From d6673746d69909f9ab52cf9fe1f9a21fca202e0c Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Thu, 30 Jul 2020 11:12:35 +0300 Subject: RDMA: Remove constant domain argument from flow creation call The "domain" argument is constant and modern device (mlx5) doesn't support anything except IB_FLOW_DOMAIN_USER, so delete this extra parameter and simplify code. Link: https://lore.kernel.org/r/20200730081235.1581127-4-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index c0b2fa7e9b95..55dfe0e9d1d6 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1863,17 +1863,6 @@ enum ib_flow_spec_type { #define IB_FLOW_SPEC_LAYER_MASK 0xF0 #define IB_FLOW_SPEC_SUPPORT_LAYERS 10 -/* Flow steering rule priority is set according to it's domain. - * Lower domain value means higher priority. - */ -enum ib_flow_domain { - IB_FLOW_DOMAIN_USER, - IB_FLOW_DOMAIN_ETHTOOL, - IB_FLOW_DOMAIN_RFS, - IB_FLOW_DOMAIN_NIC, - IB_FLOW_DOMAIN_NUM /* Must be last */ -}; - enum ib_flow_flags { IB_FLOW_ATTR_FLAGS_DONT_TRAP = 1UL << 1, /* Continue match, no steal */ IB_FLOW_ATTR_FLAGS_EGRESS = 1UL << 2, /* Egress flow */ @@ -2471,7 +2460,7 @@ struct ib_device_ops { void (*dealloc_xrcd)(struct ib_xrcd *xrcd, struct ib_udata *udata); struct ib_flow *(*create_flow)(struct ib_qp *qp, struct ib_flow_attr *flow_attr, - int domain, struct ib_udata *udata); + struct ib_udata *udata); int (*destroy_flow)(struct ib_flow *flow_id); struct ib_flow_action *(*create_flow_action_esp)( struct ib_device *device, -- cgit v1.2.3 From a4e6a1dd57469d6ecee084db1507d3e37908d1e2 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Fri, 31 Jul 2020 09:04:20 +0300 Subject: RDMA/efa: Introduce SRD RNR retry This patch introduces the ability to configure SRD QPs with the RNR retry parameter when issuing a modify QP command. In addition, a capability bit was added to report support to the userspace library. Link: https://lore.kernel.org/r/20200731060420.17053-5-galpress@amazon.com Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe --- include/uapi/rdma/efa-abi.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/uapi/rdma/efa-abi.h b/include/uapi/rdma/efa-abi.h index 507a2862bedb..f89fbb5b1e8d 100644 --- a/include/uapi/rdma/efa-abi.h +++ b/include/uapi/rdma/efa-abi.h @@ -105,6 +105,7 @@ struct efa_ibv_create_ah_resp { enum { EFA_QUERY_DEVICE_CAPS_RDMA_READ = 1 << 0, + EFA_QUERY_DEVICE_CAPS_RNR_RETRY = 1 << 1, }; struct efa_ibv_ex_query_device_resp { -- cgit v1.2.3 From c0f4979e90eaa4cd4a63ae95fbf8e65a85150a5f Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Wed, 5 Aug 2020 14:28:52 -0300 Subject: RDMA/cm: Remove unused cm_class Previous commits removed all references to the /sys/class/infiniband_cm/ directory represented by the cm_class symbol. Remove the directory and cm_class. Fixes: a1a8e4a85cf7 ("rdma: Delete the ib_ucm module") Link: https://lore.kernel.org/r/0-v1-90096a98c476+205-remove_cm_leftovers_jgg@nvidia.com Signed-off-by: Jason Gunthorpe --- include/rdma/ib_cm.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_cm.h b/include/rdma/ib_cm.h index 382427add677..e23eb357b761 100644 --- a/include/rdma/ib_cm.h +++ b/include/rdma/ib_cm.h @@ -14,9 +14,6 @@ #include #include -/* ib_cm and ib_user_cm modules share /sys/class/infiniband_cm */ -extern struct class cm_class; - enum ib_cm_state { IB_CM_IDLE, IB_CM_LISTEN, -- cgit v1.2.3 From b3d03daa7cd19a91266dc2cea3587dcf60e7a1f0 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Aug 2020 09:53:11 -0400 Subject: RDMA/core: Move the rdma_show_ib_cm_event() macro Refactor: Make it globally available in the utilities header. Link: https://lore.kernel.org/r/159767239131.2968.9520990257041764685.stgit@klimt.1015granger.net Signed-off-by: Chuck Lever Signed-off-by: Jason Gunthorpe --- include/trace/events/rdma.h | 41 ++++++++++++++++++++++++++++++++++++++++- include/trace/events/rpcrdma.h | 1 + 2 files changed, 41 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/trace/events/rdma.h b/include/trace/events/rdma.h index aa19afc73a4e..81bb454fc288 100644 --- a/include/trace/events/rdma.h +++ b/include/trace/events/rdma.h @@ -6,7 +6,6 @@ /* * enum ib_event_type, from include/rdma/ib_verbs.h */ - #define IB_EVENT_LIST \ ib_event(CQ_ERR) \ ib_event(QP_FATAL) \ @@ -90,6 +89,46 @@ IB_WC_STATUS_LIST #define rdma_show_wc_status(x) \ __print_symbolic(x, IB_WC_STATUS_LIST) +/* + * enum ib_cm_event_type, from include/rdma/ib_cm.h + */ +#define IB_CM_EVENT_LIST \ + ib_cm_event(REQ_ERROR) \ + ib_cm_event(REQ_RECEIVED) \ + ib_cm_event(REP_ERROR) \ + ib_cm_event(REP_RECEIVED) \ + ib_cm_event(RTU_RECEIVED) \ + ib_cm_event(USER_ESTABLISHED) \ + ib_cm_event(DREQ_ERROR) \ + ib_cm_event(DREQ_RECEIVED) \ + ib_cm_event(DREP_RECEIVED) \ + ib_cm_event(TIMEWAIT_EXIT) \ + ib_cm_event(MRA_RECEIVED) \ + ib_cm_event(REJ_RECEIVED) \ + ib_cm_event(LAP_ERROR) \ + ib_cm_event(LAP_RECEIVED) \ + ib_cm_event(APR_RECEIVED) \ + ib_cm_event(SIDR_REQ_ERROR) \ + ib_cm_event(SIDR_REQ_RECEIVED) \ + ib_cm_event_end(SIDR_REP_RECEIVED) + +#undef ib_cm_event +#undef ib_cm_event_end + +#define ib_cm_event(x) TRACE_DEFINE_ENUM(IB_CM_##x); +#define ib_cm_event_end(x) TRACE_DEFINE_ENUM(IB_CM_##x); + +IB_CM_EVENT_LIST + +#undef ib_cm_event +#undef ib_cm_event_end + +#define ib_cm_event(x) { IB_CM_##x, #x }, +#define ib_cm_event_end(x) { IB_CM_##x, #x } + +#define rdma_show_ib_cm_event(x) \ + __print_symbolic(x, IB_CM_EVENT_LIST) + /* * enum rdma_cm_event_type, from include/rdma/rdma_cm.h */ diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index abe942225637..b6aad52beb62 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -13,6 +13,7 @@ #include #include #include +#include #include /** -- cgit v1.2.3 From 7c4b1ab9f16732fb921b3f11cd127fa65f26ad5c Mon Sep 17 00:00:00 2001 From: Mark Zhang Date: Tue, 18 Aug 2020 14:52:45 +0300 Subject: IB/mlx5: Add DCT RoCE LAG support When DCT QPs work in RoCE LAG mode: 1. DCT creation is allowed only when it is supported 2. The "port" of a DCT QP is assigned in a round-robin way Link: https://lore.kernel.org/r/20200818115245.700581-3-leon@kernel.org Signed-off-by: Mark Zhang Reviewed-by: Maor Gottlieb Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/linux/mlx5/mlx5_ifc.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index de1ffb4804d6..aee25e4fb2cc 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -1430,7 +1430,8 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 log_bf_reg_size[0x5]; - u8 reserved_at_270[0x8]; + u8 reserved_at_270[0x6]; + u8 lag_dct[0x2]; u8 lag_tx_port_affinity[0x1]; u8 reserved_at_279[0x2]; u8 lag_master[0x1]; -- cgit v1.2.3 From d114c6feedfe0600c19b9f9479a4026354d1f7fd Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 18 Aug 2020 15:05:18 +0300 Subject: RDMA/cma: Add missing locking to rdma_accept() In almost all cases rdma_accept() is called under the handler_mutex by ULPs from their handler callbacks. The one exception was ucma which did not get the handler_mutex. To improve the understand-ability of the locking scheme obtain the mutex for ucma as well. This improves how ucma works by allowing it to directly use handler_mutex for some of its internal locking against the handler callbacks intead of the global file->mut lock. There does not seem to be a serious bug here, other than a DISCONNECT event can be delivered concurrently with accept succeeding. Link: https://lore.kernel.org/r/20200818120526.702120-7-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/rdma_cm.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index cf5da2ae49bf..c1334c9a7aa8 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -253,6 +253,8 @@ int rdma_listen(struct rdma_cm_id *id, int backlog); int __rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, const char *caller); +void rdma_lock_handler(struct rdma_cm_id *id); +void rdma_unlock_handler(struct rdma_cm_id *id); int __rdma_accept_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, const char *caller, struct rdma_ucm_ece *ece); @@ -270,6 +272,9 @@ int __rdma_accept_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, * In the case of error, a reject message is sent to the remote side and the * state of the qp associated with the id is modified to error, such that any * previously posted receive buffers would be flushed. + * + * This function is for use by kernel ULPs and must be called from under the + * handler callback. */ #define rdma_accept(id, conn_param) \ __rdma_accept((id), (conn_param), KBUILD_MODNAME) -- cgit v1.2.3 From 5f9e2822d12fe5050da5db0e65924d5ddc86bf29 Mon Sep 17 00:00:00 2001 From: Bob Pearson Date: Thu, 20 Aug 2020 17:46:23 -0500 Subject: RDMA/rxe: Fix style warnings Fixed several minor checkpatch warnings in existing rxe source. Link: https://lore.kernel.org/r/20200820224638.3212-3-rpearson@hpe.com Signed-off-by: Bob Pearson Signed-off-by: Jason Gunthorpe --- include/uapi/rdma/rdma_user_rxe.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/uapi/rdma/rdma_user_rxe.h b/include/uapi/rdma/rdma_user_rxe.h index aae2e696bb38..d8f2e0e46dab 100644 --- a/include/uapi/rdma/rdma_user_rxe.h +++ b/include/uapi/rdma/rdma_user_rxe.h @@ -99,8 +99,8 @@ struct rxe_send_wr { struct ib_mr *mr; __aligned_u64 reserved; }; - __u32 key; - __u32 access; + __u32 key; + __u32 access; } reg; } wr; }; @@ -112,7 +112,7 @@ struct rxe_sge { }; struct mminfo { - __aligned_u64 offset; + __aligned_u64 offset; __u32 size; __u32 pad; }; -- cgit v1.2.3 From 074bf2c2c7a1f5722eabccee7d919fe20ada0cda Mon Sep 17 00:00:00 2001 From: Weihang Li Date: Fri, 21 Aug 2020 17:31:29 +0800 Subject: RDMA/hns: Get udp sport num dynamically instead of using a fixed value The UDP source port number in RoCE v2 is used to create entropy for network routers (ECMP), load balancers and 802.3ad link aggregation switching that are not aware of RoCE IB headers. Considering that the IB core has achieved a new interface to get a hashed value of it, the fixed value of it in QPC and UD WQE in hns driver could be fixed and the port number is to be set dynamically now. For QPC of RC, the value could be hashed from flow_lable if the user pass it in or from remote qpn and local qpn. For WQE of UD, it is set according to fl or as a random value. Link: https://lore.kernel.org/r/1598002289-8611-1-git-send-email-liweihang@huawei.com Signed-off-by: Weihang Li Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 55dfe0e9d1d6..c868609a4ffa 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -4706,6 +4706,7 @@ bool rdma_dev_access_netns(const struct ib_device *device, const struct net *net); #define IB_ROCE_UDP_ENCAP_VALID_PORT_MIN (0xC000) +#define IB_ROCE_UDP_ENCAP_VALID_PORT_MAX (0xFFFF) #define IB_GRH_FLOWLABEL_MASK (0x000FFFFF) /** -- cgit v1.2.3 From 61690d01db32eb1f94adc9ac2b8bb741d34e4671 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 25 Aug 2020 15:17:08 -0300 Subject: RDMA/umem: Fix signature of stub ib_umem_find_best_pgsz() The original function returns unsigned long and 0 on failure. Fixes: 4a35339958f1 ("RDMA/umem: Add API to find best driver supported page size in an MR") Link: https://lore.kernel.org/r/0-v1-982a13cc5c6d+501ae-fix_best_pgsz_stub_jgg@nvidia.com Reviewed-by: Gal Pressman Acked-by: Shiraz Saleem Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 71f573a418bf..07a764eb692e 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -68,10 +68,11 @@ static inline int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offs size_t length) { return -EINVAL; } -static inline int ib_umem_find_best_pgsz(struct ib_umem *umem, - unsigned long pgsz_bitmap, - unsigned long virt) { - return -EINVAL; +static inline unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem, + unsigned long pgsz_bitmap, + unsigned long virt) +{ + return 0; } #endif /* CONFIG_INFINIBAND_USER_MEM */ -- cgit v1.2.3 From f553246f7f794675da1794ae7ee07d1f35e561ae Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Wed, 2 Sep 2020 11:17:08 +0300 Subject: RDMA/core: Change how failing destroy is handled during uobj abort Currently it triggers a WARN_ON and then goes ahead and destroys the uobject anyhow, leaking any driver memory. The only place that leaks driver memory should be during FD close() in uverbs_destroy_ufile_hw(). Drivers are only allowed to fail destroy uobjects if they guarantee destroy will eventually succeed. uverbs_destroy_ufile_hw() provides the loop to give the driver that chance. Link: https://lore.kernel.org/r/20200902081708.746631-1-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index c868609a4ffa..3781d36df31c 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1463,11 +1463,6 @@ enum rdma_remove_reason { RDMA_REMOVE_DRIVER_REMOVE, /* uobj is being cleaned-up before being committed */ RDMA_REMOVE_ABORT, - /* - * uobj has been fully created, with the uobj->object set, but is being - * cleaned up before being comitted - */ - RDMA_REMOVE_ABORT_HWOBJ, }; struct ib_rdmacg_object { -- cgit v1.2.3 From 91a7c58fce065506fd98954d27694e5d83c96638 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:13 +0300 Subject: RDMA: Restore ability to fail on PD deallocate The IB verbs objects are counted by the kernel and ib_core ensures that deallocate PD will success so it will be called once all other objects that depends on PD will be released. This is achieved by managing various reference counters on such objects. The mlx5 driver didn't follow this standard flow when allowed DEVX objects that are not managed by ib_core to be interleaved with the ones under ib_core responsibility. In such interleaved scenarios deallocate command can fail and ib_core will leave uobject in internal DB and attempt to clean it later to free resources anyway. This change partially restores returned value from dealloc_pd() for all drivers, but keeping in mind that non-DEVX devices and kernel verbs paths shouldn't fail. Fixes: 21a428a019c9 ("RDMA: Handle PD allocations by IB/core") Link: https://lore.kernel.org/r/20200907120921.476363-2-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 3781d36df31c..a49aaebaed58 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2398,7 +2398,7 @@ struct ib_device_ops { void (*mmap_free)(struct rdma_user_mmap_entry *entry); void (*disassociate_ucontext)(struct ib_ucontext *ibcontext); int (*alloc_pd)(struct ib_pd *pd, struct ib_udata *udata); - void (*dealloc_pd)(struct ib_pd *pd, struct ib_udata *udata); + int (*dealloc_pd)(struct ib_pd *pd, struct ib_udata *udata); int (*create_ah)(struct ib_ah *ah, struct rdma_ah_init_attr *attr, struct ib_udata *udata); int (*modify_ah)(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); @@ -3456,12 +3456,7 @@ struct ib_pd *__ib_alloc_pd(struct ib_device *device, unsigned int flags, #define ib_alloc_pd(device, flags) \ __ib_alloc_pd((device), (flags), KBUILD_MODNAME) -/** - * ib_dealloc_pd_user - Deallocate kernel/user PD - * @pd: The protection domain - * @udata: Valid user data or NULL for kernel objects - */ -void ib_dealloc_pd_user(struct ib_pd *pd, struct ib_udata *udata); +int ib_dealloc_pd_user(struct ib_pd *pd, struct ib_udata *udata); /** * ib_dealloc_pd - Deallocate kernel PD @@ -3471,7 +3466,9 @@ void ib_dealloc_pd_user(struct ib_pd *pd, struct ib_udata *udata); */ static inline void ib_dealloc_pd(struct ib_pd *pd) { - ib_dealloc_pd_user(pd, NULL); + int ret = ib_dealloc_pd_user(pd, NULL); + + WARN_ONCE(ret, "Destroy of kernel PD shouldn't fail"); } enum rdma_create_ah_flags { -- cgit v1.2.3 From 9a9ebf8cd72b809405ad571fb2f635ffc9df2420 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:14 +0300 Subject: RDMA: Restore ability to fail on AH destroy Like any other IB verbs objects, AH are refcounted by ib_core. The release of those objects are controlled by ib_core with promise that AH destroy can't fail. Being SW object for now, this change makes dealloc_ah() to behave like any other destroy IB flows. Fixes: d345691471b4 ("RDMA: Handle AH allocations by IB/core") Link: https://lore.kernel.org/r/20200907120921.476363-3-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index a49aaebaed58..71b145f569a8 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2403,7 +2403,7 @@ struct ib_device_ops { struct ib_udata *udata); int (*modify_ah)(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); int (*query_ah)(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); - void (*destroy_ah)(struct ib_ah *ah, u32 flags); + int (*destroy_ah)(struct ib_ah *ah, u32 flags); int (*create_srq)(struct ib_srq *srq, struct ib_srq_init_attr *srq_init_attr, struct ib_udata *udata); @@ -3596,9 +3596,11 @@ int rdma_destroy_ah_user(struct ib_ah *ah, u32 flags, struct ib_udata *udata); * * NOTE: for user ah use rdma_destroy_ah_user with valid udata! */ -static inline int rdma_destroy_ah(struct ib_ah *ah, u32 flags) +static inline void rdma_destroy_ah(struct ib_ah *ah, u32 flags) { - return rdma_destroy_ah_user(ah, flags, NULL); + int ret = rdma_destroy_ah_user(ah, flags, NULL); + + WARN_ONCE(ret, "Destroy of kernel AH shouldn't fail"); } struct ib_srq *ib_create_srq_user(struct ib_pd *pd, -- cgit v1.2.3 From 119181d1d4327d3259ab25aa0ea3d3bc364afcdc Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:16 +0300 Subject: RDMA: Restore ability to fail on SRQ destroy In similar way to other IB objects, restore the ability to return error on SRQ destroy. Strictly speaking, this change is not necessary, and provided here to ensure a symmetrical interface like other destroy functions. Fixes: 68e326dea1db ("RDMA: Handle SRQ allocations by IB/core") Link: https://lore.kernel.org/r/20200907120921.476363-5-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 71b145f569a8..40bb6a455960 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2411,7 +2411,7 @@ struct ib_device_ops { enum ib_srq_attr_mask srq_attr_mask, struct ib_udata *udata); int (*query_srq)(struct ib_srq *srq, struct ib_srq_attr *srq_attr); - void (*destroy_srq)(struct ib_srq *srq, struct ib_udata *udata); + int (*destroy_srq)(struct ib_srq *srq, struct ib_udata *udata); struct ib_qp *(*create_qp)(struct ib_pd *pd, struct ib_qp_init_attr *qp_init_attr, struct ib_udata *udata); @@ -3654,9 +3654,11 @@ int ib_destroy_srq_user(struct ib_srq *srq, struct ib_udata *udata); * * NOTE: for user srq use ib_destroy_srq_user with valid udata! */ -static inline int ib_destroy_srq(struct ib_srq *srq) +static inline void ib_destroy_srq(struct ib_srq *srq) { - return ib_destroy_srq_user(srq, NULL); + int ret = ib_destroy_srq_user(srq, NULL); + + WARN_ONCE(ret, "Destroy of kernel SRQ shouldn't fail"); } /** -- cgit v1.2.3 From 7e3c66c9a989d5b53387ceebc88b9e4a9b1d6434 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:17 +0300 Subject: RDMA/core: Delete function indirection for alloc/free kernel CQ The ib_alloc_cq*() and ib_free_cq*() are solely kernel verbs to manage CQs and doesn't need extra indirection just to call same functions with constant parameter NULL as udata. Link: https://lore.kernel.org/r/20200907120921.476363-6-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 62 +++++-------------------------------------------- 1 file changed, 6 insertions(+), 56 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 40bb6a455960..b74fd1a5ccb6 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -3802,46 +3802,15 @@ static inline int ib_post_recv(struct ib_qp *qp, return qp->device->ops.post_recv(qp, recv_wr, bad_recv_wr ? : &dummy); } -struct ib_cq *__ib_alloc_cq_user(struct ib_device *dev, void *private, - int nr_cqe, int comp_vector, - enum ib_poll_context poll_ctx, - const char *caller, struct ib_udata *udata); - -/** - * ib_alloc_cq_user: Allocate kernel/user CQ - * @dev: The IB device - * @private: Private data attached to the CQE - * @nr_cqe: Number of CQEs in the CQ - * @comp_vector: Completion vector used for the IRQs - * @poll_ctx: Context used for polling the CQ - * @udata: Valid user data or NULL for kernel objects - */ -static inline struct ib_cq *ib_alloc_cq_user(struct ib_device *dev, - void *private, int nr_cqe, - int comp_vector, - enum ib_poll_context poll_ctx, - struct ib_udata *udata) -{ - return __ib_alloc_cq_user(dev, private, nr_cqe, comp_vector, poll_ctx, - KBUILD_MODNAME, udata); -} - -/** - * ib_alloc_cq: Allocate kernel CQ - * @dev: The IB device - * @private: Private data attached to the CQE - * @nr_cqe: Number of CQEs in the CQ - * @comp_vector: Completion vector used for the IRQs - * @poll_ctx: Context used for polling the CQ - * - * NOTE: for user cq use ib_alloc_cq_user with valid udata! - */ +struct ib_cq *__ib_alloc_cq(struct ib_device *dev, void *private, int nr_cqe, + int comp_vector, enum ib_poll_context poll_ctx, + const char *caller); static inline struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private, int nr_cqe, int comp_vector, enum ib_poll_context poll_ctx) { - return ib_alloc_cq_user(dev, private, nr_cqe, comp_vector, poll_ctx, - NULL); + return __ib_alloc_cq(dev, private, nr_cqe, comp_vector, poll_ctx, + KBUILD_MODNAME); } struct ib_cq *__ib_alloc_cq_any(struct ib_device *dev, void *private, @@ -3863,26 +3832,7 @@ static inline struct ib_cq *ib_alloc_cq_any(struct ib_device *dev, KBUILD_MODNAME); } -/** - * ib_free_cq_user - Free kernel/user CQ - * @cq: The CQ to free - * @udata: Valid user data or NULL for kernel objects - * - * NOTE: This function shouldn't be called on shared CQs. - */ -void ib_free_cq_user(struct ib_cq *cq, struct ib_udata *udata); - -/** - * ib_free_cq - Free kernel CQ - * @cq: The CQ to free - * - * NOTE: for user cq use ib_free_cq_user with valid udata! - */ -static inline void ib_free_cq(struct ib_cq *cq) -{ - ib_free_cq_user(cq, NULL); -} - +void ib_free_cq(struct ib_cq *cq); int ib_process_cq_direct(struct ib_cq *cq, int budget); /** -- cgit v1.2.3 From 43d781b9fa562f0c6e50f62c870fbfeb9dc85213 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:18 +0300 Subject: RDMA: Allow fail of destroy CQ Like any other verbs objects, CQ shouldn't fail during destroy, but mlx5_ib didn't follow this contract with mixed IB verbs objects with DEVX. Such mix causes to the situation where FW and kernel are fully interdependent on the reference counting of each side. Kernel verbs and drivers that don't have DEVX flows shouldn't fail. Fixes: e39afe3d6dbd ("RDMA: Convert CQ allocations to be under core responsibility") Link: https://lore.kernel.org/r/20200907120921.476363-7-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index b74fd1a5ccb6..bec05baaeaed 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2423,7 +2423,7 @@ struct ib_device_ops { int (*create_cq)(struct ib_cq *cq, const struct ib_cq_init_attr *attr, struct ib_udata *udata); int (*modify_cq)(struct ib_cq *cq, u16 cq_count, u16 cq_period); - void (*destroy_cq)(struct ib_cq *cq, struct ib_udata *udata); + int (*destroy_cq)(struct ib_cq *cq, struct ib_udata *udata); int (*resize_cq)(struct ib_cq *cq, int cqe, struct ib_udata *udata); struct ib_mr *(*get_dma_mr)(struct ib_pd *pd, int mr_access_flags); struct ib_mr *(*reg_user_mr)(struct ib_pd *pd, u64 start, u64 length, @@ -3890,7 +3890,9 @@ int ib_destroy_cq_user(struct ib_cq *cq, struct ib_udata *udata); */ static inline void ib_destroy_cq(struct ib_cq *cq) { - ib_destroy_cq_user(cq, NULL); + int ret = ib_destroy_cq_user(cq, NULL); + + WARN_ONCE(ret, "Destroy of kernel CQ shouldn't fail"); } /** -- cgit v1.2.3 From d0c45c8556e57342d44c9548763609ffcc4e3866 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:19 +0300 Subject: RDMA: Change XRCD destroy return value Update XRCD destroy flow to allow command failure. Fixes: 28ad5f65c314 ("RDMA: Move XRCD to be under ib_core responsibility") Link: https://lore.kernel.org/r/20200907120921.476363-8-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index bec05baaeaed..e25e25ad8d2c 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2452,7 +2452,7 @@ struct ib_device_ops { int (*attach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); int (*detach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); int (*alloc_xrcd)(struct ib_xrcd *xrcd, struct ib_udata *udata); - void (*dealloc_xrcd)(struct ib_xrcd *xrcd, struct ib_udata *udata); + int (*dealloc_xrcd)(struct ib_xrcd *xrcd, struct ib_udata *udata); struct ib_flow *(*create_flow)(struct ib_qp *qp, struct ib_flow_attr *flow_attr, struct ib_udata *udata); -- cgit v1.2.3 From add53535fb3033c249d9327ae3e7c36d3382bbd1 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:20 +0300 Subject: RDMA: Restore ability to return error for destroy WQ Make this interface symmetrical to other destroy paths. Fixes: a49b1dc7ae44 ("RDMA: Convert destroy_wq to be void") Link: https://lore.kernel.org/r/20200907120921.476363-9-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index e25e25ad8d2c..23209bca8c1b 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2480,7 +2480,7 @@ struct ib_device_ops { struct ib_wq *(*create_wq)(struct ib_pd *pd, struct ib_wq_init_attr *init_attr, struct ib_udata *udata); - void (*destroy_wq)(struct ib_wq *wq, struct ib_udata *udata); + int (*destroy_wq)(struct ib_wq *wq, struct ib_udata *udata); int (*modify_wq)(struct ib_wq *wq, struct ib_wq_attr *attr, u32 wq_attr_mask, struct ib_udata *udata); struct ib_rwq_ind_table *(*create_rwq_ind_table)( @@ -4316,7 +4316,7 @@ struct net_device *ib_device_netdev(struct ib_device *dev, u8 port); struct ib_wq *ib_create_wq(struct ib_pd *pd, struct ib_wq_init_attr *init_attr); -int ib_destroy_wq(struct ib_wq *wq, struct ib_udata *udata); +int ib_destroy_wq_user(struct ib_wq *wq, struct ib_udata *udata); int ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *attr, u32 wq_attr_mask); int ib_destroy_rwq_ind_table(struct ib_rwq_ind_table *wq_ind_table); -- cgit v1.2.3 From 71ff3f6268644783d556fab10066f1027030135f Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 7 Sep 2020 15:09:21 +0300 Subject: RDMA: Make counters destroy symmetrical Change counters to return failure like any other verbs destroy, however this flow shouldn't return error at all. Link: https://lore.kernel.org/r/20200907120921.476363-10-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 23209bca8c1b..29a36ec662b5 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2498,7 +2498,7 @@ struct ib_device_ops { struct uverbs_attr_bundle *attrs); int (*create_counters)(struct ib_counters *counters, struct uverbs_attr_bundle *attrs); - void (*destroy_counters)(struct ib_counters *counters); + int (*destroy_counters)(struct ib_counters *counters); int (*read_counters)(struct ib_counters *counters, struct ib_counters_read_attr *counters_read_attr, struct uverbs_attr_bundle *attrs); -- cgit v1.2.3 From 3361c29e9279e682c0e9a7d7461b4e3bbc77830b Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 4 Sep 2020 19:41:44 -0300 Subject: RDMA/umem: Use simpler logic for ib_umem_find_best_pgsz() The calculation in rdma_find_pg_bit() is fairly complicated, and the function is never called anywhere else. Inline a simpler version into ib_umem_find_best_pgsz() Link: https://lore.kernel.org/r/3-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 24 ------------------------ 1 file changed, 24 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 29a36ec662b5..97658d706aa1 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -3335,30 +3335,6 @@ static inline bool rdma_cap_read_inv(struct ib_device *dev, u32 port_num) return rdma_protocol_iwarp(dev, port_num); } -/** - * rdma_find_pg_bit - Find page bit given address and HW supported page sizes - * - * @addr: address - * @pgsz_bitmap: bitmap of HW supported page sizes - */ -static inline unsigned int rdma_find_pg_bit(unsigned long addr, - unsigned long pgsz_bitmap) -{ - unsigned long align; - unsigned long pgsz; - - align = addr & -addr; - - /* Find page bit such that addr is aligned to the highest supported - * HW page size - */ - pgsz = pgsz_bitmap & ~(-align << 1); - if (!pgsz) - return __ffs(pgsz_bitmap); - - return __fls(pgsz); -} - /** * rdma_core_cap_opa_port - Return whether the RDMA Port is OPA or not. * @device: Device -- cgit v1.2.3 From ebc24096c4c40017d9f9b0fddc5d69b94ad10369 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 4 Sep 2020 19:41:45 -0300 Subject: RDMA/umem: Add rdma_umem_for_each_dma_block() This helper does the same as rdma_for_each_block(), except it works on a umem. This simplifies most of the call sites. Link: https://lore.kernel.org/r/4-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Acked-by: Miguel Ojeda Acked-by: Shiraz Saleem Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'include') diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 07a764eb692e..b880512ba95f 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -40,6 +40,26 @@ static inline size_t ib_umem_num_pages(struct ib_umem *umem) PAGE_SHIFT; } +static inline void __rdma_umem_block_iter_start(struct ib_block_iter *biter, + struct ib_umem *umem, + unsigned long pgsz) +{ + __rdma_block_iter_start(biter, umem->sg_head.sgl, umem->nmap, pgsz); +} + +/** + * rdma_umem_for_each_dma_block - iterate over contiguous DMA blocks of the umem + * @umem: umem to iterate over + * @pgsz: Page size to split the list into + * + * pgsz must be <= PAGE_SIZE or computed by ib_umem_find_best_pgsz(). The + * returned DMA blocks will be aligned to pgsz and span the range: + * ALIGN_DOWN(umem->address, pgsz) to ALIGN(umem->address + umem->length, pgsz) + */ +#define rdma_umem_for_each_dma_block(umem, biter, pgsz) \ + for (__rdma_umem_block_iter_start(biter, umem, pgsz); \ + __rdma_block_iter_next(biter);) + #ifdef CONFIG_INFINIBAND_USER_MEM struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, -- cgit v1.2.3 From a665aca89a411115e35ea937c2d3fb2ee4f5a701 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 4 Sep 2020 19:41:47 -0300 Subject: RDMA/umem: Split ib_umem_num_pages() into ib_umem_num_dma_blocks() ib_umem_num_pages() should only be used by things working with the SGL in CPU pages directly. Drivers building DMA lists should use the new ib_num_dma_blocks() which returns the number of blocks rdma_umem_for_each_block() will return. To make this general for DMA drivers requires a different implementation. Computing DMA block count based on umem->address only works if the requested page size is < PAGE_SIZE and/or the IOVA == umem->address. Instead the number of DMA pages should be computed in the IOVA address space, not umem->address. Thus the IOVA has to be stored inside the umem so it can be used for these calculations. For now set it to umem->address by default and fix it up if ib_umem_find_best_pgsz() was called. This allows drivers to be converted to ib_umem_num_dma_blocks() safely. Link: https://lore.kernel.org/r/6-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem.h | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index b880512ba95f..cf001d4e0a4f 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -17,6 +17,7 @@ struct ib_umem_odp; struct ib_umem { struct ib_device *ibdev; struct mm_struct *owning_mm; + u64 iova; size_t length; unsigned long address; u32 writable : 1; @@ -33,11 +34,17 @@ static inline int ib_umem_offset(struct ib_umem *umem) return umem->address & ~PAGE_MASK; } +static inline size_t ib_umem_num_dma_blocks(struct ib_umem *umem, + unsigned long pgsz) +{ + return (size_t)((ALIGN(umem->iova + umem->length, pgsz) - + ALIGN_DOWN(umem->iova, pgsz))) / + pgsz; +} + static inline size_t ib_umem_num_pages(struct ib_umem *umem) { - return (ALIGN(umem->address + umem->length, PAGE_SIZE) - - ALIGN_DOWN(umem->address, PAGE_SIZE)) >> - PAGE_SHIFT; + return ib_umem_num_dma_blocks(umem, PAGE_SIZE); } static inline void __rdma_umem_block_iter_start(struct ib_block_iter *biter, @@ -55,6 +62,8 @@ static inline void __rdma_umem_block_iter_start(struct ib_block_iter *biter, * pgsz must be <= PAGE_SIZE or computed by ib_umem_find_best_pgsz(). The * returned DMA blocks will be aligned to pgsz and span the range: * ALIGN_DOWN(umem->address, pgsz) to ALIGN(umem->address + umem->length, pgsz) + * + * Performs exactly ib_umem_num_dma_blocks() iterations. */ #define rdma_umem_for_each_dma_block(umem, biter, pgsz) \ for (__rdma_umem_block_iter_start(biter, umem, pgsz); \ -- cgit v1.2.3 From 81655d3c4a95f05495f3213acbbc5423fc4dbdfb Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 4 Sep 2020 19:41:56 -0300 Subject: RDMA/mlx4: Use ib_umem_num_dma_blocks() For the calls linked to mlx4_ib_umem_calc_optimal_mtt_size() use ib_umem_num_dma_blocks() inside the function, it is just some weird static default. All other places are just using it with PAGE_SIZE, switch to ib_umem_num_dma_blocks(). As this is the last call site, remove ib_umem_num_count(). Link: https://lore.kernel.org/r/15-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index cf001d4e0a4f..70597508c765 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -74,7 +74,6 @@ static inline void __rdma_umem_block_iter_start(struct ib_block_iter *biter, struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, size_t size, int access); void ib_umem_release(struct ib_umem *umem); -int ib_umem_page_count(struct ib_umem *umem); int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset, size_t length); unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem, @@ -92,7 +91,6 @@ static inline struct ib_umem *ib_umem_get(struct ib_device *device, return ERR_PTR(-EINVAL); } static inline void ib_umem_release(struct ib_umem *umem) { } -static inline int ib_umem_page_count(struct ib_umem *umem) { return 0; } static inline int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset, size_t length) { return -EINVAL; -- cgit v1.2.3 From b3003a74456f0c1f614a46c07e16abe33bfdd087 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 4 Sep 2020 19:41:57 -0300 Subject: RDMA/qedr: Remove fbo and zbva from the MR MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit zbva is always false, so fbo is never read. A 'zero-based-virtual-address' is simply IOVA == 0, and the driver already supports this. Link: https://lore.kernel.org/r/16-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Acked-by: Michal Kalderon  Signed-off-by: Jason Gunthorpe --- include/linux/qed/qed_rdma_if.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/qed/qed_rdma_if.h b/include/linux/qed/qed_rdma_if.h index f464d85e88a4..aeb242cefebf 100644 --- a/include/linux/qed/qed_rdma_if.h +++ b/include/linux/qed/qed_rdma_if.h @@ -242,10 +242,8 @@ struct qed_rdma_register_tid_in_params { bool pbl_two_level; u8 pbl_page_size_log; u8 page_size_log; - u32 fbo; u64 length; u64 vaddr; - bool zbva; bool phy_mr; bool dma_mr; -- cgit v1.2.3 From b60b9c0274007fe68439cd9cefdabbd7fb8a2ce6 Mon Sep 17 00:00:00 2001 From: Bob Pearson Date: Thu, 3 Sep 2020 17:40:34 -0500 Subject: RDMA/core: Added missing WR and WC opcodes Add work completion opcodes to a new ib_uverbs_wc_opcode enum in ib_user_verbs.h. This plays the same role as ib_uverbs_wr_opcode documenting the opcodes in the user space API. Assigned the IB_WC_XXX opcodes in ib_verbs.h to the IB_UVERBS_WC_XXX where they are defined. This follows the same pattern as the IB_WR_XXX opcodes. This fixes an incorrect value for LSO that had crept in but is not currently being used. Added a missing IB_WR_BIND_MW opcode in ib_verbs.h. Link: https://lore.kernel.org/r/20200903224039.437391-2-rpearson@hpe.com Signed-off-by: Bob Pearson Reviewed-by: Jason Gunthorpe Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 16 +++++++++------- include/uapi/rdma/ib_user_verbs.h | 11 +++++++++++ 2 files changed, 20 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 97658d706aa1..5ae1d9849881 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -952,13 +952,14 @@ enum ib_wc_status { const char *__attribute_const__ ib_wc_status_msg(enum ib_wc_status status); enum ib_wc_opcode { - IB_WC_SEND, - IB_WC_RDMA_WRITE, - IB_WC_RDMA_READ, - IB_WC_COMP_SWAP, - IB_WC_FETCH_ADD, - IB_WC_LSO, - IB_WC_LOCAL_INV, + IB_WC_SEND = IB_UVERBS_WC_SEND, + IB_WC_RDMA_WRITE = IB_UVERBS_WC_RDMA_WRITE, + IB_WC_RDMA_READ = IB_UVERBS_WC_RDMA_READ, + IB_WC_COMP_SWAP = IB_UVERBS_WC_COMP_SWAP, + IB_WC_FETCH_ADD = IB_UVERBS_WC_FETCH_ADD, + IB_WC_BIND_MW = IB_UVERBS_WC_BIND_MW, + IB_WC_LOCAL_INV = IB_UVERBS_WC_LOCAL_INV, + IB_WC_LSO = IB_UVERBS_WC_TSO, IB_WC_REG_MR, IB_WC_MASKED_COMP_SWAP, IB_WC_MASKED_FETCH_ADD, @@ -1291,6 +1292,7 @@ enum ib_wr_opcode { IB_WR_RDMA_READ = IB_UVERBS_WR_RDMA_READ, IB_WR_ATOMIC_CMP_AND_SWP = IB_UVERBS_WR_ATOMIC_CMP_AND_SWP, IB_WR_ATOMIC_FETCH_AND_ADD = IB_UVERBS_WR_ATOMIC_FETCH_AND_ADD, + IB_WR_BIND_MW = IB_UVERBS_WR_BIND_MW, IB_WR_LSO = IB_UVERBS_WR_TSO, IB_WR_SEND_WITH_INV = IB_UVERBS_WR_SEND_WITH_INV, IB_WR_RDMA_READ_WITH_INV = IB_UVERBS_WR_RDMA_READ_WITH_INV, diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index 0474c7400268..456438c18c2c 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -457,6 +457,17 @@ struct ib_uverbs_poll_cq { __u32 ne; }; +enum ib_uverbs_wc_opcode { + IB_UVERBS_WC_SEND = 0, + IB_UVERBS_WC_RDMA_WRITE = 1, + IB_UVERBS_WC_RDMA_READ = 2, + IB_UVERBS_WC_COMP_SWAP = 3, + IB_UVERBS_WC_FETCH_ADD = 4, + IB_UVERBS_WC_BIND_MW = 5, + IB_UVERBS_WC_LOCAL_INV = 6, + IB_UVERBS_WC_TSO = 7, +}; + struct ib_uverbs_wc { __aligned_u64 wr_id; __u32 status; -- cgit v1.2.3 From 97fb3e33474897c35b7cfe6ff0e16a3e172d9249 Mon Sep 17 00:00:00 2001 From: Michal Kalderon Date: Wed, 2 Sep 2020 19:57:38 +0300 Subject: qede: Notify qedr when mtu has changed MTU change on ethtool is currently not supported for iWARP. Notify qedr driver so that appropriate logging can take place. Link: https://lore.kernel.org/r/20200902165741.8355-6-michal.kalderon@marvell.com Signed-off-by: Michal Kalderon Signed-off-by: Jason Gunthorpe --- include/linux/qed/qede_rdma.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/qed/qede_rdma.h b/include/linux/qed/qede_rdma.h index 072da2f6da37..0d5564a59a59 100644 --- a/include/linux/qed/qede_rdma.h +++ b/include/linux/qed/qede_rdma.h @@ -20,7 +20,8 @@ enum qede_rdma_event { QEDE_UP, QEDE_DOWN, QEDE_CHANGE_ADDR, - QEDE_CLOSE + QEDE_CLOSE, + QEDE_CHANGE_MTU, }; struct qede_rdma_event_work { @@ -54,6 +55,7 @@ void qede_rdma_dev_event_open(struct qede_dev *dev); void qede_rdma_dev_event_close(struct qede_dev *dev); void qede_rdma_dev_remove(struct qede_dev *dev, bool recovery); void qede_rdma_event_changeaddr(struct qede_dev *edr); +void qede_rdma_event_change_mtu(struct qede_dev *edev); #else static inline int qede_rdma_dev_add(struct qede_dev *dev, -- cgit v1.2.3 From 639bf4415cadff4c18e13aa5cb0dba2d443e3aa7 Mon Sep 17 00:00:00 2001 From: Aharon Landau Date: Thu, 17 Sep 2020 12:02:21 +0300 Subject: net/mlx5: Refactor query port speed functions The functions mlx5_query_port_link_width_oper and mlx5_query_port_ib_proto_oper are always called together, so combine them to a new function called mlx5_query_port_oper to avoid duplication. And while the mlx5i_get_port_settings is the same as mlx5_query_port_oper therefore let's remove it. According to the IB spec link_width_oper and ib_proto_oper should be u16 and not as written u8, so perform casting as a preparation to cross-RDMA patch which will fix that type for all drivers in the RDMA subsystem. Fixes: ada68c31ba9c ("net/mlx5: Introduce a new header file for physical port functions") Signed-off-by: Aharon Landau Reviewed-by: Michael Guralnik Signed-off-by: Leon Romanovsky --- include/linux/mlx5/port.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 2d45a6af52a4..4d33ae0c2d97 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -133,10 +133,9 @@ enum mlx5e_connector_type { int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps); int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys, int ptys_size, int proto_mask, u8 local_port); -int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev, - u8 *link_width_oper, u8 local_port); -int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev, - u8 *proto_oper, u8 local_port); + +int mlx5_query_ib_port_oper(struct mlx5_core_dev *dev, u16 *link_width_oper, + u16 *proto_oper, u8 local_port); void mlx5_toggle_port_link(struct mlx5_core_dev *dev); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status); -- cgit v1.2.3 From e27014bdb47eb435f78573685f4196c07329f1f7 Mon Sep 17 00:00:00 2001 From: Aharon Landau Date: Thu, 17 Sep 2020 12:02:22 +0300 Subject: RDMA/mlx5: Delete duplicated mlx5_ptys_width enum Combine two same enums to avoid duplication. Signed-off-by: Aharon Landau Reviewed-by: Michael Guralnik Signed-off-by: Leon Romanovsky --- include/linux/mlx5/port.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include') diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 4d33ae0c2d97..23edd2db4803 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -125,6 +125,14 @@ enum mlx5e_connector_type { MLX5E_CONNECTOR_TYPE_NUMBER, }; +enum mlx5_ptys_width { + MLX5_PTYS_WIDTH_1X = 1 << 0, + MLX5_PTYS_WIDTH_2X = 1 << 1, + MLX5_PTYS_WIDTH_4X = 1 << 2, + MLX5_PTYS_WIDTH_8X = 1 << 3, + MLX5_PTYS_WIDTH_12X = 1 << 4, +}; + #define MLX5E_PROT_MASK(link_mode) (1 << link_mode) #define MLX5_GET_ETH_PROTO(reg, out, ext, field) \ (ext ? MLX5_GET(reg, out, ext_##field) : \ -- cgit v1.2.3 From d18bb3e15201918b8d07e85a6e010ca5ed28dad5 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Wed, 2 Sep 2020 11:16:22 +0300 Subject: RDMA: Clean MW allocation and free flows Move allocation and destruction of memory windows under ib_core responsibility and clean drivers to ensure that no updates to MW ib_core structures are done in driver layer. Link: https://lore.kernel.org/r/20200902081623.746359-2-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 5ae1d9849881..a292647c3db3 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2448,8 +2448,7 @@ struct ib_device_ops { unsigned int *sg_offset); int (*check_mr_status)(struct ib_mr *mr, u32 check_mask, struct ib_mr_status *mr_status); - struct ib_mw *(*alloc_mw)(struct ib_pd *pd, enum ib_mw_type type, - struct ib_udata *udata); + int (*alloc_mw)(struct ib_mw *mw, struct ib_udata *udata); int (*dealloc_mw)(struct ib_mw *mw); int (*attach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); int (*detach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); @@ -2610,6 +2609,7 @@ struct ib_device_ops { DECLARE_RDMA_OBJ_SIZE(ib_ah); DECLARE_RDMA_OBJ_SIZE(ib_counters); DECLARE_RDMA_OBJ_SIZE(ib_cq); + DECLARE_RDMA_OBJ_SIZE(ib_mw); DECLARE_RDMA_OBJ_SIZE(ib_pd); DECLARE_RDMA_OBJ_SIZE(ib_srq); DECLARE_RDMA_OBJ_SIZE(ib_ucontext); -- cgit v1.2.3 From c0a6b5ecc5b7dd028c2921415ea036074a8f8b00 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Wed, 2 Sep 2020 11:16:23 +0300 Subject: RDMA: Convert RWQ table logic to ib_core allocation scheme Move struct ib_rwq_ind_table allocation to ib_core. Link: https://lore.kernel.org/r/20200902081623.746359-3-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index a292647c3db3..fc9a55c140fa 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2484,10 +2484,9 @@ struct ib_device_ops { int (*destroy_wq)(struct ib_wq *wq, struct ib_udata *udata); int (*modify_wq)(struct ib_wq *wq, struct ib_wq_attr *attr, u32 wq_attr_mask, struct ib_udata *udata); - struct ib_rwq_ind_table *(*create_rwq_ind_table)( - struct ib_device *device, - struct ib_rwq_ind_table_init_attr *init_attr, - struct ib_udata *udata); + int (*create_rwq_ind_table)(struct ib_rwq_ind_table *ib_rwq_ind_table, + struct ib_rwq_ind_table_init_attr *init_attr, + struct ib_udata *udata); int (*destroy_rwq_ind_table)(struct ib_rwq_ind_table *wq_ind_table); struct ib_dm *(*alloc_dm)(struct ib_device *device, struct ib_ucontext *context, @@ -2611,6 +2610,7 @@ struct ib_device_ops { DECLARE_RDMA_OBJ_SIZE(ib_cq); DECLARE_RDMA_OBJ_SIZE(ib_mw); DECLARE_RDMA_OBJ_SIZE(ib_pd); + DECLARE_RDMA_OBJ_SIZE(ib_rwq_ind_table); DECLARE_RDMA_OBJ_SIZE(ib_srq); DECLARE_RDMA_OBJ_SIZE(ib_ucontext); DECLARE_RDMA_OBJ_SIZE(ib_xrcd); @@ -4297,7 +4297,6 @@ struct ib_wq *ib_create_wq(struct ib_pd *pd, int ib_destroy_wq_user(struct ib_wq *wq, struct ib_udata *udata); int ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *attr, u32 wq_attr_mask); -int ib_destroy_rwq_ind_table(struct ib_rwq_ind_table *wq_ind_table); int ib_map_mr_sg(struct ib_mr *mr, struct scatterlist *sg, int sg_nents, unsigned int *sg_offset, unsigned int page_size); -- cgit v1.2.3 From 9d8feb460adb90e89e64101ce2c1cfcd548a6d83 Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Mon, 31 Aug 2020 10:01:48 +0300 Subject: RDMA/mlx5: Add sw_owner_v2 bit capability Added sw_owner_v2 which will be enabled for future devices, replacing sw_owner bit. Signed-off-by: Alex Vesker Signed-off-by: Leon Romanovsky --- include/linux/mlx5/mlx5_ifc.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index de1ffb4804d6..3061ceebbaf3 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -420,7 +420,8 @@ struct mlx5_ifc_flow_table_prop_layout_bits { u8 reserved_at_1a[0x2]; u8 ipsec_encrypt[0x1]; u8 ipsec_decrypt[0x1]; - u8 reserved_at_1e[0x2]; + u8 sw_owner_v2[0x1]; + u8 reserved_at_1f[0x1]; u8 termination_table_raw_traffic[0x1]; u8 reserved_at_21[0x1]; -- cgit v1.2.3 From 376ceb31ff879c2ee6b48eef841d6fa7720f6f43 Mon Sep 17 00:00:00 2001 From: Aharon Landau Date: Thu, 17 Sep 2020 12:02:23 +0300 Subject: RDMA: Fix link active_speed size According to the IB spec active_speed size should be u16 and not u8 as before. Changing it to allow further extensions in offered speeds. Link: https://lore.kernel.org/r/20200917090223.1018224-4-leon@kernel.org Signed-off-by: Aharon Landau Reviewed-by: Michael Guralnik Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index c0b2fa7e9b95..a7e203bcb012 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -535,7 +535,8 @@ enum ib_port_speed { IB_SPEED_FDR10 = 8, IB_SPEED_FDR = 16, IB_SPEED_EDR = 32, - IB_SPEED_HDR = 64 + IB_SPEED_HDR = 64, + IB_SPEED_NDR = 128, }; /** @@ -669,7 +670,7 @@ struct ib_port_attr { u8 subnet_timeout; u8 init_type_reply; u8 active_width; - u8 active_speed; + u16 active_speed; u8 phys_state; u16 port_cap_flags2; }; @@ -4410,7 +4411,7 @@ void ib_drain_rq(struct ib_qp *qp); void ib_drain_sq(struct ib_qp *qp); void ib_drain_qp(struct ib_qp *qp); -int ib_get_eth_speed(struct ib_device *dev, u8 port_num, u8 *speed, u8 *width); +int ib_get_eth_speed(struct ib_device *dev, u8 port_num, u16 *speed, u8 *width); static inline u8 *rdma_ah_retrieve_dmac(struct rdma_ah_attr *attr) { -- cgit v1.2.3 From 13ef5539def732dc7b9c58c320d97a0a95b52634 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Tue, 22 Sep 2020 12:11:04 +0300 Subject: RDMA/restrack: Count references to the verbs objects Refactor the restrack code to make sure the kref inside the restrack entry properly kref's the object in which it is embedded. This slight change is needed for future conversions of MR and QP which are refcounted before the release and kfree. The ideal flow from ib_core perspective as follows: * Allocate ib_* structure with rdma_zalloc_*. * Set everything that is known to ib_core to that newly created object. * Initialize kref with restrack help * Call to driver specific allocation functions. * Insert into restrack DB .... * Return and release restrack with restrack_put. Largely this means a rdma_restrack_new() should be called near allocating the containing structure. Link: https://lore.kernel.org/r/20200922091106.2152715-4-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/restrack.h | 7 ------- 1 file changed, 7 deletions(-) (limited to 'include') diff --git a/include/rdma/restrack.h b/include/rdma/restrack.h index 7682d1bcf789..d7c166237939 100644 --- a/include/rdma/restrack.h +++ b/include/rdma/restrack.h @@ -110,13 +110,6 @@ int rdma_restrack_count(struct ib_device *dev, void rdma_restrack_kadd(struct rdma_restrack_entry *res); void rdma_restrack_uadd(struct rdma_restrack_entry *res); -/** - * rdma_restrack_del() - delete object from the reource tracking database - * @res: resource entry - * @type: actual type of object to operate - */ -void rdma_restrack_del(struct rdma_restrack_entry *res); - /** * rdma_is_kernel_res() - check the owner of resource * @res: resource entry -- cgit v1.2.3 From c34a23c28c6b0045b1f21649de30f68da72547af Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Tue, 22 Sep 2020 12:11:05 +0300 Subject: RDMA/restrack: Simplify restrack tracking in kernel flows Have a single rdma_restrack_add() that adds an entry, there is no reason to split the user/kernel here, the rdma_restrack_set_task() is responsible for this difference. This patch prepares the code to the future requirement of making restrack is mandatory for managing ib objects. Link: https://lore.kernel.org/r/20200922091106.2152715-5-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/restrack.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/rdma/restrack.h b/include/rdma/restrack.h index d7c166237939..db59e208f5e8 100644 --- a/include/rdma/restrack.h +++ b/include/rdma/restrack.h @@ -107,7 +107,6 @@ struct rdma_restrack_entry { int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type); -void rdma_restrack_kadd(struct rdma_restrack_entry *res); void rdma_restrack_uadd(struct rdma_restrack_entry *res); /** -- cgit v1.2.3 From b09c4d70122091c1865cb63a9c4dad1a94a8e339 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Tue, 22 Sep 2020 12:11:06 +0300 Subject: RDMA/restrack: Improve readability in task name management Use rdma_restrack_set_name() and rdma_restrack_parent_name() instead of tricky uses of rdma_restrack_attach_task()/rdma_restrack_uadd(). This uniformly makes all restracks add'd using rdma_restrack_add(). Link: https://lore.kernel.org/r/20200922091106.2152715-6-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/rdma_cm.h | 47 ++++++++++++++--------------------------------- include/rdma/restrack.h | 13 +------------ 2 files changed, 15 insertions(+), 45 deletions(-) (limited to 'include') diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index c1334c9a7aa8..c672ae1da26b 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -110,11 +110,14 @@ struct rdma_cm_id { u8 port_num; }; -struct rdma_cm_id *__rdma_create_id(struct net *net, - rdma_cm_event_handler event_handler, - void *context, enum rdma_ucm_port_space ps, - enum ib_qp_type qp_type, - const char *caller); +struct rdma_cm_id * +__rdma_create_kernel_id(struct net *net, rdma_cm_event_handler event_handler, + void *context, enum rdma_ucm_port_space ps, + enum ib_qp_type qp_type, const char *caller); +struct rdma_cm_id *rdma_create_user_id(rdma_cm_event_handler event_handler, + void *context, + enum rdma_ucm_port_space ps, + enum ib_qp_type qp_type); /** * rdma_create_id - Create an RDMA identifier. @@ -132,9 +135,9 @@ struct rdma_cm_id *__rdma_create_id(struct net *net, * The event handler callback serializes on the id's mutex and is * allowed to sleep. */ -#define rdma_create_id(net, event_handler, context, ps, qp_type) \ - __rdma_create_id((net), (event_handler), (context), (ps), (qp_type), \ - KBUILD_MODNAME) +#define rdma_create_id(net, event_handler, context, ps, qp_type) \ + __rdma_create_kernel_id(net, event_handler, context, ps, qp_type, \ + KBUILD_MODNAME) /** * rdma_destroy_id - Destroys an RDMA identifier. @@ -250,34 +253,12 @@ int rdma_connect_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, */ int rdma_listen(struct rdma_cm_id *id, int backlog); -int __rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, - const char *caller); +int rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param); void rdma_lock_handler(struct rdma_cm_id *id); void rdma_unlock_handler(struct rdma_cm_id *id); -int __rdma_accept_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, - const char *caller, struct rdma_ucm_ece *ece); - -/** - * rdma_accept - Called to accept a connection request or response. - * @id: Connection identifier associated with the request. - * @conn_param: Information needed to establish the connection. This must be - * provided if accepting a connection request. If accepting a connection - * response, this parameter must be NULL. - * - * Typically, this routine is only called by the listener to accept a connection - * request. It must also be called on the active side of a connection if the - * user is performing their own QP transitions. - * - * In the case of error, a reject message is sent to the remote side and the - * state of the qp associated with the id is modified to error, such that any - * previously posted receive buffers would be flushed. - * - * This function is for use by kernel ULPs and must be called from under the - * handler callback. - */ -#define rdma_accept(id, conn_param) \ - __rdma_accept((id), (conn_param), KBUILD_MODNAME) +int rdma_accept_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, + struct rdma_ucm_ece *ece); /** * rdma_notify - Notifies the RDMA CM of an asynchronous event that has diff --git a/include/rdma/restrack.h b/include/rdma/restrack.h index db59e208f5e8..d3a1cc5be7bc 100644 --- a/include/rdma/restrack.h +++ b/include/rdma/restrack.h @@ -106,14 +106,11 @@ struct rdma_restrack_entry { int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type); - -void rdma_restrack_uadd(struct rdma_restrack_entry *res); - /** * rdma_is_kernel_res() - check the owner of resource * @res: resource entry */ -static inline bool rdma_is_kernel_res(struct rdma_restrack_entry *res) +static inline bool rdma_is_kernel_res(const struct rdma_restrack_entry *res) { return !res->user; } @@ -130,14 +127,6 @@ int __must_check rdma_restrack_get(struct rdma_restrack_entry *res); */ int rdma_restrack_put(struct rdma_restrack_entry *res); -/** - * rdma_restrack_set_task() - set the task for this resource - * @res: resource entry - * @caller: kernel name, the current task will be used if the caller is NULL. - */ -void rdma_restrack_set_task(struct rdma_restrack_entry *res, - const char *caller); - /* * Helper functions for rdma drivers when filling out * nldev driver attributes. -- cgit v1.2.3 From 09a5f210f67eea4a2176820c3bc398747a564705 Mon Sep 17 00:00:00 2001 From: Wenpeng Liang Date: Wed, 16 Sep 2020 16:43:24 +0800 Subject: RDMA/hns: Add support for CQE in size of 64 Bytes The new version of RoCEE supports using CQE in size of 32B or 64B. The performance of bus can be improved by using larger size of CQE. Link: https://lore.kernel.org/r/1600245806-56321-3-git-send-email-liweihang@huawei.com Signed-off-by: Wenpeng Liang Signed-off-by: Weihang Li Signed-off-by: Jason Gunthorpe --- include/uapi/rdma/hns-abi.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/rdma/hns-abi.h b/include/uapi/rdma/hns-abi.h index eb76b38a00d4..9ec85f76e9ac 100644 --- a/include/uapi/rdma/hns-abi.h +++ b/include/uapi/rdma/hns-abi.h @@ -39,6 +39,8 @@ struct hns_roce_ib_create_cq { __aligned_u64 buf_addr; __aligned_u64 db_addr; + __u32 cqe_size; + __u32 reserved; }; struct hns_roce_ib_create_cq_resp { @@ -73,7 +75,7 @@ struct hns_roce_ib_create_qp_resp { struct hns_roce_ib_alloc_ucontext_resp { __u32 qp_tab_size; - __u32 reserved; + __u32 cqe_size; }; struct hns_roce_ib_alloc_pd_resp { -- cgit v1.2.3 From a6f0b08dbaf289c3c57284e16ac8043140f2139b Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 29 Sep 2020 13:09:07 -0300 Subject: RDMA/core: Remove ucontext->closing Nothing reads this any more, and the reason for its existence has passed due to the deferred fput() scheme. Fixes: 8ea1f989aa07 ("drivers/IB,usnic: reduce scope of mmap_sem") Link: https://lore.kernel.org/r/0-v1-df64ff042436+42-uctx_closing_jgg@nvidia.com Reviewed-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index f18502984e6f..5ad997346f7f 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1477,12 +1477,6 @@ struct ib_rdmacg_object { struct ib_ucontext { struct ib_device *device; struct ib_uverbs_file *ufile; - /* - * 'closing' can be read by the driver only during a destroy callback, - * it is set when we are closing the file descriptor and indicates - * that mm_sem may be locked. - */ - bool closing; bool cleanup_retryable; -- cgit v1.2.3 From 36f30e486dce22345c2dd3a3ba439c12cd67f6ba Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Wed, 30 Sep 2020 19:38:25 +0300 Subject: IB/core: Improve ODP to use hmm_range_fault() Move to use hmm_range_fault() instead of get_user_pags_remote() to improve performance in a few aspects: This includes: - Dropping the need to allocate and free memory to hold its output - No need any more to use put_page() to unpin the pages - The logic to detect contiguous pages is done based on the returned order, no need to run per page and evaluate. In addition, moving to use hmm_range_fault() enables to reduce page faults in the system with it's snapshot mode, this will be introduced in next patches from this series. As part of this, cleanup some flows and use the required data structures to work with hmm_range_fault(). Link: https://lore.kernel.org/r/20200930163828.1336747-2-leon@kernel.org Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem_odp.h | 21 ++++++++------------- 1 file changed, 8 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index d16d2c17e733..a53b62ac8a9d 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -14,17 +14,13 @@ struct ib_umem_odp { struct mmu_interval_notifier notifier; struct pid *tgid; + /* An array of the pfns included in the on-demand paging umem. */ + unsigned long *pfn_list; + /* - * An array of the pages included in the on-demand paging umem. - * Indices of pages that are currently not mapped into the device will - * contain NULL. - */ - struct page **page_list; - /* - * An array of the same size as page_list, with DMA addresses mapped - * for pages the pages in page_list. The lower two bits designate - * access permissions. See ODP_READ_ALLOWED_BIT and - * ODP_WRITE_ALLOWED_BIT. + * An array with DMA addresses mapped for pfns in pfn_list. + * The lower two bits designate access permissions. + * See ODP_READ_ALLOWED_BIT and ODP_WRITE_ALLOWED_BIT. */ dma_addr_t *dma_list; /* @@ -97,9 +93,8 @@ ib_umem_odp_alloc_child(struct ib_umem_odp *root_umem, unsigned long addr, const struct mmu_interval_notifier_ops *ops); void ib_umem_odp_release(struct ib_umem_odp *umem_odp); -int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset, - u64 bcnt, u64 access_mask, - unsigned long current_seq); +int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 start_offset, + u64 bcnt, u64 access_mask); void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset, u64 bound); -- cgit v1.2.3 From 8bfafde08693cc2e740134c3e32b40254f5d82ae Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Wed, 30 Sep 2020 19:38:26 +0300 Subject: IB/core: Enable ODP sync without faulting Enable ODP sync without faulting, this improves performance by reducing the number of page faults in the system. The gain from this option is that the device page table can be aligned with the presented pages in the CPU page table without causing page faults. As of that, the overhead on data path from hardware point of view to trigger a fault which end-up by calling the driver to bring the pages will be dropped. Link: https://lore.kernel.org/r/20200930163828.1336747-3-leon@kernel.org Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_umem_odp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index a53b62ac8a9d..0844c1d05ac6 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -94,7 +94,7 @@ ib_umem_odp_alloc_child(struct ib_umem_odp *root_umem, unsigned long addr, void ib_umem_odp_release(struct ib_umem_odp *umem_odp); int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 start_offset, - u64 bcnt, u64 access_mask); + u64 bcnt, u64 access_mask, bool fault); void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset, u64 bound); -- cgit v1.2.3 From 677cf51f71c97bcf98852aa2077d7289bc73e3b3 Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Wed, 30 Sep 2020 19:38:27 +0300 Subject: RDMA/mlx5: Extend advice MR to support non faulting mode Extend advice MR to support non faulting mode, this can improve performance by increasing the populated page tables in the device. Link: https://lore.kernel.org/r/20200930163828.1336747-4-leon@kernel.org Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/uapi/rdma/ib_user_ioctl_verbs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/uapi/rdma/ib_user_ioctl_verbs.h b/include/uapi/rdma/ib_user_ioctl_verbs.h index 5debab45ebcb..fb8cdb38198b 100644 --- a/include/uapi/rdma/ib_user_ioctl_verbs.h +++ b/include/uapi/rdma/ib_user_ioctl_verbs.h @@ -208,6 +208,7 @@ enum ib_uverbs_read_counters_flags { enum ib_uverbs_advise_mr_advice { IB_UVERBS_ADVISE_MR_ADVICE_PREFETCH, IB_UVERBS_ADVISE_MR_ADVICE_PREFETCH_WRITE, + IB_UVERBS_ADVISE_MR_ADVICE_PREFETCH_NO_FAULT, }; enum ib_uverbs_advise_mr_flag { -- cgit v1.2.3 From a4947e84f23474803b62a2759b5808147e4e15f9 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sun, 13 Sep 2020 13:29:28 +0300 Subject: overflow: Include header file with SIZE_MAX declaration The various array_size functions use SIZE_MAX define, but missed limits.h causes to failure to compile code that needs overflow.h. In file included from drivers/infiniband/core/uverbs_std_types_device.c:6: ./include/linux/overflow.h: In function 'array_size': ./include/linux/overflow.h:258:10: error: 'SIZE_MAX' undeclared (first use in this function) 258 | return SIZE_MAX; | ^~~~~~~~ Fixes: 610b15c50e86 ("overflow.h: Add allocation size calculation helpers") Link: https://lore.kernel.org/r/20200913102928.134985-1-leon@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/linux/overflow.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/overflow.h b/include/linux/overflow.h index 93fcef105061..ff3c48f0abc5 100644 --- a/include/linux/overflow.h +++ b/include/linux/overflow.h @@ -3,6 +3,7 @@ #define __LINUX_OVERFLOW_H #include +#include /* * In the fallback code below, we need to compute the minimum and -- cgit v1.2.3 From 1c15b4f2a42ff6697767c22c8ff5f9bcc22fdbe5 Mon Sep 17 00:00:00 2001 From: Avihai Horon Date: Wed, 23 Sep 2020 19:50:13 +0300 Subject: RDMA/core: Modify enum ib_gid_type and enum rdma_network_type Separate IB_GID_TYPE_IB and IB_GID_TYPE_ROCE to two different values, so enum ib_gid_type will match the gid types of the new query GID table API which will be introduced in the following patches. This change in enum ib_gid_type requires to change also enum rdma_network_type by separating RDMA_NETWORK_IB and RDMA_NETWORK_ROCE_V1 values. Link: https://lore.kernel.org/r/20200923165015.2491894-3-leon@kernel.org Signed-off-by: Avihai Horon Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 5ad997346f7f..3b61fba531d0 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -138,10 +138,9 @@ union ib_gid { extern union ib_gid zgid; enum ib_gid_type { - /* If link layer is Ethernet, this is RoCE V1 */ IB_GID_TYPE_IB = 0, - IB_GID_TYPE_ROCE = 0, - IB_GID_TYPE_ROCE_UDP_ENCAP = 1, + IB_GID_TYPE_ROCE = 1, + IB_GID_TYPE_ROCE_UDP_ENCAP = 2, IB_GID_TYPE_SIZE }; @@ -180,7 +179,7 @@ rdma_node_get_transport(unsigned int node_type); enum rdma_network_type { RDMA_NETWORK_IB, - RDMA_NETWORK_ROCE_V1 = RDMA_NETWORK_IB, + RDMA_NETWORK_ROCE_V1, RDMA_NETWORK_IPV4, RDMA_NETWORK_IPV6 }; @@ -190,9 +189,10 @@ static inline enum ib_gid_type ib_network_to_gid_type(enum rdma_network_type net if (network_type == RDMA_NETWORK_IPV4 || network_type == RDMA_NETWORK_IPV6) return IB_GID_TYPE_ROCE_UDP_ENCAP; - - /* IB_GID_TYPE_IB same as RDMA_NETWORK_ROCE_V1 */ - return IB_GID_TYPE_IB; + else if (network_type == RDMA_NETWORK_ROCE_V1) + return IB_GID_TYPE_ROCE; + else + return IB_GID_TYPE_IB; } static inline enum rdma_network_type @@ -201,6 +201,9 @@ rdma_gid_attr_network_type(const struct ib_gid_attr *attr) if (attr->gid_type == IB_GID_TYPE_IB) return RDMA_NETWORK_IB; + if (attr->gid_type == IB_GID_TYPE_ROCE) + return RDMA_NETWORK_ROCE_V1; + if (ipv6_addr_v4mapped((struct in6_addr *)&attr->gid)) return RDMA_NETWORK_IPV4; else -- cgit v1.2.3 From c4b4d548fabc15b9c5db9f61204dd0c608414d2d Mon Sep 17 00:00:00 2001 From: Avihai Horon Date: Wed, 23 Sep 2020 19:50:14 +0300 Subject: RDMA/core: Introduce new GID table query API Introduce rdma_query_gid_table which enables querying all the GID tables of a given device and copying the attributes of all valid GID entries to a provided buffer. This API provides a faster way to query a GID table using single call and will be used in libibverbs to improve current approach that requires multiple calls to open, close and read multiple sysfs files for a single GID table entry. Link: https://lore.kernel.org/r/20200923165015.2491894-4-leon@kernel.org Signed-off-by: Avihai Horon Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_cache.h | 3 +++ include/uapi/rdma/ib_user_ioctl_verbs.h | 8 ++++++++ 2 files changed, 11 insertions(+) (limited to 'include') diff --git a/include/rdma/ib_cache.h b/include/rdma/ib_cache.h index 66a8f369a2fa..bae29f50adff 100644 --- a/include/rdma/ib_cache.h +++ b/include/rdma/ib_cache.h @@ -110,5 +110,8 @@ const struct ib_gid_attr *rdma_get_gid_attr(struct ib_device *device, u8 port_num, int index); void rdma_put_gid_attr(const struct ib_gid_attr *attr); void rdma_hold_gid_attr(const struct ib_gid_attr *attr); +ssize_t rdma_query_gid_table(struct ib_device *device, + struct ib_uverbs_gid_entry *entries, + size_t max_entries); #endif /* _IB_CACHE_H */ diff --git a/include/uapi/rdma/ib_user_ioctl_verbs.h b/include/uapi/rdma/ib_user_ioctl_verbs.h index fb8cdb38198b..14820082de5e 100644 --- a/include/uapi/rdma/ib_user_ioctl_verbs.h +++ b/include/uapi/rdma/ib_user_ioctl_verbs.h @@ -251,4 +251,12 @@ enum rdma_driver_id { RDMA_DRIVER_SIW, }; +struct ib_uverbs_gid_entry { + __aligned_u64 gid[2]; + __u32 gid_index; + __u32 port_num; + __u32 gid_type; + __u32 netdev_ifindex; /* It is 0 if there is no netdev associated with it */ +}; + #endif -- cgit v1.2.3 From 9f85cbe50aa044a46f0a22fda323fa27b80c82da Mon Sep 17 00:00:00 2001 From: Avihai Horon Date: Wed, 23 Sep 2020 19:50:15 +0300 Subject: RDMA/uverbs: Expose the new GID query API to user space Expose the query GID table and entry API to user space by adding two new methods and method handlers to the device object. This API provides a faster way to query a GID table using single call and will be used in libibverbs to improve current approach that requires multiple calls to open, close and read multiple sysfs files for a single GID table entry. Link: https://lore.kernel.org/r/20200923165015.2491894-5-leon@kernel.org Signed-off-by: Avihai Horon Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 6 +++--- include/uapi/rdma/ib_user_ioctl_cmds.h | 16 ++++++++++++++++ include/uapi/rdma/ib_user_ioctl_verbs.h | 6 ++++++ 3 files changed, 25 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 3b61fba531d0..ce935d70fdc8 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -138,9 +138,9 @@ union ib_gid { extern union ib_gid zgid; enum ib_gid_type { - IB_GID_TYPE_IB = 0, - IB_GID_TYPE_ROCE = 1, - IB_GID_TYPE_ROCE_UDP_ENCAP = 2, + IB_GID_TYPE_IB = IB_UVERBS_GID_TYPE_IB, + IB_GID_TYPE_ROCE = IB_UVERBS_GID_TYPE_ROCE_V1, + IB_GID_TYPE_ROCE_UDP_ENCAP = IB_UVERBS_GID_TYPE_ROCE_V2, IB_GID_TYPE_SIZE }; diff --git a/include/uapi/rdma/ib_user_ioctl_cmds.h b/include/uapi/rdma/ib_user_ioctl_cmds.h index 99dcabf61a71..7968a1845355 100644 --- a/include/uapi/rdma/ib_user_ioctl_cmds.h +++ b/include/uapi/rdma/ib_user_ioctl_cmds.h @@ -70,6 +70,8 @@ enum uverbs_methods_device { UVERBS_METHOD_QUERY_PORT, UVERBS_METHOD_GET_CONTEXT, UVERBS_METHOD_QUERY_CONTEXT, + UVERBS_METHOD_QUERY_GID_TABLE, + UVERBS_METHOD_QUERY_GID_ENTRY, }; enum uverbs_attrs_invoke_write_cmd_attr_ids { @@ -352,4 +354,18 @@ enum uverbs_attrs_async_event_create { UVERBS_ATTR_ASYNC_EVENT_ALLOC_FD_HANDLE, }; +enum uverbs_attrs_query_gid_table_cmd_attr_ids { + UVERBS_ATTR_QUERY_GID_TABLE_ENTRY_SIZE, + UVERBS_ATTR_QUERY_GID_TABLE_FLAGS, + UVERBS_ATTR_QUERY_GID_TABLE_RESP_ENTRIES, + UVERBS_ATTR_QUERY_GID_TABLE_RESP_NUM_ENTRIES, +}; + +enum uverbs_attrs_query_gid_entry_cmd_attr_ids { + UVERBS_ATTR_QUERY_GID_ENTRY_PORT, + UVERBS_ATTR_QUERY_GID_ENTRY_GID_INDEX, + UVERBS_ATTR_QUERY_GID_ENTRY_FLAGS, + UVERBS_ATTR_QUERY_GID_ENTRY_RESP_ENTRY, +}; + #endif diff --git a/include/uapi/rdma/ib_user_ioctl_verbs.h b/include/uapi/rdma/ib_user_ioctl_verbs.h index 14820082de5e..22483799cd07 100644 --- a/include/uapi/rdma/ib_user_ioctl_verbs.h +++ b/include/uapi/rdma/ib_user_ioctl_verbs.h @@ -251,6 +251,12 @@ enum rdma_driver_id { RDMA_DRIVER_SIW, }; +enum ib_uverbs_gid_type { + IB_UVERBS_GID_TYPE_IB, + IB_UVERBS_GID_TYPE_ROCE_V1, + IB_UVERBS_GID_TYPE_ROCE_V2, +}; + struct ib_uverbs_gid_entry { __aligned_u64 gid[2]; __u32 gid_index; -- cgit v1.2.3 From 07da1223ec939982497db3caccd6215b55acc35c Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Sun, 4 Oct 2020 18:43:37 +0300 Subject: lib/scatterlist: Add support in dynamic allocation of SG table from pages Extend __sg_alloc_table_from_pages to support dynamic allocation of SG table from pages. It should be used by drivers that can't supply all the pages at one time. This function returns the last populated SGE in the table. Users should pass it as an argument to the function from the second call and forward. As before, nents will be equal to the number of populated SGEs (chunks). With this new extension, drivers can benefit the optimization of merging contiguous pages without a need to allocate all pages in advance and hold them in a large buffer. E.g. with the Infiniband driver that allocates a single page for hold the pages. For 1TB memory registration, the temporary buffer would consume only 4KB, instead of 2GB. Link: https://lore.kernel.org/r/20201004154340.1080481-2-leon@kernel.org Signed-off-by: Maor Gottlieb Reviewed-by: Christoph Hellwig Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- include/linux/scatterlist.h | 38 ++++++++++++++++++++++---------------- 1 file changed, 22 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h index 45cf7b69d852..36c47e7e66a2 100644 --- a/include/linux/scatterlist.h +++ b/include/linux/scatterlist.h @@ -165,6 +165,22 @@ static inline void sg_set_buf(struct scatterlist *sg, const void *buf, #define for_each_sgtable_dma_sg(sgt, sg, i) \ for_each_sg((sgt)->sgl, sg, (sgt)->nents, i) +static inline void __sg_chain(struct scatterlist *chain_sg, + struct scatterlist *sgl) +{ + /* + * offset and length are unused for chain entry. Clear them. + */ + chain_sg->offset = 0; + chain_sg->length = 0; + + /* + * Set lowest bit to indicate a link pointer, and make sure to clear + * the termination bit if it happens to be set. + */ + chain_sg->page_link = ((unsigned long) sgl | SG_CHAIN) & ~SG_END; +} + /** * sg_chain - Chain two sglists together * @prv: First scatterlist @@ -178,18 +194,7 @@ static inline void sg_set_buf(struct scatterlist *sg, const void *buf, static inline void sg_chain(struct scatterlist *prv, unsigned int prv_nents, struct scatterlist *sgl) { - /* - * offset and length are unused for chain entry. Clear them. - */ - prv[prv_nents - 1].offset = 0; - prv[prv_nents - 1].length = 0; - - /* - * Set lowest bit to indicate a link pointer, and make sure to clear - * the termination bit if it happens to be set. - */ - prv[prv_nents - 1].page_link = ((unsigned long) sgl | SG_CHAIN) - & ~SG_END; + __sg_chain(&prv[prv_nents - 1], sgl); } /** @@ -286,10 +291,11 @@ void sg_free_table(struct sg_table *); int __sg_alloc_table(struct sg_table *, unsigned int, unsigned int, struct scatterlist *, unsigned int, gfp_t, sg_alloc_fn *); int sg_alloc_table(struct sg_table *, unsigned int, gfp_t); -int __sg_alloc_table_from_pages(struct sg_table *sgt, struct page **pages, - unsigned int n_pages, unsigned int offset, - unsigned long size, unsigned int max_segment, - gfp_t gfp_mask); +struct scatterlist *__sg_alloc_table_from_pages(struct sg_table *sgt, + struct page **pages, unsigned int n_pages, unsigned int offset, + unsigned long size, unsigned int max_segment, + struct scatterlist *prv, unsigned int left_pages, + gfp_t gfp_mask); int sg_alloc_table_from_pages(struct sg_table *sgt, struct page **pages, unsigned int n_pages, unsigned int offset, unsigned long size, gfp_t gfp_mask); -- cgit v1.2.3 From e0477b34d9d11c1a7b1f80bfdbcdc8952ce2adb7 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Thu, 8 Oct 2020 11:27:52 +0300 Subject: RDMA: Explicitly pass in the dma_device to ib_register_device The code in setup_dma_device has become rather convoluted, move all of this to the drivers. Drives now pass in a DMA capable struct device which will be used to setup DMA, or drivers must fully configure the ibdev for DMA and pass in NULL. Other than setting the masks in rvt all drivers were doing this already anyhow. mthca, mlx4 and mlx5 were already setting up maximum DMA segment size for DMA based on their hardweare limits in: __mthca_init_one() dma_set_max_seg_size (1G) __mlx4_init_one() dma_set_max_seg_size (1G) mlx5_pci_init() set_dma_caps() dma_set_max_seg_size (2G) Other non software drivers (except usnic) were extended to UINT_MAX [1, 2] instead of 2G as was before. [1] https://lore.kernel.org/linux-rdma/20200924114940.GE9475@nvidia.com/ [2] https://lore.kernel.org/linux-rdma/20200924114940.GE9475@nvidia.com/ Link: https://lore.kernel.org/r/20201008082752.275846-1-leon@kernel.org Link: https://lore.kernel.org/r/6b2ed339933d066622d5715903870676d8cc523a.1602590106.git.mchehab+huawei@kernel.org Suggested-by: Christoph Hellwig Signed-off-by: Parav Pandit Signed-off-by: Leon Romanovsky Reviewed-by: Christoph Hellwig Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Jason Gunthorpe --- include/rdma/ib_verbs.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index ce935d70fdc8..9bf6c319a670 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -2782,7 +2782,8 @@ void ib_dealloc_device(struct ib_device *device); void ib_get_device_fw_str(struct ib_device *device, char *str); -int ib_register_device(struct ib_device *device, const char *name); +int ib_register_device(struct ib_device *device, const char *name, + struct device *dma_device); void ib_unregister_device(struct ib_device *device); void ib_unregister_driver(enum rdma_driver_id driver_id); void ib_unregister_device_and_put(struct ib_device *device); -- cgit v1.2.3 From e0d696d201dd5d31813787d9b61a42fc459eee89 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Thu, 15 Oct 2020 20:42:18 -0300 Subject: RDMA/rxe: Move the definitions for rxe_av.network_type to uAPI RXE was wrongly using an internal kernel enum as part of its uAPI, split this out into a dedicated uAPI enum just for RXE. It only uses the IPv4 and IPv6 values. This was exposed by changing the internal kernel enum definition which broke RXE. Fixes: 1c15b4f2a42f ("RDMA/core: Modify enum ib_gid_type and enum rdma_network_type") Signed-off-by: Jason Gunthorpe --- include/uapi/rdma/rdma_user_rxe.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/uapi/rdma/rdma_user_rxe.h b/include/uapi/rdma/rdma_user_rxe.h index d8f2e0e46dab..e591d8c1f3cf 100644 --- a/include/uapi/rdma/rdma_user_rxe.h +++ b/include/uapi/rdma/rdma_user_rxe.h @@ -39,6 +39,11 @@ #include #include +enum { + RXE_NETWORK_TYPE_IPV4 = 1, + RXE_NETWORK_TYPE_IPV6 = 2, +}; + union rxe_gid { __u8 raw[16]; struct { @@ -57,6 +62,7 @@ struct rxe_global_route { struct rxe_av { __u8 port_num; + /* From RXE_NETWORK_TYPE_* */ __u8 network_type; __u8 dmac[6]; struct rxe_global_route grh; -- cgit v1.2.3