similar to: [RFC] Implement Batched (group) ticket lock

On 05/28/2014 08:16 AM, Raghavendra K T wrote: > > TODO: > - we need an intelligent way to nullify the effect of batching for baremetal > (because extra cmpxchg is not required). To do this, you will need to have 2 slightly different algorithms depending on the paravirt_ticketlocks_enabled jump label. > > - My kernbench/ebizzy test on baremetal (32 cpu +ht sandybridge) did

On 05/28/2014 08:16 AM, Raghavendra K T wrote: This patch looks very promising. > TODO: > - we need an intelligent way to nullify the effect of batching for baremetal > (because extra cmpxchg is not required). On (larger?) NUMA systems, the unfairness may be a nice performance benefit, reducing cache line bouncing through the system, and it could well outweigh the extra cmpxchg at

In virtualized environment there are mainly three problems related to spinlocks that affects performance. 1. LHP (lock holder preemption) 2. Lock Waiter Preemption (LWP) 3. Starvation/fairness Though Ticketlocks solve fairness problem it worsens LWP, LHP problems. Though pv-ticketlocks tried to address these problems we can further improve at the cost of relaxed fairness. The following patch

In virtualized environment there are mainly three problems related to spinlocks that affects performance. 1. LHP (lock holder preemption) 2. Lock Waiter Preemption (LWP) 3. Starvation/fairness Though Ticketlocks solve fairness problem it worsens LWP, LHP problems. Though pv-ticketlocks tried to address these problems we can further improve at the cost of relaxed fairness. The following patch

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which is

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which is

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which is

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which is

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

Paravirt spinlock clears slowpath flag after doing unlock. As explained by Linus currently it does: prev = *lock; add_smp(&lock->tickets.head, TICKET_LOCK_INC); /* add_smp() is a full mb() */ if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) __ticket_unlock_slowpath(lock, prev); which

To speed up paravirtualized spinlock handling when running on bare metal introduce a new pvops function "spin_unlock". This is a simple add instruction (possibly with lock prefix) when the kernel is running on bare metal. As the patched instruction includes a lock prefix in some configurations annotate this location to be subject to lock prefix patching. This is working even if

Paravirtualized spinlocks produce some overhead even if the kernel is running on bare metal. The main reason are the more complex locking and unlocking functions. Especially unlocking is no longer just one instruction but so complex that it is no longer inlined. This patch series addresses this issue by adding two more pvops functions to reduce the size of the inlined spinlock functions. When

Paravirtualized spinlocks produce some overhead even if the kernel is running on bare metal. The main reason are the more complex locking and unlocking functions. Especially unlocking is no longer just one instruction but so complex that it is no longer inlined. This patch series addresses this issue by adding two more pvops functions to reduce the size of the inlined spinlock functions. When

* Raghavendra K T <raghavendra.kt at linux.vnet.ibm.com> [2015-02-15 11:25:44]: Resending the V5 with smp_mb__after_atomic() change without bumping up revision ---8<--- >From 0b9ecde30e3bf5b5b24009fd2ac5fc7ac4b81158 Mon Sep 17 00:00:00 2001 From: Raghavendra K T <raghavendra.kt at linux.vnet.ibm.com> Date: Fri, 6 Feb 2015 16:44:11 +0530 Subject: [PATCH RESEND V5] x86 spinlock:

* Raghavendra K T <raghavendra.kt at linux.vnet.ibm.com> [2015-02-15 11:25:44]: Resending the V5 with smp_mb__after_atomic() change without bumping up revision ---8<--- >From 0b9ecde30e3bf5b5b24009fd2ac5fc7ac4b81158 Mon Sep 17 00:00:00 2001 From: Raghavendra K T <raghavendra.kt at linux.vnet.ibm.com> Date: Fri, 6 Feb 2015 16:44:11 +0530 Subject: [PATCH RESEND V5] x86 spinlock:

On 02/06/2015 06:49 AM, Raghavendra K T wrote: > Paravirt spinlock clears slowpath flag after doing unlock. > As explained by Linus currently it does: > prev = *lock; > add_smp(&lock->tickets.head, TICKET_LOCK_INC); > > /* add_smp() is a full mb() */ > > if (unlikely(lock->tickets.tail &