* pxa168 and LINUX_REBOOT_CMD_POWER_OFF
@ 2010-08-22 9:21 Jaya Kumar
2010-08-22 9:52 ` Marek Vasut
2010-08-22 10:43 ` Eric Miao
0 siblings, 2 replies; 5+ messages in thread
From: Jaya Kumar @ 2010-08-22 9:21 UTC (permalink / raw)
To: linux-arm-kernel
Hi Arm friends,
I have a quick question about pxa168 and implementing support for
power off. I am playing with a pxa168 board (I think it might be based
off Aspenite) that appears to have no power off support, ie: I see all
the normal shutdown messages but nothing happens, even the console
stays active. I see userspace send:
sys_reboot (... LINUX_REBOOT_CMD_POWER_OFF )
to which the kernel responds with:
[ 87.248429] System halted.
which is coming from:
316void kernel_halt(void)
317{
318 kernel_shutdown_prepare(SYSTEM_HALT);
319 sysdev_shutdown();
320 printk(KERN_EMERG "System halted.\n");
321 machine_halt();
322}
I noticed that we have:
366 /* Instead of trying to make the power_off code look like
367 * halt when pm_power_off is not set do it the easy way.
368 */
369 if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off)
370 cmd = LINUX_REBOOT_CMD_HALT;
371
which suggests that I'm missing pm_power_off which is why the command
got demoted to a halt. I suppose the correct path it should take if we
really want it to power off is:
392 case LINUX_REBOOT_CMD_POWER_OFF:
393 kernel_power_off();
394 unlock_kernel();
395 do_exit(0);
396 break;
and looking in arch/arm/mach-mmp/ , I see that other board drivers
implement pm_power_off handlers that use an EC to shutdown. But I
don't think there is an EC on this board. There's a MAX8660 PMIC but
I'm not sure that should be involved in shutdown. Coming back to
Aspenite, I tried to figure out how it achieves shutdown and I don't
see it having a pm_power_off. I don't have an aspenite board to
compare with but I was wondering if shutdown works on it and if so,
some pointers to code that is involved would help me figure out how to
make shutdown work on this board. I'd appreciate any advice.
Thanks,
jaya
^ permalink raw reply [flat|nested] 5+ messages in thread* pxa168 and LINUX_REBOOT_CMD_POWER_OFF 2010-08-22 9:21 pxa168 and LINUX_REBOOT_CMD_POWER_OFF Jaya Kumar @ 2010-08-22 9:52 ` Marek Vasut 2010-08-22 10:43 ` Eric Miao 1 sibling, 0 replies; 5+ messages in thread From: Marek Vasut @ 2010-08-22 9:52 UTC (permalink / raw) To: linux-arm-kernel Dne Ne 22. srpna 2010 11:21:31 Jaya Kumar napsal(a): > Hi Arm friends, > > I have a quick question about pxa168 and implementing support for > power off. I am playing with a pxa168 board (I think it might be based > off Aspenite) that appears to have no power off support, ie: I see all > the normal shutdown messages but nothing happens, even the console > stays active. I see userspace send: > > sys_reboot (... LINUX_REBOOT_CMD_POWER_OFF ) > > to which the kernel responds with: > [ 87.248429] System halted. > > which is coming from: > > 316void kernel_halt(void) > 317{ > 318 kernel_shutdown_prepare(SYSTEM_HALT); > 319 sysdev_shutdown(); > 320 printk(KERN_EMERG "System halted.\n"); > 321 machine_halt(); > 322} > > I noticed that we have: > 366 /* Instead of trying to make the power_off code look like > 367 * halt when pm_power_off is not set do it the easy way. > 368 */ > 369 if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off) > 370 cmd = LINUX_REBOOT_CMD_HALT; > 371 > > which suggests that I'm missing pm_power_off which is why the command > got demoted to a halt. I suppose the correct path it should take if we > really want it to power off is: > > 392 case LINUX_REBOOT_CMD_POWER_OFF: > 393 kernel_power_off(); > 394 unlock_kernel(); > 395 do_exit(0); > 396 break; > > and looking in arch/arm/mach-mmp/ , I see that other board drivers > implement pm_power_off handlers that use an EC to shutdown. But I > don't think there is an EC on this board. There's a MAX8660 PMIC but > I'm not sure that should be involved in shutdown. Coming back to > Aspenite, I tried to figure out how it achieves shutdown and I don't > see it having a pm_power_off. I don't have an aspenite board to > compare with but I was wondering if shutdown works on it and if so, > some pointers to code that is involved would help me figure out how to > make shutdown work on this board. I'd appreciate any advice. There's no "poweroff" support on ARM I believe. The kernel just hangs at certain point and that's all (and this can be overridden on some platforms). Cheers > > Thanks, > jaya > > _______________________________________________ > linux-arm-kernel mailing list > linux-arm-kernel at lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel ^ permalink raw reply [flat|nested] 5+ messages in thread
* pxa168 and LINUX_REBOOT_CMD_POWER_OFF 2010-08-22 9:21 pxa168 and LINUX_REBOOT_CMD_POWER_OFF Jaya Kumar 2010-08-22 9:52 ` Marek Vasut @ 2010-08-22 10:43 ` Eric Miao 2010-08-22 15:01 ` Haojian Zhuang 1 sibling, 1 reply; 5+ messages in thread From: Eric Miao @ 2010-08-22 10:43 UTC (permalink / raw) To: linux-arm-kernel On Sun, Aug 22, 2010 at 5:21 PM, Jaya Kumar <jayakumar.lkml@gmail.com> wrote: > Hi Arm friends, > > I have a quick question about pxa168 and implementing support for > power off. I am playing with a pxa168 board (I think it might be based > off Aspenite) that appears to have no power off support, ie: I see all > the normal shutdown messages but nothing happens, even the console > stays active. I see userspace send: > > sys_reboot (... LINUX_REBOOT_CMD_POWER_OFF ) > > to which the kernel responds with: > [ ? 87.248429] System halted. > > which is coming from: > > ?316void kernel_halt(void) > ?317{ > ?318 ? ? ? ?kernel_shutdown_prepare(SYSTEM_HALT); > ?319 ? ? ? ?sysdev_shutdown(); > ?320 ? ? ? ?printk(KERN_EMERG "System halted.\n"); > ?321 ? ? ? ?machine_halt(); > ?322} > > I noticed that we have: > ?366 ? ? ? ?/* Instead of trying to make the power_off code look like > ?367 ? ? ? ? * halt when pm_power_off is not set do it the easy way. > ?368 ? ? ? ? */ > ?369 ? ? ? ?if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off) > ?370 ? ? ? ? ? ? ? ?cmd = LINUX_REBOOT_CMD_HALT; > ?371 > > which suggests that I'm missing pm_power_off which is why the command > got demoted to a halt. I suppose the correct path it should take if we > really want it to power off is: > > ?392 ? ? ? ?case LINUX_REBOOT_CMD_POWER_OFF: > ?393 ? ? ? ? ? ? ? ?kernel_power_off(); > ?394 ? ? ? ? ? ? ? ?unlock_kernel(); > ?395 ? ? ? ? ? ? ? ?do_exit(0); > ?396 ? ? ? ? ? ? ? ?break; > > and looking in arch/arm/mach-mmp/ , I see that other board drivers > implement pm_power_off handlers that use an EC to shutdown. But I > don't think there is an EC on this board. There's a MAX8660 PMIC but > I'm not sure that should be involved in shutdown. Coming back to > Aspenite, I tried to figure out how it achieves shutdown and I don't > see it having a pm_power_off. I don't have an aspenite board to > compare with but I was wondering if shutdown works on it and if so, > some pointers to code that is involved would help me figure out how to > make shutdown work on this board. I'd appreciate any advice. Yes, normally have to use an external EC for shutdown. And I believe it isn't implemented in the version of aspenite.c in mainline, probably a good idea to check Marvell's latest release on their extranet. And I believe power off normally means deep-sleep for PXA processor without an EC. Haojian, Can we help check this? > > Thanks, > jaya > > _______________________________________________ > linux-arm-kernel mailing list > linux-arm-kernel at lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel > ^ permalink raw reply [flat|nested] 5+ messages in thread
* pxa168 and LINUX_REBOOT_CMD_POWER_OFF 2010-08-22 10:43 ` Eric Miao @ 2010-08-22 15:01 ` Haojian Zhuang 2010-08-29 5:08 ` Jaya Kumar 0 siblings, 1 reply; 5+ messages in thread From: Haojian Zhuang @ 2010-08-22 15:01 UTC (permalink / raw) To: linux-arm-kernel On Sun, Aug 22, 2010 at 6:43 PM, Eric Miao <eric.y.miao@gmail.com> wrote: > On Sun, Aug 22, 2010 at 5:21 PM, Jaya Kumar <jayakumar.lkml@gmail.com> wrote: >> Hi Arm friends, >> >> I have a quick question about pxa168 and implementing support for >> power off. I am playing with a pxa168 board (I think it might be based >> off Aspenite) that appears to have no power off support, ie: I see all >> the normal shutdown messages but nothing happens, even the console >> stays active. I see userspace send: >> >> sys_reboot (... LINUX_REBOOT_CMD_POWER_OFF ) >> >> to which the kernel responds with: >> [ ? 87.248429] System halted. >> >> which is coming from: >> >> ?316void kernel_halt(void) >> ?317{ >> ?318 ? ? ? ?kernel_shutdown_prepare(SYSTEM_HALT); >> ?319 ? ? ? ?sysdev_shutdown(); >> ?320 ? ? ? ?printk(KERN_EMERG "System halted.\n"); >> ?321 ? ? ? ?machine_halt(); >> ?322} >> >> I noticed that we have: >> ?366 ? ? ? ?/* Instead of trying to make the power_off code look like >> ?367 ? ? ? ? * halt when pm_power_off is not set do it the easy way. >> ?368 ? ? ? ? */ >> ?369 ? ? ? ?if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off) >> ?370 ? ? ? ? ? ? ? ?cmd = LINUX_REBOOT_CMD_HALT; >> ?371 >> >> which suggests that I'm missing pm_power_off which is why the command >> got demoted to a halt. I suppose the correct path it should take if we >> really want it to power off is: >> >> ?392 ? ? ? ?case LINUX_REBOOT_CMD_POWER_OFF: >> ?393 ? ? ? ? ? ? ? ?kernel_power_off(); >> ?394 ? ? ? ? ? ? ? ?unlock_kernel(); >> ?395 ? ? ? ? ? ? ? ?do_exit(0); >> ?396 ? ? ? ? ? ? ? ?break; >> >> and looking in arch/arm/mach-mmp/ , I see that other board drivers >> implement pm_power_off handlers that use an EC to shutdown. But I >> don't think there is an EC on this board. There's a MAX8660 PMIC but >> I'm not sure that should be involved in shutdown. Coming back to >> Aspenite, I tried to figure out how it achieves shutdown and I don't >> see it having a pm_power_off. I don't have an aspenite board to >> compare with but I was wondering if shutdown works on it and if so, >> some pointers to code that is involved would help me figure out how to >> make shutdown work on this board. I'd appreciate any advice. > > Yes, normally have to use an external EC for shutdown. And I believe > it isn't implemented in the version of aspenite.c in mainline, probably a > good idea to check Marvell's latest release on their extranet. > > And I believe power off normally means deep-sleep for PXA processor > without an EC. > > Haojian, > > Can we help check this? > Before talking about power off state, I need to clarify some background. Max8660 is a PMIC compatible with PXA3xx series. And there's no EC to control power on Aspenite board. There's all eight power domains can be controlled by max8660. V1~V5 is controlled by SYS_EN signal. V6 & V7 is controlled by I2C bus. V8 is VCC_BBATT domain that should be always ON. So we can't shutdown all power on Aspenite since V8 shoudl be always ON. So the key is SYS_EN. In Aspenite board, one signal of GPIO expander is named EXP2_SYS_DIS_N. If we put this signal to low, the SYS_EN signal will do a negative transition. In marvell's reference code, we didn't implement power off mode in Aspenite board. We implemented system hibernation mode that is used to save information to DDR and shutdown most of power of board. If you like, you can control EXP2_SYS_DIS_N signal to implement your power off mode. One thing should be noticed, there's still a little power consumption on Aspenite board even this signal is controlled to low. So there's no purely power off mode in Aspenite board. Thanks Haojian ^ permalink raw reply [flat|nested] 5+ messages in thread
* pxa168 and LINUX_REBOOT_CMD_POWER_OFF 2010-08-22 15:01 ` Haojian Zhuang @ 2010-08-29 5:08 ` Jaya Kumar 0 siblings, 0 replies; 5+ messages in thread From: Jaya Kumar @ 2010-08-29 5:08 UTC (permalink / raw) To: linux-arm-kernel On Sun, Aug 22, 2010 at 11:01 PM, Haojian Zhuang <haojian.zhuang@gmail.com> wrote: > > There's all eight power domains can be controlled by max8660. V1~V5 is > controlled by SYS_EN signal. V6 & V7 is controlled by I2C bus. V8 is > VCC_BBATT domain that should be always ON. So we can't shutdown all > power on Aspenite since V8 shoudl be always ON. > > So the key is SYS_EN. In Aspenite board, one signal of GPIO expander > is named EXP2_SYS_DIS_N. If we put this signal to low, the SYS_EN > signal will do a negative transition. Thanks Haojian, Eric, Marek. That was useful info. Regards, jaya ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2010-08-29 5:08 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2010-08-22 9:21 pxa168 and LINUX_REBOOT_CMD_POWER_OFF Jaya Kumar 2010-08-22 9:52 ` Marek Vasut 2010-08-22 10:43 ` Eric Miao 2010-08-22 15:01 ` Haojian Zhuang 2010-08-29 5:08 ` Jaya Kumar
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.