| Message ID | 20230215160252.14672-2-nrb@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | s390: nmi: fix virtual-physical address confusion | expand |
On Wed, 15 Feb 2023 17:02:52 +0100 Nico Boehr <nrb@linux.ibm.com> wrote: > When a machine check is received while in SIE, it is reinjected into the > guest in some cases. The respective code needs to access the sie_block, > which is taken from the backed up R14. > > Since reinjection only occurs while we are in SIE (i.e. between the > labels sie_entry and sie_leave in entry.S and thus if CIF_MCCK_GUEST is > set), the backed up R14 will always contain a physical address in > s390_backup_mcck_info. > > This currently works, because virtual and physical addresses are > the same. > > Add phys_to_virt() to resolve the virtual-physical confusion. > > Signed-off-by: Nico Boehr <nrb@linux.ibm.com> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com> > --- > arch/s390/kernel/nmi.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c > index 5dbf274719a9..322160328866 100644 > --- a/arch/s390/kernel/nmi.c > +++ b/arch/s390/kernel/nmi.c > @@ -347,7 +347,7 @@ static void notrace s390_backup_mcck_info(struct pt_regs *regs) > > /* r14 contains the sie block, which was set in sie64a */ > struct kvm_s390_sie_block *sie_block = > - (struct kvm_s390_sie_block *) regs->gprs[14]; > + (struct kvm_s390_sie_block *)phys_to_virt(regs->gprs[14]); > > if (sie_block == NULL) > /* Something's seriously wrong, stop system. */
On Wed, Feb 15, 2023 at 05:02:52PM +0100, Nico Boehr wrote: > When a machine check is received while in SIE, it is reinjected into the > guest in some cases. The respective code needs to access the sie_block, > which is taken from the backed up R14. > > Since reinjection only occurs while we are in SIE (i.e. between the > labels sie_entry and sie_leave in entry.S and thus if CIF_MCCK_GUEST is > set), the backed up R14 will always contain a physical address in > s390_backup_mcck_info. > > This currently works, because virtual and physical addresses are > the same. > > Add phys_to_virt() to resolve the virtual-physical confusion. > > Signed-off-by: Nico Boehr <nrb@linux.ibm.com> > --- > arch/s390/kernel/nmi.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c > index 5dbf274719a9..322160328866 100644 > --- a/arch/s390/kernel/nmi.c > +++ b/arch/s390/kernel/nmi.c > @@ -347,7 +347,7 @@ static void notrace s390_backup_mcck_info(struct pt_regs *regs) > > /* r14 contains the sie block, which was set in sie64a */ > struct kvm_s390_sie_block *sie_block = > - (struct kvm_s390_sie_block *) regs->gprs[14]; > + (struct kvm_s390_sie_block *)phys_to_virt(regs->gprs[14]); Casting to (struct kvm_s390_sie_block *) is not superfluous, since phys_to_virt() returns (void *). > > if (sie_block == NULL) > /* Something's seriously wrong, stop system. */ Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
On Thu, Feb 16, 2023 at 07:17:07AM +0100, Alexander Gordeev wrote:
> Casting to (struct kvm_s390_sie_block *) is not superfluous,
s/not//
On 2/16/23 10:38, Alexander Gordeev wrote: > On Thu, Feb 16, 2023 at 07:17:07AM +0100, Alexander Gordeev wrote: >> Casting to (struct kvm_s390_sie_block *) is not superfluous, > > s/not// Do you want to pick this up or should it go through the kvm tree?
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index 5dbf274719a9..322160328866 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -347,7 +347,7 @@ static void notrace s390_backup_mcck_info(struct pt_regs *regs) /* r14 contains the sie block, which was set in sie64a */ struct kvm_s390_sie_block *sie_block = - (struct kvm_s390_sie_block *) regs->gprs[14]; + (struct kvm_s390_sie_block *)phys_to_virt(regs->gprs[14]); if (sie_block == NULL) /* Something's seriously wrong, stop system. */
When a machine check is received while in SIE, it is reinjected into the guest in some cases. The respective code needs to access the sie_block, which is taken from the backed up R14. Since reinjection only occurs while we are in SIE (i.e. between the labels sie_entry and sie_leave in entry.S and thus if CIF_MCCK_GUEST is set), the backed up R14 will always contain a physical address in s390_backup_mcck_info. This currently works, because virtual and physical addresses are the same. Add phys_to_virt() to resolve the virtual-physical confusion. Signed-off-by: Nico Boehr <nrb@linux.ibm.com> --- arch/s390/kernel/nmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)