Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

[-- Attachment #1: Type: text/plain, Size: 5452 bytes --]

Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
> Hi,
> 
> I'm using the Odroid U2 to test some features implemented in
> GStreamer.
> The parts I need mostly works in mainline kernel, my main annoyance
> ist
> that reboot hangs.
> 
> Is this issue known ? I'm booting from SD card is that matters. Any
> advise would be appreciated. I'm not list member, so please keep me
> in
> CC. I'm sharing my .config, in case that is my fault.
> 
> regards,
> Nicolas


If that helps:

[root@odroidu2-nicolas nicolas]# echo 1 > /proc/sys/kernel/sysrq
[root@odroidu2-nicolas nicolas]# echo b > /proc/sysrq-trigger
[  153.605514] sysrq: SysRq : Resetting
[  153.605665] CPU0: stopping
[  153.606194] CPU: 0 PID: 247 Comm: NetworkManager Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
[  153.614764] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  153.620922] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
[  153.628605] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
[  153.635799] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
[  153.643171] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
[  153.650722] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
[  153.658168] Exception stack(0xec91dc78 to 0xec91dcc0)
[  153.663207] dc60:                                                       ee4263b0 edc6e400
[  153.671383] dc80: 00000000 000021f0 ee4263b0 ffffe000 0010200d 00000000 ee4263b0 ee426268
[  153.679541] dca0: ee4261c0 ec91dd08 00000003 ec91dcc8 c06cfc30 c0159f68 60000053 ffffffff
[  153.687711] [<c010b7d4>] (__irq_svc) from [<c0159f68>] (do_raw_spin_unlock+0x7c/0x11c)
[  153.695608] [<c0159f68>] (do_raw_spin_unlock) from [<c06cfc30>] (_raw_spin_unlock+0x8/0x3c)
[  153.703951] [<c06cfc30>] (_raw_spin_unlock) from [<c0210a34>] (__find_get_block_slow+0xd4/0x1a0)
[  153.712710] [<c0210a34>] (__find_get_block_slow) from [<c0211d04>] (__find_get_block+0x170/0x2dc)
[  153.721560] [<c0211d04>] (__find_get_block) from [<c02124ac>] (__getblk_gfp+0x24/0x54)
[  153.729455] [<c02124ac>] (__getblk_gfp) from [<c0251b0c>] (ext4_getblk+0x6c/0x174)
[  153.737000] [<c0251b0c>] (ext4_getblk) from [<c0251c20>] (ext4_bread+0xc/0x98)
[  153.744210] [<c0251c20>] (ext4_bread) from [<c025a9d8>] (__ext4_read_dirblock+0x2c/0x3b0)
[  153.752366] [<c025a9d8>] (__ext4_read_dirblock) from [<c025ad90>] (dx_probe+0x34/0x55c)
[  153.760350] [<c025ad90>] (dx_probe) from [<c025c828>] (ext4_find_entry+0x450/0x5f4)
[  153.767991] [<c025c828>] (ext4_find_entry) from [<c025fbcc>] (ext4_unlink+0x20/0x258)
[  153.775815] [<c025fbcc>] (ext4_unlink) from [<c01ee210>] (vfs_unlink+0xb8/0x190)
[  153.783188] [<c01ee210>] (vfs_unlink) from [<c01f186c>] (do_unlinkat+0x214/0x26c)
[  153.790655] [<c01f186c>] (do_unlinkat) from [<c01076c0>] (ret_fast_syscall+0x0/0x3c)
[  153.798355] CPU2: stopping
[  153.801062] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
[  153.809022] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  153.815156] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
[  153.822858] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
[  153.830053] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
[  153.837429] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
[  153.844977] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
[  153.852425] Exception stack(0xee8b7f88 to 0xee8b7fd0)
[  153.857476] 7f80:                   00000001 00000000 00000000 c01141c0 ee8b6000 c0b02454
[  153.865640] 7fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb c0b024c0 00003e4e ee8b7fd8
[  153.873787] 7fc0: c01081b0 c01081b4 600f0053 ffffffff
[  153.878840] [<c010b7d4>] (__irq_svc) from [<c01081b4>] (arch_cpu_idle+0x38/0x3c)
[  153.886234] [<c01081b4>] (arch_cpu_idle) from [<c0153578>] (cpu_startup_entry+0x1c0/0x238)
[  153.894463] [<c0153578>] (cpu_startup_entry) from [<401015ac>] (0x401015ac)
[  153.901384] CPU1: stopping
[  153.904089] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
[  153.912054] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  153.918188] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
[  153.925889] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
[  153.933086] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
[  153.940462] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
[  153.948010] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
[  153.955457] Exception stack(0xee8b5f88 to 0xee8b5fd0)
[  153.960508] 5f80:                   00000001 00000000 00000000 c01141c0 ee8b4000 c0b02454
[  153.968672] 5fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb c0b024c0 00003e62 ee8b5fd8
[  153.976819] 5fc0: c01081b0 c01081b4 60070053 ffffffff
[  153.981869] [<c010b7d4>] (__irq_svc) from [<c01081b4>] (arch_cpu_idle+0x38/0x3c)
[  153.989260] [<c01081b4>] (arch_cpu_idle) from [<c0153578>] (cpu_startup_entry+0x1c0/0x238)
[  153.997490] [<#

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Tobias Jakobi]

Hi Nicolas,

do you have the reset and poweroff drivers activated?

CONFIG_POWER_RESET_SYSCON=y
CONFIG_POWER_RESET_SYSCON_POWEROFF=y

With best wishes,
Tobias


Nicolas Dufresne wrote:
> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>> Hi,
>>
>> I'm using the Odroid U2 to test some features implemented in
>> GStreamer.
>> The parts I need mostly works in mainline kernel, my main annoyance
>> ist
>> that reboot hangs.
>>
>> Is this issue known ? I'm booting from SD card is that matters. Any
>> advise would be appreciated. I'm not list member, so please keep me
>> in
>> CC. I'm sharing my .config, in case that is my fault.
>>
>> regards,
>> Nicolas
> 
> 
> If that helps:
> 
> [root@odroidu2-nicolas nicolas]# echo 1 > /proc/sys/kernel/sysrq
> [root@odroidu2-nicolas nicolas]# echo b > /proc/sysrq-trigger
> [  153.605514] sysrq: SysRq : Resetting
> [  153.605665] CPU0: stopping
> [  153.606194] CPU: 0 PID: 247 Comm: NetworkManager Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
> [  153.614764] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
> [  153.620922] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
> [  153.628605] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
> [  153.635799] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
> [  153.643171] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
> [  153.650722] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
> [  153.658168] Exception stack(0xec91dc78 to 0xec91dcc0)
> [  153.663207] dc60:                                                       ee4263b0 edc6e400
> [  153.671383] dc80: 00000000 000021f0 ee4263b0 ffffe000 0010200d 00000000 ee4263b0 ee426268
> [  153.679541] dca0: ee4261c0 ec91dd08 00000003 ec91dcc8 c06cfc30 c0159f68 60000053 ffffffff
> [  153.687711] [<c010b7d4>] (__irq_svc) from [<c0159f68>] (do_raw_spin_unlock+0x7c/0x11c)
> [  153.695608] [<c0159f68>] (do_raw_spin_unlock) from [<c06cfc30>] (_raw_spin_unlock+0x8/0x3c)
> [  153.703951] [<c06cfc30>] (_raw_spin_unlock) from [<c0210a34>] (__find_get_block_slow+0xd4/0x1a0)
> [  153.712710] [<c0210a34>] (__find_get_block_slow) from [<c0211d04>] (__find_get_block+0x170/0x2dc)
> [  153.721560] [<c0211d04>] (__find_get_block) from [<c02124ac>] (__getblk_gfp+0x24/0x54)
> [  153.729455] [<c02124ac>] (__getblk_gfp) from [<c0251b0c>] (ext4_getblk+0x6c/0x174)
> [  153.737000] [<c0251b0c>] (ext4_getblk) from [<c0251c20>] (ext4_bread+0xc/0x98)
> [  153.744210] [<c0251c20>] (ext4_bread) from [<c025a9d8>] (__ext4_read_dirblock+0x2c/0x3b0)
> [  153.752366] [<c025a9d8>] (__ext4_read_dirblock) from [<c025ad90>] (dx_probe+0x34/0x55c)
> [  153.760350] [<c025ad90>] (dx_probe) from [<c025c828>] (ext4_find_entry+0x450/0x5f4)
> [  153.767991] [<c025c828>] (ext4_find_entry) from [<c025fbcc>] (ext4_unlink+0x20/0x258)
> [  153.775815] [<c025fbcc>] (ext4_unlink) from [<c01ee210>] (vfs_unlink+0xb8/0x190)
> [  153.783188] [<c01ee210>] (vfs_unlink) from [<c01f186c>] (do_unlinkat+0x214/0x26c)
> [  153.790655] [<c01f186c>] (do_unlinkat) from [<c01076c0>] (ret_fast_syscall+0x0/0x3c)
> [  153.798355] CPU2: stopping
> [  153.801062] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
> [  153.809022] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
> [  153.815156] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
> [  153.822858] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
> [  153.830053] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
> [  153.837429] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
> [  153.844977] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
> [  153.852425] Exception stack(0xee8b7f88 to 0xee8b7fd0)
> [  153.857476] 7f80:                   00000001 00000000 00000000 c01141c0 ee8b6000 c0b02454
> [  153.865640] 7fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb c0b024c0 00003e4e ee8b7fd8
> [  153.873787] 7fc0: c01081b0 c01081b4 600f0053 ffffffff
> [  153.878840] [<c010b7d4>] (__irq_svc) from [<c01081b4>] (arch_cpu_idle+0x38/0x3c)
> [  153.886234] [<c01081b4>] (arch_cpu_idle) from [<c0153578>] (cpu_startup_entry+0x1c0/0x238)
> [  153.894463] [<c0153578>] (cpu_startup_entry) from [<401015ac>] (0x401015ac)
> [  153.901384] CPU1: stopping
> [  153.904089] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.7.0-rc7-00034-gb569927-dirty #9
> [  153.912054] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
> [  153.918188] [<c010e480>] (unwind_backtrace) from [<c010ad18>] (show_stack+0x10/0x14)
> [  153.925889] [<c010ad18>] (show_stack) from [<c032bf38>] (dump_stack+0x78/0x8c)
> [  153.933086] [<c032bf38>] (dump_stack) from [<c010d2a8>] (handle_IPI+0x198/0x1ac)
> [  153.940462] [<c010d2a8>] (handle_IPI) from [<c010150c>] (gic_handle_irq+0x94/0x98)
> [  153.948010] [<c010150c>] (gic_handle_irq) from [<c010b7d4>] (__irq_svc+0x54/0x90)
> [  153.955457] Exception stack(0xee8b5f88 to 0xee8b5fd0)
> [  153.960508] 5f80:                   00000001 00000000 00000000 c01141c0 ee8b4000 c0b02454
> [  153.968672] 5fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb c0b024c0 00003e62 ee8b5fd8
> [  153.976819] 5fc0: c01081b0 c01081b4 60070053 ffffffff
> [  153.981869] [<c010b7d4>] (__irq_svc) from [<c01081b4>] (arch_cpu_idle+0x38/0x3c)
> [  153.989260] [<c01081b4>] (arch_cpu_idle) from [<c0153578>] (cpu_startup_entry+0x1c0/0x238)
> [  153.997490] [<#
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

Le jeudi 14 juillet 2016 à 15:45 -0400, Nicolas Dufresne a écrit :
> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
> > Hi,
> > 
> > I'm using the Odroid U2 to test some features implemented in
> > GStreamer.
> > The parts I need mostly works in mainline kernel, my main annoyance
> > ist
> > that reboot hangs.
> > 
> > Is this issue known ? I'm booting from SD card is that matters. Any
> > advise would be appreciated. I'm not list member, so please keep me
> > in
> > CC. I'm sharing my .config, in case that is my fault.

As suggested by Javier on IRC, here's the boot log:

http://paste.fedoraproject.org/391219/14685276/

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

[-- Attachment #1: Type: text/plain, Size: 6346 bytes --]

Le jeudi 14 juillet 2016 à 22:01 +0200, Tobias Jakobi a écrit :
> Hi Nicolas,
> 
> do you have the reset and poweroff drivers activated?
> 
> CONFIG_POWER_RESET_SYSCON=y
> CONFIG_POWER_RESET_SYSCON_POWEROFF=y

Yes, both.

> 
> With best wishes,
> Tobias
> 
> 
> Nicolas Dufresne wrote:
> > Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
> > > Hi,
> > > 
> > > I'm using the Odroid U2 to test some features implemented in
> > > GStreamer.
> > > The parts I need mostly works in mainline kernel, my main
> > > annoyance
> > > ist
> > > that reboot hangs.
> > > 
> > > Is this issue known ? I'm booting from SD card is that matters.
> > > Any
> > > advise would be appreciated. I'm not list member, so please keep
> > > me
> > > in
> > > CC. I'm sharing my .config, in case that is my fault.
> > > 
> > > regards,
> > > Nicolas
> > 
> > 
> > If that helps:
> > 
> > [root@odroidu2-nicolas nicolas]# echo 1 > /proc/sys/kernel/sysrq
> > [root@odroidu2-nicolas nicolas]# echo b > /proc/sysrq-trigger
> > [  153.605514] sysrq: SysRq : Resetting
> > [  153.605665] CPU0: stopping
> > [  153.606194] CPU: 0 PID: 247 Comm: NetworkManager Not tainted
> > 4.7.0-rc7-00034-gb569927-dirty #9
> > [  153.614764] Hardware name: SAMSUNG EXYNOS (Flattened Device
> > Tree)
> > [  153.620922] [<c010e480>] (unwind_backtrace) from [<c010ad18>]
> > (show_stack+0x10/0x14)
> > [  153.628605] [<c010ad18>] (show_stack) from [<c032bf38>]
> > (dump_stack+0x78/0x8c)
> > [  153.635799] [<c032bf38>] (dump_stack) from [<c010d2a8>]
> > (handle_IPI+0x198/0x1ac)
> > [  153.643171] [<c010d2a8>] (handle_IPI) from [<c010150c>]
> > (gic_handle_irq+0x94/0x98)
> > [  153.650722] [<c010150c>] (gic_handle_irq) from [<c010b7d4>]
> > (__irq_svc+0x54/0x90)
> > [  153.658168] Exception stack(0xec91dc78 to 0xec91dcc0)
> > [  153.663207]
> > dc60:                                                       ee4263b
> > 0 edc6e400
> > [  153.671383] dc80: 00000000 000021f0 ee4263b0 ffffe000 0010200d
> > 00000000 ee4263b0 ee426268
> > [  153.679541] dca0: ee4261c0 ec91dd08 00000003 ec91dcc8 c06cfc30
> > c0159f68 60000053 ffffffff
> > [  153.687711] [<c010b7d4>] (__irq_svc) from [<c0159f68>]
> > (do_raw_spin_unlock+0x7c/0x11c)
> > [  153.695608] [<c0159f68>] (do_raw_spin_unlock) from [<c06cfc30>]
> > (_raw_spin_unlock+0x8/0x3c)
> > [  153.703951] [<c06cfc30>] (_raw_spin_unlock) from [<c0210a34>]
> > (__find_get_block_slow+0xd4/0x1a0)
> > [  153.712710] [<c0210a34>] (__find_get_block_slow) from
> > [<c0211d04>] (__find_get_block+0x170/0x2dc)
> > [  153.721560] [<c0211d04>] (__find_get_block) from [<c02124ac>]
> > (__getblk_gfp+0x24/0x54)
> > [  153.729455] [<c02124ac>] (__getblk_gfp) from [<c0251b0c>]
> > (ext4_getblk+0x6c/0x174)
> > [  153.737000] [<c0251b0c>] (ext4_getblk) from [<c0251c20>]
> > (ext4_bread+0xc/0x98)
> > [  153.744210] [<c0251c20>] (ext4_bread) from [<c025a9d8>]
> > (__ext4_read_dirblock+0x2c/0x3b0)
> > [  153.752366] [<c025a9d8>] (__ext4_read_dirblock) from
> > [<c025ad90>] (dx_probe+0x34/0x55c)
> > [  153.760350] [<c025ad90>] (dx_probe) from [<c025c828>]
> > (ext4_find_entry+0x450/0x5f4)
> > [  153.767991] [<c025c828>] (ext4_find_entry) from [<c025fbcc>]
> > (ext4_unlink+0x20/0x258)
> > [  153.775815] [<c025fbcc>] (ext4_unlink) from [<c01ee210>]
> > (vfs_unlink+0xb8/0x190)
> > [  153.783188] [<c01ee210>] (vfs_unlink) from [<c01f186c>]
> > (do_unlinkat+0x214/0x26c)
> > [  153.790655] [<c01f186c>] (do_unlinkat) from [<c01076c0>]
> > (ret_fast_syscall+0x0/0x3c)
> > [  153.798355] CPU2: stopping
> > [  153.801062] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.7.0-rc7-
> > 00034-gb569927-dirty #9
> > [  153.809022] Hardware name: SAMSUNG EXYNOS (Flattened Device
> > Tree)
> > [  153.815156] [<c010e480>] (unwind_backtrace) from [<c010ad18>]
> > (show_stack+0x10/0x14)
> > [  153.822858] [<c010ad18>] (show_stack) from [<c032bf38>]
> > (dump_stack+0x78/0x8c)
> > [  153.830053] [<c032bf38>] (dump_stack) from [<c010d2a8>]
> > (handle_IPI+0x198/0x1ac)
> > [  153.837429] [<c010d2a8>] (handle_IPI) from [<c010150c>]
> > (gic_handle_irq+0x94/0x98)
> > [  153.844977] [<c010150c>] (gic_handle_irq) from [<c010b7d4>]
> > (__irq_svc+0x54/0x90)
> > [  153.852425] Exception stack(0xee8b7f88 to 0xee8b7fd0)
> > [  153.857476] 7f80:                   00000001 00000000 00000000
> > c01141c0 ee8b6000 c0b02454
> > [  153.865640] 7fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb
> > c0b024c0 00003e4e ee8b7fd8
> > [  153.873787] 7fc0: c01081b0 c01081b4 600f0053 ffffffff
> > [  153.878840] [<c010b7d4>] (__irq_svc) from [<c01081b4>]
> > (arch_cpu_idle+0x38/0x3c)
> > [  153.886234] [<c01081b4>] (arch_cpu_idle) from [<c0153578>]
> > (cpu_startup_entry+0x1c0/0x238)
> > [  153.894463] [<c0153578>] (cpu_startup_entry) from [<401015ac>]
> > (0x401015ac)
> > [  153.901384] CPU1: stopping
> > [  153.904089] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.7.0-rc7-
> > 00034-gb569927-dirty #9
> > [  153.912054] Hardware name: SAMSUNG EXYNOS (Flattened Device
> > Tree)
> > [  153.918188] [<c010e480>] (unwind_backtrace) from [<c010ad18>]
> > (show_stack+0x10/0x14)
> > [  153.925889] [<c010ad18>] (show_stack) from [<c032bf38>]
> > (dump_stack+0x78/0x8c)
> > [  153.933086] [<c032bf38>] (dump_stack) from [<c010d2a8>]
> > (handle_IPI+0x198/0x1ac)
> > [  153.940462] [<c010d2a8>] (handle_IPI) from [<c010150c>]
> > (gic_handle_irq+0x94/0x98)
> > [  153.948010] [<c010150c>] (gic_handle_irq) from [<c010b7d4>]
> > (__irq_svc+0x54/0x90)
> > [  153.955457] Exception stack(0xee8b5f88 to 0xee8b5fd0)
> > [  153.960508] 5f80:                   00000001 00000000 00000000
> > c01141c0 ee8b4000 c0b02454
> > [  153.968672] 5fa0: 00000000 00000000 c0a623f8 c0b024b8 c0b3ecdb
> > c0b024c0 00003e62 ee8b5fd8
> > [  153.976819] 5fc0: c01081b0 c01081b4 60070053 ffffffff
> > [  153.981869] [<c010b7d4>] (__irq_svc) from [<c01081b4>]
> > (arch_cpu_idle+0x38/0x3c)
> > [  153.989260] [<c01081b4>] (arch_cpu_idle) from [<c0153578>]
> > (cpu_startup_entry+0x1c0/0x238)
> > [  153.997490] [<#
> > 
> 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/14/2016 10:33 PM, Nicolas Dufresne wrote:
> Le jeudi 14 juillet 2016 à 22:01 +0200, Tobias Jakobi a écrit :
>> Hi Nicolas,
>>
>> do you have the reset and poweroff drivers activated?
>>
>> CONFIG_POWER_RESET_SYSCON=y
>> CONFIG_POWER_RESET_SYSCON_POWEROFF=y
> 
> Yes, both.
> 
>>
>> With best wishes,
>> Tobias
>>
>>
>> Nicolas Dufresne wrote:
>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>> Hi,
>>>>
>>>> I'm using the Odroid U2 to test some features implemented in
>>>> GStreamer.
>>>> The parts I need mostly works in mainline kernel, my main
>>>> annoyance
>>>> ist
>>>> that reboot hangs.
>>>>
>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>> Any
>>>> advise would be appreciated. I'm not list member, so please keep
>>>> me
>>>> in
>>>> CC. I'm sharing my .config, in case that is my fault.

First of all - use just command "reboot" to restart. It should work. If
it doesn't, please let us now.

I was able to reproduce your issue - OOPS in emergency restart - also in
my configuration (U3 + eMMC + network boot).

It is weird... I'll take a look at it.

Best regards,
Krzysztog

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>> Nicolas Dufresne wrote:
>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>> Hi,
>>>>>
>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>> GStreamer.
>>>>> The parts I need mostly works in mainline kernel, my main
>>>>> annoyance
>>>>> ist
>>>>> that reboot hangs.
>>>>>
>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>> Any
>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>> me
>>>>> in
>>>>> CC. I'm sharing my .config, in case that is my fault.
> 
> First of all - use just command "reboot" to restart. It should work. If
> it doesn't, please let us now.
> 
> I was able to reproduce your issue - OOPS in emergency restart - also in
> my configuration (U3 + eMMC + network boot).
> 
> It is weird... I'll take a look at it.

Actually I did not encounter OOPS of yours. I was able to reproduce a
failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...

The failure seems to be present for long time, at least since v4.2. I am
in progress of testing more releases.

ext4 backtrace present in your OOPs might be related to improper reboot.
Even for emergency reboot you should first sync, then remount ro and
finally reboot (so sysrq: sub).

Best Regards,
Krzysztof

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Tobias Jakobi]

Hey,

Krzysztof Kozlowski wrote:
> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>> Nicolas Dufresne wrote:
>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>> Hi,
>>>>>>
>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>> GStreamer.
>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>> annoyance
>>>>>> ist
>>>>>> that reboot hangs.
>>>>>>
>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>> Any
>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>> me
>>>>>> in
>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>
>> First of all - use just command "reboot" to restart. It should work. If
>> it doesn't, please let us now.
>>
>> I was able to reproduce your issue - OOPS in emergency restart - also in
>> my configuration (U3 + eMMC + network boot).
>>
>> It is weird... I'll take a look at it.
> 
> Actually I did not encounter OOPS of yours. I was able to reproduce a
> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
I'm not sure how much that helps, but I can't reproduce any of this on
my X2 (which is quite closely related to the U2/U3).
I'm only encountering non-working reboot when applying Marek's latest
IOMMU runpm patchset.


> The failure seems to be present for long time, at least since v4.2. I am
> in progress of testing more releases.
At least not for the X2, and I'm booting from a sdcard like Nicolas.
IIRC then pwrseq only plays a role for booting from eMMC.


> ext4 backtrace present in your OOPs might be related to improper reboot.
> Even for emergency reboot you should first sync, then remount ro and
> finally reboot (so sysrq: sub).
I currently don't have access to my UART adapater, but I'll try to
reproduce the Oops by issuing a reboot SysRq directly.


With best wishes,
Tobias


> Best Regards,
> Krzysztof
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/15/2016 03:15 PM, Tobias Jakobi wrote:
> Hey,
> 
> Krzysztof Kozlowski wrote:
>> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>>> Nicolas Dufresne wrote:
>>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>>> Hi,
>>>>>>>
>>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>>> GStreamer.
>>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>>> annoyance
>>>>>>> ist
>>>>>>> that reboot hangs.
>>>>>>>
>>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>>> Any
>>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>>> me
>>>>>>> in
>>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>>
>>> First of all - use just command "reboot" to restart. It should work. If
>>> it doesn't, please let us now.
>>>
>>> I was able to reproduce your issue - OOPS in emergency restart - also in
>>> my configuration (U3 + eMMC + network boot).
>>>
>>> It is weird... I'll take a look at it.
>>
>> Actually I did not encounter OOPS of yours. I was able to reproduce a
>> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
> I'm not sure how much that helps, but I can't reproduce any of this on
> my X2 (which is quite closely related to the U2/U3).

You mean sysrq reboot works fine?

Best regards,
Krzysztof

> I'm only encountering non-working reboot when applying Marek's latest
> IOMMU runpm patchset.
> 
> 
>> The failure seems to be present for long time, at least since v4.2. I am
>> in progress of testing more releases.
> At least not for the X2, and I'm booting from a sdcard like Nicolas.
> IIRC then pwrseq only plays a role for booting from eMMC.
> 
> 
>> ext4 backtrace present in your OOPs might be related to improper reboot.
>> Even for emergency reboot you should first sync, then remount ro and
>> finally reboot (so sysrq: sub).
> I currently don't have access to my UART adapater, but I'll try to
> reproduce the Oops by issuing a reboot SysRq directly.
> 
> 
> With best wishes,
> Tobias
> 
> 
>> Best Regards,
>> Krzysztof
>>
> 
> 
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Tobias Jakobi]

Krzysztof Kozlowski wrote:
> On 07/15/2016 03:15 PM, Tobias Jakobi wrote:
>> Hey,
>>
>> Krzysztof Kozlowski wrote:
>>> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>>>> Nicolas Dufresne wrote:
>>>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>>>> GStreamer.
>>>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>>>> annoyance
>>>>>>>> ist
>>>>>>>> that reboot hangs.
>>>>>>>>
>>>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>>>> Any
>>>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>>>> me
>>>>>>>> in
>>>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>>>
>>>> First of all - use just command "reboot" to restart. It should work. If
>>>> it doesn't, please let us now.
>>>>
>>>> I was able to reproduce your issue - OOPS in emergency restart - also in
>>>> my configuration (U3 + eMMC + network boot).
>>>>
>>>> It is weird... I'll take a look at it.
>>>
>>> Actually I did not encounter OOPS of yours. I was able to reproduce a
>>> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
>> I'm not sure how much that helps, but I can't reproduce any of this on
>> my X2 (which is quite closely related to the U2/U3).
> 
> You mean sysrq reboot works fine?
No, I haven't checked that yet. But normal reboot and poweroff via init
works properly.

- Tobias


> Best regards,
> Krzysztof
> 
>> I'm only encountering non-working reboot when applying Marek's latest
>> IOMMU runpm patchset.
>>
>>
>>> The failure seems to be present for long time, at least since v4.2. I am
>>> in progress of testing more releases.
>> At least not for the X2, and I'm booting from a sdcard like Nicolas.
>> IIRC then pwrseq only plays a role for booting from eMMC.
>>
>>
>>> ext4 backtrace present in your OOPs might be related to improper reboot.
>>> Even for emergency reboot you should first sync, then remount ro and
>>> finally reboot (so sysrq: sub).
>> I currently don't have access to my UART adapater, but I'll try to
>> reproduce the Oops by issuing a reboot SysRq directly.
>>
>>
>> With best wishes,
>> Tobias
>>
>>
>>> Best Regards,
>>> Krzysztof
>>>
>>
>>
>>
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/15/2016 03:20 PM, Krzysztof Kozlowski wrote:
> On 07/15/2016 03:15 PM, Tobias Jakobi wrote:
>> Hey,
>>
>> Krzysztof Kozlowski wrote:
>>> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>>>> Nicolas Dufresne wrote:
>>>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>>>> GStreamer.
>>>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>>>> annoyance
>>>>>>>> ist
>>>>>>>> that reboot hangs.
>>>>>>>>
>>>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>>>> Any
>>>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>>>> me
>>>>>>>> in
>>>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>>>
>>>> First of all - use just command "reboot" to restart. It should work. If
>>>> it doesn't, please let us now.
>>>>
>>>> I was able to reproduce your issue - OOPS in emergency restart - also in
>>>> my configuration (U3 + eMMC + network boot).
>>>>
>>>> It is weird... I'll take a look at it.
>>>
>>> Actually I did not encounter OOPS of yours. I was able to reproduce a
>>> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
>> I'm not sure how much that helps, but I can't reproduce any of this on
>> my X2 (which is quite closely related to the U2/U3).
> 
> You mean sysrq reboot works fine?

BTW, I bisected this to merge commit appearing in v4.3 merge window,
bringing in cpufreq-dt to Exynos4412. There was a reboot issue related
to cpufreq frequency and voltage. Bartlomiej fixed it in 1605b60ad064
("ARM: dts: add suspend opp to exynos4412"). However the opp-suspend and
cpufreq-suspend are activated through syscore shutdown.

Unfortunately it seems that syscore shutdown is not called during
emergency reboot.

It looks like more generic problem.

Best regards,
Krzysztof

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/15/2016 03:23 PM, Tobias Jakobi wrote:
> Krzysztof Kozlowski wrote:
>> On 07/15/2016 03:15 PM, Tobias Jakobi wrote:
>>> Hey,
>>>
>>> Krzysztof Kozlowski wrote:
>>>> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>>>>> Nicolas Dufresne wrote:
>>>>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>>>>> GStreamer.
>>>>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>>>>> annoyance
>>>>>>>>> ist
>>>>>>>>> that reboot hangs.
>>>>>>>>>
>>>>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>>>>> Any
>>>>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>>>>> me
>>>>>>>>> in
>>>>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>>>>
>>>>> First of all - use just command "reboot" to restart. It should work. If
>>>>> it doesn't, please let us now.
>>>>>
>>>>> I was able to reproduce your issue - OOPS in emergency restart - also in
>>>>> my configuration (U3 + eMMC + network boot).
>>>>>
>>>>> It is weird... I'll take a look at it.
>>>>
>>>> Actually I did not encounter OOPS of yours. I was able to reproduce a
>>>> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
>>> I'm not sure how much that helps, but I can't reproduce any of this on
>>> my X2 (which is quite closely related to the U2/U3).
>>
>> You mean sysrq reboot works fine?
> No, I haven't checked that yet. But normal reboot and poweroff via init
> works properly.

Yeah, these work fine. It is all about emergency restart.

Best regards,
Krzysztof

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

[-- Attachment #1: Type: text/plain, Size: 675 bytes --]

Le vendredi 15 juillet 2016 à 09:55 +0200, Krzysztof Kozlowski a
écrit :
> 
> First of all - use just command "reboot" to restart. It should work.
> If
> it doesn't, please let us now.

It does not work with reboot either, though no OOPS, just hang. Is the
trace at sysrq reboot really a OOPS? I believed it was just a dump of
the 3 other CPU at the moment the request was made.

> 
> I was able to reproduce your issue - OOPS in emergency restart - also
> in
> my configuration (U3 + eMMC + network boot).
> 
> It is weird... I'll take a look at it.

Btw, my configuration is U2 + SD boot. Any ways I can gather provide
better informations ?

Nicolas

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Tobias Jakobi]

Krzysztof Kozlowski wrote:
> On 07/15/2016 03:23 PM, Tobias Jakobi wrote:
>> Krzysztof Kozlowski wrote:
>>> On 07/15/2016 03:15 PM, Tobias Jakobi wrote:
>>>> Hey,
>>>>
>>>> Krzysztof Kozlowski wrote:
>>>>> On 07/15/2016 09:55 AM, Krzysztof Kozlowski wrote:
>>>>>>>> Nicolas Dufresne wrote:
>>>>>>>>> Le jeudi 14 juillet 2016 à 15:37 -0400, Nicolas Dufresne a écrit :
>>>>>>>>>> Hi,
>>>>>>>>>>
>>>>>>>>>> I'm using the Odroid U2 to test some features implemented in
>>>>>>>>>> GStreamer.
>>>>>>>>>> The parts I need mostly works in mainline kernel, my main
>>>>>>>>>> annoyance
>>>>>>>>>> ist
>>>>>>>>>> that reboot hangs.
>>>>>>>>>>
>>>>>>>>>> Is this issue known ? I'm booting from SD card is that matters.
>>>>>>>>>> Any
>>>>>>>>>> advise would be appreciated. I'm not list member, so please keep
>>>>>>>>>> me
>>>>>>>>>> in
>>>>>>>>>> CC. I'm sharing my .config, in case that is my fault.
>>>>>>
>>>>>> First of all - use just command "reboot" to restart. It should work. If
>>>>>> it doesn't, please let us now.
>>>>>>
>>>>>> I was able to reproduce your issue - OOPS in emergency restart - also in
>>>>>> my configuration (U3 + eMMC + network boot).
>>>>>>
>>>>>> It is weird... I'll take a look at it.
>>>>>
>>>>> Actually I did not encounter OOPS of yours. I was able to reproduce a
>>>>> failure to reboot on U3. XU3 reboots fine. It might be related to pwrseq...
>>>> I'm not sure how much that helps, but I can't reproduce any of this on
>>>> my X2 (which is quite closely related to the U2/U3).
>>>
>>> You mean sysrq reboot works fine?
>> No, I haven't checked that yet. But normal reboot and poweroff via init
>> works properly.
> 
> Yeah, these work fine. It is all about emergency restart.
I can confirm that this indeed doesn't work.

chidori ~ # echo 1 > /proc/sys/kernel/sysrq
chidori ~ # echo s > /proc/sysrq-trigger
chidori ~ # echo u > /proc/sysrq-trigger
chidori ~ # echo b > /proc/sysrq-trigger
<board hangs, only red power LED on>

I guess Nicolas is seeing a different or another problem here though.

With best wishes,
Tobias


> Best regards,
> Krzysztof
> 
> 
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

[-- Attachment #1: Type: text/plain, Size: 770 bytes --]

Le vendredi 15 juillet 2016 à 16:33 +0200, Tobias Jakobi a écrit :
> I can confirm that this indeed doesn't work.
> 
> chidori ~ # echo 1 > /proc/sys/kernel/sysrq
> chidori ~ # echo s > /proc/sysrq-trigger
> chidori ~ # echo u > /proc/sysrq-trigger
> chidori ~ # echo b > /proc/sysrq-trigger
> <board hangs, only red power LED on>
> 
> I guess Nicolas is seeing a different or another problem here though.

Well for me both method hangs. On an extra note, reset inside Uboot
works, but reboot in Linux hang. I tried adding a trace
in syscon_restart_handle(), but the trace is never displayed. I'm
wondering if that handler is ever called, or if it's just that the
serial console was shut down already.

Any idea what else should I trace ?

Nicolas

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Marian Mihailescu]

I can provide more information, i've been battling with this for a long time.
Tried on U3 and U3+ with similar results:

1) samsung tizen 4.1 kernel reboots OK.

2) i tried 4.4, 4.6, 4.7 (from Tobias's repository) all have the reboot issue.
i've put some printk in the syscon reboot. it looks like it gets registered,
but i never saw the printk put in the actual notifier function.

3) the issue is the same for emmc and sd-card, i've tried both

4) IT'S VERY RANDOM

* i've got sysrq rebooting OK, i've got sysrq ending in a kernel crash
(like Nicolas above)
* I've got rebooting working ok sometimes, but only from tty1 console or
sometimes ssh - NEVER from ttySAC1 serial console.
rebooting from serial console always resulted in hang after systemd stops.
* after switching from emmc to sd-card, there are always problems.
i had sd-card rebooting ok (from tty1) for a few hours, then after i put in
an emmc card => hell broke loose. putting sd-card back => still had issues,
although sd was fine before.
* many times, after reboot hang or switching from emmc to sd
or the other way around => kernel would crash on boot for a few times.

I reverted the SYSCON driver changes, and went back to the reboot code
in the exynos pmu.c file like 4.1 kernel => got the same issues.

I thought it's a systemd issue (after seeing echo b > /proc/sysrq-trigger
working ok) and tried the kernel on a ubuntu 14.04 image (with upstart)
=> got the same issues.

I think it might be u-boot.

PS: please keep me in the CC from here on as well.

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/15/2016 05:56 PM, Nicolas Dufresne wrote:
> Le vendredi 15 juillet 2016 à 16:33 +0200, Tobias Jakobi a écrit :
>> I can confirm that this indeed doesn't work.
>>
>> chidori ~ # echo 1 > /proc/sys/kernel/sysrq
>> chidori ~ # echo s > /proc/sysrq-trigger
>> chidori ~ # echo u > /proc/sysrq-trigger
>> chidori ~ # echo b > /proc/sysrq-trigger
>> <board hangs, only red power LED on>
>>
>> I guess Nicolas is seeing a different or another problem here though.
> 
> Well for me both method hangs. On an extra note, reset inside Uboot
> works, but reboot in Linux hang. I tried adding a trace
> in syscon_restart_handle(), but the trace is never displayed. I'm
> wondering if that handler is ever called, or if it's just that the
> serial console was shut down already.
> 
> Any idea what else should I trace ?

This is weird... reboot should work. Just to confirm it - can you test
following patch in both scenarios (this should fix both of them... at
least on Odroid U3 but your board might have different SPL/BL1):

-- >8 --
diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
index 58ad48e7b8f7..7d963d36e96d 100644
--- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
+++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
@@ -423,7 +423,8 @@

                        buck2_reg: BUCK2 {
                                regulator-name = "vdd_arm";
-                               regulator-min-microvolt = <900000>;
+                               //regulator-min-microvolt = <900000>;
+                               regulator-min-microvolt = <1000000>;
                                regulator-max-microvolt = <1350000>;
                                regulator-always-on;
                                regulator-boot-on;

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Krzysztof Kozlowski]

On 07/19/2016 01:29 PM, Krzysztof Kozlowski wrote:
> On 07/15/2016 05:56 PM, Nicolas Dufresne wrote:
>> Le vendredi 15 juillet 2016 à 16:33 +0200, Tobias Jakobi a écrit :
>>> I can confirm that this indeed doesn't work.
>>>
>>> chidori ~ # echo 1 > /proc/sys/kernel/sysrq
>>> chidori ~ # echo s > /proc/sysrq-trigger
>>> chidori ~ # echo u > /proc/sysrq-trigger
>>> chidori ~ # echo b > /proc/sysrq-trigger
>>> <board hangs, only red power LED on>
>>>
>>> I guess Nicolas is seeing a different or another problem here though.
>>
>> Well for me both method hangs. On an extra note, reset inside Uboot
>> works, but reboot in Linux hang. I tried adding a trace
>> in syscon_restart_handle(), but the trace is never displayed. I'm
>> wondering if that handler is ever called, or if it's just that the
>> serial console was shut down already.
>>
>> Any idea what else should I trace ?
> 
> This is weird... reboot should work. Just to confirm it - can you test
> following patch in both scenarios (this should fix both of them... at
> least on Odroid U3 but your board might have different SPL/BL1):

I dig a little bit and emergency reboot on Odroid U3 (and family) is not
solvable in an easy and clean way.

The problem is: reset is a software reset of AP which means that clocks
go to default values. If cpufreq is active and CPU regulator voltage is
low, then it will fail to boot because clock frequency is high but
voltage is low, too low.

Possible solutions would be:
1. Reset PMIC values. This does not happen on software reset. I tried
also hardware reset through watchdog - also nothing.

2. Set regulator to safe value (same as for suspend, opp-suspend). This
cannot be used in emergency path because it is a sleeping, long
operation. I even wrote simpler cpufreq restart handler but still this
is not accepted sleeping operation. Also on Odroid U3 the CPU regulator
voltage cannot be controlled through GPIO/DVS because... they are wired
to VDD.

3. Extend emergency reboot path with execution of some new
syscore_emergency_shutdown() op. That is quite intrusive and requires
changing in many archs... Not really applicable.

4. Change the minimum voltage of buck2 to 1.0V on Odroid. Set is as
constrain, always. Not nice, but should help... Is it worth the effort?

Best regards,
Krzysztof

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Nicolas Dufresne]

Hi Krzysztof,

I'm leaving for a long vacation (6w) today and didn't find the time to
test your patch. I'd like to thank you for looking into this issue. I
will we able to test this patch at my return.

I do understand now (and with the following mail), that this hardware
have problematic constraints that are specific to the Hardkernel
integration.

regards,
Nicolas

Le mardi 19 juillet 2016 à 13:29 +0200, Krzysztof Kozlowski a écrit :
> On 07/15/2016 05:56 PM, Nicolas Dufresne wrote:
> > Le vendredi 15 juillet 2016 à 16:33 +0200, Tobias Jakobi a écrit :
> > > I can confirm that this indeed doesn't work.
> > > 
> > > chidori ~ # echo 1 > /proc/sys/kernel/sysrq
> > > chidori ~ # echo s > /proc/sysrq-trigger
> > > chidori ~ # echo u > /proc/sysrq-trigger
> > > chidori ~ # echo b > /proc/sysrq-trigger
> > > <board hangs, only red power LED on>
> > > 
> > > I guess Nicolas is seeing a different or another problem here
> > > though.
> > 
> > Well for me both method hangs. On an extra note, reset inside Uboot
> > works, but reboot in Linux hang. I tried adding a trace
> > in syscon_restart_handle(), but the trace is never displayed. I'm
> > wondering if that handler is ever called, or if it's just that the
> > serial console was shut down already.
> > 
> > Any idea what else should I trace ?
> 
> This is weird... reboot should work. Just to confirm it - can you
> test
> following patch in both scenarios (this should fix both of them... at
> least on Odroid U3 but your board might have different SPL/BL1):
> 
> -- >8 --
> diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> index 58ad48e7b8f7..7d963d36e96d 100644
> --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> @@ -423,7 +423,8 @@
> 
>                         buck2_reg: BUCK2 {
>                                 regulator-name = "vdd_arm";
> -                               regulator-min-microvolt = <900000>;
> +                               //regulator-min-microvolt = <900000>;
> +                               regulator-min-microvolt = <1000000>;
>                                 regulator-max-microvolt = <1350000>;
>                                 regulator-always-on;
>                                 regulator-boot-on;
> 
> 
> 
> 

Re: Odroid U2 Kernel 4.7rc7+ Hang on reboot

[Marian Mihailescu]

Krzysztof Kozlowski <k.kozlowski <at> samsung.com> writes:

> 
> On 07/15/2016 05:56 PM, Nicolas Dufresne wrote:
> > Le vendredi 15 juillet 2016 à 16:33 +0200, Tobias Jakobi a écrit :
> >> I can confirm that this indeed doesn't work.
> >>
> >> chidori ~ # echo 1 > /proc/sys/kernel/sysrq
> >> chidori ~ # echo s > /proc/sysrq-trigger
> >> chidori ~ # echo u > /proc/sysrq-trigger
> >> chidori ~ # echo b > /proc/sysrq-trigger
> >> <board hangs, only red power LED on>
> >>
> >> I guess Nicolas is seeing a different or another problem here 
though.
> > 
> > Well for me both method hangs. On an extra note, reset inside Uboot
> > works, but reboot in Linux hang. I tried adding a trace
> > in syscon_restart_handle(), but the trace is never displayed. I'm
> > wondering if that handler is ever called, or if it's just that the
> > serial console was shut down already.
> > 
> > Any idea what else should I trace ?
> 
> This is weird... reboot should work. Just to confirm it - can you test
> following patch in both scenarios (this should fix both of them... at
> least on Odroid U3 but your board might have different SPL/BL1):
> 

Hi Krzysztof,

Thanks for looking into this.
While Nicolas is away, I can do whatever testing you need.
I can confirm that with this patch, I can now reboot both from tty1 and 
ttySAC1. I haven't done any extensive testing, but I could reboot 
without issues several times in a row.

I do remember with previous kernels (even 3.x) there would be issues 
with reboot if the governor was anything else than performance. This is 
why my config only has performance enabled - but I was still having 
reboot issues. Do you expect using performance gov + this patch to fully 
solve the reboot issue?

Thanks,
Marian