sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Anatoly Pugachev]

Hello!

Trying to boot latest git kernel 4.14.0-12995-g0c86a6bd85ff does not
boot on my sparc64 LDOM [1], kernel config [2].

If I revert latest git sparc tree merge, with

$ git show 1deab8ce2c91e3b16563b7a7ea150f82334262ec --oneline
1deab8ce2c91 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc

mator@ttip:linux-2.6$ git revert -n -m 1
1deab8ce2c91e3b16563b7a7ea150f82334262ec

it is able to boot successfully.

Can you please teach me on how to git bisect merge commit? Thanks.

[1]. http://paste.debian.net/plain/997067
[2]. http://paste.debian.net/plain/997071

LDOM console boot log (first 100 lines. Full log in the URL[1] above):

Loading Linux 4.14.0-12995-g0c86a6bd85ff ...
Loading initial ramdisk ...

[    0.000017] PROMLIB: Sun IEEE Boot Prom 'OBP 4.38.8 2017/02/22 13:51'
[    0.000028] PROMLIB: Root node compatible: sun4v
[    0.000070] Linux version 4.14.0-12995-g0c86a6bd85ff (mator@ttip)
(gcc version 7.2.0 (Debian 7.2.0-16)) #523 SMP Wed Nov 22 18:03:05 MSK
2017
[    0.000358] bootconsole [earlyprom0] enabled
[    0.000404] ARCH: SUN4V
[    0.000452] Ethernet address: 00:14:4f:fa:06:f2
[    0.000504] MM: PAGE_OFFSET is 0xfff8000000000000 (max_phys_bits = 47)
[    0.000568] MM: VMALLOC [0x0000000100000000 --> 0x0006000000000000]
[    0.000630] MM: VMEMMAP [0x0006000000000000 --> 0x000c000000000000]
[    0.012719] Kernel: Using 5 locked TLB entries for main kernel image.
[    0.012787] Remapping the kernel...
[    0.012816] done.
[    0.020800] kmemleak: Kernel memory leak detector disabled
[    0.044638] OF stdout device is: /virtual-devices@100/console@1
[    0.044700] PROM: Built device tree with 74487 bytes of memory.
[    0.044758] MDESC: Size is 30976 bytes.
[    0.044865] PLATFORM: banner-name [SPARC T5-2]
[    0.044912] PLATFORM: name [ORCL,SPARC-T5-2]
[    0.044958] PLATFORM: hostid [84fa06f2]
[    0.044997] PLATFORM: serial# [0035260e]
[    0.045041] PLATFORM: stick-frequency [3b9aca00]
[    0.045086] PLATFORM: mac-address [144ffa06f2]
[    0.045132] PLATFORM: watchdog-resolution [1000 ms]
[    0.045181] PLATFORM: watchdog-max-timeout [31536000000 ms]
[    0.045236] PLATFORM: max-cpus [1024]
[    0.045324] Top of RAM: 0x40bf8b4000, Total RAM: 0x7ff37c000
[    0.045380] Memory hole size: 232453MB
[    0.168499] Allocated 40960 bytes for kernel page tables.
[    0.168615] Zone ranges:
[    0.168657]   Normal   [mem 0x0000000040400000-0x00000040bf8b3fff]
[    0.168724] Movable zone start for each node
[    0.168771] Early memory node ranges
[    0.168809]   node   0: [mem 0x0000000040400000-0x000000007febffff]
[    0.168869]   node   0: [mem 0x000000007ff40000-0x000000007ff47fff]
[    0.168927]   node   0: [mem 0x0000000080000000-0x000000077fffffff]
[    0.168988]   node   0: [mem 0x0000002000000000-0x000000202fffffff]
[    0.169045]   node   0: [mem 0x0000004030000000-0x00000040bf8b3fff]
[    0.169107] Initmem setup node 0 [mem 0x0000000040400000-0x00000040bf8b3fff]
[    0.169200] Unable to handle kernel NULL pointer dereference
[    0.169258] tsk->{mm,active_mm}->context = 0000000000000000
[    0.169311] tsk->{mm,active_mm}->pgd = fff8004030002000
[    0.169364]               \|/ ____ \|/
[    0.169364]               "@'/ .. \`@"
[    0.169364]               /_| \__/ |_\
[    0.169364]                  \__U_/
[    0.169488] swapper(0): Oops [#1]
[    0.169535] CPU: 0 PID: 0 Comm: swapper Not tainted
4.14.0-12995-g0c86a6bd85ff #523
[    0.169607] task: 0000000000c06c80 task.stack: 0000000000be8000
[    0.169663] TSTATE: 0000000080e01600 TPC: 00000000009f8c04 TNPC:
00000000009f8c08 Y: 0000000f    Not tainted
[    0.169756] TPC: <minmax_running_max+0x4/0xa0>
[    0.169801] g0: 0000000000000000 g1: 0000000001fb4000 g2:
0000000000080000 g3: 0000000001512730
[    0.169879] g4: 0000000000c06c80 g5: 0000000000000000 g6:
0000000000be8000 g7: 0000000000d16400
[    0.169957] o0: 0000000000cac148 o1: 0000000000b701f0 o2:
00000000014c8970 o3: 0000000000000000
[    0.170037] o4: 000000000000000e o5: 00000000014b9c00 sp:
0000000000beb0b1 ret_pc: 00000000004cfc74
[    0.170121] RPC: <lockdep_init_map+0x14/0x40>
[    0.170166] l0: 000000000000000e l1: 0000000000c29c00 l2:
0000000000000000 l3: 0000000000000002
[    0.170243] l4: 000000000000002d l5: 000000000000000c l6:
000000000000001b l7: 0000000000c571d0
[    0.171323] i0: 0000000000000018 i1: 0000000000b701f0 i2:
00000000014c8970 i3: 0000000000000000
[    0.171400] i4: 000000000000000e i5: 00000000c0000000 i6:
0000000000beb171 i7: 00000000005cd34c
[    0.171482] I7: <zone_batchsize.isra.7+0x4c/0x80>
[    0.171528] Call Trace:
[    0.171560]  [00000000005cd34c] zone_batchsize.isra.7+0x4c/0x80
[    0.171620]  [0000000000cfb8f8] free_area_init_node+0x4ac/0x51c
[    0.171681]  [0000000000cdc968] free_area_init_nodes+0x548/0x574
[    0.171744]  [0000000000ccef1c] paging_init+0xf7c/0xf98
[    0.171796]  [0000000000cc96bc] setup_arch+0x284/0x708
[    0.171848]  [0000000000cc6718] start_kernel+0x60/0x4c8
[    0.171900]  [0000000000cc9394] start_early_boot+0x2c8/0x2d8
[    0.171958]  [0000000000a08500] tlb_fixup_done+0x4c/0x6c
[    0.172011]  [000000000011d8a4] 0x11d8a4
[    0.172053] Disabling lock debugging due to kernel taint
[    0.172103] Caller[00000000005cd34c]: zone_batchsize.isra.7+0x4c/0x80
[    0.172162] Caller[0000000000cfb8f8]: free_area_init_node+0x4ac/0x51c
[    0.172223] Caller[0000000000cdc968]: free_area_init_nodes+0x548/0x574
[    0.172282] Caller[0000000000ccef1c]: paging_init+0xf7c/0xf98
[    0.172335] Caller[0000000000cc96bc]: setup_arch+0x284/0x708
[    0.172388] Caller[0000000000cc6718]: start_kernel+0x60/0x4c8
[    0.172443] Caller[0000000000cc9394]: start_early_boot+0x2c8/0x2d8
[    0.172499] Caller[0000000000a08500]: tlb_fixup_done+0x4c/0x6c
[    0.172553] Caller[000000000011d8a4]: 0x11d8a4
[    0.172597] Instruction DUMP:
[    0.172600]  01000000
[    0.172632]  01000000
[    0.172658]  9de3bf40
[    0.172685] <c2062004>
[    0.172712]  f627a7fb
[    0.172737]  80a0401b
[    0.172764]  08400007
[    0.172792]  f427a7f7
[    0.172818]  c2062010
[    0.172845]
[    0.172890] Kernel panic - not syncing: Attempted to kill the idle task!
[    0.173132] Unable to handle kernel NULL pointer dereference
[    0.173186] tsk->{mm,active_mm}->context = 0000000000000000
[    0.173238] tsk->{mm,active_mm}->pgd = fff8004030002000

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Anatoly Pugachev]

On Wed, Nov 22, 2017 at 6:40 PM, Anatoly Pugachev <matorola@gmail.com> wrote:
> Hello!
>
> Trying to boot latest git kernel 4.14.0-12995-g0c86a6bd85ff does not
> boot on my sparc64 LDOM [1], kernel config [2].

4.15.0-rc1 a bit different OOPS, but still does not boot:

Loading Linux 4.15.0-rc1 ...
Loading initial ramdisk ...

[    0.000018] PROMLIB: Sun IEEE Boot Prom 'OBP 4.38.8 2017/02/22 13:51'
[    0.000030] PROMLIB: Root node compatible: sun4v
[    0.000074] Linux version 4.15.0-rc1 (mator@ttip) (gcc version
7.2.0 (Debian 7.2.0-16)) #527 SMP Mon Nov 27 12:44:52 MSK 2017
[    0.000367] bootconsole [earlyprom0] enabled
[    0.000414] ARCH: SUN4V
[    0.000467] Ethernet address: 00:14:4f:fa:06:f2
[    0.000524] MM: PAGE_OFFSET is 0xfff8000000000000 (max_phys_bits = 47)
[    0.000591] MM: VMALLOC [0x0000000100000000 --> 0x0006000000000000]
[    0.001681] MM: VMEMMAP [0x0006000000000000 --> 0x000c000000000000]
[    0.014136] Kernel: Using 5 locked TLB entries for main kernel image.
[    0.014224] Remapping the kernel...
[    0.014254] done.
[    0.022778] kmemleak: Kernel memory leak detector disabled
[    0.048367] OF stdout device is: /virtual-devices@100/console@1
[    0.048447] PROM: Built device tree with 74487 bytes of memory.
[    0.048513] MDESC: Size is 30976 bytes.
[    0.048634] PLATFORM: banner-name [SPARC T5-2]
[    0.048687] PLATFORM: name [ORCL,SPARC-T5-2]
[    0.048733] PLATFORM: hostid [84fa06f2]
[    0.048775] PLATFORM: serial# [0035260e]
[    0.048816] PLATFORM: stick-frequency [3b9aca00]
[    0.048864] PLATFORM: mac-address [144ffa06f2]
[    0.048911] PLATFORM: watchdog-resolution [1000 ms]
[    0.048964] PLATFORM: watchdog-max-timeout [31536000000 ms]
[    0.049022] PLATFORM: max-cpus [1024]
[    0.049131] Top of RAM: 0x40bf8b4000, Total RAM: 0x7ff37c000
[    0.049195] Memory hole size: 232453MB
[    0.173656] Allocated 40960 bytes for kernel page tables.
[    0.173782] Zone ranges:
[    0.173826]   Normal   [mem 0x0000000040400000-0x00000040bf8b3fff]
[    0.173898] Movable zone start for each node
[    0.174970] Early memory node ranges
[    0.175029]   node   0: [mem 0x0000000040400000-0x000000007febffff]
[    0.175100]   node   0: [mem 0x000000007ff40000-0x000000007ff47fff]
[    0.175161]   node   0: [mem 0x0000000080000000-0x000000077fffffff]
[    0.175221]   node   0: [mem 0x0000002000000000-0x000000202fffffff]
[    0.176318]   node   0: [mem 0x0000004030000000-0x00000040bf8b3fff]
[    0.176400] Initmem setup node 0 [mem 0x0000000040400000-0x00000040bf8b3fff]
[    0.176522] Kernel unaligned access at TPC[9f8a10]
minmax_subwin_update+0x30/0x100
[    0.176608] Unable to handle kernel NULL pointer dereference in mna handler
[    0.176614]  at virtual address 0000000000000009
[    0.176728] current->{active_,}mm->context = 0000000000000000
[    0.176787] current->{active_,}mm->pgd = fff8004030002000
[    0.176844]               \|/ ____ \|/
[    0.176844]               "@'/ .. \`@"
[    0.176844]               /_| \__/ |_\
[    0.176844]                  \__U_/
[    0.178015] swapper(0): Oops [#1]
[    0.178070] CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1 #527
[    0.178133] task: 0000000000c06c80 task.stack: 0000000000be8000
[    0.178190] TSTATE: 0000000080e01607 TPC: 00000000009f8a10 TNPC:
00000000009f8a14 Y: 0000000f    Not tainted
[    0.179325] TPC: <minmax_subwin_update+0x30/0x100>
[    0.179388] g0: 000000000000002d g1: 0000000001fb4000 g2:
0000000000080000 g3: 0000000001512630
[    0.179468] g4: 0000000000c06c80 g5: 0000000000000000 g6:
0000000000be8000 g7: 0000000000d16400
[    0.180578] o0: 0000000000000018 o1: 0000000000b702c8 o2:
00000000014c8968 o3: 0000000000000000
[    0.180681] o4: 000000000000000e o5: 00000000c0000000 sp:
0000000000beb171 ret_pc: 00000000005cd1cc
[    0.180773] RPC: <zone_batchsize.isra.7+0x4c/0x80>
[    0.180824] l0: 0000000000800000 l1: 0000000000d6d000 l2:
0000000000000000 l3: 0000000000cbc2c8
[    0.181938] l4: 0000000000bafe98 l5: 00000000fef195e1 l6:
0000000000c07538 l7: 09740cf2593a0d3c
[    0.182033] i0: 0000000000000001 i1: 0000000000b702c8 i2:
00000000014c8968 i3: 0000000000bebb00
[    0.182113] i4: 0000000000bebb08 i5: 0000000000000000 i6:
0000000000beb221 i7: 0000000000cfb8d8
[    0.183234] I7: <free_area_init_node+0x4ac/0x51c>
[    0.183299] Call Trace:
[    0.183334]  [0000000000cfb8d8] free_area_init_node+0x4ac/0x51c
[    0.183396]  [0000000000cdc960] free_area_init_nodes+0x548/0x574
[    0.184488]  [0000000000ccef1c] paging_init+0xf7c/0xf98
[    0.184557]  [0000000000cc96bc] setup_arch+0x284/0x708
[    0.184615]  [0000000000cc6718] start_kernel+0x60/0x4c8
[    0.184669]  [0000000000cc9394] start_early_boot+0x2c8/0x2d8
[    0.184731]  [0000000000a08400] tlb_fixup_done+0x4c/0x6c
[    0.185821]  [000000000011d8a4] 0x11d8a4
[    0.185878] Disabling lock debugging due to kernel taint
[    0.185945] Caller[0000000000cfb8d8]: free_area_init_node+0x4ac/0x51c
[    0.186012] Caller[0000000000cdc960]: free_area_init_nodes+0x548/0x574
[    0.187103] Caller[0000000000ccef1c]: paging_init+0xf7c/0xf98
[    0.187174] Caller[0000000000cc96bc]: setup_arch+0x284/0x708
[    0.187234] Caller[0000000000cc6718]: start_kernel+0x60/0x4c8
[    0.187289] Caller[0000000000cc9394]: start_early_boot+0x2c8/0x2d8
[    0.188377] Caller[0000000000a08400]: tlb_fixup_done+0x4c/0x6c
[    0.188448] Caller[000000000011d8a4]: 0x11d8a4
[    0.188495] Instruction DUMP:
[    0.188499]  fa062014
[    0.188531]  c4260000
[    0.188558]  c2262004
[    0.188586] <f8262008>
[    0.188614]  fa26200c
[    0.189669]  ce068000
[    0.189710]  ce262010
[    0.189739]  c606a004
[    0.189769]  c6262014
[    0.189798]
[    0.189845] Kernel panic - not syncing: Attempted to kill the idle task!

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Meelis Roos]

> [    0.176608] Unable to handle kernel NULL pointer dereference in mna handler
> [    0.176614]  at virtual address 0000000000000009
> [    0.176728] current->{active_,}mm->context = 0000000000000000
> [    0.176787] current->{active_,}mm->pgd = fff8004030002000


> [    0.188495] Instruction DUMP:
> [    0.188499]  fa062014
> [    0.188531]  c4260000
> [    0.188558]  c2262004
> [    0.188586] <f8262008>

I am seeing similar problem on V480, with kernel crosscompiled from 
server with gcc (and other up to date userland from Debian sparc64).

Sometimes it happesn after I change a configuration option (like rmeove 
cassini from being built in) and does not go away when I change the 
option back.

-- 
Meelis Roos (mroos@linux.ee)

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Meelis Roos <mroos@linux.ee>
Date: Mon, 27 Nov 2017 13:37:07 +0200 (EET)

>> [    0.176608] Unable to handle kernel NULL pointer dereference in mna handler
>> [    0.176614]  at virtual address 0000000000000009
>> [    0.176728] current->{active_,}mm->context = 0000000000000000
>> [    0.176787] current->{active_,}mm->pgd = fff8004030002000
> 
> 
>> [    0.188495] Instruction DUMP:
>> [    0.188499]  fa062014
>> [    0.188531]  c4260000
>> [    0.188558]  c2262004
>> [    0.188586] <f8262008>
> 
> I am seeing similar problem on V480, with kernel crosscompiled from 
> server with gcc (and other up to date userland from Debian sparc64).
> 
> Sometimes it happesn after I change a configuration option (like rmeove 
> cassini from being built in) and does not go away when I change the 
> option back.

If someone can bisect this it would help tremendously.

I've been away from my sparc64 systems for a while and I hope I can get
started working on these regressions again soon.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Anatoly Pugachev]

On Mon, Nov 27, 2017 at 1:20 PM, Anatoly Pugachev <matorola@gmail.com> wrote:
> On Wed, Nov 22, 2017 at 6:40 PM, Anatoly Pugachev <matorola@gmail.com> wrote:
>> Hello!
>>
>> Trying to boot latest git kernel 4.14.0-12995-g0c86a6bd85ff does not
>> boot on my sparc64 LDOM [1], kernel config [2].
>
> 4.15.0-rc1 a bit different OOPS, but still does not boot:

tried with debian sparc64 default kernel config file, and git kernel
v4.15-rc1-24-g43570f0383d6
still this LDOM doesn't boot. Kernel boot log is below.

Going to try git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc.git
to boot and bisect...

Loading Linux 4.15.0-rc1+ ...
Loading initial ramdisk ...

[    0.000014] PROMLIB: Sun IEEE Boot Prom 'OBP 4.38.8 2017/02/22 13:51'
[    0.000018] PROMLIB: Root node compatible: sun4v
[    0.000050] Linux version 4.15.0-rc1+ (mator@ttip) (gcc version
7.2.0 (Debian 7.2.0-16)) #528 SMP Wed Nov 29 11:28:51 MSK 2017
[    0.000304] bootconsole [earlyprom0] enabled
[    0.000344] ARCH: SUN4V
[    0.000386] Ethernet address: 00:14:4f:fa:06:f2
[    0.000434] MM: PAGE_OFFSET is 0xfff8000000000000 (max_phys_bits = 47)
[    0.000494] MM: VMALLOC [0x0000000100000000 --> 0x0006000000000000]
[    0.000551] MM: VMEMMAP [0x0006000000000000 --> 0x000c000000000000]
[    0.010204] Kernel: Using 3 locked TLB entries for main kernel image.
[    0.010274] Remapping the kernel...
[    0.010294] done.
[    0.040080] OF stdout device is: /virtual-devices@100/console@1
[    0.040150] PROM: Built device tree with 68631 bytes of memory.
[    0.040212] MDESC: Size is 30976 bytes.
[    0.040310] PLATFORM: banner-name [SPARC T5-2]
[    0.040355] PLATFORM: name [ORCL,SPARC-T5-2]
[    0.040395] PLATFORM: hostid [84fa06f2]
[    0.040432] PLATFORM: serial# [0035260e]
[    0.040469] PLATFORM: stick-frequency [3b9aca00]
[    0.040512] PLATFORM: mac-address [144ffa06f2]
[    0.040553] PLATFORM: watchdog-resolution [1000 ms]
[    0.040599] PLATFORM: watchdog-max-timeout [31536000000 ms]
[    0.040649] PLATFORM: max-cpus [1024]
[    0.040747] Top of RAM: 0x40bf8b8000, Total RAM: 0x7ff380000
[    0.040804] Memory hole size: 232453MB
[    0.068288] Allocated 40960 bytes for kernel page tables.
[    0.068391] Zone ranges:
[    0.068428]   Normal   [mem 0x0000000040400000-0x00000040bf8b7fff]
[    0.068491] Movable zone start for each node
[    0.068532] Early memory node ranges
[    0.068568]   node   0: [mem 0x0000000040400000-0x000000007febffff]
[    0.068624]   node   0: [mem 0x000000007ff40000-0x000000007ff47fff]
[    0.068681]   node   0: [mem 0x0000000080000000-0x000000077fffffff]
[    0.068738]   node   0: [mem 0x0000002000000000-0x000000202fffffff]
[    0.068794]   node   0: [mem 0x0000004030000000-0x00000040bf8b7fff]
[    0.068851] Initmem setup node 0 [mem 0x0000000040400000-0x00000040bf8b7fff]
[    0.068940] Kernel unaligned access at TPC[9b75bc] secpath_set+0x3c/0x80
[    0.069009] Unable to handle kernel NULL pointer dereference in mna handler
[    0.069011]  at virtual address 0000000000000069
[    0.069115] current->{active_,}mm->context = 0000000000000000
[    0.069167] current->{active_,}mm->pgd = fff8004030002000
[    0.069219]               \|/ ____ \|/
[    0.069219]               "@'/ .. \`@"
[    0.069219]               /_| \__/ |_\
[    0.069219]                  \__U_/
[    0.069346] swapper(0): Oops [#1]
[    0.069382] CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1+ #528
[    0.069441] task: 0000000000c121c0 task.stack: 0000000000bf8000
[    0.069496] TSTATE: 0000000080e01607 TPC: 00000000009b75bc TNPC:
00000000009b75c0 Y: 00000170    Not tainted
[    0.069582] TPC: <secpath_set+0x3c/0x80>
[    0.069621] g0: 000000000000002d g1: fffffffffffffff4 g2:
0000000000080000 g3: 0000000000d06400
[    0.069698] g4: 0000000000c121c0 g5: 0000000000000000 g6:
0000000000bf8000 g7: 0000000000d06400
[    0.069776] o0: 0000000000000018 o1: 0000000000bfbaa8 o2:
0000000000000088 o3: 0000000000ca5980
[    0.069853] o4: 0000000000000000 o5: 0000000000c53a98 sp:
0000000000bfb171 ret_pc: 00000000005a32ec
[    0.069939] RPC: <zone_batchsize.isra.8+0x4c/0x80>
[    0.069984] l0: 0000000000800000 l1: 0000000000d63000 l2:
0000000000000000 l3: 0000000000cb83c8
[    0.070061] l4: 0000000000b96cd8 l5: 00000000fef195e1 l6:
0000000000000082 l7: 0000000000d6eee8
[    0.070137] i0: 0000000000000001 i1: 0000000000b53658 i2:
0000000000000000 i3: 0000000000bfbb00
[    0.070214] i4: 0000000000bfbb08 i5: 0000000000000018 i6:
0000000000bfb221 i7: 0000000000cfb200
[    0.070294] I7: <free_area_init_node+0x468/0x4cc>
[    0.070338] Call Trace:
[    0.070366]  [0000000000cfb200] free_area_init_node+0x468/0x4cc
[    0.070424]  [0000000000cd8c7c] free_area_init_nodes+0x548/0x568
[    0.070482]  [0000000000ccae70] paging_init+0xf74/0xf90
[    0.070533]  [0000000000cc562c] setup_arch+0x284/0x6ec
[    0.070581]  [0000000000cc2728] start_kernel+0x5c/0x4a0
[    0.070631]  [0000000000cc5304] start_early_boot+0x2b8/0x2c8
[    0.070686]  [0000000000a41000] tlb_fixup_done+0x4c/0x6c
[    0.070737]  [000000000011d8a4] 0x11d8a4
[    0.070775] Disabling lock debugging due to kernel taint
[    0.070827] Caller[0000000000cfb200]: free_area_init_node+0x468/0x4cc
[    0.070886] Caller[0000000000cd8c7c]: free_area_init_nodes+0x548/0x568
[    0.070945] Caller[0000000000ccae70]: paging_init+0xf74/0xf90
[    0.070998] Caller[0000000000cc562c]: setup_arch+0x284/0x6ec
[    0.071051] Caller[0000000000cc2728]: start_kernel+0x5c/0x4a0
[    0.071105] Caller[0000000000cc5304]: start_early_boot+0x2b8/0x2c8
[    0.071162] Caller[0000000000a41000]: tlb_fixup_done+0x4c/0x6c
[    0.071217] Caller[000000000011d8a4]: 0x11d8a4
[    0.071259] Instruction DUMP:
[    0.071261]  ba100008
[    0.071292]  02c2000c
[    0.071317]  82103ff4
[    0.071342] <f85e2068>
[    0.071368]  02c70007
[    0.071392]  9210001c
[    0.071417]  4002014a
[    0.071443]  90102001
[    0.071468]  80a22000
[    0.071494]
[    0.071538] Kernel panic - not syncing: Attempted to kill the idle task!
[    0.071679] Unable to handle kernel NULL pointer dereference
...

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Anatoly Pugachev <matorola@gmail.com>
Date: Wed, 29 Nov 2017 11:56:49 +0300

> [    0.068851] Initmem setup node 0 [mem 0x0000000040400000-0x00000040bf8b7fff]

We're setting up the free memory areas in early bootup...

> [    0.068940] Kernel unaligned access at TPC[9b75bc] secpath_set+0x3c/0x80
> [    0.069009] Unable to handle kernel NULL pointer dereference in mna handler
> [    0.069011]  at virtual address 0000000000000069
> [    0.069115] current->{active_,}mm->context = 0000000000000000
> [    0.069167] current->{active_,}mm->pgd = fff8004030002000
> [    0.069219]               \|/ ____ \|/
> [    0.069219]               "@'/ .. \`@"
> [    0.069219]               /_| \__/ |_\
> [    0.069219]                  \__U_/
> [    0.069346] swapper(0): Oops [#1]
> [    0.069382] CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1+ #528
> [    0.069441] task: 0000000000c121c0 task.stack: 0000000000bf8000
> [    0.069496] TSTATE: 0000000080e01607 TPC: 00000000009b75bc TNPC:
> 00000000009b75c0 Y: 00000170    Not tainted
> [    0.069582] TPC: <secpath_set+0x3c/0x80>

And we're suddenly crashing in completely unrelated IPSEC networking
code.

Something is seriously wrong here.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Anatoly Pugachev <matorola@gmail.com>
Date: Mon, 27 Nov 2017 13:20:12 +0300

> [    0.175161]   node   0: [mem 0x0000000080000000-0x000000077fffffff]
> [    0.175221]   node   0: [mem 0x0000002000000000-0x000000202fffffff]
> [    0.176318]   node   0: [mem 0x0000004030000000-0x00000040bf8b3fff]
> [    0.176400] Initmem setup node 0 [mem 0x0000000040400000-0x00000040bf8b3fff]

For the record, same pattern as your other crash:

> [    0.176522] Kernel unaligned access at TPC[9f8a10]
> minmax_subwin_update+0x30/0x100
> [    0.176608] Unable to handle kernel NULL pointer dereference in mna handler
> [    0.176614]  at virtual address 0000000000000009
> [    0.176728] current->{active_,}mm->context = 0000000000000000
> [    0.176787] current->{active_,}mm->pgd = fff8004030002000
> [    0.176844]               \|/ ____ \|/
> [    0.176844]               "@'/ .. \`@"
> [    0.176844]               /_| \__/ |_\
> [    0.176844]                  \__U_/
> [    0.178015] swapper(0): Oops [#1]
> [    0.178070] CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1 #527
> [    0.178133] task: 0000000000c06c80 task.stack: 0000000000be8000
> [    0.178190] TSTATE: 0000000080e01607 TPC: 00000000009f8a10 TNPC:
> 00000000009f8a14 Y: 0000000f    Not tainted
> [    0.179325] TPC: <minmax_subwin_update+0x30/0x100>

We are initializing free memory, then we end up at a totally unrelated
piece of code.

Something is causing the cpu to call some random address and we need
to figure out why.

I've built the current kernel on a SunBlade-2500 with gcc-4.4.5 and I
couldn't reproduce this crash.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Anatoly Pugachev]

On Wed, Nov 29, 2017 at 9:48 PM, David Miller <davem@davemloft.net> wrote:
>
> We are initializing free memory, then we end up at a totally unrelated
> piece of code.
>
> Something is causing the cpu to call some random address and we need
> to figure out why.
>
> I've built the current kernel on a SunBlade-2500 with gcc-4.4.5 and I
> couldn't reproduce this crash.

David, using your git sparc kernel tree, bisect log:

mator@ttip:~/sparc$ git remote -v
origin  git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc.git (fetch)
origin  git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc.git (push)

mator@ttip:~/sparc$ git bisect good
46ad8d2d22c17e2b577adec55ae87161666a3267 is the first bad commit
commit 46ad8d2d22c17e2b577adec55ae87161666a3267
Author: Vijay Kumar <vijay.ac.kumar@oracle.com>
Date:   Wed Oct 11 12:50:06 2017 -0600

    sparc64: Use sparc optimized fls and __fls for T4 and above

    For T4 and above, patch fls and __fls functions
    at the boot time to use lzcnt instruction.

    Signed-off-by: Vijay Kumar <vijay.ac.kumar@oracle.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>

:040000 040000 ce2dbe8d19a2aa2cd1fae96c53d1424c5a381da9
54eb6cc99041f28bbe77b573cd7b2d55b8b0f5eb M      arch

mator@ttip:~/sparc$ git bisect log
git bisect start
# bad: [0c86a6bd85ff0629cd2c5141027fc1c8bb6cde9c] Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
git bisect bad 0c86a6bd85ff0629cd2c5141027fc1c8bb6cde9c
# good: [39dae59d66acd86d1de24294bd2f343fd5e7a625] Linux 4.14-rc8
git bisect good 39dae59d66acd86d1de24294bd2f343fd5e7a625
# good: [5bbcc0f595fadb4cac0eddc4401035ec0bd95b09] Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
git bisect good 5bbcc0f595fadb4cac0eddc4401035ec0bd95b09
# good: [441692aafc1731087bbaf657a8b6059d95c2a6df] Merge branch
'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm
git bisect good 441692aafc1731087bbaf657a8b6059d95c2a6df
# good: [bec04432cb9036dedf89140c102b5ac03e4b3626] Merge branch 'next'
of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux
git bisect good bec04432cb9036dedf89140c102b5ac03e4b3626
# good: [bfe0feb143151cce76411bb211c3eec0f6cff7ba] amdgpu/dc: convert
dc_stream_state to kref.
git bisect good bfe0feb143151cce76411bb211c3eec0f6cff7ba
# good: [2ce079f04d5914dae14fdc8618f804cc0d2a1b8f] Merge tag
'kbuild-misc-v4.15' of
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
git bisect good 2ce079f04d5914dae14fdc8618f804cc0d2a1b8f
# good: [817002475046737877915d87889a012d851650fa] Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
git bisect good 817002475046737877915d87889a012d851650fa
# bad: [020aae3ee58c1af0e7ffc4e2cc9fe4dc630338cb] ima: do not update
security.ima if appraisal status is not INTEGRITY_PASS
git bisect bad 020aae3ee58c1af0e7ffc4e2cc9fe4dc630338cb
# good: [f2645fa317b8905b8934f06a0601d5b7fa66aba0] platform/x86:
dell-smbios-wmi: introduce userspace interface
git bisect good f2645fa317b8905b8934f06a0601d5b7fa66aba0
# bad: [07c455ee222f3ad219c2835d05a175a326a138fb] Merge tag
'platform-drivers-x86-v4.15-1' of
git://git.infradead.org/linux-platform-drivers-x86
git bisect bad 07c455ee222f3ad219c2835d05a175a326a138fb
# bad: [1deab8ce2c91e3b16563b7a7ea150f82334262ec] Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc
git bisect bad 1deab8ce2c91e3b16563b7a7ea150f82334262ec
# good: [70cbec0c53739e170ea9c418e740f007ab5bca52] sparc64: SPARC
optimized fls function
git bisect good 70cbec0c53739e170ea9c418e740f007ab5bca52
# bad: [68fa10dce0bfe263000ea3fd147ed74f5a4342c2] sparc/led: Convert
timers to use timer_setup()
git bisect bad 68fa10dce0bfe263000ea3fd147ed74f5a4342c2
# bad: [46ad8d2d22c17e2b577adec55ae87161666a3267] sparc64: Use sparc
optimized fls and __fls for T4 and above
git bisect bad 46ad8d2d22c17e2b577adec55ae87161666a3267
# good: [2b41ce5df20b698436526fcd8478094140b91706] sparc64: SPARC
optimized __fls function
git bisect good 2b41ce5df20b698436526fcd8478094140b91706
# first bad commit: [46ad8d2d22c17e2b577adec55ae87161666a3267]
sparc64: Use sparc optimized fls and __fls for T4 and above

reverting just 46ad8d2d22c17e2b577adec55ae87161666a3267 on upstream
git kernel (git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git)
I was able to boot kernel v4.15-rc1-24-g43570f0383d6 successfully.

So, can we blame 46ad8d2d22c17e2b577adec55ae87161666a3267 and revert it ?

Thanks.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Frans van Berckel]

On Wed, 2017-11-29 at 13:48 -0500, David Miller wrote:
> From: Anatoly Pugachev <matorola@gmail.com>
> Date: Mon, 27 Nov 2017 13:20:12 +0300
> 
> > [    0.175161]   node   0: [mem 0x0000000080000000-
> 0x000000077fffffff]
> > [    0.175221]   node   0: [mem 0x0000002000000000-
> 0x000000202fffffff]
> > [    0.176318]   node   0: [mem 0x0000004030000000-
> 0x00000040bf8b3fff]
> > [    0.176400] Initmem setup node 0 [mem 0x0000000040400000-
> 0x00000040bf8b3fff]
> 
> For the record, same pattern as your other crash:
> 
> > [    0.176522] Kernel unaligned access at TPC[9f8a10]
> > minmax_subwin_update+0x30/0x100
> > [    0.176608] Unable to handle kernel NULL pointer dereference in
> mna handler
> > [    0.176614]  at virtual address 0000000000000009
> > [    0.176728] current->{active_,}mm->context = 0000000000000000
> > [    0.176787] current->{active_,}mm->pgd = fff8004030002000
> > [    0.176844]               \|/ ____ \|/
> > [    0.176844]               "@'/ .. \`@"
> > [    0.176844]               /_| \__/ |_\
> > [    0.176844]                  \__U_/
> > [    0.178015] swapper(0): Oops [#1]
> > [    0.178070] CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1
> #527
> > [    0.178133] task: 0000000000c06c80 task.stack: 0000000000be8000
> > [    0.178190] TSTATE: 0000000080e01607 TPC: 00000000009f8a10 TNPC:
> > 00000000009f8a14 Y: 0000000f    Not tainted
> > [    0.179325] TPC: <minmax_subwin_update+0x30/0x100>
> 
> We are initializing free memory, then we end up at a totally
> unrelated
> piece of code.
> 
> Something is causing the cpu to call some random address and we need
> to figure out why.
> 
> I've built the current kernel on a SunBlade-2500 with gcc-4.4.5 and I
> couldn't reproduce this crash.

Do you have a spare partition, or extra free disk, for use? Sparc64 is
what some of us are running. Maybe it helps reproducing. ;-)

https://cdimage.debian.org/cdimage/ports/

Thanks,


Frans van Berckel

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Frans van Berckel <fberckel@xs4all.nl>
Date: Wed, 29 Nov 2017 19:59:11 +0100

> Do you have a spare partition, or extra free disk, for use? Sparc64 is
> what some of us are running. Maybe it helps reproducing. ;-)
> 
> https://cdimage.debian.org/cdimage/ports/

That's nice, but this appears to be a Niagara and later specific problem
which is why I can't reproduce it.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Anatoly Pugachev <matorola@gmail.com>
Date: Wed, 29 Nov 2017 21:58:21 +0300

> On Wed, Nov 29, 2017 at 9:48 PM, David Miller <davem@davemloft.net> wrote:
>>
>> We are initializing free memory, then we end up at a totally unrelated
>> piece of code.
>>
>> Something is causing the cpu to call some random address and we need
>> to figure out why.
>>
>> I've built the current kernel on a SunBlade-2500 with gcc-4.4.5 and I
>> couldn't reproduce this crash.
> 
> David, using your git sparc kernel tree, bisect log:
> 
> mator@ttip:~/sparc$ git remote -v
> origin  git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc.git (fetch)
> origin  git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc.git (push)
> 
> mator@ttip:~/sparc$ git bisect good
> 46ad8d2d22c17e2b577adec55ae87161666a3267 is the first bad commit
> commit 46ad8d2d22c17e2b577adec55ae87161666a3267
> Author: Vijay Kumar <vijay.ac.kumar@oracle.com>
> Date:   Wed Oct 11 12:50:06 2017 -0600
> 
>     sparc64: Use sparc optimized fls and __fls for T4 and above

Thanks for bisecting, I'll take a look.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: David Miller <davem@davemloft.net>
Date: Wed, 29 Nov 2017 14:26:44 -0500 (EST)

>> mator@ttip:~/sparc$ git bisect good
>> 46ad8d2d22c17e2b577adec55ae87161666a3267 is the first bad commit
>> commit 46ad8d2d22c17e2b577adec55ae87161666a3267
>> Author: Vijay Kumar <vijay.ac.kumar@oracle.com>
>> Date:   Wed Oct 11 12:50:06 2017 -0600
>> 
>>     sparc64: Use sparc optimized fls and __fls for T4 and above
> 
> Thanks for bisecting, I'll take a look.

The problem is that the distance between the generic sparc64 fls/__fls
symbols and the NG4 optimized ones are too large for the branch patch.
It needs to fit in a 19-bit displacement field.

When there are library routines and patching, you have to group them
near eachother.  This means if the generic routine is "lib-y" you
have to make the patch code "lib-y" too.

In this case, the NG4fls.o object is placed into "obj-y" which is
too far from the "lib-y" code to fit into the disp19 relocation
of the patch branch.

Vijay, why did you put the NG4fls.o into obj-y?

Anatoly, please try this patch:

diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
index 0f0f76b..063556f 100644
--- a/arch/sparc/lib/Makefile
+++ b/arch/sparc/lib/Makefile
@@ -19,7 +19,7 @@ lib-$(CONFIG_SPARC32) += muldi3.o bitext.o cmpdi2.o
 lib-$(CONFIG_SPARC64) += multi3.o
 lib-$(CONFIG_SPARC64) += fls.o
 lib-$(CONFIG_SPARC64) += fls64.o
-obj-$(CONFIG_SPARC64) += NG4fls.o
+lib-$(CONFIG_SPARC64) += NG4fls.o
 
 lib-$(CONFIG_SPARC64) += copy_page.o clear_page.o bzero.o
 lib-$(CONFIG_SPARC64) += csum_copy.o csum_copy_from_user.o csum_copy_to_user.o

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Anatoly Pugachev]

On Wed, Nov 29, 2017 at 10:37 PM, David Miller <davem@davemloft.net> wrote:
>
> Anatoly, please try this patch:
>
> diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
> index 0f0f76b..063556f 100644
> --- a/arch/sparc/lib/Makefile
> +++ b/arch/sparc/lib/Makefile
> @@ -19,7 +19,7 @@ lib-$(CONFIG_SPARC32) += muldi3.o bitext.o cmpdi2.o
>  lib-$(CONFIG_SPARC64) += multi3.o
>  lib-$(CONFIG_SPARC64) += fls.o
>  lib-$(CONFIG_SPARC64) += fls64.o
> -obj-$(CONFIG_SPARC64) += NG4fls.o
> +lib-$(CONFIG_SPARC64) += NG4fls.o
>
>  lib-$(CONFIG_SPARC64) += copy_page.o clear_page.o bzero.o
>  lib-$(CONFIG_SPARC64) += csum_copy.o csum_copy_from_user.o csum_copy_to_user.o


Yes, this patch fixes problem. Thanks.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[David Miller]

From: Anatoly Pugachev <matorola@gmail.com>
Date: Wed, 29 Nov 2017 23:05:19 +0300

> On Wed, Nov 29, 2017 at 10:37 PM, David Miller <davem@davemloft.net> wrote:
>>
>> Anatoly, please try this patch:
>>
>> diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
>> index 0f0f76b..063556f 100644
>> --- a/arch/sparc/lib/Makefile
>> +++ b/arch/sparc/lib/Makefile
>> @@ -19,7 +19,7 @@ lib-$(CONFIG_SPARC32) += muldi3.o bitext.o cmpdi2.o
>>  lib-$(CONFIG_SPARC64) += multi3.o
>>  lib-$(CONFIG_SPARC64) += fls.o
>>  lib-$(CONFIG_SPARC64) += fls64.o
>> -obj-$(CONFIG_SPARC64) += NG4fls.o
>> +lib-$(CONFIG_SPARC64) += NG4fls.o
>>
>>  lib-$(CONFIG_SPARC64) += copy_page.o clear_page.o bzero.o
>>  lib-$(CONFIG_SPARC64) += csum_copy.o csum_copy_from_user.o csum_copy_to_user.o
> 
> 
> Yes, this patch fixes problem. Thanks.

Thanks, I'll push this upstream.

Re: sparc64: git kernel 4.14.0-12995-g0c86a6bd85ff does not boot

[Vijay Kumar]

On 11/29/2017 1:37 PM, David Miller wrote:
> Vijay, why did you put the NG4fls.o into obj-y?
>
> Anatoly, please try this patch:
>
> diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
> index 0f0f76b..063556f 100644
> --- a/arch/sparc/lib/Makefile
> +++ b/arch/sparc/lib/Makefile
> @@ -19,7 +19,7 @@ lib-$(CONFIG_SPARC32) += muldi3.o bitext.o cmpdi2.o
>   lib-$(CONFIG_SPARC64) += multi3.o
>   lib-$(CONFIG_SPARC64) += fls.o
>   lib-$(CONFIG_SPARC64) += fls64.o
> -obj-$(CONFIG_SPARC64) += NG4fls.o
> +lib-$(CONFIG_SPARC64) += NG4fls.o
>   
>   lib-$(CONFIG_SPARC64) += copy_page.o clear_page.o bzero.o
>   lib-$(CONFIG_SPARC64) += csum_copy.o csum_copy_from_user.o csum_copy_to_user.o
Reviewed-by: Vijay Kumar <vijay.ac.kumar@oracle.com>

You are right, obj-y is not required.

Thanks,
Vijay