[PATCH v4 2/6] powerpc: move spinlock implementation to simple_spinlock

To prepare for queued spinlocks. This is a simple rename except to update preprocessor guard name and a file reference. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/simple_spinlock.h| 288 ++ .../include/asm/simple_spinlock_types.h | 21 ++ arch/powerpc

[PATCH v4 6/6] powerpc: implement smp_cond_load_relaxed

This implements smp_cond_load_relaed with the slowpath busy loop using the preferred SMT priority pattern. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/barrier.h | 14 ++ 1 file changed, 14 insertions(+) diff --git a/arch/powerpc/include/asm/barrier.h b/arch/powerpc

[PATCH v4 1/6] powerpc/pseries: move some PAPR paravirt functions to their own file

These functions will be used by queued spinlock implementation, and may be useful elsewhere too, so move them out of spinlock.h. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 59 + arch/powerpc/include/asm/spinlock.h | 24 +--- arch

[PATCH v4 5/6] powerpc/qspinlock: optimised atomic_try_cmpxchg_lock that adds the lock hint

This brings the behaviour of the uncontended fast path back to roughly equivalent to simple spinlocks -- a single atomic op with lock hint. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/atomic.h| 28 arch/powerpc/include/asm/qspinlock.h | 2 +- 2

[PATCH v4 3/6] powerpc/64s: implement queued spinlocks and rwlocks

regressions haven't been analysed very well yet, there are a lot of things that can be tuned, particularly the paravirtualised locking, but the numbers already look like a good net win even on relatively small systems. Acked-by: Peter Zijlstra (Intel) Signed-off-by: Nicholas Piggin --- arch/po

[PATCH v4 4/6] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

investigated and improved in future. Performance results can be found in the commit which added queued spinlocks. Acked-by: Peter Zijlstra (Intel) Acked-by: Waiman Long Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 28 arch/powerpc/include/asm

[PATCH v4 0/6] powerpc: queued spinlocks and rwlocks

cool technology and great results especially on the big systems but even on smaller ones there are nice gains. Thanks Waiman and everyone who developed it. Thanks, Nick Nicholas Piggin (6): powerpc/pseries: move some PAPR paravirt functions to their own file powerpc: move spinlock implemen

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Waiman Long's message of July 24, 2020 12:29 am: > On 7/23/20 9:30 AM, Nicholas Piggin wrote: >>> I would prefer to extract out the pending bit handling code out into a >>> separate helper function which can be overridden by the arch code >>> inst

Re: [PATCH v3 5/6] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

Excerpts from Michael Ellerman's message of July 9, 2020 8:53 pm: > Nicholas Piggin writes: > >> Signed-off-by: Nicholas Piggin >> --- >> arch/powerpc/include/asm/paravirt.h | 28 >> arch/powerpc/include/asm/qspinlock.h | 66 +

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Waiman Long's message of July 22, 2020 12:36 am: > On 7/21/20 7:08 AM, Nicholas Piggin wrote: >> diff --git a/arch/powerpc/include/asm/qspinlock.h >> b/arch/powerpc/include/asm/qspinlock.h >> index b752d34517b3..26d8766a1106 100644 >> --- a/arch

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Peter Zijlstra's message of July 9, 2020 6:31 pm: > On Wed, Jul 08, 2020 at 07:54:34PM -0400, Waiman Long wrote: >> On 7/8/20 4:41 AM, Peter Zijlstra wrote: >> > On Tue, Jul 07, 2020 at 03:57:06PM +1000, Nicholas Piggin wrote: >> > > Yes, powerpc c

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Peter Zijlstra's message of July 8, 2020 6:41 pm: > On Tue, Jul 07, 2020 at 03:57:06PM +1000, Nicholas Piggin wrote: >> Yes, powerpc could certainly get more performance out of the slow >> paths, and then there are a few parameters to tune. > Sorry for the

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Waiman Long's message of July 8, 2020 1:33 pm: > On 7/7/20 1:57 AM, Nicholas Piggin wrote: >> Yes, powerpc could certainly get more performance out of the slow >> paths, and then there are a few parameters to tune. >> >> We don't have a good alte

Re: [PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

Excerpts from Waiman Long's message of July 7, 2020 4:39 am: > On 7/6/20 12:35 AM, Nicholas Piggin wrote: >> v3 is updated to use __pv_queued_spin_unlock, noticed by Waiman (thank you). >> >> Thanks, >> Nick >> >> Nicholas Piggin (6): >>

[PATCH v3 4/6] powerpc/64s: implement queued spinlocks and rwlocks

. Signed-off-by: Nicholas Piggin --- arch/powerpc/Kconfig | 13 arch/powerpc/include/asm/Kbuild | 2 ++ arch/powerpc/include/asm/qspinlock.h | 25 +++ arch/powerpc/include/asm/spinlock.h | 5 + arch/powerpc/include

[PATCH v3 5/6] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 28 arch/powerpc/include/asm/qspinlock.h | 66 +++ arch/powerpc/include/asm/qspinlock_paravirt.h | 7 ++ arch/powerpc/platforms/pseries/Kconfig| 5 ++ arch/powerpc

[PATCH v3 6/6] powerpc/qspinlock: optimised atomic_try_cmpxchg_lock that adds the lock hint

This brings the behaviour of the uncontended fast path back to roughly equivalent to simple spinlocks -- a single atomic op with lock hint. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/atomic.h| 28 arch/powerpc/include/asm/qspinlock.h | 2 +- 2

[PATCH v3 2/6] powerpc/pseries: move some PAPR paravirt functions to their own file

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 61 + arch/powerpc/include/asm/spinlock.h | 24 +--- arch/powerpc/lib/locks.c| 12 +++--- 3 files changed, 68 insertions(+), 29 deletions(-) create mode 100644 arch/powerpc

[PATCH v3 3/6] powerpc: move spinlock implementation to simple_spinlock

To prepare for queued spinlocks. This is a simple rename except to update preprocessor guard name and a file reference. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/simple_spinlock.h| 292 ++ .../include/asm/simple_spinlock_types.h | 21 ++ arch/powerpc

[PATCH v3 1/6] powerpc/powernv: must include hvcall.h to get PAPR defines

An include goes away in future patches which breaks compilation without this. Signed-off-by: Nicholas Piggin --- arch/powerpc/platforms/powernv/pci-ioda-tce.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/platforms/powernv/pci-ioda-tce.c b/arch/powerpc/platforms/powernv/pci

[PATCH v3 0/6] powerpc: queued spinlocks and rwlocks

v3 is updated to use __pv_queued_spin_unlock, noticed by Waiman (thank you). Thanks, Nick Nicholas Piggin (6): powerpc/powernv: must include hvcall.h to get PAPR defines powerpc/pseries: move some PAPR paravirt functions to their own file powerpc: move spinlock implementation to

Re: [PATCH v2 5/6] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

Excerpts from Waiman Long's message of July 6, 2020 5:00 am: > On 7/3/20 3:35 AM, Nicholas Piggin wrote: >> Signed-off-by: Nicholas Piggin >> --- >> arch/powerpc/include/asm/paravirt.h | 28 ++ >> arch/powerpc/inclu

[PATCH v2 6/6] powerpc/qspinlock: optimised atomic_try_cmpxchg_lock that adds the lock hint

This brings the behaviour of the uncontended fast path back to roughly equivalent to simple spinlocks -- a single atomic op with lock hint. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/atomic.h| 28 arch/powerpc/include/asm/qspinlock.h | 2 +- 2

[PATCH v2 0/6] powerpc: queued spinlocks and rwlocks

v2 is updated to account for feedback from Will, Peter, and Waiman (thank you), and trims off a couple of RFC and unrelated patches. Thanks, Nick Nicholas Piggin (6): powerpc/powernv: must include hvcall.h to get PAPR defines powerpc/pseries: move some PAPR paravirt functions to their own

[PATCH v2 3/6] powerpc: move spinlock implementation to simple_spinlock

To prepare for queued spinlocks. This is a simple rename except to update preprocessor guard name and a file reference. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/simple_spinlock.h| 292 ++ .../include/asm/simple_spinlock_types.h | 21 ++ arch/powerpc

[PATCH v2 1/6] powerpc/powernv: must include hvcall.h to get PAPR defines

An include goes away in future patches which breaks compilation without this. Signed-off-by: Nicholas Piggin --- arch/powerpc/platforms/powernv/pci-ioda-tce.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/platforms/powernv/pci-ioda-tce.c b/arch/powerpc/platforms/powernv/pci

[PATCH v2 2/6] powerpc/pseries: move some PAPR paravirt functions to their own file

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 61 + arch/powerpc/include/asm/spinlock.h | 24 +--- arch/powerpc/lib/locks.c| 12 +++--- 3 files changed, 68 insertions(+), 29 deletions(-) create mode 100644 arch/powerpc

[PATCH v2 4/6] powerpc/64s: implement queued spinlocks and rwlocks

. Signed-off-by: Nicholas Piggin --- arch/powerpc/Kconfig | 13 arch/powerpc/include/asm/Kbuild | 2 ++ arch/powerpc/include/asm/qspinlock.h | 25 +++ arch/powerpc/include/asm/spinlock.h | 5 + arch/powerpc/include

[PATCH v2 5/6] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 28 ++ arch/powerpc/include/asm/qspinlock.h | 55 +++ arch/powerpc/include/asm/qspinlock_paravirt.h | 5 ++ arch/powerpc/platforms/pseries/Kconfig| 5 ++ arch/powerpc

Re: [PATCH 5/8] powerpc/64s: implement queued spinlocks and rwlocks

Excerpts from Will Deacon's message of July 2, 2020 8:35 pm: > On Thu, Jul 02, 2020 at 08:25:43PM +1000, Nicholas Piggin wrote: >> Excerpts from Will Deacon's message of July 2, 2020 6:02 pm: >> > On Thu, Jul 02, 2020 at 05:48:36PM +1000, Nicholas Piggin wrote: >

Re: [PATCH 2/8] powerpc/pseries: use smp_rmb() in H_CONFER spin yield

Excerpts from Peter Zijlstra's message of July 2, 2020 6:28 pm: > On Thu, Jul 02, 2020 at 05:48:33PM +1000, Nicholas Piggin wrote: >> There is no need for rmb(), this allows faster lwsync here. > > Since you determined this; I'm thinking you actually understand the &

Re: [PATCH 5/8] powerpc/64s: implement queued spinlocks and rwlocks

Excerpts from Will Deacon's message of July 2, 2020 6:02 pm: > On Thu, Jul 02, 2020 at 05:48:36PM +1000, Nicholas Piggin wrote: >> diff --git a/arch/powerpc/include/asm/qspinlock.h >> b/arch/powerpc/include/asm/qspinlock.h >> new file mode 100644 >> index 00

[PATCH 6/8] powerpc/pseries: implement paravirt qspinlocks for SPLPAR

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 23 arch/powerpc/include/asm/qspinlock.h | 55 +++ arch/powerpc/include/asm/qspinlock_paravirt.h | 5 ++ arch/powerpc/platforms/pseries/Kconfig| 5 ++ arch/powerpc

[PATCH 7/8] powerpc/qspinlock: optimised atomic_try_cmpxchg_lock that adds the lock hint

This brings the behaviour of the uncontended fast path back to roughly equivalent to simple spinlocks -- a single atomic op with lock hint. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/atomic.h| 28 arch/powerpc/include/asm/qspinlock.h | 2 +- 2

[PATCH 8/8] powerpc/64s: remove paravirt from simple spinlocks (RFC only)

RFC until we settle on queued spinlocks for 64s and remove the option to go back to simple locks. If other sub-archs want to keep simple spinlocks, the code can be nicely simplified. --- arch/powerpc/include/asm/simple_spinlock.h | 61 +--- arch/powerpc/kvm/book3s_hv_rm_mmu.c

[PATCH 2/8] powerpc/pseries: use smp_rmb() in H_CONFER spin yield

There is no need for rmb(), this allows faster lwsync here. Signed-off-by: Nicholas Piggin --- arch/powerpc/lib/locks.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/lib/locks.c b/arch/powerpc/lib/locks.c index 6440d5943c00..47a530de733e 100644 --- a/arch

[PATCH 4/8] powerpc: move spinlock implementation to simple_spinlock

To prepare for queued spinlocks. This is a simple rename except to update preprocessor guard name and a file reference. Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/simple_spinlock.h| 292 ++ .../include/asm/simple_spinlock_types.h | 21 ++ arch/powerpc

[PATCH 5/8] powerpc/64s: implement queued spinlocks and rwlocks

. Signed-off-by: Nicholas Piggin --- arch/powerpc/Kconfig | 13 + arch/powerpc/include/asm/Kbuild | 2 ++ arch/powerpc/include/asm/qspinlock.h | 20 arch/powerpc/include/asm/spinlock.h | 5 + arch/powerpc/include/asm

[PATCH 3/8] powerpc/pseries: move some PAPR paravirt functions to their own file

Signed-off-by: Nicholas Piggin --- arch/powerpc/include/asm/paravirt.h | 61 + arch/powerpc/include/asm/spinlock.h | 24 +--- arch/powerpc/lib/locks.c| 12 +++--- 3 files changed, 68 insertions(+), 29 deletions(-) create mode 100644 arch/powerpc

[PATCH 0/8] powerpc: queued spinlocks and rwlocks

airly soon. Thanks, Nick Nicholas Piggin (8): powerpc/powernv: must include hvcall.h to get PAPR defines powerpc/pseries: use smp_rmb() in H_CONFER spin yield powerpc/pseries: move some PAPR paravirt functions to their own file powerpc: move spinlock implementation to simple_spinlock p

[PATCH 1/8] powerpc/powernv: must include hvcall.h to get PAPR defines

An include goes away in future patches which breaks compilation without this. Signed-off-by: Nicholas Piggin --- arch/powerpc/platforms/powernv/pci-ioda-tce.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/platforms/powernv/pci-ioda-tce.c b/arch/powerpc/platforms/powernv/pci

Re: [PATCH 2/5] stop_machine: yield CPU during stop machine

On Fri, 21 Oct 2016 14:05:36 +0200 Peter Zijlstra wrote: > On Fri, Oct 21, 2016 at 01:58:55PM +0200, Christian Borntraeger wrote: > > stop_machine can take a very long time if the hypervisor does > > overcommitment for guest CPUs. When waiting for "the one", lets > > give up our CPU by using the