search for: rwsem

Btrfs uses an rwsem to control access to its extent tree. Threads will hold a read lock on this rwsem while they scan the extent tree, and if need_resched() they will drop the lock and schedule. The transaction commit needs to take a write lock for this rwsem for a very short period to switch out the commit roots....

As the vcpu_is_preempted() call is pretty costly compared with other checks within mutex_spin_on_owner() and rwsem_spin_on_owner(), they are done at a reduce frequency of once every 256 iterations. Signed-off-by: Waiman Long <longman at redhat.com> --- kernel/locking/mutex.c | 5 ++++- kernel/locking/rwsem-xadd.c | 6 ++++-- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/kernel/lock...

...wner 8.45% sched-messaging [kernel.vmlinux] [k] mutex_unlock 4.12% sched-messaging [kernel.vmlinux] [k] system_call 3.01% sched-messaging [kernel.vmlinux] [k] system_call_common 2.83% sched-messaging [kernel.vmlinux] [k] copypage_power7 2.64% sched-messaging [kernel.vmlinux] [k] rwsem_spin_on_owner 2.00% sched-messaging [kernel.vmlinux] [k] osq_lock after patch: 9.99% sched-messaging [kernel.vmlinux] [k] mutex_unlock 5.28% sched-messaging [unknown] [H] 0xc0000000000768e0 4.27% sched-messaging [kernel.vmlinux] [k] __copy_tofrom_user_power7 3.77% sched-me...

...in process (950) terminated with status 1 ====================================================== [ INFO: possible circular locking dependency detected ] 3.8.0-rc2-expert #1 Not tainted ------------------------------------------------------- Xorg/1193 is trying to acquire lock: ((fb_notifier_list).rwsem){++++.+}, at: [<ffffffff810697c1>] __blocking_notifier_call_chain+0x51/0xc0 but task is already holding lock: (console_lock){+.+.+.}, at: [<ffffffff81263f95>] do_fb_ioctl+0x2e5/0x5f0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -...

The freeze rwsem is taken by sb_start_intwrite() and dropped during the commit_ or end_transaction(). In the async case, that happens in a worker thread. Tell lockdep the calling thread is releasing ownership of the rwsem and the async thread is picking it up. Josef and I worked out a more complicated solution t...

On Wed, Feb 08, 2017 at 01:00:25PM -0500, Waiman Long wrote: > As the vcpu_is_preempted() call is pretty costly compared with other > checks within mutex_spin_on_owner() and rwsem_spin_on_owner(), they > are done at a reduce frequency of once every 256 iterations. That's just disgusting.

On Wed, Feb 08, 2017 at 01:00:25PM -0500, Waiman Long wrote: > As the vcpu_is_preempted() call is pretty costly compared with other > checks within mutex_spin_on_owner() and rwsem_spin_on_owner(), they > are done at a reduce frequency of once every 256 iterations. That's just disgusting.

It was found when running fio sequential write test with a XFS ramdisk on a 2-socket x86-64 system, the %CPU times as reported by perf were as follows: 71.27% 0.28% fio [k] down_write 70.99% 0.01% fio [k] call_rwsem_down_write_failed 69.43% 1.18% fio [k] rwsem_down_write_failed 65.51% 54.57% fio [k] osq_lock 9.72% 7.99% fio [k] __raw_callee_save___kvm_vcpu_is_preempted 4.16% 4.16% fio [k] __kvm_vcpu_is_preempted So making vcpu_is_preempted() a callee-save function has a pretty high cost assoc...

It was found when running fio sequential write test with a XFS ramdisk on a 2-socket x86-64 system, the %CPU times as reported by perf were as follows: 71.27% 0.28% fio [k] down_write 70.99% 0.01% fio [k] call_rwsem_down_write_failed 69.43% 1.18% fio [k] rwsem_down_write_failed 65.51% 54.57% fio [k] osq_lock 9.72% 7.99% fio [k] __raw_callee_save___kvm_vcpu_is_preempted 4.16% 4.16% fio [k] __kvm_vcpu_is_preempted So making vcpu_is_preempted() a callee-save function has a pretty high cost assoc...

change from v2: no code change, fix typos, update some comments change from v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report before patch: 18.09% sched-messaging [kernel.vmli...

change from v2: no code change, fix typos, update some comments change from v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report before patch: 18.09% sched-messaging [kernel.vmli...

...915/i915_gem_request.c | 2 +- drivers/vhost/net.c | 4 ++-- kernel/locking/mcs_spinlock.h | 4 ++-- kernel/locking/mutex.c | 4 ++-- kernel/locking/osq_lock.c | 6 +++--- kernel/locking/qrwlock.c | 6 +++--- kernel/locking/rwsem-xadd.c | 4 ++-- lib/lockref.c | 2 +- 8 files changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 8832f8e..383d134 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c...

...s(-) diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 6116e3b..c54e9f8 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -464,6 +464,9 @@ long drm_ioctl(struct file *filp, } else memset(kdata, 0, usize); + if (dev->driver->ioctls_need_rwsem) + down_read(&dev->ioctls_rwsem); + if (ioctl->flags & DRM_UNLOCKED) retcode = func(dev, kdata, file_priv); else { @@ -472,6 +475,9 @@ long drm_ioctl(struct file *filp, mutex_unlock(&drm_global_mutex); } + if (dev->driver->ioctls_need_rwsem) + up_read...

...c1>] SyS_ioctl+0x81/0xa0 [ 2372.462991] [<ffffffff817045a4>] tracesys+0xdd/0xe2 ==================================================== It''s because that we don''t do the right thing when checking if it''s ok to tell lockdep that we''re trying to release the rwsem. If the trans handle''s type is TRANS_ATTACH, we won''t acquire the freeze rwsem, but as TRANS_ATTACH fits the check (trans < TRANS_JOIN_NOLOCK), we''ll release the freeze rwsem, which makes lockdep complains a lot. Reported-by: Ma Jianpeng <majianpeng@gmail.com>...

...3 12:41:25 sp2 kernel: d9d18000 bf384f01 00000ceb 0000b2f9 00000001 d9d1810c c2013ac4 edc5de40 Apr 3 12:41:25 sp2 kernel: 08515c98 c6cb37b8 c2014464 c200cc80 00000020 00000000 00000000 00000000 Apr 3 12:41:25 sp2 kernel: Call Trace: Apr 3 12:41:25 sp2 kernel: [<c0622f16>] rwsem_down_write_failed+0x126/0x141 Apr 3 12:41:25 sp2 kernel: [<c0439989>] .text.lock.rwsem+0x2b/0x3a Apr 3 12:41:25 sp2 kernel: [<c046aa6a>] sys_mprotect+0xbd/0x1eb Apr 3 12:41:25 sp2 kernel: [<c0404f4b>] syscall_call+0x7/0xb Apr 3 12:41:25 sp2 kernel: ====================...

change from v3: add x86 vcpu preempted check patch change from v2: no code change, fix typos, update some comments change from v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report 18.09% sched-messaging [kernel.vmlinux] [k] osq_...

change from v3: add x86 vcpu preempted check patch change from v2: no code change, fix typos, update some comments change from v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report 18.09% sched-messaging [kernel.vmlinux] [k] osq_...

change fomr v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report 18.09% sched-messaging [kernel.vmlinux] [k] osq_...

change fomr v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report 18.09% sched-messaging [kernel.vmlinux] [k] osq_...

...patch change from v3: add x86 vcpu preempted check patch change from v2: no code change, fix typos, update some comments change from v1: a simplier definition of default vcpu_is_preempted skip mahcine type check on ppc, and add config. remove dedicated macro. add one patch to drop overload of rwsem_spin_on_owner and mutex_spin_on_owner. add more comments thanks boqun and Peter's suggestion. This patch set aims to fix lock holder preemption issues. test-case: perf record -a perf bench sched messaging -g 400 -p && perf report 18.09% sched-messaging [kernel.vmlinux] [k] osq_...