leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

KVM: s390: Limit sthyi execution

Browse Source 
Store hypervisor information is a valid instruction not only in
supervisor state but also in problem state, i.e. the guest's
userspace. Its execution is not only computational and memory
intensive, but also has to get hold of the ipte lock to write to the
guest's memory.

This lock is not intended to be held often and long, especially not
from the untrusted guest userspace. Therefore we apply rate limiting
of sthyi executions per VM.

Signed-off-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Acked-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
This commit is contained in:
Janosch Frank 2016-05-10 15:03:42 +02:00 committed by Christian Borntraeger
parent 95ca2cb579
commit 7d0a5e6241
3 changed files with 14 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
arch/s390/include/asm/kvm_host.h
View File

@ -652,6 +652,7 @@ struct kvm_arch{
wait_queue_head_t ipte_wq;
int ipte_lock_count;
struct mutex ipte_mutex;
struct ratelimit_state sthyi_limit;
spinlock_t start_stop_lock;
struct sie_page2 *sie_page2;
struct kvm_s390_cpu_model model;

2
arch/s390/kvm/kvm-s390.c
View File

@ -1151,6 +1151,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
rc = -ENOMEM;
ratelimit_state_init(&kvm->arch.sthyi_limit, 5 * HZ, 500);
kvm->arch.use_esca = 0; /* start with basic SCA */
rwlock_init(&kvm->arch.sca_lock);
kvm->arch.sca = (struct bsca_block *) get_zeroed_page(GFP_KERNEL);

11
arch/s390/kvm/sthyi.c
View File

@ -12,6 +12,7 @@
#include <linux/errno.h>
#include <linux/pagemap.h>
#include <linux/vmalloc.h>
#include <linux/ratelimit.h>
#include <asm/kvm_host.h>
#include <asm/asm-offsets.h>
@ -403,6 +404,16 @@ int handle_sthyi(struct kvm_vcpu *vcpu)
u64 code, addr, cc = 0;
struct sthyi_sctns *sctns = NULL;
/*
* STHYI requires extensive locking in the higher hypervisors
* and is very computational/memory expensive. Therefore we
* ratelimit the executions per VM.
*/
if (!__ratelimit(&vcpu->kvm->arch.sthyi_limit)) {
kvm_s390_retry_instr(vcpu);
return 0;
}
kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
code = vcpu->run->s.regs.gprs[reg1];
addr = vcpu->run->s.regs.gprs[reg2];