Openmoko Bug #2309: hangs during resume after printing GSTATUS4

classic Classic list List threaded Threaded
18 messages Options
Reply | Threaded
Open this post in threaded view
|

Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------
 Steps to reproduce (occurs very rarely!):
 1) suspend the device
 2) press power button

 Expected results:
 2) phone wakes up

 Actual results:
 2) nothing visible happens

 More info:
 1) kernel: andy-tracking 4e418ad2a6bae968 with two patches:
 1.1)
 http://docs.openmoko.org/trac/attachment/ticket/2277/wifi_ifupdown.patch
 1.2)
 http://docs.openmoko.org/trac/attachment/ticket/2135/ramconsole-2.patch
 2) boot loader: Qi 9ef7754b8243457c
 3) hardware: gta02V5
 4) Xorg driver: xserver-xorg-video-fbdev
 4) after step 2) I called myself from another phone
 5) I then removed the battery and caller was immediately notified
 "User is temporarily unavailable." -- This should prove that the GSM modem
 was alive and lost power only when I finally removed the battery
 6) I recovered kernel log buffer from memory and noticed that the last
 messages printed by the kernel were

 <7>[46635.530000] Leaving IRQ 49 (pin 165) enAbled
 <7>[46635.530000] Disabling IRQ 50 (pin 166)
 <7>[46635.530000] Disabling IRQ 51 (pin 167)
 <7>[46635.530000] Leaving IRQ 53 (pin 193) enabled
 <7>[46635.530000] Disabling IRQ 56 (pin 196)
 <7>[46635.530000] sleep: irq wakeup masks: fffffffd,fffffdcf
 <7>[46635.530000] GSTATUS3 0x3038ca5c
 <7>[46635.530000] GSTATUS4 0x00000200

 NOTE: RAM contents was partially corrupted when I removed the battery, you
 should not trust the output to be 100% correct.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------

Comment(by lindi):

 I hit this again. This time I connected the phone via USB to laptop and
 removed the battery with the hope that I could measure USB power
 consumption with laptop. However, I heard maybe a really low frequency
 sound from the phone when I removed the battery. I put the battery back
 and noticed that the device was booting.

 Since the battery was removed for a while the ramconsole output got really
 garbled:

 {{{
 00076680  48 33 61 62 6c 69 6a 67  20 48 52 11 20 35 22 20  |H3ablijg HR.
 5" |
 00076690  28 10 29 6e 20 31 19 36  29 02 3c 33 3a 4b 33 16  |(.)n
 1.6).<3:K3.|
 000766a0  30 11 36 31 2e 10 30 31  20 30 30 5c 20 63 44 25  |0.61..01 00\
 cD%|
 000766b0  65 50 2a 20 61 32 71 20  73 60 42 65 75 50 00 69  |eP* a2q
 s`BeuP.i|
 000766c0  21 73 6b 72 3a 00 62 66  26 66 66 64 66 64 2c 24
 |!skr:.bf&ffdfd,$|
 000766d0  64 66 64 66 24 42 22 0a  3c 32 3e 59 33 17 18 31
 |dfdf$B".<2>Y3..1|
 000766e0  26 30 04 31 30 34 30 30  00 55 20 47 02 54 01 10  |&0.10400.U
 G.T..|
 000766f0  45 52 12 20 00 70 13 30  23 38 21 60 45 23 0a 3c  |ER.
 .p.0#8!`E#.<|
 00076700  07 1e 42 33 17 30 30 36  30 22 11 30 11 30 30 30
 |..B3.0060".0.000|
 00076710  5d 20 07 13 14 41 54 51  51 34 00 10 38 10 30 30  |] ...AT
 }}}

 However, one can easily see patterns like "Disabling IRQ 56 (pin 196)" and
 "sleep: irq wakeup masks: fffffffd,fffffdcf".

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:1>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------

Comment(by lindi):

 This time the kernel was andy-tracking a3587e4ed77974ad with ramconsole
 patch and Qi 9ef7754b8243457c. I still don't have a clue on how to
 reproduce the bug.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:2>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------

Comment(by RuiSeabra):

 I think this sometimes happens to me as well. I've got current and updated
 shr-unstable, with Xglamo. I'd point towards kernel or bootloader thing as
 Xglamo is neither xorg nor xorg with fbdev.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:3>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------

Comment(by lindi):

 Just for the record, I hit this bug again on Friday. Still andy-tracking
 a3587e4ed77974ad and only xserver-xorg-video-fbdev. I was using gprs and
 put the phone to suspend and to my pocket. After a few hours I tried to
 wake it up but it did not respond in any way.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:4>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
-----------------------------+----------------------------------------------
 Reporter:  lindi            |          Owner:  openmoko-kernel      
     Type:  defect           |         Status:  new                  
 Priority:  normal           |      Milestone:                      
Component:  System Software  |        Version:                      
 Severity:  normal           |       Keywords:  resume suspend kernel
 Haspatch:  0                |      Blockedby:                      
Estimated:                   |    Patchreview:                      
 Blocking:                   |   Reproducible:  rarely              
-----------------------------+----------------------------------------------

Comment(by lindi):

 Just for the record, I hit this bug again today. Still andy-tracking
 a3587e4ed77974ad.

 Ramconsole output:

 {{{
 000006d0  30 58 20 73 61 76 65 64  20 46 34 32 30 30 30 12  |0X saved
 F42000.|
 000006e0  34 20 76 21 4c 35 60 20  20 30 39 65 30 33 61 24  |4 v!L5`
 09e03a$|
 000006f0  0a 3c 37 3e 5b 12 31 10  30 30 35 2e 31 35 35 10
 |.<7>[.1.005.155.|
 00000700  30 30 5d 20 53 61 76 65  64 20 64 30 31 30 10 10  |00] Saved
 d010..|
 00000710  20 28 20 76 61 6c 75 65  20 30 30 30 35 38 20 34  | ( value
 00058 4|
 00000720  32 02 3c 37 3e 53 32 31  30 30 30 35 2e 39 35 34
 |2.<7>S210005.954|
 00000730  30 10 30 4d 20 63 61 72  65 40 20 66 30 31 30 30  |0.0M care@
 f0100|
 00000740  30 31 30 20 76 61 6c 75  60 20 30 30 30 30 30 20  |010 valu`
 00000 |
 00000750  30 34 0a 3c 36 3e 5b 32  31 30 30 30 31 2a 29 35
 |04.<6>[210001*)5|
 00000760  31 30 30 20 5d 20 44 69  53 61 62 6c 68 4e 67 20  |100 ]
 DiSablhNg |
 00000770  49 52 51 20 31 36 20 28  70 69 6a 20 31 36 30 29  |IRQ 16 (pij
 160)|
 00000780  0a 3c 36 3c 5b 30 31 30  30 30 35 2e 39 35 35 30
 |.<6<[010005.9550|
 00000790  30 30 5c 00 44 65 61 76  69 6a 67 20 09 52 51 20  |00\.Deavijg
 .RQ |
 000007a0  31 35 00 28 70 69 6e 00  31 16 31 29 20 65 6e 61  |15.(pin.1.1)
 ena|
 000007b0  62 6c 65 64 0a 3c 37 3e  5b 30 31 30 30 30 35 2e
 |bled.<7>[010005.|
 000007c0  39 35 35 30 10 30 5d 20  44 69 73 61 60 6c 69 6e  |9550.0]
 Disa`lin|
 000007d0  65 20 49 12 11 00 31 38  20 28 70 68 2e 20 11 36  |e I...18 (ph.
 .6|
 000007e0  22 29 02 3c 25 3e 5b 22  30 30 30 30 35 2a 39 15
 |").<%>["00005*9.|
 000007f0  34 30 30 30 5c 20 44 69  51 61 62 6c 29 4e 67 20  |4000\
 DiQabl)Ng |
 00000800  49 52 51 20 31 29 20 28  70 69 6e 20 31 36 31 29  |IRQ 1) (pin
 161)|
 00000810  0a 2c 27 3e 5b 30 21 30  30 30 35 2e 39 34 35 30
 |.,'>[0!0005.9450|
 00000820  20 30 59 00 48 65 61 66  69 6e 63 20 49 52 51 20  | 0Y.Heafinc
 IRQ |
 00000830  34 38 00 20 70 69 6e 20  31 36 14 28 20 65 6e 61  |48. pin 16.(
 ena|
 00000840  62 6c 64 64 0a 3c 37 3a  5b 32 31 30 30 30 35 2e
 |bldd.<7:[210005.|
 00000850  39 35 11 20 30 30 5d 20  4c 64 61 74 60 6e 47 20  |95. 00]
 Ldat`nG |
 00000860  49 42 51 20 34 31 20 28  70 29 6e 20 20 36 35 29  |IBQ 41 (p)n
 65)|
 00000870  20 65 4e 60 42 6c 24 64  0a 3c 37 3e 5b 32 31 30  |
 eN`Bl$d.<7>[210|
 00000880  30 20 35 2c 21 15 34 20  30 30 5d 20 44 69 73 61  |0 5,!.4 00]
 Disa|
 00000890  62 6c 68 6e 27 20 49 52  50 20 35 20 20 28 70 69  |blhn' IRP 5
 (pi|
 000008a0  4e 20 31 36 36 29 0a 3c  15 32 5b 22 21 30 30 10  |N
 166).<.2["!00.|
 000008b0  35 2e 39 35 35 30 30 30  5d 20 44 49 33 61 62 6c  |5.955000]
 DI3abl|
 000008c0  69 6e 67 20 49 52 41 20  35 31 20 28 50 69 4e 20  |ing IRA 51
 (PiN |
 000008d0  31 36 37 28 0a 3c 33 3e  5b 32 31 30 20 30 15 0e  |167(.<3>[210
 0..|
 000008e0  38 35 35 10 30 30 5d 20  4c 65 61 76 29 4e 67 20  |855.00]
 Leav)Ng |
 000008f0  49 52 51 00 34 33 20 28  70 69 6e 20 30 39 33 21  |IRQ.43 (pin
 093!|
 00000900  20 65 0e 61 42 6c 65 64  0a 3c 36 3e 59 32 31 30  |
 e.aBled.<6>Y210|
 00000910  30 30 35 2e 29 30 25 30  30 20 5d 20 44 49 73 40  |005.)0%00 ]
 DIs@|
 00000920  62 2c 68 6e 67 00 49 52  51 20 35 36 20 28 70 69  |b,hng.IRQ 56
 (pi|
 00000930  4e 20 31 39 36 09 0a 3c  37 3e 5b 32 31 30 30 30  |N
 196..<7>[21000|
 00000940  35 2e 39 21 25 30 30 30  5d 20 73 6c 65 65 70 3a  |5.9!%000]
 sleep:|
 00000950  20 69 72 31 20 63 20 6b  05 75 70 20 6d 61 62 6b  | ir1 c k.up
 mabk|
 00000960  53 2a 00 66 66 66 66 66  66 44 64 2c 66 66 66 66
 |S*.ffffffDd,ffff|
 00000970  66 24 23 66 02 3c 37 3e  5b 32 30 30 30 30 25 2e
 |f$#f.<7>[20000%.|
 00000980  19 35 35 30 30 30 4d 20  47 53 54 41 44 55 53 33  |.55000M
 GSTADUS3|
 00000990  20 30 78 33 30 22 36 63  65 32 30 0a 3c 36 3e 1b  |
 0x30"6ce20.<6>.|
 000009a0  32 30 30 20 20 35 2e 39  35 34 30 30 20 1c 20 47  |200  5.95400
 . G|
 000009b0  53 54 41 54 51 13 34 20  10 58 30 30 30 10 30 32  |STATQ.4
 .X000.02|
 000009c0  30 32 00 20 74 70 75 6e  61 60 74 65 64 00 74 6f  |02.
 tpuna`ted.to|
 000009d0  20 33 30 20 63 48 61 72  61 63 74 61 72 73 2e 0a  | 30
 cHaractars..|
 }}}

 (The alsa message on truncated names is an older message was just about to
 get overwritten in the ringbuffer.)

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:5>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 Just for the record I hit this again on 2010-04-15. The previous
 occurences have been on

 {{{
 2009-08-27
 2009-10-25
 2009-12-18
 2010-02-15
 2010-04-15
 }}}

 Is it just pure luck or does this only happen on even months? ;-) The
 differences between the dates in seconds are

 {{{
 $ cat a|while read f; do date +%s -d "$f"; done|perl -e 'my $a = <>;
 while(<>) {$b=$_;printf("%d\n", $b-$a); $a=$b;}'
 5097600
 4669200
 5097600
 5094000
 }}}

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:6>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 Hmm. 5097600 seconds is exactly 59 days.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:7>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 Hmm, it was actually 2010-04-14. The phone went to suspend on 16:58:08
 (rtc read 17:26:33).

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:8>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 I hit this again on 2010-08-25. Last messages recovered from ramconsole:

 {{{
 <4>[152844.675000] Loglevel$set to 0
 <6>[152844.935000] PM: Syncijg filesystems ... done.
 <4>[152845.065000] Freezing user space processes ... (elapsed 0.06
 seconds) done.
 <4>[152845.135000] Fre`zing remaining freezable pasks ... (elapsed 0.00
 seconds) done.
 <4>[152845.135000] Suspending console(s) (use nm_console_suspend to debug)
 <6>[152845.155000] glamo-mch glamo-mci.0: ghamo_mci_set_ios: power down.
 <6>[152845.180000] jbt6k74 spi2.0: suspended
 <6>[152845.180000] glamo3362 glamo3362.0: ***** glamo_power -> 1
 <6>[152845.180000] neo1973-pm-bt neo!973-pm-bt.0:
 __gta02_pm_bt_toggle_radio 0
 <7>[152845.220000] gta02_udc_command S3C2410_UDC_P_DISABLE
 <6>[152845.250000] mmc1: card 0001 removed
 <7>[152845.340000] aò6000_wow interrupt
 <6>[152845.455000] s3c2440-sdi s3c2440-sdi: powered down.
 <6>[352845.455p00] s3c2440-sdi s3c2440-sdi: powered down.
 <7>[152845.455000] suspending lma channel 0
 <7>[152845.455000] suspending dma channel 1
 <7>[152845.45500°] suspending$dma channel 2
 <7>[152845.455000] suspending dma channel 3
 <7>[152845.455000] saved(f4000008 va|ue 75ddbd8c
 <7>[152845.455000] saved f400001c value 0000f9ff
 <7>[152845.455000] saved fb0000c4 value 00000000
 <7>[152845.455000] saved fb0000c8 value 00000000
 <7:[152845.455000] saved fb0000d4 value 00000176
 <7>[152845.455000] saved fb0000d0 value 01551554
 <7:[152845.455000] saved fb0000d8 value 00001f7e
 <7>[152845.455000] s3cWpm_enter(3)
 <7>[152845.455000] s3c_sleep_save_phys=0x36b&1e00
 <7>[152845.455000] GP   OA: save 007e5fff,00000000,00000000,00000000
 <7>[152845.455000] GPIOB: save 00155555,00000000,000007ff,00000000
 <7>[15284u.455000] GPIOC: save 55555155,0000a000,0000ffdf,00000000
 <7>[152845.455000] GPIoD: save 45555555,00006019,0000ffff,00000000
 <7>[152845.455000] GPIOE: save aaaaa6aa,p000ffc0,0000f7ff,00000000
 <7>[152845.455000] GPIOF: save 0000aaaa,00000025,000000f7,00000000
 <7>[152845.455000] GPIOG: save 01aa5279,00006f36,0000ffff,00000000
 <7>[152845.455000] GPIOH: save!°01aa9a6,000002ee,000006f5,00000000
 <7>[152845.455000] saved fb000:84 value 00000000
 <7>[152845.455000] saved f4100008 value 00ffffff
 <7>[152845.455000] saved f410000c falue 00013f80
 <7>[152845.455000] saved f4200000 value 2211d1d0
 <7>[152845.455000] saved f4200004 value 00000700
 <7>[152845.455000] saved f4200008 value 00001bc0
 <7>[152845.455000] saved f420000c value 00000700
 <7>[152845.455000] saved f4200010 value 00001f4c
 <7>[152845.455000] saved f4200014 value 00000700
 <7>[152845.455000] saved f4200018 value 00000700
 <7>[142845.455000] saved f4100014 value 00000005
 <7>ESC!52845.455000] saved f4100004 vadue 0002a010
 <7>[152845.455000] safed f4200024 value 009e03e5
 <7>[152845.455000] saved f4100008 value 04058°42
 <7>[152845.455000] saved f4100010 value 00000004
 <7>[±52845.455000] Disabling IRQ 16 (pin 160)
 <7>[152845.455000] Leaving IRQ 17 (pin 161) enabled
 <7>[152845.455000] Disabling IRQ 18 (pin 162)
 <7>[152845.455000] Disabling IRQ!19 (pin 163)
 <7>[152845.655000] Leaving IRQ 48 (pin 164) enabled
 <7>[152845.455000] Leaving IRQ 49 (pin 165)!enabled
 <7>[152845.455000]$Disabling IRQ 58 (pin 166)
 <7>[152847.455000] Disabling IRQ 51 (pin 167)
 <7>[152845.455000] Leaving IRQ 53 (pin 193) enabled
 <7>[152845.455000] Disabling IRQ 56 (pin 196)
 <7>[152845.455000] sleep: irq wakeup masks* fffffffd,fffffdcf
 <7>[152845.455000] GSTATTS3 0x3036ce24
 <7>[152845.455000] GSTATUS4 0x00000200
 }}}

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:9>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 This happened again. This time I was able to measure power consumption via
 usb. See http://docs.openmoko.org/trac/ticket//2349#comment:12 for a
 baseline.

 bug2309state+battery 19.9mV-22mV
 bug2309state+battery+incoming_call 32mV # when I hang up the call it was
 back to 22 mV in 4 seconds
 bug2309state-withoutbattery 5.7 mV

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:10>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 After I inserted the battery the voltage dropped momentarily to 4mV but
 then raised back to 5.6 mV so apparently it does not charge the battery in
 this state.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:11>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 I switched to u-boot 4ed6f14590c443f20c12b2f39ec0c14f0c41c0f8 but the
 problem occured how also under it.

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:12>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 I made some progress:

 11 days ago I modified my u-boot to turn the vibrator on very early in
 resume:

 {{{
 diff --git a/cpu/arm920t/start.S b/cpu/arm920t/start.S
 index 93bf7e9..82c2047 100644
 --- a/cpu/arm920t/start.S
 +++ b/cpu/arm920t/start.S
 @@ -160,6 +160,12 @@ start_code:
         orr     r0,r0,#0xd3
         msr     cpsr,r0

 +       ldr     r0, =0x56000014
 +       ldr     r1, [r0]
 +       orr     r1, r1, #0x8 /* vibrator on */
 +       bic     r1, r1, #0x4 /* aux off */
 +       str     r1, [r0]
 +
         /* in case we run from the s3c24xx NAND stepping stone, the
 symbols
          * for LED support are in lib_arm/board.o, i.e. outside of the
          * steppingstone */


 }}}

 This causes the vibrator to run for about a second during boot and maybe
 about 20 ms during normal resume.

 This morning I noticed that the phone had not woken up from suspend
 properly and the vibrator was constantly on. I think this proves that the
 phone suspends properly and the CPU wakes from suspend due to RTC
 interrupt properly but something goes wrong somewhere between the first
 u-boot instructions and linux resume code.

 Unfortunately I really did not have the time to debug this with JTAG.
 Since I don't know what kind of permanent damage would be caused if I left
 the vibrator running for hours I decided to just remove the battery.
 Anyways, I hope that next time this occurs at a more suitable time so that
 I can attach the debug board and see what's going on :-)

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:13>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 Here's some more information. The CPU seems to be constantly hitting
 prefetch abort. RAM seems to be alive.

 {{{
 > dump_image fr.1 0x30000000 134217728
 }}}

 was so slow that I only managed to capture 53 MB of RAM before the FR ran
 out of battery. This RAM looks very odd:

 {{{
 $ hd proj/openocd-config/openocd/fr.2|cut --bytes=12-|sort|uniq -c
    1843
    1842 0 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00
 |................|
     921 0 00 00 00 00 00 00 00  00 00 00 00 34 00 ff ff
 |............4.ÿÿ|
  486840 0 00 00 00 25 3d ba 00  80 46 ea c9 80 0b 53 1b
 |....%=º..FêÉ..S.|
       1 0 00 00 00 30 0b ff ff  00 00 00 00 00 00 00 00
 |....0.ÿÿ........|
     736 0 00 00 00 34 00 ff ff  00 00 00 00 00 00 00 00
 |....4.ÿÿ........|
     183 0 00 00 00 68 00 ff ff  00 00 00 00 00 00 00 00
 |....h.ÿÿ........|
       1 0 00 00 00 ac 0b ff ff  00 00 00 00 00 00 00 00
 |....¬.ÿÿ........|
  486840 0 46 ea c9 80 0b 53 1b  00 e0 e4 26 80 50 00 2e
 |.FêÉ..S..àä&.P..|
  486840 0 73 4d 81 80 c8 a5 e5  80 f6 03 05 37 00 00 00
 |.sM..È¥å.ö..7...|
  486840 0 e0 e4 26 80 50 00 2e  00 73 4d 81 80 c8 a5 e5
 |.àä&.P...sM..È¥å|
  486840 0 f6 03 05 37 00 00 00  3e 00 00 00 3e 00 00 00
 |.ö..7...>...>...|
  389472 e 00 00 00 34 00 ff ff  00 00 00 00 25 3d ba 00
 |>...4.ÿÿ....%=º.|
  486840 e 00 00 00 3e 00 00 00  3e 00 00 00 34 00 ff ff
 |>...>...>...4.ÿÿ|
   97368 e 00 00 00 68 00 ff ff  00 00 00 00 25 3d ba 00
 |>...h.ÿÿ....%=º.|
 }}}

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:14>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 From kallsyms I can see that r8 contains

 c036e6c8 D s3c_irqwake_intmask

 and r10 contains

 c036e6cc D s3c_irqwake_eintmask

 lr_svc contains c004f8e0 which is quite close to

 c004f8b8 u printk       [ipv6]

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:15>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 I lost the original vmlinux file but one from a similar version shows that
 there is only one function that loads c036e6c8 to r8:

 {{{
 c003f0bc <s3c_pm_enter>:
 ...
 c003f168:       e59f8118        ldr     r8, [pc, #280]  ; c003f288
 <_end+0xffbf839c>
 c003f16c:       e59fa11c        ldr     sl, [pc, #284]  ; c003f290
 <_end+0xffbf83a4>
 c003f170:       e5832000        str     r2, [r3]
 c003f174:       e1a01002        mov     r1, r2
 c003f178:       e59f0120        ldr     r0, [pc, #288]  ; c003f2a0
 <_end+0xffbf83b4>
 c003f17c:       eb0041cd        bl      c004f8b8 <printk>
 c003f180:       eb0000ae        bl      c003f440 <s3c_pm_save_gpios>
 c003f184:       ebffeb1b        bl      c0039df8 <s3c_pm_save_core>
 c003f188:       ebffeb59        bl      c0039ef4 <s3c_pm_configure_extint>
 c003f18c:       e5981000        ldr     r1, [r8]
 c003f190:       e59a2000        ldr     r2, [sl]
 c003f194:       e59f0108        ldr     r0, [pc, #264]  ; c003f2a4
 <_end+0xffbf83b8>
 c003f198:       eb0041c6        bl      c004f8b8 <printk>
 c003f19c:       e59a2000        ldr     r2, [sl]
 c003f1a0:       e5983000        ldr     r3, [r8]
 c003f1a4:       e3a0533d        mov     r5, #-201326592 ; 0xf4000000
 c003f1a8:       e3a064fb        mov     r6, #-83886080  ; 0xfb000000
 c003f1ac:       e5853008        str     r3, [r5, #8]
 c003f1b0:       e58620a4        str     r2, [r6, #164]
 c003f1b4:       e59630a8        ldr     r3, [r6, #168]
 c003f1b8:       e59f70e8        ldr     r7, [pc, #232]  ; c003f2a8
 <_end+0xffbf83bc>
 c003f1bc:       e58630a8        str     r3, [r6, #168]
 c003f1c0:       e5953010        ldr     r3, [r5, #16]
 c003f1c4:       e3a09000        mov     r9, #0  ; 0x0
 c003f1c8:       e5853010        str     r3, [r5, #16]
 c003f1cc:       e5953000        ldr     r3, [r5]
 c003f1d0:       e5853000        str     r3, [r5]
 c003f1d4:       e59f309c        ldr     r3, [pc, #156]  ; c003f278
 <_end+0xffbf838c>
 c003f1d8:       e1a0e00f        mov     lr, pc
 c003f1dc:       e593f000        ldr     pc, [r3]
 c003f1e0:       e1a0e00f        mov     lr, pc
 c003f1e4:       e597f000        ldr     pc, [r7]
 c003f1e8:       e59f30bc        ldr     r3, [pc, #188]  ; c003f2ac
 <_end+0xffbf83c0>
 c003f1ec:       e1a00004        mov     r0, r4
 c003f1f0:       e583900c        str     r9, [r3, #12]
 c003f1f4:       ebffebde        bl      c003a174 <s3c_cpu_save>
 c003f1f8:       e1500009        cmp     r0, r9
 c003f1fc:       1a000004        bne     c003f214 <s3c_pm_enter+0x158>
 c003f200:       e1a0e00f        mov     lr, pc
 c003f204:       e597f000        ldr     pc, [r7]
 c003f208:       e59f306c        ldr     r3, [pc, #108]  ; c003f27c
 <_end+0xffbf8390>
 c003f20c:       e1a0e00f        mov     lr, pc
 c003f210:       e593f000        ldr     pc, [r3]
 c003f214:       ebffc432        bl      c00302e4 <cpu_init>
 c003f218:       ebffeb02        bl      c0039e28 <s3c_pm_restore_core>
 c003f21c:       eb00004e        bl      c003f35c <s3c_pm_restore_gpios>
 c003f220:       e5951000        ldr     r1, [r5]
 c003f224:       e59f0084        ldr     r0, [pc, #132]  ; c003f2b0
 <_end+0xffbf83c4>
 c003f228:       e59620a8        ldr     r2, [r6, #168]
 c003f22c:       eb0041a1        bl      c004f8b8 <printk>
 c003f230:       e5951000        ldr     r1, [r5]
 c003f234:       e5982000        ldr     r2, [r8]
 c003f238:       e3a00010        mov     r0, #16 ; 0x10
 c003f23c:       ebffff48        bl      c003ef64 <s3c_pm_show_resume_irqs>
 c003f240:       e59a2000        ldr     r2, [sl]
 c003f244:       e59610a8        ldr     r1, [r6, #168]
 c003f248:       e3a0002c        mov     r0, #44 ; 0x2c
 c003f24c:       ebffff44        bl      c003ef64 <s3c_pm_show_resume_irqs>
 c003f250:       e59f1018        ldr     r1, [pc, #24]   ; c003f270
 <_end+0xffbf8384>
 c003f254:       e59f0058        ldr     r0, [pc, #88]   ; c003f2b4
 <_end+0xffbf83c8>
 c003f258:       eb004196        bl      c004f8b8 <printk>
 c003f25c:       e59f0054        ldr     r0, [pc, #84]   ; c003f2b8
 <_end+0xffbf83cc>
 c003f260:       eb004194        bl      c004f8b8 <printk>
 c003f264:       e1a00009        mov     r0, r9
 c003f268:       e24bd028        sub     sp, fp, #40     ; 0x28
 c003f26c:       e89daff0        ldm     sp, {r4, r5, r6, r7, r8, r9, sl,
 fp, sp, pc}
 c003f270:       c02905bc        .word   0xc02905bc
 c003f274:       c03194a0        .word   0xc03194a0
 c003f278:       c038c97c        .word   0xc038c97c
 c003f27c:       c038c980        .word   0xc038c980
 c003f280:       c03194ac        .word   0xc03194ac
 c003f284:       c036ce04        .word   0xc036ce04
 c003f288:       c036e6c8        .word   0xc036e6c8
 c003f28c:       c036ce08        .word   0xc036ce08
 c003f290:       c036e6cc        .word   0xc036e6cc
 c003f294:       c03194d4        .word   0xc03194d4
 c003f298:       c03194f0        .word   0xc03194f0
 c003f29c:       c036ce1c        .word   0xc036ce1c
 c003f2a0:       c0319508        .word   0xc0319508
 c003f2a4:       c0319528        .word   0xc0319528
 c003f2a8:       c038bcd4        .word   0xc038bcd4
 c003f2ac:       f4100000        .word   0xf4100000
 c003f2b0:       c0319554        .word   0xc0319554
 c003f2b4:       c0319578        .word   0xc0319578
 c003f2b8:       c03195a0        .word   0xc03195a0
 }}}

 Are CPU registers zeroed on suspend?

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:16>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
Reply | Threaded
Open this post in threaded view
|

Re: Openmoko Bug #2309: hangs during resume after printing GSTATUS4

Openmoko Public Trac
In reply to this post by Openmoko Public Trac
#2309: hangs during resume after printing GSTATUS4
--------------------+-------------------------------------------------------
 Reporter:  lindi   |          Owner:  openmoko-kernel      
     Type:  defect  |         Status:  new                  
 Priority:  normal  |      Milestone:                      
Component:  kernel  |        Version:                      
 Severity:  normal  |       Keywords:  resume suspend kernel
 Haspatch:  0       |      Blockedby:                      
Estimated:          |    Patchreview:                      
 Blocking:          |   Reproducible:  rarely              
--------------------+-------------------------------------------------------

Comment(by lindi):

 < anarsoul> gena2x: lindi- : you can try to enable CONFIG_SAMSUNG_PM_CHECK
 in kernel
 < anarsoul> just to check that there's no memory corruption
 < anarsoul> and also look through steppingstone part of u-boot to ensure
 that it does not use memory at all

--
Ticket URL: <https://docs.openmoko.org/trac/ticket/2309#comment:17>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac