- Kernel - mailweb.openeuler.org

[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

[PATCH openEuler-1.0-LTS] 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/I8YHBT -------------------------------- 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: 453eaea60e92 ("sched/qos: Add qos_tg_{throttle,unthrottle}_{up,down}") Signed-off-by: Zhao Wenhui <zhaowenhui8(a)huawei.com> --- kernel/sched/fair.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index ce6bb4304949..6d0ec315f7be 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -116,7 +116,7 @@ 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); #endif #ifdef CONFIG_CFS_BANDWIDTH @@ -5218,7 +5218,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_held(&rq->lock); #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

[PATCH OLK-6.6] tlb: reserve fields for struct mmu_gather
by Jinjiang Tu 24 Jan '24

24 Jan '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8YJQK -------------------------------- reserve fields for struct mmu_gather. Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- include/asm-generic/tlb.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h index 129a3a759976..1a3191e844b7 100644 --- a/include/asm-generic/tlb.h +++ b/include/asm-generic/tlb.h @@ -14,6 +14,7 @@ #include <linux/mmu_notifier.h> #include <linux/swap.h> #include <linux/hugetlb_inline.h> +#include <linux/kabi.h> #include <asm/tlbflush.h> #include <asm/cacheflush.h> @@ -348,6 +349,8 @@ struct mmu_gather { unsigned int page_size; #endif #endif + KABI_RESERVE(1) + KABI_RESERVE(2) }; void tlb_flush_mmu(struct mmu_gather *tlb); -- 2.25.1

3 2

[PATCH OLK-6.6] arch/mm/fault: fix major fault accounting when retrying under per-VMA lock
by Tong Tiangen 24 Jan '24

24 Jan '24

From: Suren Baghdasaryan <surenb(a)google.com> mainline inclusion from mainline-v6.7 commit 46e714c729c8d1d8110bc0545d7ffe8a759c9dc0 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8YKL3 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- A test [1] in Android test suite started failing after [2] was merged. It turns out that after handling a major fault under per-VMA lock, the process major fault counter does not register that fault as major. Before [2] read faults would be done under mmap_lock, in which case FAULT_FLAG_TRIED flag is set before retrying. That in turn causes mm_account_fault() to account the fault as major once retry completes. With per-VMA locks we often retry because a fault can't be handled without locking the whole mm using mmap_lock. Therefore such retries do not set FAULT_FLAG_TRIED flag. This logic does not work after [2] because we can now handle read major faults under per-VMA lock and upon retry the fact there was a major fault gets lost. Fix this by setting FAULT_FLAG_TRIED after retrying under per-VMA lock if VM_FAULT_MAJOR was returned. Ideally we would use an additional VM_FAULT bit to indicate the reason for the retry (could not handle under per-VMA lock vs other reason) but this simpler solution seems to work, so keeping it simple. [1] https://cs.android.com/android/platform/superproject/+/master:test/vts-test… [2] https://lore.kernel.org/all/20231006195318.4087158-6-willy@infradead.org/ Link: https://lkml.kernel.org/r/20231226214610.109282-1-surenb@google.com Fixes: 12214eba1992 ("mm: handle read faults under the VMA lock") Signed-off-by: Suren Baghdasaryan <surenb(a)google.com> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: Alexander Gordeev <agordeev(a)linux.ibm.com> Cc: Andy Lutomirski <luto(a)kernel.org> Cc: Catalin Marinas <catalin.marinas(a)arm.com> Cc: Christophe Leroy <christophe.leroy(a)csgroup.eu> Cc: Dave Hansen <dave.hansen(a)linux.intel.com> Cc: Gerald Schaefer <gerald.schaefer(a)linux.ibm.com> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Palmer Dabbelt <palmer(a)dabbelt.com> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Will Deacon <will(a)kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> conflict: arch/s390/mm/fault.c Signed-off-by: Tong Tiangen <tongtiangen(a)huawei.com> --- arch/arm64/mm/fault.c | 2 ++ arch/powerpc/mm/fault.c | 2 ++ arch/riscv/mm/fault.c | 2 ++ arch/s390/mm/fault.c | 3 +++ arch/x86/mm/fault.c | 2 ++ 5 files changed, 11 insertions(+) diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index 171810268f6f..c376e58e8cf0 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -633,6 +633,8 @@ static int __kprobes do_page_fault(unsigned long far, unsigned long esr, goto done; } count_vm_vma_lock_event(VMA_LOCK_RETRY); + if (fault & VM_FAULT_MAJOR) + mm_flags |= FAULT_FLAG_TRIED; /* Quick path to respond to signals */ if (fault_signal_pending(fault, regs)) { diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c index b1723094d464..ec23164ad768 100644 --- a/arch/powerpc/mm/fault.c +++ b/arch/powerpc/mm/fault.c @@ -496,6 +496,8 @@ static int ___do_page_fault(struct pt_regs *regs, unsigned long address, goto done; } count_vm_vma_lock_event(VMA_LOCK_RETRY); + if (fault & VM_FAULT_MAJOR) + flags |= FAULT_FLAG_TRIED; if (fault_signal_pending(fault, regs)) return user_mode(regs) ? 0 : SIGBUS; diff --git a/arch/riscv/mm/fault.c b/arch/riscv/mm/fault.c index 90d4ba36d1d0..081339ddf47e 100644 --- a/arch/riscv/mm/fault.c +++ b/arch/riscv/mm/fault.c @@ -304,6 +304,8 @@ void handle_page_fault(struct pt_regs *regs) goto done; } count_vm_vma_lock_event(VMA_LOCK_RETRY); + if (fault & VM_FAULT_MAJOR) + flags |= FAULT_FLAG_TRIED; if (fault_signal_pending(fault, regs)) { if (!user_mode(regs)) diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index b678295931c3..f5463535013a 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -424,6 +424,9 @@ static inline vm_fault_t do_exception(struct pt_regs *regs, int access) goto out; } count_vm_vma_lock_event(VMA_LOCK_RETRY); + if (fault & VM_FAULT_MAJOR) + flags |= FAULT_FLAG_TRIED; + /* Quick path to respond to signals */ if (fault_signal_pending(fault, regs)) { fault = VM_FAULT_SIGNAL; diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index ab778eac1952..679b09cfe241 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -1370,6 +1370,8 @@ void do_user_addr_fault(struct pt_regs *regs, goto done; } count_vm_vma_lock_event(VMA_LOCK_RETRY); + if (fault & VM_FAULT_MAJOR) + flags |= FAULT_FLAG_TRIED; /* Quick path to respond to signals */ if (fault_signal_pending(fault, regs)) { -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] fs:/dcache.c: fix negative dentry limit not complete problem
by Long Li 24 Jan '24

24 Jan '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8WPQW CVE: NA -------------------------------- Restrictions of negative dentry can avoid softlock when too many dentry in memory, but restrictions before do not always take effect. For example, removing files which has been created would not enter retain_dentry(), because the dentry of file maybe has been added to lru list of superblock, it caused that fast_dput() reutrn true in last dput(). So, add restriction logic for the file which has been added to the lru list in fast_dput(), it prevents the negative dentry exceeding the limit when deleting existing file. Factor out negative dentry limit logic into limit_negative_dentry(), make the code more clean. Fixes: 7ba5d5d0e8b5 ("fs/dcache.c: avoid softlock since too many negative dentry") Signed-off-by: Long Li <leo.lilong(a)huawei.com> --- fs/dcache.c | 58 +++++++++++++++++++++++++++++++++-------------------- 1 file changed, 36 insertions(+), 22 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index ef1641398eda..e52e3abdf378 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -636,10 +636,36 @@ static inline struct dentry *lock_parent(struct dentry *dentry) return __lock_parent(dentry); } -static inline bool retain_dentry(struct dentry *dentry) +/* + * Return true if dentry is negative and exceed negative dentry limit. + */ +static inline bool limit_negative_dentry(struct dentry *dentry) { struct dentry *parent; + parent = dentry->d_parent; + if (unlikely(!parent)) + return false; + + WARN_ON((atomic_read(&parent->d_neg_dnum) < 0)); + if (!dentry->d_inode) { + if (!(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT)) { + unsigned int flags = READ_ONCE(dentry->d_flags); + + flags |= DCACHE_NEGATIVE_ACCOUNT; + WRITE_ONCE(dentry->d_flags, flags); + atomic_inc(&parent->d_neg_dnum); + } + + if (atomic_read(&parent->d_neg_dnum) >= NEG_DENTRY_LIMIT) + return true; + } + + return false; +} + +static inline bool retain_dentry(struct dentry *dentry) +{ WARN_ON(d_in_lookup(dentry)); /* Unreachable? Get rid of it */ @@ -654,27 +680,9 @@ static inline bool retain_dentry(struct dentry *dentry) return false; } - if (unlikely(!dentry->d_parent)) - goto noparent; - - parent = dentry->d_parent; - /* Return false if it's negative */ - WARN_ON((atomic_read(&parent->d_neg_dnum) < 0)); - if (!dentry->d_inode) { - if (!(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT)) { - unsigned flags = READ_ONCE(dentry->d_flags); - - flags |= DCACHE_NEGATIVE_ACCOUNT; - WRITE_ONCE(dentry->d_flags, flags); - atomic_inc(&parent->d_neg_dnum); - } - } - - if (!dentry->d_inode && - atomic_read(&parent->d_neg_dnum) >= NEG_DENTRY_LIMIT) + if (unlikely(limit_negative_dentry(dentry))) return false; -noparent: /* retain; LRU fodder */ dentry->d_lockref.count--; if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) @@ -808,8 +816,14 @@ static inline bool fast_dput(struct dentry *dentry) d_flags &= DCACHE_REFERENCED | DCACHE_LRU_LIST | DCACHE_DISCONNECTED; /* Nothing to do? Dropping the reference was all we needed? */ - if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) - return true; + if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) { + /* + * If dentry is negative and need to be limitted, it maybe need + * to be killed, so shouldn't fast put and retain in memory. + */ + if (likely(!limit_negative_dentry(dentry))) + return true; + } /* * Not the fast normal case? Get the lock. We've already decremented -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS] 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/I8YHBT -------------------------------- 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: 453eaea60e92 ("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 ce6bb4304949..81e3106962e3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -116,6 +116,7 @@ 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); #endif @@ -5218,7 +5219,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_held(&rq->lock); #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

[PATCH OLK-6.6 v2 0/4] reserve space for arm64 related structures.
by Yuntao Liu 24 Jan '24

24 Jan '24

Yuntao Liu (4): kabi: reserve space for struct efi.h:efi kabi: reserve space for struct extable.h:exception_table_entry kabi: reserve space for struct fb.h:fb_deferred_io fb.h:fb_info kabi: reserve space for struct processor.h:thread_struct arch/arm64/include/asm/extable.h | 3 +++ arch/arm64/include/asm/processor.h | 9 +++++++++ include/linux/efi.h | 3 +++ include/linux/fb.h | 7 +++++++ 4 files changed, 22 insertions(+) -- 2.34.1

2 5

openEuler Kernel SIG双周例会
by openEuler conference 24 Jan '24

24 Jan '24

您好！ Kernel 邀请您参加 2024-01-26 14:00 召开的WeLink会议(自动录制) 会议主题：openEuler Kernel SIG双周例会会议内容： 1. 进展update 2. 议题征集中（新增议题可直接回复本邮件申报，或者填入会议看板）会议链接：https://bmeeting.huaweicloud.com:36443/#/j/967727869 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 更多资讯尽在：https://www.openeuler.org/zh/ Hello! Kernel invites you to attend the WeLink conference(auto recording) will be held at 2024-01-26 14:00, The subject of the conference is openEuler Kernel SIG双周例会, Summary: 1. 进展update 2. 议题征集中（新增议题可直接回复本邮件申报，或者填入会议看板） You can join the meeting at https://bmeeting.huaweicloud.com:36443/#/j/967727869. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. More information: https://www.openeuler.org/en/

1 0

[openeuler:OLK-6.6 1689/2572] include/linux/suspend.h:36:60: error: expected ')' before numeric constant
by kernel test robot 24 Jan '24

24 Jan '24

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: 77caa0f8fcb7c7abbb34b5b49ae26c34c042fdc7 commit: 1f46f3a128f21c190281c72cad99267bcc177e34 [1689/2572] PCI: PM: fix pcie mrrs restoring config: loongarch-allnoconfig (https://download.01.org/0day-ci/archive/20240124/202401241339.EjsOdYEo-lkp@…) compiler: loongarch64-linux-gcc (GCC) 13.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240124/202401241339.EjsOdYEo-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/202401241339.EjsOdYEo-lkp@intel.com/ All errors (new ones prefixed by >>): In file included from drivers/pci/pci.c:35: >> include/linux/suspend.h:36:60: error: expected ')' before numeric constant 36 | #define PM_SUSPEND_ON ((__force suspend_state_t) 0) | ^ include/linux/suspend.h:341:34: note: in expansion of macro 'PM_SUSPEND_ON' 341 | #define pm_suspend_target_state (PM_SUSPEND_ON) | ^~~~~~~~~~~~~ drivers/pci/pci.c:177:17: note: in expansion of macro 'pm_suspend_target_state' 177 | suspend_state_t pm_suspend_target_state; | ^~~~~~~~~~~~~~~~~~~~~~~ vim +36 include/linux/suspend.h 95d9ffbe01fb21 Rafael J. Wysocki 2007-10-18 35 95d9ffbe01fb21 Rafael J. Wysocki 2007-10-18 @36 #define PM_SUSPEND_ON ((__force suspend_state_t) 0) 690cbb90a709c1 Rafael J. Wysocki 2017-08-10 37 #define PM_SUSPEND_TO_IDLE ((__force suspend_state_t) 1) 7e73c5ae6e7991 Zhang Rui 2013-02-06 38 #define PM_SUSPEND_STANDBY ((__force suspend_state_t) 2) 95d9ffbe01fb21 Rafael J. Wysocki 2007-10-18 39 #define PM_SUSPEND_MEM ((__force suspend_state_t) 3) 690cbb90a709c1 Rafael J. Wysocki 2017-08-10 40 #define PM_SUSPEND_MIN PM_SUSPEND_TO_IDLE 95d9ffbe01fb21 Rafael J. Wysocki 2007-10-18 41 #define PM_SUSPEND_MAX ((__force suspend_state_t) 4) 95d9ffbe01fb21 Rafael J. Wysocki 2007-10-18 42 :::::: The code at line 36 was first introduced by commit :::::: 95d9ffbe01fb21d524c86bf77871255066bc6e55 PM: Move definition of struct pm_ops to suspend.h :::::: TO: Rafael J. Wysocki <rjw(a)sisk.pl> :::::: CC: Linus Torvalds <torvalds(a)woody.linux-foundation.org> -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[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