PXA not wakeup after suspen..

Hi,

Currently i am working on suspend and resume.
Basically, i am using the mainstoneii BSP from intel and do some changes to
fit to my platform...

The suspend part working fine... however, not able to resume(even processor
wakeup)... However the GPIO reset is working fine... Do anyone face this
problem? Any ideas...

Best regards,
Thomas

--
Message posted via PocketPCJunkies.com
http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1

randika
Fri Apr 11 01:46:00 PDT 2008


Hi James,

Have you configured the registers related to wakeup sources correctely?

check on PWER, PRER, PFER regs.
check whether your switch is working correctly as well.

regards,
rand

"JamesCool via PocketPCJunkies.com" wrote:

> Hi,
>
> Currently i am working on suspend and resume.
> Basically, i am using the mainstoneii BSP from intel and do some changes to
> fit to my platform...
>
> The suspend part working fine... however, not able to resume(even processor
> wakeup)... However the GPIO reset is working fine... Do anyone face this
> problem? Any ideas...
>
> Best regards,
> Thomas
>
> --
> Message posted via PocketPCJunkies.com
> http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1
>
>

Bruce
Fri Apr 11 05:37:42 PDT 2008

How do you know that suspend works fine? One indication that suspend is not
working is that the CPU won't wake back up, so it may not be working fine.

--
Bruce Eitman (eMVP)
Senior Engineer
beitman AT applieddata DOT net

Applied Data Systems
www.applieddata.net
An ISO 9001:2000 Registered Company
Microsoft WEP Gold-level Member

"JamesCool via PocketPCJunkies.com" <u42351@uwe> wrote in message
news:82825eb6ea42b@uwe...
> Hi,
>
> Currently i am working on suspend and resume.
> Basically, i am using the mainstoneii BSP from intel and do some changes
> to
> fit to my platform...
>
> The suspend part working fine... however, not able to resume(even
> processor
> wakeup)... However the GPIO reset is working fine... Do anyone face this
> problem? Any ideas...
>
> Best regards,
> Thomas
>
> --
> Message posted via PocketPCJunkies.com
> http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1
>

StefanoBianconi
Fri Apr 11 08:18:01 PDT 2008

Hi Thomas,
I'm facing the same problem, I resolved this issue with the Errata 50 on
PXA270 Specification Updates: The processor does not exit from sleep/deepsleep
mode. Entering in sleep mode at 91MHz, solve the problem.

regards

Stefano

"randika" wrote:

>
> Hi James,
>
> Have you configured the registers related to wakeup sources correctely?
>
> check on PWER, PRER, PFER regs.
> check whether your switch is working correctly as well.
>
> regards,
> rand
>
> "JamesCool via PocketPCJunkies.com" wrote:
>
> > Hi,
> >
> > Currently i am working on suspend and resume.
> > Basically, i am using the mainstoneii BSP from intel and do some changes to
> > fit to my platform...
> >
> > The suspend part working fine... however, not able to resume(even processor
> > wakeup)... However the GPIO reset is working fine... Do anyone face this
> > problem? Any ideas...
> >
> > Best regards,
> > Thomas
> >
> > --
> > Message posted via PocketPCJunkies.com
> > http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1
> >
> >

JamesCool
Fri Apr 11 17:36:45 PDT 2008

Hi Bruce,

Basically, the suspend code is directly using the Mainstoneii BSp... I assume
it work fine.. So, may i know any ways to check whether the suspend work fine?


Bruce Eitman [eMVP] wrote:
>How do you know that suspend works fine? One indication that suspend is not
>working is that the CPU won't wake back up, so it may not be working fine.
>
>> Hi,
>>
>[quoted text clipped - 10 lines]
>> Best regards,
>> Thomas

--
Message posted via PocketPCJunkies.com
http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1

JamesCool
Sat Apr 12 02:18:25 PDT 2008

Hi,

How you add the frequency change code before suspend?
How you able to get the virtual address for CCCR?

Best regards,

Stefano Bianconi wrote:
>Hi Thomas,
>I'm facing the same problem, I resolved this issue with the Errata 50 on
>PXA270 Specification Updates: The processor does not exit from sleep/deepsleep
>mode. Entering in sleep mode at 91MHz, solve the problem.
>
>regards
>
>Stefano
>
>> Hi James,
>>
>[quoted text clipped - 19 lines]
>> > Best regards,
>> > Thomas

--
Message posted via PocketPCJunkies.com
http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1

Bruce
Sat Apr 12 06:29:11 PDT 2008

It has been a long time since I have needed to check this, but RAM goes into
auto refresh and there is a way to check that by probing the pins. The
current usage will also go way down, but keep in mind as drivers power down,
so does current usage.
--
Bruce Eitman (eMVP)
Senior Engineer
beitman AT applieddata DOT net

Applied Data Systems
www.applieddata.net
An ISO 9001:2000 Registered Company
Microsoft WEP Gold-level Member

"JamesCool via PocketPCJunkies.com" <u42351@uwe> wrote in message
news:828af7ffac583@uwe...
> Hi Bruce,
>
> Basically, the suspend code is directly using the Mainstoneii BSp... I
> assume
> it work fine.. So, may i know any ways to check whether the suspend work
> fine?
>
>
> Bruce Eitman [eMVP] wrote:
>>How do you know that suspend works fine? One indication that suspend is
>>not
>>working is that the CPU won't wake back up, so it may not be working fine.
>>
>>> Hi,
>>>
>>[quoted text clipped - 10 lines]
>>> Best regards,
>>> Thomas
>
> --
> Message posted via PocketPCJunkies.com
> http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1
>

JamesCool
Tue Apr 15 22:01:44 PDT 2008

Hi,

Now the processor wakeup and resume start... However, the OS seems stop ....
Below is the debug message i get...


..............OSD_DebugCommsResume.........................
Xllp_SuspendResume: State Restored: ResumePhase2Data=xa00ff420->xa00ff420
Xllp_SuspendResume: Wake Reason: PEDR=x00000002, PKSR=x00000000,
ISCR=x00000000, IER=x00000000
Xllp_SuspendResume: Exit: (entry_CPSR=xa0000093)
Xllp_SuspendResume: Exit: RCNR=x00000001, RTAR=x00000000, RTSR=x00000000,
PICR=x00000000
Xllp_SuspendResume: Exit: RYAR1=x000fa622, RDAR1=x005a0000
Xllp_SuspendResume: Exit: OSCR0=x8251edb3, OSMR0=x8250a9fd, OIER=x00000001
Xllp_SuspendResume: Exit: OMCR4=x00000000, OMCR5=x00000000, OMCR6=x00000000,
OMCR7=x00000000
OEMPowerOff: Xllp_SuspendAndResume(x804397a0) returned x00000000
OEMPowerOff: g_oalWakeSource => SYSWAKE_POWER_BUTTON(x00000403)
0x87d1ec64: Back from OEMPowerOff
0x87d1ec64: Calling device manager power proc.
0x87d1ec64: +FYS_PowerUp
0x87d1ec64: send low battery patch
0x87d1ec64: -FYS_PowerUp
0x87d1ec64: Calling GWES power proc.
0x87d1ec64: Returning to normally scheduled programming.

The system just stop here, may i know what is goes wrong?

Best regards,
James

JamesCool wrote:
>Hi,
>
>Currently i am working on suspend and resume.
>Basically, i am using the mainstoneii BSP from intel and do some changes to
>fit to my platform...
>
>The suspend part working fine... however, not able to resume(even processor
>wakeup)... However the GPIO reset is working fine... Do anyone face this
>problem? Any ideas...
>
>Best regards,
>Thomas

--
Message posted via PocketPCJunkies.com
http://www.pocketpcjunkies.com/Uwe/Forums.aspx/wince-pb/200804/1

Michel
Tue Apr 15 23:04:54 PDT 2008

Looks like your timer tick never fires. The timer tick is the interrupt
that tickles the scheduler. If that one never fires, the scheduler never
schedules another thread. What was IER when you entered suspend?

Good luck,

Michel Verhagen, eMVP
Check out my blog: http://GuruCE.com/blog

GuruCE Ltd.
Microsoft Embedded Partner
http://GuruCE.com
Consultancy, training and development services.

JamesCool via PocketPCJunkies.com wrote:
> Hi,
>
> Now the processor wakeup and resume start... However, the OS seems stop ....
> Below is the debug message i get...
>
>
> ...............OSD_DebugCommsResume.........................
> Xllp_SuspendResume: State Restored: ResumePhase2Data=xa00ff420->xa00ff420
> Xllp_SuspendResume: Wake Reason: PEDR=x00000002, PKSR=x00000000,
> ISCR=x00000000, IER=x00000000
> Xllp_SuspendResume: Exit: (entry_CPSR=xa0000093)
> Xllp_SuspendResume: Exit: RCNR=x00000001, RTAR=x00000000, RTSR=x00000000,
> PICR=x00000000
> Xllp_SuspendResume: Exit: RYAR1=x000fa622, RDAR1=x005a0000
> Xllp_SuspendResume: Exit: OSCR0=x8251edb3, OSMR0=x8250a9fd, OIER=x00000001
> Xllp_SuspendResume: Exit: OMCR4=x00000000, OMCR5=x00000000, OMCR6=x00000000,
> OMCR7=x00000000
> OEMPowerOff: Xllp_SuspendAndResume(x804397a0) returned x00000000
> OEMPowerOff: g_oalWakeSource => SYSWAKE_POWER_BUTTON(x00000403)
> 0x87d1ec64: Back from OEMPowerOff
> 0x87d1ec64: Calling device manager power proc.
> 0x87d1ec64: +FYS_PowerUp
> 0x87d1ec64: send low battery patch
> 0x87d1ec64: -FYS_PowerUp
> 0x87d1ec64: Calling GWES power proc.
> 0x87d1ec64: Returning to normally scheduled programming.
>
> The system just stop here, may i know what is goes wrong?
>
> Best regards,
> James
>
> JamesCool wrote:
>> Hi,
>>
>> Currently i am working on suspend and resume.
>> Basically, i am using the mainstoneii BSP from intel and do some changes to
>> fit to my platform...
>>
>> The suspend part working fine... however, not able to resume(even processor
>> wakeup)... However the GPIO reset is working fine... Do anyone face this
>> problem? Any ideas...
>>
>> Best regards,
>> Thomas
>

JamesCool
Wed Apr 16 00:29:36 PDT 2008

Hi,

Please see the below debug message before entering suspend

0x87d19950: Powering Off system:
0x87d19950: Calling GWES power proc.
0x87d19950: Calling device manager power proc.
0x87d19950: +FYS_PowerDown
0x87d19950: -FYS_PowerDown
0x87d19950: Calling OEMPowerOff...
OEMPowerOff->Xllp_SuspendAndResume(x804397a0)
Xllp_SuspendResume: BSP_MAINSTONE not defined
Xllp_SuspendResume: Entry: RCNR=x00000222, RTAR=x00000000, RTSR=x00001020,
PICR=x00000000
Xllp_SuspendResume: Entry: RYAR1=x000fa622, RDAR1=x005a0000
Xllp_SuspendResume: Entry: OSCR0=x6985798e, OSMR0=x69846d4e, OIER=x00000001
Xllp_SuspendResume: Entry: OSMR0=x69846d4e, OSMR1=x00000000, OSMR2=x00000000,
OSMR3=x00000000
Xllp_SuspendResume: Entry: OSMR4=x00000000, OSMR5=x00000000, OSMR6=x00000000,
OSMR7=x00000000
Xllp_SuspendResume: Entry: OSMR8=x00000000, OSMR9=x00000000, OSMR10=x00000000,
OSMR11=x00000000
Xllp_SuspendResume: Entry: ResumePhase2Data @ xa00ff420->xa00ff420
Xllp_SuspendResume: Entry: ICPR=x04000000, ICMR=x86000900
Xllp_SuspendResume: Saving mode specific core registers
Xllp_SuspendResume: FIQ(x11) saved: x804393bc-x8043939c, (8)
Xllp_SuspendResume: IRQ(x12) saved: x8043939c-x80439390, (3)
Xllp_SuspendResume: UND(x1b) saved: x80439390-x80439384, (3)
Xllp_SuspendResume: ABT(x17) saved: x80439384-x80439378, (3)
Xllp_SuspendResume: SVC(x13) saved: x80439378-x8043936c, (3)
Xllp_SuspendResume: SYS(x1f) saved: x8043936c-x80439360, (3)
Xllp_SuspendResume: mode specific core registers saved
Xllp_SuspendResume: FIQ@x8043939c (8)
x804393b8:xfffffbff
x804393b4:xffff0900
x804393b0:xdfff97df
x804393ac:xfffcc7ff
x804393a8:xdff77eff
x804393a4:xedfffb6f
x804393a0:xfffffbfe
x8043939c:x68000056
Xllp_SuspendResume: IRQ@x80439390 (3)
x80439398:xffffc800
x80439394:xffff0800
x80439390:x6000001f
Xllp_SuspendResume: UND@x80439384 (3)
x8043938c:x8041611c
x80439388:xffffc800
x80439384:x6000001f
Xllp_SuspendResume: ABT@x80439378 (3)
x80439380:x8011a224
x8043937c:xffff1000
x80439378:x2000001f
Xllp_SuspendResume: SVC@x8043936c (3)
x80439374:x801aede8
x80439370:xffffc728
x8043936c:x6000001f
Xllp_SuspendResume: SYS@x80439360 (3)
x80439368:x80128f7c
x80439364:x0624f4e0
x80439360:x600000df
Xllp_SuspendResume: After Core Save: ResumePhase2Data=xa00ff420->xa00ff420
Xllp_SuspendResume: pMEMCRegs->MDREFR=0x2003a018
Xllp_SuspendResume: pMEMCRegs->MDREFR Cleared
Xllp_SuspendResume: pPWRRegs->PSPR =0xa00ff420
Xllp_SuspendResume: pGPIORegs->GPDR0=0xc061e800
Xllp_SuspendResume: clkMgr->oscc=0xf
Xllp_SuspendResume: pwrMgr->PCFR=0x61
Xllp_SuspendResume: pwrMgr->PSLR=0xcc500000
Xllp_SuspendResume: PWER=x00000000->x00000003, aaPRER=x00000000->x00000003,
PFER=x00000000->x00000003, PKWR=x00000000->x0000000

Thanks,
James

Michel Verhagen (eMVP) wrote:
>Looks like your timer tick never fires. The timer tick is the interrupt
>that tickles the scheduler. If that one never fires, the scheduler never
>schedules another thread. What was IER when you entered suspend?
>
>Good luck,
>
>Michel Verhagen, eMVP
>Check out my blog: http://GuruCE.com/blog
>
> GuruCE Ltd.
> Microsoft Embedded Partner
> http://GuruCE.com
> Consultancy, training and development services.
>
>> Hi,
>>
>[quoted text clipped - 39 lines]
>>> Best regards,
>>> Thomas

--
Message posted via http://www.pocketpcjunkies.com

JamesCool
Wed Apr 16 01:17:59 PDT 2008

HI Micheal,

I had add a debug message at oeminterrupthandler(0 funtion for IRQ_OSMR0...
I can see, after resumming, there is this OSMR0 interrupt being fired all the
way like before entering sleep...

Is that possible the os scheduler lose track after resuming?

Best regards,
James
Michel Verhagen (eMVP) wrote:
>Looks like your timer tick never fires. The timer tick is the interrupt
>that tickles the scheduler. If that one never fires, the scheduler never
>schedules another thread. What was IER when you entered suspend?
>
>Good luck,
>
>Michel Verhagen, eMVP
>Check out my blog: http://GuruCE.com/blog
>
> GuruCE Ltd.
> Microsoft Embedded Partner
> http://GuruCE.com
> Consultancy, training and development services.
>
>> Hi,
>>
>[quoted text clipped - 39 lines]
>>> Best regards,
>>> Thomas

--
Message posted via http://www.pocketpcjunkies.com