While booting linux-next [next-20190603] on a POWER9 LPAR following
BUG is encountered and the boot fails.
If I revert the following 2 patches I no longer see this BUG message
07031d37b2f9 ( mm/vmalloc.c: switch to WARN_ON() and move it under unlink_va() )
728e0fbf263e ( mm/vmalloc.c: get rid of one single unlink_va() when merge )
[ 1.130734] ------------[ cut here ]------------
[ 1.130745] kernel BUG at mm/vmalloc.c:470!
[ 1.130753] Oops: Exception in kernel mode, sig: 5 [#1]
[ 1.130761] LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
[ 1.130768] Modules linked in: ibmvscsi(+) ibmveth scsi_transport_srp
[ 1.130781] CPU: 23 PID: 127 Comm: kworker/23:0 Not tainted 5.2.0-rc3-next-20190603 #1
[ 1.130796] Workqueue: events do_free_init
[ 1.130803] NIP: c0000000003bcc48 LR: c0000000003bcbc4 CTR: c0000000003bc860
[ 1.130812] REGS: c0000018ec7877e0 TRAP: 0700 Not tainted (5.2.0-rc3-next-20190603)
[ 1.130820] MSR: 800000010282b033 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE,TM[E]> CR: 24008888 XER: 20040005
[ 1.130837] CFAR: c0000000003bcc6c IRQMASK: 0
[ 1.130837] GPR00: c0000000003bd0e4 c0000018ec787a70 c000000001929400 0000000000000001
[ 1.130837] GPR04: c0000018e8d2e8f0 c0000018e8d2e8f0 c008000006570000 c008000006550000
[ 1.130837] GPR08: c008000009670000 c0000018e8d0e098 c0000018e8d0e5f8 c0000018e8ca93f0
[ 1.130837] GPR12: c0000000003bc860 c00000001ec4e200 c000000000160188 c0000018f4044d00
[ 1.130837] GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
[ 1.130837] GPR20: 0000000000000000 0000000000000000 c000000001485108 c0000000019620d8
[ 1.130837] GPR24: c000000001962464 0000000000000800 0000000000000001 0000000000001800
[ 1.130837] GPR28: c000000001ccc0a8 c000000001ccc080 c0000018e8d2e8b8 c0000018e8d2e8d8
[ 1.130899] NIP [c0000000003bcc48] __free_vmap_area+0xf8/0x480
[ 1.130906] LR [c0000000003bcbc4] __free_vmap_area+0x74/0x480
[ 1.130912] Call Trace:
[ 1.130916] [c0000018ec787a70] [c0000000003bcf70] __free_vmap_area+0x420/0x480 (unreliable)
[ 1.130924] [c0000018ec787ac0] [c0000000003bd0e4] __purge_vmap_area_lazy+0x114/0x1e0
[ 1.130932] [c0000018ec787b10] [c0000000003bef44] _vm_unmap_aliases+0x1a4/0x210
[ 1.130939] [c0000018ec787b90] [c0000000003c1c48] __vunmap+0xe8/0x220
[ 1.130946] [c0000018ec787c20] [c00000000021102c] module_memfree+0x3c/0x50
[ 1.130953] [c0000018ec787c40] [c0000000002110ac] do_free_init+0x6c/0xa0
[ 1.130964] [c0000018ec787c70] [c000000000156df0] process_one_work+0x260/0x520
[ 1.130976] [c0000018ec787d10] [c000000000157138] worker_thread+0x88/0x5f0
[ 1.130985] [c0000018ec787db0] [c000000000160328] kthread+0x1a8/0x1b0
[ 1.130996] [c0000018ec787e20] [c00000000000ba54] ret_from_kernel_thread+0x5c/0x68
[ 1.131004] Instruction dump:
[ 1.131011] e9292c90 2fa90000 419e0378 e8fe0000 e8de0008 60000000 e909ffe8 e949ffe0
[ 1.131021] 7fa74040 409c0014 7faa3040 409c002c <0fe00000> 60000000 7faa3040 409cfff4
[ 1.131032] ---[ end trace b0b43434aedbb78e ]—
Have attached the boot log for reference.
Thanks
-Sachin
Hi Sachin,
On Tue, 4 Jun 2019 14:45:43 +0530 Sachin Sant <[email protected]> wrote:
>
> While booting linux-next [next-20190603] on a POWER9 LPAR following
> BUG is encountered and the boot fails.
>
> If I revert the following 2 patches I no longer see this BUG message
>
> 07031d37b2f9 ( mm/vmalloc.c: switch to WARN_ON() and move it under unlink_va() )
> 728e0fbf263e ( mm/vmalloc.c: get rid of one single unlink_va() when merge )
This latter patch has been fixed in today's linux-next ...
--
Cheers,
Stephen Rothwell
> On 04-Jun-2019, at 3:59 PM, Stephen Rothwell <[email protected]> wrote:
>
> Hi Sachin,
>
> On Tue, 4 Jun 2019 14:45:43 +0530 Sachin Sant <[email protected]> wrote:
>>
>> While booting linux-next [next-20190603] on a POWER9 LPAR following
>> BUG is encountered and the boot fails.
>>
>> If I revert the following 2 patches I no longer see this BUG message
>>
>> 07031d37b2f9 ( mm/vmalloc.c: switch to WARN_ON() and move it under unlink_va() )
>> 728e0fbf263e ( mm/vmalloc.c: get rid of one single unlink_va() when merge )
>
> This latter patch has been fixed in today's linux-next …
Thanks Stephen.
With today’s next (20190604) I no longer see this issue.
Thanks
-Sachin
Hi Sachin,
On Tue, 4 Jun 2019 19:09:26 +0530 Sachin Sant <[email protected]> wrote:
>
> With today’s next (20190604) I no longer see this issue.
Excellent, thanks for verifying.
--
Cheers,
Stephen Rothwell