]> git.baikalelectronics.ru Git - kernel.git/commit
KVM: s390: pv: avoid double free of sida page
authorClaudio Imbrenda <imbrenda@linux.ibm.com>
Mon, 20 Sep 2021 13:24:50 +0000 (15:24 +0200)
committerChristian Borntraeger <borntraeger@de.ibm.com>
Mon, 25 Oct 2021 07:20:38 +0000 (09:20 +0200)
commit87b97a23b71d81e623b488ddde20f3765224f02b
tree7754a919724458c5cae701b701177d2c80f86c7e
parenteeca85b682e22d59bbc922aa290a8159f2cadb93
KVM: s390: pv: avoid double free of sida page

If kvm_s390_pv_destroy_cpu is called more than once, we risk calling
free_page on a random page, since the sidad field is aliased with the
gbea, which is not guaranteed to be zero.

This can happen, for example, if userspace calls the KVM_PV_DISABLE
IOCTL, and it fails, and then userspace calls the same IOCTL again.
This scenario is only possible if KVM has some serious bug or if the
hardware is broken.

The solution is to simply return successfully immediately if the vCPU
was already non secure.

Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Fixes: 511ead40324d7036649fae652be4b4c2304f35f0 ("KVM: S390: protvirt: Introduce instruction data area bounce buffer")
Reviewed-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20210920132502.36111-3-imbrenda@linux.ibm.com>
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
arch/s390/kvm/pv.c