- Kernel - mailweb.openeuler.org

[PATCH 0/2] EEVDF null pointer
by Zicheng Qu 04 Nov '25

04 Nov '25

Zicheng Qu (2): EEVDF null pointer reproduction zero_vruntime kernel/sched/debug.c | 8 +- kernel/sched/fair.c | 170 +++++++++++++++++----------- kernel/sched/sched.h | 2 +- null_reproduction_test/Makefile | 9 ++ null_reproduction_test/fullcpu.c | 12 ++ null_reproduction_test/make.sh | 17 +++ null_reproduction_test/test.sh | 34 ++++++ null_reproduction_test/test_sched.c | 141 +++++++++++++++++++++++ 8 files changed, 325 insertions(+), 68 deletions(-) create mode 100644 null_reproduction_test/Makefile create mode 100644 null_reproduction_test/fullcpu.c create mode 100755 null_reproduction_test/make.sh create mode 100755 null_reproduction_test/test.sh create mode 100644 null_reproduction_test/test_sched.c -- 2.34.1

1 2

[PATCH openEuler-1.0-LTS] tty: vt: Fix vc_origin buffer copy overflow in fbcon_prepare_logo()
by Chen Jun 04 Nov '25

04 Nov '25

From: Wang ShaoBo <bobo.shaobowang(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID4U8F ---------------------------------------------------------------------- I got some KASAN report as below: BUG: KASAN: slab-use-after-free in fbcon_prepare_logo+0x61e/0xc90 Read of size 14 at addr ffff88812c9a4c38 by task syz.0.3549/19016 CPU: 0 PID: 19016 Comm: syz.0.3549 Not tainted 6.6.0+ #80 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_address_description.constprop.0+0x6b/0x3d0 ? fbcon_prepare_logo+0x61e/0xc90 print_report+0xba/0x280 ? fbcon_prepare_logo+0x61e/0xc90 ? kasan_addr_to_slab+0xd/0xa0 ? fbcon_prepare_logo+0x61e/0xc90 kasan_report+0xaf/0xe0 ? fbcon_prepare_logo+0x61e/0xc90 kasan_check_range+0x100/0x1c0 __asan_memcpy+0x23/0x60 fbcon_prepare_logo+0x61e/0xc90 fbcon_init+0xeb9/0x1db0 ? __pfx_drm_fb_helper_set_par+0x10/0x10 visual_init+0x310/0x5c0 do_bind_con_driver.isra.0+0x627/0xbd0 store_bind+0x60b/0x710 ? __pfx_store_bind+0x10/0x10 dev_attr_store+0x5a/0x90 ? __pfx_dev_attr_store+0x10/0x10 sysfs_kf_write+0x145/0x1b0 kernfs_fop_write_iter+0x367/0x580 ? __pfx_sysfs_kf_write+0x10/0x10 new_sync_write+0x1b1/0x2d0 ? __pfx_new_sync_write+0x10/0x10 ? rb_commit+0x121/0x910 ? avc_policy_seqno+0xe/0x20 ? selinux_file_permission+0x129/0x5d0 ? security_file_permission+0xa8/0x700 vfs_write+0x71a/0x960 ksys_write+0x12e/0x260 fbcon_init() -> vc_resize() //success resize vc_origin buffer size=224=7(cols)*2*16(rows) -> bcon_prepare_logo(vc, info, old_cols, old_rows, new_cols, new_rows) //old_cols=256，old_rows=4，new_cols=7，new_rows=16 There happened to be a vc_origin buffer copy overflow error in fbcon_prepare_logo(), scrolling screen down when using old cols after vc resize would trigger out of lower bound of vc_origin buffer. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> Acked-by: Thomas Zimmermann <tzimmermann(a)suse.de> Signed-off-by: Chen Jun <chenjun102(a)huawei.com> --- drivers/video/fbdev/core/fbcon.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index f5178eefec64..83a4949e2497 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -592,6 +592,8 @@ static void fbcon_prepare_logo(struct vc_data *vc, struct fb_info *info, /* We can scroll screen down */ r = q - step - cols; for (cnt = rows - logo_lines; cnt > 0; cnt--) { + if (r < (unsigned short *) vc->vc_origin) + break; scr_memcpyw(r + step, r, vc->vc_size_row); r -= cols; } -- 2.22.0

2 1

[PATCH OLK-5.10 0/3] fix CVE-2025-40080
by Zheng Qixing 04 Nov '25

04 Nov '25

fix CVE-2025-40080 Eric Dumazet (2): net: use sk_is_tcp() in more places nbd: restrict sockets to TCP and UDP John Fastabend (1): bpf: syzkaller found null ptr deref in unix_bpf proto add drivers/block/nbd.c | 8 ++++++++ include/net/sock.h | 10 ++++++++++ net/core/skbuff.c | 6 ++---- net/core/sock.c | 6 ++---- net/core/sock_map.c | 6 ------ 5 files changed, 22 insertions(+), 14 deletions(-) -- 2.39.2

2 4

[PATCH openEuler-1.0-LTS 0/2] *** fix CVE-2023-53454 ***
by Lin Ruifeng 04 Nov '25

04 Nov '25

*** fix CVE-2023-53454 *** Charles Han (1): HID: multitouch: Add NULL check in mt_input_configured Rahul Rameshbabu (1): HID: multitouch: Correct devm device reference for hidinput input_dev name drivers/hid/hid-multitouch.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) -- 2.22.0

2 3

[PATCH openEuler-1.0-LTS] PM: sleep: core: Clear power.must_resume in noirq suspend error path
by Xinyu Zheng 04 Nov '25

04 Nov '25

From: "Rafael J. Wysocki" <rafael.j.wysocki(a)intel.com> mainline inclusion from mainline-v6.18-rc1 commit be82483d1b60baf6747884bd74cb7de484deaf76 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID4AQL Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ---------------------------------------------------------------------- If system suspend is aborted in the "noirq" phase (for instance, due to an error returned by one of the device callbacks), power.is_noirq_suspended will not be set for some devices and device_resume_noirq() will return early for them. Consequently, noirq resume callbacks will not run for them at all because the noirq suspend callbacks have not run for them yet. If any of them has power.must_resume set and late suspend has been skipped for it (due to power.smart_suspend), early resume should be skipped for it either, or its state may become inconsistent (for instance, if the early resume assumes that it will always follow noirq resume). Make that happen by clearing power.must_resume in device_resume_noirq() for devices with power.is_noirq_suspended clear that have been left in suspend by device_suspend_late(), which will subsequently cause device_resume_early() to leave the device in suspend and avoid changing its state. Fixes: 0d4b54c6fee8 ("PM / core: Add LEAVE_SUSPENDED driver flag") Link: https://lore.kernel.org/linux-pm/5d692b81-6f58-4e86-9cb0-ede69a09d799@rowla… Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com> Reviewed-by: Ulf Hansson <ulf.hansson(a)linaro.org> Link: https://patch.msgid.link/3381776.aeNJFYEL58@rafael.j.wysocki Conflicts: drivers/base/power/main.c [context conflict and function rename] Signed-off-by: Xinyu Zheng <zhengxinyu6(a)huawei.com> --- drivers/base/power/main.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index fe7506aae13c..00d9353927ae 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -625,8 +625,20 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (dev->power.syscore || dev->power.direct_complete) goto Out; - if (!dev->power.is_noirq_suspended) + if (!dev->power.is_noirq_suspended) { + /* + * This means that system suspend has been aborted in the noirq + * phase before invoking the noirq suspend callback for the + * device, so if device_suspend_late() has left it in suspend, + * device_resume_early() should leave it in suspend either in + * case the early resume of it depends on the noirq resume that + * has not run. + */ + if ((dev_pm_smart_suspend_and_suspended(dev)) + dev->power.must_resume = false; + goto Out; + } dpm_wait_for_superior(dev, async); -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] tty: vt: Fix vc_origin buffer copy overflow in fbcon_prepare_logo()
by Chen Jun 04 Nov '25

04 Nov '25

From: Wang ShaoBo <bobo.shaobowang(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID4U8F ---------------------------------------------------------------------- I got some KASAN report as below: BUG: KASAN: slab-use-after-free in fbcon_prepare_logo+0x61e/0xc90 Read of size 14 at addr ffff88812c9a4c38 by task syz.0.3549/19016 CPU: 0 PID: 19016 Comm: syz.0.3549 Not tainted 6.6.0+ #80 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_address_description.constprop.0+0x6b/0x3d0 ? fbcon_prepare_logo+0x61e/0xc90 print_report+0xba/0x280 ? fbcon_prepare_logo+0x61e/0xc90 ? kasan_addr_to_slab+0xd/0xa0 ? fbcon_prepare_logo+0x61e/0xc90 kasan_report+0xaf/0xe0 ? fbcon_prepare_logo+0x61e/0xc90 kasan_check_range+0x100/0x1c0 __asan_memcpy+0x23/0x60 fbcon_prepare_logo+0x61e/0xc90 fbcon_init+0xeb9/0x1db0 ? __pfx_drm_fb_helper_set_par+0x10/0x10 visual_init+0x310/0x5c0 do_bind_con_driver.isra.0+0x627/0xbd0 store_bind+0x60b/0x710 ? __pfx_store_bind+0x10/0x10 dev_attr_store+0x5a/0x90 ? __pfx_dev_attr_store+0x10/0x10 sysfs_kf_write+0x145/0x1b0 kernfs_fop_write_iter+0x367/0x580 ? __pfx_sysfs_kf_write+0x10/0x10 new_sync_write+0x1b1/0x2d0 ? __pfx_new_sync_write+0x10/0x10 ? rb_commit+0x121/0x910 ? avc_policy_seqno+0xe/0x20 ? selinux_file_permission+0x129/0x5d0 ? security_file_permission+0xa8/0x700 vfs_write+0x71a/0x960 ksys_write+0x12e/0x260 fbcon_init() -> vc_resize() //success resize vc_origin buffer size=224=7(cols)*2*16(rows) -> bcon_prepare_logo(vc, info, old_cols, old_rows, new_cols, new_rows) //old_cols=256，old_rows=4，new_cols=7，new_rows=16 There happened to be a vc_origin buffer copy overflow error in fbcon_prepare_logo(), scrolling screen down when using old cols after vc resize would trigger out of lower bound of vc_origin buffer. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> Acked-by: Thomas Zimmermann <tzimmermann(a)suse.de> --- drivers/video/fbdev/core/fbcon.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index f5178eefec64..83a4949e2497 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -592,6 +592,8 @@ static void fbcon_prepare_logo(struct vc_data *vc, struct fb_info *info, /* We can scroll screen down */ r = q - step - cols; for (cnt = rows - logo_lines; cnt > 0; cnt--) { + if (r < (unsigned short *) vc->vc_origin) + break; scr_memcpyw(r + step, r, vc->vc_size_row); r -= cols; } -- 2.22.0

2 1

[openeuler:OLK-6.6 3098/3098] kernel/bpf-rvi/common_kfuncs.c:12:32: warning: no previous prototype for 'bpf_mem_cgroup_from_task'
by kernel test robot 04 Nov '25

04 Nov '25

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: a29fc03bd0ddaf7388cf31604ef5bd9807585109 commit: ebd40bbe89d81531c7df459a827d7608c3de57bc [3098/3098] bpf-rvi: Add bpf_mem_cgroup_from_task() kfunc config: x86_64-randconfig-001-20251104 (https://download.01.org/0day-ci/archive/20251104/202511040801.Uk13PqHb-lkp@…) compiler: gcc-14 (Debian 14.2.0-19) 14.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251104/202511040801.Uk13PqHb-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202511040801.Uk13PqHb-lkp@intel.com/ All warnings (new ones prefixed by >>): >> kernel/bpf-rvi/common_kfuncs.c:12:32: warning: no previous prototype for 'bpf_mem_cgroup_from_task' [-Wmissing-prototypes] 12 | __bpf_kfunc struct mem_cgroup *bpf_mem_cgroup_from_task(struct task_struct *p) | ^~~~~~~~~~~~~~~~~~~~~~~~ vim +/bpf_mem_cgroup_from_task +12 kernel/bpf-rvi/common_kfuncs.c 7 8 /* 9 * Common support kfuncs 10 */ 11 > 12 __bpf_kfunc struct mem_cgroup *bpf_mem_cgroup_from_task(struct task_struct *p) 13 { 14 return mem_cgroup_from_task(p); 15 } 16 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-6.6 3075/3075] drivers/ub/ubus/enum.c:1097:6: warning: no previous prototype for function 'ub_enum_clear_ent_list'
by kernel test robot 04 Nov '25

04 Nov '25

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: a29fc03bd0ddaf7388cf31604ef5bd9807585109 commit: 1a64cb7c7288cac262e7b3fd8d4dcd49b8a5bde5 [3075/3075] ub:ubus: Support for UB Bus Controller Enumeration config: arm64-allmodconfig (https://download.01.org/0day-ci/archive/20251104/202511040855.Fu1HC1a9-lkp@…) compiler: clang version 19.1.7 (https://github.com/llvm/llvm-project cd708029e0b2869e80abe31ddb175f7c35361f90) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251104/202511040855.Fu1HC1a9-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202511040855.Fu1HC1a9-lkp@intel.com/ All warnings (new ones prefixed by >>): drivers/ub/ubus/enum.c:996:5: warning: no previous prototype for function 'ub_cfg_read_guid' [-Wmissing-prototypes] 996 | int ub_cfg_read_guid(struct ub_entity *uent) | ^ drivers/ub/ubus/enum.c:996:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 996 | int ub_cfg_read_guid(struct ub_entity *uent) | ^ | static >> drivers/ub/ubus/enum.c:1097:6: warning: no previous prototype for function 'ub_enum_clear_ent_list' [-Wmissing-prototypes] 1097 | void ub_enum_clear_ent_list(struct list_head *dev_list) | ^ drivers/ub/ubus/enum.c:1097:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 1097 | void ub_enum_clear_ent_list(struct list_head *dev_list) | ^ | static drivers/ub/ubus/enum.c:1187:5: warning: no previous prototype for function 'ub_enum_probe' [-Wmissing-prototypes] 1187 | int ub_enum_probe(void) | ^ drivers/ub/ubus/enum.c:1187:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 1187 | int ub_enum_probe(void) | ^ | static drivers/ub/ubus/enum.c:1211:6: warning: no previous prototype for function 'ub_enum_remove' [-Wmissing-prototypes] 1211 | void ub_enum_remove(void) | ^ drivers/ub/ubus/enum.c:1211:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 1211 | void ub_enum_remove(void) | ^ | static 4 warnings generated. vim +/ub_enum_clear_ent_list +1097 drivers/ub/ubus/enum.c 995 > 996 int ub_cfg_read_guid(struct ub_entity *uent) 997 { 998 u32 val = 0; 999 int i, ret; 1000 1001 for (i = 0; i < UB_GUID_DW_NUM; i++) { 1002 ret = ub_cfg_read_dword(uent, UB_GUID + i * sizeof(u32), &val); 1003 if (ret) 1004 return ret; 1005 1006 uent->guid.dw[i] = val; 1007 } 1008 1009 return 0; 1010 } 1011 1012 static void ub_enum_destroy_entity(struct ub_entity *uent) 1013 { 1014 message_remove_device(uent); 1015 1016 if (is_primary(uent)) 1017 ub_ubc_put(uent->ubc); 1018 1019 kfree(uent); 1020 } 1021 1022 static struct ub_entity *ub_enum_create_uent(struct ub_bus_controller *ubc) 1023 { 1024 struct ub_entity *uent; 1025 int ret; 1026 1027 uent = ub_alloc_ent(); 1028 if (!uent) 1029 return NULL; 1030 1031 uent->pue = uent; 1032 uent->entity_idx = 0; 1033 uent->ubc = ub_ubc_get(ubc); 1034 uent->upi = UB_CP_UPI; 1035 ret = message_probe_device(uent); 1036 if (ret) { 1037 dev_err(&ubc->dev, "enum msg probe dev failed, ret=%d\n", ret); 1038 goto err_out; 1039 } 1040 1041 return uent; 1042 err_out: 1043 ub_ubc_put(ubc); 1044 kfree(uent); 1045 return NULL; 1046 } 1047 1048 static struct ub_entity *ub_enum_create_bus_controller(struct ub_bus_controller *ubc) 1049 { 1050 struct ub_entity *uent; 1051 int ret; 1052 1053 uent = ub_enum_create_uent(ubc); 1054 if (!uent) 1055 return (struct ub_entity *)ERR_PTR(-ENOMEM); 1056 1057 ubc->uent = uent; 1058 ret = ub_cfg_read_guid(uent); 1059 if (ret) { 1060 dev_err(&ubc->dev, "read guid failed, ret=%d\n", ret); 1061 goto err_out; 1062 } 1063 1064 uent->topo_rank = 0; 1065 uent->dev.parent = &ubc->dev; 1066 1067 return uent; 1068 err_out: 1069 ub_enum_destroy_entity(uent); 1070 return (struct ub_entity *)ERR_PTR(ret); 1071 } 1072 1073 static void ub_enum_topo_ent_uninit(struct ub_entity *uent) 1074 { 1075 if (is_primary(uent)) { 1076 ub_route_clear(uent); /* alloc in route_config */ 1077 ub_cna_free(uent); /* alloc in na_cfg */ 1078 ub_ports_unset(uent); /* alloc in scan_device */ 1079 } 1080 } 1081 1082 static int ub_enum_and_configure_ent(struct ub_entity *uent, void *buf) 1083 { 1084 int ret; 1085 1086 ret = ub_enum_ent(uent, buf); 1087 if (ret) 1088 return ret; 1089 1090 ret = ub_enum_na_cfg(uent, buf); 1091 if (ret) 1092 ub_ports_unset(uent); 1093 1094 return ret; 1095 } 1096 > 1097 void ub_enum_clear_ent_list(struct list_head *dev_list) 1098 { 1099 struct ub_entity *uent, *tmp; 1100 struct ub_port *port; 1101 1102 list_for_each_entry_safe_reverse(uent, tmp, dev_list, node) { 1103 list_del(&uent->node); 1104 for_each_uent_port(port, uent) 1105 ub_port_disconnect(port); 1106 1107 ub_cna_free(uent); 1108 ub_ports_unset(uent); 1109 ub_enum_destroy_entity(uent); 1110 } 1111 } 1112 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-6.6 3098/3098] drivers/iommu/hisilicon/ummu-core/core_tid.c:62:1: warning: no previous prototype for function 'default_manager_alloc'
by kernel test robot 04 Nov '25

04 Nov '25

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: a29fc03bd0ddaf7388cf31604ef5bd9807585109 commit: 83106717e2b7400d5e65a361975b20a706ceddd4 [3098/3098] iommu/ummu-core: add UMMU Token ID operation interfaces config: arm64-allmodconfig (https://download.01.org/0day-ci/archive/20251104/202511040838.PRCEsIJz-lkp@…) compiler: clang version 19.1.7 (https://github.com/llvm/llvm-project cd708029e0b2869e80abe31ddb175f7c35361f90) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251104/202511040838.PRCEsIJz-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202511040838.PRCEsIJz-lkp@intel.com/ All warnings (new ones prefixed by >>): >> drivers/iommu/hisilicon/ummu-core/core_tid.c:62:1: warning: no previous prototype for function 'default_manager_alloc' [-Wmissing-prototypes] 62 | default_manager_alloc(struct ummu_core_device *core_device, u32 max_tid, | ^ drivers/iommu/hisilicon/ummu-core/core_tid.c:61:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 61 | struct ummu_tid_manager * | ^ | static >> drivers/iommu/hisilicon/ummu-core/core_tid.c:75:6: warning: no previous prototype for function 'default_manager_free' [-Wmissing-prototypes] 75 | void default_manager_free(struct ummu_tid_manager *manager) | ^ drivers/iommu/hisilicon/ummu-core/core_tid.c:75:1: note: declare 'static' if the function is not intended to be used outside of this translation unit 75 | void default_manager_free(struct ummu_tid_manager *manager) | ^ | static 2 warnings generated. vim +/default_manager_alloc +62 drivers/iommu/hisilicon/ummu-core/core_tid.c 60 61 struct ummu_tid_manager * > 62 default_manager_alloc(struct ummu_core_device *core_device, u32 max_tid, 63 u32 min_tid) 64 { 65 struct ummu_tid_manager *manager; 66 67 manager = kzalloc(sizeof(struct ummu_tid_manager), GFP_KERNEL); 68 if (!manager) 69 return NULL; 70 71 xa_init_flags(&manager->token_ids, XA_FLAGS_ALLOC); 72 return manager; 73 } 74 > 75 void default_manager_free(struct ummu_tid_manager *manager) 76 { 77 struct ummu_tid_data *data; 78 unsigned long index; 79 80 if (!manager) { 81 pr_err("invalid param.\n"); 82 return; 83 } 84 /* all tid have to be released by the user */ 85 if (WARN_ON(!xa_empty(&manager->token_ids))) { 86 pr_warn("tid resource was not released.\n"); 87 xa_for_each(&manager->token_ids, index, data) 88 manager->ops->free_tid(manager, index); 89 } 90 xa_destroy(&manager->token_ids); 91 kfree(manager); 92 } 93 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-6.6 3098/3098] net/oenetcls/oenetcls_flow.c:140:48: error: 'struct net_device' has no member named 'rx_cpu_rmap'
by kernel test robot 04 Nov '25

04 Nov '25

Hi Wang, First bad commit (maybe != root cause): tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: a29fc03bd0ddaf7388cf31604ef5bd9807585109 commit: 22d4075bf5ef29ba4b329f954ac28a7de1d69a65 [3098/3098] net/oenetcls: remove oenetcls trace hook config: x86_64-randconfig-121-20251104 (https://download.01.org/0day-ci/archive/20251104/202511040811.Ogiq35Kv-lkp@…) compiler: gcc-14 (Debian 14.2.0-19) 14.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251104/202511040811.Ogiq35Kv-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202511040811.Ogiq35Kv-lkp@intel.com/ All errors (new ones prefixed by >>): net/oenetcls/oenetcls_flow.c:18:6: warning: no previous prototype for 'is_oecls_config_netdev' [-Wmissing-prototypes] 18 | bool is_oecls_config_netdev(const char *name) | ^~~~~~~~~~~~~~~~~~~~~~ net/oenetcls/oenetcls_flow.c: In function 'set_oecls_cpu': >> net/oenetcls/oenetcls_flow.c:140:48: error: 'struct net_device' has no member named 'rx_cpu_rmap' 140 | if (!skb_rx_queue_recorded(skb) || !dev->rx_cpu_rmap || | ^~ >> net/oenetcls/oenetcls_flow.c:161:29: error: 'const struct net_device_ops' has no member named 'ndo_rx_flow_steer' 161 | rc = dev->netdev_ops->ndo_rx_flow_steer(dev, skb, rxq_index, flow_id); | ^~ Kconfig warnings: (for reference only) WARNING: unmet direct dependencies detected for PTP_1588_CLOCK Depends on [n]: NET [=y] && POSIX_TIMERS [=n] Selected by [m]: - SXE [=m] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] - SXE_VF [=m] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] vim +140 net/oenetcls/oenetcls_flow.c 4bed6ba0e88f50 Wang Liang 2025-08-26 17 4bed6ba0e88f50 Wang Liang 2025-08-26 @18 bool is_oecls_config_netdev(const char *name) 4bed6ba0e88f50 Wang Liang 2025-08-26 19 { 4bed6ba0e88f50 Wang Liang 2025-08-26 20 struct oecls_netdev_info *netdev_info; 4bed6ba0e88f50 Wang Liang 2025-08-26 21 int netdev_loop; 4bed6ba0e88f50 Wang Liang 2025-08-26 22 4bed6ba0e88f50 Wang Liang 2025-08-26 23 for_each_oecls_netdev(netdev_loop, netdev_info) 4bed6ba0e88f50 Wang Liang 2025-08-26 24 if (strcmp(netdev_info->dev_name, name) == 0) 4bed6ba0e88f50 Wang Liang 2025-08-26 25 return true; 4bed6ba0e88f50 Wang Liang 2025-08-26 26 4bed6ba0e88f50 Wang Liang 2025-08-26 27 return false; 4bed6ba0e88f50 Wang Liang 2025-08-26 28 } 4bed6ba0e88f50 Wang Liang 2025-08-26 29 22d4075bf5ef29 Wang Liang 2025-09-12 30 static bool _oecls_timeout(struct net_device *dev, u16 rxq_index, 22d4075bf5ef29 Wang Liang 2025-09-12 31 u32 flow_id, u16 filter_id) 4bed6ba0e88f50 Wang Liang 2025-08-26 32 { 4bed6ba0e88f50 Wang Liang 2025-08-26 33 struct netdev_rx_queue *rxqueue = dev->_rx + rxq_index; 4bed6ba0e88f50 Wang Liang 2025-08-26 34 struct oecls_dev_flow_table *flow_table; 4bed6ba0e88f50 Wang Liang 2025-08-26 35 struct oecls_dev_flow *rflow; 4bed6ba0e88f50 Wang Liang 2025-08-26 36 bool expire = true; 4bed6ba0e88f50 Wang Liang 2025-08-26 37 unsigned int cpu; 4bed6ba0e88f50 Wang Liang 2025-08-26 38 4bed6ba0e88f50 Wang Liang 2025-08-26 39 rcu_read_lock(); 4bed6ba0e88f50 Wang Liang 2025-08-26 40 flow_table = rcu_dereference(rxqueue->oecls_ftb); 4bed6ba0e88f50 Wang Liang 2025-08-26 41 if (flow_table && flow_id <= flow_table->mask) { 4bed6ba0e88f50 Wang Liang 2025-08-26 42 rflow = &flow_table->flows[flow_id]; 4bed6ba0e88f50 Wang Liang 2025-08-26 43 cpu = READ_ONCE(rflow->cpu); 4bed6ba0e88f50 Wang Liang 2025-08-26 44 oecls_debug("dev:%s, rxq:%d, flow_id:%u, filter_id:%d/%d, cpu:%d\n", dev->name, 4bed6ba0e88f50 Wang Liang 2025-08-26 45 rxq_index, flow_id, filter_id, rflow->filter, cpu); 4bed6ba0e88f50 Wang Liang 2025-08-26 46 4bed6ba0e88f50 Wang Liang 2025-08-26 47 if (rflow->filter == filter_id && cpu < nr_cpu_ids) { 4bed6ba0e88f50 Wang Liang 2025-08-26 48 if (time_before(jiffies, rflow->timeout + OECLS_TIMEOUT)) { 4bed6ba0e88f50 Wang Liang 2025-08-26 49 expire = false; 4bed6ba0e88f50 Wang Liang 2025-08-26 50 } else { 4bed6ba0e88f50 Wang Liang 2025-08-26 51 rflow->isvalid = 0; 4bed6ba0e88f50 Wang Liang 2025-08-26 52 WRITE_ONCE(rflow->cpu, OECLS_NO_CPU); 4bed6ba0e88f50 Wang Liang 2025-08-26 53 } 4bed6ba0e88f50 Wang Liang 2025-08-26 54 } 4bed6ba0e88f50 Wang Liang 2025-08-26 55 } 4bed6ba0e88f50 Wang Liang 2025-08-26 56 rcu_read_unlock(); 4bed6ba0e88f50 Wang Liang 2025-08-26 57 oecls_debug("%s, dev:%s, rxq:%d, flow_id:%u, filter_id:%d, expire:%d\n", __func__, 4bed6ba0e88f50 Wang Liang 2025-08-26 58 dev->name, rxq_index, flow_id, filter_id, expire); 22d4075bf5ef29 Wang Liang 2025-09-12 59 return expire; 4bed6ba0e88f50 Wang Liang 2025-08-26 60 } 4bed6ba0e88f50 Wang Liang 2025-08-26 61 22d4075bf5ef29 Wang Liang 2025-09-12 62 static void _oecls_flow_update(struct sock *sk) 4bed6ba0e88f50 Wang Liang 2025-08-26 63 { 4bed6ba0e88f50 Wang Liang 2025-08-26 64 struct oecls_sock_flow_table *tb; 4bed6ba0e88f50 Wang Liang 2025-08-26 65 unsigned int hash, index; 4bed6ba0e88f50 Wang Liang 2025-08-26 66 u32 val; 4bed6ba0e88f50 Wang Liang 2025-08-26 67 u32 cpu = raw_smp_processor_id(); 4bed6ba0e88f50 Wang Liang 2025-08-26 68 4bed6ba0e88f50 Wang Liang 2025-08-26 69 if (sk->sk_state != TCP_ESTABLISHED) 4bed6ba0e88f50 Wang Liang 2025-08-26 70 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 71 4bed6ba0e88f50 Wang Liang 2025-08-26 72 if (check_appname(current->comm)) 4bed6ba0e88f50 Wang Liang 2025-08-26 73 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 74 4bed6ba0e88f50 Wang Liang 2025-08-26 75 rcu_read_lock(); 4bed6ba0e88f50 Wang Liang 2025-08-26 76 tb = rcu_dereference(oecls_sock_flow_table); 4bed6ba0e88f50 Wang Liang 2025-08-26 77 hash = READ_ONCE(sk->sk_rxhash); 4bed6ba0e88f50 Wang Liang 2025-08-26 78 if (tb && hash) { 4bed6ba0e88f50 Wang Liang 2025-08-26 79 index = hash & tb->mask; 4bed6ba0e88f50 Wang Liang 2025-08-26 80 val = hash & ~oecls_cpu_mask; 4bed6ba0e88f50 Wang Liang 2025-08-26 81 val |= cpu; 4bed6ba0e88f50 Wang Liang 2025-08-26 82 4bed6ba0e88f50 Wang Liang 2025-08-26 83 if (READ_ONCE(tb->ents[index]) != val) { 4bed6ba0e88f50 Wang Liang 2025-08-26 84 WRITE_ONCE(tb->ents[index], val); 4bed6ba0e88f50 Wang Liang 2025-08-26 85 4bed6ba0e88f50 Wang Liang 2025-08-26 86 oecls_debug("[%s] sk:%p, hash:0x%x, index:0x%x, val:0x%x, cpu:%d\n", 4bed6ba0e88f50 Wang Liang 2025-08-26 87 current->comm, sk, hash, index, val, cpu); 4bed6ba0e88f50 Wang Liang 2025-08-26 88 } 4bed6ba0e88f50 Wang Liang 2025-08-26 89 } 4bed6ba0e88f50 Wang Liang 2025-08-26 90 rcu_read_unlock(); 4bed6ba0e88f50 Wang Liang 2025-08-26 91 } 4bed6ba0e88f50 Wang Liang 2025-08-26 92 4bed6ba0e88f50 Wang Liang 2025-08-26 93 static int flow_get_queue_idx(struct net_device *dev, int nid, struct sk_buff *skb) 4bed6ba0e88f50 Wang Liang 2025-08-26 94 { 4bed6ba0e88f50 Wang Liang 2025-08-26 95 struct oecls_netdev_info *netdev_info; 4bed6ba0e88f50 Wang Liang 2025-08-26 96 int netdev_loop; 4bed6ba0e88f50 Wang Liang 2025-08-26 97 u32 hash, index; 4bed6ba0e88f50 Wang Liang 2025-08-26 98 struct oecls_numa_info *numa_info; 4bed6ba0e88f50 Wang Liang 2025-08-26 99 struct oecls_numa_bound_dev_info *bound_dev = NULL; 4bed6ba0e88f50 Wang Liang 2025-08-26 100 int rxq_id, rxq_num, i; 4bed6ba0e88f50 Wang Liang 2025-08-26 101 4bed6ba0e88f50 Wang Liang 2025-08-26 102 numa_info = get_oecls_numa_info(nid); 4bed6ba0e88f50 Wang Liang 2025-08-26 103 if (!numa_info) 4bed6ba0e88f50 Wang Liang 2025-08-26 104 return -1; 4bed6ba0e88f50 Wang Liang 2025-08-26 105 4bed6ba0e88f50 Wang Liang 2025-08-26 106 for_each_oecls_netdev(netdev_loop, netdev_info) { 4bed6ba0e88f50 Wang Liang 2025-08-26 107 if (strcmp(netdev_info->dev_name, dev->name) == 0) { 4bed6ba0e88f50 Wang Liang 2025-08-26 108 bound_dev = &numa_info->bound_dev[netdev_loop]; 4bed6ba0e88f50 Wang Liang 2025-08-26 109 break; 4bed6ba0e88f50 Wang Liang 2025-08-26 110 } 4bed6ba0e88f50 Wang Liang 2025-08-26 111 } 4bed6ba0e88f50 Wang Liang 2025-08-26 112 4bed6ba0e88f50 Wang Liang 2025-08-26 113 if (!bound_dev) 4bed6ba0e88f50 Wang Liang 2025-08-26 114 return -1; 4bed6ba0e88f50 Wang Liang 2025-08-26 115 rxq_num = bitmap_weight(bound_dev->bitmap_rxq, OECLS_MAX_RXQ_NUM_PER_DEV); 4bed6ba0e88f50 Wang Liang 2025-08-26 116 if (rxq_num == 0) 4bed6ba0e88f50 Wang Liang 2025-08-26 117 return -1; 4bed6ba0e88f50 Wang Liang 2025-08-26 118 4bed6ba0e88f50 Wang Liang 2025-08-26 119 hash = skb_get_hash(skb); 4bed6ba0e88f50 Wang Liang 2025-08-26 120 index = hash % rxq_num; 4bed6ba0e88f50 Wang Liang 2025-08-26 121 4bed6ba0e88f50 Wang Liang 2025-08-26 122 i = 0; 4bed6ba0e88f50 Wang Liang 2025-08-26 123 for_each_set_bit(rxq_id, bound_dev->bitmap_rxq, OECLS_MAX_RXQ_NUM_PER_DEV) 4bed6ba0e88f50 Wang Liang 2025-08-26 124 if (index == i++) 4bed6ba0e88f50 Wang Liang 2025-08-26 125 return rxq_id; 4bed6ba0e88f50 Wang Liang 2025-08-26 126 4bed6ba0e88f50 Wang Liang 2025-08-26 127 return -1; 4bed6ba0e88f50 Wang Liang 2025-08-26 128 } 4bed6ba0e88f50 Wang Liang 2025-08-26 129 4bed6ba0e88f50 Wang Liang 2025-08-26 130 static void set_oecls_cpu(struct net_device *dev, struct sk_buff *skb, 4bed6ba0e88f50 Wang Liang 2025-08-26 131 struct oecls_dev_flow *old_rflow, int old_rxq_id, u16 next_cpu) 4bed6ba0e88f50 Wang Liang 2025-08-26 132 { 4bed6ba0e88f50 Wang Liang 2025-08-26 133 struct netdev_rx_queue *rxqueue; 4bed6ba0e88f50 Wang Liang 2025-08-26 134 struct oecls_dev_flow_table *dtb; 4bed6ba0e88f50 Wang Liang 2025-08-26 135 struct oecls_dev_flow *rflow; 4bed6ba0e88f50 Wang Liang 2025-08-26 136 u32 flow_id, hash; 4bed6ba0e88f50 Wang Liang 2025-08-26 137 u16 rxq_index; 4bed6ba0e88f50 Wang Liang 2025-08-26 138 int rc; 4bed6ba0e88f50 Wang Liang 2025-08-26 139 4bed6ba0e88f50 Wang Liang 2025-08-26 @140 if (!skb_rx_queue_recorded(skb) || !dev->rx_cpu_rmap || 4bed6ba0e88f50 Wang Liang 2025-08-26 141 !(dev->features & NETIF_F_NTUPLE)) 4bed6ba0e88f50 Wang Liang 2025-08-26 142 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 143 4bed6ba0e88f50 Wang Liang 2025-08-26 144 rxq_index = flow_get_queue_idx(dev, cpu_to_node(next_cpu), skb); 4bed6ba0e88f50 Wang Liang 2025-08-26 145 if (rxq_index == skb_get_rx_queue(skb) || rxq_index < 0) 4bed6ba0e88f50 Wang Liang 2025-08-26 146 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 147 4bed6ba0e88f50 Wang Liang 2025-08-26 148 rxqueue = dev->_rx + rxq_index; 4bed6ba0e88f50 Wang Liang 2025-08-26 149 dtb = rcu_dereference(rxqueue->oecls_ftb); 4bed6ba0e88f50 Wang Liang 2025-08-26 150 if (!dtb) 4bed6ba0e88f50 Wang Liang 2025-08-26 151 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 152 4bed6ba0e88f50 Wang Liang 2025-08-26 153 hash = skb_get_hash(skb); 4bed6ba0e88f50 Wang Liang 2025-08-26 154 flow_id = hash & dtb->mask; 4bed6ba0e88f50 Wang Liang 2025-08-26 155 rflow = &dtb->flows[flow_id]; 4bed6ba0e88f50 Wang Liang 2025-08-26 156 if (rflow->isvalid && rflow->cpu == next_cpu) { 4bed6ba0e88f50 Wang Liang 2025-08-26 157 rflow->timeout = jiffies; 4bed6ba0e88f50 Wang Liang 2025-08-26 158 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 159 } 4bed6ba0e88f50 Wang Liang 2025-08-26 160 4bed6ba0e88f50 Wang Liang 2025-08-26 @161 rc = dev->netdev_ops->ndo_rx_flow_steer(dev, skb, rxq_index, flow_id); 4bed6ba0e88f50 Wang Liang 2025-08-26 162 oecls_debug("skb:%p, rxq:%d, hash:0x%x, flow_id:%u, old_rxq_id:%d, next_cpu:%d, rc:%d\n", 4bed6ba0e88f50 Wang Liang 2025-08-26 163 skb, rxq_index, hash, flow_id, old_rxq_id, next_cpu, rc); 4bed6ba0e88f50 Wang Liang 2025-08-26 164 if (rc < 0) 4bed6ba0e88f50 Wang Liang 2025-08-26 165 return; 4bed6ba0e88f50 Wang Liang 2025-08-26 166 4bed6ba0e88f50 Wang Liang 2025-08-26 167 rflow->filter = rc; 4bed6ba0e88f50 Wang Liang 2025-08-26 168 rflow->isvalid = 1; 4bed6ba0e88f50 Wang Liang 2025-08-26 169 rflow->timeout = jiffies; 4bed6ba0e88f50 Wang Liang 2025-08-26 170 if (old_rflow->filter == rflow->filter) 4bed6ba0e88f50 Wang Liang 2025-08-26 171 old_rflow->filter = OECLS_NO_FILTER; 4bed6ba0e88f50 Wang Liang 2025-08-26 172 rflow->cpu = next_cpu; 4bed6ba0e88f50 Wang Liang 2025-08-26 173 } 4bed6ba0e88f50 Wang Liang 2025-08-26 174 :::::: The code at line 140 was first introduced by commit :::::: 4bed6ba0e88f50484fd5fb06bd993727b981b718 net/oenetcls: introduce oenetcls for network optimization :::::: TO: Wang Liang <wangliang74(a)huawei.com> :::::: CC: Wang Liang <wangliang74(a)huawei.com> -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0