AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[Tuomas Tynkkynen]

Hello,

Booting Linus's master of today with the usual arm64 defconfig fails for
me on the Hikey960 board. I've bisected it down to:

commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
Author: Will Deacon <will.deacon@arm.com>
Date:   Tue Jun 19 17:55:28 2018 +0100

     arm64: insn: Don't fallback on nosync path for general insn patching

Here's the boot log with earlycon enabled after which the boot hangs:

[    0.000000] Booting Linux on physical CPU 0x0000000000 [0x410fd034]
[    0.000000] Linux version 4.18.0-rc3-00017-g693350a79980 (tmtynkky@duuni) (gcc version 8.2.0 (Buildroot 2018.08-git-00807-g0184fa3)) #17 SMP PREEMPT Thu Aug 16 02:18:59 EEST 2018
[    0.000000] Machine model: HiKey960
[    0.000000] efi: Getting EFI parameters from FDT:
[    0.000000] efi: EFI v2.60 by EDK II
[    0.000000] efi:  MEMATTR=0xbe65d698
[    0.000000] cma: Reserved 32 MiB at 0x00000000bbc00000
[    0.000000] earlycon: pl11 at MMIO 0x00000000fff32000 (options '115200n8')
[    0.000000] bootconsole [pl11] enabled
[    0.000000] NUMA: No NUMA configuration found
[    0.000000] NUMA: Faking a node at [mem 0x0000000000000000-0x00000000bfffffff]
[    0.000000] NUMA: NODE_DATA [mem 0xbffbf240-0xbffc09ff]
[    0.000000] Zone ranges:
[    0.000000]   DMA32    [mem 0x0000000000000000-0x00000000bfffffff]
[    0.000000]   Normal   empty
[    0.000000] Movable zone start for each node
[    0.000000] Early memory node ranges
[    0.000000]   node   0: [mem 0x0000000000000000-0x000000001abfffff]
[    0.000000]   node   0: [mem 0x000000001ad88000-0x0000000031ffffff]
[    0.000000]   node   0: [mem 0x0000000032101000-0x000000003dffffff]
[    0.000000]   node   0: [mem 0x0000000040000000-0x0000000089b7ffff]
[    0.000000]   node   0: [mem 0x0000000089cc0000-0x00000000b9bdffff]
[    0.000000]   node   0: [mem 0x00000000b9be0000-0x00000000b9c7ffff]
[    0.000000]   node   0: [mem 0x00000000b9c80000-0x00000000b9d6ffff]
[    0.000000]   node   0: [mem 0x00000000b9d70000-0x00000000ba12ffff]
[    0.000000]   node   0: [mem 0x00000000ba130000-0x00000000ba130fff]
[    0.000000]   node   0: [mem 0x00000000ba131000-0x00000000ba134fff]
[    0.000000]   node   0: [mem 0x00000000ba135000-0x00000000bf0fffff]
[    0.000000]   node   0: [mem 0x00000000bf100000-0x00000000bf13ffff]
[    0.000000]   node   0: [mem 0x00000000bf140000-0x00000000bf14ffff]
[    0.000000]   node   0: [mem 0x00000000bf150000-0x00000000bf19ffff]
[    0.000000]   node   0: [mem 0x00000000bf1a0000-0x00000000bfffffff]
[    0.000000] Initmem setup node 0 [mem 0x0000000000000000-0x00000000bfffffff]
[    0.000000] On node 0 totalpages: 777271
[    0.000000]   DMA32 zone: 12288 pages used for memmap
[    0.000000]   DMA32 zone: 0 pages reserved
[    0.000000]   DMA32 zone: 777271 pages, LIFO batch:31
[    0.000000] psci: probing for conduit method from DT.
[    0.000000] psci: PSCIv1.1 detected in firmware.
[    0.000000] psci: Using standard PSCI v0.2 function IDs
[    0.000000] psci: MIGRATE_INFO_TYPE not supported.
[    0.000000] psci: SMC Calling Convention v1.1
[    0.000000] random: get_random_bytes called from start_kernel+0x94/0x400 with crng_init=0
[    0.000000] percpu: Embedded 23 pages/cpu @(____ptrval____) s56024 r8192 d29992 u94208
[    0.000000] pcpu-alloc: s56024 r8192 d29992 u94208 alloc=23*4096
[    0.000000] pcpu-alloc: [0] 0 [0] 1 [0] 2 [0] 3 [0] 4 [0] 5 [0] 6 [0] 7
[    0.000000] Detected VIPT I-cache on CPU0
[    0.000000] CPU features: detected: Kernel page table isolation (KPTI)
[    0.000000] CPU features: enabling workaround for ARM erratum 845719
[    0.000000] Speculative Store Bypass Disable mitigation not required
[    0.000000] Built 1 zonelists, mobility grouping on.  Total pages: 764983
[    0.000000] Policy zone: DMA32
[    0.000000] Kernel command line: BOOT_IMAGE=/boot/linux-next-kernel/Image loglevel=8 console=ttyAMA6,115200n8 root=/dev/mmcblk0p2 clk_ignore_unused earlycon
[    0.000000] Memory: 2998448K/3109084K available (10428K kernel code, 1352K rwdata, 7108K rodata, 1216K init, 379K bss, 77868K reserved, 32768K cma-reserved)
[    0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=8, Nodes=1
[    0.000000] Preemptible hierarchical RCU implementation.
[    0.000000] 	RCU restricting CPUs from NR_CPUS=64 to nr_cpu_ids=8.
[    0.000000] 	Tasks RCU enabled.
[    0.000000] RCU: Adjusting geometry for rcu_fanout_leaf=16, nr_cpu_ids=8
[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0
[    0.000000] GIC: Using split EOI/Deactivate mode
[    0.000000] arch_timer: cp15 timer(s) running at 1.92MHz (phys).
[    0.000000] clocksource: arch_sys_counter: mask: 0xffffffffffffff max_cycles: 0x1c570327e, max_idle_ns: 1763180809505 ns
[    0.000004] sched_clock: 56 bits at 1920kHz, resolution 520ns, wraps every 4398046510977ns
[    0.008472] clocksource: arm,sp804: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 58334160866034 ns
[    0.018760] Console: colour dummy device 80x25
[    0.023349] Calibrating delay loop (skipped), value calculated using timer frequency.. 3.84 BogoMIPS (lpj=7680)
[    0.033596] pid_max: default: 32768 minimum: 301
[    0.038337] Security Framework initialized
[    0.043842] Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes)
[    0.051817] Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes)
[    0.059009] Mount-cache hash table entries: 8192 (order: 4, 65536 bytes)
[    0.065827] Mountpoint-cache hash table entries: 8192 (order: 4, 65536 bytes)
[    0.097149] ASID allocator initialised with 32768 entries
[    0.102693] Hierarchical SRCU implementation.
[    0.112378] Remapping and enabling EFI services.
[    0.121086] smp: Bringing up secondary CPUs ...
PU1: Booted secondary processor 0x0000000001 [0x410fd034]
[    0.165989] Detected VIPT I-cache on CPU2
[    0.166013] CPU2: Booted secondary processor 0x0000000002 [0x410fd034]
[    0.186002] Detected VIPT I-cache on CPU3
[    0.186020] CPU3: Booted secondary processor 0x0000000003 [0x410fd034]
[    0.207565] CPU features: enabling workaround for ARM erratum 858921
[    0.207575] Detected VIPT I-cache on CPU4
[    0.207583] CPU features: SANITY CHECK: Unexpected variation in SYS_ID_AA64MMFR0_EL1. Boot CPU: 0x00000000001122, CPU4: 0x00000000101122
[    0.207614] CPU features: Unsupported CPU feature variation detected.
[    0.207643] BUG: scheduling while atomic: swapper/4/0/0x00000002
[    0.207645] Modules linked in:
[    0.207652] CPU: 4 PID: 0 Comm: swapper/4 Tainted: G S                4.18.0-rc3-00017-g693350a79980 #17
[    0.207654] Hardware name: HiKey960 (DT)
[    0.207656] Call trace:
[    0.207670]  dump_backtrace+0x0/0x150
[    0.207674]  show_stack+0x14/0x20
[    0.207680]  dump_stack+0x90/0xbc
[    0.207685]  __schedule_bug+0x58/0x70
[    0.207690]  __schedule+0x51c/0x5b0
[    0.207693]  schedule+0x38/0xa0
[    0.207694]  schedule_timeout+0x23c/0x340
[    0.207698]  wait_for_common+0x140/0x170
[    0.207701]  wait_for_completion+0x14/0x20
[    0.207704]  __stop_cpus+0x68/0xa0
[    0.207706]  stop_cpus+0x40/0x60
[    0.207708]  stop_machine_cpuslocked+0xd4/0xf0
[    0.207711]  aarch64_insn_patch_text+0x48/0x70
[    0.207715]  arch_jump_label_transform+0x48/0x80
[    0.207718]  __jump_label_update+0xa4/0xf0
[    0.207721]  jump_label_update+0xe8/0x120
[    0.207724]  static_key_enable_cpuslocked+0x7c/0xc0
[    0.207730]  arch_timer_check_ool_workaround+0x190/0x1f0
[    0.207732]  arch_timer_starting_cpu+0xd8/0x290
[    0.207737]  cpuhp_invoke_callback+0x90/0x220
[    0.207739]  notify_cpu_starting+0x68/0xa0
[    0.207742]  secondary_start_kernel+0xe8/0x160
[    0.207754] bad: scheduling from the idle thread!
[    0.207756] CPU: 4 PID: 0 Comm: swapper/4 Tainted: G S      W         4.18.0-rc3-00017-g693350a79980 #17
[    0.207757] Hardware name: HiKey960 (DT)
[    0.207758] Call trace:
[    0.207761]  dump_backtrace+0x0/0x150
[    0.207764]  show_stack+0x14/0x20
[    0.207766]  dump_stack+0x90/0xbc
[    0.207769]  dequeue_task_idle+0x24/0x40
[    0.207771]  deactivate_task+0x70/0xc0
[    0.207774]  __schedule+0x2f8/0x5b0
[    0.207776]  schedule+0x38/0xa0
[    0.207778]  schedule_timeout+0x23c/0x340
[    0.207780]  wait_for_common+0x140/0x170
[    0.207783]  wait_for_completion+0x14/0x20
[    0.207784]  __stop_cpus+0x68/0xa0
[    0.207786]  stop_cpus+0x40/0x60
[    0.207788]  stop_machine_cpuslocked+0xd4/0xf0
[    0.207790]  aarch64_insn_patch_text+0x48/0x70
[    0.207792]  arch_jump_label_transform+0x48/0x80
[    0.207794]  __jump_label_update+0xa4/0xf0
[    0.207796]  jump_label_update+0xe8/0x120
[    0.207799]  static_key_enable_cpuslocked+0x7c/0xc0
[    0.207801]  arch_timer_check_ool_workaround+0x190/0x1f0
[    0.207804]  arch_timer_starting_cpu+0xd8/0x290
[    0.207806]  cpuhp_invoke_callback+0x90/0x220
[    0.207808]  notify_cpu_starting+0x68/0xa0
[    0.207810]  secondary_start_kernel+0xe8/0x160
[    0.207819] bad: scheduling from the idle thread!
[    0.207821] CPU: 4 PID: 0 Comm: swapper/4 Tainted: G S      W         4.18.0-rc3-00017-g693350a79980 #17
[    0.207822] Hardware name: HiKey960 (DT)
[    0.207823] Call trace:
[    0.207826]  dump_backtrace+0x0/0x150
[    0.207828]  show_stack+0x14/0x20
[    0.207830]  dump_stack+0x90/0xbc
[    0.207832]  dequeue_task_idle+0x24/0x40
[    0.207834]  deactivate_task+0x70/0xc0
[    0.207836]  __schedule+0x2f8/0x5b0
[    0.207839]  schedule+0x38/0xa0
[    0.207841]  schedule_timeout+0x23c/0x340
[    0.207843]  wait_for_common+0x140/0x170
[    0.207845]  wait_for_completion+0x14/0x20
[    0.207847]  __stop_cpus+0x68/0xa0
[    0.207848]  stop_cpus+0x40/0x60
[    0.207851]  stop_machine_cpuslocked+0xd4/0xf0
[    0.207853]  aarch64_insn_patch_text+0x48/0x70
[    0.207855]  arch_jump_label_transform+0x48/0x80
[    0.207857]  __jump_label_update+0xa4/0xf0
[    0.207859]  jump_label_update+0xe8/0x120
[    0.207861]  static_key_enable_cpuslocked+0x7c/0xc0
[    0.207864]  arch_timer_check_ool_workaround+0x190/0x1f0
[    0.207866]  arch_timer_starting_cpu+0xd8/0x290
[    0.207868]  cpuhp_invoke_callback+0x90/0x220
[    0.207870]  notify_cpu_starting+0x68/0xa0
[    0.207872]  secondary_start_kernel+0xe8/0x160
[    1.226177] CPU4: failed to come online
[    1.633614] CPU4: failed in unknown state : 0x0

Any idea what could be the problem?

Thanks in advance!
- Tuomas

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[leo.yan]

On Thu, Aug 16, 2018 at 02:38:51AM +0300, Tuomas Tynkkynen wrote:
> Hello,
> 
> Booting Linus's master of today with the usual arm64 defconfig fails for
> me on the Hikey960 board. I've bisected it down to:
> 
> commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
> Author: Will Deacon <will.deacon@arm.com>
> Date:   Tue Jun 19 17:55:28 2018 +0100
> 
>     arm64: insn: Don't fallback on nosync path for general insn patching
> 
> Here's the boot log with earlycon enabled after which the boot hangs:

[...]

> [    0.207583] CPU features: SANITY CHECK: Unexpected variation in SYS_ID_AA64MMFR0_EL1. Boot CPU: 0x00000000001122, CPU4: 0x00000000101122
> [    0.207614] CPU features: Unsupported CPU feature variation detected.
> [    0.207643] BUG: scheduling while atomic: swapper/4/0/0x00000002
> [    0.207645] Modules linked in:
> [    0.207652] CPU: 4 PID: 0 Comm: swapper/4 Tainted: G S                4.18.0-rc3-00017-g693350a79980 #17
> [    0.207654] Hardware name: HiKey960 (DT)
> [    0.207656] Call trace:
> [    0.207670]  dump_backtrace+0x0/0x150
> [    0.207674]  show_stack+0x14/0x20
> [    0.207680]  dump_stack+0x90/0xbc
> [    0.207685]  __schedule_bug+0x58/0x70
> [    0.207690]  __schedule+0x51c/0x5b0
> [    0.207693]  schedule+0x38/0xa0
> [    0.207694]  schedule_timeout+0x23c/0x340
> [    0.207698]  wait_for_common+0x140/0x170
> [    0.207701]  wait_for_completion+0x14/0x20
> [    0.207704]  __stop_cpus+0x68/0xa0
> [    0.207706]  stop_cpus+0x40/0x60
> [    0.207708]  stop_machine_cpuslocked+0xd4/0xf0
> [    0.207711]  aarch64_insn_patch_text+0x48/0x70
> [    0.207715]  arch_jump_label_transform+0x48/0x80
> [    0.207718]  __jump_label_update+0xa4/0xf0
> [    0.207721]  jump_label_update+0xe8/0x120
> [    0.207724]  static_key_enable_cpuslocked+0x7c/0xc0
> [    0.207730]  arch_timer_check_ool_workaround+0x190/0x1f0
> [    0.207732]  arch_timer_starting_cpu+0xd8/0x290
> [    0.207737]  cpuhp_invoke_callback+0x90/0x220
> [    0.207739]  notify_cpu_starting+0x68/0xa0
> [    0.207742]  secondary_start_kernel+0xe8/0x160

From this part log, it's quite likely related with CPU4 (CA73) with
arch timer workaround ERRATUM_858921, before I reported another
similiar lockup issue: https://lkml.org/lkml/2017/7/24/346.

Thanks,
Leo Yan

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[Will Deacon]

Hi Tuomas, [+John]

On Thu, Aug 16, 2018 at 02:38:51AM +0300, Tuomas Tynkkynen wrote:
> Booting Linus's master of today with the usual arm64 defconfig fails for
> me on the Hikey960 board. I've bisected it down to:
> 
> commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
> Author: Will Deacon <will.deacon@arm.com>
> Date:   Tue Jun 19 17:55:28 2018 +0100
> 
>     arm64: insn: Don't fallback on nosync path for general insn patching
> 
> Here's the boot log with earlycon enabled after which the boot hangs:

Thanks for the boot log, that's really helpful. Please can you try the diff
below?

Out of interest -- do you know if Hikey960 is used by any boot-testing
farms? The patch above has been in linux-next for ages, but we didn't see
any complaints until it hit mainline :(

Cheers,

Will

--->8

diff --git a/arch/arm64/kernel/jump_label.c b/arch/arm64/kernel/jump_label.c
index c2dd1ad3e648..e0756416e567 100644
--- a/arch/arm64/kernel/jump_label.c
+++ b/arch/arm64/kernel/jump_label.c
@@ -36,7 +36,7 @@ void arch_jump_label_transform(struct jump_entry *entry,
 		insn = aarch64_insn_gen_nop();
 	}
 
-	aarch64_insn_patch_text(&addr, &insn, 1);
+	aarch64_insn_patch_text_nosync(addr, insn);
 }
 
 void arch_jump_label_transform_static(struct jump_entry *entry,

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[Valentin Schneider]

Hi,

On 16/08/18 11:22, Will Deacon wrote:
> Hi Tuomas, [+John]
> 
> On Thu, Aug 16, 2018 at 02:38:51AM +0300, Tuomas Tynkkynen wrote:
>> Booting Linus's master of today with the usual arm64 defconfig fails for
>> me on the Hikey960 board. I've bisected it down to:
>>
>> commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
>> Author: Will Deacon <will.deacon@arm.com>
>> Date:   Tue Jun 19 17:55:28 2018 +0100
>>
>>     arm64: insn: Don't fallback on nosync path for general insn patching
>>
>> Here's the boot log with earlycon enabled after which the boot hangs:
> 
> Thanks for the boot log, that's really helpful. Please can you try the diff
> below?
> 

I get the same "BUG: scheduling while atomic" when running the current master
(f91e654474d413201ae578820fb63f8a811f6c4e), and confirmed that I don't get
it on my setup when checking out the direct parent of 693350a7998018391852c48f68956cf0f855b2b9.

Tried out the diff and the board doesn't seem to die anymore, so FWIW:

Tested-by: Valentin Schneider <valentin.schneider@arm.com>

> Out of interest -- do you know if Hikey960 is used by any boot-testing
> farms? The patch above has been in linux-next for ages, but we didn't see
> any complaints until it hit mainline :(
> 
> Cheers,
> 
> Will
> 
> --->8
> 
> diff --git a/arch/arm64/kernel/jump_label.c b/arch/arm64/kernel/jump_label.c
> index c2dd1ad3e648..e0756416e567 100644
> --- a/arch/arm64/kernel/jump_label.c
> +++ b/arch/arm64/kernel/jump_label.c
> @@ -36,7 +36,7 @@ void arch_jump_label_transform(struct jump_entry *entry,
>  		insn = aarch64_insn_gen_nop();
>  	}
>  
> -	aarch64_insn_patch_text(&addr, &insn, 1);
> +	aarch64_insn_patch_text_nosync(addr, insn);
>  }
>  
>  void arch_jump_label_transform_static(struct jump_entry *entry,
> 

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[John Stultz]

On Thu, Aug 16, 2018 at 3:22 AM, Will Deacon <will.deacon@arm.com> wrote:
> Hi Tuomas, [+John]
>
> On Thu, Aug 16, 2018 at 02:38:51AM +0300, Tuomas Tynkkynen wrote:
>> Booting Linus's master of today with the usual arm64 defconfig fails for
>> me on the Hikey960 board. I've bisected it down to:
>>
>> commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
>> Author: Will Deacon <will.deacon@arm.com>
>> Date:   Tue Jun 19 17:55:28 2018 +0100
>>
>>     arm64: insn: Don't fallback on nosync path for general insn patching
>>
>> Here's the boot log with earlycon enabled after which the boot hangs:
>
> Thanks for the boot log, that's really helpful. Please can you try the diff
> below?
>
> Out of interest -- do you know if Hikey960 is used by any boot-testing
> farms? The patch above has been in linux-next for ages, but we didn't see
> any complaints until it hit mainline :(

Unfortunately not yet. HiKey960 upstreaming is ongoing, and the UFS
support patches should land in the 4.19 merge window.
After which, it will enable more mainline testing as folks won't have
to carry patches just to get the system to boot.

> diff --git a/arch/arm64/kernel/jump_label.c b/arch/arm64/kernel/jump_label.c
> index c2dd1ad3e648..e0756416e567 100644
> --- a/arch/arm64/kernel/jump_label.c
> +++ b/arch/arm64/kernel/jump_label.c
> @@ -36,7 +36,7 @@ void arch_jump_label_transform(struct jump_entry *entry,
>                 insn = aarch64_insn_gen_nop();
>         }
>
> -       aarch64_insn_patch_text(&addr, &insn, 1);
> +       aarch64_insn_patch_text_nosync(addr, insn);
>  }
>
>  void arch_jump_label_transform_static(struct jump_entry *entry,

Yes, this works for me as well.

Tested-by: John Stultz <john.stultz@linaro.org>

thanks so much!
-john

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[Tuomas Tynkkynen]

Hi,

On 08/16/2018 01:22 PM, Will Deacon wrote:
> Hi Tuomas, [+John]
> 
> On Thu, Aug 16, 2018 at 02:38:51AM +0300, Tuomas Tynkkynen wrote:
>> Booting Linus's master of today with the usual arm64 defconfig fails for
>> me on the Hikey960 board. I've bisected it down to:
>>
>> commit 693350a7998018391852c48f68956cf0f855b2b9 (HEAD, refs/bisect/bad)
>> Author: Will Deacon <will.deacon@arm.com>
>> Date:   Tue Jun 19 17:55:28 2018 +0100
>>
>>      arm64: insn: Don't fallback on nosync path for general insn patching
>>
>> Here's the boot log with earlycon enabled after which the boot hangs:
> 
> Thanks for the boot log, that's really helpful. Please can you try the diff
> below?
> 

Thanks for the quick response, that indeed fixes it. If you need more tags:

Reported-by: Tuomas Tynkkynen <tuomas@tuxera.com>
Tested-by: Tuomas Tynkkynen <tuomas@tuxera.com>

- Tuomas

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[Tuomas Tynkkynen]

Hi,

On 08/16/2018 08:57 PM, John Stultz wrote:
> On Thu, Aug 16, 2018 at 3:22 AM, Will Deacon <will.deacon@arm.com> wrote:
>> Hi Tuomas, [+John]
...
>>
>> Out of interest -- do you know if Hikey960 is used by any boot-testing
>> farms? The patch above has been in linux-next for ages, but we didn't see
>> any complaints until it hit mainline :(
> 
> Unfortunately not yet. HiKey960 upstreaming is ongoing, and the UFS
> support patches should land in the 4.19 merge window.
> After which, it will enable more mainline testing as folks won't have
> to carry patches just to get the system to boot.
> 

A microSD rootfs gets me a working system on e.g. v4.17... except that
even there the system hangs if I don't have 'clk_ignore_unused' on
the kernel command line. I wonder if this is a known problem?

- Tuomas

Re: AArch64 boot failure on Hikey960 on latest master after "arm64: insn: Don't fallback on nosync path for general insn patching"

[John Stultz]

On Thu, Aug 16, 2018 at 11:37 AM, Tuomas Tynkkynen
<tuomas.tynkkynen@iki.fi> wrote:
> Hi,
>
> On 08/16/2018 08:57 PM, John Stultz wrote:
>>
>> On Thu, Aug 16, 2018 at 3:22 AM, Will Deacon <will.deacon@arm.com> wrote:
>>>
>>> Hi Tuomas, [+John]
>
> ...
>>>
>>>
>>> Out of interest -- do you know if Hikey960 is used by any boot-testing
>>> farms? The patch above has been in linux-next for ages, but we didn't see
>>> any complaints until it hit mainline :(
>>
>>
>> Unfortunately not yet. HiKey960 upstreaming is ongoing, and the UFS
>> support patches should land in the 4.19 merge window.
>> After which, it will enable more mainline testing as folks won't have
>> to carry patches just to get the system to boot.
>>
>
> A microSD rootfs gets me a working system on e.g. v4.17... except that
> even there the system hangs if I don't have 'clk_ignore_unused' on
> the kernel command line. I wonder if this is a known problem?

Oh, good point. I'm usually focusing on AOSP, so I've not tried
booting from the uSD slot.

As for the  clk_ignore_unused, I've not seen it, but I'm usually
testing with a much larger patch stack targeting AOSP (only
occasionally trying with a smaller set to validate patches being
prepped for upstream).  So if you want to file a bug, please do so
here: https://bugs.96boards.org/

Appreciate your testing! Please do feel free to reach out/cc me if you
run into other issues.

thanks
-john