On Thu, 2014-08-21 at 03:32AM +0200, Daniel Lezcano wrote:
> On 08/20/2014 10:41 PM, Soren Brinkmann wrote:
> >The hotplug code contains only a single function, which is an SMP
> >function. Move that to platsmp.c where all other SMP runctions reside.
> >That allows removing hotplug.c and declaring the cpu_die function
> >static.
> >
> >Signed-off-by: Soren Brinkmann <soren.brinkmann@xxxxxxxxxx>
> >---
> > arch/arm/mach-zynq/Makefile | 1 -
> > arch/arm/mach-zynq/common.h | 3 +--
> > arch/arm/mach-zynq/hotplug.c | 17 -----------------
> > arch/arm/mach-zynq/platsmp.c | 18 ++++++++++++++++++
> > 4 files changed, 19 insertions(+), 20 deletions(-)
> >
> >diff --git a/arch/arm/mach-zynq/Makefile b/arch/arm/mach-zynq/Makefile
> >index 820dff6e1eba..c85fb3f7d5cd 100644
> >--- a/arch/arm/mach-zynq/Makefile
> >+++ b/arch/arm/mach-zynq/Makefile
> >@@ -6,5 +6,4 @@
> > obj-y := common.o slcr.o pm.o
> > CFLAGS_REMOVE_hotplug.o =-march=armv6k
> > CFLAGS_hotplug.o =-Wa,-march=armv7-a -mcpu=cortex-a9
> >-obj-$(CONFIG_HOTPLUG_CPU) += hotplug.o
> > obj-$(CONFIG_SMP) += headsmp.o platsmp.o
> >diff --git a/arch/arm/mach-zynq/common.h b/arch/arm/mach-zynq/common.h
> >index c0773e87e83c..e6bb12c50a23 100644
> >--- a/arch/arm/mach-zynq/common.h
> >+++ b/arch/arm/mach-zynq/common.h
> >@@ -39,8 +39,7 @@ extern struct smp_operations zynq_smp_ops __initdata;
> >
> > extern void __iomem *zynq_scu_base;
> >
> >-/* Hotplug */
> >-extern void zynq_platform_cpu_die(unsigned int cpu);
> >+int zynq_pm_late_init(void);
> >
> > int zynq_pm_late_init(void);
> >
> >diff --git a/arch/arm/mach-zynq/hotplug.c b/arch/arm/mach-zynq/hotplug.c
> >index fe44a05677e2..b685c89f11e4 100644
> >--- a/arch/arm/mach-zynq/hotplug.c
> >+++ b/arch/arm/mach-zynq/hotplug.c
> >@@ -12,20 +12,3 @@
> > */
> > #include <asm/proc-fns.h>
> >
> >-/*
> >- * platform-specific code to shutdown a CPU
> >- *
> >- * Called with IRQs disabled
> >- */
> >-void zynq_platform_cpu_die(unsigned int cpu)
> >-{
> >- zynq_slcr_cpu_state_write(cpu, true);
> >-
> >- /*
> >- * there is no power-control hardware on this platform, so all
> >- * we can do is put the core into WFI; this is safe as the calling
> >- * code will have already disabled interrupts
> >- */
> >- for (;;)
> >- cpu_do_idle();
> >-}
> >diff --git a/arch/arm/mach-zynq/platsmp.c b/arch/arm/mach-zynq/platsmp.c
> >index f77f7ca4c45b..04e578718aa2 100644
> >--- a/arch/arm/mach-zynq/platsmp.c
> >+++ b/arch/arm/mach-zynq/platsmp.c
> >@@ -132,6 +132,24 @@ static int zynq_cpu_kill(unsigned cpu)
> > zynq_slcr_cpu_stop(cpu);
> > return 1;
> > }
> >+
> >+/*
> >+ * platform-specific code to shutdown a CPU
> >+ *
> >+ * Called with IRQs disabled
> >+ */
> >+static void zynq_platform_cpu_die(unsigned int cpu)
> >+{
> >+ zynq_slcr_cpu_state_write(cpu, true);
> >+
> >+ /*
> >+ * there is no power-control hardware on this platform, so all
> >+ * we can do is put the core into WFI; this is safe as the calling
> >+ * code will have already disabled interrupts
> >+ */
> >+ for (;;)
> >+ cpu_do_idle();
>
> IIUC, the cpu_do_idle() will flush the L1 cache and then call the
> WFI. It makes sense if we are about to power down the core. So I am
> wondering if we can just call wfi() instead.
I'm not sure - it's not that trivial to trace that through the sources.
But I think cpu_do_idle ends up in cpu_v7_do_idle which is just:
ENTRY(cpu_v7_do_idle)
dsb @ WFI may enter a low-power mode
wfi
ret lr
ENDPROC(cpu_v7_do_idle)
I think that is what we want here.
Thanks,
Sören
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
