- Kernel - mailweb.openeuler.org

[PATCH OLK-5.10] RDMA/hns: Fix mbox timing out by adding retry mechanism
by Junxian Huang 12 Mar '25

12 Mar '25

maillist inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSLDS CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?h=wip… ---------------------------------------------------------------------- If a QP is modified to error state and a flush CQE process is triggered, the subsequent QP destruction mbox can still be successfully posted but will be blocked in HW until the flush CQE process finishes. This causes further mbox posting timeouts in driver. The blocking time is related to QP depth. Considering an extreme case where SQ depth and RQ depth are both 32K, the blocking time can reach about 135ms. This patch adds a retry mechanism for mbox posting. For each try, FW waits 15ms for HW to complete the previous mbox, otherwise return a timeout error code to driver. Counting other time consumption in FW, set 8 tries for mbox posting and a 5ms time gap before each retry to increase to a sufficient timeout limit. Fixes: 0425e3e6e0c7 ("RDMA/hns: Support flush cqe for hip08 in kernel space") Signed-off-by: Junxian Huang <huangjunxian6(a)hisilicon.com> Link: https://patch.msgid.link/20250208105930.522796-1-huangjunxian6@hisilicon.com Signed-off-by: Leon Romanovsky <leon(a)kernel.org> Signed-off-by: Xinghai Cen <cenxinghai(a)h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 64 ++++++++++++++++------ drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 2 + 2 files changed, 50 insertions(+), 16 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 88a10930447b..3813b318f1c3 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -1395,10 +1395,8 @@ static u32 hns_roce_cmdq_tx_timeout(u16 opcode, u32 tx_timeout) return tx_timeout; } -static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u16 opcode) +static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u32 tx_timeout) { - struct hns_roce_v2_priv *priv = hr_dev->priv; - u32 tx_timeout = hns_roce_cmdq_tx_timeout(opcode, priv->cmq.tx_timeout); u32 timeout = 0; do { @@ -1408,8 +1406,9 @@ static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u16 opcode) } while (++timeout < tx_timeout); } -static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, - struct hns_roce_cmq_desc *desc, int num) +static int __hns_roce_cmq_send_one(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc, + int num, u32 tx_timeout) { struct hns_roce_v2_priv *priv = hr_dev->priv; struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq; @@ -1418,8 +1417,6 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, int ret; int i; - spin_lock_bh(&csq->lock); - tail = csq->head; for (i = 0; i < num; i++) { @@ -1433,22 +1430,17 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, atomic64_inc(&hr_dev->dfx_cnt[HNS_ROCE_DFX_CMDS_CNT]); - hns_roce_wait_csq_done(hr_dev, le16_to_cpu(desc->opcode)); + hns_roce_wait_csq_done(hr_dev, tx_timeout); if (hns_roce_cmq_csq_done(hr_dev)) { ret = 0; for (i = 0; i < num; i++) { /* check the result of hardware write back */ - desc[i] = csq->desc[tail++]; + desc_ret = le16_to_cpu(csq->desc[tail++].retval); if (tail == csq->desc_num) tail = 0; - - desc_ret = le16_to_cpu(desc[i].retval); if (likely(desc_ret == CMD_EXEC_SUCCESS)) continue; - dev_err_ratelimited(hr_dev->dev, - "Cmdq IO error, opcode = 0x%x, return = 0x%x.\n", - desc->opcode, desc_ret); ret = hns_roce_cmd_err_convert_errno(desc_ret); } } else { @@ -1463,14 +1455,54 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, ret = -EAGAIN; } - spin_unlock_bh(&csq->lock); - if (ret) atomic64_inc(&hr_dev->dfx_cnt[HNS_ROCE_DFX_CMDS_ERR_CNT]); return ret; } +static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc, int num) +{ + struct hns_roce_v2_priv *priv = hr_dev->priv; + struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq; + u16 opcode = le16_to_cpu(desc->opcode); + u32 tx_timeout = hns_roce_cmdq_tx_timeout(opcode, priv->cmq.tx_timeout); + u8 try_cnt = HNS_ROCE_OPC_POST_MB_TRY_CNT; + u32 rsv_tail; + int ret; + int i; + + while (try_cnt) { + try_cnt--; + + spin_lock_bh(&csq->lock); + rsv_tail = csq->head; + ret = __hns_roce_cmq_send_one(hr_dev, desc, num, tx_timeout); + if (opcode == HNS_ROCE_OPC_POST_MB && ret == -ETIME && + try_cnt) { + spin_unlock_bh(&csq->lock); + mdelay(HNS_ROCE_OPC_POST_MB_RETRY_GAP_MSEC); + continue; + } + + for (i = 0; i < num; i++) { + desc[i] = csq->desc[rsv_tail++]; + if (rsv_tail == csq->desc_num) + rsv_tail = 0; + } + spin_unlock_bh(&csq->lock); + break; + } + + if (ret) + dev_err_ratelimited(hr_dev->dev, + "Cmdq IO error, opcode = 0x%x, return = %d.\n", + opcode, ret); + + return ret; +} + static int hns_roce_cmq_send(struct hns_roce_dev *hr_dev, struct hns_roce_cmq_desc *desc, int num) { diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index c59e40649454..dbc405885ea7 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -243,6 +243,8 @@ enum hns_roce_opcode_type { }; #define HNS_ROCE_OPC_POST_MB_TIMEOUT 35000 +#define HNS_ROCE_OPC_POST_MB_TRY_CNT 8 +#define HNS_ROCE_OPC_POST_MB_RETRY_GAP_MSEC 5 struct hns_roce_cmdq_tx_timeout_map { u16 opcode; u32 tx_timeout; -- 2.33.0

2 1

[PATCH OLK-6.6] perf: Configure BRBE correctly on VHE host
by Yushan Wang 12 Mar '25

12 Mar '25

From: Yicong Yang <yangyicong(a)hisilicon.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSI0M CVE: NA ---------------------------------------------------------------------- The BRBE support backported of v18 didn't configure the BRBCR_EL1 correctly which will miss some key functions like FZP during sampling. In VHE mode with MDCR_EL2.HPMN set to PMCR_EL0.N, the counters are controlled by BRBCR_EL1 rather than BRBCR_EL2 (which writes to BRBCR_EL1 are redirected to). Use the same value for both register except keep EL1 and EL0 recording disabled in guests. The fix is backported from the v19 version [1]. [1] https://lore.kernel.org/linux-arm-kernel/ 20250202-arm-brbe-v19-v19-0-1c1300802385(a)kernel.org/ T/#m887b1dcf7f6784b7293626952052f27b68784beb fixes: 6c6848e7e00c ("drivers: perf: arm_pmuv3: Enable branch stack sampling via FEAT_BRBE") Signed-off-by: Yicong Yang <yangyicong(a)hisilicon.com> Signed-off-by: Qizhi Zhang <zhangqizhi3(a)h-partners.com> --- drivers/perf/arm_brbe.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/perf/arm_brbe.c b/drivers/perf/arm_brbe.c index d795e8fd646f..22e4828feec8 100644 --- a/drivers/perf/arm_brbe.c +++ b/drivers/perf/arm_brbe.c @@ -874,6 +874,16 @@ void armv8pmu_branch_enable(struct arm_pmu *arm_pmu) brbcr &= ~BRBCR_ELx_CONFIG_MASK; brbcr |= branch_type_to_brbcr(cpuc->branch_sample_type); write_sysreg_s(brbcr, SYS_BRBCR_EL1); + + /* + * In VHE mode with MDCR_EL2.HPMN set to PMCR_EL0.N, the counters are + * controlled by BRBCR_EL1 rather than BRBCR_EL2 (which writes to + * BRBCR_EL1 are redirected to). Use the same value for both register + * except keep EL1 and EL0 recording disabled in guests. + */ + if (is_kernel_in_hyp_mode()) + write_sysreg_s(brbcr & ~(BRBCR_ELx_ExBRE | BRBCR_ELx_E0BRE), SYS_BRBCR_EL12); + isb(); } @@ -886,6 +896,11 @@ void armv8pmu_branch_disable(void) brbcr &= ~(BRBCR_ELx_E0BRE | BRBCR_ELx_ExBRE); brbfcr |= BRBFCR_EL1_PAUSED; write_sysreg_s(brbcr, SYS_BRBCR_EL1); + + /* See the comment in armv8pmu_branch_enable() */ + if (is_kernel_in_hyp_mode()) + write_sysreg_s(brbcr, SYS_BRBCR_EL12); + write_sysreg_s(brbfcr, SYS_BRBFCR_EL1); isb(); } -- 2.33.0

2 1

[openeuler:OLK-6.6] BUILD REGRESSION 2f1a9354fdf9708fa38bd6916f84034bc1510c16
by kernel test robot 12 Mar '25

12 Mar '25

tree/branch: https://gitee.com/openeuler/kernel.git OLK-6.6 branch HEAD: 2f1a9354fdf9708fa38bd6916f84034bc1510c16 !15423 [openEuler-24.03-LTS][linux-6.6.y sync] Backport 6.6.69-6.6.70 LTS Additional Patches Error/Warning ids grouped by kconfigs: recent_errors |-- arm64-alldefconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-allmodconfig | |-- include-linux-fortify-string.h:warning:call-to-__write_overflow_field-declared-with-warning-attribute:detected-write-beyond-size-of-field-(1st-parameter)-maybe-use-struct_group() | |-- mm-memcontrol.c:warning:no-previous-prototype-for-function-hisi_oom_recover | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_limit_mbytes_sysctl_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_reclaim_enable_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_reclaim_sysctl_handler | |-- mm-share_pool.c:warning:Function-parameter-or-member-node_id-not-described-in-sp_area_alloc | |-- mm-share_pool.c:warning:Function-parameter-or-member-spg_id-not-described-in-mg_sp_unshare | |-- mm-share_pool.c:warning:duplicate-section-name-Return | |-- mm-share_pool.c:warning:expecting-prototype-for-mp_sp_group_id_by_pid().-Prototype-was-for-mg_sp_group_id_by_pid()-instead | |-- mm-share_pool.c:warning:variable-is_hugepage-set-but-not-used | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-allnoconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-randconfig-001-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-randconfig-002-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-randconfig-003-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-randconfig-004-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- arm64-randconfig-051-20250312 | |-- Documentation-devicetree-bindings-dma-snps-dw-axi-dmac.yaml:snps-priority:multiple-incompatible-types:uint32-array | `-- Documentation-devicetree-bindings-net-snps-dwmac.yaml:snps-priority:multiple-incompatible-types:uint32 |-- arm64-randconfig-052-20250312 | |-- Documentation-devicetree-bindings-dma-snps-dw-axi-dmac.yaml:snps-priority:multiple-incompatible-types:uint32-array | `-- Documentation-devicetree-bindings-net-snps-dwmac.yaml:snps-priority:multiple-incompatible-types:uint32 |-- arm64-randconfig-053-20250312 | |-- Documentation-devicetree-bindings-dma-snps-dw-axi-dmac.yaml:snps-priority:multiple-incompatible-types:uint32-array | `-- Documentation-devicetree-bindings-net-snps-dwmac.yaml:snps-priority:multiple-incompatible-types:uint32 |-- arm64-randconfig-054-20250312 | |-- Documentation-devicetree-bindings-dma-snps-dw-axi-dmac.yaml:snps-priority:multiple-incompatible-types:uint32-array | `-- Documentation-devicetree-bindings-net-snps-dwmac.yaml:snps-priority:multiple-incompatible-types:uint32 |-- arm64-randconfig-055-20250312 | |-- Documentation-devicetree-bindings-dma-snps-dw-axi-dmac.yaml:snps-priority:multiple-incompatible-types:uint32-array | `-- Documentation-devicetree-bindings-net-snps-dwmac.yaml:snps-priority:multiple-incompatible-types:uint32 |-- loongarch-allmodconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_limit_mbytes_sysctl_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_reclaim_enable_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_reclaim_sysctl_handler | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- loongarch-allnoconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- loongarch-allyesconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_limit_mbytes_sysctl_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_reclaim_enable_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-cache_reclaim_sysctl_handler | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- loongarch-randconfig-001-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- loongarch-randconfig-002-20250311 | `-- include-linux-mmzone.h:error:error-Allocator-MAX_ORDER-exceeds-SECTION_SIZE |-- x86_64-allnoconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-allyesconfig | |-- include-linux-fortify-string.h:warning:call-to-__write_overflow_field-declared-with-warning-attribute:detected-write-beyond-size-of-field-(1st-parameter)-maybe-use-struct_group() | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_limit_mbytes_sysctl_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_reclaim_enable_handler | |-- mm-page_cache_limit.c:warning:no-previous-prototype-for-function-cache_reclaim_sysctl_handler | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-001-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-002-20250311 | |-- drivers-cpufreq-acpi-cpufreq.c:warning:sched_set_itmt-defined-but-not-used | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-003-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-004-20250311 | |-- ld.lld:error:undefined-symbol:arch_kexec_protect_crashkres | |-- ld.lld:error:undefined-symbol:machine_crash_shutdown | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-005-20250311 | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-buildonly-randconfig-006-20250311 | |-- drivers-cpufreq-acpi-cpufreq.c:warning:sched_set_itmt-defined-but-not-used | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial |-- x86_64-defconfig | |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used | |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task | |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead | `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial `-- x86_64-randconfig-161-20250312 |-- include-linux-fortify-string.h:warning:call-to-__write_overflow_field-declared-with-warning-attribute:detected-write-beyond-size-of-field-(1st-parameter)-maybe-use-struct_group() |-- mm-compaction.c-(null)()-warn:(struct-ctl_table)-proc_handler-cannot-be-NULL.-Expression:vm_compaction-proc_handler |-- mm-compaction.c-(null)()-warn:(struct-ctl_table)-procname-cannot-be-NULL.-Expression:vm_compaction-procname |-- mm-memory.c:warning:variable-nr_pages-set-but-not-used |-- mm-oom_kill.c:warning:Function-parameter-or-member-oc-not-described-in-oom_next_task |-- mm-oom_kill.c:warning:Function-parameter-or-member-points-not-described-in-oom_next_task |-- mm-oom_kill.c:warning:Function-parameter-or-member-task-not-described-in-oom_next_task |-- mm-oom_kill.c:warning:expecting-prototype-for-We-choose-the-task-in-low().-Prototype-was-for-oom_next_task()-instead `-- mm-vmalloc.c:warning:Function-parameter-or-member-pgoff-not-described-in-remap_vmalloc_hugepage_range_partial elapsed time: 823m configs tested: 20 configs skipped: 123 tested configs: arm64 alldefconfig gcc-14.2.0 arm64 allmodconfig clang-18 arm64 allnoconfig gcc-14.2.0 arm64 randconfig-001-20250311 gcc-14.2.0 arm64 randconfig-002-20250311 gcc-14.2.0 arm64 randconfig-003-20250311 gcc-14.2.0 arm64 randconfig-004-20250311 gcc-14.2.0 loongarch allmodconfig gcc-14.2.0 loongarch allnoconfig gcc-14.2.0 loongarch randconfig-001-20250311 gcc-14.2.0 loongarch randconfig-002-20250311 gcc-14.2.0 x86_64 allnoconfig clang-19 x86_64 allyesconfig clang-19 x86_64 buildonly-randconfig-001-20250311 gcc-12 x86_64 buildonly-randconfig-002-20250311 gcc-12 x86_64 buildonly-randconfig-003-20250311 clang-19 x86_64 buildonly-randconfig-004-20250311 clang-19 x86_64 buildonly-randconfig-005-20250311 gcc-12 x86_64 buildonly-randconfig-006-20250311 gcc-12 x86_64 defconfig gcc-11 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6 0/4] some readahead improvement
by Tong Tiangen 11 Mar '25

11 Mar '25

Jan Kara (3): readahead: make sure sync readahead reads needed page readahead: don't shorten readahead window in read_pages() readahead: properly shorten readahead when falling back to do_page_cache_ra() Yafang Shao (1): mm/readahead: fix large folio support in async readahead mm/readahead.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-) -- 2.25.1

2 5

[PATCH openEuler-1.0-LTS] hrtimers: Handle CPU state correctly on hotplug
by Zeng Heng 11 Mar '25

11 Mar '25

From: Koichiro Den <koichiro.den(a)canonical.com> stable inclusion from stable-v6.6.74 commit a5cbbea145b400e40540c34816d16d36e0374fbc category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBLWTN CVE: CVE-2024-57951 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 2f8dea1692eef2b7ba6a256246ed82c365fdc686 upstream. Consider a scenario where a CPU transitions from CPUHP_ONLINE to halfway through a CPU hotunplug down to CPUHP_HRTIMERS_PREPARE, and then back to CPUHP_ONLINE: Since hrtimers_prepare_cpu() does not run, cpu_base.hres_active remains set to 1 throughout. However, during a CPU unplug operation, the tick and the clockevents are shut down at CPUHP_AP_TICK_DYING. On return to the online state, for instance CFS incorrectly assumes that the hrtick is already active, and the chance of the clockevent device to transition to oneshot mode is also lost forever for the CPU, unless it goes back to a lower state than CPUHP_HRTIMERS_PREPARE once. This round-trip reveals another issue; cpu_base.online is not set to 1 after the transition, which appears as a WARN_ON_ONCE in enqueue_hrtimer(). Aside of that, the bulk of the per CPU state is not reset either, which means there are dangling pointers in the worst case. Address this by adding a corresponding startup() callback, which resets the stale per CPU state and sets the online flag. [ tglx: Make the new callback unconditionally available, remove the online modification in the prepare() callback and clear the remaining state in the starting callback instead of the prepare callback ] Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Signed-off-by: Koichiro Den <koichiro.den(a)canonical.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/all/20241220134421.3809834-1-koichiro.den@canonical… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/hrtimer.h kernel/cpu.c kernel/time/hrtimer.c [Fix context conflict.] Signed-off-by: Zeng Heng <zengheng4(a)huawei.com> --- include/linux/hrtimer.h | 1 + kernel/time/hrtimer.c | 9 +++++++++ 2 files changed, 10 insertions(+) diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 542b4fa2cda9..27f504fc36d2 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -507,6 +507,7 @@ extern void __init hrtimers_init(void); extern void sysrq_timer_list_show(void); int hrtimers_prepare_cpu(unsigned int cpu); +int hrtimers_cpu_starting(unsigned int cpu); #ifdef CONFIG_HOTPLUG_CPU int hrtimers_dead_cpu(unsigned int cpu); #else diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index ce8fe5adafb0..c7817f3e814e 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1882,6 +1882,14 @@ int hrtimers_prepare_cpu(unsigned int cpu) } cpu_base->cpu = cpu; + return 0; +} + +int hrtimers_cpu_starting(unsigned int cpu) +{ + struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); + + /* Clear out any left over state from a CPU down operation */ cpu_base->active_bases = 0; cpu_base->hres_active = 0; cpu_base->hang_detected = 0; @@ -1974,6 +1982,7 @@ int hrtimers_dead_cpu(unsigned int scpu) void __init hrtimers_init(void) { hrtimers_prepare_cpu(smp_processor_id()); + hrtimers_cpu_starting(smp_processor_id()); open_softirq(HRTIMER_SOFTIRQ, hrtimer_run_softirq); } -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] hrtimers: Handle CPU state correctly on hotplug
by Zeng Heng 11 Mar '25

11 Mar '25

From: Koichiro Den <koichiro.den(a)canonical.com> stable inclusion from stable-v6.6.74 commit a5cbbea145b400e40540c34816d16d36e0374fbc category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBLWTN CVE: CVE-2024-57951 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 2f8dea1692eef2b7ba6a256246ed82c365fdc686 upstream. Consider a scenario where a CPU transitions from CPUHP_ONLINE to halfway through a CPU hotunplug down to CPUHP_HRTIMERS_PREPARE, and then back to CPUHP_ONLINE: Since hrtimers_prepare_cpu() does not run, cpu_base.hres_active remains set to 1 throughout. However, during a CPU unplug operation, the tick and the clockevents are shut down at CPUHP_AP_TICK_DYING. On return to the online state, for instance CFS incorrectly assumes that the hrtick is already active, and the chance of the clockevent device to transition to oneshot mode is also lost forever for the CPU, unless it goes back to a lower state than CPUHP_HRTIMERS_PREPARE once. This round-trip reveals another issue; cpu_base.online is not set to 1 after the transition, which appears as a WARN_ON_ONCE in enqueue_hrtimer(). Aside of that, the bulk of the per CPU state is not reset either, which means there are dangling pointers in the worst case. Address this by adding a corresponding startup() callback, which resets the stale per CPU state and sets the online flag. [ tglx: Make the new callback unconditionally available, remove the online modification in the prepare() callback and clear the remaining state in the starting callback instead of the prepare callback ] Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Signed-off-by: Koichiro Den <koichiro.den(a)canonical.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/all/20241220134421.3809834-1-koichiro.den@canonical… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/hrtimer.h kernel/cpu.c kernel/time/hrtimer.c [Fix context conflict.] Signed-off-by: Zeng Heng <zengheng4(a)huawei.com> --- include/linux/hrtimer.h | 1 + kernel/time/hrtimer.c | 10 ++++++++++ 2 files changed, 11 insertions(+) diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 542b4fa2cda9..27f504fc36d2 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -507,6 +507,7 @@ extern void __init hrtimers_init(void); extern void sysrq_timer_list_show(void); int hrtimers_prepare_cpu(unsigned int cpu); +int hrtimers_cpu_starting(unsigned int cpu); #ifdef CONFIG_HOTPLUG_CPU int hrtimers_dead_cpu(unsigned int cpu); #else diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index ce8fe5adafb0..300eb4039f26 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1882,6 +1882,15 @@ int hrtimers_prepare_cpu(unsigned int cpu) } cpu_base->cpu = cpu; + hrtimer_cpu_base_init_expiry_lock(cpu_base); + return 0; +} + +int hrtimers_cpu_starting(unsigned int cpu) +{ + struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); + + /* Clear out any left over state from a CPU down operation */ cpu_base->active_bases = 0; cpu_base->hres_active = 0; cpu_base->hang_detected = 0; @@ -1974,6 +1983,7 @@ int hrtimers_dead_cpu(unsigned int scpu) void __init hrtimers_init(void) { hrtimers_prepare_cpu(smp_processor_id()); + hrtimers_cpu_starting(smp_processor_id()); open_softirq(HRTIMER_SOFTIRQ, hrtimer_run_softirq); } -- 2.25.1

2 1

[PATCH] RDMA/hns: Fix mbox timing out by adding retry mechanism
by Junxian Huang 11 Mar '25

11 Mar '25

maillist inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSLDS CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?h=wip… ---------------------------------------------------------------------- If a QP is modified to error state and a flush CQE process is triggered, the subsequent QP destruction mbox can still be successfully posted but will be blocked in HW until the flush CQE process finishes. This causes further mbox posting timeouts in driver. The blocking time is related to QP depth. Considering an extreme case where SQ depth and RQ depth are both 32K, the blocking time can reach about 135ms. This patch adds a retry mechanism for mbox posting. For each try, FW waits 15ms for HW to complete the previous mbox, otherwise return a timeout error code to driver. Counting other time consumption in FW, set 8 tries for mbox posting and a 5ms time gap before each retry to increase to a sufficient timeout limit. Fixes: 0425e3e6e0c7 ("RDMA/hns: Support flush cqe for hip08 in kernel space") Signed-off-by: Junxian Huang <huangjunxian6(a)hisilicon.com> Link: https://patch.msgid.link/20250208105930.522796-1-huangjunxian6@hisilicon.com Signed-off-by: Leon Romanovsky <leon(a)kernel.org> Signed-off-by: Xinghai Cen <cenxinghai(a)h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 64 ++++++++++++++++------ drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 2 + 2 files changed, 50 insertions(+), 16 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 88a10930447b..3813b318f1c3 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -1395,10 +1395,8 @@ static u32 hns_roce_cmdq_tx_timeout(u16 opcode, u32 tx_timeout) return tx_timeout; } -static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u16 opcode) +static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u32 tx_timeout) { - struct hns_roce_v2_priv *priv = hr_dev->priv; - u32 tx_timeout = hns_roce_cmdq_tx_timeout(opcode, priv->cmq.tx_timeout); u32 timeout = 0; do { @@ -1408,8 +1406,9 @@ static void hns_roce_wait_csq_done(struct hns_roce_dev *hr_dev, u16 opcode) } while (++timeout < tx_timeout); } -static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, - struct hns_roce_cmq_desc *desc, int num) +static int __hns_roce_cmq_send_one(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc, + int num, u32 tx_timeout) { struct hns_roce_v2_priv *priv = hr_dev->priv; struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq; @@ -1418,8 +1417,6 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, int ret; int i; - spin_lock_bh(&csq->lock); - tail = csq->head; for (i = 0; i < num; i++) { @@ -1433,22 +1430,17 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, atomic64_inc(&hr_dev->dfx_cnt[HNS_ROCE_DFX_CMDS_CNT]); - hns_roce_wait_csq_done(hr_dev, le16_to_cpu(desc->opcode)); + hns_roce_wait_csq_done(hr_dev, tx_timeout); if (hns_roce_cmq_csq_done(hr_dev)) { ret = 0; for (i = 0; i < num; i++) { /* check the result of hardware write back */ - desc[i] = csq->desc[tail++]; + desc_ret = le16_to_cpu(csq->desc[tail++].retval); if (tail == csq->desc_num) tail = 0; - - desc_ret = le16_to_cpu(desc[i].retval); if (likely(desc_ret == CMD_EXEC_SUCCESS)) continue; - dev_err_ratelimited(hr_dev->dev, - "Cmdq IO error, opcode = 0x%x, return = 0x%x.\n", - desc->opcode, desc_ret); ret = hns_roce_cmd_err_convert_errno(desc_ret); } } else { @@ -1463,14 +1455,54 @@ static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, ret = -EAGAIN; } - spin_unlock_bh(&csq->lock); - if (ret) atomic64_inc(&hr_dev->dfx_cnt[HNS_ROCE_DFX_CMDS_ERR_CNT]); return ret; } +static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc, int num) +{ + struct hns_roce_v2_priv *priv = hr_dev->priv; + struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq; + u16 opcode = le16_to_cpu(desc->opcode); + u32 tx_timeout = hns_roce_cmdq_tx_timeout(opcode, priv->cmq.tx_timeout); + u8 try_cnt = HNS_ROCE_OPC_POST_MB_TRY_CNT; + u32 rsv_tail; + int ret; + int i; + + while (try_cnt) { + try_cnt--; + + spin_lock_bh(&csq->lock); + rsv_tail = csq->head; + ret = __hns_roce_cmq_send_one(hr_dev, desc, num, tx_timeout); + if (opcode == HNS_ROCE_OPC_POST_MB && ret == -ETIME && + try_cnt) { + spin_unlock_bh(&csq->lock); + mdelay(HNS_ROCE_OPC_POST_MB_RETRY_GAP_MSEC); + continue; + } + + for (i = 0; i < num; i++) { + desc[i] = csq->desc[rsv_tail++]; + if (rsv_tail == csq->desc_num) + rsv_tail = 0; + } + spin_unlock_bh(&csq->lock); + break; + } + + if (ret) + dev_err_ratelimited(hr_dev->dev, + "Cmdq IO error, opcode = 0x%x, return = %d.\n", + opcode, ret); + + return ret; +} + static int hns_roce_cmq_send(struct hns_roce_dev *hr_dev, struct hns_roce_cmq_desc *desc, int num) { diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index c59e40649454..dbc405885ea7 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -243,6 +243,8 @@ enum hns_roce_opcode_type { }; #define HNS_ROCE_OPC_POST_MB_TIMEOUT 35000 +#define HNS_ROCE_OPC_POST_MB_TRY_CNT 8 +#define HNS_ROCE_OPC_POST_MB_RETRY_GAP_MSEC 5 struct hns_roce_cmdq_tx_timeout_map { u16 opcode; u32 tx_timeout; -- 2.33.0

1 0

[PATCH OLK-6.6] vrf: use RCU protection in l3mdev_l3_out()
by Liu Jian 11 Mar '25

11 Mar '25

From: Eric Dumazet <edumazet(a)google.com> stable inclusion from stable-v6.6.79 commit c40cb5c03e37552d6eff963187109e2c3f78ef6f category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBPCA6 CVE: CVE-2025-21791 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ------------------------------------------------- [ Upstream commit 6d0ce46a93135d96b7fa075a94a88fe0da8e8773 ] l3mdev_l3_out() can be called without RCU being held: raw_sendmsg() ip_push_pending_frames() ip_send_skb() ip_local_out() __ip_local_out() l3mdev_ip_out() Add rcu_read_lock() / rcu_read_unlock() pair to avoid a potential UAF. Fixes: a8e3e1a9f020 ("net: l3mdev: Add hook to output path") Signed-off-by: Eric Dumazet <edumazet(a)google.com> Reviewed-by: David Ahern <dsahern(a)kernel.org> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Link: https://patch.msgid.link/20250207135841.1948589-7-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- include/net/l3mdev.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index ba8dc2523c59..eb5ef1b2e059 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -204,10 +204,12 @@ struct sk_buff *l3mdev_l3_out(struct sock *sk, struct sk_buff *skb, u16 proto) if (netif_is_l3_slave(dev)) { struct net_device *master; + rcu_read_lock(); master = netdev_master_upper_dev_get_rcu(dev); if (master && master->l3mdev_ops->l3mdev_l3_out) skb = master->l3mdev_ops->l3mdev_l3_out(master, sk, skb, proto); + rcu_read_unlock(); } return skb; -- 2.34.1

2 1

[PATCH OLK-5.10] vrf: use RCU protection in l3mdev_l3_out()
by Liu Jian 11 Mar '25

11 Mar '25

From: Eric Dumazet <edumazet(a)google.com> mainline inclusion from mainline-v6.14-rc3 commit 6d0ce46a93135d96b7fa075a94a88fe0da8e8773 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBPCA6 CVE: CVE-2025-21791 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------------------------- l3mdev_l3_out() can be called without RCU being held: raw_sendmsg() ip_push_pending_frames() ip_send_skb() ip_local_out() __ip_local_out() l3mdev_ip_out() Add rcu_read_lock() / rcu_read_unlock() pair to avoid a potential UAF. Fixes: a8e3e1a9f020 ("net: l3mdev: Add hook to output path") Signed-off-by: Eric Dumazet <edumazet(a)google.com> Reviewed-by: David Ahern <dsahern(a)kernel.org> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Link: https://patch.msgid.link/20250207135841.1948589-7-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- include/net/l3mdev.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index ba8dc2523c59..eb5ef1b2e059 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -204,10 +204,12 @@ struct sk_buff *l3mdev_l3_out(struct sock *sk, struct sk_buff *skb, u16 proto) if (netif_is_l3_slave(dev)) { struct net_device *master; + rcu_read_lock(); master = netdev_master_upper_dev_get_rcu(dev); if (master && master->l3mdev_ops->l3mdev_l3_out) skb = master->l3mdev_ops->l3mdev_l3_out(master, sk, skb, proto); + rcu_read_unlock(); } return skb; -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] vrf: use RCU protection in l3mdev_l3_out()
by Liu Jian 11 Mar '25

11 Mar '25

From: Eric Dumazet <edumazet(a)google.com> mainline inclusion from mainline-v6.14-rc3 commit 6d0ce46a93135d96b7fa075a94a88fe0da8e8773 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBPCA6 CVE: CVE-2025-21791 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------------------------- l3mdev_l3_out() can be called without RCU being held: raw_sendmsg() ip_push_pending_frames() ip_send_skb() ip_local_out() __ip_local_out() l3mdev_ip_out() Add rcu_read_lock() / rcu_read_unlock() pair to avoid a potential UAF. Fixes: a8e3e1a9f020 ("net: l3mdev: Add hook to output path") Signed-off-by: Eric Dumazet <edumazet(a)google.com> Reviewed-by: David Ahern <dsahern(a)kernel.org> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Link: https://patch.msgid.link/20250207135841.1948589-7-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- include/net/l3mdev.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index a421e2fa06e3..3300f1440158 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -181,10 +181,12 @@ struct sk_buff *l3mdev_l3_out(struct sock *sk, struct sk_buff *skb, u16 proto) if (netif_is_l3_slave(dev)) { struct net_device *master; + rcu_read_lock(); master = netdev_master_upper_dev_get_rcu(dev); if (master && master->l3mdev_ops->l3mdev_l3_out) skb = master->l3mdev_ops->l3mdev_l3_out(master, sk, skb, proto); + rcu_read_unlock(); } return skb; -- 2.34.1

2 1