings
RDMA/rxe: Add SPDX hdrs to rxe source files
RDMA/rxe: Address an issue with hardened user copy
RDMA/core: Added missing WR and WC opcodes
RDMA/rxe: Remove duplicate entries in
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 25 Sep 2020 09:57:29 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/33d04c66f5799befa7c0c618be387541d0c311a3 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Only one small bug fix for rxe found by blktests. The following changes since commit 856deb866d16e29bd65952e0289066f6078af773: Linux 5.9-rc5 (2020-09-13 16:06:00 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 4aa1615268a8ac2b20096211d3f9ac53874067d7: RDMA/core: Fix ordering of CQ pool destruction (2020-09-14 15:20:18 -0300) RDMA third 5.9-rc pull request One fix for a bug that blktests hits when using rxe: - Tear down the CQ pool before waiting for all references to go away Jason Gunthorpe (1): RDMA/core: Fix ordering of CQ pool destruction drivers/infiniband/core/device.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 10 Sep 2020 20:38:10 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/b1df2a0783f3d80d6d37102eb90f06727113c7dc Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Another small RC update, nothing too interesting, 5.9 is looking good so far. Thanks, Jason The following changes since commit d012a7190fc1fd72ed48911e77ca97ba4521bccd: Linux 5.9-rc2 (2020-08-23 14:08:43 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 0b089c1ef7047652b13b4cdfdb1e0e7dbdb8c9ab: IB/isert: Fix unaligned immediate-data handling (2020-09-09 13:46:03 -0300) RDMA second 5.9-rc pull request A number of driver bug fixes and a few recent regressions: - Several bug fixes for bnxt_re. Crashing, incorrect data reported, and corruption on new HW - Memory leak and crash in rxe - Fix sysfs corruption in rxe if the netdev name is too long - Fix a crash on error unwind in the new cq_pool code - Fix kobject panics in rtrs by working device lifetime properly - Fix a data corruption bug in iser target related to misaligned buffers Dinghao Liu (1): RDMA/rxe: Fix memleak in rxe_mem_init_user Kamal Heib (2): RDMA/rxe: Fix panic when calling kmem_cache_create() RDMA/core: Fix reported speed and width Mark Bloch (1): RDMA/mlx4: Read pkey table length instead of hardcoded value Md Haris Iqbal (2): RDMA/rtrs-srv: Replace device_register with device_initialize and device_add RDMA/rtrs-srv: Set .release function for rtrs srv device during device init Naresh Kumar PBS (3): RDMA/bnxt_re: Static NQ depth allocation RDMA/bnxt_re: Restrict the max_gids to 256 RDMA/bnxt_re: Fix driver crash on unaligned PSN entry address Sagi Grimberg (1): IB/isert: Fix unaligned immediate-data handling Selvin Xavier (3): RDMA/bnxt_re: Do not report transparent vlan from QP1 RDMA/bnxt_re: Fix the qp table indexing RDMA/bnxt_re: Remove the qp from list only if the qp destroy succeeds Xi Wang (1): RDMA/core: Fix unsafe linked list traversal after failing to allocate CQ Yi Zhang (1): RDMA/rxe: Fix the parent sysfs read when the interface has 15 chars YueHaibing (1): RDMA/bnxt_re: Remove set but not used variable 'qplib_ctx' drivers/infiniband/core/cq.c | 4 +- drivers/infiniband/core/verbs.c | 2 +- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 43 - drivers/infiniband/hw/bnxt_re/main.c | 5 +- drivers/infiniband/hw/bnxt_re/qplib_fp.c | 26 +--- drivers/infiniband/hw/bnxt_re/qplib_rcfw.c | 10 +-- drivers/infiniband/hw/bnxt_re/qplib_rcfw.h | 5 ++ drivers/infiniband/hw/bnxt_re/qplib_sp.c | 2 +- drivers/infiniband/hw/bnxt_re/qplib_sp.h | 1 + drivers/infiniband/hw/mlx4/main.c| 3 +- drivers/infiniband/sw/rxe/rxe.c | 4 ++ drivers/infiniband/sw/rxe/rxe.h | 2 + drivers/infiniband/sw/rxe/rxe_mr.c | 1 + drivers/infiniband/sw/rxe/rxe_sysfs.c| 5 ++ drivers/infiniband/sw/rxe/rxe_verbs.c| 2 +- drivers/infiniband/ulp/isert/ib_isert.c | 93 ++-- drivers/infiniband/ulp/isert/ib_isert.h | 41 +--- drivers/infiniband/ulp/rtrs/rtrs-srv-sysfs.c | 16 ++--- drivers/infiniband/ulp/rtrs/rtrs-srv.c | 9 +++ 19 files changed, 168 insertions(+), 106 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 21 Aug 2020 11:06:12 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/cd02217a5d813e2bbec984a9eeb8280ff290a150 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request Nothing exciting, just a few bug fixes and a MAINTAINERS file update. There is a small merge conflict in the .mailmap, it has been sorted in your tree and Leon's entry was moved. It should be resolved by adding the new line at the new sorted position: Leon Romanovsky Leon Romanovsky +Leon Romanovsky Thanks, Jason The following changes since commit 9123e3a74ec7b934a4a099e98af6a61c2f80bbf5: Linux 5.9-rc1 (2020-08-16 13:04:57 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to f6da70d99c96256f8be0cbb4dd72d45d622c7823: MAINTAINERS: Update Mellanox and Cumulus Network addresses to new domain (2020-08-21 10:48:48 -0300) RDMA first 5.9rc pull request One regression from 5.8 and a few bugs from earlier kernels. - Various spelling corrections in kernel prints - Bug fixes in hfi1 and bntx_re - Revert a 5.8 patch in hns - Batch update for Mellanox and Cumulus maintainers emails Colin Ian King (3): RDMA/hns: Fix spelling mistake "epmty" -> "empty" RDMA/usnic: Fix spelling mistake "transistion" -> "transition" RDMA/core: Fix spelling mistake "Could't" -> "Couldn't" Kaike Wan (1): RDMA/hfi1: Correct an interlock issue for TID RDMA WRITE request Leon Romanovsky (1): MAINTAINERS: Update Mellanox and Cumulus Network addresses to new domain Selvin Xavier (1): RDMA/bnxt_re: Do not add user qps to flushlist Weihang Li (1): Revert "RDMA/hns: Reserve one sge in order to avoid local length error" .mailmap| 2 + MAINTAINERS | 58 ++--- drivers/infiniband/core/device.c| 2 +- drivers/infiniband/hw/bnxt_re/main.c| 3 +- drivers/infiniband/hw/hfi1/tid_rdma.c | 1 + drivers/infiniband/hw/hns/hns_roce_device.h | 2 - drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 2 +- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 9 ++--- drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 4 +- drivers/infiniband/hw/hns/hns_roce_qp.c | 5 +-- drivers/infiniband/hw/hns/hns_roce_srq.c| 2 +- drivers/infiniband/hw/usnic/usnic_ib_main.c | 2 +- 12 files changed, 45 insertions(+), 47 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 6 Aug 2020 15:27:32 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/d7806bbd22cabc3e3b0a985cfcffa29cf156bb30 Thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, These are the proposed RDMA patches for 5.9. A quiet cycle after the larger 5.8 effort. Substantially cleanup and driver work with a few smaller features this time. There is one conflict with Kees's removal of uninitialized_var - resolve by taking the RDMA version of the conflicting hunks to delete the local variables. The tag for-linus-merged with my merge resolution to your tree is also available for reference. Thanks, Jason The following changes since commit 2d1b69ed65ee033aa541518cc9f6a815296ac493: net/mlx5: kTLS, Improve TLS params layout structures (2020-06-27 13:50:46 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 23fcc7dee2c6aba1060558683988263851e74bab: RDMA/mlx5: Fix flow destination setting for RDMA TX flow table (2020-08-05 21:09:39 -0300) RDMA 5.9 merge window pull request Smaller set of RDMA updates. A smaller number of 'big topics' with the majority of changes being driver updates. - Driver updates for hfi1, rxe, mlx5, hns, qedr, usnic, bnxt_re - Removal of dead or redundant code across the drivers - RAW resource tracker dumps to include a device specific data blob for device objects to aide device debugging - Further advance the IOCTL interface, remove the ability to turn it off. Add QUERY_CONTEXT, QUERY_MR, and QUERY_PD commands - Remove stubs related to devices with no pkey table - A shared CQ scheme to allow multiple ULPs to share the CQ rings of a device to give higher performance - Several more static checker, syzkaller and rare crashers fixed Bolarinwa Olayemi Saheed (1): IB/hfi1: Convert PCIBIOS_* errors to generic -E* errors Christophe JAILLET (1): RDMA/usnic: switch from 'pci_' to 'dma_' API Chuck Lever (1): RDMA/core: Clean up tracepoint headers Colton Lewis (1): RDMA: Correct trivial kernel-doc inconsistencies Danil Kipnis (1): RDMA/rtrs-clt: add an additional random 8 seconds before reconnecting Daria Velikovsky (1): RDMA/mlx5: Init dest_type when create flow Devesh Sharma (6): RDMA/bnxt_re: introduce wqe mode to select execution path RDMA/bnxt_re: introduce a function to allocate swq RDMA/bnxt_re: Pull psn buffer dynamically based on prod RDMA/bnxt_re: Add helper data structures RDMA/bnxt_re: Change wr posting logic to accommodate variable wqes RDMA/bnxt_re: Update maintainers for Broadcom rdma driver Eli Cohen (3): net/mlx5: Support setting access rights of dma addresses net/mlx5: Add VDPA interface type to supported enumerations net/mlx5: Add interface changes required for VDPA Eric Dumazet (1): RDMA/umem: Add a schedule point in ib_umem_get() Gal Pressman (8): RDMA/core: Check for error instead of success in alloc MR function RDMA/core: Remove ib_alloc_mr_user function RDMA: Remove the udata parameter from alloc_mr callback RDMA/mlx5: Remove unused to_mibmr function RDMA/efa: Expose maximum TX doorbell batch RDMA/efa: Expose minimum SQ size RDMA/efa: User/kernel compatibility handshake mechanism RDMA/efa: Add EFA 0xefa1 PCI ID Gustavo A. R. Silva (2): IB/hfi1: Remove unnecessary fall-through markings IB/hfi1: Use fallthrough pseudo-keyword Jack Wang (1): RDMA/rtrs: remove WQ_MEM_RECLAIM for rtrs_wq Jason Gunthorpe (9): Merge branch 'raw_dumps' into rdma.git for-next RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah() RDMA/core: Fix bogus WARN_ON during ib_unregister_device_queued() Merge branch 'mlx5_ipoib_qpn' into rdma.git for-next Merge branch 'mlx5_uar' into rdma.git /for-next RDMA/cma: Simplify DEVICE_REMOVAL for internal_id RDMA/cma: Using the standard locking pattern when delivering the removal event RDMA/cma: Remove unneeded locking for req paths RDMA/cma: Execute rdma_cm destruction from a handler properly Jing Xiangfeng (1): IB/srpt: Remove WARN_ON from srpt_cm_req_recv Kamal Heib (16): RDMA/ipoib: Return void from ipoib_mcast_stop_thread() RDMA/hfi1: Remove hfi1_create_qp declaration RDMA/rxe: Remove unused rxe_mem_map_pages RDMA/ipoib: Return void from ipoib_ib_dev_stop() RDMA/rxe: Drop pointless checks in rxe_init_ports RDMA/rxe: Return void from rxe_init_port_param() RDMA/rxe: Return void from rxe_mem_init_dma() RDMA/rxe: Remove rxe_link_layer() RDMA/core: Expose pkeys sysfs files only if pkey_tbl_len is set RDMA/core: Allocate the pkey cache only if the pkey_tbl_len is set RDMA/core: Remove query_pkey from the mandatory ops RDMA/siw: Remove the query_pkey callback RDMA/cxgb4: Remove the query_pkey callback RDMA/i40iw: Remove the query_pkey callback RDMA/qedr:
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 31 Jul 2020 12:17:02 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/ae2911de2eb5dc9ccb88c6502c776a8fbb7acc67 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Still a few regressions from the merge window and the usual collection of fixes. The following changes since commit 92ed301919932f13b9172e525674157e983d: Linux 5.8-rc7 (2020-07-26 14:14:06 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to fb448ce87a4a9482b084e67faf804aec79ed9b43: RDMA/core: Free DIM memory in error unwind (2020-07-30 11:03:33 -0300) RDMA fourth 5.8 rc pull request Two more merge window regressions, a corruption bug in hfi1 and a few other small fixes. - Missing user input validation regression in ucma - Disallowing a previously allowed user combination regression in mlx5 - ODP prefetch memory leaking triggerable by userspace - Memory corruption in hf1 due to faulty ring buffer logic - Missed mutex initialization crash in mlx5 - Two small defects with RDMA DIM Jason Gunthorpe (2): RDMA/cm: Add min length checks to user structure copies RDMA/mlx5: Fix prefetch memory leak if get_prefetchable_mr fails Leon Romanovsky (4): RDMA/mlx5: Allow providing extra scatter CQE QP flag RDMA/mlx5: Initialize QP mutex for the debug kernels RDMA/core: Stop DIM before destroying CQ RDMA/core: Free DIM memory in error unwind Mike Marciniszyn (1): IB/rdmavt: Fix RQ counting issues causing use of an invalid RWQE drivers/infiniband/core/cq.c | 14 +++--- drivers/infiniband/core/ucma.c| 4 drivers/infiniband/hw/mlx5/odp.c | 5 ++--- drivers/infiniband/hw/mlx5/qp.c | 29 - drivers/infiniband/sw/rdmavt/qp.c | 33 - drivers/infiniband/sw/rdmavt/rc.c | 4 +--- include/rdma/rdmavt_qp.h | 19 +++ 7 files changed, 57 insertions(+), 51 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 24 Jul 2020 14:47:39 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/fcef1046eb1b78c98105e9b68e48df6022c23a06 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Things remain quiet here, a few more of the usual fixes. The following changes since commit 11ba468877bb23f28956a35e896356252d63c983: Linux 5.8-rc5 (2020-07-12 16:34:50 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to a862192e9227ad46e0447fd0a03869ba1b30d16f: RDMA/mlx5: Prevent prefetch from racing with implicit destruction (2020-07-21 13:51:35 -0300) RDMA third 5.8 rc pull request One merge window regression, some corruption bugs in HNS and a few more syzkaller fixes. - Two long standing syzkaller races - Fix incorrect HW configuration in HNS - Restore accidentally dropped locking in IB CM - Fix ODP prefetch bug added in the big rework several versions ago Jason Gunthorpe (1): RDMA/mlx5: Prevent prefetch from racing with implicit destruction Leon Romanovsky (1): RDMA/core: Fix race in rdma_alloc_commit_uobject() Maor Gottlieb (2): RDMA/mlx5: Use xa_lock_irq when access to SRQ table RDMA/cm: Protect access to remote_sidr_table Weihang Li (1): RDMA/hns: Fix wrong assignment of lp_pktn_ini in QPC Xi Wang (1): RDMA/hns: Fix wrong PBL offset when VA is not aligned to PAGE_SIZE drivers/infiniband/core/cm.c | 2 ++ drivers/infiniband/core/rdma_core.c| 6 +++--- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 34 +++--- drivers/infiniband/hw/hns/hns_roce_mr.c| 2 +- drivers/infiniband/hw/mlx5/odp.c | 22 --- drivers/infiniband/hw/mlx5/srq_cmd.c | 4 ++-- 6 files changed, 49 insertions(+), 21 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 10 Jul 2020 14:58:06 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/aa0c9086b40c17a7ad94425b3b70dd1fdd7497bf Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Second rc pull request Fairly normal here, things are actually somewhat quiet when look at the next release. Thanks, Jason The following changes since commit 9ebcfadb0610322ac537dd7aa5d9cbc2b2894c68: Linux 5.8-rc3 (2020-06-28 15:00:24 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 0a03715068794e4b524f66ebbf412ab1f2933f3f: RDMA/mlx5: Set PD pointers for the error flow unwind (2020-07-08 20:15:59 -0300) RDMA second 5.8 rc pull request Small update, a few more merge window bugs and normal driver bug fixes: - Two merge window regressions in mlx5: a error path bug found by syzkaller and some lost code during a rework preventing ipoib from working in some configurations - Silence clang compilation warning in OPA related code - Fix a long standing race condition in ib_nl for ACM - Resolve when the HFI1 is shutdown Aya Levin (1): IB/mlx5: Fix 50G per lane indication Divya Indi (1): IB/sa: Resolv use-after-free in ib_nl_make_request() Kaike Wan (2): IB/hfi1: Do not destroy hfi1_wq when the device is shut down IB/hfi1: Do not destroy link_wq when the device is shut down Kamal Heib (1): RDMA/siw: Fix reporting vendor_part_id Leon Romanovsky (2): RDMA/mlx5: Fix legacy IPoIB QP initialization RDMA/mlx5: Set PD pointers for the error flow unwind Nathan Chancellor (1): IB/hfi1: Add explicit cast OPA_MTU_8192 to 'enum ib_mtu' drivers/infiniband/core/sa_query.c| 38 --- drivers/infiniband/hw/hfi1/init.c | 37 +- drivers/infiniband/hw/hfi1/qp.c | 7 +-- drivers/infiniband/hw/hfi1/tid_rdma.c | 5 - drivers/infiniband/hw/mlx5/main.c | 2 +- drivers/infiniband/hw/mlx5/qp.c | 7 ++- drivers/infiniband/sw/siw/siw_main.c | 3 ++- 7 files changed, 63 insertions(+), 36 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 25 Jun 2020 14:56:10 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/87d93e9a91c76bcb45112d863ef72aab41e01879 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request Collected fixes from the last four weeks, fixing several regressions in areas that had big changes this cycle. Thanks, Jason The following changes since commit b3a9e3b9622ae10064826dccb4f7a52bd88c7407: Linux 5.8-rc1 (2020-06-14 12:45:04 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 38fd98afeeb79d3b148db49f81f2ec6a37a4ee00: IB/hfi1: Add atomic triggered sleep/wakeup (2020-06-24 16:13:38 -0300) RDMA first 5.8 rc pull request Several regression fixes from work that landed in the merge window, particularly in the mlx5 driver: - Various static checker and warning fixes - General bug fixes in rvt, qedr, hns, mlx5 and hfi1 - Several regression fixes related to the ECE and QP changes in last cycle - Fixes for a few long standing crashers in CMA, uverbs ioctl, and xrc Aditya Pakki (1): RDMA/rvt: Fix potential memory leak caused by rvt_alloc_rq Colin Ian King (1): RDMA/mlx5: Remove duplicated assignment to resp.response_length Dennis Dalessandro (2): IB/hfi1: Restore kfree in dummy_netdev cleanup IB/hfi1: Fix module use count flaw due to leftover module put calls Fan Guo (1): RDMA/mad: Fix possible memory leak in ib_mad_post_receive_mads() Gal Pressman (1): RDMA/efa: Set maximum pkeys device attribute Kieran Bingham (1): RDMA/hfi1: Fix trivial mis-spelling of 'descriptor' Leon Romanovsky (6): RDMA/core: Annotate CMA unlock helper routine RDMA/mlx5: Add missed RST2INIT and INIT2INIT steps during ECE handshake RDMA/core: Check that type_attrs is not NULL prior access RDMA/mlx5: Don't access ib_qp fields in internal destroy QP path RDMA/mlx5: Remove ECE limitation from the RAW_PACKET QPs RDMA/mlx5: Protect from kernel crash if XRC_TGT doesn't have udata Maor Gottlieb (1): RDMA/mlx5: Fix remote gid value in query QP Mark Zhang (2): RDMA/cma: Protect bind_list and listen_list while finding matching cm id RDMA/counter: Query a counter before release Max Gurtovoy (1): RDMA/mlx5: Fix integrity enabled QP creation Michal Kalderon (1): RDMA/qedr: Fix KASAN: use-after-free in ucma_event_handler+0x532 Mike Marciniszyn (2): IB/hfi1: Correct -EBUSY handling in tx code IB/hfi1: Add atomic triggered sleep/wakeup Shay Drory (1): IB/mad: Fix use after free when destroying MAD agent Tom Seewald (2): RDMA/mlx5: Fix -Wformat warning in check_ucmd_data() RDMA/siw: Fix pointer-to-int-cast warning in siw_rx_pbl() Yangyang Li (2): RDMA/hns: Fix a calltrace when registering MR from userspace RDMA/hns: Fix an cmd queue issue when resetting drivers/infiniband/core/cm.c| 1 + drivers/infiniband/core/cma.c | 18 + drivers/infiniband/core/counters.c | 4 +- drivers/infiniband/core/mad.c | 3 +- drivers/infiniband/core/rdma_core.c | 36 ++ drivers/infiniband/hw/efa/efa_verbs.c | 1 + drivers/infiniband/hw/hfi1/debugfs.c| 19 + drivers/infiniband/hw/hfi1/iowait.h | 2 +- drivers/infiniband/hw/hfi1/ipoib.h | 6 ++ drivers/infiniband/hw/hfi1/ipoib_tx.c | 104 ++-- drivers/infiniband/hw/hfi1/netdev_rx.c | 2 +- drivers/infiniband/hw/hfi1/verbs_txreq.h| 2 +- drivers/infiniband/hw/hns/hns_roce_device.h | 7 +- drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 4 +- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 17 ++--- drivers/infiniband/hw/hns/hns_roce_mr.c | 5 +- drivers/infiniband/hw/mlx5/qp.c | 50 ++--- drivers/infiniband/hw/mlx5/qpc.c| 8 +++ drivers/infiniband/hw/qedr/qedr_iw_cm.c | 13 +++- drivers/infiniband/sw/rdmavt/qp.c | 6 +- drivers/infiniband/sw/siw/siw_qp_rx.c | 3 +- include/linux/mlx5/mlx5_ifc.h | 10 +-- 22 files changed, 197 insertions(+), 124 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 4 Jun 2020 16:51:31 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/242b23319809e05170b3cc0d44d3b4bd202bb073 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.8.
A more active cycle than most of the recent past. In particular the new RNBD
block driver was Ack'd by Jens and is flowing through RDMA due to it also
introducing a new ULP.
Also a heads up, Mellanox has been acquired by NVIDIA and you'll probably see
a new email address from me and others this next cycle.
Thanks,
Jason
The following changes since commit 6b646a7e4af69814dd1a3340fca0f02d4977420d:
net/mlx5: Add ability to read and write ECE options (2020-05-27 14:08:28
+0300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to fba97dc7fc76b2c9a909fa0b3786d30a9899f5cf:
RDMA/cm: Spurious WARNING triggered in cm_destroy_id() (2020-06-03 15:48:18
-0300)
RDMA 5.8 merge window pull request
A few large, long discussed works this time. The RNBD block driver has
been posted for nearly two years now, and the removal of FMR has been a
recurring discussion theme for a long time. The usual smattering of
features and bug fixes.
- Various small driver bugs fixes in rxe, mlx5, hfi1, and efa
- Continuing driver cleanups in bnxt_re, hns
- Big cleanup of mlx5 QP creation flows
- More consistent use of src port and flow label when LAG is used and a
mlx5 implementation
- Additional set of cleanups for IB CM
- 'RNBD' network block driver and target. This is a network block RDMA
device specific to ionos's cloud environment. It brings strong multipath
and resiliency capabilities.
- Accelerated IPoIB for HFI1
- QP/WQ/SRQ ioctl migration for uverbs, and support for multiple async fds
- Support for exchanging the new IBTA defiend ECE data during RDMA CM
exchanges
- Removal of the very old and insecure FMR interface from all ULPs and
drivers. FRWR should be preferred for at least a decade now.
Aharon Landau (2):
RDMA/mlx5: Verify that QP is created with RQ or SQ
RDMA/mlx5: Add init2init as a modify command
Bart Van Assche (4):
RDMA/srp: Make the channel count configurable per target
RDMA/srpt: Make debug output more detailed
RDMA/srpt: Reduce max_recv_sge to 1
RDMA/srpt: Increase max_send_sge
Chuck Lever (1):
RDMA/core: Move and rename trace_cm_id_create()
Colin Ian King (3):
RDMA/mlx5: Remove duplicated assignment to variable rcqe_sz
RDMA/hns: remove duplicate assignment to pointer raq
IB/hfi1: Fix spelling mistake "enought" -> "enough"
Dan Carpenter (5):
RDMA/rtrs: Fix some signedness bugs in error handling
RDMA/rtrs: Fix a couple off by one bugs in rtrs_srv_rdma_done()
block/rnbd: Fix an IS_ERR() vs NULL check in find_or_create_sess()
IB/hfi1: Fix hfi1_netdev_rx_init() error handling
RDMA/hns: Uninitialized variable in modify_qp_init_to_rtr()
Danil Kipnis (3):
rnbd/rtrs: Pass max segment size from blk user to the rdma library
RDMA/rnbd: Fix compilation error when CONFIG_MODULES is disabled
RDMA/rtrs: Get rid of the do_next_path while_next_path macros
Danit Goldberg (1):
RDMA/cm: Remove unused store to ret in cm_rej_handler
Daria Velikovsky (1):
RDMA/mlx5: Add support for drop action in DV steering
Devesh Sharma (4):
RDMA/bnxt_re: Reduce device page size detection code
RDMA/bnxt_re: Update missing hsi data structures
RDMA/bnxt_re: Simplify obtaining queue entry from hw ring
RDMA/bnxt_re: Remove dead code from rcfw
Gal Pressman (6):
RDMA/efa: Report create CQ error counter
RDMA/efa: Count mmap failures
RDMA/efa: Count admin commands errors
RDMA/efa: Fix setting of wrong bit in get/set_feature commands
RDMA/efa: Report host information to the device
RDMA/mlx5: Remove FMR leftovers
Gary Leshner (6):
IB/hfi1: Add functions to transmit datagram ipoib packets
IB/hfi1: Add the transmit side of a datagram ipoib RDMA netdev
IB/hfi1: Remove module parameter for KDETH qpns
IB/{rdmavt, hfi1}: Implement creation of accelerated UD QPs
IB/{hfi1, ipoib, rdma}: Broadcast ping sent packets which exceeded mtu
size
IB/ipoib: Add capability to switch between datagram and connected mode
Grzegorz Andrejczuk (6):
IB/hfi1: RSM rules for AIP
IB/hfi1: Rename num_vnic_contexts as num_netdev_contexts
IB/hfi1: Add interrupt handler functions for accelerated ipoib
IB/hfi1: Add rx functions for dummy netdev
IB/hfi1: Activate the dummy netdev
IB/hfi1: Add packet histogram trace event
Gustavo A. R. Silva (4):
IB/rdmavt: Replace zero-length array with flexible-array
RDMA/rtrs: client: Fix function return on success
RDMA/siw: Replace one-element array and use struct_size() helper
RDMA/core: Use sizeof_field() helper
Israel Rukshin (1):
RDMA/iser: Remove support
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 29 May 2020 11:15:56 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/6ff64d2537f5b445177c30a2fc7779a6f2107ed5 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Third rc pull request Nothing profound here, just a last set of long standing bug fixes for 5.7 The following changes since commit b9bbe6ed63b2b9f2c9ee5cbd0f2c946a2723f4ce: Linux 5.7-rc6 (2020-05-17 16:48:37 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 1acba6a817852d4aa7916d5c4f2c82f702ee9224: IB/ipoib: Fix double free of skb in case of multicast traffic in CM mode (2020-05-27 21:14:09 -0300) Third RDMA 5.7 rc pull request A few bug fixes: - Incorrect error unwind in qib and pvrdma - User triggerable NULL pointer crash in mlx5 with ODP prefetch - syzkaller RCU race in uverbs - Rare double free crash in ipoib Jason Gunthorpe (1): RDMA/core: Fix double destruction of uobject Kaike Wan (1): IB/qib: Call kobject_put() when kobject_init_and_add() fails Maor Gottlieb (1): RDMA/mlx5: Fix NULL pointer dereference in destroy_prefetch_work Qiushi Wu (1): RDMA/pvrdma: Fix missing pci disable in pvrdma_pci_probe() Valentine Fatiev (1): IB/ipoib: Fix double free of skb in case of multicast traffic in CM mode drivers/infiniband/core/rdma_core.c| 20 +--- drivers/infiniband/hw/mlx5/mr.c| 1 + drivers/infiniband/hw/qib/qib_sysfs.c | 9 + drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c | 2 +- drivers/infiniband/ulp/ipoib/ipoib.h | 4 drivers/infiniband/ulp/ipoib/ipoib_cm.c| 15 +-- drivers/infiniband/ulp/ipoib/ipoib_ib.c| 9 +++-- drivers/infiniband/ulp/ipoib/ipoib_main.c | 10 ++ include/rdma/uverbs_std_types.h| 2 +- 9 files changed, 47 insertions(+), 25 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 15 May 2020 16:13:54 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/d5dfe4f1b44ed532653c2335267ad9599c8a698e Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Second rc pull request One regression, some syzkaller crashers and other long standing bugs for RDMA. Thanks, Jason The following changes since commit 0e698dfa282211e414076f9dc7e83c1c288314fd: Linux 5.7-rc4 (2020-05-03 14:56:04 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to ccfdbaa5cf4601b9b71601893029dcc9245c002b: RDMA/uverbs: Move IB_EVENT_DEVICE_FATAL to destroy_uobj (2020-05-12 17:02:25 -0300) Second RDMA 5.7 rc pull request A few minor bug fixes for user visible defects, and one regression: - Various bugs from static checkers and syzkaller - Add missing error checking in mlx4 - Prevent RTNL lock recursion in i40iw - Fix segfault in cxgb4 in peer abort cases - Fix a regression added in 5.7 where the IB_EVENT_DEVICE_FATAL could be lost, and wasn't delivered to all the FDs Dan Carpenter (1): i40iw: Fix error handling in i40iw_manage_arp_cache() Denis V. Lunev (1): IB/i40iw: Remove bogus call to netdev_master_upper_dev_get() Jack Morgenstein (2): IB/mlx4: Test return value of calls to ib_get_cached_pkey IB/core: Fix potential NULL pointer dereference in pkey cache Jason Gunthorpe (2): RDMA/uverbs: Do not discard the IB_EVENT_DEVICE_FATAL event RDMA/uverbs: Move IB_EVENT_DEVICE_FATAL to destroy_uobj Maor Gottlieb (1): RDMA/core: Fix double put of resource Mike Marciniszyn (1): IB/hfi1: Fix another case where pq is left on waitlist Potnuri Bharat Teja (1): RDMA/iw_cxgb4: Fix incorrect function parameters Sudip Mukherjee (1): RDMA/rxe: Always return ERR_PTR from rxe_create_mmap_info() drivers/infiniband/core/cache.c| 7 +++-- drivers/infiniband/core/nldev.c| 3 +-- drivers/infiniband/core/rdma_core.c| 3 ++- drivers/infiniband/core/uverbs.h | 4 +++ drivers/infiniband/core/uverbs_main.c | 12 +++-- .../infiniband/core/uverbs_std_types_async_fd.c| 30 +- drivers/infiniband/hw/cxgb4/cm.c | 7 +++-- drivers/infiniband/hw/hfi1/user_sdma.c | 4 --- drivers/infiniband/hw/i40iw/i40iw_cm.c | 8 -- drivers/infiniband/hw/i40iw/i40iw_hw.c | 2 +- drivers/infiniband/hw/mlx4/qp.c| 14 +++--- drivers/infiniband/sw/rxe/rxe_mmap.c | 2 +- drivers/infiniband/sw/rxe/rxe_queue.c | 11 +--- 13 files changed, 68 insertions(+), 39 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Tue, 28 Apr 2020 16:59:14 -0300: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/edb98d162418e90d6d2c1cec42e09be0375cfd88 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request A few regressions and some syzkaller crashers for RDMA. Thanks, Jason The following changes since commit 8f3d9f354286745c751374f5f1fcafee6b3f3136: Linux 5.7-rc1 (2020-04-12 12:35:55 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to f0abc761bbb9418876cc4d1ebc473e4ea6352e42: RDMA/core: Fix race between destroy and release FD object (2020-04-24 15:40:41 -0300) First RDMA 5.7 rc pull request A couple of regressions were found in rc1, as well as another set of races and bugs : - A regression where RDMA_CM_EVENT_REJECTED was lost in some cases - Bad error handling in the CM, uverbs, rvt, siw and i40iw - Kernel stack memory leak to user space in mlx4 - Missing data in a uapi query for mlx5 - Three races found by syzkaller in the ib core code Aharon Landau (1): RDMA/mlx5: Set GRH fields in query QP on RoCE Alaa Hleihel (1): RDMA/mlx4: Initialize ib_spec on the stack Colin Ian King (1): i40iw: fix null pointer dereference on a null wqe pointer Dan Carpenter (1): RDMA/cm: Fix an error check in cm_alloc_id_priv() Jason Gunthorpe (2): RDMA/siw: Fix potential siw_mem refcnt leak in siw_fastreg_mr() RDMA/uverbs: Fix a race with disassociate and exit_mmap() Leon Romanovsky (4): RDMA/cm: Fix missing RDMA_CM_EVENT_REJECTED event after receiving REJ message RDMA/core: Prevent mixed use of FDs between shared ufiles RDMA/core: Fix overwriting of uobj in case of error RDMA/core: Fix race between destroy and release FD object Sudip Mukherjee (1): IB/rdmavt: Always return ERR_PTR from rvt_create_mmap_info() drivers/infiniband/core/cm.c | 26 ++ drivers/infiniband/core/rdma_core.c | 9 - drivers/infiniband/core/uverbs_main.c| 4 drivers/infiniband/hw/i40iw/i40iw_ctrl.c | 2 +- drivers/infiniband/hw/mlx4/main.c| 3 ++- drivers/infiniband/hw/mlx5/qp.c | 4 +++- drivers/infiniband/sw/rdmavt/cq.c| 4 ++-- drivers/infiniband/sw/rdmavt/mmap.c | 4 ++-- drivers/infiniband/sw/rdmavt/qp.c| 4 ++-- drivers/infiniband/sw/rdmavt/srq.c | 4 ++-- drivers/infiniband/sw/siw/siw_qp_tx.c| 15 +++ 11 files changed, 47 insertions(+), 32 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Wed, 9 Oct 2019 14:28:15 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/8a8c600de5dc1d9a7f4b83269fddc80ebd3dd045 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request The usual collection of driver bug fixes, and a few regressions from the merge window. Nothing particularly worrisome. The following changes since commit 54ecb8f7028c5eb3d740bb82b0f1d90f2df63c5c: Linux 5.4-rc1 (2019-09-30 10:35:40 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 0417791536ae1e28d7f0418f1d20048ec4d3c6cf: RDMA/mlx5: Add missing synchronize_srcu() for MW cases (2019-10-04 15:54:22 -0300) RDMA subsystem updates for 5.4-rc Various minor bug fixes posted in the last couple of weeks: - Various missed memory frees and error unwind bugs - Fix regressions in a few iwarp drivers from 5.4 patches - A few regressions added in past kernels - Squash a number of races in mlx5 ODP code Adit Ranadive (1): RDMA/vmw_pvrdma: Free SRQ only once Bart Van Assche (1): RDMA/iwcm: Fix a lock inversion issue Christophe JAILLET (1): RDMA/core: Fix an error handling path in 'res_get_common_doit()' Greg KH (1): RDMA/cxgb4: Do not dma memory off of the stack Jack Morgenstein (1): RDMA/cm: Fix memory leak in cm_add/remove_one Jason Gunthorpe (6): RDMA/mlx5: Do not allow rereg of a ODP MR RDMA/mlx5: Fix a race with mlx5_ib_update_xlt on an implicit MR RDMA/odp: Lift umem_mutex out of ib_umem_odp_unmap_dma_pages() RDMA/mlx5: Order num_pending_prefetch properly with synchronize_srcu RDMA/mlx5: Put live in the correct place for ODP MRs RDMA/mlx5: Add missing synchronize_srcu() for MW cases Krishnamraju Eraparaju (1): RDMA/siw: Fix serialization issue in write_space() Leon Romanovsky (1): RDMA/nldev: Reshuffle the code to avoid need to rebind QP in error path Michal Kalderon (1): RDMA/core: Fix use after free and refcnt leak on ndev in_device in iwarp_query_port Mohamad Heib (1): IB/core: Fix wrong iterating on ports Navid Emamdoost (1): RDMA/hfi1: Prevent memory leak in sdma_init Potnuri Bharat Teja (1): RDMA/iw_cxgb4: fix SRQ access from dump_qp() Shiraz, Saleem (1): RDMA/i40iw: Associate ibdev to netdev before IB device registration drivers/infiniband/core/cm.c | 3 ++ drivers/infiniband/core/cma.c | 3 +- drivers/infiniband/core/device.c | 9 ++-- drivers/infiniband/core/nldev.c | 12 ++--- drivers/infiniband/core/security.c| 2 +- drivers/infiniband/core/umem_odp.c| 6 ++- drivers/infiniband/hw/cxgb4/device.c | 7 ++- drivers/infiniband/hw/cxgb4/mem.c | 28 ++- drivers/infiniband/hw/cxgb4/qp.c | 10 +--- drivers/infiniband/hw/hfi1/sdma.c | 5 +- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 4 ++ drivers/infiniband/hw/mlx5/devx.c | 58 +++ drivers/infiniband/hw/mlx5/mlx5_ib.h | 3 +- drivers/infiniband/hw/mlx5/mr.c | 68 +++ drivers/infiniband/hw/mlx5/odp.c | 58 ++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c | 2 - drivers/infiniband/sw/siw/siw_qp.c| 15 -- drivers/net/ethernet/mellanox/mlx5/core/mr.c | 8 +--- 18 files changed, 154 insertions(+), 147 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 19 Sep 2019 16:34:46 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/018c6837f3e63b45163d55a1668d9f8e6fdecf6e Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.4
This cycle has proved to be quiet with a focus on bug fix and cleanup style
patches, and less on major new functionality. I am aware of no conflicts.
The following changes since commit f74c2bb98776e2de508f4d607cd519873065118e:
Linux 5.3-rc8 (2019-09-08 13:33:15 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 3eca7fc2d8d1275d9cf0c709f0937becbfcf6d96:
RDMA: Fix double-free in srq creation error flow (2019-09-16 14:37:38 -0300)
RDMA subsystem updates for 5.4
This cycle mainly saw lots of bug fixes and clean up code across the core
code and several drivers, few new functional changes were made.
- Many cleanup and bug fixes for hns
- Various small bug fixes and cleanups in hfi1, mlx5, usnic, qed,
bnxt_re, efa
- Share the query_port code between all the iWarp drivers
- General rework and cleanup of the ODP MR umem code to fit better with
the mmu notifier get/put scheme
- Support rdma netlink in non init_net name spaces
- mlx5 support for XRC devx and DC ODP
Andrey Konovalov (1):
RDMA/mlx4: Untag user pointers in mlx4_get_umem_mr
Arnd Bergmann (1):
RDMA/usnic: Avoid overly large buffers on stack
Bernard Metzler (2):
RDMA/siw: Fix page address mapping in TX path
RDMA/siw: Relax from kmap_atomic() use in TX path
Chuck Lever (1):
rdma: Enable ib_alloc_cq to spread work over a device's comp_vectors
Chuhong Yuan (1):
IB/usnic: Use dev_get_drvdata
Colin Ian King (3):
RDMA/hns: Fix comparison of unsigned long variable 'end' with less than
zero
RDMA/core: fix spelling mistake "Nelink" -> "Netlink"
RDMA/bnxt_re: Fix spelling mistake "missin_resp" -> "missing_resp"
Dan Carpenter (1):
RDMA/hns: Fix some white space check_mtu_validate()
Danit Goldberg (2):
IB/mlx5: Use the original address for the page during free_pages
IB/mlx5: Free mpi in mp_slave mode
Doug Ledford (3):
Merge branch 'wip/dl-for-rc' into wip/dl-for-next
Merge remote-tracking branch 'mlx5-next/mlx5-next' into wip/dl-for-next
Merge remote-tracking branch 'mlx5-next/mlx5-next' into for-next
Gal Pressman (6):
RDMA/efa: Expose device statistics
RDMA/core: Introduce ratelimited ibdev printk functions
RDMA/efa: Rate limit admin queue error prints
RDMA/efa: Remove umem check on dereg MR flow
RDMA/efa: Use existing FIELD_SIZEOF macro
RDMA/efa: Fix incorrect error print
Guoqing Jiang (1):
Documentation/infiniband: update name of some functions
Hariprasad Kelam (2):
RDMA/qib: Unneeded variable ret
RDMA/qedr: Remove Unneeded variable rc
Håkon Bugge (1):
RDMA/cma: Fix false error message
Ira Weiny (1):
IB/hfi1: Define variables as unsigned long to fix KASAN warning
Jack Morgenstein (1):
RDMA: Fix double-free in srq creation error flow
Jason Gunthorpe (16):
RDMA: Make most headers compile stand alone
RDMA/odp: Use the common interval tree library instead of generic
RDMA/odp: Iterate over the whole rbtree directly
RDMA/odp: Make it clearer when a umem is an implicit ODP umem
RMDA/odp: Consolidate umem_odp initialization
RDMA/odp: Make the three ways to create a umem_odp clear
RDMA/odp: Split creating a umem_odp from ib_umem_get
RDMA/odp: Provide ib_umem_odp_release() to undo the allocs
RDMA/odp: Check for overflow when computing the umem_odp end
RDMA/odp: Use kvcalloc for the dma_list and page_list
RDMA/mlx5: Use ib_umem_start instead of umem.address
RDMA/mlx5: Use odp instead of mr->umem in pagefault_mr
Merge branch 'odp_fixes' into rdma.git for-next
Merge branch 'mlx5-odp-dc' into rdma.git for-next
RDMA/odp: Add missing cast for 32 bit
Merge tag 'v5.3-rc8' into rdma.git for-next
Joe Perches (1):
mlx5: Fix formats with line continuation whitespace
Kaike Wan (3):
IB/hfi1: Do not update hcrc for a KDETH packet during fault injection
IB/hfi1: Add traces for TID RDMA READ
IB/{rdmavt, hfi1, qib}: Add a counter for credit waits
Kamal Heib (4):
RDMA: Introduce ib_port_phys_state enum
RDMA/cxgb3: Use ib_device_set_netdev()
RDMA/core: Add common iWARP query port
RDMA/{cxgb3, cxgb4, i40iw}: Remove common code
Lang Cheng (10):
RDMA/hns: Clean up unnecessary initial assignment
RDMA/hns: Update some comments style
RDMA/hns: Handling the error return value of hem function
RDMA/hns: Split bool statement and assign statement
RDMA/hns: Refactor irq request code
RDMA/hns: Remove unnecessary kzalloc
RDMA/hns: Remove unuseful member
RDMA/hns: Modify the data structure of hns_roce_av
RDMA/hns: Fix cast from
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Tue, 30 Jul 2019 12:15:01 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/32a024b9a9f3b40f84bc55a6dd35eaa770ea26a4 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request The usual collection of driver bug fixes, and fewer siw static checker warnings than I feared. Thanks, Jason The following changes since commit 5f9e832c137075045d15cd6899ab0505cfb2ca4b: Linus 5.3-rc1 (2019-07-21 14:05:38 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to b7165bd0d6cbb93732559be6ea8774653b204480: IB/mlx5: Fix RSS Toeplitz setup to be aligned with the HW specification (2019-07-25 11:45:48 -0300) 5.3 rc RDMA pull request A few regression and bug fixes for the patches merged in the last cycle: - hns fixes a subtle crash from the ib core SGL rework - hfi1 fixes various error handling, oops and protocol errors - bnxt_re fixes a regression where nvmeof doesn't work on some configurations - mlx5 fixes a serious 'use after free' bug in how MR caching is handled - Some edge case crashers in the new statistic core code - More siw static checker fixups Chuhong Yuan (1): IB/mlx5: Replace kfree with kvfree John Fleck (1): IB/hfi1: Check for error on call to alloc_rsm_map_table Kaike Wan (3): IB/hfi1: Unreserve a flushed OPFN request IB/hfi1: Field not zero-ed when allocating TID flow memory IB/hfi1: Drop all TID RDMA READ RESP packets after r_next_psn Mao Wenan (1): RDMA/siw: Remove set but not used variables 'rv' Moni Shoua (1): IB/mlx5: Prevent concurrent MR updates during invalidation Parav Pandit (2): IB/core: Fix querying total rdma stats IB/counters: Always initialize the port counter object Selvin Xavier (1): RDMA/bnxt_re: Honor vlan_id in GID entry comparison Wei Yongjun (1): RDMA/siw: Fix error return code in siw_init_module() Xi Wang (1): RDMA/hns: Fix sg offset non-zero issue Yishai Hadas (5): IB/mlx5: Fix unreg_umr to ignore the mkey state IB/mlx5: Use direct mkey destroy command upon UMR unreg failure IB/mlx5: Move MRs to a kernel PD when freeing them to the MR cache IB/mlx5: Fix clean_mr() to work in the expected order IB/mlx5: Fix RSS Toeplitz setup to be aligned with the HW specification drivers/infiniband/core/counters.c| 11 +--- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 7 +++-- drivers/infiniband/hw/bnxt_re/qplib_res.c | 13 +++--- drivers/infiniband/hw/bnxt_re/qplib_res.h | 2 +- drivers/infiniband/hw/bnxt_re/qplib_sp.c | 14 ++ drivers/infiniband/hw/bnxt_re/qplib_sp.h | 7 - drivers/infiniband/hw/hfi1/chip.c | 11 ++-- drivers/infiniband/hw/hfi1/rc.c | 2 -- drivers/infiniband/hw/hfi1/tid_rdma.c | 43 ++- drivers/infiniband/hw/hns/hns_roce_db.c | 15 ++- drivers/infiniband/hw/mlx5/mlx5_ib.h | 1 + drivers/infiniband/hw/mlx5/mr.c | 23 ++--- drivers/infiniband/hw/mlx5/odp.c | 7 ++--- drivers/infiniband/hw/mlx5/qp.c | 13 ++ drivers/infiniband/sw/siw/siw_cm.c| 3 +-- drivers/infiniband/sw/siw/siw_main.c | 1 + include/rdma/rdmavt_qp.h | 9 +++ 17 files changed, 89 insertions(+), 93 deletions(-)
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Mon, 15 Jul 2019 15:26:22 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/2a3c389a0fde49b241430df806a34276568cfb29 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.3
This has been a more exciting merge window than usual, the late merge of the
new siw driver, combined with the absence of the 0-Day testing service and the
July long weekend created a stream of build fixes for siw this week. I'm
hoping we got them all, but if there are more I will send them for rc1.
There is one non-textual conflict with an API change in netdev that will
require this patch in the merge commit to resolve:
diff --git a/drivers/infiniband/sw/siw/siw_cm.c
b/drivers/infiniband/sw/siw/siw_cm.c
index 43f7f12e5f7f81..a7cde98e73e8c8 100644
--- a/drivers/infiniband/sw/siw/siw_cm.c
+++ b/drivers/infiniband/sw/siw/siw_cm.c
@@ -1963,6 +1963,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
if (id->local_addr.ss_family == AF_INET) {
struct in_device *in_dev = in_dev_get(dev);
struct sockaddr_in s_laddr, *s_raddr;
+ const struct in_ifaddr *ifa;
memcpy(_laddr, >local_addr, sizeof(s_laddr));
s_raddr = (struct sockaddr_in *)>remote_addr;
@@ -1973,8 +1974,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
_raddr->sin_addr, ntohs(s_raddr->sin_port));
rtnl_lock();
- for_ifa(in_dev)
- {
+ in_dev_for_each_ifa_rtnl(ifa, in_dev) {
if (ipv4_is_zeronet(s_laddr.sin_addr.s_addr) ||
s_laddr.sin_addr.s_addr == ifa->ifa_address) {
s_laddr.sin_addr.s_addr = ifa->ifa_address;
@@ -1986,7 +1986,6 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
listeners++;
}
}
- endfor_ifa(in_dev);
rtnl_unlock();
in_dev_put(in_dev);
} else if (id->local_addr.ss_family == AF_INET6) {
The tag for-linus-merged with my merge resolution to your tree is also
available to pull.
The following changes since commit f8efee08dd9d41ab71010e9b16c9ead51753b7d6:
net/mlx5: Add rts2rts_qp_counters_set_id field in hca cap (2019-07-04
21:36:33 +0300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 0b043644c0ca601cb19943a81aa1f1455dbe9461:
RMDA/siw: Require a 64 bit arch (2019-07-12 12:12:06 -0300)
5.3 Merge window RDMA pull request
A smaller cycle this time. Notably we see another new driver, 'Soft
iWarp', and the deletion of an ancient unused driver for nes.
- Revise and simplify the signature offload RDMA MR APIs
- More progress on hoisting object allocation boiler plate code out of the
drivers
- Driver bug fixes and revisions for hns, hfi1, efa, cxgb4, qib, i40iw
- Tree wide cleanups: struct_size, put_user_page, xarray, rst doc conversion
- Removal of obsolete ib_ucm chardev and nes driver
- netlink based discovery of chardevs and autoloading of the modules
providing them
- Move more of the rdamvt/hfi1 uapi to include/uapi/rdma
- New driver 'siw' for software based iWarp running on top of netdev,
much like rxe's software RoCE.
- mlx5 feature to report events in their raw devx format to userspace
- Expose per-object counters through rdma tool
- Adaptive interrupt moderation for RDMA (DIM), sharing the DIM core
from netdev
Bart Van Assche (1):
RDMA/srp: Accept again source addresses that do not have a port number
Bernard Metzler (12):
rdma/siw: iWarp wire packet format
rdma/siw: main include file
rdma/siw: network and RDMA core interface
rdma/siw: connection management
rdma/siw: application interface
rdma/siw: application buffer management
rdma/siw: queue pair methods
rdma/siw: transmit path
rdma/siw: receive path
rdma/siw: completion queue methods
rdma/siw: addition to kernel build environment
RDMA/siw: Remove unnecessary kthread create/destroy printouts
Colin Ian King (4):
RDMA/hns: fix inverted logic of readl read and shift
RDMA/hns: fix potential integer overflow on left shift
RDMA/hns: fix spelling mistake "attatch" -> "attach"
RDMA/uverbs: remove redundant assignment to variable ret
Dag Moxnes (1):
RDMA/core: Fix race when resolving IP address
Dan Carpenter (2):
RDMA/uverbs: check for allocation failure in uapi_add_elm()
RDMA/hns: Fix an error code in hns_roce_set_user_sq_size()
Daniel Kranzdorf (1):
RDMA/efa: Entropy in admin commands id
Danit Goldberg (1):
IB/mlx5: Report correctly tag matching rendezvous capability
Dennis Dalessandro (2):
IB/hfi1: Remove extra brackets from an if
IB/hfi1: No need to use try_module_get for debugfs
Doug Ledford (3):
Merge
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 6 Jun 2019 20:14:24 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/6e38335dcc70f03faba26bf1260ee024d930afe1 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
Things are looking pretty quiet here in RDMA, not too many bug fixes rolling
in right now. Here is the first batch of proposed rc fixes.
Thanks,
Jason
The following changes since commit a188339ca5a396acc588e5851ed7e19f66b0ebd9:
Linux 5.2-rc1 (2019-05-19 15:47:09 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 4f240dfec6bcc852b124ea7c419fb590949fbd4c:
RDMA/efa: Remove MAYEXEC flag check from mmap flow (2019-05-29 13:13:03 -0300)
5.2 First rc pull request
The usual driver bug fixes and fixes for a couple of regressions introduced in
5.2:
- Fix a race on bootup with RDMA device renaming and srp. SRP also needs to
rename its internal sys files
- Fix a memory leak in hns
- Don't leak resources in efa on certain error unwinds
- Don't panic in certain error unwinds in ib_register_device
- Various small user visible bug fix patches for the hfi and efa drivers
- Fix the 32 bit compilation break
Gal Pressman (2):
RDMA/uverbs: Pass udata on uverbs error unwind
RDMA/efa: Remove MAYEXEC flag check from mmap flow
Jason Gunthorpe (1):
RDMA/core: Clear out the udata before error unwind
Kamal Heib (1):
RDMA/core: Fix panic when port_data isn't initialized
Kamenee Arumugam (1):
IB/hfi1: Validate page aligned for a given virtual address
Leon Romanovsky (2):
RDMA/srp: Rename SRP sysfs name after IB device rename trigger
RDMA/hns: Fix PD memory leak for internal allocation
Michal Kubecek (1):
mlx5: avoid 64-bit division
Mike Marciniszyn (3):
IB/rdmavt: Fix alloc_qpn() WARN_ON()
IB/hfi1: Insure freeze_work work_struct is canceled on shutdown
IB/{qib, hfi1, rdmavt}: Correct ibv_devinfo max_mr value
drivers/infiniband/core/device.c | 49 +++
drivers/infiniband/core/rdma_core.h | 2 ++
drivers/infiniband/core/uverbs_cmd.c | 30 +++-
drivers/infiniband/core/uverbs_std_types_cq.c | 2 +-
drivers/infiniband/core/uverbs_std_types_mr.c | 2 +-
drivers/infiniband/hw/efa/efa_verbs.c | 1 -
drivers/infiniband/hw/hfi1/chip.c | 1 +
drivers/infiniband/hw/hfi1/user_exp_rcv.c | 3 ++
drivers/infiniband/hw/hfi1/verbs.c| 2 --
drivers/infiniband/hw/hns/hns_roce_hw_v1.c| 1 +
drivers/infiniband/hw/mlx5/cmd.c | 9 +++--
drivers/infiniband/hw/mlx5/main.c | 2 +-
drivers/infiniband/hw/qib/qib_verbs.c | 2 --
drivers/infiniband/sw/rdmavt/mr.c | 2 ++
drivers/infiniband/sw/rdmavt/qp.c | 3 +-
drivers/infiniband/ulp/srp/ib_srp.c | 18 +-
include/rdma/ib_verbs.h | 1 +
17 files changed, 95 insertions(+), 35 deletions(-)
signature.asc
Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Wed, 15 May 2019 00:46:58 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/5ac94332248ee017964ba368cdda4ce647e3aba7 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, These are the 2nd proposed RDMA patches for 5.2. As requested the main reason to send this is to fix the gcc 9.1 warnings that many people may hit now that FC30 is out. Thanks, Jason The following changes since commit b79656ed44c6865e17bcd93472ec39488bcc4984: RDMA/ipoib: Allow user space differentiate between valid dev_port (2019-05-07 16:13:23 -0300) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to c191f93454bcc92810b9c8cdb895a452a57948c2: net/mlx5: Set completion EQs as shared resources (2019-05-14 10:22:09 -0300) 5.2 Merge Window second pull request This is being sent to get a fix for the gcc 9.1 build warnings, and I've also pulled in some bug fix patches that were posted in the last two weeks. - Avoid the gcc 9.1 warning about overflowing a union member - Fix the wrong callback type for a single response netlink to doit - Bug fixes from more usage of the mlx5 devx interface Jason Gunthorpe (1): RDMA: Directly cast the sockaddr union to sockaddr Parav Pandit (1): RDMA/core: Change system parameters callback from dumpit to doit Yishai Hadas (2): IB/mlx5: Verify DEVX general object type correctly net/mlx5: Set completion EQs as shared resources drivers/infiniband/core/addr.c | 16 drivers/infiniband/core/nldev.c | 27 +++ drivers/infiniband/hw/mlx5/devx.c| 13 ++--- drivers/infiniband/hw/ocrdma/ocrdma_ah.c | 5 ++--- drivers/infiniband/hw/ocrdma/ocrdma_hw.c | 5 ++--- drivers/net/ethernet/mellanox/mlx5/core/eq.c | 3 +++ include/linux/mlx5/mlx5_ifc.h| 2 +- include/uapi/rdma/rdma_netlink.h | 2 +- 8 files changed, 42 insertions(+), 31 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 9 May 2019 13:37:23 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/dce45af5c2e9e85f22578f2f8065f225f5d11764 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 5.2.
There is a small conflict with the mlx5-next (net-next) tree resolved with
this hunk:
diff --cc drivers/infiniband/hw/mlx5/main.c
index 687f99172037be,fae6a6a1fbea12..abac70ad5c7c46
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@@ -200,12 -172,18 +200,12 @@@ static int mlx5_netdev_event(struct not
switch (event) {
case NETDEV_REGISTER:
+ /* Should already be registered during the load */
+ if (ibdev->is_rep)
+ break;
write_lock(>netdev_lock);
- if (ndev->dev.parent == >pdev->dev)
- if (ibdev->rep) {
- struct mlx5_eswitch *esw = ibdev->mdev->priv.eswitch;
- struct net_device *rep_ndev;
-
- rep_ndev = mlx5_ib_get_rep_netdev(esw,
-ibdev->rep->vport);
- if (rep_ndev == ndev)
- roce->netdev = ndev;
- } else if (ndev->dev.parent == mdev->device) {
++ if (ndev->dev.parent == mdev->device)
roce->netdev = ndev;
- }
write_unlock(>netdev_lock);
break;
I also have some notifications from StephenR about minor conflicts with
patches in akpm's tree that you haven't merged yet. Here are his resolutions:
https://lore.kernel.org/lkml/20190509153805.6dfbb...@canb.auug.org.au/
https://lore.kernel.org/lkml/20190506204824.11a7b...@canb.auug.org.au/
https://lore.kernel.org/lkml/20190506203710.10b08...@canb.auug.org.au/
The tag for-linus-merged with my merge resolution for the mlx5-next
branch is also available to pull.
The following changes since commit 96780e4f46b2fc0fc5ae2b95957002e2c42b11d3:
net/mlx5: Introduce new TIR creation core API (2019-04-24 12:33:37 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to b79656ed44c6865e17bcd93472ec39488bcc4984:
RDMA/ipoib: Allow user space differentiate between valid dev_port (2019-05-07
16:13:23 -0300)
5.2 Merge Window pull request
This has been a smaller cycle than normal. One new driver was accepted,
which is unusual, and at least one more driver remains in review on the
list.
- Driver fixes for hns, hfi1, nes, rxe, i40iw, mlx5, cxgb4, vmw_pvrdma
- Many patches from MatthewW converting radix tree and IDR users to use
xarray
- Introduction of tracepoints to the MAD layer
- Build large SGLs at the start for DMA mapping and get the driver to
split them
- Generally clean SGL handling code throughout the subsystem
- Support for restricting RDMA devices to net namespaces for containers
- Progress to remove object allocation boilerplate code from drivers
- Change in how the mlx5 driver shows representor ports linked to VFs
- mlx5 uapi feature to access the on chip SW ICM memory
- Add a new driver for 'EFA'. This is HW that supports user space packet
processing through QPs in Amazon's cloud
Ariel Levkovich (5):
IB/mlx5: Expose TIR ICM address to user space
IB/mlx5: Support device memory type attribute
IB/mlx5: Warn on allocated MEMIC buffers during cleanup
IB/mlx5: Add steering SW ICM device memory type
IB/mlx5: Device resource control for privileged DEVX user
Artemy Kovalyov (3):
IB/mlx5: WQE dump jumps over first 16 bytes
IB/mlx5: Compare only index part of a memory window rkey
IB/core: Set qp->real_qp before it may be accessed
Bart Van Assche (6):
RDMA/uverbs: Add a __user annotation to a pointer
RDMA/uverbs: Annotate uverbs_request_next_ptr() return value as a __user
pointer
RDMA/uverbs: Allow the compiler to verify declaration and definition
consistency
IB/mlx5: Declare devx_async_cmd_event_fops static
IB/hfi1: Fix two format strings
IB/qib: Remove a set-but-not-used variable
Chengguang Xu (1):
infiniband/qib: Fix typo in comment
Colin Ian King (6):
RDMA/nes: remove redundant check on udata
IB/iser: remove uninitialized variable len
opa_vnic: fix check on record->event, incorrect operator used
RDMA/mlx5: Check for error return in flow_rule rather than err
RDMA/cxgb4: Fix null pointer dereference on alloc_skb failure
RDMA/cxgb4: Fix spelling mistake "immedate" -> "immediate"
Dennis Dalessandro (1):
IB/core, ipoib: Do not overreact to SM LID change event
Devesh Sharma (1):
RDMA/ocrdma: Remove use of idr use pci bdf instead
Enrico Weigelt, metux IT consult (1):
drivers: infiniband: Fix whitespace in kconfig
Erez Alfasi (1):
RDMA: Use __packed annotation instead of __attribute__ ((packed))
Gal Pressman
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sun, Apr 28, 2019 at 05:09:08PM -0700, Linus Torvalds wrote: > On Sun, Apr 28, 2019 at 4:49 PM Jason Gunthorpe wrote: > > > > It is for high availability - we have situations where the hardware > > can fault and needs some kind of destructive recovery. For instance a > > firmware reboot, or a VM migration. > > > > In these designs there may be multiple cards in the system and the > > userspace application could be using both. Just because one card > > crashed we can't send SIGBUS and kill the application, that breaks the > > HA design. > > Why can't this magical application that is *so* special that it is HA > and does magic mmap's of special rdma areas just catch the SIGBUS? > > Honestly, the whole "it's for HA" excuse stinks. It stinks because you > now silently just replace the mapping with *garbage*. That's not HA, > that's just random. This should only used in cases where user space only writes to the BAR page (it is an interrupt to the device essentially), so it doesn't care that the pages are now garbage, we just need to redirect the writes away from the bar. However I think someone later on added a readable counter BAR pages to one of the devices :( So even that ideal wasn't respected. > Wouldn't it be a lot better to just get the SIGBUS, and then that > magical application knows that "oh, it's gone", and it could - in its > SIGBUS handler - just do the dummy anonymous mmap() with /dev/zero it > if it wants to? This does sound more appealing, and probably should have been done instead. All this VMA stuff has been a big pain in the long run Thanks, Jason
Re: [GIT PULL] Please pull RDMA subsystem changes
On Mon, 2019-04-29 at 11:42 -0400, Doug Ledford wrote:
> On Mon, 2019-04-29 at 08:40 +, Jason Gunthorpe wrote:
> > On Mon, Apr 29, 2019 at 08:09:47AM +0200, Heiko Carstens wrote:
> > > On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> > > > Hi Linus,
> > > >
> > > > Third rc pull request
> > > >
> > > > Nothing particularly special here. There is a small merge conflict
> > > > with Adrea's mm_still_valid patches which is resolved as below:
> > > ...
> > > > Jason Gunthorpe (3):
> > > > RDMA/mlx5: Do not allow the user to write to the clock page
> > > > RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> > > > RDMA/ucontext: Fix regression with disassociate
> > >
> > > This doesn't compile. The patch below would fix it, but not sure if
> > > this is what is intended:
> > >
> > > drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
> > > drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault'
> > > has no member named 'vm_start'
> > >vmf->page = ZERO_PAGE(vmf->vm_start);
> > > ^~
> > > diff --git a/drivers/infiniband/core/uverbs_main.c
> > > b/drivers/infiniband/core/uverbs_main.c
> > > index 7843e89235c3..65fe89b3fa2d 100644
> > > +++ b/drivers/infiniband/core/uverbs_main.c
> > > @@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault
> > > *vmf)
> > >
> > > /* Read only pages can just use the system zero page. */
> > > if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
> > > - vmf->page = ZERO_PAGE(vmf->vm_start);
> > > + vmf->page = ZERO_PAGE(vmf->vma->vm_start);
> > > get_page(vmf->page);
> > > return 0;
> > > }
> > >
> >
> > Thanks Heiko, this looks right to me.
> >
> > I'm surprised to be seeing this at this point, these patches should
> > have been seen by 0 day for several days now, and they were in
> > linux-next already too..
> >
> > Doug, can you send this to Linus today?
>
> Yep.
>
Done.
--
Doug Ledford
GPG KeyID: B826A3330E572FDD
Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
signature.asc
Description: This is a digitally signed message part
Re: [GIT PULL] Please pull RDMA subsystem changes
On Mon, 2019-04-29 at 08:40 +, Jason Gunthorpe wrote:
> On Mon, Apr 29, 2019 at 08:09:47AM +0200, Heiko Carstens wrote:
> > On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> > > Hi Linus,
> > >
> > > Third rc pull request
> > >
> > > Nothing particularly special here. There is a small merge conflict
> > > with Adrea's mm_still_valid patches which is resolved as below:
> > ...
> > > Jason Gunthorpe (3):
> > > RDMA/mlx5: Do not allow the user to write to the clock page
> > > RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> > > RDMA/ucontext: Fix regression with disassociate
> >
> > This doesn't compile. The patch below would fix it, but not sure if
> > this is what is intended:
> >
> > drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
> > drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault' has
> > no member named 'vm_start'
> >vmf->page = ZERO_PAGE(vmf->vm_start);
> > ^~
> > diff --git a/drivers/infiniband/core/uverbs_main.c
> > b/drivers/infiniband/core/uverbs_main.c
> > index 7843e89235c3..65fe89b3fa2d 100644
> > +++ b/drivers/infiniband/core/uverbs_main.c
> > @@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault *vmf)
> >
> > /* Read only pages can just use the system zero page. */
> > if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
> > - vmf->page = ZERO_PAGE(vmf->vm_start);
> > + vmf->page = ZERO_PAGE(vmf->vma->vm_start);
> > get_page(vmf->page);
> > return 0;
> > }
> >
>
> Thanks Heiko, this looks right to me.
>
> I'm surprised to be seeing this at this point, these patches should
> have been seen by 0 day for several days now, and they were in
> linux-next already too..
>
> Doug, can you send this to Linus today?
Yep.
--
Doug Ledford
GPG KeyID: B826A3330E572FDD
Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
signature.asc
Description: This is a digitally signed message part
Re: [GIT PULL] Please pull RDMA subsystem changes
On Mon, Apr 29, 2019 at 11:00:40AM +0200, Michal Kubecek wrote:
> On Mon, Apr 29, 2019 at 08:40:37AM +, Jason Gunthorpe wrote:
> > On Mon, Apr 29, 2019 at 08:09:47AM +0200, Heiko Carstens wrote:
> > > On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> > > > Hi Linus,
> > > >
> > > > Third rc pull request
> > > >
> > > > Nothing particularly special here. There is a small merge conflict
> > > > with Adrea's mm_still_valid patches which is resolved as below:
> > > ...
> > > > Jason Gunthorpe (3):
> > > > RDMA/mlx5: Do not allow the user to write to the clock page
> > > > RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> > > > RDMA/ucontext: Fix regression with disassociate
> > >
> > > This doesn't compile. The patch below would fix it, but not sure if
> > > this is what is intended:
> > >
> > > drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
> > > drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault'
> > > has no member named 'vm_start'
> > >vmf->page = ZERO_PAGE(vmf->vm_start);
> > > ^~
> > > diff --git a/drivers/infiniband/core/uverbs_main.c
> > > b/drivers/infiniband/core/uverbs_main.c
> > > index 7843e89235c3..65fe89b3fa2d 100644
> > > +++ b/drivers/infiniband/core/uverbs_main.c
> > > @@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault
> > > *vmf)
> > >
> > > /* Read only pages can just use the system zero page. */
> > > if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
> > > - vmf->page = ZERO_PAGE(vmf->vm_start);
> > > + vmf->page = ZERO_PAGE(vmf->vma->vm_start);
> > > get_page(vmf->page);
> > > return 0;
> > > }
> > >
> >
> > Thanks Heiko, this looks right to me.
> >
> > I'm surprised to be seeing this at this point, these patches should
> > have been seen by 0 day for several days now, and they were in
> > linux-next already too..
>
> Most architectures have versions of ZERO_PAGE() which ignore the
> argument so that the code builds anyway. I'm not sure if 0-day also
> tests s390x builds (which is where I ran into this).
According to 0-build results for this patch, the answer is yes, it builds.
s390default_defconfig
And it compiles uverbs_main.c (CONFIG_INFINIBAND_USER_ACCESS)
kernel git:(rdma-next) grep INFIN arch/s390/configs/debug_defconfig
CONFIG_INFINIBAND=m
CONFIG_INFINIBAND_USER_ACCESS=m
CONFIG_MLX4_INFINIBAND=m
CONFIG_MLX5_INFINIBAND=m
Thanks
>
> Michal Kubecek
Re: [GIT PULL] Please pull RDMA subsystem changes
On Mon, Apr 29, 2019 at 08:40:37AM +, Jason Gunthorpe wrote:
> On Mon, Apr 29, 2019 at 08:09:47AM +0200, Heiko Carstens wrote:
> > On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> > > Hi Linus,
> > >
> > > Third rc pull request
> > >
> > > Nothing particularly special here. There is a small merge conflict
> > > with Adrea's mm_still_valid patches which is resolved as below:
> > ...
> > > Jason Gunthorpe (3):
> > > RDMA/mlx5: Do not allow the user to write to the clock page
> > > RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> > > RDMA/ucontext: Fix regression with disassociate
> >
> > This doesn't compile. The patch below would fix it, but not sure if
> > this is what is intended:
> >
> > drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
> > drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault' has
> > no member named 'vm_start'
> >vmf->page = ZERO_PAGE(vmf->vm_start);
> > ^~
> > diff --git a/drivers/infiniband/core/uverbs_main.c
> > b/drivers/infiniband/core/uverbs_main.c
> > index 7843e89235c3..65fe89b3fa2d 100644
> > +++ b/drivers/infiniband/core/uverbs_main.c
> > @@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault *vmf)
> >
> > /* Read only pages can just use the system zero page. */
> > if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
> > - vmf->page = ZERO_PAGE(vmf->vm_start);
> > + vmf->page = ZERO_PAGE(vmf->vma->vm_start);
> > get_page(vmf->page);
> > return 0;
> > }
> >
>
> Thanks Heiko, this looks right to me.
>
> I'm surprised to be seeing this at this point, these patches should
> have been seen by 0 day for several days now, and they were in
> linux-next already too..
Most architectures have versions of ZERO_PAGE() which ignore the
argument so that the code builds anyway. I'm not sure if 0-day also
tests s390x builds (which is where I ran into this).
Michal Kubecek
Re: [GIT PULL] Please pull RDMA subsystem changes
On Mon, Apr 29, 2019 at 08:09:47AM +0200, Heiko Carstens wrote:
> On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> > Hi Linus,
> >
> > Third rc pull request
> >
> > Nothing particularly special here. There is a small merge conflict
> > with Adrea's mm_still_valid patches which is resolved as below:
> ...
> > Jason Gunthorpe (3):
> > RDMA/mlx5: Do not allow the user to write to the clock page
> > RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> > RDMA/ucontext: Fix regression with disassociate
>
> This doesn't compile. The patch below would fix it, but not sure if
> this is what is intended:
>
> drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
> drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault' has no
> member named 'vm_start'
>vmf->page = ZERO_PAGE(vmf->vm_start);
> ^~
> diff --git a/drivers/infiniband/core/uverbs_main.c
> b/drivers/infiniband/core/uverbs_main.c
> index 7843e89235c3..65fe89b3fa2d 100644
> +++ b/drivers/infiniband/core/uverbs_main.c
> @@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault *vmf)
>
> /* Read only pages can just use the system zero page. */
> if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
> - vmf->page = ZERO_PAGE(vmf->vm_start);
> + vmf->page = ZERO_PAGE(vmf->vma->vm_start);
> get_page(vmf->page);
> return 0;
> }
>
Thanks Heiko, this looks right to me.
I'm surprised to be seeing this at this point, these patches should
have been seen by 0 day for several days now, and they were in
linux-next already too..
Doug, can you send this to Linus today?
Thanks,
Jason
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sun, Apr 28, 2019 at 11:52:12AM +, Jason Gunthorpe wrote:
> Hi Linus,
>
> Third rc pull request
>
> Nothing particularly special here. There is a small merge conflict
> with Adrea's mm_still_valid patches which is resolved as below:
...
> Jason Gunthorpe (3):
> RDMA/mlx5: Do not allow the user to write to the clock page
> RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
> RDMA/ucontext: Fix regression with disassociate
This doesn't compile. The patch below would fix it, but not sure if
this is what is intended:
drivers/infiniband/core/uverbs_main.c: In function 'rdma_umap_fault':
drivers/infiniband/core/uverbs_main.c:898:28: error: 'struct vm_fault' has no
member named 'vm_start'
vmf->page = ZERO_PAGE(vmf->vm_start);
^~
diff --git a/drivers/infiniband/core/uverbs_main.c
b/drivers/infiniband/core/uverbs_main.c
index 7843e89235c3..65fe89b3fa2d 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -895,7 +895,7 @@ static vm_fault_t rdma_umap_fault(struct vm_fault *vmf)
/* Read only pages can just use the system zero page. */
if (!(vmf->vma->vm_flags & (VM_WRITE | VM_MAYWRITE))) {
- vmf->page = ZERO_PAGE(vmf->vm_start);
+ vmf->page = ZERO_PAGE(vmf->vma->vm_start);
get_page(vmf->page);
return 0;
}
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sun, Apr 28, 2019 at 4:49 PM Jason Gunthorpe wrote: > > It is for high availability - we have situations where the hardware > can fault and needs some kind of destructive recovery. For instance a > firmware reboot, or a VM migration. > > In these designs there may be multiple cards in the system and the > userspace application could be using both. Just because one card > crashed we can't send SIGBUS and kill the application, that breaks the > HA design. Why can't this magical application that is *so* special that it is HA and does magic mmap's of special rdma areas just catch the SIGBUS? Honestly, the whole "it's for HA" excuse stinks. It stinks because you now silently just replace the mapping with *garbage*. That's not HA, that's just random. Wouldn't it be a lot better to just get the SIGBUS, and then that magical application knows that "oh, it's gone", and it could - in its SIGBUS handler - just do the dummy anonymous mmap() with /dev/zero it if it wants to? Whatever. It really sounds like this is yet another horrible back for bad interfaces for all these "super-special high-end enterprise people". I hope that some day enterprise people will wake up and realize that "enterprise" seems to be often a code name for "lots of random hacks". Linus
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sun, Apr 28, 2019 at 09:59:56AM -0700, Linus Torvalds wrote: > On Sun, Apr 28, 2019 at 4:52 AM Jason Gunthorpe wrote: > > > > Nothing particularly special here. There is a small merge conflict > > with Adrea's mm_still_valid patches which is resolved as below: > > I still don't understand *why* you play the crazy VM games to begin with. > > What's wrong with just returning SIGBUS? Why does that > rdma_umap_fault() not just look like this one-liner: > > return VM_FAULT_SIGBUS; > > without the crazy parts? Nobody ever explained why you'd want to have > that silly "let's turn it into a bogus anonymous mapping". There was a big thread where I went over the use case with Andrea, but I guess that was private.. It is for high availability - we have situations where the hardware can fault and needs some kind of destructive recovery. For instance a firmware reboot, or a VM migration. In these designs there may be multiple cards in the system and the userspace application could be using both. Just because one card crashed we can't send SIGBUS and kill the application, that breaks the HA design. So.. the kernel makes the BAR VMA into a 'dummy' and sends an async notification to the application. The use of the BAR memory by userspace is all 'write only' so it doesn't really care. When it sees the async notification it safely cleans up the userspace side of things. An more modern VM example of where this gets used is on VM systems using SRIO-V pass through of a raw RDMA device. When it is time to migrate the VM then the hypervisor causes the SRIO-V instance to fault and be removed from the guest kernel, then migrates and attaches a new RDMA SRIO-V instance. The user space is expected to see the failure, maintain state, then recover onto the new device. The only alternative that has come up would be to delay the kernel side until the application cleans up and deletes the VMA, but people generally don't like this as it degrades the recovery time and has the usual problems with blocking the kernel on userspace. When this was created I'm not sure people explored more creative ideas like trying to handle/ignore the SIGBUS in userspace - unfortunately it has been so long now that we are probably stuck doing this as part of the UAPI. I've been trying to make it less crufty over the last year based on remarks from yourself and Andrea, but I'm still stuck with this basic requirement that the VMA shouldn't fault or touch the BAR after the hardware is released by the kernel. Thanks, Jason
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Sun, 28 Apr 2019 11:52:12 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/14f974d7f0f1f93d8c35f496ae774ba0f1b3389a Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
Re: [GIT PULL] Please pull RDMA subsystem changes
On Sun, Apr 28, 2019 at 4:52 AM Jason Gunthorpe wrote: > > Nothing particularly special here. There is a small merge conflict > with Adrea's mm_still_valid patches which is resolved as below: I still don't understand *why* you play the crazy VM games to begin with. What's wrong with just returning SIGBUS? Why does that rdma_umap_fault() not just look like this one-liner: return VM_FAULT_SIGBUS; without the crazy parts? Nobody ever explained why you'd want to have that silly "let's turn it into a bogus anonymous mapping". I've pulled this, but why do the rdma and SCSI people always do these strange and pointless things? Linus
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
Third rc pull request
Nothing particularly special here. There is a small merge conflict
with Adrea's mm_still_valid patches which is resolved as below:
diff --cc drivers/infiniband/core/uverbs_main.c
index db20b6e0f253c9,f2e7ffe6fc5466..7843e89235c34b
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@@ -1039,7 -992,9 +1039,9 @@@ void uverbs_user_mmap_disassociate(stru
* at a time to get the lock ordering right. Typically there
* will only be one mm, so no big deal.
*/
- down_write(>mmap_sem);
+ down_read(>mmap_sem);
+ if (!mmget_still_valid(mm))
+ goto skip_mm;
mutex_lock(>umap_lock);
list_for_each_entry_safe (priv, next_priv, >umaps,
list) {
@@@ -1051,9 -1006,11 +1053,10 @@@
zap_vma_ptes(vma, vma->vm_start,
vma->vm_end - vma->vm_start);
- vma->vm_flags &= ~(VM_SHARED | VM_MAYSHARE);
}
mutex_unlock(>umap_lock);
+ skip_mm:
- up_write(>mmap_sem);
+ up_read(>mmap_sem);
mmput(mm);
}
}
The tag for-linus-merged with my merge resolution to your tree is also
available to pull.
Thanks,
Jason
The following changes since commit dc4060a5dc2557e6b5aa813bf5b73677299d62d2:
Linux 5.1-rc5 (2019-04-14 15:17:41 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 2557fabd6e29f349bfa0ac13f38ac98aa5eafc74:
RDMA/hns: Bugfix for mapping user db (2019-04-25 10:40:04 -0300)
5.1 Third RC pull request
One core bug fix and a few driver ones
- FRWR memory registration for hfi1/qib didn't work with with some iovas
causing a NFSoRDMA failure regression due to a fix in the NFS side
- A command flow error in mlx5 allowed user space to send a corrupt
command (and also smash the kernel stack we've since learned)
- Fix a regression and some bugs with device hot unplug that was
discovered while reviewing Andrea's patches
- hns has a failure if the user asks for certain QP configurations
Guy Levi (1):
IB/mlx5: Fix scatter to CQE in DCT QP creation
Jason Gunthorpe (3):
RDMA/mlx5: Do not allow the user to write to the clock page
RDMA/mlx5: Use rdma_user_map_io for mapping BAR pages
RDMA/ucontext: Fix regression with disassociate
Josh Collier (1):
IB/rdmavt: Fix frwr memory registration
Lijun Ou (1):
RDMA/hns: Bugfix for mapping user db
drivers/infiniband/core/uverbs.h| 1 +
drivers/infiniband/core/uverbs_main.c | 52 +++--
drivers/infiniband/hw/hns/hns_roce_qp.c | 2 +-
drivers/infiniband/hw/mlx5/main.c | 12
drivers/infiniband/hw/mlx5/qp.c | 11 ---
drivers/infiniband/sw/rdmavt/mr.c | 17 ++-
include/uapi/rdma/mlx5-abi.h| 1 +
7 files changed, 76 insertions(+), 20 deletions(-)
(diffstat from tag for-linus-merged)
signature.asc
Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Wed, 10 Apr 2019 18:46:23 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/582549e3fbe137eb6ce9be591aca25ca36b4 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Second rc pull request The following changes since commit 8c2ffd9174779014c3fe1f96d9dc3641d9175f00: Linux 5.1-rc2 (2019-03-24 14:02:26 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to d737b25b1ae0540ba13cbd45ebb9b58a1d6d7f0d: IB/hfi1: Do not flush send queue in the TID RDMA second leg (2019-04-10 15:09:30 -0300) 5.1 Second RC pull request Several driver bug fixes posted in the last several weeks - Several bug fixes for the hfi1 driver 'TID RDMA' functionality merged into 5.1. Since TID RDMA is on by default these all seem to be regressions. - Wrong software permission checks on memory in mlx5 - Memory leak in vmw_pvrdma during driver remove - Several bug fixes for hns driver features merged into 5.1 Kaike Wan (5): IB/hfi1: Failed to drain send queue when QP is put into error state IB/hfi1: Clear the IOWAIT pending bits when QP is put into error state IB/hfi1: Eliminate opcode tests on mr deref IB/hfi1: Fix the allocation of RSM table IB/hfi1: Do not flush send queue in the TID RDMA second leg Kamal Heib (1): RDMA/vmw_pvrdma: Fix memory leak on pvrdma_pci_remove Lijun Ou (1): RDMA/hns: Fix bug that caused srq creation to fail Moni Shoua (1): IB/mlx5: Reset access mask when looping inside page fault handler Yangyang Li (1): RDMA/hns: Bugfix for SCC hem free drivers/infiniband/hw/hfi1/chip.c | 26 +++-- drivers/infiniband/hw/hfi1/qp.c| 4 +++- drivers/infiniband/hw/hfi1/rc.c| 4 ++-- drivers/infiniband/hw/hfi1/tid_rdma.c | 31 +++--- drivers/infiniband/hw/hns/hns_roce_hem.c | 6 +++-- drivers/infiniband/hw/hns/hns_roce_mr.c| 4 ++-- drivers/infiniband/hw/hns/hns_roce_qp.c| 3 --- drivers/infiniband/hw/mlx5/odp.c | 3 ++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c | 2 ++ 9 files changed, 42 insertions(+), 41 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Mon, 18 Mar 2019 01:04:12 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/01c8d80383d9076ab4fbebc3e60ae9abc70f70d5 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request - nothing too special, just some driver bug fixes that have been sitting during the merge window. Thanks, Jason The following changes since commit 9e98c678c2d6ae3a17cb2de55d17f69dddaa231b: Linux 5.1-rc1 (2019-03-17 14:22:26 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to ec4fe4bcc584b55e24e8d1768f5510a62c0fd619: i40iw: Avoid panic when handling the inetdev event (2019-03-17 21:40:40 -0300) 5.1 First RC pull request Several driver bug fixes post in the last three weeks - First part of a race condition fix in mlx4 with CATAS errors - Bad interaction with FW causing resource leaks in the mlx5 DCT flow - Bad reporting of link speed/width in new mlx5 devices - Userspace triggable OOPs in i40iw Aya Levin (1): IB/mlx5: Fix mapping of link-mode to IB width and speed Feng Tang (1): i40iw: Avoid panic when handling the inetdev event Jack Morgenstein (1): IB/mlx4: Fix race condition between catas error reset and aliasguid flows Yishai Hadas (2): net/mlx5: Fix DCT creation bad flow IB/mlx5: Use mlx5 core to create/destroy a DEVX DCT drivers/infiniband/hw/i40iw/i40iw_utils.c| 12 - drivers/infiniband/hw/mlx4/alias_GUID.c | 2 +- drivers/infiniband/hw/mlx5/devx.c| 34 + drivers/infiniband/hw/mlx5/main.c| 7 +++ drivers/infiniband/hw/mlx5/qp.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/qp.c | 72 +++- include/linux/mlx5/qp.h | 3 +- 7 files changed, 89 insertions(+), 45 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 7 Mar 2019 01:34:16 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/a50243b1ddcdd766d0d17fbfeeb1a22e62fdc461 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, These are the proposed RDMA patches for 5.1. There is a small conflict with the net tree in drivers/infiniband/hw/mlx4/Kconfig that is resolved by deleting both lines. There may also be conflicts with Matt's xarray tree that changes some of the xarray APIs. RDMA has gained some new users of those APIs this cycle. Looking at the patchworks I'm thinking we might have a second merge window pull request with some bug fixes and other rcish things that were sent a little too late for this. The next merge window may also see up to three new RDMA drivers, which is quite unusual. Thanks, Jason The tag for-linus-merged with my merge resolution to your tree is also available to pull. The following changes since commit 08e8676f1607925adf36e399f0faa8ce3b10bb86: IB/mlx5: Add support for 50Gbps per lane link modes (2019-02-14 12:14:42 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to fca22e7e595f1799cfbfdfa13e16d48ece0d136c: net/mlx5: ODP support for XRC transport is not enabled by default in FW (2019-03-06 15:53:09 -0400) 5.1 Merge Window Pull Request This has been a slightly more active cycle than normal with ongoing core changes and quite a lot of collected driver updates. - Various driver fixes for bnxt_re, cxgb4, hns, mlx5, pvrdma, rxe - A new data transfer mode for HFI1 giving higher performance - Significant functional and bug fix update to the mlx5 On-Demand-Paging MR feature - A chip hang reset recovery system for hns - Change mm->pinned_vm to an atomic64 - Update bnxt_re to support a new 57500 chip - A sane netlink 'rdma link add' method for creating rxe devices and fixing the various unregistration race conditions in rxe's unregister flow - Allow lookup up objects by an ID over netlink - Various reworking of the core to driver interface: * Drivers should not assume umem SGLs are in PAGE_SIZE chunks * ucontext is accessed via udata not other means * Start to make the core code responsible for object memory allocation * Drivers should convert struct device to struct ib_device via a helper * Drivers have more tools to avoid use after unregister problems Adit Ranadive (1): RDMA/vmw_pvrdma: Support upto 64-bit PFNs Bart Van Assche (5): IB/core: Declare local functions 'static' IB/mlx5: Declare local functions 'static' RDMA/srp: Increase max_segment_size IB/core: Remove ib_sg_dma_address() and ib_sg_dma_len() IB/mlx5: Do not use hw_access_flags for be and CPU data Colin Ian King (2): RDMA/bnxt_re: fix or'ing of data into an uninitialized struct member RDMA: Fix allocation failure on pointer pd Dan Carpenter (2): RDMA/bnxt_re: fix a size calculation RDMA/core: Fix a WARN() message Daniel Jurgens (4): IB/core: Unregister notifier before freeing MAD security IB/core: Fix potential memory leak while creating MAD agents IB/core: Eliminate a hole in MAD agent struct IB/core: Don't register each MAD agent for LSM notifier Danit Goldberg (1): IB/cma: Define option to set ack timeout and pack tos_set Davidlohr Bueso (8): mm: make mm->pinned_vm an atomic64 counter drivers/mic/scif: do not use mmap_sem drivers/IB,qib: optimize mmap_sem usage drivers/IB,hfi1: do not se mmap_sem drivers/IB,usnic: reduce scope of mmap_sem drivers/IB,core: reduce scope of mmap_sem Documentation/infiniband: update from locked to pinned_vm drivers/IB,qib: Fix pinned/locked limit check in qib_get_user_pages() Devesh Sharma (9): RDMA/bnxt_re: Add chip context to identify 57500 series RDMA/bnxt_re: Add 64bit doorbells for 57500 series RDMA/bnxt_re: Skip backing store allocation for 57500 series RDMA/bnxt_re: Enable GSI QP support for 57500 series RDMA/bnxt_re: Add extended psn structure for 57500 adapters RDMA/bnxt_re: Update kernel user abi to pass chip context RDMA/bnxt_en: Enable RDMA driver support for 57500 chip bnxt_re: fix the regression due to changes in alloc_pbl bnxt_re: Clean cq for kernel consumers only Doug Ledford (6): Merge branch 'opfn' into hfi1-tid Merge branch 'tid-read' into hfi1-tid Merge branch 'tid-write' into hfi1-tid Merge branch 'hfi1-tid' into wip/dl-for-next Merge branch 'wip/dl-for-next' into for-next Merge branch 'for-next' of git://git.kernel.org/.../rdma/rdma into for-next Erez Alfasi (1): IB/ipoib: Use __func__ instead of function's name Gal Pressman (6): IB/usnic: Fix out of bounds index check in query pkey RDMA/ocrdma: Fix out of bounds index check in query pkey RDMA/qedr: Fix out of bounds index check in query pkey RDMA: Add indication for in kernel API
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Thu, 21 Feb 2019 23:07:20 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/168bd29830e8ebbffcd70d2af50249dca088e1a8 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Small set of three regression fixing patches, things are looking pretty good here. Thanks The following changes since commit 8834f5600cf3c8db365e18a3d5cac2c2780c81e5: Linux 5.0-rc5 (2019-02-03 13:48:04 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to f09ef134a7ca3f0d2ce485a757f5b79809ebb803: iw_cxgb4: cq/qp mask depends on bar2 pages in a host page (2019-02-15 09:39:39 -0700) Third 5.0 rc pull request Three smallish patches fixing regressions in v5.0: - Fix cxgb4 to work again with non-4k page sizes - NULL pointer oops in SRP during sg_reset Bart Van Assche (1): RDMA/srp: Rework SCSI device reset handling Raju Rangoju (2): cxgb4: Export sge_host_page_size to ulds iw_cxgb4: cq/qp mask depends on bar2 pages in a host page drivers/infiniband/hw/cxgb4/device.c | 15 +-- drivers/infiniband/ulp/srp/ib_srp.c| 10 -- drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c | 1 + drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h | 1 + 4 files changed, 15 insertions(+), 12 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 1 Feb 2019 17:41:11 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/5eeb63359b1ec4914d9898e24aa357ff930e6ee1 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
Second rc pull request
Still expecting some more bug fixes for new stuff merged in 5.0, but
at this rate that stuff is looking likely to go into 5.1.
The following changes since commit 49a57857aeea06ca831043acbb0fa5e0f50602fd:
Linux 5.0-rc3 (2019-01-21 13:14:44 +1300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to 7b21b69ab203136fdc153c7707fa6c409e523c2e:
IB/uverbs: Fix OOPs in uverbs_user_mmap_disassociate (2019-01-29 13:57:22
-0700)
Second 5.0 rc pull request
Still not much going on, the usual set of OOP's and driver updates this time.
- Fix two uapi breakage regressions in mlx5 drivers
- Various oops fixes in hfi1, mlx4, umem, uverbs, and ipoib
- A protocol bug fix for hfi1 preventing it from implementing the verbs
API properly, and a compatability fix for EXEC STACK user programs
- Fix missed refcounting in the 'advise_mr' patches merged this
cycle.
- Fix wrong use of the uABI in the hns SRQ patches merged this cycle.
Artemy Kovalyov (1):
RDMA/umem: Add missing initialization of owning_mm
Brian Welty (1):
IB/{hfi1, qib}: Fix WC.byte_len calculation for UD_SEND_WITH_IMM
Feras Daoud (1):
IB/ipoib: Fix for use-after-free in ipoib_cm_tx_start
Jack Morgenstein (1):
IB/mlx4: Fix using wrong function to destroy sqp AHs under SRIOV
Jason Gunthorpe (2):
RDMA/device: Expose ib_device_try_get(()
IB/mlx5: Fix how advise_mr() launches async work
Lijun Ou (1):
RDMA/hns: Update the kernel header file of hns
Mark Bloch (2):
RDMA/mlx5: Fix check for supported user flags when creating a QP
RDMA/mlx5: Fix flow creation on representors
Michael J. Ruhl (1):
IB/hfi1: Remove overly conservative VM_EXEC flag check
Mike Marciniszyn (1):
IB/hfi1: Add limit test for RC/UC send via loopback
Yishai Hadas (3):
IB/uverbs: Fix OOPs upon device disassociation
IB/uverbs: Fix ioctl query port to consider device disassociation
IB/uverbs: Fix OOPs in uverbs_user_mmap_disassociate
drivers/infiniband/core/core_priv.h | 1 -
drivers/infiniband/core/device.c | 13 +---
drivers/infiniband/core/umem_odp.c| 3 +++
drivers/infiniband/core/uverbs_main.c | 25 +++
drivers/infiniband/core/uverbs_std_types_device.c | 8 +++-
drivers/infiniband/hw/hfi1/file_ops.c | 2 +-
drivers/infiniband/hw/hfi1/ud.c | 1 -
drivers/infiniband/hw/hns/hns_roce_srq.c | 10 +++--
drivers/infiniband/hw/mlx4/mad.c | 6 +++---
drivers/infiniband/hw/mlx5/flow.c | 3 +--
drivers/infiniband/hw/mlx5/odp.c | 10 -
drivers/infiniband/hw/mlx5/qp.c | 16 ---
drivers/infiniband/hw/qib/qib_ud.c| 1 -
drivers/infiniband/sw/rdmavt/qp.c | 7 ++-
drivers/infiniband/ulp/ipoib/ipoib.h | 1 -
drivers/infiniband/ulp/ipoib/ipoib_cm.c | 3 +--
include/rdma/ib_verbs.h | 24 --
include/uapi/rdma/hns-abi.h | 5 +
18 files changed, 97 insertions(+), 42 deletions(-)
signature.asc
Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 18 Jan 2019 03:56:44 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/d7393226d15add056285c8fc86723d54d7e0c77d Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Just a small batch of patches for -rc today. Thanks The following changes since commit bfeffd155283772bbe78c6a05dec7c0128ee500c: Linux 5.0-rc1 (2019-01-06 17:08:20 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to d6f4a21f309dfe10a5693ad236358dd6fcc46f7a: RDMA/uverbs: Mark ioctl responses with UVERBS_ATTR_F_VALID_OUTPUT (2019-01-14 14:02:22 -0700) First 5.0 rc pull request Not much so far, but I'm feeling like the 2nd PR -rc will be larger than this. We have the usual batch of bugs and two fixes to code merged this cycle. - Restore valgrind support for the ioctl verbs interface merged this window, and fix a missed error code on an error path from that conversion - A user reported crash on obsolete mthca hardware - pvrdma was using the wrong command opcode toward the hypervisor - NULL pointer crash regression when dumping rdma-cm over netlink - Be conservative about exposing the global rkey Adit Ranadive (1): RDMA/vmw_pvrdma: Return the correct opcode when creating WR Gal Pressman (1): RDMA/uverbs: Fix post send success return value in case of error Jason Gunthorpe (1): RDMA/uverbs: Mark ioctl responses with UVERBS_ATTR_F_VALID_OUTPUT Leon Romanovsky (2): RDMA/nldev: Don't expose unsafe global rkey to regular user RDMA/mthca: Clear QP objects during their allocation Steve Wise (1): RDMA/cma: Add cm_id restrack resource based on kernel or user cm_id type drivers/infiniband/core/cma.c| 5 ++- drivers/infiniband/core/nldev.c | 4 -- drivers/infiniband/core/rdma_core.h | 2 + drivers/infiniband/core/uverbs_cmd.c | 11 - drivers/infiniband/core/uverbs_ioctl.c | 62 ++-- drivers/infiniband/core/uverbs_main.c| 1 + drivers/infiniband/hw/mthca/mthca_provider.c | 4 +- drivers/infiniband/hw/vmw_pvrdma/pvrdma.h| 35 +++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c | 6 +++ include/uapi/rdma/vmw_pvrdma-abi.h | 1 + 10 files changed, 109 insertions(+), 22 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Fri, 4 Jan 2019 05:00:10 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/3954e1d0310e30e743431b58918825c4d4fe8812 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, Second merge window pull request This could also be applied after rc1. It is a small collection of fixes posted over the holiday break. The merge error with advise_mr is worth fixing right away so the other, more minor, things are coming along for the ride. The following changes since commit 5d24ae67a961c51beb255a28c9c417d9710247c2: Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma (2018-12-28 14:57:10 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 9c6260de505b63638dd86fcc33849b17f6146d94: infiniband/qedr: Potential null ptr dereference of qp (2019-01-02 16:08:38 -0700) 4.21 merge window 2nd pull request Over the break a few defects were found, so this is a -rc style pull request of various small things that have been posted. - An attempt to shorten RCU grace period driven delays showed crashes during heavier testing, and has been entirely reverted - A missed merge/rebase error between the advise_mr and ib_device_ops series - Some small static analysis driven fixes from Julia and Aditya - Missed ability to create a XRC_INI in the devx verbs interop series Aditya Pakki (2): infiniband: bnxt_re: qplib: Check the return value of send_message infiniband/qedr: Potential null ptr dereference of qp Julia Lawall (1): IB/ipoib: drop useless LIST_HEAD Leon Romanovsky (1): Revert "IB/mlx5: Fix long EEH recover time with NVMe offloads" Moni Shoua (1): IB/core: Add advise_mr to the list of known ops Yishai Hadas (1): IB/mlx5: Allow XRC INI usage via verbs in DEVX context drivers/infiniband/core/device.c | 1 + drivers/infiniband/hw/bnxt_re/qplib_sp.c | 5 ++--- drivers/infiniband/hw/mlx5/mr.c | 19 +++ drivers/infiniband/hw/mlx5/qp.c | 3 ++- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 2 ++ drivers/infiniband/ulp/ipoib/ipoib_ib.c | 1 - 6 files changed, 10 insertions(+), 21 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
The pull request you sent on Mon, 24 Dec 2018 22:16:14 +: > git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/5d24ae67a961c51beb255a28c9c417d9710247c2 Thank you! -- Deet-doot-dot, I am a bot. https://korg.wiki.kernel.org/userdoc/prtracker
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
These are the proposed RDMA patches for 4.21.
There is a conflict with the net tree merged in 4.20 and the Mellanox shared
tree. DaveM has a resolution for this in his net-next tree already, but if you
merge RDMA first, here is the resolution:
struct mlx5_ifc_flow_table_eswitch_cap_bits {
- u8 reserved_at_0[0x1c];
- u8 fdb_multi_path_to_table[0x1];
- u8 reserved_at_1d[0x1];
+ u8 reserved_at_0[0x1a];
u8 multi_fdb_encap[0x1];
- u8 reserved_at_1f[0x1e1];
+ u8 reserved_at_1b[0x1];
+ u8 fdb_multi_path_to_table[0x1];
+ u8 reserved_at_1d[0x3];
+
+ u8 reserved_at_20[0x1e0];
There is also a conflict in -next with the NFS tree, in this case the correct
resolution is to continue to delete net/sunrpc/xprtrdma/fmr_ops.c
Things got a bit busy in a rush up to the US holiday season, so we may see a
busier early -rc cycle, I've seen a few thing that suggest there are some new
bugs now. The major changes to the uapi will hopefully encourage all the
driver maintainer do to early testing.
This PR is generated off v4.20, typically we've waited until net-next is
merged. For this reason you will see below a large number of patches that
duplicate ones in net-next from the Mellanox shared branch.
The tag for-linus-merged with my merge resolution to your tree is also
available to pull.
The following changes since commit 40e020c129cfc991e8ab4736d2665351ffd1468d:
Linux 4.20-rc6 (2018-12-09 15:31:00 -0800)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to f617e5ffe04fd46010b618c9eeadaa04588704c9:
RDMA/srpt: Use kmem_cache_free() instead of kfree() (2018-12-22 16:07:47
-0700)
4.21 merge window pull request
This has been a fairly typical cycle, with the usual sorts of driver
updates. Several series continue to come through which improve and
modernize various parts of the core code, and we finally are starting to
get the uAPI command interface cleaned up.
- Various driver fixes for bnxt_re, cxgb3/4, hfi1, hns, i40iw, mlx4, mlx5,
qib, rxe, usnic
- Rework the entire syscall flow for uverbs to be able to run over
ioctl(). Finally getting past the historic bad choice to use write()
for command execution
- More functional coverage with the mlx5 'devx' user API
- Start of the HFI1 series for 'TID RDMA'
- SRQ support in the hns driver
- Support for new IBTA defined 2x lane widths
- A big series to consolidate all the driver function pointers into
a big struct and have drivers provide a 'static const' version of the
struct instead of open coding initialization
- New 'advise_mr' uAPI to control device caching/loading of page tables
- Support for inline data in SRPT
- Modernize how umad uses the driver core and creates cdev's and sysfs
files
- First steps toward removing 'uobject' from the view of the drivers
Alex Vesker (1):
IB/mlx5: Enable TX on a DEVX flow table
Allen Pais (1):
RDMA/hns: prefer dma_zalloc_coherent to dma_alloc_coherent/memse
Andrew Boyer (2):
RDMA/rxe: Distinguish between down links and disabled links
RDMA/rxe: Add link_down, rdma_sends, rdma_recvs stats counters
Artemy Kovalyov (1):
IB/mlx5: Allow modify AV in DCI QP to RTR
Ashutosh Dixit (1):
IB/hfi1: Consider LMC in 16B/bypass ingress packet check
Bart Van Assche (15):
include/scsi/srp.h: Move response flag definitions into this file
include/scsi/srp.h: Add support for immediate data
RDMA/srp: Document srp_parse_in() arguments
RDMA/srp: Handle large SCSI CDBs correctly
RDMA/srp: Propagate ib_post_send() failures to the SCSI mid-layer
RDMA/srp: Move srp_rdma_ch.max_ti_iu_len declaration
RDMA/srp: Rework handling of the maximum information unit length
RDMA/srp: Add support for immediate data
RDMA/srpt: Fix a use-after-free in the channel release code
RDMA/srpt: Improve coding style conformance
RDMA/srpt: Join split strings
RDMA/srpt: Make kernel-doc headers complete
RDMA/srpt: Remove driver version and release date
RDMA/srpt: Rework the srpt_alloc_srq() error path
RDMA/srpt: Add support for immediate data
Chuck Lever (1):
rxe: IB_WR_REG_MR does not capture MR's iova field
Colin Ian King (3):
IB/qib: fix spelling mistake "colescing" -> "coalescing"
RDMA/drivers: Fix spelling mistake "initalize" -> "initialize"
IB/usnic: fix spelling mistake "miniumum" -> "minimum"
Dan Carpenter (4):
IB/qib: Fix an error code in qib_sdma_verbs_send()
RDMA/hns: Fix an error code in hns_roce_create_srq()
IB/uverbs: Signedness bug in UVERBS_HANDLER()
RDMA/mlx5: Signedness bug in UVERBS_HANDLER()
Daniel Jurgens (1):
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request This is a bit later than usual for our first -rc but I'm not seeing anything worry-some in the RDMA tree right now. Quiet so far this -rc cycle, only a few internal driver related bugs and a small series fixing ODP bugs found by more advanced testing. The following changes since commit 651022382c7f8da46cb4872a545ee1da6d097d2a: Linux 4.20-rc1 (2018-11-04 15:37:52 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 7bca603a69c0c239654a8f0bcb99e1a60b30040c: RDMA/mlx5: Initialize return variable in case pagefault was skipped (2018-11-29 15:16:45 -0700) First rc pull request A set of small driver and core code fixes: - Small series fixing longtime user triggerable bugs in the ODP processing inside mlx5 and core code - Various small driver malfunctions and crashes (use after, free, error unwind, implementation bugs) - A misfunction of the RDMA GID cache that can be triggered by the administrator Artemy Kovalyov (3): IB/mlx5: Skip non-ODP MR when handling a page fault IB/umem: Set correct address to the invalidation function IB/mlx5: Fix page fault handling for MW Kamal Heib (1): RDMA/rdmavt: Fix rvt_create_ah function signature Leon Romanovsky (1): RDMA/mlx5: Initialize return variable in case pagefault was skipped Majd Dibbiny (1): RDMA/mlx5: Fix fence type for IB_WR_LOCAL_INV WR Michael Guralnik (1): IB/mlx5: Avoid load failure due to unknown link width Parav Pandit (1): RDMA/core: Add GIDs while changing MAC addr only for registered ndev Sagi Grimberg (1): iser: set sector for ambiguous mr status errors Selvin Xavier (2): RDMA/bnxt_re: Fix system hang when registration with L2 driver fails RDMA/bnxt_re: Avoid accessing the device structure after it is freed Yishai Hadas (1): net/mlx5: Fix XRC SRQ umem valid bits Yixian Liu (1): RDMA/hns: Bugfix pbl configuration for rereg mr Yonatan Cohen (1): IB/mlx5: Fix XRC QP support after introducing extended atomic drivers/infiniband/core/roce_gid_mgmt.c| 6 +- drivers/infiniband/core/umem_odp.c | 20 ++--- drivers/infiniband/hw/bnxt_re/main.c | 3 + drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 128 ++--- drivers/infiniband/hw/mlx5/main.c | 29 +++ drivers/infiniband/hw/mlx5/odp.c | 10 +++ drivers/infiniband/hw/mlx5/qp.c| 22 ++--- drivers/infiniband/sw/rdmavt/ah.c | 4 +- drivers/infiniband/sw/rdmavt/ah.h | 3 +- drivers/infiniband/ulp/iser/iser_verbs.c | 7 +- include/linux/mlx5/mlx5_ifc.h | 12 ++- 11 files changed, 121 insertions(+), 123 deletions(-) signature.asc Description: PGP signature
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus, First rc pull request This is a bit later than usual for our first -rc but I'm not seeing anything worry-some in the RDMA tree right now. Quiet so far this -rc cycle, only a few internal driver related bugs and a small series fixing ODP bugs found by more advanced testing. The following changes since commit 651022382c7f8da46cb4872a545ee1da6d097d2a: Linux 4.20-rc1 (2018-11-04 15:37:52 -0800) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 7bca603a69c0c239654a8f0bcb99e1a60b30040c: RDMA/mlx5: Initialize return variable in case pagefault was skipped (2018-11-29 15:16:45 -0700) First rc pull request A set of small driver and core code fixes: - Small series fixing longtime user triggerable bugs in the ODP processing inside mlx5 and core code - Various small driver malfunctions and crashes (use after, free, error unwind, implementation bugs) - A misfunction of the RDMA GID cache that can be triggered by the administrator Artemy Kovalyov (3): IB/mlx5: Skip non-ODP MR when handling a page fault IB/umem: Set correct address to the invalidation function IB/mlx5: Fix page fault handling for MW Kamal Heib (1): RDMA/rdmavt: Fix rvt_create_ah function signature Leon Romanovsky (1): RDMA/mlx5: Initialize return variable in case pagefault was skipped Majd Dibbiny (1): RDMA/mlx5: Fix fence type for IB_WR_LOCAL_INV WR Michael Guralnik (1): IB/mlx5: Avoid load failure due to unknown link width Parav Pandit (1): RDMA/core: Add GIDs while changing MAC addr only for registered ndev Sagi Grimberg (1): iser: set sector for ambiguous mr status errors Selvin Xavier (2): RDMA/bnxt_re: Fix system hang when registration with L2 driver fails RDMA/bnxt_re: Avoid accessing the device structure after it is freed Yishai Hadas (1): net/mlx5: Fix XRC SRQ umem valid bits Yixian Liu (1): RDMA/hns: Bugfix pbl configuration for rereg mr Yonatan Cohen (1): IB/mlx5: Fix XRC QP support after introducing extended atomic drivers/infiniband/core/roce_gid_mgmt.c| 6 +- drivers/infiniband/core/umem_odp.c | 20 ++--- drivers/infiniband/hw/bnxt_re/main.c | 3 + drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 128 ++--- drivers/infiniband/hw/mlx5/main.c | 29 +++ drivers/infiniband/hw/mlx5/odp.c | 10 +++ drivers/infiniband/hw/mlx5/qp.c| 22 ++--- drivers/infiniband/sw/rdmavt/ah.c | 4 +- drivers/infiniband/sw/rdmavt/ah.h | 3 +- drivers/infiniband/ulp/iser/iser_verbs.c | 7 +- include/linux/mlx5/mlx5_ifc.h | 12 ++- 11 files changed, 121 insertions(+), 123 deletions(-) signature.asc Description: PGP signature
Re: [GIT PULL] Please pull RDMA subsystem changes
On Thu, Oct 25, 2018 at 2:21 PM Jason Gunthorpe wrote: > > This pull request is following your recommendation from 4.19, the > for-linus tag has no merge, a for-linus-merged tag has my merge > resolution for your reference, and the diffstat below has been > replaced with the diffstat from the for-linus-merged tag. Thanks, looks good. Pulled, Linus
Re: [GIT PULL] Please pull RDMA subsystem changes
On Thu, Oct 25, 2018 at 2:21 PM Jason Gunthorpe wrote: > > This pull request is following your recommendation from 4.19, the > for-linus tag has no merge, a for-linus-merged tag has my merge > resolution for your reference, and the diffstat below has been > replaced with the diffstat from the for-linus-merged tag. Thanks, looks good. Pulled, Linus
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
Here are the RDMA patches for 4.20.
This time there is one simple git contextual conflict and a conflict
that causes a compile failure. The latter is fixed with this patch:
diff --git a/drivers/infiniband/hw/mlx5/flow.c
b/drivers/infiniband/hw/mlx5/flow.c
index e57435cb6d96..f86cdcafdafc 100644
--- a/drivers/infiniband/hw/mlx5/flow.c
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -160,7 +160,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
ret = -EINVAL;
goto err_out;
}
- flow_act.has_flow_tag = true;
+ flow_act.flags |= FLOW_ACT_HAS_TAG;
}
flow_handler = mlx5_ib_raw_fs_rule_add(dev, fs_matcher, _act,
This appears to have arisen because a patch was sent to netdev
changing infiniband code that should have gone to the shared tree. The
appropriate people have been reminded about the process.
Otherwise it has been a reasonable cycle here with not too much major
change.
This pull request is following your recommendation from 4.19, the
for-linus tag has no merge, a for-linus-merged tag has my merge
resolution for your reference, and the diffstat below has been
replaced with the diffstat from the for-linus-merged tag.
The following changes since commit 94a04d1d3d3681adde1a3e022b25dbac7b345b7e:
net/mlx5: Expose DC scatter to CQE capability bit (2018-10-16 21:55:29 +0300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to a60109dc9a954ef9eddba6577e2d2e9e7952e487:
IB/mlx5: Add support for extended atomic operations (2018-10-17 11:53:23
-0400)
First merge window pull request
This has been a smaller cycle with many of the commits being smallish code
fixes and improvements across the drivers.
- Driver updates for bnxt_re, cxgb4, hfi1, hns, mlx5, nes, qedr, and rxe
- Memory window support in hns
- mlx5 user API 'flow mutate/steering' allows accessing the full packet
mangling and matching machinery from user space
- Support inter-working with verbs API calls in the 'devx' mlx5 user API, and
provide options to use devx with less privilege
- Modernize the use of syfs and the device interface to use attribute groups
and cdev properly for uverbs, and clean up some of the core code's device list
management
- More progress on net namespaces for RDMA devices
- Consolidate driver BAR mmapping support into core code helpers and rework
how RDMA holds poitners to mm_struct for get_user_pages cases
- First pass to use 'dev_name' instead of ib_device->name
- Device renaming for RDMA devices
Alex Estrin (2):
IB/sa: simplify return code logic for ib_nl_send_msg()
IB/hfi1: Add mtu check for operational data VLs
Arnd Bergmann (1):
RDMA: Fix dependencies for rdma_user_mmap_io
Arseny Maslennikov (3):
Documentation/ABI: document /sys/class/net/*/dev_port
IB/ipoib: Use dev_port to expose network interface port numbers
IB/ipoib: Log sysfs 'dev_id' accesses from userspace
Artemy Kovalyov (1):
IB/mlx5: Fix MR cache initialization
Brian Welty (1):
IB/{hfi1, qib, rdmavt}: Move copy SGE logic into rdmavt
Chuck Lever (2):
RDMA/core: Document CM @event_handler function
RDMA/core: Document QP @event_handler function
Colin Ian King (3):
IB/usnic: fix spelling mistake "unvalid" -> "invalid"
IB/mthca: remove redundant inner check of mdev->mthca_flags
RDMA/hns: fix spelling mistake "reseved" -> "reserved"
Dan Carpenter (1):
RDMA/hns: Fix an error code in hns_roce_v2_init_eq_table()
Denis Drozdov (1):
IB/ipoib: Clear IPCB before icmp_send
Dennis Dalessandro (3):
IB/hfi1: Rework file list in Makefile
IB/hfi1: Prepare resource waits for dual leg
IB/hfi1: Ensure ucast_dlid access doesnt exceed bounds
Devesh Sharma (3):
RDMA/bnxt_re: Drop L2 async events silently
RDMA/bnxt_re: Fix qp async event reporting
RDMA/bnxt_re: Limit max_pkey to 16 bit value
Doug Ledford (4):
Merge branch 'mlx5-vport-loopback' into rdma.get
RDMA/umem: Minor optimizations
RDMA/umem: Fix potential addition overflow
Merge remote-tracking branch 'mlx5-next' into for-next
Gal Pressman (1):
RDMA/mlx5: Remove extraneous error check
Guy Levi (1):
IB/uverbs: Add IDRs array attribute type to ioctl() interface
Hannes Reinecke (1):
ib_srp: Remove WARN_ON in srp_terminate_io()
Håkon Bugge (3):
RDMA/i40iw: Fix incorrect iterator type
IB/mlx4: Enable debug print of SMPs
IB/mlx4: Add port and TID to MAD debug print
Igor Stoppa (1):
IB/srp: Remove unnecessary unlikely()
Israel Rukshin (1):
IB/iser: Fix possible NULL deref at iser_inv_desc()
Jack Morgenstein (1):
IB/core: Add an unbound WQ
[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,
Here are the RDMA patches for 4.20.
This time there is one simple git contextual conflict and a conflict
that causes a compile failure. The latter is fixed with this patch:
diff --git a/drivers/infiniband/hw/mlx5/flow.c
b/drivers/infiniband/hw/mlx5/flow.c
index e57435cb6d96..f86cdcafdafc 100644
--- a/drivers/infiniband/hw/mlx5/flow.c
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -160,7 +160,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
ret = -EINVAL;
goto err_out;
}
- flow_act.has_flow_tag = true;
+ flow_act.flags |= FLOW_ACT_HAS_TAG;
}
flow_handler = mlx5_ib_raw_fs_rule_add(dev, fs_matcher, _act,
This appears to have arisen because a patch was sent to netdev
changing infiniband code that should have gone to the shared tree. The
appropriate people have been reminded about the process.
Otherwise it has been a reasonable cycle here with not too much major
change.
This pull request is following your recommendation from 4.19, the
for-linus tag has no merge, a for-linus-merged tag has my merge
resolution for your reference, and the diffstat below has been
replaced with the diffstat from the for-linus-merged tag.
The following changes since commit 94a04d1d3d3681adde1a3e022b25dbac7b345b7e:
net/mlx5: Expose DC scatter to CQE capability bit (2018-10-16 21:55:29 +0300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus
for you to fetch changes up to a60109dc9a954ef9eddba6577e2d2e9e7952e487:
IB/mlx5: Add support for extended atomic operations (2018-10-17 11:53:23
-0400)
First merge window pull request
This has been a smaller cycle with many of the commits being smallish code
fixes and improvements across the drivers.
- Driver updates for bnxt_re, cxgb4, hfi1, hns, mlx5, nes, qedr, and rxe
- Memory window support in hns
- mlx5 user API 'flow mutate/steering' allows accessing the full packet
mangling and matching machinery from user space
- Support inter-working with verbs API calls in the 'devx' mlx5 user API, and
provide options to use devx with less privilege
- Modernize the use of syfs and the device interface to use attribute groups
and cdev properly for uverbs, and clean up some of the core code's device list
management
- More progress on net namespaces for RDMA devices
- Consolidate driver BAR mmapping support into core code helpers and rework
how RDMA holds poitners to mm_struct for get_user_pages cases
- First pass to use 'dev_name' instead of ib_device->name
- Device renaming for RDMA devices
Alex Estrin (2):
IB/sa: simplify return code logic for ib_nl_send_msg()
IB/hfi1: Add mtu check for operational data VLs
Arnd Bergmann (1):
RDMA: Fix dependencies for rdma_user_mmap_io
Arseny Maslennikov (3):
Documentation/ABI: document /sys/class/net/*/dev_port
IB/ipoib: Use dev_port to expose network interface port numbers
IB/ipoib: Log sysfs 'dev_id' accesses from userspace
Artemy Kovalyov (1):
IB/mlx5: Fix MR cache initialization
Brian Welty (1):
IB/{hfi1, qib, rdmavt}: Move copy SGE logic into rdmavt
Chuck Lever (2):
RDMA/core: Document CM @event_handler function
RDMA/core: Document QP @event_handler function
Colin Ian King (3):
IB/usnic: fix spelling mistake "unvalid" -> "invalid"
IB/mthca: remove redundant inner check of mdev->mthca_flags
RDMA/hns: fix spelling mistake "reseved" -> "reserved"
Dan Carpenter (1):
RDMA/hns: Fix an error code in hns_roce_v2_init_eq_table()
Denis Drozdov (1):
IB/ipoib: Clear IPCB before icmp_send
Dennis Dalessandro (3):
IB/hfi1: Rework file list in Makefile
IB/hfi1: Prepare resource waits for dual leg
IB/hfi1: Ensure ucast_dlid access doesnt exceed bounds
Devesh Sharma (3):
RDMA/bnxt_re: Drop L2 async events silently
RDMA/bnxt_re: Fix qp async event reporting
RDMA/bnxt_re: Limit max_pkey to 16 bit value
Doug Ledford (4):
Merge branch 'mlx5-vport-loopback' into rdma.get
RDMA/umem: Minor optimizations
RDMA/umem: Fix potential addition overflow
Merge remote-tracking branch 'mlx5-next' into for-next
Gal Pressman (1):
RDMA/mlx5: Remove extraneous error check
Guy Levi (1):
IB/uverbs: Add IDRs array attribute type to ioctl() interface
Hannes Reinecke (1):
ib_srp: Remove WARN_ON in srp_terminate_io()
Håkon Bugge (3):
RDMA/i40iw: Fix incorrect iterator type
IB/mlx4: Enable debug print of SMPs
IB/mlx4: Add port and TID to MAD debug print
Igor Stoppa (1):
IB/srp: Remove unnecessary unlikely()
Israel Rukshin (1):
IB/iser: Fix possible NULL deref at iser_inv_desc()
Jack Morgenstein (1):
IB/core: Add an unbound WQ
