From bfb1a7c91fb7758273b4a8d735313d9cc388b502 Mon Sep 17 00:00:00 2001 From: Nick Desaulniers Date: Wed, 2 Feb 2022 12:55:53 -0800 Subject: x86/bug: Merge annotate_reachable() into _BUG_FLAGS() asm In __WARN_FLAGS(), we had two asm statements (abbreviated): asm volatile("ud2"); asm volatile(".pushsection .discard.reachable"); These pair of statements are used to trigger an exception, but then help objtool understand that for warnings, control flow will be restored immediately afterwards. The problem is that volatile is not a compiler barrier. GCC explicitly documents this: > Note that the compiler can move even volatile asm instructions > relative to other code, including across jump instructions. Also, no clobbers are specified to prevent instructions from subsequent statements from being scheduled by compiler before the second asm statement. This can lead to instructions from subsequent statements being emitted by the compiler before the second asm statement. Providing a scheduling model such as via -march= options enables the compiler to better schedule instructions with known latencies to hide latencies from data hazards compared to inline asm statements in which latencies are not estimated. If an instruction gets scheduled by the compiler between the two asm statements, then objtool will think that it is not reachable, producing a warning. To prevent instructions from being scheduled in between the two asm statements, merge them. Also remove an unnecessary unreachable() asm annotation from BUG() in favor of __builtin_unreachable(). objtool is able to track that the ud2 from BUG() terminates control flow within the function. Link: https://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html#Volatile Link: https://github.com/ClangBuiltLinux/linux/issues/1483 Signed-off-by: Nick Desaulniers Signed-off-by: Josh Poimboeuf Link: https://lore.kernel.org/r/20220202205557.2260694-1-ndesaulniers@google.com --- include/linux/compiler.h | 21 +++++---------------- 1 file changed, 5 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 429dcebe2b99..0f7fd205ab7e 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -117,14 +117,6 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, */ #define __stringify_label(n) #n -#define __annotate_reachable(c) ({ \ - asm volatile(__stringify_label(c) ":\n\t" \ - ".pushsection .discard.reachable\n\t" \ - ".long " __stringify_label(c) "b - .\n\t" \ - ".popsection\n\t" : : "i" (c)); \ -}) -#define annotate_reachable() __annotate_reachable(__COUNTER__) - #define __annotate_unreachable(c) ({ \ asm volatile(__stringify_label(c) ":\n\t" \ ".pushsection .discard.unreachable\n\t" \ @@ -133,24 +125,21 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, }) #define annotate_unreachable() __annotate_unreachable(__COUNTER__) -#define ASM_UNREACHABLE \ - "999:\n\t" \ - ".pushsection .discard.unreachable\n\t" \ - ".long 999b - .\n\t" \ +#define ASM_REACHABLE \ + "998:\n\t" \ + ".pushsection .discard.reachable\n\t" \ + ".long 998b - .\n\t" \ ".popsection\n\t" /* Annotate a C jump table to allow objtool to follow the code flow */ #define __annotate_jump_table __section(".rodata..c_jump_table") #else -#define annotate_reachable() #define annotate_unreachable() +# define ASM_REACHABLE #define __annotate_jump_table #endif -#ifndef ASM_UNREACHABLE -# define ASM_UNREACHABLE -#endif #ifndef unreachable # define unreachable() do { \ annotate_unreachable(); \ -- cgit v1.2.3-71-gd317 From 6bf625a4140f24b490766043b307f8252519578b Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Sun, 6 Feb 2022 11:36:56 -0800 Subject: Drivers: hv: vmbus: Rework use of DMA_BIT_MASK(64) Using DMA_BIT_MASK(64) as an initializer for a global variable causes problems with Clang 12.0.1. The compiler doesn't understand that value 64 is excluded from the shift at compile time, resulting in a build error. While this is a compiler problem, avoid the issue by setting up the dma_mask memory as part of struct hv_device, and initialize it using dma_set_mask(). Reported-by: Nathan Chancellor Reported-by: Vitaly Chikunov Reported-by: Jakub Kicinski Fixes: 743b237c3a7b ("scsi: storvsc: Add Isolation VM support for storvsc driver") Signed-off-by: Michael Kelley Reviewed-by: Nathan Chancellor Tested-by: Nathan Chancellor Link: https://lore.kernel.org/r/1644176216-12531-1-git-send-email-mikelley@microsoft.com Signed-off-by: Wei Liu --- drivers/hv/vmbus_drv.c | 4 ++-- include/linux/hyperv.h | 1 + 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index 34a4fd21bdf5..12a2b37e87f3 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -2082,7 +2082,6 @@ struct hv_device *vmbus_device_create(const guid_t *type, return child_device_obj; } -static u64 vmbus_dma_mask = DMA_BIT_MASK(64); /* * vmbus_device_register - Register the child device */ @@ -2123,8 +2122,9 @@ int vmbus_device_register(struct hv_device *child_device_obj) } hv_debug_add_dev_dir(child_device_obj); - child_device_obj->device.dma_mask = &vmbus_dma_mask; child_device_obj->device.dma_parms = &child_device_obj->dma_parms; + child_device_obj->device.dma_mask = &child_device_obj->dma_mask; + dma_set_mask(&child_device_obj->device, DMA_BIT_MASK(64)); return 0; err_kset_unregister: diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index f565a8938836..fe2e0179ed51 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1262,6 +1262,7 @@ struct hv_device { struct vmbus_channel *channel; struct kset *channels_kset; struct device_dma_parameters dma_parms; + u64 dma_mask; /* place holder to keep track of the dir for hv device in debugfs */ struct dentry *debug_dir; -- cgit v1.2.3-71-gd317 From cb1f65c1e1424a4b5e4a86da8aa3b8fd8459c8ec Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 4 Feb 2022 18:35:22 +0100 Subject: PM: s2idle: ACPI: Fix wakeup interrupts handling After commit e3728b50cd9b ("ACPI: PM: s2idle: Avoid possible race related to the EC GPE") wakeup interrupts occurring immediately after the one discarded by acpi_s2idle_wake() may be missed. Moreover, if the SCI triggers again immediately after the rearming in acpi_s2idle_wake(), that wakeup may be missed too. The problem is that pm_system_irq_wakeup() only calls pm_system_wakeup() when pm_wakeup_irq is 0, but that's not the case any more after the interrupt causing acpi_s2idle_wake() to run until pm_wakeup_irq is cleared by the pm_wakeup_clear() call in s2idle_loop(). However, there may be wakeup interrupts occurring in that time frame and if that happens, they will be missed. To address that issue first move the clearing of pm_wakeup_irq to the point at which it is known that the interrupt causing acpi_s2idle_wake() to tun will be discarded, before rearming the SCI for wakeup. Moreover, because that only reduces the size of the time window in which the issue may manifest itself, allow pm_system_irq_wakeup() to register two second wakeup interrupts in a row and, when discarding the first one, replace it with the second one. [Of course, this assumes that only one wakeup interrupt can be discarded in one go, but currently that is the case and I am not aware of any plans to change that.] Fixes: e3728b50cd9b ("ACPI: PM: s2idle: Avoid possible race related to the EC GPE") Cc: 5.4+ # 5.4+ Signed-off-by: Rafael J. Wysocki --- drivers/acpi/sleep.c | 1 + drivers/base/power/wakeup.c | 41 ++++++++++++++++++++++++++++++++++------- include/linux/suspend.h | 4 ++-- kernel/power/main.c | 5 ++++- kernel/power/process.c | 2 +- kernel/power/suspend.c | 2 -- 6 files changed, 42 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index fac7c9d4c9a1..d4fbea91ab6b 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -758,6 +758,7 @@ bool acpi_s2idle_wake(void) return true; } + pm_wakeup_clear(acpi_sci_irq); rearm_wake_irq(acpi_sci_irq); } diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index 99bda0da23a8..8666590201c9 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -34,7 +34,8 @@ suspend_state_t pm_suspend_target_state; bool events_check_enabled __read_mostly; /* First wakeup IRQ seen by the kernel in the last cycle. */ -unsigned int pm_wakeup_irq __read_mostly; +static unsigned int wakeup_irq[2] __read_mostly; +static DEFINE_RAW_SPINLOCK(wakeup_irq_lock); /* If greater than 0 and the system is suspending, terminate the suspend. */ static atomic_t pm_abort_suspend __read_mostly; @@ -942,19 +943,45 @@ void pm_system_cancel_wakeup(void) atomic_dec_if_positive(&pm_abort_suspend); } -void pm_wakeup_clear(bool reset) +void pm_wakeup_clear(unsigned int irq_number) { - pm_wakeup_irq = 0; - if (reset) + raw_spin_lock_irq(&wakeup_irq_lock); + + if (irq_number && wakeup_irq[0] == irq_number) + wakeup_irq[0] = wakeup_irq[1]; + else + wakeup_irq[0] = 0; + + wakeup_irq[1] = 0; + + raw_spin_unlock_irq(&wakeup_irq_lock); + + if (!irq_number) atomic_set(&pm_abort_suspend, 0); } void pm_system_irq_wakeup(unsigned int irq_number) { - if (pm_wakeup_irq == 0) { - pm_wakeup_irq = irq_number; + unsigned long flags; + + raw_spin_lock_irqsave(&wakeup_irq_lock, flags); + + if (wakeup_irq[0] == 0) + wakeup_irq[0] = irq_number; + else if (wakeup_irq[1] == 0) + wakeup_irq[1] = irq_number; + else + irq_number = 0; + + raw_spin_unlock_irqrestore(&wakeup_irq_lock, flags); + + if (irq_number) pm_system_wakeup(); - } +} + +unsigned int pm_wakeup_irq(void) +{ + return wakeup_irq[0]; } /** diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 3e8ecdebe601..300273ff40cc 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -497,14 +497,14 @@ extern void ksys_sync_helper(void); /* drivers/base/power/wakeup.c */ extern bool events_check_enabled; -extern unsigned int pm_wakeup_irq; extern suspend_state_t pm_suspend_target_state; extern bool pm_wakeup_pending(void); extern void pm_system_wakeup(void); extern void pm_system_cancel_wakeup(void); -extern void pm_wakeup_clear(bool reset); +extern void pm_wakeup_clear(unsigned int irq_number); extern void pm_system_irq_wakeup(unsigned int irq_number); +extern unsigned int pm_wakeup_irq(void); extern bool pm_get_wakeup_count(unsigned int *count, bool block); extern bool pm_save_wakeup_count(unsigned int count); extern void pm_wakep_autosleep_enabled(bool set); diff --git a/kernel/power/main.c b/kernel/power/main.c index 44169f3081fd..7e646079fbeb 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -504,7 +504,10 @@ static ssize_t pm_wakeup_irq_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return pm_wakeup_irq ? sprintf(buf, "%u\n", pm_wakeup_irq) : -ENODATA; + if (!pm_wakeup_irq()) + return -ENODATA; + + return sprintf(buf, "%u\n", pm_wakeup_irq()); } power_attr_ro(pm_wakeup_irq); diff --git a/kernel/power/process.c b/kernel/power/process.c index b7e7798637b8..11b570fcf049 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -134,7 +134,7 @@ int freeze_processes(void) if (!pm_freezing) atomic_inc(&system_freezing_cnt); - pm_wakeup_clear(true); + pm_wakeup_clear(0); pr_info("Freezing user space processes ... "); pm_freezing = true; error = try_to_freeze_tasks(true); diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 80cc1f0f502b..6fcdee7e87a5 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -136,8 +136,6 @@ static void s2idle_loop(void) break; } - pm_wakeup_clear(false); - s2idle_enter(); } -- cgit v1.2.3-71-gd317 From 0764db9b49c932b89ee4d9e3236dff4bb07b4a66 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 11 Feb 2022 16:32:32 -0800 Subject: mm: memcg: synchronize objcg lists with a dedicated spinlock Alexander reported a circular lock dependency revealed by the mmap1 ltp test: LOCKDEP_CIRCULAR (suite: ltp, case: mtest06 (mmap1)) WARNING: possible circular locking dependency detected 5.17.0-20220113.rc0.git0.f2211f194038.300.fc35.s390x+debug #1 Not tainted ------------------------------------------------------ mmap1/202299 is trying to acquire lock: 00000001892c0188 (css_set_lock){..-.}-{2:2}, at: obj_cgroup_release+0x4a/0xe0 but task is already holding lock: 00000000ca3b3818 (&sighand->siglock){-.-.}-{2:2}, at: force_sig_info_to_task+0x38/0x180 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&sighand->siglock){-.-.}-{2:2}: __lock_acquire+0x604/0xbd8 lock_acquire.part.0+0xe2/0x238 lock_acquire+0xb0/0x200 _raw_spin_lock_irqsave+0x6a/0xd8 __lock_task_sighand+0x90/0x190 cgroup_freeze_task+0x2e/0x90 cgroup_migrate_execute+0x11c/0x608 cgroup_update_dfl_csses+0x246/0x270 cgroup_subtree_control_write+0x238/0x518 kernfs_fop_write_iter+0x13e/0x1e0 new_sync_write+0x100/0x190 vfs_write+0x22c/0x2d8 ksys_write+0x6c/0xf8 __do_syscall+0x1da/0x208 system_call+0x82/0xb0 -> #0 (css_set_lock){..-.}-{2:2}: check_prev_add+0xe0/0xed8 validate_chain+0x736/0xb20 __lock_acquire+0x604/0xbd8 lock_acquire.part.0+0xe2/0x238 lock_acquire+0xb0/0x200 _raw_spin_lock_irqsave+0x6a/0xd8 obj_cgroup_release+0x4a/0xe0 percpu_ref_put_many.constprop.0+0x150/0x168 drain_obj_stock+0x94/0xe8 refill_obj_stock+0x94/0x278 obj_cgroup_charge+0x164/0x1d8 kmem_cache_alloc+0xac/0x528 __sigqueue_alloc+0x150/0x308 __send_signal+0x260/0x550 send_signal+0x7e/0x348 force_sig_info_to_task+0x104/0x180 force_sig_fault+0x48/0x58 __do_pgm_check+0x120/0x1f0 pgm_check_handler+0x11e/0x180 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&sighand->siglock); lock(css_set_lock); lock(&sighand->siglock); lock(css_set_lock); *** DEADLOCK *** 2 locks held by mmap1/202299: #0: 00000000ca3b3818 (&sighand->siglock){-.-.}-{2:2}, at: force_sig_info_to_task+0x38/0x180 #1: 00000001892ad560 (rcu_read_lock){....}-{1:2}, at: percpu_ref_put_many.constprop.0+0x0/0x168 stack backtrace: CPU: 15 PID: 202299 Comm: mmap1 Not tainted 5.17.0-20220113.rc0.git0.f2211f194038.300.fc35.s390x+debug #1 Hardware name: IBM 3906 M04 704 (LPAR) Call Trace: dump_stack_lvl+0x76/0x98 check_noncircular+0x136/0x158 check_prev_add+0xe0/0xed8 validate_chain+0x736/0xb20 __lock_acquire+0x604/0xbd8 lock_acquire.part.0+0xe2/0x238 lock_acquire+0xb0/0x200 _raw_spin_lock_irqsave+0x6a/0xd8 obj_cgroup_release+0x4a/0xe0 percpu_ref_put_many.constprop.0+0x150/0x168 drain_obj_stock+0x94/0xe8 refill_obj_stock+0x94/0x278 obj_cgroup_charge+0x164/0x1d8 kmem_cache_alloc+0xac/0x528 __sigqueue_alloc+0x150/0x308 __send_signal+0x260/0x550 send_signal+0x7e/0x348 force_sig_info_to_task+0x104/0x180 force_sig_fault+0x48/0x58 __do_pgm_check+0x120/0x1f0 pgm_check_handler+0x11e/0x180 INFO: lockdep is turned off. In this example a slab allocation from __send_signal() caused a refilling and draining of a percpu objcg stock, resulted in a releasing of another non-related objcg. Objcg release path requires taking the css_set_lock, which is used to synchronize objcg lists. This can create a circular dependency with the sighandler lock, which is taken with the locked css_set_lock by the freezer code (to freeze a task). In general it seems that using css_set_lock to synchronize objcg lists makes any slab allocations and deallocation with the locked css_set_lock and any intervened locks risky. To fix the problem and make the code more robust let's stop using css_set_lock to synchronize objcg lists and use a new dedicated spinlock instead. Link: https://lkml.kernel.org/r/Yfm1IHmoGdyUR81T@carbon.dhcp.thefacebook.com Fixes: bf4f059954dc ("mm: memcg/slab: obj_cgroup API") Signed-off-by: Roman Gushchin Reported-by: Alexander Egorenkov Tested-by: Alexander Egorenkov Reviewed-by: Waiman Long Acked-by: Tejun Heo Reviewed-by: Shakeel Butt Reviewed-by: Jeremy Linton Tested-by: Jeremy Linton Cc: Johannes Weiner Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 5 +++-- mm/memcontrol.c | 10 +++++----- 2 files changed, 8 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index b72d75141e12..0abbd685703b 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -219,7 +219,7 @@ struct obj_cgroup { struct mem_cgroup *memcg; atomic_t nr_charged_bytes; union { - struct list_head list; + struct list_head list; /* protected by objcg_lock */ struct rcu_head rcu; }; }; @@ -315,7 +315,8 @@ struct mem_cgroup { #ifdef CONFIG_MEMCG_KMEM int kmemcg_id; struct obj_cgroup __rcu *objcg; - struct list_head objcg_list; /* list of inherited objcgs */ + /* list of inherited objcgs, protected by objcg_lock */ + struct list_head objcg_list; #endif MEMCG_PADDING(_pad2_); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 09d342c7cbd0..36e9f38c919d 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -254,7 +254,7 @@ struct mem_cgroup *vmpressure_to_memcg(struct vmpressure *vmpr) } #ifdef CONFIG_MEMCG_KMEM -extern spinlock_t css_set_lock; +static DEFINE_SPINLOCK(objcg_lock); bool mem_cgroup_kmem_disabled(void) { @@ -298,9 +298,9 @@ static void obj_cgroup_release(struct percpu_ref *ref) if (nr_pages) obj_cgroup_uncharge_pages(objcg, nr_pages); - spin_lock_irqsave(&css_set_lock, flags); + spin_lock_irqsave(&objcg_lock, flags); list_del(&objcg->list); - spin_unlock_irqrestore(&css_set_lock, flags); + spin_unlock_irqrestore(&objcg_lock, flags); percpu_ref_exit(ref); kfree_rcu(objcg, rcu); @@ -332,7 +332,7 @@ static void memcg_reparent_objcgs(struct mem_cgroup *memcg, objcg = rcu_replace_pointer(memcg->objcg, NULL, true); - spin_lock_irq(&css_set_lock); + spin_lock_irq(&objcg_lock); /* 1) Ready to reparent active objcg. */ list_add(&objcg->list, &memcg->objcg_list); @@ -342,7 +342,7 @@ static void memcg_reparent_objcgs(struct mem_cgroup *memcg, /* 3) Move already reparented objcgs to the parent's list */ list_splice(&memcg->objcg_list, &parent->objcg_list); - spin_unlock_irq(&css_set_lock); + spin_unlock_irq(&objcg_lock); percpu_ref_kill(&objcg->refcnt); } -- cgit v1.2.3-71-gd317 From 8913c61001482378d4ed8cc577b17c1ba3e847e4 Mon Sep 17 00:00:00 2001 From: Peng Liu Date: Fri, 11 Feb 2022 16:32:35 -0800 Subject: kfence: make test case compatible with run time set sample interval The parameter kfence_sample_interval can be set via boot parameter and late shell command, which is convenient for automated tests and KFENCE parameter optimization. However, KFENCE test case just uses compile-time CONFIG_KFENCE_SAMPLE_INTERVAL, which will make KFENCE test case not run as users desired. Export kfence_sample_interval, so that KFENCE test case can use run-time-set sample interval. Link: https://lkml.kernel.org/r/20220207034432.185532-1-liupeng256@huawei.com Signed-off-by: Peng Liu Reviewed-by: Marco Elver Cc: Alexander Potapenko Cc: Dmitry Vyukov Cc: Jonathan Corbet Cc: Sumit Semwal Cc: Christian Knig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kfence.h | 2 ++ mm/kfence/core.c | 3 ++- mm/kfence/kfence_test.c | 8 ++++---- 3 files changed, 8 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kfence.h b/include/linux/kfence.h index 4b5e3679a72c..f49e64222628 100644 --- a/include/linux/kfence.h +++ b/include/linux/kfence.h @@ -17,6 +17,8 @@ #include #include +extern unsigned long kfence_sample_interval; + /* * We allocate an even number of pages, as it simplifies calculations to map * address to metadata indices; effectively, the very first page serves as an diff --git a/mm/kfence/core.c b/mm/kfence/core.c index 5ad40e3add45..13128fa13062 100644 --- a/mm/kfence/core.c +++ b/mm/kfence/core.c @@ -47,7 +47,8 @@ static bool kfence_enabled __read_mostly; -static unsigned long kfence_sample_interval __read_mostly = CONFIG_KFENCE_SAMPLE_INTERVAL; +unsigned long kfence_sample_interval __read_mostly = CONFIG_KFENCE_SAMPLE_INTERVAL; +EXPORT_SYMBOL_GPL(kfence_sample_interval); /* Export for test modules. */ #ifdef MODULE_PARAM_PREFIX #undef MODULE_PARAM_PREFIX diff --git a/mm/kfence/kfence_test.c b/mm/kfence/kfence_test.c index a22b1af85577..50dbb815a2a8 100644 --- a/mm/kfence/kfence_test.c +++ b/mm/kfence/kfence_test.c @@ -268,13 +268,13 @@ static void *test_alloc(struct kunit *test, size_t size, gfp_t gfp, enum allocat * 100x the sample interval should be more than enough to ensure we get * a KFENCE allocation eventually. */ - timeout = jiffies + msecs_to_jiffies(100 * CONFIG_KFENCE_SAMPLE_INTERVAL); + timeout = jiffies + msecs_to_jiffies(100 * kfence_sample_interval); /* * Especially for non-preemption kernels, ensure the allocation-gate * timer can catch up: after @resched_after, every failed allocation * attempt yields, to ensure the allocation-gate timer is scheduled. */ - resched_after = jiffies + msecs_to_jiffies(CONFIG_KFENCE_SAMPLE_INTERVAL); + resched_after = jiffies + msecs_to_jiffies(kfence_sample_interval); do { if (test_cache) alloc = kmem_cache_alloc(test_cache, gfp); @@ -608,7 +608,7 @@ static void test_gfpzero(struct kunit *test) int i; /* Skip if we think it'd take too long. */ - KFENCE_TEST_REQUIRES(test, CONFIG_KFENCE_SAMPLE_INTERVAL <= 100); + KFENCE_TEST_REQUIRES(test, kfence_sample_interval <= 100); setup_test_cache(test, size, 0, NULL); buf1 = test_alloc(test, size, GFP_KERNEL, ALLOCATE_ANY); @@ -739,7 +739,7 @@ static void test_memcache_alloc_bulk(struct kunit *test) * 100x the sample interval should be more than enough to ensure we get * a KFENCE allocation eventually. */ - timeout = jiffies + msecs_to_jiffies(100 * CONFIG_KFENCE_SAMPLE_INTERVAL); + timeout = jiffies + msecs_to_jiffies(100 * kfence_sample_interval); do { void *objects[100]; int i, num = kmem_cache_alloc_bulk(test_cache, GFP_ATOMIC, ARRAY_SIZE(objects), -- cgit v1.2.3-71-gd317 From 5891cd5ec46c2c2eb6427cb54d214b149635dd0e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 11 Feb 2022 12:06:23 -0800 Subject: net_sched: add __rcu annotation to netdev->qdisc syzbot found a data-race [1] which lead me to add __rcu annotations to netdev->qdisc, and proper accessors to get LOCKDEP support. [1] BUG: KCSAN: data-race in dev_activate / qdisc_lookup_rcu write to 0xffff888168ad6410 of 8 bytes by task 13559 on cpu 1: attach_default_qdiscs net/sched/sch_generic.c:1167 [inline] dev_activate+0x2ed/0x8f0 net/sched/sch_generic.c:1221 __dev_open+0x2e9/0x3a0 net/core/dev.c:1416 __dev_change_flags+0x167/0x3f0 net/core/dev.c:8139 rtnl_configure_link+0xc2/0x150 net/core/rtnetlink.c:3150 __rtnl_newlink net/core/rtnetlink.c:3489 [inline] rtnl_newlink+0xf4d/0x13e0 net/core/rtnetlink.c:3529 rtnetlink_rcv_msg+0x745/0x7e0 net/core/rtnetlink.c:5594 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff888168ad6410 of 8 bytes by task 13560 on cpu 0: qdisc_lookup_rcu+0x30/0x2e0 net/sched/sch_api.c:323 __tcf_qdisc_find+0x74/0x3a0 net/sched/cls_api.c:1050 tc_del_tfilter+0x1c7/0x1350 net/sched/cls_api.c:2211 rtnetlink_rcv_msg+0x5ba/0x7e0 net/core/rtnetlink.c:5585 netlink_rcv_skb+0x14e/0x250 net/netlink/af_netlink.c:2494 rtnetlink_rcv+0x18/0x20 net/core/rtnetlink.c:5612 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x602/0x6d0 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x728/0x850 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmsg+0x195/0x230 net/socket.c:2496 __do_sys_sendmsg net/socket.c:2505 [inline] __se_sys_sendmsg net/socket.c:2503 [inline] __x64_sys_sendmsg+0x42/0x50 net/socket.c:2503 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0xffffffff85dee080 -> 0xffff88815d96ec00 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 13560 Comm: syz-executor.2 Not tainted 5.17.0-rc3-syzkaller-00116-gf1baf68e1383-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 470502de5bdb ("net: sched: unlock rules update API") Signed-off-by: Eric Dumazet Cc: Vlad Buslov Reported-by: syzbot Cc: Jamal Hadi Salim Cc: Cong Wang Cc: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- net/core/rtnetlink.c | 6 ++++-- net/sched/cls_api.c | 6 +++--- net/sched/sch_api.c | 22 ++++++++++++---------- net/sched/sch_generic.c | 29 ++++++++++++++++------------- 5 files changed, 36 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e490b84732d1..8b5a314db167 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2158,7 +2158,7 @@ struct net_device { struct netdev_queue *_tx ____cacheline_aligned_in_smp; unsigned int num_tx_queues; unsigned int real_num_tx_queues; - struct Qdisc *qdisc; + struct Qdisc __rcu *qdisc; unsigned int tx_queue_len; spinlock_t tx_global_lock; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 710da8a36729..2fb8eb6791e8 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1699,6 +1699,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, { struct ifinfomsg *ifm; struct nlmsghdr *nlh; + struct Qdisc *qdisc; ASSERT_RTNL(); nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags); @@ -1716,6 +1717,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (tgt_netnsid >= 0 && nla_put_s32(skb, IFLA_TARGET_NETNSID, tgt_netnsid)) goto nla_put_failure; + qdisc = rtnl_dereference(dev->qdisc); if (nla_put_string(skb, IFLA_IFNAME, dev->name) || nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) || nla_put_u8(skb, IFLA_OPERSTATE, @@ -1735,8 +1737,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, #endif put_master_ifindex(skb, dev) || nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) || - (dev->qdisc && - nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) || + (qdisc && + nla_put_string(skb, IFLA_QDISC, qdisc->ops->id)) || nla_put_ifalias(skb, dev) || nla_put_u32(skb, IFLA_CARRIER_CHANGES, atomic_read(&dev->carrier_up_count) + diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 5f0f346b576f..5ce1208a6ea3 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1044,7 +1044,7 @@ static int __tcf_qdisc_find(struct net *net, struct Qdisc **q, /* Find qdisc */ if (!*parent) { - *q = dev->qdisc; + *q = rcu_dereference(dev->qdisc); *parent = (*q)->handle; } else { *q = qdisc_lookup_rcu(dev, TC_H_MAJ(*parent)); @@ -2587,7 +2587,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) parent = tcm->tcm_parent; if (!parent) - q = dev->qdisc; + q = rtnl_dereference(dev->qdisc); else q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent)); if (!q) @@ -2962,7 +2962,7 @@ static int tc_dump_chain(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; if (!tcm->tcm_parent) - q = dev->qdisc; + q = rtnl_dereference(dev->qdisc); else q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent)); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 179825a3b2fd..e3c0e8ea2dbb 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -301,7 +301,7 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) if (!handle) return NULL; - q = qdisc_match_from_root(dev->qdisc, handle); + q = qdisc_match_from_root(rtnl_dereference(dev->qdisc), handle); if (q) goto out; @@ -320,7 +320,7 @@ struct Qdisc *qdisc_lookup_rcu(struct net_device *dev, u32 handle) if (!handle) return NULL; - q = qdisc_match_from_root(dev->qdisc, handle); + q = qdisc_match_from_root(rcu_dereference(dev->qdisc), handle); if (q) goto out; @@ -1082,10 +1082,10 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, skip: if (!ingress) { notify_and_destroy(net, skb, n, classid, - dev->qdisc, new); + rtnl_dereference(dev->qdisc), new); if (new && !new->ops->attach) qdisc_refcount_inc(new); - dev->qdisc = new ? : &noop_qdisc; + rcu_assign_pointer(dev->qdisc, new ? : &noop_qdisc); if (new && new->ops->attach) new->ops->attach(new); @@ -1451,7 +1451,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { - q = dev->qdisc; + q = rtnl_dereference(dev->qdisc); } if (!q) { NL_SET_ERR_MSG(extack, "Cannot find specified qdisc on specified device"); @@ -1540,7 +1540,7 @@ replay: q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { - q = dev->qdisc; + q = rtnl_dereference(dev->qdisc); } /* It may be default qdisc, ignore it */ @@ -1762,7 +1762,8 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) s_q_idx = 0; q_idx = 0; - if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx, + if (tc_dump_qdisc_root(rtnl_dereference(dev->qdisc), + skb, cb, &q_idx, s_q_idx, true, tca[TCA_DUMP_INVISIBLE]) < 0) goto done; @@ -2033,7 +2034,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, } else if (qid1) { qid = qid1; } else if (qid == 0) - qid = dev->qdisc->handle; + qid = rtnl_dereference(dev->qdisc)->handle; /* Now qid is genuine qdisc handle consistent * both with parent and child. @@ -2044,7 +2045,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, portid = TC_H_MAKE(qid, portid); } else { if (qid == 0) - qid = dev->qdisc->handle; + qid = rtnl_dereference(dev->qdisc)->handle; } /* OK. Locate qdisc */ @@ -2205,7 +2206,8 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) s_t = cb->args[0]; t = 0; - if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t, true) < 0) + if (tc_dump_tclass_root(rtnl_dereference(dev->qdisc), + skb, tcm, cb, &t, s_t, true) < 0) goto done; dev_queue = dev_ingress_queue(dev); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index f893d9a81b01..5bab9f8b8f45 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -1164,30 +1164,33 @@ static void attach_default_qdiscs(struct net_device *dev) if (!netif_is_multiqueue(dev) || dev->priv_flags & IFF_NO_QUEUE) { netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL); - dev->qdisc = txq->qdisc_sleeping; - qdisc_refcount_inc(dev->qdisc); + qdisc = txq->qdisc_sleeping; + rcu_assign_pointer(dev->qdisc, qdisc); + qdisc_refcount_inc(qdisc); } else { qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT, NULL); if (qdisc) { - dev->qdisc = qdisc; + rcu_assign_pointer(dev->qdisc, qdisc); qdisc->ops->attach(qdisc); } } + qdisc = rtnl_dereference(dev->qdisc); /* Detect default qdisc setup/init failed and fallback to "noqueue" */ - if (dev->qdisc == &noop_qdisc) { + if (qdisc == &noop_qdisc) { netdev_warn(dev, "default qdisc (%s) fail, fallback to %s\n", default_qdisc_ops->id, noqueue_qdisc_ops.id); dev->priv_flags |= IFF_NO_QUEUE; netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL); - dev->qdisc = txq->qdisc_sleeping; - qdisc_refcount_inc(dev->qdisc); + qdisc = txq->qdisc_sleeping; + rcu_assign_pointer(dev->qdisc, qdisc); + qdisc_refcount_inc(qdisc); dev->priv_flags ^= IFF_NO_QUEUE; } #ifdef CONFIG_NET_SCHED - if (dev->qdisc != &noop_qdisc) - qdisc_hash_add(dev->qdisc, false); + if (qdisc != &noop_qdisc) + qdisc_hash_add(qdisc, false); #endif } @@ -1217,7 +1220,7 @@ void dev_activate(struct net_device *dev) * and noqueue_qdisc for virtual interfaces */ - if (dev->qdisc == &noop_qdisc) + if (rtnl_dereference(dev->qdisc) == &noop_qdisc) attach_default_qdiscs(dev); if (!netif_carrier_ok(dev)) @@ -1383,7 +1386,7 @@ static int qdisc_change_tx_queue_len(struct net_device *dev, void dev_qdisc_change_real_num_tx(struct net_device *dev, unsigned int new_real_tx) { - struct Qdisc *qdisc = dev->qdisc; + struct Qdisc *qdisc = rtnl_dereference(dev->qdisc); if (qdisc->ops->change_real_num_tx) qdisc->ops->change_real_num_tx(qdisc, new_real_tx); @@ -1447,7 +1450,7 @@ static void dev_init_scheduler_queue(struct net_device *dev, void dev_init_scheduler(struct net_device *dev) { - dev->qdisc = &noop_qdisc; + rcu_assign_pointer(dev->qdisc, &noop_qdisc); netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc); if (dev_ingress_queue(dev)) dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); @@ -1475,8 +1478,8 @@ void dev_shutdown(struct net_device *dev) netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc); if (dev_ingress_queue(dev)) shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); - qdisc_put(dev->qdisc); - dev->qdisc = &noop_qdisc; + qdisc_put(rtnl_dereference(dev->qdisc)); + rcu_assign_pointer(dev->qdisc, &noop_qdisc); WARN_ON(timer_pending(&dev->watchdog_timer)); } -- cgit v1.2.3-71-gd317