[PATCH v2] powerpc: always enable queued spinlocks for 64s, disable for others

[PATCH v2] powerpc: always enable queued spinlocks for 64s, disable for others

[Nicholas Piggin]

Queued spinlocks have shown to have good performance and fairness
properties even on smaller (2 socket) POWER systems. This selects
them automatically for 64s. For other platforms they are de-selected,
the standard spinlock is far simpler and smaller code, and single
chips with a handful of cores is unlikely to show any improvement.

CONFIG_EXPERT still allows this to be changed, e.g., to help debug
performance or correctness issues.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
Since v1, made the condition simpler as suggested by Christophe.
Verified this works as expected for a bunch of old configs
(e.g., updates old !EXPERT configs to PPC_QUEUED_SPINLOCKS=y if they
were previously =n).

Thanks,
Nick

 arch/powerpc/Kconfig | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 107bb4319e0e..eebb4a8c156c 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -505,18 +505,14 @@ config HOTPLUG_CPU
 	  Say N if you are unsure.
 
 config PPC_QUEUED_SPINLOCKS
-	bool "Queued spinlocks"
+	bool "Queued spinlocks" if EXPERT
 	depends on SMP
+	default PPC_BOOK3S_64
 	help
 	  Say Y here to use queued spinlocks which give better scalability and
 	  fairness on large SMP and NUMA systems without harming single threaded
 	  performance.
 
-	  This option is currently experimental, the code is more complex and
-	  less tested so it defaults to "N" for the moment.
-
-	  If unsure, say "N".
-
 config ARCH_CPU_PROBE_RELEASE
 	def_bool y
 	depends on HOTPLUG_CPU
-- 
2.23.0

Re: [PATCH v2] powerpc: always enable queued spinlocks for 64s, disable for others

[Michael Ellerman]

On Mon, 18 Jan 2021 22:34:51 +1000, Nicholas Piggin wrote:
> Queued spinlocks have shown to have good performance and fairness
> properties even on smaller (2 socket) POWER systems. This selects
> them automatically for 64s. For other platforms they are de-selected,
> the standard spinlock is far simpler and smaller code, and single
> chips with a handful of cores is unlikely to show any improvement.
> 
> CONFIG_EXPERT still allows this to be changed, e.g., to help debug
> performance or correctness issues.

Applied to powerpc/next.

[1/1] powerpc: Always enable queued spinlocks for 64s, disable for others
      https://git.kernel.org/powerpc/c/c9f3401313a5089f100d7d1ef4b75cd7b49b2190

cheers