[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <5722C834.1030209@linux.vnet.ibm.com>
Date: Fri, 29 Apr 2016 10:34:28 +0800
From: Pan Xinhui <xinhui@...ux.vnet.ibm.com>
To: Waiman Long <waiman.long@....com>
CC: linux-kernel@...r.kernel.org,
virtualization@...ts.linux-foundation.org,
linuxppc-dev@...ts.ozlabs.org, peterz@...radead.org,
Boqun Feng <boqun.feng@...il.com>,
Paul Mackerras <paulus@...ba.org>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
Michael Ellerman <mpe@...erman.id.au>, jeremy@...p.org,
chrisw@...s-sol.org, akataria@...are.com, rusty@...tcorp.com.au,
gwshan@...ux.vnet.ibm.com, adam.buchbinder@...il.com,
tglx@...utronix.de, marc.zyngier@....com
Subject: Re: [PATCH resend] powerpc: enable qspinlock and its virtualization
support
On 2016年04月29日 05:07, Waiman Long wrote:
> On 04/28/2016 06:55 AM, Pan Xinhui wrote:
>> From: Pan Xinhui<xinhui.pan@...ux.vnet.ibm.com>
>>
>> This patch aims to enable qspinlock on PPC. And on pseries platform, it also support
>> paravirt qspinlock.
>>
>> Signed-off-by: Pan Xinhui<xinhui.pan@...ux.vnet.ibm.com>
>> ---
>> arch/powerpc/include/asm/qspinlock.h | 37 +++++++++++++++
>> arch/powerpc/include/asm/qspinlock_paravirt.h | 36 +++++++++++++++
>> .../powerpc/include/asm/qspinlock_paravirt_types.h | 13 ++++++
>> arch/powerpc/include/asm/spinlock.h | 31 ++++++++-----
>> arch/powerpc/include/asm/spinlock_types.h | 4 ++
>> arch/powerpc/kernel/paravirt.c | 52 ++++++++++++++++++++++
>> arch/powerpc/lib/locks.c | 32 +++++++++++++
>> arch/powerpc/platforms/pseries/setup.c | 5 +++
>> 8 files changed, 198 insertions(+), 12 deletions(-)
>> create mode 100644 arch/powerpc/include/asm/qspinlock.h
>> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt.h
>> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt_types.h
>> create mode 100644 arch/powerpc/kernel/paravirt.c
>>
>>
>
> This is just an enablement patch. You will also need a patch to activate qspinlock for, at lease, some PPC configs. Right?
>
yep, I want to enable these config and makefile at last.
it just looks like
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index 2da380f..ae7c2f1 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -50,6 +50,7 @@ obj-$(CONFIG_PPC_970_NAP) += idle_power4.o
obj-$(CONFIG_PPC_P7_NAP) += idle_power7.o
procfs-y := proc_powerpc.o
obj-$(CONFIG_PROC_FS) += $(procfs-y)
+obj-$(CONFIG_PARAVIRT_SPINLOCKS) += paravirt.o
rtaspci-$(CONFIG_PPC64)-$(CONFIG_PCI) := rtas_pci.o
obj-$(CONFIG_PPC_RTAS) += rtas.o rtas-rtc.o $(rtaspci-y-y)
obj-$(CONFIG_PPC_RTAS_DAEMON) += rtasd.o
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index bec90fb..46632e4 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -21,6 +21,7 @@ config PPC_PSERIES
select HOTPLUG_CPU if SMP
select ARCH_RANDOM
select PPC_DOORBELL
+ select ARCH_USE_QUEUED_SPINLOCKS
default y
config PPC_SPLPAR
@@ -127,3 +128,11 @@ config HV_PERF_CTRS
systems. 24x7 is available on Power 8 systems.
If unsure, select Y.
+
+config PARAVIRT_SPINLOCKS
+ bool "Paravirtialization support for qspinlock"
+ depends on PPC_SPLPAR && QUEUED_SPINLOCKS
+ default y
+ help
+ If platform supports virtualization, for example PowerVM, this option
+ can let guest have a better performace.
--
2.4.3
> It has dependency on the pv_wait() patch that I sent out extend the parameter list. Some performance data on how PPC system will perform with and without qspinlock will also be helpful data points.
>
For now, pv_wait defined in ppc is static inline void pv_wait(u8 *ptr, u8 val)
My plan is that waiting your patch goes in kernel tree first, then I send out another patch to extend the parameter list.
yes, I need copy some performance data in my patch's comments.
thanks
xinhui
> Cheers,
> Longman
>
Powered by blists - more mailing lists