| Message ID | 1455877513-7680-1-git-send-email-geert+renesas@glider.be (mailing list archive) |
|---|---|
| State | RFC |
| Delegated to: | Simon Horman |
| Headers | show |
Hi Geert, On 2016-02-19 11:25:13 +0100, Geert Uytterhoeven wrote: > When doing s2ram, emev2/kzm9d crashes during system suspend with > > CPU1: smp_ops.cpu_die() returned, trying to resuscitate > Unable to handle kernel NULL pointer dereference at virtual address 00000104 > > As of commit d62242d7f63d6c87 ("ARM: shmobile: Remove partial CPU > Hotplug from EMEV2"), there's no longer an smp_ops.cpu_die() callback. > > Add standard SCU-based hotplug callbacks like on other Cortex-A9 SoCs to > fix the crash during suspend. > > Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> > --- > This fixes the crash during the suspend phase, but I have no idea if > resume actually works. I didn't manage to resume the machine after > enabling wake-up on ttyS0 (remote access only). > > Can anyone with physical access to emev2/kzm9d please check if pressing > a button resumes the system? As discussed on IRC I tested this and it fixes the crash when suspending to ram but unfortunately I can't get it to resume. / # evtest /dev/input/event0 Input driver version is 1.0.1 Input device ID: bus 0x19 vendor 0x1 product 0x1 version 0x100 Input device name: "gpio_keys" Supported events: Event type 0 (EV_SYN) Event type 1 (EV_KEY) Event code 2 (KEY_1) Event code 3 (KEY_2) Event code 4 (KEY_3) Event code 5 (KEY_4) Properties: Testing ... (interrupt to exit) Event: time 308.056274, type 1 (EV_KEY), code 4 (KEY_3), value 1 Event: time 308.056274, -------------- SYN_REPORT ------------ Event: time 311.046234, type 1 (EV_KEY), code 4 (KEY_3), value 0 Event: time 311.046234, -------------- SYN_REPORT ------------ Event: time 324.426239, type 1 (EV_KEY), code 4 (KEY_3), value 1 Event: time 324.426239, -------------- SYN_REPORT ------------ Event: time 327.306243, type 1 (EV_KEY), code 4 (KEY_3), value 0 Event: time 327.306243, -------------- SYN_REPORT ------------ ^C / # echo 0 > /sys/module/printk/parameters/console_suspend / # echo mem > /sys/power/state [ 349.980000] PM: Syncing filesystems ... done. [ 349.990000] Freezing user space processes ... (elapsed 0.005 seconds) done. [ 350.000000] Freezing remaining freezable tasks ... (elapsed 0.002 seconds) done. [ 350.010000] PM: suspend of devices complete after 1.098 msecs [ 350.020000] PM: late suspend of devices complete after 0.671 msecs [ 350.020000] PM: noirq suspend of devices complete after 0.671 msecs [ 350.030000] Disabling non-boot CPUs ... [ 350.030000] CPU1: shutdown > > Thanks! > --- > arch/arm/mach-shmobile/smp-emev2.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/arch/arm/mach-shmobile/smp-emev2.c b/arch/arm/mach-shmobile/smp-emev2.c > index 3a732199cf5e98c0..e6e8f2298a6a4786 100644 > --- a/arch/arm/mach-shmobile/smp-emev2.c > +++ b/arch/arm/mach-shmobile/smp-emev2.c > @@ -53,4 +53,9 @@ static void __init emev2_smp_prepare_cpus(unsigned int max_cpus) > const struct smp_operations emev2_smp_ops __initconst = { > .smp_prepare_cpus = emev2_smp_prepare_cpus, > .smp_boot_secondary = emev2_boot_secondary, > +#ifdef CONFIG_HOTPLUG_CPU > + .cpu_can_disable = shmobile_smp_cpu_can_disable, > + .cpu_die = shmobile_smp_scu_cpu_die, > + .cpu_kill = shmobile_smp_scu_cpu_kill, > +#endif > }; > -- > 1.9.1 > -- Regards, Niklas Söderlund
Hi Niklas, On Fri, Feb 19, 2016 at 2:24 PM, Niklas Söderlund <niklas.soderlund@ragnatech.se> wrote: > On 2016-02-19 11:25:13 +0100, Geert Uytterhoeven wrote: >> When doing s2ram, emev2/kzm9d crashes during system suspend with >> >> CPU1: smp_ops.cpu_die() returned, trying to resuscitate >> Unable to handle kernel NULL pointer dereference at virtual address 00000104 >> >> As of commit d62242d7f63d6c87 ("ARM: shmobile: Remove partial CPU >> Hotplug from EMEV2"), there's no longer an smp_ops.cpu_die() callback. >> >> Add standard SCU-based hotplug callbacks like on other Cortex-A9 SoCs to >> fix the crash during suspend. >> >> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> >> --- >> This fixes the crash during the suspend phase, but I have no idea if >> resume actually works. I didn't manage to resume the machine after >> enabling wake-up on ttyS0 (remote access only). >> >> Can anyone with physical access to emev2/kzm9d please check if pressing >> a button resumes the system? > > As discussed on IRC I tested this and it fixes the crash when suspending > to ram but unfortunately I can't get it to resume. Thanks a lot for testing! Let's drop this patch for now. Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
diff --git a/arch/arm/mach-shmobile/smp-emev2.c b/arch/arm/mach-shmobile/smp-emev2.c index 3a732199cf5e98c0..e6e8f2298a6a4786 100644 --- a/arch/arm/mach-shmobile/smp-emev2.c +++ b/arch/arm/mach-shmobile/smp-emev2.c @@ -53,4 +53,9 @@ static void __init emev2_smp_prepare_cpus(unsigned int max_cpus) const struct smp_operations emev2_smp_ops __initconst = { .smp_prepare_cpus = emev2_smp_prepare_cpus, .smp_boot_secondary = emev2_boot_secondary, +#ifdef CONFIG_HOTPLUG_CPU + .cpu_can_disable = shmobile_smp_cpu_can_disable, + .cpu_die = shmobile_smp_scu_cpu_die, + .cpu_kill = shmobile_smp_scu_cpu_kill, +#endif };
When doing s2ram, emev2/kzm9d crashes during system suspend with CPU1: smp_ops.cpu_die() returned, trying to resuscitate Unable to handle kernel NULL pointer dereference at virtual address 00000104 As of commit d62242d7f63d6c87 ("ARM: shmobile: Remove partial CPU Hotplug from EMEV2"), there's no longer an smp_ops.cpu_die() callback. Add standard SCU-based hotplug callbacks like on other Cortex-A9 SoCs to fix the crash during suspend. Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> --- This fixes the crash during the suspend phase, but I have no idea if resume actually works. I didn't manage to resume the machine after enabling wake-up on ttyS0 (remote access only). Can anyone with physical access to emev2/kzm9d please check if pressing a button resumes the system? Thanks! --- arch/arm/mach-shmobile/smp-emev2.c | 5 +++++ 1 file changed, 5 insertions(+)