[PATCH v3 0/9] s390x: even more hardening fixes

Eric Farman posted 9 patches 3 days, 23 hours ago
hw/intc/s390_flic_kvm.c |  8 +++++---
hw/s390x/css.c          | 28 ++++++++++++++++++++--------
hw/s390x/sclp.c         |  6 ++++++
hw/s390x/virtio-ccw.c   | 29 ++++++++++++++++++++++++++++-
hw/vfio/ap.c            |  4 ++++
hw/vfio/ccw.c           |  2 ++
6 files changed, 65 insertions(+), 12 deletions(-)
[PATCH v3 0/9] s390x: even more hardening fixes
Posted by Eric Farman 3 days, 23 hours ago
A small pile of additional hardening fixes.

v2->v3:
 - Collect reviews/acks (Thanks Christian, Matt!)
 - [CB] Patch 7: Rename realloc variable to new_buf
 - [CB] Patch 8: Move check of at_byte before check of command chaining flag

v1->v2: https://lore.kernel.org/qemu-devel/20260921184305.3791311-1-farman@linux.ibm.com/
 - Collect reviews/acks (Thanks Matt, Christian, Tony!)
 - [MR] Add fixes/stable tags to patches 1, 2, 8
 - [CB, MR] Drop fixes/stable tags from patches 5, 6
 - [CB] Patch 2: Unwind indicators upon re-set, instead of rejecting it
 - [MR] Patch 7: Use a temporary variable on g_try_realloc() call
 - [MR] Patch 8: Use CCW datastream to increment no-data count

v1: https://lore.kernel.org/qemu-devel/20260916203242.2509787-1-farman@linux.ibm.com/

Eric Farman (9):
  virtio-ccw: Limit ISC to MAX_ISC
  virtio-ccw: Do not overwrite existing indicators
  virtio-ccw: Fix rc check in read configuration path
  vfio/ap: Clear error object pointer after non-fatal errors
  vfio/ccw: Clear error object pointer after non-fatal errors
  hw/intc/s390x: Fix payload to KVM_HAS_DEVICE_ATTR ioctl
  hw/intc/s390x: Fix check after g_try_realloc
  s390x/css: Use datastream pointer for checking no-data CCWs
  s390x/sclp: Reject short headers in PV case

 hw/intc/s390_flic_kvm.c |  8 +++++---
 hw/s390x/css.c          | 28 ++++++++++++++++++++--------
 hw/s390x/sclp.c         |  6 ++++++
 hw/s390x/virtio-ccw.c   | 29 ++++++++++++++++++++++++++++-
 hw/vfio/ap.c            |  4 ++++
 hw/vfio/ccw.c           |  2 ++
 6 files changed, 65 insertions(+), 12 deletions(-)

-- 
2.53.0
Re: [PATCH v3 0/9] s390x: even more hardening fixes
Posted by Eric Farman 1 day, 6 hours ago

On 9/22/26 4:15 PM, Eric Farman wrote:
> A small pile of additional hardening fixes.
> 
> v2->v3:
>   - Collect reviews/acks (Thanks Christian, Matt!)
>   - [CB] Patch 7: Rename realloc variable to new_buf
>   - [CB] Patch 8: Move check of at_byte before check of command chaining flag
> 
> v1->v2: https://lore.kernel.org/qemu-devel/20260921184305.3791311-1-farman@linux.ibm.com/
>   - Collect reviews/acks (Thanks Matt, Christian, Tony!)
>   - [MR] Add fixes/stable tags to patches 1, 2, 8
>   - [CB, MR] Drop fixes/stable tags from patches 5, 6
>   - [CB] Patch 2: Unwind indicators upon re-set, instead of rejecting it
>   - [MR] Patch 7: Use a temporary variable on g_try_realloc() call
>   - [MR] Patch 8: Use CCW datastream to increment no-data count
> 
> v1: https://lore.kernel.org/qemu-devel/20260916203242.2509787-1-farman@linux.ibm.com/
> 
> Eric Farman (9):
>    virtio-ccw: Limit ISC to MAX_ISC
>    virtio-ccw: Do not overwrite existing indicators
>    virtio-ccw: Fix rc check in read configuration path
>    vfio/ap: Clear error object pointer after non-fatal errors
>    vfio/ccw: Clear error object pointer after non-fatal errors
>    hw/intc/s390x: Fix payload to KVM_HAS_DEVICE_ATTR ioctl
>    hw/intc/s390x: Fix check after g_try_realloc
>    s390x/css: Use datastream pointer for checking no-data CCWs
>    s390x/sclp: Reject short headers in PV case
> 
>   hw/intc/s390_flic_kvm.c |  8 +++++---
>   hw/s390x/css.c          | 28 ++++++++++++++++++++--------
>   hw/s390x/sclp.c         |  6 ++++++
>   hw/s390x/virtio-ccw.c   | 29 ++++++++++++++++++++++++++++-
>   hw/vfio/ap.c            |  4 ++++
>   hw/vfio/ccw.c           |  2 ++
>   6 files changed, 65 insertions(+), 12 deletions(-)
> 

Applied for 11.2; thank you!
Re: [PATCH v3 0/9] s390x: even more hardening fixes
Posted by Christian Borntraeger 3 days, 11 hours ago
Am 22.09.26 um 22:15 schrieb Eric Farman:
> A small pile of additional hardening fixes.
> 
> v2->v3:
>   - Collect reviews/acks (Thanks Christian, Matt!)
>   - [CB] Patch 7: Rename realloc variable to new_buf
>   - [CB] Patch 8: Move check of at_byte before check of command chaining flag
> 
> v1->v2: https://lore.kernel.org/qemu-devel/20260921184305.3791311-1-farman@linux.ibm.com/
>   - Collect reviews/acks (Thanks Matt, Christian, Tony!)
>   - [MR] Add fixes/stable tags to patches 1, 2, 8
>   - [CB, MR] Drop fixes/stable tags from patches 5, 6
>   - [CB] Patch 2: Unwind indicators upon re-set, instead of rejecting it
>   - [MR] Patch 7: Use a temporary variable on g_try_realloc() call
>   - [MR] Patch 8: Use CCW datastream to increment no-data count
> 
> v1: https://lore.kernel.org/qemu-devel/20260916203242.2509787-1-farman@linux.ibm.com/
> 
> Eric Farman (9):
>    virtio-ccw: Limit ISC to MAX_ISC
>    virtio-ccw: Do not overwrite existing indicators
>    virtio-ccw: Fix rc check in read configuration path
>    vfio/ap: Clear error object pointer after non-fatal errors
>    vfio/ccw: Clear error object pointer after non-fatal errors
>    hw/intc/s390x: Fix payload to KVM_HAS_DEVICE_ATTR ioctl
>    hw/intc/s390x: Fix check after g_try_realloc
>    s390x/css: Use datastream pointer for checking no-data CCWs
>    s390x/sclp: Reject short headers in PV case
Series looks sane to me.