mbox series

[4.9,00/24] V4.9 backport of 32-bit arm spectre patches

Message ID 20181031135713.2873-1-dave.long@linaro.org
Headers show
Series V4.9 backport of 32-bit arm spectre patches | expand

Message

David Long Oct. 31, 2018, 1:56 p.m. UTC 
From: "David A. Long" <dave.long@linaro.org>


V4.9 backport of spectre patches from Russell M. King's spectre branch.
Patches not yet in upstream are excluded.

Marc Zyngier (2):
  ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17
  ARM: KVM: invalidate icache on guest exit for Cortex-A15

Russell King (22):
  ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs
  ARM: bugs: prepare processor bug infrastructure
  ARM: bugs: hook processor bug checking into SMP and suspend paths
  ARM: bugs: add support for per-processor bug checking
  ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre
  ARM: spectre-v2: harden branch predictor on context switches
  ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit
  ARM: spectre-v2: harden user aborts in kernel space
  ARM: spectre-v2: add firmware based hardening
  ARM: spectre-v2: warn about incorrect context switching functions
  ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15
  ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling
  ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1
  ARM: spectre-v1: add speculation barrier (csdb) macros
  ARM: spectre-v1: add array_index_mask_nospec() implementation
  ARM: spectre-v1: fix syscall entry
  ARM: signal: copy registers using __copy_from_user()
  ARM: vfp: use __copy_from_user() when restoring VFP state
  ARM: oabi-compat: copy semops using __copy_from_user()
  ARM: use __inttype() in get_user()
  ARM: spectre-v1: use get_user() for __get_user()
  ARM: spectre-v1: mitigate user accesses

 arch/arm/include/asm/assembler.h   |  12 ++
 arch/arm/include/asm/barrier.h     |  32 ++++++
 arch/arm/include/asm/bugs.h        |   6 +-
 arch/arm/include/asm/cp15.h        |   3 +
 arch/arm/include/asm/cputype.h     |   8 ++
 arch/arm/include/asm/kvm_asm.h     |   2 -
 arch/arm/include/asm/kvm_host.h    |  14 ++-
 arch/arm/include/asm/kvm_mmu.h     |  23 +++-
 arch/arm/include/asm/proc-fns.h    |   4 +
 arch/arm/include/asm/system_misc.h |  15 +++
 arch/arm/include/asm/thread_info.h |   4 +-
 arch/arm/include/asm/uaccess.h     |  26 +++--
 arch/arm/kernel/Makefile           |   1 +
 arch/arm/kernel/bugs.c             |  18 +++
 arch/arm/kernel/entry-common.S     |  18 ++-
 arch/arm/kernel/entry-header.S     |  25 +++++
 arch/arm/kernel/signal.c           |  55 ++++-----
 arch/arm/kernel/smp.c              |   4 +
 arch/arm/kernel/suspend.c          |   2 +
 arch/arm/kernel/sys_oabi-compat.c  |   8 +-
 arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-
 arch/arm/lib/copy_from_user.S      |   9 ++
 arch/arm/mm/Kconfig                |  23 ++++
 arch/arm/mm/Makefile               |   2 +-
 arch/arm/mm/fault.c                |   3 +
 arch/arm/mm/proc-macros.S          |   3 +-
 arch/arm/mm/proc-v7-2level.S       |   6 -
 arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++
 arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------
 arch/arm/vfp/vfpmodule.c           |  17 ++-
 30 files changed, 674 insertions(+), 107 deletions(-)
 create mode 100644 arch/arm/kernel/bugs.c
 create mode 100644 arch/arm/mm/proc-v7-bugs.c

-- 
2.17.1

Comments

Florian Fainelli Oct. 31, 2018, 9:23 p.m. UTC | #1 
Hi David,

On 10/31/18 6:56 AM, David Long wrote:
> From: "David A. Long" <dave.long@linaro.org>

> 

> V4.9 backport of spectre patches from Russell M. King's spectre branch.

> Patches not yet in upstream are excluded.


Thanks for submitting those patches!

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>


Boot tested on a Brahma-B15 based platform and did not see any
regressions or issues, seeing about the same hackbench performance
before and after.

Test:
#!/bin/sh
for i in $(seq 0 9)
do
	hackbench 13 process 10000
done

before:

min: 140.800
max: 142.571
avg: 141.7233

after:

min: 140.004
max: 141.600
avg: 141.0242


> 

> Marc Zyngier (2):

>   ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17

>   ARM: KVM: invalidate icache on guest exit for Cortex-A15

> 

> Russell King (22):

>   ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs

>   ARM: bugs: prepare processor bug infrastructure

>   ARM: bugs: hook processor bug checking into SMP and suspend paths

>   ARM: bugs: add support for per-processor bug checking

>   ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre

>   ARM: spectre-v2: harden branch predictor on context switches

>   ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit

>   ARM: spectre-v2: harden user aborts in kernel space

>   ARM: spectre-v2: add firmware based hardening

>   ARM: spectre-v2: warn about incorrect context switching functions

>   ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15

>   ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling

>   ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1

>   ARM: spectre-v1: add speculation barrier (csdb) macros

>   ARM: spectre-v1: add array_index_mask_nospec() implementation

>   ARM: spectre-v1: fix syscall entry

>   ARM: signal: copy registers using __copy_from_user()

>   ARM: vfp: use __copy_from_user() when restoring VFP state

>   ARM: oabi-compat: copy semops using __copy_from_user()

>   ARM: use __inttype() in get_user()

>   ARM: spectre-v1: use get_user() for __get_user()

>   ARM: spectre-v1: mitigate user accesses

> 

>  arch/arm/include/asm/assembler.h   |  12 ++

>  arch/arm/include/asm/barrier.h     |  32 ++++++

>  arch/arm/include/asm/bugs.h        |   6 +-

>  arch/arm/include/asm/cp15.h        |   3 +

>  arch/arm/include/asm/cputype.h     |   8 ++

>  arch/arm/include/asm/kvm_asm.h     |   2 -

>  arch/arm/include/asm/kvm_host.h    |  14 ++-

>  arch/arm/include/asm/kvm_mmu.h     |  23 +++-

>  arch/arm/include/asm/proc-fns.h    |   4 +

>  arch/arm/include/asm/system_misc.h |  15 +++

>  arch/arm/include/asm/thread_info.h |   4 +-

>  arch/arm/include/asm/uaccess.h     |  26 +++--

>  arch/arm/kernel/Makefile           |   1 +

>  arch/arm/kernel/bugs.c             |  18 +++

>  arch/arm/kernel/entry-common.S     |  18 ++-

>  arch/arm/kernel/entry-header.S     |  25 +++++

>  arch/arm/kernel/signal.c           |  55 ++++-----

>  arch/arm/kernel/smp.c              |   4 +

>  arch/arm/kernel/suspend.c          |   2 +

>  arch/arm/kernel/sys_oabi-compat.c  |   8 +-

>  arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-

>  arch/arm/lib/copy_from_user.S      |   9 ++

>  arch/arm/mm/Kconfig                |  23 ++++

>  arch/arm/mm/Makefile               |   2 +-

>  arch/arm/mm/fault.c                |   3 +

>  arch/arm/mm/proc-macros.S          |   3 +-

>  arch/arm/mm/proc-v7-2level.S       |   6 -

>  arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++

>  arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------

>  arch/arm/vfp/vfpmodule.c           |  17 ++-

>  30 files changed, 674 insertions(+), 107 deletions(-)

>  create mode 100644 arch/arm/kernel/bugs.c

>  create mode 100644 arch/arm/mm/proc-v7-bugs.c

> 



-- 
Florian
David Long Nov. 2, 2018, 1:18 a.m. UTC | #2 
On 10/31/18 9:56 AM, David Long wrote:
> From: "David A. Long" <dave.long@linaro.org>

> 

> V4.9 backport of spectre patches from Russell M. King's spectre branch.

> Patches not yet in upstream are excluded.

> 

> Marc Zyngier (2):

>    ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17

>    ARM: KVM: invalidate icache on guest exit for Cortex-A15

> 

> Russell King (22):

>    ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs

>    ARM: bugs: prepare processor bug infrastructure

>    ARM: bugs: hook processor bug checking into SMP and suspend paths

>    ARM: bugs: add support for per-processor bug checking

>    ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre

>    ARM: spectre-v2: harden branch predictor on context switches

>    ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit

>    ARM: spectre-v2: harden user aborts in kernel space

>    ARM: spectre-v2: add firmware based hardening

>    ARM: spectre-v2: warn about incorrect context switching functions

>    ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15

>    ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling

>    ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1

>    ARM: spectre-v1: add speculation barrier (csdb) macros

>    ARM: spectre-v1: add array_index_mask_nospec() implementation

>    ARM: spectre-v1: fix syscall entry

>    ARM: signal: copy registers using __copy_from_user()

>    ARM: vfp: use __copy_from_user() when restoring VFP state

>    ARM: oabi-compat: copy semops using __copy_from_user()

>    ARM: use __inttype() in get_user()

>    ARM: spectre-v1: use get_user() for __get_user()

>    ARM: spectre-v1: mitigate user accesses

> 

>   arch/arm/include/asm/assembler.h   |  12 ++

>   arch/arm/include/asm/barrier.h     |  32 ++++++

>   arch/arm/include/asm/bugs.h        |   6 +-

>   arch/arm/include/asm/cp15.h        |   3 +

>   arch/arm/include/asm/cputype.h     |   8 ++

>   arch/arm/include/asm/kvm_asm.h     |   2 -

>   arch/arm/include/asm/kvm_host.h    |  14 ++-

>   arch/arm/include/asm/kvm_mmu.h     |  23 +++-

>   arch/arm/include/asm/proc-fns.h    |   4 +

>   arch/arm/include/asm/system_misc.h |  15 +++

>   arch/arm/include/asm/thread_info.h |   4 +-

>   arch/arm/include/asm/uaccess.h     |  26 +++--

>   arch/arm/kernel/Makefile           |   1 +

>   arch/arm/kernel/bugs.c             |  18 +++

>   arch/arm/kernel/entry-common.S     |  18 ++-

>   arch/arm/kernel/entry-header.S     |  25 +++++

>   arch/arm/kernel/signal.c           |  55 ++++-----

>   arch/arm/kernel/smp.c              |   4 +

>   arch/arm/kernel/suspend.c          |   2 +

>   arch/arm/kernel/sys_oabi-compat.c  |   8 +-

>   arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-

>   arch/arm/lib/copy_from_user.S      |   9 ++

>   arch/arm/mm/Kconfig                |  23 ++++

>   arch/arm/mm/Makefile               |   2 +-

>   arch/arm/mm/fault.c                |   3 +

>   arch/arm/mm/proc-macros.S          |   3 +-

>   arch/arm/mm/proc-v7-2level.S       |   6 -

>   arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++

>   arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------

>   arch/arm/vfp/vfpmodule.c           |  17 ++-

>   30 files changed, 674 insertions(+), 107 deletions(-)

>   create mode 100644 arch/arm/kernel/bugs.c

>   create mode 100644 arch/arm/mm/proc-v7-bugs.c

> 


kvm-unit-test'ing of this results in a hypervisor panic that doesn't 
happen without the patches. This needs to be figured out before it is 
accepted into stable. Looks like a V2 will be needed. Clearly kernelci 
testing alone is not sufficient when dealing with kvm changes.

-dl
Marc Zyngier Nov. 2, 2018, 8:54 a.m. UTC | #3 
On 02/11/18 01:18, David Long wrote:
> On 10/31/18 9:56 AM, David Long wrote:

>> From: "David A. Long" <dave.long@linaro.org>

>>

>> V4.9 backport of spectre patches from Russell M. King's spectre branch.

>> Patches not yet in upstream are excluded.

>>

>> Marc Zyngier (2):

>>    ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17

>>    ARM: KVM: invalidate icache on guest exit for Cortex-A15

>>

>> Russell King (22):

>>    ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs

>>    ARM: bugs: prepare processor bug infrastructure

>>    ARM: bugs: hook processor bug checking into SMP and suspend paths

>>    ARM: bugs: add support for per-processor bug checking

>>    ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre

>>    ARM: spectre-v2: harden branch predictor on context switches

>>    ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit

>>    ARM: spectre-v2: harden user aborts in kernel space

>>    ARM: spectre-v2: add firmware based hardening

>>    ARM: spectre-v2: warn about incorrect context switching functions

>>    ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15

>>    ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling

>>    ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1

>>    ARM: spectre-v1: add speculation barrier (csdb) macros

>>    ARM: spectre-v1: add array_index_mask_nospec() implementation

>>    ARM: spectre-v1: fix syscall entry

>>    ARM: signal: copy registers using __copy_from_user()

>>    ARM: vfp: use __copy_from_user() when restoring VFP state

>>    ARM: oabi-compat: copy semops using __copy_from_user()

>>    ARM: use __inttype() in get_user()

>>    ARM: spectre-v1: use get_user() for __get_user()

>>    ARM: spectre-v1: mitigate user accesses

>>

>>   arch/arm/include/asm/assembler.h   |  12 ++

>>   arch/arm/include/asm/barrier.h     |  32 ++++++

>>   arch/arm/include/asm/bugs.h        |   6 +-

>>   arch/arm/include/asm/cp15.h        |   3 +

>>   arch/arm/include/asm/cputype.h     |   8 ++

>>   arch/arm/include/asm/kvm_asm.h     |   2 -

>>   arch/arm/include/asm/kvm_host.h    |  14 ++-

>>   arch/arm/include/asm/kvm_mmu.h     |  23 +++-

>>   arch/arm/include/asm/proc-fns.h    |   4 +

>>   arch/arm/include/asm/system_misc.h |  15 +++

>>   arch/arm/include/asm/thread_info.h |   4 +-

>>   arch/arm/include/asm/uaccess.h     |  26 +++--

>>   arch/arm/kernel/Makefile           |   1 +

>>   arch/arm/kernel/bugs.c             |  18 +++

>>   arch/arm/kernel/entry-common.S     |  18 ++-

>>   arch/arm/kernel/entry-header.S     |  25 +++++

>>   arch/arm/kernel/signal.c           |  55 ++++-----

>>   arch/arm/kernel/smp.c              |   4 +

>>   arch/arm/kernel/suspend.c          |   2 +

>>   arch/arm/kernel/sys_oabi-compat.c  |   8 +-

>>   arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-

>>   arch/arm/lib/copy_from_user.S      |   9 ++

>>   arch/arm/mm/Kconfig                |  23 ++++

>>   arch/arm/mm/Makefile               |   2 +-

>>   arch/arm/mm/fault.c                |   3 +

>>   arch/arm/mm/proc-macros.S          |   3 +-

>>   arch/arm/mm/proc-v7-2level.S       |   6 -

>>   arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++

>>   arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------

>>   arch/arm/vfp/vfpmodule.c           |  17 ++-

>>   30 files changed, 674 insertions(+), 107 deletions(-)

>>   create mode 100644 arch/arm/kernel/bugs.c

>>   create mode 100644 arch/arm/mm/proc-v7-bugs.c

>>

> 

> kvm-unit-test'ing of this results in a hypervisor panic that doesn't 

> happen without the patches. This needs to be figured out before it is 

> accepted into stable. Looks like a V2 will be needed. Clearly kernelci 

> testing alone is not sufficient when dealing with kvm changes.


How about posting the panic message, a description of what you were
doing when that happened, and details of the configuration (HW used,
Thumb-2 or not...)? If you cannot perform the analysis yourself, at
least give us enough information to help you.

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...
Russell King (Oracle) Nov. 2, 2018, 11:28 a.m. UTC | #4 
On Thu, Nov 01, 2018 at 09:18:02PM -0400, David Long wrote:
> On 10/31/18 9:56 AM, David Long wrote:

> >From: "David A. Long" <dave.long@linaro.org>

> >

> >V4.9 backport of spectre patches from Russell M. King's spectre branch.

> >Patches not yet in upstream are excluded.

> >

> >Marc Zyngier (2):

> >   ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17

> >   ARM: KVM: invalidate icache on guest exit for Cortex-A15

> >

> >Russell King (22):

> >   ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs

> >   ARM: bugs: prepare processor bug infrastructure

> >   ARM: bugs: hook processor bug checking into SMP and suspend paths

> >   ARM: bugs: add support for per-processor bug checking

> >   ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre

> >   ARM: spectre-v2: harden branch predictor on context switches

> >   ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit

> >   ARM: spectre-v2: harden user aborts in kernel space

> >   ARM: spectre-v2: add firmware based hardening

> >   ARM: spectre-v2: warn about incorrect context switching functions

> >   ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15

> >   ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling

> >   ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1

> >   ARM: spectre-v1: add speculation barrier (csdb) macros

> >   ARM: spectre-v1: add array_index_mask_nospec() implementation

> >   ARM: spectre-v1: fix syscall entry

> >   ARM: signal: copy registers using __copy_from_user()

> >   ARM: vfp: use __copy_from_user() when restoring VFP state

> >   ARM: oabi-compat: copy semops using __copy_from_user()

> >   ARM: use __inttype() in get_user()

> >   ARM: spectre-v1: use get_user() for __get_user()

> >   ARM: spectre-v1: mitigate user accesses

> >

> >  arch/arm/include/asm/assembler.h   |  12 ++

> >  arch/arm/include/asm/barrier.h     |  32 ++++++

> >  arch/arm/include/asm/bugs.h        |   6 +-

> >  arch/arm/include/asm/cp15.h        |   3 +

> >  arch/arm/include/asm/cputype.h     |   8 ++

> >  arch/arm/include/asm/kvm_asm.h     |   2 -

> >  arch/arm/include/asm/kvm_host.h    |  14 ++-

> >  arch/arm/include/asm/kvm_mmu.h     |  23 +++-

> >  arch/arm/include/asm/proc-fns.h    |   4 +

> >  arch/arm/include/asm/system_misc.h |  15 +++

> >  arch/arm/include/asm/thread_info.h |   4 +-

> >  arch/arm/include/asm/uaccess.h     |  26 +++--

> >  arch/arm/kernel/Makefile           |   1 +

> >  arch/arm/kernel/bugs.c             |  18 +++

> >  arch/arm/kernel/entry-common.S     |  18 ++-

> >  arch/arm/kernel/entry-header.S     |  25 +++++

> >  arch/arm/kernel/signal.c           |  55 ++++-----

> >  arch/arm/kernel/smp.c              |   4 +

> >  arch/arm/kernel/suspend.c          |   2 +

> >  arch/arm/kernel/sys_oabi-compat.c  |   8 +-

> >  arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-

> >  arch/arm/lib/copy_from_user.S      |   9 ++

> >  arch/arm/mm/Kconfig                |  23 ++++

> >  arch/arm/mm/Makefile               |   2 +-

> >  arch/arm/mm/fault.c                |   3 +

> >  arch/arm/mm/proc-macros.S          |   3 +-

> >  arch/arm/mm/proc-v7-2level.S       |   6 -

> >  arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++

> >  arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------

> >  arch/arm/vfp/vfpmodule.c           |  17 ++-

> >  30 files changed, 674 insertions(+), 107 deletions(-)

> >  create mode 100644 arch/arm/kernel/bugs.c

> >  create mode 100644 arch/arm/mm/proc-v7-bugs.c

> >

> 

> kvm-unit-test'ing of this results in a hypervisor panic that doesn't happen

> without the patches. This needs to be figured out before it is accepted into

> stable. Looks like a V2 will be needed. Clearly kernelci testing alone is

> not sufficient when dealing with kvm changes.


I've discovered in the last few days that kernelci boot testing is
next to useless - it bases its pass/fail result on whether it gets to
a shell prompt, which can happen even if the kernel hits a BUG() or
warning that doesn't prevent the system getting to a shell prompt.

For example, see:

01:08:40.181846  [    9.309984] Unable to handle kernel paging request at virtual address e7fddef0

which is the kernel hitting a BUG() in:

https://storage.kernelci.org/rmk/to-build/v4.16-38-g9fa10446d304/arm/multi_v7_defconfig/lab-collabora/boot-exynos5800-peach-pi.html

and that results in a "pass" result for that boot test.

So, don't believe a "pass" result from kernelci at the moment, it's
meaningless in determining whether anything has broken.  The only
way around this is to manually read each and every boot log, which
is tedious, or run your own tests on local systems.

I've reported this to info@kernelci.org earlier this week, and am
waiting for a response.

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
According to speedtest.net: 11.9Mbps down 500kbps up
David Long Nov. 2, 2018, 5:22 p.m. UTC | #5 
On 11/2/18 4:54 AM, Marc Zyngier wrote:
> On 02/11/18 01:18, David Long wrote:

>> On 10/31/18 9:56 AM, David Long wrote:

>>> From: "David A. Long" <dave.long@linaro.org>

>>>

>>> V4.9 backport of spectre patches from Russell M. King's spectre branch.

>>> Patches not yet in upstream are excluded.

>>>

>>> Marc Zyngier (2):

>>>     ARM: KVM: invalidate BTB on guest exit for Cortex-A12/A17

>>>     ARM: KVM: invalidate icache on guest exit for Cortex-A15

>>>

>>> Russell King (22):

>>>     ARM: add more CPU part numbers for Cortex and Brahma B15 CPUs

>>>     ARM: bugs: prepare processor bug infrastructure

>>>     ARM: bugs: hook processor bug checking into SMP and suspend paths

>>>     ARM: bugs: add support for per-processor bug checking

>>>     ARM: spectre: add Kconfig symbol for CPUs vulnerable to Spectre

>>>     ARM: spectre-v2: harden branch predictor on context switches

>>>     ARM: spectre-v2: add Cortex A8 and A15 validation of the IBE bit

>>>     ARM: spectre-v2: harden user aborts in kernel space

>>>     ARM: spectre-v2: add firmware based hardening

>>>     ARM: spectre-v2: warn about incorrect context switching functions

>>>     ARM: spectre-v2: KVM: invalidate icache on guest exit for Brahma B15

>>>     ARM: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling

>>>     ARM: KVM: report support for SMCCC_ARCH_WORKAROUND_1

>>>     ARM: spectre-v1: add speculation barrier (csdb) macros

>>>     ARM: spectre-v1: add array_index_mask_nospec() implementation

>>>     ARM: spectre-v1: fix syscall entry

>>>     ARM: signal: copy registers using __copy_from_user()

>>>     ARM: vfp: use __copy_from_user() when restoring VFP state

>>>     ARM: oabi-compat: copy semops using __copy_from_user()

>>>     ARM: use __inttype() in get_user()

>>>     ARM: spectre-v1: use get_user() for __get_user()

>>>     ARM: spectre-v1: mitigate user accesses

>>>

>>>    arch/arm/include/asm/assembler.h   |  12 ++

>>>    arch/arm/include/asm/barrier.h     |  32 ++++++

>>>    arch/arm/include/asm/bugs.h        |   6 +-

>>>    arch/arm/include/asm/cp15.h        |   3 +

>>>    arch/arm/include/asm/cputype.h     |   8 ++

>>>    arch/arm/include/asm/kvm_asm.h     |   2 -

>>>    arch/arm/include/asm/kvm_host.h    |  14 ++-

>>>    arch/arm/include/asm/kvm_mmu.h     |  23 +++-

>>>    arch/arm/include/asm/proc-fns.h    |   4 +

>>>    arch/arm/include/asm/system_misc.h |  15 +++

>>>    arch/arm/include/asm/thread_info.h |   4 +-

>>>    arch/arm/include/asm/uaccess.h     |  26 +++--

>>>    arch/arm/kernel/Makefile           |   1 +

>>>    arch/arm/kernel/bugs.c             |  18 +++

>>>    arch/arm/kernel/entry-common.S     |  18 ++-

>>>    arch/arm/kernel/entry-header.S     |  25 +++++

>>>    arch/arm/kernel/signal.c           |  55 ++++-----

>>>    arch/arm/kernel/smp.c              |   4 +

>>>    arch/arm/kernel/suspend.c          |   2 +

>>>    arch/arm/kernel/sys_oabi-compat.c  |   8 +-

>>>    arch/arm/kvm/hyp/hyp-entry.S       | 110 +++++++++++++++++-

>>>    arch/arm/lib/copy_from_user.S      |   9 ++

>>>    arch/arm/mm/Kconfig                |  23 ++++

>>>    arch/arm/mm/Makefile               |   2 +-

>>>    arch/arm/mm/fault.c                |   3 +

>>>    arch/arm/mm/proc-macros.S          |   3 +-

>>>    arch/arm/mm/proc-v7-2level.S       |   6 -

>>>    arch/arm/mm/proc-v7-bugs.c         | 174 +++++++++++++++++++++++++++++

>>>    arch/arm/mm/proc-v7.S              | 154 +++++++++++++++++++------

>>>    arch/arm/vfp/vfpmodule.c           |  17 ++-

>>>    30 files changed, 674 insertions(+), 107 deletions(-)

>>>    create mode 100644 arch/arm/kernel/bugs.c

>>>    create mode 100644 arch/arm/mm/proc-v7-bugs.c

>>>

>>

>> kvm-unit-test'ing of this results in a hypervisor panic that doesn't

>> happen without the patches. This needs to be figured out before it is

>> accepted into stable. Looks like a V2 will be needed. Clearly kernelci

>> testing alone is not sufficient when dealing with kvm changes.

> 

> How about posting the panic message, a description of what you were

> doing when that happened, and details of the configuration (HW used,

> Thumb-2 or not...)? If you cannot perform the analysis yourself, at

> least give us enough information to help you.

> 

> Thanks,

> 

> 	M.

> 


The goal of my email was to make sure this didn't end up going out as-is 
in the next v4.9-stable, not to beg help debugging. But I can see how 
the mail might have been interpreted differently.  The intent was that I 
was going to do the "figuring out", and ask for help when and if I 
needed it.

If anyone is interested though:  The test is the kvm-unit-tests run on 
an exynos arndale 5250, built from the default config for that platform 
plus turning on most of the virtualization config lines I could find, 
and not trying to turn on any thumb-2. The problem goes away if I remove 
patches 12/24 and 13/24 (and probably just one of those). The kernel 
panic messages are below:


> [ 1388.419157] Kernel panic - not syncing: 

> [ 1388.419157] HYP panic: UNDEF PC:40000000 CPSR:000001d3

> [ 1388.426742] CPU: 0 PID: 1345 Comm: qemu-system-arm Not tainted 4.9.135-00024-g0cf93698e984 #1

> [ 1388.435242] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)

> [ 1388.441324] [<c0222bd4>] (unwind_backtrace) from [<c021f8d8>] (show_stack+0x10/0x14)

> [ 1388.449049] [<c021f8d8>] (show_stack) from [<c044a1b0>] (dump_stack+0x78/0x8c)

> [ 1388.456254] [<c044a1b0>] (dump_stack) from [<c02b02b4>] (panic+0xdc/0x258)

> [ 1388.463104] [<c02b02b4>] (panic) from [<c020fa98>] (kvm_arch_vcpu_ioctl_run+0xa4/0x468)

> [ 1388.471096] [<c020fa98>] (kvm_arch_vcpu_ioctl_run) from [<c0208dc4>] (kvm_vcpu_ioctl+0x374/0x6fc)

> [ 1388.479950] [<c0208dc4>] (kvm_vcpu_ioctl) from [<c03100ac>] (do_vfs_ioctl+0x9c/0x7e4)

> [ 1388.487760] [<c03100ac>] (do_vfs_ioctl) from [<c0310828>] (SyS_ioctl+0x34/0x58)

> [ 1388.495052] [<c0310828>] (SyS_ioctl) from [<c021c8c0>] (ret_fast_syscall+0x0/0x40)

> [ 1388.502604] CPU1: stopping

> [ 1388.505285] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.9.135-00024-g0cf93698e984 #1

> [ 1388.513014] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)

> [ 1388.519091] [<c0222bd4>] (unwind_backtrace) from [<c021f8d8>] (show_stack+0x10/0x14)

> [ 1388.526820] [<c021f8d8>] (show_stack) from [<c044a1b0>] (dump_stack+0x78/0x8c)

> [ 1388.534025] [<c044a1b0>] (dump_stack) from [<c0221e98>] (handle_IPI+0x198/0x1ac)

> [ 1388.541401] [<c0221e98>] (handle_IPI) from [<c0201540>] (gic_handle_irq+0x94/0x98)

> [ 1388.548953] [<c0201540>] (gic_handle_irq) from [<c0220378>] (__irq_svc+0x58/0x8c)

> [ 1388.556415] Exception stack(0xee89bf50 to 0xee89bf98)

> [ 1388.561443] bf40:                                     00000000 c10308b4 00000001 2e193000

> [ 1388.569610] bf60: ffffe000 c1003bf4 00000000 00000143 00000000 eeffb148 49344906 c10308b4

> [ 1388.577768] bf80: fffffff5 ee89bfa0 c065b03c c065b12c 600d0013 ffffffff

> [ 1388.584365] [<c0220378>] (__irq_svc) from [<c065b12c>] (cpuidle_enter_state+0x264/0x320)

> [ 1388.592443] [<c065b12c>] (cpuidle_enter_state) from [<c026a1c4>] (cpu_startup_entry+0x168/0x228)

> [ 1388.601208] [<c026a1c4>] (cpu_startup_entry) from [<402016ec>] (0x402016ec)

> [ 1388.608161] ---[ end Kernel panic - not syncing: 

> [ 1388.608161] HYP panic: UNDEF PC:40000000 CPSR:000001d3

> [ 1388.619517] ------------[ cut here ]------------

> [ 1388.622669] WARNING: CPU: 0 PID: 1345 at kernel/workqueue.c:857 wq_worker_waking_up+0x78/0x80

> [ 1388.631175] Modules linked in: s5p_mfc videobuf2_dma_contig v4l2_common videobuf2_memops videobuf2_v4l2 videobuf2_core vid

> eodev media

> [ 1388.643146] CPU: 0 PID: 1345 Comm: qemu-system-arm Not tainted 4.9.135-00024-g0cf93698e984 #1

> [ 1388.651660] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)

> [ 1388.657735] [<c0222bd4>] (unwind_backtrace) from [<c021f8d8>] (show_stack+0x10/0x14)

> [ 1388.665467] [<c021f8d8>] (show_stack) from [<c044a1b0>] (dump_stack+0x78/0x8c)

> [ 1388.672670] [<c044a1b0>] (dump_stack) from [<c0230070>] (__warn+0xe8/0x100)

> [ 1388.679608] [<c0230070>] (__warn) from [<c0230138>] (warn_slowpath_null+0x20/0x28)

> [ 1388.687164] [<c0230138>] (warn_slowpath_null) from [<c0246b28>] (wq_worker_waking_up+0x78/0x80)

> [ 1388.695850] [<c0246b28>] (wq_worker_waking_up) from [<c0251cd0>] (ttwu_do_activate+0x58/0x70)

> [ 1388.704354] [<c0251cd0>] (ttwu_do_activate) from [<c0253898>] (try_to_wake_up+0x19c/0x290)

> [ 1388.712598] [<c0253898>] (try_to_wake_up) from [<c0269a84>] (autoremove_wake_function+0xc/0x34)

> [ 1388.721280] [<c0269a84>] (autoremove_wake_function) from [<c0269494>] (__wake_up_common+0x4c/0x80)

> [ 1388.730220] [<c0269494>] (__wake_up_common) from [<c0269500>] (__wake_up+0x38/0x4c)

> [ 1388.737858] [<c0269500>] (__wake_up) from [<c06325b0>] (i2c_s3c_irq_nextbyte+0x488/0x4bc)

> [ 1388.746016] [<c06325b0>] (i2c_s3c_irq_nextbyte) from [<c063332c>] (s3c24xx_i2c_irq+0x34/0x78)

> [ 1388.754523] [<c063332c>] (s3c24xx_i2c_irq) from [<c0278688>] (__handle_irq_event_percpu+0x50/0x11c)

> [ 1388.763550] [<c0278688>] (__handle_irq_event_percpu) from [<c0278770>] (handle_irq_event_percpu+0x1c/0x58)

> [ 1388.773184] [<c0278770>] (handle_irq_event_percpu) from [<c02787e4>] (handle_irq_event+0x38/0x5c)

> [ 1388.782039] [<c02787e4>] (handle_irq_event) from [<c027bb70>] (handle_fasteoi_irq+0xd0/0x1a0)

> [ 1388.790544] [<c027bb70>] (handle_fasteoi_irq) from [<c0277984>] (generic_handle_irq+0x24/0x34)

> [ 1388.799137] [<c0277984>] (generic_handle_irq) from [<c0277eac>] (__handle_domain_irq+0x7c/0xec)

> [ 1388.807816] [<c0277eac>] (__handle_domain_irq) from [<c0201500>] (gic_handle_irq+0x54/0x98)

> [ 1388.816149] [<c0201500>] (gic_handle_irq) from [<c0220378>] (__irq_svc+0x58/0x8c)

> [ 1388.823611] Exception stack(0xed63be00 to 0xed63be48)

> [ 1388.828641] be00: 00002bee 00000007 fac81000 c0680298 c10836d8 00005dbf c26f7ba3 000000c8

> [ 1388.836806] be20: c0c04390 00000063 199996c0 00000000 00000007 ed63be50 c0222558 c0447fec

> [ 1388.844963] be40: 80000153 ffffffff

> [ 1388.848432] [<c0220378>] (__irq_svc) from [<c0447fec>] (__timer_delay+0x44/0x58)

> [ 1388.855818] [<c0447fec>] (__timer_delay) from [<c02b0418>] (panic+0x240/0x258)

> [ 1388.863022] [<c02b0418>] (panic) from [<c020fa98>] (kvm_arch_vcpu_ioctl_run+0xa4/0x468)

> [ 1388.871009] [<c020fa98>] (kvm_arch_vcpu_ioctl_run) from [<c0208dc4>] (kvm_vcpu_ioctl+0x374/0x6fc)

> [ 1388.879862] [<c0208dc4>] (kvm_vcpu_ioctl) from [<c03100ac>] (do_vfs_ioctl+0x9c/0x7e4)

> [ 1388.887673] [<c03100ac>] (do_vfs_ioctl) from [<c0310828>] (SyS_ioctl+0x34/0x58)

> [ 1388.894965] [<c0310828>] (SyS_ioctl) from [<c021c8c0>] (ret_fast_syscall+0x0/0x40)

> [ 1388.902512] ---[ end trace 9f81df9f2aa3f954 ]---



Thanks,
-dl