January 2024 - Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] pciehp: fix a race between pciehp and removing operations by sysfs
by Xiongfeng Wang 25 Jan '24

25 Jan '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8YPFU CVE: NA ------------------------------------------------- When I run a stress test about pcie hotplug and removing operations by sysfs, I got a hange task, and the following call trace is printed. INFO: task irq/746-pciehp:41551 blocked for more than 120 seconds. Tainted: P W OE 4.19.25- "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. irq/746-pciehp D 0 41551 2 0x00000228 Call trace: __switch_to+0x94/0xe8 __schedule+0x270/0x8b0 schedule+0x2c/0x88 schedule_preempt_disabled+0x14/0x20 __mutex_lock.isra.1+0x1fc/0x540 __mutex_lock_slowpath+0x24/0x30 mutex_lock+0x80/0xa8 pci_lock_rescan_remove+0x20/0x28 pciehp_configure_device+0x30/0x140 pciehp_handle_presence_or_link_change+0x35c/0x4b0 pciehp_ist+0x1cc/0x1d0 irq_thread_fn+0x30/0x80 irq_thread+0x128/0x200 kthread+0x134/0x138 ret_from_fork+0x10/0x18 INFO: task bash:6424 blocked for more than 120 seconds. Tainted: P W OE 4.19.25- "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. bash D 0 6424 2231 0x00000200 Call trace: __switch_to+0x94/0xe8 __schedule+0x270/0x8b0 schedule+0x2c/0x88 schedule_timeout+0x224/0x448 wait_for_common+0x198/0x2a0 wait_for_completion+0x28/0x38 kthread_stop+0x60/0x190 __free_irq+0x1c0/0x348 free_irq+0x40/0x88 pcie_shutdown_notification+0x54/0x80 pciehp_remove+0x30/0x50 pcie_port_remove_service+0x3c/0x58 device_release_driver_internal+0x1b4/0x250 device_release_driver+0x28/0x38 bus_remove_device+0xd4/0x160 device_del+0x128/0x348 device_unregister+0x24/0x78 remove_iter+0x48/0x58 device_for_each_child+0x6c/0xb8 pcie_port_device_remove+0x2c/0x48 pcie_portdrv_remove+0x5c/0x68 pci_device_remove+0x48/0xd8 device_release_driver_internal+0x1b4/0x250 device_release_driver+0x28/0x38 pci_stop_bus_device+0x84/0xb8 pci_stop_and_remove_bus_device_locked+0x24/0x40 remove_store+0xa4/0xb8 dev_attr_store+0x44/0x60 sysfs_kf_write+0x58/0x80 kernfs_fop_write+0xe8/0x1f0 __vfs_write+0x60/0x190 vfs_write+0xac/0x1c0 ksys_write+0x6c/0xd8 __arm64_sys_write+0x24/0x30 el0_svc_common+0xa0/0x180 el0_svc_handler+0x38/0x78 el0_svc+0x8/0xc When we remove a slot by sysfs. 'pci_stop_and_remove_bus_device_locked()' will be called. This function will get the global mutex lock 'pci_rescan_remove_lock', and remove the slot. If the irq thread 'pciehp_ist' is still running, we will wait until it exits. If a pciehp interrupt happens immediately after we remove the slot by sysfs, but before we free the pciehp irq in 'pci_stop_and_remove_bus_device_locked()'. 'pciehp_ist' will hung because the global mutex lock 'pci_rescan_remove_lock' is held by the sysfs operation. But the sysfs operation is waiting for the pciehp irq thread 'pciehp_ist' ends. Then a hung task occurs. So this two kinds of operation, removing through attention buttion and removing through /sys/devices/pci***/remove, should not be excuted at the same time. This patch add a global variable to mark that one of these operations is under processing. When this variable is set, if another operation is requested, it will be rejected. We use a global variable 'slot_being_removed_rescaned' to mark whether a slot is being removed or rescaned. This will cause a slot hotplug operation is delayed if another slot is being remove or rescaned. But if these two slots are under different root ports, they should not influence each other. This patch make the flag 'slot_being_removed_rescanned' per root port so that one slot hotplug operation doesn't influence slots below another root port. We record the root port in struct pci_dev when the pci device is initialized and added into the system instead of using 'pcie_find_root_port()' to find the root port when we need it. Because iterating the pci tree needs the protection of 'pci_lock_rescan_remove()'. This will make the problem more complexed because the lock is very coarse-grained. We don't need to worry about 'use-after-free' because child pci devices are always removed before the root port device is removed. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Conflicts: drivers/pci/hotplug/pciehp_ctrl.c drivers/pci/hotplug/pciehp_hpc.c include/linux/pci.h Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- drivers/pci/hotplug/pciehp.h | 5 ++ drivers/pci/hotplug/pciehp_ctrl.c | 40 ++++++++++++++++ drivers/pci/hotplug/pciehp_hpc.c | 79 ++++++++++++++++++++++++++----- drivers/pci/pci-sysfs.c | 22 +++++++++ drivers/pci/probe.c | 5 ++ include/linux/pci.h | 8 ++++ include/linux/workqueue.h | 2 + 7 files changed, 150 insertions(+), 11 deletions(-) diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index e0a614acee05..dad48d151ce2 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -199,6 +199,11 @@ static inline const char *slot_name(struct controller *ctrl) return hotplug_slot_name(&ctrl->hotplug_slot); } +static inline struct pci_dev *ctrl_dev(struct controller *ctrl) +{ + return ctrl->pcie->port; +} + static inline struct controller *to_ctrl(struct hotplug_slot *hotplug_slot) { return container_of(hotplug_slot, struct controller, hotplug_slot); diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index dcdbfcf404dd..458ab6f6379c 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -143,6 +143,8 @@ void pciehp_queue_pushbutton_work(struct work_struct *work) { struct controller *ctrl = container_of(work, struct controller, button_work.work); + int events = ctrl->button_work.data; + struct pci_dev *rpdev = ctrl_dev(ctrl)->rpdev; mutex_lock(&ctrl->state_lock); switch (ctrl->state) { @@ -153,6 +155,15 @@ void pciehp_queue_pushbutton_work(struct work_struct *work) pciehp_request(ctrl, PCI_EXP_SLTSTA_PDC); break; default: + if (events) { + atomic_or(events, &ctrl->pending_events); + if (!pciehp_poll_mode) + irq_wake_thread(ctrl->pcie->irq, ctrl); + } else { + if (rpdev) + clear_bit(0, + &rpdev->slot_being_removed_rescanned); + } break; } mutex_unlock(&ctrl->state_lock); @@ -160,6 +171,8 @@ void pciehp_queue_pushbutton_work(struct work_struct *work) void pciehp_handle_button_press(struct controller *ctrl) { + struct pci_dev *rpdev = ctrl_dev(ctrl)->rpdev; + mutex_lock(&ctrl->state_lock); switch (ctrl->state) { case OFF_STATE: @@ -176,6 +189,7 @@ void pciehp_handle_button_press(struct controller *ctrl) /* blink power indicator and turn off attention */ pciehp_set_indicators(ctrl, PCI_EXP_SLTCTL_PWR_IND_BLINK, PCI_EXP_SLTCTL_ATTN_IND_OFF); + ctrl->button_work.data = 0; schedule_delayed_work(&ctrl->button_work, 5 * HZ); break; case BLINKINGOFF_STATE: @@ -199,10 +213,14 @@ void pciehp_handle_button_press(struct controller *ctrl) ctrl_info(ctrl, "Slot(%s): Button press: canceling request to power on\n", slot_name(ctrl)); } + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); break; default: ctrl_err(ctrl, "Slot(%s): Button press: ignoring invalid state %#x\n", slot_name(ctrl), ctrl->state); + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); break; } mutex_unlock(&ctrl->state_lock); @@ -210,6 +228,8 @@ void pciehp_handle_button_press(struct controller *ctrl) void pciehp_handle_disable_request(struct controller *ctrl) { + struct pci_dev *rpdev = ctrl_dev(ctrl)->rpdev; + mutex_lock(&ctrl->state_lock); switch (ctrl->state) { case BLINKINGON_STATE: @@ -221,11 +241,14 @@ void pciehp_handle_disable_request(struct controller *ctrl) mutex_unlock(&ctrl->state_lock); ctrl->request_result = pciehp_disable_slot(ctrl, SAFE_REMOVAL); + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); } void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events) { int present, link_active; + struct pci_dev *rpdev = ctrl_dev(ctrl)->rpdev; /* * If the slot is on and presence or link has changed, turn it off. @@ -266,6 +289,8 @@ void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events) slot_name(ctrl)); } mutex_unlock(&ctrl->state_lock); + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); return; } @@ -288,6 +313,8 @@ void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events) mutex_unlock(&ctrl->state_lock); break; } + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); } static int __pciehp_enable_slot(struct controller *ctrl) @@ -408,6 +435,14 @@ int pciehp_sysfs_enable_slot(struct hotplug_slot *hotplug_slot) int pciehp_sysfs_disable_slot(struct hotplug_slot *hotplug_slot) { struct controller *ctrl = to_ctrl(hotplug_slot); + struct pci_dev *rpdev = ctrl_dev(ctrl)->rpdev; + + if (rpdev && test_and_set_bit(0, + &rpdev->slot_being_removed_rescanned)) { + ctrl_info(ctrl, "Slot(%s): Slot is being removed or rescanned, please try later!\n", + slot_name(ctrl)); + return -EINVAL; + } mutex_lock(&ctrl->state_lock); switch (ctrl->state) { @@ -418,6 +453,8 @@ int pciehp_sysfs_disable_slot(struct hotplug_slot *hotplug_slot) wait_event(ctrl->requester, !atomic_read(&ctrl->pending_events) && !ctrl->ist_running); + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); return ctrl->request_result; case POWEROFF_STATE: ctrl_info(ctrl, "Slot(%s): Already in powering off state\n", @@ -436,5 +473,8 @@ int pciehp_sysfs_disable_slot(struct hotplug_slot *hotplug_slot) } mutex_unlock(&ctrl->state_lock); + if (rpdev) + clear_bit(0, &rpdev->slot_being_removed_rescanned); + return -ENODEV; } diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index fd713abdfb9f..a0fdcd4dc358 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -45,11 +45,6 @@ static const struct dmi_system_id inband_presence_disabled_dmi_table[] = { {} }; -static inline struct pci_dev *ctrl_dev(struct controller *ctrl) -{ - return ctrl->pcie->port; -} - static irqreturn_t pciehp_isr(int irq, void *dev_id); static irqreturn_t pciehp_ist(int irq, void *dev_id); static int pciehp_poll(void *data); @@ -694,6 +689,7 @@ static irqreturn_t pciehp_ist(int irq, void *dev_id) { struct controller *ctrl = (struct controller *)dev_id; struct pci_dev *pdev = ctrl_dev(ctrl); + struct pci_dev *rpdev = pdev->rpdev; irqreturn_t ret; u32 events; @@ -716,8 +712,20 @@ static irqreturn_t pciehp_ist(int irq, void *dev_id) } /* Check Attention Button Pressed */ - if (events & PCI_EXP_SLTSTA_ABP) - pciehp_handle_button_press(ctrl); + if (events & PCI_EXP_SLTSTA_ABP) { + if (!rpdev || (rpdev && !test_and_set_bit(0, + &rpdev->slot_being_removed_rescanned))) + pciehp_handle_button_press(ctrl); + else { + if (ctrl->state == BLINKINGOFF_STATE || + ctrl->state == BLINKINGON_STATE) + pciehp_handle_button_press(ctrl); + else + ctrl_info(ctrl, "Slot(%s): Slot operation failed because a remove or" + " rescan operation is under processing, please try later!\n", + slot_name(ctrl)); + } + } /* Check Power Fault Detected */ if (events & PCI_EXP_SLTSTA_PFD) { @@ -741,10 +749,59 @@ static irqreturn_t pciehp_ist(int irq, void *dev_id) * or Data Link Layer State Changed events. */ down_read_nested(&ctrl->reset_lock, ctrl->depth); - if (events & DISABLE_SLOT) - pciehp_handle_disable_request(ctrl); - else if (events & (PCI_EXP_SLTSTA_PDC | PCI_EXP_SLTSTA_DLLSC)) - pciehp_handle_presence_or_link_change(ctrl, events); + if (events & DISABLE_SLOT) { + if (!rpdev || (rpdev && !test_and_set_bit(0, + &rpdev->slot_being_removed_rescanned))) + pciehp_handle_disable_request(ctrl); + else { + if (ctrl->state == BLINKINGOFF_STATE || + ctrl->state == BLINKINGON_STATE) + pciehp_handle_disable_request(ctrl); + else { + ctrl_info(ctrl, "Slot(%s): DISABLE_SLOT event in remove or rescan process!\n", + slot_name(ctrl)); + /* + * we use the work_struct private data to store + * the event type + */ + ctrl->button_work.data = DISABLE_SLOT; + /* + * If 'work.timer' is pending, schedule the work will + * cause BUG_ON(). + */ + if (!timer_pending(&ctrl->button_work.timer)) + schedule_delayed_work(&ctrl->button_work, 3 * HZ); + else + ctrl_info(ctrl, "Slot(%s): Didn't schedule delayed_work because timer is pending!\n", + slot_name(ctrl)); + } + } + } else if (events & (PCI_EXP_SLTSTA_PDC | PCI_EXP_SLTSTA_DLLSC)) { + if (!rpdev || (rpdev && !test_and_set_bit(0, + &rpdev->slot_being_removed_rescanned))) + pciehp_handle_presence_or_link_change(ctrl, events); + else { + if (ctrl->state == BLINKINGOFF_STATE || + ctrl->state == BLINKINGON_STATE) + pciehp_handle_presence_or_link_change(ctrl, + events); + else { + /* + * When we are removing or rescanning through + * sysfs, suprise link down/up happens. So we + * will handle this event 3 seconds later. + */ + ctrl_info(ctrl, "Slot(%s): Surprise link down/up in remove or rescan process!\n", + slot_name(ctrl)); + ctrl->button_work.data = events & (PCI_EXP_SLTSTA_PDC | PCI_EXP_SLTSTA_DLLSC); + if (!timer_pending(&ctrl->button_work.timer)) + schedule_delayed_work(&ctrl->button_work, 3 * HZ); + else + ctrl_info(ctrl, "Slot(%s): Didn't schedule delayed_work because timer is pending!\n", + slot_name(ctrl)); + } + } + } up_read(&ctrl->reset_lock); ret = IRQ_HANDLED; diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 3317b9354716..156f97c9c696 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -483,12 +483,34 @@ static ssize_t remove_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { unsigned long val; + struct pci_dev *rpdev = to_pci_dev(dev)->rpdev; if (kstrtoul(buf, 0, &val) < 0) return -EINVAL; + if (rpdev && test_and_set_bit(0, + &rpdev->slot_being_removed_rescanned)) { + pr_info("Slot is being removed or rescanned, please try later!\n"); + return -EINVAL; + } + + /* + * if 'dev' is root port itself, 'pci_stop_and_remove_bus_device()' may + * free the 'rpdev', but we need to clear + * 'rpdev->slot_being_removed_rescanned' in the end. So get 'rpdev' to + * avoid possible 'use-after-free'. + */ + if (rpdev) + pci_dev_get(rpdev); + if (val && device_remove_file_self(dev, attr)) pci_stop_and_remove_bus_device_locked(to_pci_dev(dev)); + + if (rpdev) { + clear_bit(0, &rpdev->slot_being_removed_rescanned); + pci_dev_put(rpdev); + } + return count; } static DEVICE_ATTR_IGNORE_LOCKDEP(remove, 0220, NULL, diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 43159965e09e..9dfb9d22b1f6 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -2578,6 +2578,11 @@ void pci_device_add(struct pci_dev *dev, struct pci_bus *bus) /* Set up MSI IRQ domain */ pci_set_msi_domain(dev); + if (pci_pcie_type(dev) == PCI_EXP_TYPE_ROOT_PORT) + dev->rpdev = dev; + else + dev->rpdev = pcie_find_root_port(dev); + /* Notifier could use PCI capabilities */ dev->match_driver = false; ret = device_add(&dev->dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index b56417276042..bf1de133c569 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -529,6 +529,13 @@ struct pci_dev { /* These methods index pci_reset_fn_methods[] */ u8 reset_methods[PCI_NUM_RESET_METHODS]; /* In priority order */ + + /* + * This flag is only set on root ports. When a slot below a root port + * is being removed or rescanned, this flag is set. + */ + unsigned long slot_being_removed_rescanned; + struct pci_dev *rpdev; /* root port pci_dev */ }; static inline struct pci_dev *pci_physfn(struct pci_dev *dev) @@ -1081,6 +1088,7 @@ extern struct bus_type pci_bus_type; /* Do NOT directly access these two variables, unless you are arch-specific PCI * code, or PCI core code. */ extern struct list_head pci_root_buses; /* List of all known PCI buses */ + /* Some device drivers need know if PCI is initiated */ int no_pci_devices(void); diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 24b1e5070f4d..571dd2ec4381 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -115,6 +115,8 @@ struct delayed_work { /* target workqueue and CPU ->timer uses to queue ->work */ struct workqueue_struct *wq; int cpu; + /* delayed_work private data, only used in pciehp now */ + unsigned long data; }; struct rcu_work { -- 2.20.1

2 1

[openeuler:OLK-6.6 1359/2707] mm/share_pool.c:1226:21: error: implicit declaration of function 'huge_ptep_get'
by kernel test robot 25 Jan '24

25 Jan '24

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: e076c150c75c29fb99c2e7d8ae567b7ed6cce8b3 commit: 549b1f40b56511536196f7522ffa4d7b3da42337 [1359/2707] mm/sharepool: Implement mg_sp_make_share_u2k() config: arm64-randconfig-002-20240125 (https://download.01.org/0day-ci/archive/20240125/202401250925.wptU5r3b-lkp@…) compiler: aarch64-linux-gcc (GCC) 13.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240125/202401250925.wptU5r3b-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/202401250925.wptU5r3b-lkp@intel.com/ All errors (new ones prefixed by >>): mm/share_pool.c: In function 'sp_hugetlb_entry': >> mm/share_pool.c:1226:21: error: implicit declaration of function 'huge_ptep_get' [-Werror=implicit-function-declaration] 1226 | pte_t pte = huge_ptep_get(ptep); | ^~~~~~~~~~~~~ >> mm/share_pool.c:1226:21: error: invalid initializer mm/share_pool.c: At top level: mm/share_pool.c:1005:13: warning: 'sp_area_put_locked' defined but not used [-Wunused-function] 1005 | static void sp_area_put_locked(struct sp_area *spa) | ^~~~~~~~~~~~~~~~~~ mm/share_pool.c:944:24: warning: 'sp_area_get' defined but not used [-Wunused-function] 944 | static struct sp_area *sp_area_get(struct sp_group *spg, | ^~~~~~~~~~~ mm/share_pool.c:791:24: warning: 'sp_area_alloc' defined but not used [-Wunused-function] 791 | static struct sp_area *sp_area_alloc(unsigned long size, unsigned long flags, | ^~~~~~~~~~~~~ mm/share_pool.c:587:21: warning: 'spa_file' defined but not used [-Wunused-function] 587 | static struct file *spa_file(struct sp_area *spa) | ^~~~~~~~ mm/share_pool.c:582:22: warning: 'spa_size' defined but not used [-Wunused-function] 582 | static unsigned long spa_size(struct sp_area *spa) | ^~~~~~~~ mm/share_pool.c:497:13: warning: 'update_mem_usage_k2u' defined but not used [-Wunused-function] 497 | static void update_mem_usage_k2u(unsigned long size, bool inc, | ^~~~~~~~~~~~~~~~~~~~ mm/share_pool.c:485:13: warning: 'update_mem_usage_alloc' defined but not used [-Wunused-function] 485 | static void update_mem_usage_alloc(unsigned long size, bool inc, | ^~~~~~~~~~~~~~~~~~~~~~ mm/share_pool.c:438:12: warning: 'sp_group_setup_mapping_normal' defined but not used [-Wunused-function] 438 | static int sp_group_setup_mapping_normal(struct mm_struct *mm, struct sp_group *spg) | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ mm/share_pool.c:380:13: warning: 'sp_mapping_detach' defined but not used [-Wunused-function] 380 | static void sp_mapping_detach(struct sp_group *spg, struct sp_mapping *spm) | ^~~~~~~~~~~~~~~~~ mm/share_pool.c:237:13: warning: 'meminfo_init' defined but not used [-Wunused-function] 237 | static void meminfo_init(struct sp_meminfo *meminfo) | ^~~~~~~~~~~~ In file included from mm/share_pool.c:26: mm/share_pool.c:76:19: warning: 'sp_group_idr' defined but not used [-Wunused-variable] 76 | static DEFINE_IDR(sp_group_idr); | ^~~~~~~~~~~~ include/linux/idr.h:56:44: note: in definition of macro 'DEFINE_IDR' 56 | #define DEFINE_IDR(name) struct idr name = IDR_INIT(name) | ^~~~ mm/share_pool.c:73:12: warning: 'system_group_count' defined but not used [-Wunused-variable] 73 | static int system_group_count; | ^~~~~~~~~~~~~~~~~~ cc1: some warnings being treated as errors vim +/huge_ptep_get +1226 mm/share_pool.c 1221 1222 static int sp_hugetlb_entry(pte_t *ptep, unsigned long hmask, 1223 unsigned long addr, unsigned long next, 1224 struct mm_walk *walk) 1225 { > 1226 pte_t pte = huge_ptep_get(ptep); 1227 struct page *page = pte_page(pte); 1228 struct sp_walk_data *sp_walk_data; 1229 1230 if (unlikely(!pte_present(pte))) { 1231 pr_debug("the page of addr %lx unexpectedly not in RAM\n", (unsigned long)addr); 1232 return -EFAULT; 1233 } 1234 1235 sp_walk_data = walk->private; 1236 get_page(page); 1237 sp_walk_data->pages[sp_walk_data->page_count++] = page; 1238 return 0; 1239 } 1240 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6] tracing: Reserve kabi fields
by Zheng Yejian 25 Jan '24

25 Jan '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8YPFE CVE: NA -------------------------------- Signed-off-by: Zheng Yejian <zhengyejian1(a)huawei.com> --- include/linux/ftrace.h | 3 +++ include/linux/trace_events.h | 3 +++ include/linux/tracepoint-defs.h | 2 ++ kernel/trace/ring_buffer.c | 7 +++++++ kernel/trace/trace.h | 6 ++++++ 5 files changed, 21 insertions(+) diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index e8921871ef9a..56892db7de5c 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -21,6 +21,7 @@ #include <linux/fs.h> #include <asm/ftrace.h> +#include <linux/kabi.h> /* * If the arch supports passing the variable contents of @@ -340,6 +341,8 @@ struct ftrace_ops { unsigned long direct_call; #endif #endif + KABI_RESERVE(1) + KABI_RESERVE(2) }; extern struct ftrace_ops __rcu *ftrace_ops_list; diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index cf9f0c61796e..172cf91a45de 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -9,6 +9,7 @@ #include <linux/hardirq.h> #include <linux/perf_event.h> #include <linux/tracepoint.h> +#include <linux/kabi.h> struct trace_array; struct array_buffer; @@ -677,6 +678,8 @@ struct trace_event_file { atomic_t ref; /* ref count for opened files */ atomic_t sm_ref; /* soft-mode reference counter */ atomic_t tm_ref; /* trigger-mode reference counter */ + KABI_RESERVE(1) + KABI_RESERVE(2) }; #define __TRACE_EVENT_FLAGS(name, value) \ diff --git a/include/linux/tracepoint-defs.h b/include/linux/tracepoint-defs.h index 4dc4955f0fbf..b2b29dc8f259 100644 --- a/include/linux/tracepoint-defs.h +++ b/include/linux/tracepoint-defs.h @@ -10,6 +10,7 @@ #include <linux/atomic.h> #include <linux/static_key.h> +#include <linux/kabi.h> struct static_call_key; @@ -39,6 +40,7 @@ struct tracepoint { int (*regfunc)(void); void (*unregfunc)(void); struct tracepoint_func __rcu *funcs; + KABI_RESERVE(1) }; #ifdef CONFIG_HAVE_ARCH_PREL32_RELOCATIONS diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index f1ef4329343b..e70aa61539dc 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -28,6 +28,7 @@ #include <linux/oom.h> #include <asm/local.h> +#include <linux/kabi.h> /* * The "absolute" timestamp in the buffer is only 59 bits. @@ -537,6 +538,9 @@ struct ring_buffer_per_cpu { struct completion update_done; struct rb_irq_work irq_work; + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) }; struct trace_buffer { @@ -571,6 +575,9 @@ struct ring_buffer_iter { u64 page_stamp; struct ring_buffer_event *event; int missed_events; + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) }; #ifdef RB_TIME_32 diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 52123df262b2..8eebd99cbb89 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -21,6 +21,7 @@ #include <linux/workqueue.h> #include <linux/ctype.h> #include <linux/once_lite.h> +#include <linux/kabi.h> #include "pid_list.h" @@ -410,6 +411,9 @@ struct trace_array { struct cond_snapshot *cond_snapshot; #endif struct trace_func_repeats __percpu *last_func_repeats; + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) }; enum { @@ -1326,6 +1330,8 @@ struct ftrace_event_field { int size; int is_signed; int len; + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct prog_entry; -- 2.25.1

1 0

[PATCH OLK-6.6] block: remove precise_iostat
by Li Nan 25 Jan '24

25 Jan '24

hulk inclusion category: bugfix bugzilla: 189480, https://gitee.com/openeuler/kernel/issues/I8Y90H -------------------------------- Previously, we believed it would be too costly to sum all the percpu variables 'in_flight' every jiffy if we want to count io accurately. So we added switch 'precise_iostat' and set it to false be default. However, the overhead of precise_iostat is 1. inc and dev in_flight when submitting io and done io 2. read in_flight up to HZ times per second It is much lower than our previous evaluation and can be ignored. Now, turn on the switch by default. And it will be removed in the feature. Signed-off-by: Li Nan <linan122(a)huawei.com> --- block/blk-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk-core.c b/block/blk-core.c index 2612fb2eea77..3ea271135a6c 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -71,7 +71,7 @@ static struct kmem_cache *blk_requestq_cachep; */ static struct workqueue_struct *kblockd_workqueue; -static bool precise_iostat; +static bool precise_iostat = true; static int __init precise_iostat_setup(char *str) { -- 2.39.2

1 0

[openeuler:openEuler-1.0-LTS 21263/21567] drivers/iio/adc/phytium-adc.c:662:12: warning: 'phytium_adc_resume' defined but not used
by kernel test robot 25 Jan '24

25 Jan '24

tree: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS head: 0ff07578f017c011d7806126c4aaafdbf39b32fa commit: 69245da182c15b365e3b770e7ac56db0f8118165 [21263/21567] adc: add phytium adc driver config: arm64-randconfig-004-20240125 (https://download.01.org/0day-ci/archive/20240125/202401250716.GKgR8ZVW-lkp@…) compiler: aarch64-linux-gcc (GCC) 13.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240125/202401250716.GKgR8ZVW-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/202401250716.GKgR8ZVW-lkp@intel.com/ All warnings (new ones prefixed by >>): >> drivers/iio/adc/phytium-adc.c:662:12: warning: 'phytium_adc_resume' defined but not used [-Wunused-function] 662 | static int phytium_adc_resume(struct device *dev) | ^~~~~~~~~~~~~~~~~~ >> drivers/iio/adc/phytium-adc.c:651:12: warning: 'phytium_adc_suspend' defined but not used [-Wunused-function] 651 | static int phytium_adc_suspend(struct device *dev) | ^~~~~~~~~~~~~~~~~~~ vim +/phytium_adc_resume +662 drivers/iio/adc/phytium-adc.c 649 650 #ifdef CONFIG_PM > 651 static int phytium_adc_suspend(struct device *dev) 652 { 653 struct iio_dev *indio_dev = dev_get_drvdata(dev); 654 struct phytium_adc *adc = iio_priv(indio_dev); 655 656 phytium_adc_power_setup(adc, false); 657 clk_disable_unprepare(adc->adc_clk); 658 659 return 0; 660 } 661 > 662 static int phytium_adc_resume(struct device *dev) 663 { 664 struct iio_dev *indio_dev = dev_get_drvdata(dev); 665 struct phytium_adc *adc = iio_priv(indio_dev); 666 667 clk_prepare_enable(adc->adc_clk); 668 phytium_adc_power_setup(adc, true); 669 670 return 0; 671 } 672 #endif 673 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-6.6 2459/2707] ERROR: modpost: module zhaoxin-cputemp uses symbol hwmon_device_register_for_thermal from namespace HWMON_THERMAL, but does not import it.
by kernel test robot 25 Jan '24

25 Jan '24

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: e076c150c75c29fb99c2e7d8ae567b7ed6cce8b3 commit: 0b07c042b9308c44ddd204a05021af6f31af6543 [2459/2707] Add support for Zhaoxin core temperature monitoring config: x86_64-openeuler_defconfig-OLK-6.6-func (https://download.01.org/0day-ci/archive/20240125/202401250639.Uf4ncB7E-lkp@…) compiler: gcc-12 (Debian 12.2.0-14) 12.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240125/202401250639.Uf4ncB7E-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/202401250639.Uf4ncB7E-lkp@intel.com/ All errors (new ones prefixed by >>, old ones prefixed by <<): WARNING: modpost: missing MODULE_DESCRIPTION() in net/802/mrp.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_htb.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_hfsc.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_red.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_gred.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_ingress.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_sfq.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_tbf.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_teql.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_prio.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_multiq.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_netem.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_drr.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_plug.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_mqprio.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_mqprio_lib.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_choke.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/sch_qfq.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/cls_u32.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/cls_route.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/cls_fw.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/cls_basic.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/em_cmp.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/em_nbyte.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/em_u32.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/em_meta.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sched/em_text.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netlink/netlink_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nfnetlink_osf.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nf_conntrack.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nf_conntrack_netlink.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nf_conntrack_broadcast.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nf_nat.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nf_tables.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nft_fib.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nft_chain_nat.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/nft_fwd_netdev.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_rr.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_wrr.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_lc.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_wlc.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_fo.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_ovf.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_lblc.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_lblcr.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_dh.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_sh.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_sed.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_nq.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_ftp.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/netfilter/ipvs/ip_vs_pe_sip.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/netfilter/nf_defrag_ipv4.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/netfilter/nf_reject_ipv4.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/netfilter/iptable_nat.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/netfilter/iptable_raw.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/ip_tunnel.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/ipip.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/ip_gre.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/udp_tunnel.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/ip_vti.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/ah4.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/esp4.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/xfrm4_tunnel.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/tunnel4.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/inet_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/tcp_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/udp_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv4/raw_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/unix/unix_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/netfilter/ip6table_raw.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/netfilter/ip6table_nat.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/netfilter/nf_defrag_ipv6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/netfilter/nf_reject_ipv6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/ah6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/esp6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/xfrm6_tunnel.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/tunnel6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/mip6.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/sit.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ipv6/ip6_udp_tunnel.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/packet/af_packet_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sunrpc/auth_gss/auth_rpcgss.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sunrpc/auth_gss/rpcsec_gss_krb5.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/8021q/8021q.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/key/af_key.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/bridge/netfilter/ebtables.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/bridge/netfilter/ebtable_broute.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/bridge/netfilter/ebtable_filter.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/bridge/netfilter/ebtable_nat.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/bridge/bridge.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/atm/atm.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/atm/lec.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/sctp/sctp_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/tipc/diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/smc/smc_diag.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/6lowpan/6lowpan.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ieee802154/6lowpan/ieee802154_6lowpan.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/ieee802154/ieee802154_socket.o WARNING: modpost: missing MODULE_DESCRIPTION() in net/vmw_vsock/vsock_diag.o >> ERROR: modpost: module zhaoxin-cputemp uses symbol hwmon_device_register_for_thermal from namespace HWMON_THERMAL, but does not import it. -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6] mm/ksm: generalize ksm_process_profit
by Nanyong Sun 24 Jan '24

24 Jan '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8YMGX CVE: NA -------------------------------- The ksm_process_profit() was only used by procfs before, so it is built under CONFIG_PROC_FS. Now memcg with ksm also call it, so delete the CONFIG_PROC_FS compilation dependency to make it more general. Fixes: 8bafb5eac5a6 ("memcg: support ksm merge any mode per cgroup") Signed-off-by: Nanyong Sun <sunnanyong(a)huawei.com> --- include/linux/ksm.h | 2 -- mm/ksm.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index c2dd786a30e1..49779f85e1da 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -87,9 +87,7 @@ void collect_procs_ksm(struct page *page, struct list_head *to_kill, int force_early); #endif -#ifdef CONFIG_PROC_FS long ksm_process_profit(struct mm_struct *); -#endif /* CONFIG_PROC_FS */ #else /* !CONFIG_KSM */ diff --git a/mm/ksm.c b/mm/ksm.c index 7401a6c87a4b..dba6a2eb1a3c 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -3106,13 +3106,11 @@ static void wait_while_offlining(void) } #endif /* CONFIG_MEMORY_HOTREMOVE */ -#ifdef CONFIG_PROC_FS long ksm_process_profit(struct mm_struct *mm) { return (long)(mm->ksm_merging_pages + mm->ksm_zero_pages) * PAGE_SIZE - mm->ksm_rmap_items * sizeof(struct ksm_rmap_item); } -#endif /* CONFIG_PROC_FS */ #ifdef CONFIG_SYSFS /* -- 2.25.1

2 1

[PATCH OLK-6.6 v2] kabi: reserve space for bpf related structures
by Pu Lehui 24 Jan '24

24 Jan '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8UOP9 -------------------------------- reserve space for bpf related structures. Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- include/linux/bpf.h | 123 ++++++++++++++++++++++++++++++ include/linux/bpf_local_storage.h | 11 +++ include/linux/bpf_verifier.h | 40 ++++++++++ include/uapi/linux/bpf.h | 22 ++++++ 4 files changed, 196 insertions(+) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 392f581af2ce..3051926802bf 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -29,6 +29,7 @@ #include <linux/rcupdate_trace.h> #include <linux/static_call.h> #include <linux/memcontrol.h> +#include <linux/kabi.h> struct bpf_verifier_env; struct bpf_verifier_log; @@ -168,6 +169,11 @@ struct bpf_map_ops { /* bpf_iter info used to open a seq_file */ const struct bpf_iter_seq_info *iter_seq_info; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; enum { @@ -288,6 +294,11 @@ struct bpf_map { bool bypass_spec_v1; bool frozen; /* write-once; write-protected by freeze_mutex */ s64 __percpu *elem_count; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; static inline const char *btf_field_type_name(enum btf_field_type type) @@ -664,6 +675,13 @@ enum bpf_type_flag { /* DYNPTR points to xdp_buff */ DYNPTR_TYPE_XDP = BIT(16 + BPF_BASE_TYPE_BITS), +#ifdef CONFIG_KABI_RESERVE + BPF_TYPE_FLAG_RESERVE_1 = BIT(17 + BPF_BASE_TYPE_BITS), + BPF_TYPE_FLAG_RESERVE_2 = BIT(18 + BPF_BASE_TYPE_BITS), + BPF_TYPE_FLAG_RESERVE_3 = BIT(19 + BPF_BASE_TYPE_BITS), + BPF_TYPE_FLAG_RESERVE_4 = BIT(20 + BPF_BASE_TYPE_BITS), +#endif + __BPF_TYPE_FLAG_MAX, __BPF_TYPE_LAST_FLAG = __BPF_TYPE_FLAG_MAX - 1, }; @@ -714,6 +732,12 @@ enum bpf_arg_type { ARG_PTR_TO_TIMER, /* pointer to bpf_timer */ ARG_PTR_TO_KPTR, /* pointer to referenced kptr */ ARG_PTR_TO_DYNPTR, /* pointer to bpf_dynptr. See bpf_type_flag for dynptr type */ + +#ifdef CONFIG_KABI_RESERVE + BPF_ARG_TYPE_RESERVE_1, + BPF_ARG_TYPE_RESERVE_2, +#endif + __BPF_ARG_TYPE_MAX, /* Extended arg_types. */ @@ -730,6 +754,11 @@ enum bpf_arg_type { /* Pointer to valid memory of size known at compile time. */ ARG_PTR_TO_FIXED_SIZE_MEM = MEM_FIXED_SIZE | ARG_PTR_TO_MEM, +#ifdef CONFIG_KABI_RESERVE + BPF_ARG_TYPE_RESERVE_3, + BPF_ARG_TYPE_RESERVE_4, +#endif + /* This must be the last entry. Its purpose is to ensure the enum is * wide enough to hold the higher bits reserved for bpf_type_flag. */ @@ -748,6 +777,12 @@ enum bpf_return_type { RET_PTR_TO_MEM, /* returns a pointer to memory */ RET_PTR_TO_MEM_OR_BTF_ID, /* returns a pointer to a valid memory or a btf_id */ RET_PTR_TO_BTF_ID, /* returns a pointer to a btf_id */ + +#ifdef CONFIG_KABI_RESERVE + BPF_RET_TYPE_RESERVE_1, + BPF_RET_TYPE_RESERVE_2, +#endif + __BPF_RET_TYPE_MAX, /* Extended ret_types. */ @@ -760,6 +795,11 @@ enum bpf_return_type { RET_PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_BTF_ID, RET_PTR_TO_BTF_ID_TRUSTED = PTR_TRUSTED | RET_PTR_TO_BTF_ID, +#ifdef CONFIG_KABI_RESERVE + BPF_RET_TYPE_RESERVE_3, + BPF_RET_TYPE_RESERVE_4, +#endif + /* This must be the last entry. Its purpose is to ensure the enum is * wide enough to hold the higher bits reserved for bpf_type_flag. */ @@ -807,6 +847,11 @@ struct bpf_func_proto { }; int *ret_btf_id; /* return value btf_id */ bool (*allowed)(const struct bpf_prog *prog); + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* bpf_context is intentionally undefined structure. Pointer to bpf_context is @@ -866,6 +911,12 @@ enum bpf_reg_type { PTR_TO_BUF, /* reg points to a read/write buffer */ PTR_TO_FUNC, /* reg points to a bpf program function */ CONST_PTR_TO_DYNPTR, /* reg points to a const struct bpf_dynptr */ + +#ifdef CONFIG_KABI_RESERVE + BPF_REG_TYPE_RESERVE_1, + BPF_REG_TYPE_RESERVE_2, +#endif + __BPF_REG_TYPE_MAX, /* Extended reg_types. */ @@ -875,6 +926,11 @@ enum bpf_reg_type { PTR_TO_TCP_SOCK_OR_NULL = PTR_MAYBE_NULL | PTR_TO_TCP_SOCK, PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | PTR_TO_BTF_ID, +#ifdef CONFIG_KABI_RESERVE + BPF_REG_TYPE_RESERVE_3, + BPF_REG_TYPE_RESERVE_4, +#endif + /* This must be the last entry. Its purpose is to ensure the enum is * wide enough to hold the higher bits reserved for bpf_type_flag. */ @@ -942,6 +998,11 @@ struct bpf_verifier_ops { int (*btf_struct_access)(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, int off, int size); + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_prog_offload_ops { @@ -1001,6 +1062,9 @@ struct btf_func_model { u8 nr_args; u8 arg_size[MAX_BPF_FUNC_ARGS]; u8 arg_flags[MAX_BPF_FUNC_ARGS]; + + KABI_RESERVE(1) + KABI_RESERVE(2) }; /* Restore arguments before returning from trampoline to let original function @@ -1121,6 +1185,11 @@ struct bpf_tramp_image { struct rcu_head rcu; struct work_struct work; }; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_trampoline { @@ -1149,6 +1218,11 @@ struct bpf_trampoline { /* Executable image of trampoline */ struct bpf_tramp_image *cur_image; struct module *mod; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_attach_target_info { @@ -1180,6 +1254,11 @@ struct bpf_dispatcher { struct static_call_key *sc_key; void *sc_tramp; #endif + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; static __always_inline __nocfi unsigned int bpf_dispatcher_nop_func( @@ -1465,6 +1544,15 @@ struct bpf_prog_aux { struct work_struct work; struct rcu_head rcu; }; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) + KABI_RESERVE(5) + KABI_RESERVE(6) + KABI_RESERVE(7) + KABI_RESERVE(8) }; struct bpf_prog { @@ -1499,6 +1587,11 @@ struct bpf_prog { DECLARE_FLEX_ARRAY(struct sock_filter, insns); DECLARE_FLEX_ARRAY(struct bpf_insn, insnsi); }; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_array_aux { @@ -1507,6 +1600,11 @@ struct bpf_array_aux { struct bpf_map *map; struct mutex poke_mutex; struct work_struct work; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_link { @@ -1516,6 +1614,11 @@ struct bpf_link { const struct bpf_link_ops *ops; struct bpf_prog *prog; struct work_struct work; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_link_ops { @@ -1529,6 +1632,11 @@ struct bpf_link_ops { struct bpf_link_info *info); int (*update_map)(struct bpf_link *link, struct bpf_map *new_map, struct bpf_map *old_map); + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_tramp_link { @@ -1626,6 +1734,11 @@ struct bpf_struct_ops { struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS]; u32 type_id; u32 value_id; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; #if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL) @@ -1727,6 +1840,11 @@ struct bpf_array { DECLARE_FLEX_ARRAY(void *, ptrs) __aligned(8); DECLARE_FLEX_ARRAY(void __percpu *, pptrs) __aligned(8); }; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; #define BPF_COMPLEXITY_LIMIT_INSNS 1000000 /* yes. 1M insns */ @@ -1822,6 +1940,11 @@ struct bpf_prog_array_item { struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE]; u64 bpf_cookie; }; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_prog_array { diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 173ec7f43ed1..c2ae8788a5cb 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -14,6 +14,7 @@ #include <linux/hash.h> #include <linux/types.h> #include <linux/bpf_mem_alloc.h> +#include <linux/kabi.h> #include <uapi/linux/btf.h> #define BPF_LOCAL_STORAGE_CACHE_SIZE 16 @@ -59,6 +60,11 @@ struct bpf_local_storage_map { struct bpf_mem_alloc selem_ma; struct bpf_mem_alloc storage_ma; bool bpf_ma; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_local_storage_data { @@ -94,6 +100,11 @@ struct bpf_local_storage { */ struct rcu_head rcu; raw_spinlock_t lock; /* Protect adding/removing from the "list" */ + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* U16_MAX is much more than enough for sk local storage diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index b6e58dab8e27..d2b59c6d9d60 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -8,6 +8,7 @@ #include <linux/btf.h> /* for struct btf and btf_id() */ #include <linux/filter.h> /* for MAX_BPF_STACK */ #include <linux/tnum.h> +#include <linux/kabi.h> /* Maximum variable offset umax_value permitted when resolving memory accesses. * In practice this is far bigger than any realistic pointer offset; this limit @@ -225,6 +226,11 @@ struct bpf_reg_state { enum bpf_reg_liveness live; /* if (!precise && SCALAR_VALUE) min/max/tnum don't affect safety */ bool precise; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; enum bpf_stack_slot_type { @@ -273,6 +279,11 @@ struct bpf_reference_state { * exiting a callback function. */ int callback_ref; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* state of the program: @@ -306,6 +317,11 @@ struct bpf_func_state { struct bpf_reference_state *refs; int allocated_stack; struct bpf_stack_state *stack; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_idx_pair { @@ -383,6 +399,11 @@ struct bpf_verifier_state { */ struct bpf_idx_pair *jmp_history; u32 jmp_history_cnt; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; #define bpf_get_spilled_reg(slot, frame) \ @@ -490,6 +511,11 @@ struct bpf_insn_aux_data { * this instruction, regardless of any heuristics */ bool force_checkpoint; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; #define MAX_USED_MAPS 64 /* max number of maps accessed by one eBPF program */ @@ -541,6 +567,11 @@ struct bpf_subprog_info { bool tail_call_reachable; bool has_ld_abs; bool is_async_cb; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; struct bpf_verifier_env; @@ -640,6 +671,15 @@ struct bpf_verifier_env { * e.g., in reg_type_str() to generate reg_type string */ char tmp_str_buf[TMP_STR_BUF_LEN]; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) + KABI_RESERVE(5) + KABI_RESERVE(6) + KABI_RESERVE(7) + KABI_RESERVE(8) }; __printf(2, 0) void bpf_verifier_vlog(struct bpf_verifier_log *log, diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 9dd0b85549b6..56c3fc8e4123 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -944,6 +944,11 @@ enum bpf_map_type { BPF_MAP_TYPE_BLOOM_FILTER, BPF_MAP_TYPE_USER_RINGBUF, BPF_MAP_TYPE_CGRP_STORAGE, + + BPF_MAP_TYPE_RESERVE_1, + BPF_MAP_TYPE_RESERVE_2, + BPF_MAP_TYPE_RESERVE_3, + BPF_MAP_TYPE_RESERVE_4, }; /* Note that tracing related programs such as @@ -989,6 +994,11 @@ enum bpf_prog_type { BPF_PROG_TYPE_SYSCALL, /* a program that can execute syscalls */ BPF_PROG_TYPE_NETFILTER, BPF_PROG_TYPE_SCHED, + + BPF_PROG_TYPE_RESERVE_1, + BPF_PROG_TYPE_RESERVE_2, + BPF_PROG_TYPE_RESERVE_3, + BPF_PROG_TYPE_RESERVE_4, }; enum bpf_attach_type { @@ -1042,6 +1052,12 @@ enum bpf_attach_type { BPF_TCX_EGRESS, BPF_TRACE_UPROBE_MULTI, BPF_SCHED, + + BPF_ATTACH_TYPE_RESERVE_1, + BPF_ATTACH_TYPE_RESERVE_2, + BPF_ATTACH_TYPE_RESERVE_3, + BPF_ATTACH_TYPE_RESERVE_4, + __MAX_BPF_ATTACH_TYPE }; @@ -1061,6 +1077,12 @@ enum bpf_link_type { BPF_LINK_TYPE_NETFILTER = 10, BPF_LINK_TYPE_TCX = 11, BPF_LINK_TYPE_UPROBE_MULTI = 12, + + BPF_LINK_TYPE_RESERVE_1 = 13, + BPF_LINK_TYPE_RESERVE_2 = 14, + BPF_LINK_TYPE_RESERVE_3 = 15, + BPF_LINK_TYPE_RESERVE_4 = 16, + MAX_BPF_LINK_TYPE, }; -- 2.34.1

1 0

[PATCH OLK-6.6 v1 0/2] Reserve syscall entries for kabi compatibility
by Wei Li 24 Jan '24

24 Jan '24

In this patch set, we: 1. Fix map_shadow_stack() syscall reserving issue 2. Reserve 15 syscall entries for kabi compatibility Sohil Mehta (1): arch: Reserve map_shadow_stack() syscall number for all architectures Wei Li (1): kabi: Reserve syscall entries for kabi compatibility arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 16 ++++++++ arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 32 ++++++++++++++++ arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/mips/kernel/syscalls/syscall_o32.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 16 ++++++++ arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 16 ++++++++ arch/x86/entry/syscalls/syscall_64.tbl | 15 ++++++++ arch/xtensa/kernel/syscalls/syscall.tbl | 1 + include/uapi/asm-generic/unistd.h | 36 +++++++++++++++++- tools/include/uapi/asm-generic/unistd.h | 37 +++++++++++++++++-- .../arch/powerpc/entry/syscalls/syscall.tbl | 16 ++++++++ .../arch/x86/entry/syscalls/syscall_64.tbl | 15 ++++++++ 21 files changed, 207 insertions(+), 5 deletions(-) -- 2.25.1

2 3

[PATCH OLK-5.10] sched/fair: Fix qos_timer deadlock when cpuhp offline
by Zhao Wenhui 24 Jan '24

24 Jan '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8WTPH -------------------------------- When cpu hotplug offline, if qos_overload_timer_handler() concurrently running, It could trigger an ABBA deadlock. As qos_overload_timer_handler() requires rq lock, while the cpu hotplug attached firstly and waiting for the qos_timer handling, This can cause Hard LOCKUP like: [359230.788754] Call trace: [359230.788755] hrtimer_active+0x7c/0xec [359230.788757] hrtimer_cancel+0x3c/0x60 [359230.788758] unthrottle_qos_cfs_rqs+0xbc/0x110 [359230.788760] unthrottle_offline_cfs_rqs+0x40/0x150 [359230.788762] rq_offline_fair+0x60/0x70 [359230.788764] set_rq_offline.part.0+0x54/0xf4 [359230.788765] set_rq_offline+0x34/0x44 [359230.788767] rq_attach_root+0x1e8/0x260 [359230.788768] cpu_attach_domain+0x244/0x430 [359230.788770] detach_destroy_domains+0xbc/0x140 [359230.788772] partition_sched_domains_locked+0x23c/0x314 [359230.788774] rebuild_sched_domains_locked+0x1f0/0x270 [359230.788776] cpuset_hotplug_workfn+0x514/0x74c [359230.788777] process_one_work+0x34c/0x800 [359230.788779] worker_thread+0xa8/0x500 [359230.788780] kthread+0x1e0/0x220 [359230.788782] ret_from_fork+0x10/0x18 [359230.788783] Kernel panic - not syncing: Hard LOCKUP Fix it by switch to use __unthrottle_qos_cfs_rqs(), instead of unthrottle_qos_cfs_rqs() in unthrottle_offline_cfs_rqs, so that it will not trigger cancel_qos_timer() when cpu hotplug offline. Fixes: c62a5f1384b9 ("sched/qos: Add qos_tg_{throttle,unthrottle}_{up,down}") Signed-off-by: Zhao Wenhui <zhaowenhui8(a)huawei.com> --- kernel/sched/fair.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 16f6720a244c..f39e7547523c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -131,6 +131,7 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(struct hrtimer, qos_overload_timer); static DEFINE_PER_CPU(int, qos_cpu_overload); unsigned int sysctl_overload_detect_period = 5000; /* in ms */ unsigned int sysctl_offline_wait_interval = 100; /* in ms */ +static int __unthrottle_qos_cfs_rqs(int cpu); static int unthrottle_qos_cfs_rqs(int cpu); static bool qos_smt_expelled(int this_cpu); #endif @@ -5733,7 +5734,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_rq_held(rq); #ifdef CONFIG_QOS_SCHED - unthrottle_qos_cfs_rqs(cpu_of(rq)); + __unthrottle_qos_cfs_rqs(cpu_of(rq)); #endif rcu_read_lock(); -- 2.34.1

2 1

2024

2023

2022

2021

2020

2019

Kernel January 2024