From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steve Wise Subject: [PATCH RESEND v1 rdma-next 6/6] RDMA/nldev: provide detailed PD information Date: Fri, 2 Feb 2018 13:24:44 -0800 Message-ID: References: Return-path: In-Reply-To: Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: jgg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org, dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org List-Id: linux-rdma@vger.kernel.org Implement the RDMA nldev netlink interface for dumping detailed PD information. Signed-off-by: Steve Wise --- drivers/infiniband/core/nldev.c | 173 +++++++++++++++++++++++++++++++++++++++ include/uapi/rdma/rdma_netlink.h | 8 ++ 2 files changed, 181 insertions(+) diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c index 8d96f3e..42cbec0 100644 --- a/drivers/infiniband/core/nldev.c +++ b/drivers/infiniband/core/nldev.c @@ -101,6 +101,11 @@ [RDMA_NLDEV_ATTR_RES_IOVA] = { .type = NLA_U64 }, [RDMA_NLDEV_ATTR_RES_MRLEN] = { .type = NLA_U64 }, [RDMA_NLDEV_ATTR_RES_PGSIZE] = { .type = NLA_U32 }, + [RDMA_NLDEV_ATTR_RES_PD] = { .type = NLA_NESTED }, + [RDMA_NLDEV_ATTR_RES_PD_ENTRY] = { .type = NLA_NESTED }, + [RDMA_NLDEV_ATTR_RES_LOCAL_DMA_LKEY] = { .type = NLA_U32 }, + [RDMA_NLDEV_ATTR_RES_PD_FLAGS] = { .type = NLA_U32 }, + [RDMA_NLDEV_ATTR_RES_UNSAFE_GLOBAL_RKEY] = { .type = NLA_U32 }, }; static int fill_nldev_handle(struct sk_buff *msg, struct ib_device *device) @@ -499,6 +504,53 @@ static int fill_res_mr_entry(struct sk_buff *msg, return -EMSGSIZE; } +static int fill_res_pd_entry(struct sk_buff *msg, + struct ib_pd *pd) +{ + struct rdma_restrack_entry *res = &pd->res; + struct nlattr *entry_attr; + + entry_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_RES_PD_ENTRY); + if (!entry_attr) + goto out; + + if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_LOCAL_DMA_LKEY, + pd->local_dma_lkey)) + goto err; + if (nla_put_u64_64bit(msg, RDMA_NLDEV_ATTR_RES_USECNT, + atomic_read(&pd->usecnt), 0)) + goto err; + if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PD_FLAGS, pd->flags)) + goto err; + if ((pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY) && + nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_UNSAFE_GLOBAL_RKEY, + pd->unsafe_global_rkey)) + goto err; + + /* + * Existence of task means that it is user PD and netlink + * user is invited to go and read /proc/PID/comm to get name + * of the task file and res->task_com should be NULL. + */ + if (rdma_is_kernel_res(res)) { + if (nla_put_string(msg, RDMA_NLDEV_ATTR_RES_KERN_NAME, + res->kern_name)) + goto err; + } else { + if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PID, + task_pid_vnr(res->task))) + goto err; + } + + nla_nest_end(msg, entry_attr); + return 0; + +err: + nla_nest_cancel(msg, entry_attr); +out: + return -EMSGSIZE; +} + static int nldev_get_doit(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { @@ -1269,6 +1321,124 @@ static int nldev_res_get_mr_dumpit(struct sk_buff *skb, return ret; } +static int nldev_res_get_pd_dumpit(struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct nlattr *tb[RDMA_NLDEV_ATTR_MAX]; + struct rdma_restrack_entry *res; + int err, ret = 0, idx = 0; + struct nlattr *table_attr; + struct ib_device *device; + int start = cb->args[0]; + struct ib_pd *pd = NULL; + struct nlmsghdr *nlh; + u32 index; + + err = nlmsg_parse(cb->nlh, 0, tb, RDMA_NLDEV_ATTR_MAX - 1, + nldev_policy, NULL); + /* + * Right now, we are expecting the device index to get PD information, + * but it is possible to extend this code to return all devices in + * one shot by checking the existence of RDMA_NLDEV_ATTR_DEV_INDEX. + * if it doesn't exist, we will iterate over all devices. + * + * But it is not needed for now. + */ + if (err || !tb[RDMA_NLDEV_ATTR_DEV_INDEX]) + return -EINVAL; + + index = nla_get_u32(tb[RDMA_NLDEV_ATTR_DEV_INDEX]); + device = ib_device_get_by_index(index); + if (!device) + return -EINVAL; + + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, + RDMA_NL_GET_TYPE(RDMA_NL_NLDEV, RDMA_NLDEV_CMD_RES_PD_GET), + 0, NLM_F_MULTI); + + if (fill_nldev_handle(skb, device)) { + ret = -EMSGSIZE; + goto err; + } + + table_attr = nla_nest_start(skb, RDMA_NLDEV_ATTR_RES_PD); + if (!table_attr) { + ret = -EMSGSIZE; + goto err; + } + + down_read(&device->res.rwsem); + hash_for_each_possible(device->res.hash, res, node, RDMA_RESTRACK_PD) { + if (idx < start) + goto next; + + if ((rdma_is_kernel_res(res) && + task_active_pid_ns(current) != &init_pid_ns) || + (!rdma_is_kernel_res(res) && + task_active_pid_ns(current) != + task_active_pid_ns(res->task))) + /* + * 1. Kernel PDs should be visible in init namspace only + * 2. Present only PDs visible in the current namespace + */ + goto next; + + if (!rdma_restrack_get(res)) + /* + * Resource is under release now, but we are not + * relesing lock now, so it will be released in + * our next pass, once we will get ->next pointer. + */ + goto next; + + pd = container_of(res, struct ib_pd, res); + + up_read(&device->res.rwsem); + ret = fill_res_pd_entry(skb, pd); + down_read(&device->res.rwsem); + /* + * Return resource back, but it won't be released till + * the &device->res.rwsem will be released for write. + */ + rdma_restrack_put(res); + + if (ret == -EMSGSIZE) + /* + * There is a chance to optimize here. + * It can be done by using list_prepare_entry + * and list_for_each_entry_continue afterwards. + */ + break; + if (ret) + goto res_err; +next: idx++; + } + up_read(&device->res.rwsem); + + nla_nest_end(skb, table_attr); + nlmsg_end(skb, nlh); + cb->args[0] = idx; + + /* + * No more PDs to fill, cancel the message and + * return 0 to mark end of dumpit. + */ + if (!pd) + goto err; + + put_device(&device->dev); + return skb->len; + +res_err: + nla_nest_cancel(skb, table_attr); + up_read(&device->res.rwsem); + +err: + nlmsg_cancel(skb, nlh); + put_device(&device->dev); + return ret; +} + static const struct rdma_nl_cbs nldev_cb_table[RDMA_NLDEV_NUM_OPS] = { [RDMA_NLDEV_CMD_GET] = { .doit = nldev_get_doit, @@ -1304,6 +1474,9 @@ static int nldev_res_get_mr_dumpit(struct sk_buff *skb, [RDMA_NLDEV_CMD_RES_MR_GET] = { .dump = nldev_res_get_mr_dumpit, }, + [RDMA_NLDEV_CMD_RES_PD_GET] = { + .dump = nldev_res_get_pd_dumpit, + }, }; void __init nldev_init(void) diff --git a/include/uapi/rdma/rdma_netlink.h b/include/uapi/rdma/rdma_netlink.h index 6adaeaa..4aeb59c 100644 --- a/include/uapi/rdma/rdma_netlink.h +++ b/include/uapi/rdma/rdma_netlink.h @@ -246,6 +246,8 @@ enum rdma_nldev_command { RDMA_NLDEV_CMD_RES_MR_GET, /* can dump */ + RDMA_NLDEV_CMD_RES_PD_GET, /* can dump */ + RDMA_NLDEV_NUM_OPS }; @@ -400,6 +402,12 @@ enum rdma_nldev_attr { RDMA_NLDEV_ATTR_RES_MRLEN, /* u64 */ RDMA_NLDEV_ATTR_RES_PGSIZE, /* u32 */ + RDMA_NLDEV_ATTR_RES_PD, /* nested table */ + RDMA_NLDEV_ATTR_RES_PD_ENTRY, /* nested table */ + RDMA_NLDEV_ATTR_RES_LOCAL_DMA_LKEY, /* u32 */ + RDMA_NLDEV_ATTR_RES_PD_FLAGS, /* u32 */ + RDMA_NLDEV_ATTR_RES_UNSAFE_GLOBAL_RKEY, /* u32 */ + RDMA_NLDEV_ATTR_MAX }; #endif /* _UAPI_RDMA_NETLINK_H */ -- 1.8.3.1 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html