Re: [PATCH v3 3/4] ARM: EXYNOS: cpuidle: add secure firmware support to AFTR mode code
From: Bartlomiej Zolnierkiewicz
Date: Thu Jul 10 2014 - 06:02:31 EST
Hi,
On Thursday, July 10, 2014 10:27:10 AM Krzysztof Kozlowski wrote:
> On 09.07.2014 19:17, Bartlomiej Zolnierkiewicz wrote:
> > * Move cp15 registers saving to exynos_save_cp15() helper and add
> > additional helper usage to do_idle firmware method.
> >
> > * Use sysram_ns_base_addr + 0x24/0x20 addresses instead of the default
> > ones used by exynos_cpu_set_boot_vector() on boards with secure
> > firmware enabled.
> >
> > * Use do_idle firmware method instead of cpu_do_idle() on boards with
> > secure firmware enabled.
> >
> > Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@xxxxxxxxxxx>
> > Acked-by: Kyungmin Park <kyungmin.park@xxxxxxxxxxx>
> > ---
> > v3:
> > - make exynos_enter_aftr() return a value
> > - add cp15 registers handling to do_idle firmware method
> > - set sysram_ns_base_addr + 0x24/0x20 in do_idle firmware method
> > - move calling of do_idle firmware method from cpuidle-exynos.c
> > to pm.c
> >
> > arch/arm/mach-exynos/common.h | 2 +-
> > arch/arm/mach-exynos/firmware.c | 26 ++++++++++++++++++--------
> > arch/arm/mach-exynos/pm.c | 11 +++++++++--
> > drivers/cpuidle/cpuidle-exynos.c | 6 +++---
> > 4 files changed, 31 insertions(+), 14 deletions(-)
> >
> > diff --git a/arch/arm/mach-exynos/common.h b/arch/arm/mach-exynos/common.h
> > index a6a200f..0829808 100644
> > --- a/arch/arm/mach-exynos/common.h
> > +++ b/arch/arm/mach-exynos/common.h
> > @@ -170,7 +170,7 @@ extern int exynos_cpu_power_state(int cpu);
> > extern void exynos_cluster_power_down(int cluster);
> > extern void exynos_cluster_power_up(int cluster);
> > extern int exynos_cluster_power_state(int cluster);
> > -extern void exynos_enter_aftr(void);
> > +extern int exynos_enter_aftr(void);
> >
> > extern void s5p_init_cpu(void __iomem *cpuid_addr);
> > extern unsigned int samsung_rev(void);
> > diff --git a/arch/arm/mach-exynos/firmware.c b/arch/arm/mach-exynos/firmware.c
> > index 53fbf5c..163f5b9 100644
> > --- a/arch/arm/mach-exynos/firmware.c
> > +++ b/arch/arm/mach-exynos/firmware.c
> > @@ -24,13 +24,30 @@
> > #include "smc.h"
> >
> > #define EXYNOS_SLEEP_MAGIC 0x00000bad
> > +#define EXYNOS_AFTR_MAGIC 0xfcba0d10
> > #define EXYNOS_BOOT_ADDR 0x8
> > #define EXYNOS_BOOT_FLAG 0xc
> >
> > +/* For Cortex-A9 Diagnostic and Power control register */
> > +static unsigned int cp15_power;
> > +static unsigned int cp15_diag;
> > +
> > +static void exynos_save_cp15(void)
> > +{
> > + /* Save Power control and Diagnostic registers */
> > + asm ("mrc p15, 0, %0, c15, c0, 0\n"
> > + "mrc p15, 0, %1, c15, c0, 1\n"
> > + : "=r" (cp15_power), "=r" (cp15_diag) : : "cc");
>
> Hi,
>
> On Exynos3250 I encounter "Oops - undefined instruction" on this asm
> while entering AFTR:
> [ 2.277946] CPUidle CPU1: going off
> [ 2.278110] CPUidle CPU0: going AFTR
> [ 2.279478] Internal error: Oops - undefined instruction: 0 [#1]
> PREEMPT SMP ARM
>
>
> Are you sure it should be called on each SoC?
This should not be called on Exynos3250 (Cortex A7) but this is not
an issue currently with upstream since:
- Exynos3250 cpuidle support is disabled in Tomasz's earlier patch
- Exynos3250 need other patches (i.e. PMU support) to make cpuidle
work in upstream kernel anyway
- there is no support for any Exynos3250 board in upstream yet
Please also note that corresponding code for cp15 registers restore
needs also to be disabled for Cortex A7 and it is not present in
Tomasz's firmware patches yet. He is going to fix his patches to
check for Cortex A9 before restoring registers. Once it is done
I'll rebase my patches and also add Cortex A9 checking to cp15
registers saving code.
When it comes to our internal tree (that you posted log from)
please fix it with adding Cortex A9 check:
if (read_cpuid_part_number() == ARM_CPU_PART_CORTEX_A9) to
cp15 registers saving/restoring code.
Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung R&D Institute Poland
Samsung Electronics
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/