Skip to contentSkip to conversation

Unable to handle kernel paging request at virtual address 0000001bf8068693

DISCUSSION Andreas Schwab 5 participants 11 messages Sep 5, 2018 – Sep 18, 2018 RISC-V

lore
Share & tools
lore
#0 lore raw

Trying to boot git://github.com/riscv/riscv-linux#riscv-linux-4.18 on the HiFive-U I'm getting this crash:

[    0.100000] Unable to handle kernel paging request at virtual address 0000001bf8068693
[    0.100000] Oops [#1]
[    0.100000] Modules linked in:
[    0.100000] CPU: 1 PID: 5 Comm: kworker/1:0 Not tainted 4.18.0-01735-g9f6192e555 #107
[    0.100000] Workqueue: rcu_gp wait_rcu_exp_gp
[    0.100000] sepc: ffffffe00116da98 ra : ffffffe000c9399a sp : ffffffe1f6addcb0
[    0.100000]  gp : ffffffe001404160 tp : ffffffe1f6ada140 t0 : ffffffe1f6c44000
[    0.100000]  t1 : 0000000000000000 t2 : 000000000000f8e1 s0 : ffffffe1f6addcc0
[    0.100000]  s1 : ffffffe000c7ff00 a0 : 0076161bf8068693 a1 : 000000007fffffff
[    0.100000]  a2 : ffffffe0013ac5c8 a3 : 0000000000000001 a4 : 0000000000000001
[    0.100000]  a5 : 000000007fffffff a6 : ffff8f98a08a9c8d a7 : ffffffe1f7d4c4f0
[    0.100000]  s2 : ffffffe0013ac5c8 s3 : ffffffe000c740d8 s4 : 0000000000000000
[    0.100000]  s5 : 00000000ffffffff s6 : 0000000000000000 s7 : ffffffe000c740d8
[    0.100000]  s8 : 0000000000000118 s9 : 0076161bf8068693 s10: ffffffe1f6a09000
[    0.100000]  s11: ffffffe000cc919e t3 : ffffffe001399aa0 t4 : ffffffe001399aa8
[    0.100000]  t5 : 00000000000c8000 t6 : ffffffe1f6c40a4b
[    0.100000] sstatus: 0000000200000100 sbadaddr: 0000001bf8068693 scause: 000000000000000d
[    0.110000] ---[ end trace 0ddcca0cde88bd4b ]---

The crash is inside _raw_spin_lock:

ffffffe00116da90 <_raw_spin_lock>:

ffffffe00116da90:       1141                    addi    sp,sp,-16
ffffffe00116da92:       e422                    sd      s0,8(sp)
ffffffe00116da94:       0800                    addi    s0,sp,16
ffffffe00116da96:       4705                    li      a4,1
ffffffe00116da98:       411c                    lw      a5,0(a0)
ffffffe00116da9a:       2781                    sext.w  a5,a5
ffffffe00116da9c:       fff5                    bnez    a5,ffffffe00116da98 <_r>
ffffffe00116da9e:       08e527af                amoswap.w       a5,a4,(a0)
ffffffe00116daa2:       0230000f                fence   r,rw
ffffffe00116daa6:       2781                    sext.w  a5,a5
ffffffe00116daa8:       fbe5                    bnez    a5,ffffffe00116da98 <_r>
ffffffe00116daaa:       6422                    ld      s0,8(sp)
ffffffe00116daac:       0141                    addi    sp,sp,16
ffffffe00116daae:       8082                    ret

Andreas.

Signature · 3 lines
Andreas Schwab, SUSE Labs, schwab at suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE  1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."
#1 lore raw
Quoted 40 lines · Andreas Schwab

Trying to boot git://github.com/riscv/riscv-linux#riscv-linux-4.18 on the HiFive-U I'm getting this crash:

[    0.100000] Unable to handle kernel paging request at virtual address 0000001bf8068693
[    0.100000] Oops [#1]
[    0.100000] Modules linked in:
[    0.100000] CPU: 1 PID: 5 Comm: kworker/1:0 Not tainted 4.18.0-01735-g9f6192e555 #107
[    0.100000] Workqueue: rcu_gp wait_rcu_exp_gp
[    0.100000] sepc: ffffffe00116da98 ra : ffffffe000c9399a sp : ffffffe1f6addcb0
[    0.100000]  gp : ffffffe001404160 tp : ffffffe1f6ada140 t0 : ffffffe1f6c44000
[    0.100000]  t1 : 0000000000000000 t2 : 000000000000f8e1 s0 : ffffffe1f6addcc0
[    0.100000]  s1 : ffffffe000c7ff00 a0 : 0076161bf8068693 a1 : 000000007fffffff
[    0.100000]  a2 : ffffffe0013ac5c8 a3 : 0000000000000001 a4 : 0000000000000001
[    0.100000]  a5 : 000000007fffffff a6 : ffff8f98a08a9c8d a7 : ffffffe1f7d4c4f0
[    0.100000]  s2 : ffffffe0013ac5c8 s3 : ffffffe000c740d8 s4 : 0000000000000000
[    0.100000]  s5 : 00000000ffffffff s6 : 0000000000000000 s7 : ffffffe000c740d8
[    0.100000]  s8 : 0000000000000118 s9 : 0076161bf8068693 s10: ffffffe1f6a09000
[    0.100000]  s11: ffffffe000cc919e t3 : ffffffe001399aa0 t4 : ffffffe001399aa8
[    0.100000]  t5 : 00000000000c8000 t6 : ffffffe1f6c40a4b
[    0.100000] sstatus: 0000000200000100 sbadaddr: 0000001bf8068693 scause: 000000000000000d
[    0.110000] ---[ end trace 0ddcca0cde88bd4b ]---

The crash is inside _raw_spin_lock:

ffffffe00116da90 <_raw_spin_lock>:

ffffffe00116da90:       1141                    addi    sp,sp,-16
ffffffe00116da92:       e422                    sd      s0,8(sp)
ffffffe00116da94:       0800                    addi    s0,sp,16
ffffffe00116da96:       4705                    li      a4,1
ffffffe00116da98:       411c                    lw      a5,0(a0)
ffffffe00116da9a:       2781                    sext.w  a5,a5
ffffffe00116da9c:       fff5                    bnez    a5,ffffffe00116da98 <_r>
ffffffe00116da9e:       08e527af                amoswap.w       a5,a4,(a0)
ffffffe00116daa2:       0230000f                fence   r,rw
ffffffe00116daa6:       2781                    sext.w  a5,a5
ffffffe00116daa8:       fbe5                    bnez    a5,ffffffe00116da98 <_r>
ffffffe00116daaa:       6422                    ld      s0,8(sp)
ffffffe00116daac:       0141                    addi    sp,sp,16
ffffffe00116daae:       8082                    ret
Quoted 2 lines

Andreas.

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Regards,
Atish

#2 lore raw
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

#3 lore raw
Quoted 12 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

Quoted 7 lines

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

OK, it backports cleanly and passes a short rcutorture test. You guys have tested it, correct? If so, I am happy to pass it on to -stable.

Thanx, Paul

#4 lore raw
Quoted 19 lines · Paul E. McKenney
Quoted 18 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700
Quoted 8 lines
Quoted 1 line

rcu: Make expedited GPs handle CPU 0 being offline

OK, it backports cleanly and passes a short rcutorture test. You guys have tested it, correct? If so, I am happy to pass it on to -stable.

Thanx, Paul

Hi Paul,
I had it tested on top of 4.18-rc8 with RISC-V board (HighFive Unleashed).

I can try again on 4.18-stable if required.

Regards,
Atish

#5 lore raw
Quoted 25 lines · Atish Patra
Quoted 24 lines · Paul E. McKenney
Quoted 12 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

Quoted 7 lines

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

OK, it backports cleanly and passes a short rcutorture test. You guys have tested it, correct? If so, I am happy to pass it on to -stable.

Quoted 7 lines
Quoted 3 lines

Thanx, Paul

Hi Paul,
I had it tested on top of 4.18-rc8 with RISC-V board (HighFive Unleashed).

I can try again on 4.18-stable if required.

Hello, Atish,

Please do test it on 4.18-stable. The -stable maintainers get understandably annoyed when given non-working backports. ;-)

Thanx, Paul

#8 lore raw
In reply to Christoph Hellwig
Quoted 7 lines · Christoph Hellwig

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

Thanks, this got me further.

Andreas.

Signature · 3 lines
Andreas Schwab, SUSE Labs, schwab at suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE  1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."
#9 lore raw
Quoted 11 lines · Andreas Schwab
Quoted 7 lines · Christoph Hellwig

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

Thanks, this got me further.

Very good! This was against 4.18-stable, correct? If so, please let me know and I will send the request to -stable. Or you guys can send the request, your choice.

Thanx, Paul

Tested-by
#6 lore raw
In reply to Paul E. McKenney
Quoted 32 lines · Paul E. McKenney
Quoted 31 lines · Atish Patra
Quoted 19 lines · Paul E. McKenney
Quoted 18 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700
Quoted 8 lines
Quoted 1 line

rcu: Make expedited GPs handle CPU 0 being offline

OK, it backports cleanly and passes a short rcutorture test. You guys have tested it, correct? If so, I am happy to pass it on to -stable.

Thanx, Paul

Hi Paul,
I had it tested on top of 4.18-rc8 with RISC-V board (HighFive Unleashed).

Quoted 10 lines
Quoted 1 line

I can try again on 4.18-stable if required.

Hello, Atish,

Please do test it on 4.18-stable. The -stable maintainers get understandably annoyed when given non-working backports. ;-)

Thanx, Paul

Done. I have verified on 4.18-stable with HighFive Unleashed board.

FWIW,

Regards,
Atish

#7 lore raw
Quoted 40 lines · Atish Patra
Quoted 39 lines · Paul E. McKenney
Quoted 25 lines · Atish Patra
Quoted 24 lines · Paul E. McKenney
Quoted 12 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

Quoted 7 lines

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

OK, it backports cleanly and passes a short rcutorture test. You guys have tested it, correct? If so, I am happy to pass it on to -stable.

Quoted 7 lines
Quoted 3 lines

Thanx, Paul

Hi Paul,
I had it tested on top of 4.18-rc8 with RISC-V board (HighFive Unleashed).

I can try again on 4.18-stable if required.

Hello, Atish,

Please do test it on 4.18-stable. The -stable maintainers get understandably annoyed when given non-working backports. ;-)

Quoted 8 lines
Quoted 3 lines

Thanx, Paul

Done. I have verified on 4.18-stable with HighFive Unleashed board.

FWIW,

And I sent the -stable request, CCing you all. Here is hoping!

Thanx, Paul

#10 lore raw
In reply to Christoph Hellwig
Quoted 12 lines · Christoph Hellwig
Quoted 6 lines · Atish Patra

This issue is fixed in 4.19-rc1.

If you don't want to pull the latest, here is the change you need to cherry-pick to your branch.

https://lore.kernel.org/lkml/20180626192746.GJ3593 at linux.vnet.ibm.com/

Paul, I think we need this in 4.18-stable.

Also Palmer, I think we should pick it up in riscv-linux-4.18.

Quoted 7 lines

The mainline commit is:

commit fcc63543650150629c8a873cbef3578770acecd9

Author: Boqun Feng <[email protected]>
Date:   Fri Jun 15 12:06:31 2018 -0700

rcu: Make expedited GPs handle CPU 0 being offline

I added it to riscv-linux-4.18. Thanks!