[PATCH openEuler-5.10-LTS 01/16] RDMA/hns: Remove redundant DFX file and DFX ops structure
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit 40b4b79c866f category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=40b...
----------------------------------------------------------------------
There is no need to use a dedicated DXF file and DFX structure to manage the interface of the query queue context.
Link: https://lore.kernel.org/r/20220822104455.2311053-2-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/Makefile | 2 +- drivers/infiniband/hw/hns/hns_roce_device.h | 10 ++---- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 35 ++++++++++++++++--- drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 3 -- drivers/infiniband/hw/hns/hns_roce_main.c | 6 +++- drivers/infiniband/hw/hns/hns_roce_restrack.c | 35 +++++++------------ 6 files changed, 50 insertions(+), 41 deletions(-)
diff --git a/drivers/infiniband/hw/hns/Makefile b/drivers/infiniband/hw/hns/Makefile index 9f04f25d9631..a7d259238305 100644 --- a/drivers/infiniband/hw/hns/Makefile +++ b/drivers/infiniband/hw/hns/Makefile @@ -10,6 +10,6 @@ hns-roce-objs := hns_roce_main.o hns_roce_cmd.o hns_roce_pd.o \ hns_roce_cq.o hns_roce_alloc.o hns_roce_db.o hns_roce_srq.o hns_roce_restrack.o
ifdef CONFIG_INFINIBAND_HNS_HIP08 -hns-roce-hw-v2-objs := hns_roce_hw_v2.o hns_roce_hw_v2_dfx.o $(hns-roce-objs) +hns-roce-hw-v2-objs := hns_roce_hw_v2.o $(hns-roce-objs) obj-$(CONFIG_INFINIBAND_HNS) += hns-roce-hw-v2.o endif diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 2d29221055be..848f0e0e76a4 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -849,11 +849,6 @@ struct hns_roce_caps { enum cong_type cong_type; };
-struct hns_roce_dfx_hw { - int (*query_cqc_info)(struct hns_roce_dev *hr_dev, u32 cqn, - int *buffer); -}; - enum hns_roce_device_state { HNS_ROCE_DEVICE_STATE_INITED, HNS_ROCE_DEVICE_STATE_RST_DOWN, @@ -899,6 +894,7 @@ struct hns_roce_hw { int (*init_eq)(struct hns_roce_dev *hr_dev); void (*cleanup_eq)(struct hns_roce_dev *hr_dev); int (*write_srqc)(struct hns_roce_srq *srq, void *mb_buf); + int (*query_cqc)(struct hns_roce_dev *hr_dev, u32 cqn, void *buffer); const struct ib_device_ops *hns_roce_dev_ops; const struct ib_device_ops *hns_roce_dev_srq_ops; }; @@ -960,7 +956,6 @@ struct hns_roce_dev { void *priv; struct workqueue_struct *irq_workq; struct work_struct ecc_work; - const struct hns_roce_dfx_hw *dfx; u32 func_num; u32 is_vf; u32 cong_algo_tmpl_id; @@ -1223,8 +1218,7 @@ u8 hns_get_gid_index(struct hns_roce_dev *hr_dev, u8 port, int gid_index); void hns_roce_handle_device_err(struct hns_roce_dev *hr_dev); int hns_roce_init(struct hns_roce_dev *hr_dev); void hns_roce_exit(struct hns_roce_dev *hr_dev); -int hns_roce_fill_res_cq_entry(struct sk_buff *msg, - struct ib_cq *ib_cq); +int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index de7873551ffc..6b772536ab3c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -5801,6 +5801,35 @@ static int hns_roce_v2_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period) return ret; }
+static int hns_roce_v2_query_cqc(struct hns_roce_dev *hr_dev, u32 cqn, + void *buffer) +{ + struct hns_roce_v2_cq_context *context; + struct hns_roce_cmd_mailbox *mailbox; + int ret; + + mailbox = hns_roce_alloc_cmd_mailbox(hr_dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + context = mailbox->buf; + ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, + HNS_ROCE_CMD_QUERY_CQC, cqn); + if (ret) { + ibdev_err(&hr_dev->ib_dev, + "failed to process cmd when querying CQ, ret = %d.\n", + ret); + goto err_mailbox; + } + + memcpy(buffer, context, sizeof(*context)); + +err_mailbox: + hns_roce_free_cmd_mailbox(hr_dev, mailbox); + + return ret; +} + static void hns_roce_irq_work_handle(struct work_struct *work) { struct hns_roce_work *irq_work = @@ -6593,10 +6622,6 @@ static void hns_roce_v2_cleanup_eq_table(struct hns_roce_dev *hr_dev) kfree(eq_table->eq); }
-static const struct hns_roce_dfx_hw hns_roce_dfx_hw_v2 = { - .query_cqc_info = hns_roce_v2_query_cqc_info, -}; - static const struct ib_device_ops hns_roce_v2_dev_ops = { .destroy_qp = hns_roce_v2_destroy_qp, .modify_cq = hns_roce_v2_modify_cq, @@ -6637,6 +6662,7 @@ static const struct hns_roce_hw hns_roce_hw_v2 = { .init_eq = hns_roce_v2_init_eq_table, .cleanup_eq = hns_roce_v2_cleanup_eq_table, .write_srqc = hns_roce_v2_write_srqc, + .query_cqc = hns_roce_v2_query_cqc, .hns_roce_dev_ops = &hns_roce_v2_dev_ops, .hns_roce_dev_srq_ops = &hns_roce_v2_dev_srq_ops, }; @@ -6668,7 +6694,6 @@ static void hns_roce_hw_v2_get_cfg(struct hns_roce_dev *hr_dev, hr_dev->is_vf = id->driver_data; hr_dev->dev = &handle->pdev->dev; hr_dev->hw = &hns_roce_hw_v2; - hr_dev->dfx = &hns_roce_dfx_hw_v2; hr_dev->sdb_offset = ROCEE_DB_SQ_L_0_REG; hr_dev->odb_offset = hr_dev->sdb_offset;
diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index 8aa56c24e677..6bf878fe661e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -1455,9 +1455,6 @@ struct hns_roce_sccc_clr_done { __le32 rsv[5]; };
-int hns_roce_v2_query_cqc_info(struct hns_roce_dev *hr_dev, u32 cqn, - int *buffer); - int hns_roce_v2_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata);
int hns_roce_v2_destroy_qp_common(struct hns_roce_dev *hr_dev, diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 11f42cebfa40..9d4dbc3205c6 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -514,7 +514,6 @@ static const struct ib_device_ops hns_roce_dev_ops = { .destroy_ah = hns_roce_destroy_ah, .destroy_cq = hns_roce_destroy_cq, .disassociate_ucontext = hns_roce_disassociate_ucontext, - .fill_res_cq_entry = hns_roce_fill_res_cq_entry, .get_dma_mr = hns_roce_get_dma_mr, .get_link_layer = hns_roce_get_link_layer, .get_port_immutable = hns_roce_port_immutable, @@ -564,6 +563,10 @@ static const struct ib_device_ops hns_roce_dev_xrcd_ops = { INIT_RDMA_OBJ_SIZE(ib_xrcd, hns_roce_xrcd, ibxrcd), };
+static const struct ib_device_ops hns_roce_dev_restrack_ops = { + .fill_res_cq_entry = hns_roce_fill_res_cq_entry, +}; + static int hns_roce_register_device(struct hns_roce_dev *hr_dev) { int ret; @@ -642,6 +645,7 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
ib_set_device_ops(ib_dev, hr_dev->hw->hns_roce_dev_ops); ib_set_device_ops(ib_dev, &hns_roce_dev_ops); + ib_set_device_ops(ib_dev, &hns_roce_dev_restrack_ops); for (i = 0; i < hr_dev->caps.num_ports; i++) { if (!hr_dev->iboe.netdevs[i]) continue; diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 24a154d64630..83417be15d3f 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -55,45 +55,34 @@ static int hns_roce_fill_cq(struct sk_buff *msg, return -EMSGSIZE; }
-int hns_roce_fill_res_cq_entry(struct sk_buff *msg, - struct ib_cq *ib_cq) +int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq) { struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device); struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq); - struct hns_roce_v2_cq_context *context; + struct hns_roce_v2_cq_context context; struct nlattr *table_attr; int ret;
- if (!hr_dev->dfx->query_cqc_info) + if (!hr_dev->hw->query_cqc) return -EINVAL;
- context = kzalloc(sizeof(struct hns_roce_v2_cq_context), GFP_KERNEL); - if (!context) - return -ENOMEM; - - ret = hr_dev->dfx->query_cqc_info(hr_dev, hr_cq->cqn, (int *)context); + ret = hr_dev->hw->query_cqc(hr_dev, hr_cq->cqn, &context); if (ret) - goto err; + return -EINVAL;
table_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_DRIVER); - if (!table_attr) { - ret = -EMSGSIZE; - goto err; - } + if (!table_attr) + return -EMSGSIZE;
- if (hns_roce_fill_cq(msg, context)) { - ret = -EMSGSIZE; - goto err_cancel_table; - } + if (hns_roce_fill_cq(msg, &context)) + goto err;
nla_nest_end(msg, table_attr); - kfree(context);
return 0;
-err_cancel_table: - nla_nest_cancel(msg, table_attr); err: - kfree(context); - return ret; + nla_nest_cancel(msg, table_attr); + + return -EMSGSIZE; }
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit eb00b9a08b9d category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=eb0...
----------------------------------------------------------------------
Remove the resttrack attributes from the queue context held by ROCEE, and add the resttrack attributes from the queue information maintained by the driver.
For example:
$ rdma res show cq dev hns_0 cqn 14 -dd -jp [ { "ifindex": 4, "ifname": "hns_0", "cqn": 14, "cqe": 127, "users": 1, "adaptive-moderation": false, "ctxn": 8, "pid": 1524, "comm": "ib_send_bw" }, "drv_cq_depth": 128, "drv_cons_index": 0, "drv_cqe_size": 32, "drv_arm_sn": 1 }
Link: https://lore.kernel.org/r/20220822104455.2311053-3-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_restrack.c | 67 +++---------------- 1 file changed, 10 insertions(+), 57 deletions(-)
diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 83417be15d3f..2e8299784bc2 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -9,72 +9,25 @@ #include "hns_roce_device.h" #include "hns_roce_hw_v2.h"
-static int hns_roce_fill_cq(struct sk_buff *msg, - struct hns_roce_v2_cq_context *context) -{ - if (rdma_nl_put_driver_u32(msg, "state", - hr_reg_read(context, CQC_ARM_ST))) - - goto err; - - if (rdma_nl_put_driver_u32(msg, "ceqn", - hr_reg_read(context, CQC_CEQN))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "cqn", - hr_reg_read(context, CQC_CQN))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "hopnum", - hr_reg_read(context, CQC_CQE_HOP_NUM))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "pi", - hr_reg_read(context, CQC_CQ_PRODUCER_IDX))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "ci", - hr_reg_read(context, CQC_CQ_CONSUMER_IDX))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "coalesce", - hr_reg_read(context, CQC_CQ_MAX_CNT))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "period", - hr_reg_read(context, CQC_CQ_PERIOD))) - goto err; - - if (rdma_nl_put_driver_u32(msg, "cnt", - hr_reg_read(context, CQC_CQE_CNT))) - goto err; - - return 0; - -err: - return -EMSGSIZE; -} - int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq) { - struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device); struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq); - struct hns_roce_v2_cq_context context; struct nlattr *table_attr; - int ret; - - if (!hr_dev->hw->query_cqc) - return -EINVAL; - - ret = hr_dev->hw->query_cqc(hr_dev, hr_cq->cqn, &context); - if (ret) - return -EINVAL;
table_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_DRIVER); if (!table_attr) return -EMSGSIZE;
- if (hns_roce_fill_cq(msg, &context)) + if (rdma_nl_put_driver_u32(msg, "cq_depth", hr_cq->cq_depth)) + goto err; + + if (rdma_nl_put_driver_u32(msg, "cons_index", hr_cq->cons_index)) + goto err; + + if (rdma_nl_put_driver_u32(msg, "cqe_size", hr_cq->cqe_size)) + goto err; + + if (rdma_nl_put_driver_u32(msg, "arm_sn", hr_cq->arm_sn)) goto err;
nla_nest_end(msg, table_attr);
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit f2b070f36d1b category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=f2b...
----------------------------------------------------------------------
The CQ raw restrack attributes come from the queue context maintained by the ROCEE.
For example:
$ rdma res show cq dev hns_0 cqn 14 -dd -jp -r [ { "ifindex": 4, "ifname": "hns_0", "data": [ 1,0,0,0,7,0,0,0,0,0,0,0,0,82,6,0,0,82,6,0,0,82,6,0, 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0, 6,0,0,0,0,0,0,0 ] } ]
Link: https://lore.kernel.org/r/20220822104455.2311053-4-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_device.h | 1 + drivers/infiniband/hw/hns/hns_roce_main.c | 1 + drivers/infiniband/hw/hns/hns_roce_restrack.c | 39 +++++++++++++++++++ 3 files changed, 41 insertions(+)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 848f0e0e76a4..c6a31863fd72 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -1219,6 +1219,7 @@ void hns_roce_handle_device_err(struct hns_roce_dev *hr_dev); int hns_roce_init(struct hns_roce_dev *hr_dev); void hns_roce_exit(struct hns_roce_dev *hr_dev); int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq); +int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 9d4dbc3205c6..e6901ff2b6b1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -565,6 +565,7 @@ static const struct ib_device_ops hns_roce_dev_xrcd_ops = {
static const struct ib_device_ops hns_roce_dev_restrack_ops = { .fill_res_cq_entry = hns_roce_fill_res_cq_entry, + .fill_res_cq_entry_raw = hns_roce_fill_res_cq_entry_raw, };
static int hns_roce_register_device(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 2e8299784bc2..3f9c2f9dfdf6 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -9,6 +9,8 @@ #include "hns_roce_device.h" #include "hns_roce_hw_v2.h"
+#define MAX_ENTRY_NUM 256 + int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq) { struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq); @@ -39,3 +41,40 @@ int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq)
return -EMSGSIZE; } + +int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device); + struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq); + struct hns_roce_v2_cq_context context; + u32 data[MAX_ENTRY_NUM] = {}; + int offset = 0; + int ret; + + if (!hr_dev->hw->query_cqc) + return -EINVAL; + + ret = hr_dev->hw->query_cqc(hr_dev, hr_cq->cqn, &context); + if (ret) + return -EINVAL; + + data[offset++] = hr_reg_read(&context, CQC_CQ_ST); + data[offset++] = hr_reg_read(&context, CQC_SHIFT); + data[offset++] = hr_reg_read(&context, CQC_CQE_SIZE); + data[offset++] = hr_reg_read(&context, CQC_CQE_CNT); + data[offset++] = hr_reg_read(&context, CQC_CQ_PRODUCER_IDX); + data[offset++] = hr_reg_read(&context, CQC_CQ_CONSUMER_IDX); + data[offset++] = hr_reg_read(&context, CQC_DB_RECORD_EN); + data[offset++] = hr_reg_read(&context, CQC_ARM_ST); + data[offset++] = hr_reg_read(&context, CQC_CMD_SN); + data[offset++] = hr_reg_read(&context, CQC_CEQN); + data[offset++] = hr_reg_read(&context, CQC_CQ_MAX_CNT); + data[offset++] = hr_reg_read(&context, CQC_CQ_PERIOD); + data[offset++] = hr_reg_read(&context, CQC_CQE_HOP_NUM); + data[offset++] = hr_reg_read(&context, CQC_CQE_BAR_PG_SZ); + data[offset++] = hr_reg_read(&context, CQC_CQE_BUF_PG_SZ); + + ret = nla_put(msg, RDMA_NLDEV_ATTR_RES_RAW, offset * sizeof(u32), data); + + return ret; +}
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit e198d65d76e9 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=e19...
----------------------------------------------------------------------
The QP restrack attributes come from the queue information maintained by the driver.
For example:
$ rdma res show qp link hns_0 lqpn 41 -jp -dd [ { "ifindex": 4, "ifname": "hns_0", "port": 1, "lqpn": 41, "rqpn": 40, "type": "RC", "state": "RTR", "rq-psn": 12474738, "sq-psn": 0, "path-mig-state": "ARMED", "pdn": 9, "pid": 1523, "comm": "ib_send_bw" }, "drv_sq_wqe_cnt": 128, "drv_sq_max_gs": 1, "drv_rq_wqe_cnt": 512, "drv_rq_max_gs": 2, "drv_ext_sge_sge_cnt": 0 }
Link: https://lore.kernel.org/r/20220822104455.2311053-5-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_device.h | 1 + drivers/infiniband/hw/hns/hns_roce_main.c | 1 + drivers/infiniband/hw/hns/hns_roce_restrack.c | 34 +++++++++++++++++++ 3 files changed, 36 insertions(+)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index c6a31863fd72..ad327cc57359 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -1220,6 +1220,7 @@ int hns_roce_init(struct hns_roce_dev *hr_dev); void hns_roce_exit(struct hns_roce_dev *hr_dev); int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq); +int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index e6901ff2b6b1..56fb290ae55f 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -566,6 +566,7 @@ static const struct ib_device_ops hns_roce_dev_xrcd_ops = { static const struct ib_device_ops hns_roce_dev_restrack_ops = { .fill_res_cq_entry = hns_roce_fill_res_cq_entry, .fill_res_cq_entry_raw = hns_roce_fill_res_cq_entry_raw, + .fill_res_qp_entry = hns_roce_fill_res_qp_entry, };
static int hns_roce_register_device(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 3f9c2f9dfdf6..e8fef37f810d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -78,3 +78,37 @@ int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq)
return ret; } + +int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp) +{ + struct hns_roce_qp *hr_qp = to_hr_qp(ib_qp); + struct nlattr *table_attr; + + table_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_DRIVER); + if (!table_attr) + return -EMSGSIZE; + + if (rdma_nl_put_driver_u32_hex(msg, "sq_wqe_cnt", hr_qp->sq.wqe_cnt)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "sq_max_gs", hr_qp->sq.max_gs)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "rq_wqe_cnt", hr_qp->rq.wqe_cnt)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "rq_max_gs", hr_qp->rq.max_gs)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "ext_sge_sge_cnt", hr_qp->sge.sge_cnt)) + goto err; + + nla_nest_end(msg, table_attr); + + return 0; + +err: + nla_nest_cancel(msg, table_attr); + + return -EMSGSIZE; +}
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit 3e89d78b21a8 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=3e8...
----------------------------------------------------------------------
The QP raw restrack attributes come from the queue context maintained by the ROCEE.
For example:
$ rdma res show qp link hns_0 -jp -dd -r [ { "ifindex": 4, "ifname": "hns_0", "data": [ 2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,4,0,0,0, 5,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,255,156,0,0,63,156,0,0, 7,0,0,0,1,0,0,0,9,0,0,0,0,0,0,0,2,0,0,0,2,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,63,156,0, 0,0,0,0,0 ] } ]
Link: https://lore.kernel.org/r/20220822104455.2311053-6-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_device.h | 2 + drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 12 ++-- drivers/infiniband/hw/hns/hns_roce_main.c | 1 + drivers/infiniband/hw/hns/hns_roce_restrack.c | 56 +++++++++++++++++++ 4 files changed, 65 insertions(+), 6 deletions(-)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index ad327cc57359..8b0a8286fc2b 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -895,6 +895,7 @@ struct hns_roce_hw { void (*cleanup_eq)(struct hns_roce_dev *hr_dev); int (*write_srqc)(struct hns_roce_srq *srq, void *mb_buf); int (*query_cqc)(struct hns_roce_dev *hr_dev, u32 cqn, void *buffer); + int (*query_qpc)(struct hns_roce_dev *hr_dev, u32 qpn, void *buffer); const struct ib_device_ops *hns_roce_dev_ops; const struct ib_device_ops *hns_roce_dev_srq_ops; }; @@ -1221,6 +1222,7 @@ void hns_roce_exit(struct hns_roce_dev *hr_dev); int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp); +int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 6b772536ab3c..6bf1a9ccdd97 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -5344,9 +5344,8 @@ static int to_ib_qp_st(enum hns_roce_v2_qp_state state) return (state < ARRAY_SIZE(map)) ? map[state] : -1; }
-static int hns_roce_v2_query_qpc(struct hns_roce_dev *hr_dev, - struct hns_roce_qp *hr_qp, - struct hns_roce_v2_qp_context *hr_context) +static int hns_roce_v2_query_qpc(struct hns_roce_dev *hr_dev, u32 qpn, + void *buffer) { struct hns_roce_cmd_mailbox *mailbox; int ret; @@ -5356,11 +5355,11 @@ static int hns_roce_v2_query_qpc(struct hns_roce_dev *hr_dev, return PTR_ERR(mailbox);
ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, HNS_ROCE_CMD_QUERY_QPC, - hr_qp->qpn); + qpn); if (ret) goto out;
- memcpy(hr_context, mailbox->buf, hr_dev->caps.qpc_sz); + memcpy(buffer, mailbox->buf, hr_dev->caps.qpc_sz);
out: hns_roce_free_cmd_mailbox(hr_dev, mailbox); @@ -5390,7 +5389,7 @@ static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, goto done; }
- ret = hns_roce_v2_query_qpc(hr_dev, hr_qp, &context); + ret = hns_roce_v2_query_qpc(hr_dev, hr_qp->qpn, &context); if (ret) { ibdev_err(ibdev, "failed to query QPC, ret = %d.\n", ret); ret = -EINVAL; @@ -6663,6 +6662,7 @@ static const struct hns_roce_hw hns_roce_hw_v2 = { .cleanup_eq = hns_roce_v2_cleanup_eq_table, .write_srqc = hns_roce_v2_write_srqc, .query_cqc = hns_roce_v2_query_cqc, + .query_qpc = hns_roce_v2_query_qpc, .hns_roce_dev_ops = &hns_roce_v2_dev_ops, .hns_roce_dev_srq_ops = &hns_roce_v2_dev_srq_ops, }; diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 56fb290ae55f..1585fe3321f8 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -567,6 +567,7 @@ static const struct ib_device_ops hns_roce_dev_restrack_ops = { .fill_res_cq_entry = hns_roce_fill_res_cq_entry, .fill_res_cq_entry_raw = hns_roce_fill_res_cq_entry_raw, .fill_res_qp_entry = hns_roce_fill_res_qp_entry, + .fill_res_qp_entry_raw = hns_roce_fill_res_qp_entry_raw, };
static int hns_roce_register_device(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index e8fef37f810d..9bafc627864b 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -112,3 +112,59 @@ int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp)
return -EMSGSIZE; } + +int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(ib_qp->device); + struct hns_roce_qp *hr_qp = to_hr_qp(ib_qp); + struct hns_roce_v2_qp_context context; + u32 data[MAX_ENTRY_NUM] = {}; + int offset = 0; + int ret; + + if (!hr_dev->hw->query_qpc) + return -EINVAL; + + ret = hr_dev->hw->query_qpc(hr_dev, hr_qp->qpn, &context); + if (ret) + return -EINVAL; + + data[offset++] = hr_reg_read(&context, QPC_QP_ST); + data[offset++] = hr_reg_read(&context, QPC_ERR_TYPE); + data[offset++] = hr_reg_read(&context, QPC_CHECK_FLG); + data[offset++] = hr_reg_read(&context, QPC_SRQ_EN); + data[offset++] = hr_reg_read(&context, QPC_SRQN); + data[offset++] = hr_reg_read(&context, QPC_QKEY_XRCD); + data[offset++] = hr_reg_read(&context, QPC_TX_CQN); + data[offset++] = hr_reg_read(&context, QPC_RX_CQN); + data[offset++] = hr_reg_read(&context, QPC_SQ_PRODUCER_IDX); + data[offset++] = hr_reg_read(&context, QPC_SQ_CONSUMER_IDX); + data[offset++] = hr_reg_read(&context, QPC_RQ_RECORD_EN); + data[offset++] = hr_reg_read(&context, QPC_RQ_PRODUCER_IDX); + data[offset++] = hr_reg_read(&context, QPC_RQ_CONSUMER_IDX); + data[offset++] = hr_reg_read(&context, QPC_SQ_SHIFT); + data[offset++] = hr_reg_read(&context, QPC_RQWS); + data[offset++] = hr_reg_read(&context, QPC_RQ_SHIFT); + data[offset++] = hr_reg_read(&context, QPC_SGE_SHIFT); + data[offset++] = hr_reg_read(&context, QPC_SQ_HOP_NUM); + data[offset++] = hr_reg_read(&context, QPC_RQ_HOP_NUM); + data[offset++] = hr_reg_read(&context, QPC_SGE_HOP_NUM); + data[offset++] = hr_reg_read(&context, QPC_WQE_SGE_BA_PG_SZ); + data[offset++] = hr_reg_read(&context, QPC_WQE_SGE_BUF_PG_SZ); + data[offset++] = hr_reg_read(&context, QPC_RETRY_NUM_INIT); + data[offset++] = hr_reg_read(&context, QPC_RETRY_CNT); + data[offset++] = hr_reg_read(&context, QPC_SQ_CUR_PSN); + data[offset++] = hr_reg_read(&context, QPC_SQ_MAX_PSN); + data[offset++] = hr_reg_read(&context, QPC_SQ_FLUSH_IDX); + data[offset++] = hr_reg_read(&context, QPC_SQ_MAX_IDX); + data[offset++] = hr_reg_read(&context, QPC_SQ_TX_ERR); + data[offset++] = hr_reg_read(&context, QPC_SQ_RX_ERR); + data[offset++] = hr_reg_read(&context, QPC_RQ_RX_ERR); + data[offset++] = hr_reg_read(&context, QPC_RQ_TX_ERR); + data[offset++] = hr_reg_read(&context, QPC_RQ_CQE_IDX); + data[offset++] = hr_reg_read(&context, QPC_RQ_RTY_TX_ERR); + + ret = nla_put(msg, RDMA_NLDEV_ATTR_RES_RAW, offset * sizeof(u32), data); + + return ret; +}
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit dc9981ef17c6 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=dc9...
----------------------------------------------------------------------
The MR restrack attributes come from the queue information maintained by the driver.
For example:
$ rdma res show mr dev hns_0 mrn 6 -dd -jp [ { "ifindex": 4, "ifname": "hns_0", "mrn": 6, "rkey": "300", "lkey": "300", "mrlen": 131072, "pdn": 8, "pid": 1524, "comm": "ib_send_bw" }, "drv_pbl_hop_num": 2, "drv_ba_pg_shift": 14, "drv_buf_pg_shift": 12 }
Link: https://lore.kernel.org/r/20220822104455.2311053-7-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_device.h | 1 + drivers/infiniband/hw/hns/hns_roce_main.c | 1 + drivers/infiniband/hw/hns/hns_roce_restrack.c | 30 +++++++++++++++++++ 3 files changed, 32 insertions(+)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 8b0a8286fc2b..414c06767e29 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -1223,6 +1223,7 @@ int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp); int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp); +int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 1585fe3321f8..a848f77e29c8 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -568,6 +568,7 @@ static const struct ib_device_ops hns_roce_dev_restrack_ops = { .fill_res_cq_entry_raw = hns_roce_fill_res_cq_entry_raw, .fill_res_qp_entry = hns_roce_fill_res_qp_entry, .fill_res_qp_entry_raw = hns_roce_fill_res_qp_entry_raw, + .fill_res_mr_entry = hns_roce_fill_res_mr_entry, };
static int hns_roce_register_device(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 9bafc627864b..84f942e19743 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -168,3 +168,33 @@ int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp)
return ret; } + +int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr) +{ + struct hns_roce_mr *hr_mr = to_hr_mr(ib_mr); + struct nlattr *table_attr; + + table_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_DRIVER); + if (!table_attr) + return -EMSGSIZE; + + if (rdma_nl_put_driver_u32_hex(msg, "pbl_hop_num", hr_mr->pbl_hop_num)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "ba_pg_shift", + hr_mr->pbl_mtr.hem_cfg.ba_pg_shift)) + goto err; + + if (rdma_nl_put_driver_u32_hex(msg, "buf_pg_shift", + hr_mr->pbl_mtr.hem_cfg.buf_pg_shift)) + goto err; + + nla_nest_end(msg, table_attr); + + return 0; + +err: + nla_nest_cancel(msg, table_attr); + + return -EMSGSIZE; +}
From: Wenpeng Liang liangwenpeng@huawei.com
mainline inclusion from mainline-for-next commit 3d67e7e236ad category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5 CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=3d6...
----------------------------------------------------------------------
The MR raw restrack attributes come from the queue context maintained by the ROCEE.
For example:
$ rdma res show mr dev hns_0 mrn 6 -dd -jp -r [ { "ifindex": 4, "ifname": "hns_0", "data": [ 1,0,0,0,2,0,0,0,0,3,0,0,0,0,2,0,0,0,0,0,32,0,0,0,2,0,0,0, 2,0,0,0,0,0,0,0 ] } ]
Link: https://lore.kernel.org/r/20220822104455.2311053-8-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang liangwenpeng@huawei.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Zhengfeng Luo luozhengfeng@h-partners.com Reviewed-by: Yangyang Li liyangyang20@huawei.com Reviewed-by: YueHaibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- drivers/infiniband/hw/hns/hns_roce_device.h | 2 ++ drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 30 ++++++++++++++++++ drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 3 +- drivers/infiniband/hw/hns/hns_roce_main.c | 1 + drivers/infiniband/hw/hns/hns_roce_restrack.c | 31 +++++++++++++++++++ 5 files changed, 66 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 414c06767e29..1386a7720d2d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -896,6 +896,7 @@ struct hns_roce_hw { int (*write_srqc)(struct hns_roce_srq *srq, void *mb_buf); int (*query_cqc)(struct hns_roce_dev *hr_dev, u32 cqn, void *buffer); int (*query_qpc)(struct hns_roce_dev *hr_dev, u32 qpn, void *buffer); + int (*query_mpt)(struct hns_roce_dev *hr_dev, u32 key, void *buffer); const struct ib_device_ops *hns_roce_dev_ops; const struct ib_device_ops *hns_roce_dev_srq_ops; }; @@ -1224,6 +1225,7 @@ int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq); int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp); int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp); int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr); +int hns_roce_fill_res_mr_entry_raw(struct sk_buff *msg, struct ib_mr *ib_mr); struct hns_user_mmap_entry * hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address, size_t length, diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 6bf1a9ccdd97..e8ea9448195e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -5829,6 +5829,35 @@ static int hns_roce_v2_query_cqc(struct hns_roce_dev *hr_dev, u32 cqn, return ret; }
+static int hns_roce_v2_query_mpt(struct hns_roce_dev *hr_dev, u32 key, + void *buffer) +{ + struct hns_roce_v2_mpt_entry *context; + struct hns_roce_cmd_mailbox *mailbox; + int ret; + + mailbox = hns_roce_alloc_cmd_mailbox(hr_dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + context = mailbox->buf; + ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, HNS_ROCE_CMD_QUERY_MPT, + key_to_hw_index(key)); + if (ret) { + ibdev_err(&hr_dev->ib_dev, + "failed to process cmd when querying MPT, ret = %d.\n", + ret); + goto err_mailbox; + } + + memcpy(buffer, context, sizeof(*context)); + +err_mailbox: + hns_roce_free_cmd_mailbox(hr_dev, mailbox); + + return ret; +} + static void hns_roce_irq_work_handle(struct work_struct *work) { struct hns_roce_work *irq_work = @@ -6663,6 +6692,7 @@ static const struct hns_roce_hw hns_roce_hw_v2 = { .write_srqc = hns_roce_v2_write_srqc, .query_cqc = hns_roce_v2_query_cqc, .query_qpc = hns_roce_v2_query_qpc, + .query_mpt = hns_roce_v2_query_mpt, .hns_roce_dev_ops = &hns_roce_v2_dev_ops, .hns_roce_dev_srq_ops = &hns_roce_v2_dev_srq_ops, }; diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index 6bf878fe661e..16437df320ea 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -758,7 +758,8 @@ struct hns_roce_v2_mpt_entry { #define MPT_INNER_PA_VLD MPT_FIELD_LOC(71, 71) #define MPT_MW_BIND_QPN MPT_FIELD_LOC(95, 72) #define MPT_BOUND_LKEY MPT_FIELD_LOC(127, 96) -#define MPT_LEN MPT_FIELD_LOC(191, 128) +#define MPT_LEN_L MPT_FIELD_LOC(159, 128) +#define MPT_LEN_H MPT_FIELD_LOC(191, 160) #define MPT_LKEY MPT_FIELD_LOC(223, 192) #define MPT_VA MPT_FIELD_LOC(287, 224) #define MPT_PBL_SIZE MPT_FIELD_LOC(319, 288) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index a848f77e29c8..b4ce05d12f8c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -569,6 +569,7 @@ static const struct ib_device_ops hns_roce_dev_restrack_ops = { .fill_res_qp_entry = hns_roce_fill_res_qp_entry, .fill_res_qp_entry_raw = hns_roce_fill_res_qp_entry_raw, .fill_res_mr_entry = hns_roce_fill_res_mr_entry, + .fill_res_mr_entry_raw = hns_roce_fill_res_mr_entry_raw, };
static int hns_roce_register_device(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 84f942e19743..989a2af2e938 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -198,3 +198,34 @@ int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr)
return -EMSGSIZE; } + +int hns_roce_fill_res_mr_entry_raw(struct sk_buff *msg, struct ib_mr *ib_mr) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(ib_mr->device); + struct hns_roce_mr *hr_mr = to_hr_mr(ib_mr); + struct hns_roce_v2_mpt_entry context; + u32 data[MAX_ENTRY_NUM] = {}; + int offset = 0; + int ret; + + if (!hr_dev->hw->query_mpt) + return -EINVAL; + + ret = hr_dev->hw->query_mpt(hr_dev, hr_mr->key, &context); + if (ret) + return -EINVAL; + + data[offset++] = hr_reg_read(&context, MPT_ST); + data[offset++] = hr_reg_read(&context, MPT_PD); + data[offset++] = hr_reg_read(&context, MPT_LKEY); + data[offset++] = hr_reg_read(&context, MPT_LEN_L); + data[offset++] = hr_reg_read(&context, MPT_LEN_H); + data[offset++] = hr_reg_read(&context, MPT_PBL_SIZE); + data[offset++] = hr_reg_read(&context, MPT_PBL_HOP_NUM); + data[offset++] = hr_reg_read(&context, MPT_PBL_BA_PG_SZ); + data[offset++] = hr_reg_read(&context, MPT_PBL_BUF_PG_SZ); + + ret = nla_put(msg, RDMA_NLDEV_ATTR_RES_RAW, offset * sizeof(u32), data); + + return ret; +}
From: Tejun Heo tj@kernel.org
mainline inclusion from mainline-v5.13-rc1 commit f46ec84b5acbf8d7067d71a6bbdde213d4b86036 category: bugfix bugzilla: 187443, https://gitee.com/openeuler/kernel/issues/I5PFPV CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
----------------------------------------------------------------------
ioc_adjust_base_vrate() ignored vrate_min when rq_wait_pct indicates that there is QD contention. The reasoning was that QD depletion always reliably indicates device saturation and thus it's safe to override user specified vrate_min. However, this sometimes leads to unnecessary throttling, especially on really fast devices, because vrate adjustments have delays and inertia. It also confuses users because the behavior violates the explicitly specified configuration.
This patch drops the special case handling so that vrate_min is always applied.
Signed-off-by: Tejun Heo tj@kernel.org Link: https://lore.kernel.org/r/YIIo1HuyNmhDeiNx@slm.duckdns.org Signed-off-by: Jens Axboe axboe@kernel.dk
confilict: block/blk-iocost.c
Signed-off-by: Li Nan linan122@huawei.com Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- block/blk-iocost.c | 4 ---- 1 file changed, 4 deletions(-)
diff --git a/block/blk-iocost.c b/block/blk-iocost.c index fb8f959a7f32..08e4ba856e3b 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -2341,10 +2341,6 @@ static void ioc_timer_fn(struct timer_list *timer) u64 vrate = ioc->vtime_base_rate; u64 vrate_min = ioc->vrate_min, vrate_max = ioc->vrate_max;
- /* rq_wait signal is always reliable, ignore user vrate_min */ - if (rq_wait_pct > RQ_WAIT_BUSY_PCT) - vrate_min = VRATE_MIN; - /* * If vrate is out of bounds, apply clamp gradually as the * bounds can change abruptly. Otherwise, apply busy_level
From: Sasha Levin sashal@kernel.org
stable inclusion from stable-v5.4.215 commit 4051324a6dafd7053c74c475e80b3ba10ae672b0 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I5T9C3 CVE: CVE-2022-3303
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/fs...
---------------------------
[ Upstream commit 8423f0b6d513b259fdab9c9bf4aaa6188d054c2d ]
There is a small race window at snd_pcm_oss_sync() that is called from OSS PCM SNDCTL_DSP_SYNC ioctl; namely the function calls snd_pcm_oss_make_ready() at first, then takes the params_lock mutex for the rest. When the stream is set up again by another thread between them, it leads to inconsistency, and may result in unexpected results such as NULL dereference of OSS buffer as a fuzzer spotted recently.
The fix is simply to cover snd_pcm_oss_make_ready() call into the same params_lock mutex with snd_pcm_oss_make_ready_locked() variant.
Reported-and-tested-by: butt3rflyh4ck butterflyhuangxx@gmail.com Reviewed-by: Jaroslav Kysela perex@perex.cz Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/CAFcO6XN7JDM4xSXGhtusQfS2mSBcx50VJKwQpCq=WeLt57aaZ... Link: https://lore.kernel.org/r/20220905060714.22549-1-tiwai@suse.de Signed-off-by: Takashi Iwai tiwai@suse.de Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Xia Longlong xialonglong1@huawei.com Reviewed-by: Kefeng Wang wangkefeng.wang@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- sound/core/oss/pcm_oss.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index f88de74da1eb..de6f94bee50b 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -1662,13 +1662,14 @@ static int snd_pcm_oss_sync(struct snd_pcm_oss_file *pcm_oss_file) runtime = substream->runtime; if (atomic_read(&substream->mmap_count)) goto __direct; - if ((err = snd_pcm_oss_make_ready(substream)) < 0) - return err; atomic_inc(&runtime->oss.rw_ref); if (mutex_lock_interruptible(&runtime->oss.params_lock)) { atomic_dec(&runtime->oss.rw_ref); return -ERESTARTSYS; } + err = snd_pcm_oss_make_ready_locked(substream); + if (err < 0) + goto unlock; format = snd_pcm_oss_format_from(runtime->oss.format); width = snd_pcm_format_physical_width(format); if (runtime->oss.buffer_used > 0) {
From: Kunkun Jiang jiangkunkun@huawei.com
virt inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I5UN39 CVE: NA
------------------------------
Two redundant arguements are added to 'iommu_device_register' of 'include/linux/iommu.h' in the commit bbf3b39e5576 ("iommu: Introduce dirty log tracking framework"). As a result, compiling the kernel fails when the CONFIG_IOMMU_API is disabled. Delete the two redundant arguements to solve this problem.
Fixes: bbf3b39e5576 ("iommu: Introduce dirty log tracking framework") Reported-by: Yejian Zheng zhengyejian1@huawei.com Signed-off-by: Kunkun Jiang jiangkunkun@huawei.com Reviewed-by: Keqian Zhu zhukeqian1@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- include/linux/iommu.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/include/linux/iommu.h b/include/linux/iommu.h index 092384b71ab2..d993036c94c2 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -1031,9 +1031,7 @@ static inline int iommu_clear_dirty_log(struct iommu_domain *domain, return -EINVAL; }
-static inline int iommu_device_register(struct iommu_device *iommu, - const struct iommu_ops *ops, - struct device *hwdev) +static inline int iommu_device_register(struct iommu_device *iommu) { return -ENODEV; }
From: Ding Xiang dingxiang@cmss.chinamobile.com
stable inclusion from stable-v5.10.124 commit 0ca74dacfd478a9c03361b7512be48ea6b22e8a5 category: bugfix bugzilla: 186639, https://gitee.com/openeuler/kernel/issues/I58WSQ CVE: CVE-2022-1184
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=...
--------------------------------
commit bc75a6eb856cb1507fa907bf6c1eda91b3fef52f upstream.
Since dx_make_map() may return -EFSCORRUPTED now, so change "count" to be a signed integer so we can correctly check for an error code returned by dx_make_map().
Fixes: 46c116b920eb ("ext4: verify dir block before splitting it") Cc: stable@kernel.org Signed-off-by: Ding Xiang dingxiang@cmss.chinamobile.com Link: https://lore.kernel.org/r/20220530100047.537598-1-dingxiang@cmss.chinamobile... Signed-off-by: Theodore Ts'o tytso@mit.edu Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Baokun Li libaokun1@huawei.com Reviewed-by: Zhang Yi yi.zhang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- fs/ext4/namei.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index df71228d34e0..933ec7c324d5 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1848,7 +1848,8 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct dx_hash_info *hinfo) { unsigned blocksize = dir->i_sb->s_blocksize; - unsigned count, continued; + unsigned continued; + int count; struct buffer_head *bh2; ext4_lblk_t newblock; u32 hash2;
From: Lukas Czerner lczerner@redhat.com
mainline inclusion from mainline-v6.0-rc1 commit 65f8ea4cd57dbd46ea13b41dc8bac03176b04233 category: bugfix bugzilla: 186639, https://gitee.com/openeuler/kernel/issues/I58WSQ CVE: CVE-2022-1184
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
Currently ext4 directory handling code implicitly assumes that the directory blocks are always within the i_size. In fact ext4_append() will attempt to allocate next directory block based solely on i_size and the i_size is then appropriately increased after a successful allocation.
However, for this to work it requires i_size to be correct. If, for any reason, the directory inode i_size is corrupted in a way that the directory tree refers to a valid directory block past i_size, we could end up corrupting parts of the directory tree structure by overwriting already used directory blocks when modifying the directory.
Fix it by catching the corruption early in __ext4_read_dirblock().
Addresses Red-Hat-Bugzilla: #2070205 CVE: CVE-2022-1184 Signed-off-by: Lukas Czerner lczerner@redhat.com Cc: stable@vger.kernel.org Reviewed-by: Andreas Dilger adilger@dilger.ca Link: https://lore.kernel.org/r/20220704142721.157985-1-lczerner@redhat.com Signed-off-by: Theodore Ts'o tytso@mit.edu Signed-off-by: Baokun Li libaokun1@huawei.com Reviewed-by: Zhang Yi yi.zhang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- fs/ext4/namei.c | 7 +++++++ 1 file changed, 7 insertions(+)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 933ec7c324d5..c7279c20dc06 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -109,6 +109,13 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode, struct ext4_dir_entry *dirent; int is_dx_block = 0;
+ if (block >= inode->i_size) { + ext4_error_inode(inode, func, line, block, + "Attempting to read directory block (%u) that is past i_size (%llu)", + block, inode->i_size); + return ERR_PTR(-EFSCORRUPTED); + } + if (ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_EIO)) bh = ERR_PTR(-EIO); else
From: Lukas Czerner lczerner@redhat.com
mainline inclusion from mainline-v6.0-rc1 commit b8a04fe77ef1360fbf73c80fddbdfeaa9407ed1b category: bugfix bugzilla: 186639, https://gitee.com/openeuler/kernel/issues/I58WSQ CVE: CVE-2022-1184
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
ext4_append() must always allocate a new block, otherwise we run the risk of overwriting existing directory block corrupting the directory tree in the process resulting in all manner of problems later on.
Add a sanity check to see if the logical block is already allocated and error out if it is.
Cc: stable@kernel.org Signed-off-by: Lukas Czerner lczerner@redhat.com Reviewed-by: Andreas Dilger adilger@dilger.ca Link: https://lore.kernel.org/r/20220704142721.157985-2-lczerner@redhat.com Signed-off-by: Theodore Ts'o tytso@mit.edu Signed-off-by: Baokun Li libaokun1@huawei.com Reviewed-by: Zhang Yi yi.zhang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- fs/ext4/namei.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index c7279c20dc06..af237deb5ac3 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -54,6 +54,7 @@ static struct buffer_head *ext4_append(handle_t *handle, struct inode *inode, ext4_lblk_t *block) { + struct ext4_map_blocks map; struct buffer_head *bh; int err;
@@ -63,6 +64,21 @@ static struct buffer_head *ext4_append(handle_t *handle, return ERR_PTR(-ENOSPC);
*block = inode->i_size >> inode->i_sb->s_blocksize_bits; + map.m_lblk = *block; + map.m_len = 1; + + /* + * We're appending new directory block. Make sure the block is not + * allocated yet, otherwise we will end up corrupting the + * directory. + */ + err = ext4_map_blocks(NULL, inode, &map, 0); + if (err < 0) + return ERR_PTR(err); + if (err) { + EXT4_ERROR_INODE(inode, "Logical block already allocated"); + return ERR_PTR(-EFSCORRUPTED); + }
bh = ext4_bread(handle, inode, *block, EXT4_GET_BLOCKS_CREATE); if (IS_ERR(bh))
From: Jan Kara jack@suse.cz
mainline inclusion from mainline-v6.1-rc1 commit 61a1d87a324ad5e3ed27c6699dfc93218fcf3201 category: bugfix bugzilla: 186639, https://gitee.com/openeuler/kernel/issues/I58WSQ CVE: CVE-2022-1184
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
The check in __ext4_read_dirblock() for block being outside of directory size was wrong because it compared block number against directory size in bytes. Fix it.
Fixes: 65f8ea4cd57d ("ext4: check if directory block is within i_size") CVE: CVE-2022-1184 CC: stable@vger.kernel.org Signed-off-by: Jan Kara jack@suse.cz Reviewed-by: Lukas Czerner lczerner@redhat.com Link: https://lore.kernel.org/r/20220822114832.1482-1-jack@suse.cz Signed-off-by: Theodore Ts'o tytso@mit.edu Signed-off-by: Baokun Li libaokun1@huawei.com Reviewed-by: Zhang Yi yi.zhang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- fs/ext4/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index af237deb5ac3..e704d2489fac 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -125,7 +125,7 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode, struct ext4_dir_entry *dirent; int is_dx_block = 0;
- if (block >= inode->i_size) { + if (block >= inode->i_size >> inode->i_blkbits) { ext4_error_inode(inode, func, line, block, "Attempting to read directory block (%u) that is past i_size (%llu)", block, inode->i_size);
From: Zhihao Cheng chengzhihao1@huawei.com
mainline inclusion from mainline-v6.0-rc8 commit 7177dd009c7c04290891e9a534cd47d1b620bd04 category: bugfix bugzilla: 187586, https://gitee.com/openeuler/kernel/issues/I5V3VI CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/f...
--------------------------------
Following process may lead to fs corruption: 1. ext4_create(dir/foo) ext4_add_nondir ext4_add_entry ext4_dx_add_entry a. add_dirent_to_buf ext4_mark_inode_dirty ext4_handle_dirty_metadata // dir inode bh is recorded into journal b. ext4_append // dx_get_count(entries) == dx_get_limit(entries) ext4_bread(EXT4_GET_BLOCKS_CREATE) ext4_getblk ext4_map_blocks ext4_ext_map_blocks ext4_mb_new_blocks dquot_alloc_block dquot_alloc_space_nodirty inode_add_bytes // update dir's i_blocks ext4_ext_insert_extent ext4_ext_dirty // record extent bh into journal ext4_handle_dirty_metadata(bh) // record new block into journal inode->i_size += inode->i_sb->s_blocksize // new size(in mem) c. ext4_handle_dirty_dx_node(bh2) // record dir's new block(dx_node) into journal d. ext4_handle_dirty_dx_node((frame - 1)->bh) e. ext4_handle_dirty_dx_node(frame->bh) f. do_split // ret err! g. add_dirent_to_buf ext4_mark_inode_dirty(dir) // update raw_inode on disk(skipped) 2. fsck -a /dev/sdb drop last block(dx_node) which beyonds dir's i_size. /dev/sdb: recovering journal /dev/sdb contains a file system with errors, check forced. /dev/sdb: Inode 12, end of extent exceeds allowed value (logical block 128, physical block 3938, len 1) 3. fsck -fn /dev/sdb dx_node->entry[i].blk > dir->i_size Pass 2: Checking directory structure Problem in HTREE directory inode 12 (/dir): bad block number 128. Clear HTree index? no Problem in HTREE directory inode 12: block #3 has invalid depth (2) Problem in HTREE directory inode 12: block #3 has bad max hash Problem in HTREE directory inode 12: block #3 not referenced
Fix it by marking inode dirty directly inside ext4_append(). Fetch a reproducer in [Link].
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216466 Cc: stable@vger.kernel.org Conflicts: fs/ext4/namei.c [ 188c299e2a26cc3("ext4: Support for checksumming from journal triggers") is not applied. ] Signed-off-by: Zhihao Cheng chengzhihao1@huawei.com Reviewed-by: Jan Kara jack@suse.cz Link: https://lore.kernel.org/r/20220911045204.516460-1-chengzhihao1@huawei.com Signed-off-by: Theodore Ts'o tytso@mit.edu Reviewed-by: Zhang Yi yi.zhang@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- fs/ext4/namei.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index e704d2489fac..f4dd3a7ee965 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -85,14 +85,19 @@ static struct buffer_head *ext4_append(handle_t *handle, return bh; inode->i_size += inode->i_sb->s_blocksize; EXT4_I(inode)->i_disksize = inode->i_size; + err = ext4_mark_inode_dirty(handle, inode); + if (err) + goto out; BUFFER_TRACE(bh, "get_write_access"); err = ext4_journal_get_write_access(handle, bh); - if (err) { - brelse(bh); - ext4_std_error(inode->i_sb, err); - return ERR_PTR(err); - } + if (err) + goto out; return bh; + +out: + brelse(bh); + ext4_std_error(inode->i_sb, err); + return ERR_PTR(err); }
static int ext4_dx_csum_verify(struct inode *inode,
From: David Leadbeater dgl@dgl.cx
stable inclusion from stable-v5.10.143 commit e12ce30fe593dd438c5b392290ad7316befc11ca category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I5OWZ7 CVE: CVE-2022-2663
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=...
---------------------------
[ Upstream commit 0efe125cfb99e6773a7434f3463f7c2fa28f3a43 ]
Ensure the match happens in the right direction, previously the destination used was the server, not the NAT host, as the comment shows the code intended.
Additionally nf_nat_irc uses port 0 as a signal and there's no valid way it can appear in a DCC message, so consider port 0 also forged.
Fixes: 869f37d8e48f ("[NETFILTER]: nf_conntrack/nf_nat: add IRC helper port") Signed-off-by: David Leadbeater dgl@dgl.cx Signed-off-by: Pablo Neira Ayuso pablo@netfilter.org Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Liu Jian liujian56@huawei.com Reviewed-by: Yue Haibing yuehaibing@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com --- net/netfilter/nf_conntrack_irc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c index c745dd19d42f..65b5b05fe38d 100644 --- a/net/netfilter/nf_conntrack_irc.c +++ b/net/netfilter/nf_conntrack_irc.c @@ -207,8 +207,9 @@ static int help(struct sk_buff *skb, unsigned int protoff,
/* dcc_ip can be the internal OR external (NAT'ed) IP */ tuple = &ct->tuplehash[dir].tuple; - if (tuple->src.u3.ip != dcc_ip && - tuple->dst.u3.ip != dcc_ip) { + if ((tuple->src.u3.ip != dcc_ip && + ct->tuplehash[!dir].tuple.dst.u3.ip != dcc_ip) || + dcc_port == 0) { net_warn_ratelimited("Forged DCC command from %pI4: %pI4:%u\n", &tuple->src.u3.ip, &dcc_ip, dcc_port);
-
Zheng Zengkai