On 2/12/21 9:39 AM, Pierre Morel wrote:
When we receive intercept a DIAG_9C from the guest we verify
that the target real CPU associated with the virtual CPU
designated by the guest is running and if not we forward the
DIAG_9C to the target real CPU.
To avoid a diag9c storm we allow a maximal rate of diag9c forwarding.
The rate is calculated as a count per second defined as a
new parameter of the s390 kvm module: diag9c_forwarding_hz .
The default value is to not forward diag9c.
Signed-off-by: Pierre Morel <pmorel@xxxxxxxxxxxxx>
---
Documentation/virt/kvm/s390-diag.rst | 33 ++++++++++++++++++++++++++++
arch/s390/include/asm/kvm_host.h | 1 +
arch/s390/include/asm/smp.h | 1 +
arch/s390/kernel/smp.c | 1 +
arch/s390/kvm/diag.c | 31 +++++++++++++++++++++++---
arch/s390/kvm/kvm-s390.c | 6 +++++
arch/s390/kvm/kvm-s390.h | 8 +++++++
7 files changed, 78 insertions(+), 3 deletions(-)
diff --git a/Documentation/virt/kvm/s390-diag.rst b/Documentation/virt/kvm/s390-diag.rst
index eaac4864d3d6..a6371bc4ea90 100644
--- a/Documentation/virt/kvm/s390-diag.rst
+++ b/Documentation/virt/kvm/s390-diag.rst
@@ -84,3 +84,36 @@ If the function code specifies 0x501, breakpoint functions may be performed.
This function code is handled by userspace.
This diagnose function code has no subfunctions and uses no parameters.
+
+
+DIAGNOSE function code 'X'9C - Voluntary Time Slice Yield
+---------------------------------------------------------
+
+General register 1 contains the target CPU address.
+
+In a guest of a hypervisor like LPAR, KVM or z/VM using shared host CPUs,
+DIAGNOSE with function code 'X'9C may improve system performance by
+yielding the host CPU on which the guest CPU is running to be assigned
+to another guest CPU, preferably the logical CPU containing the specified
+target CPU.
+
+
+DIAG 'X'9C forwarding
++++++++++++++++++++++
+
+Under KVM, the guest operating system may send a DIAGNOSE code 'X'9C to
+the host when it fails to acquire a spinlock for a virtual CPU
+and detects that the host CPU on which the virtual guest CPU owner is
+assigned to is not running to try to get this host CPU running and
+consequently the guest virtual CPU running and freeing the lock.
+
+However, on the logical partition the real CPU on which the previously
+targeted host CPU is assign may itself not be running.
+By forwarding the DIAGNOSE code 'X'9C, initially sent by the guest,
+from the host to LPAR hypervisor, this one will hopefully schedule
+the host CPU which will let KVM run the target guest CPU.
+
+diag9c_forwarding_hz
+ KVM kernel parameter allowing to specify the maximum number of DIAGNOSE
+ 'X'9C forwarding per second in the purpose of avoiding a DIAGNOSE 'X'9C
+ forwarding storm.
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 527776a1f076..cb19508c22fb 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -456,6 +456,7 @@ struct kvm_vcpu_stat {
u64 diagnose_44;
u64 diagnose_9c;
u64 diagnose_9c_ignored;
+ u64 diagnose_9c_forward;
u64 diagnose_258;
u64 diagnose_308;
u64 diagnose_500;
diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h
index 01e360004481..e317fd4866c1 100644
--- a/arch/s390/include/asm/smp.h
+++ b/arch/s390/include/asm/smp.h
@@ -63,5 +63,6 @@ extern void __noreturn cpu_die(void);
extern void __cpu_die(unsigned int cpu);
extern int __cpu_disable(void);
extern void schedule_mcck_handler(void);
+void notrace smp_yield_cpu(int cpu);
#endif /* __ASM_SMP_H */
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 27c763014114..15e207a671fd 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -422,6 +422,7 @@ void notrace smp_yield_cpu(int cpu)
asm volatile("diag %0,0,0x9c"
: : "d" (pcpu_devices[cpu].address));
}
+EXPORT_SYMBOL(smp_yield_cpu);
/*
* Send cpus emergency shutdown signal. This gives the cpus the
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index 5b8ec1c447e1..34cf41fa6fa2 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -150,6 +150,19 @@ static int __diag_time_slice_end(struct kvm_vcpu *vcpu)
return 0;
}
+static unsigned int forward_cnt;
+static unsigned long cur_slice;
+
+static int diag9c_forwarding_overrun(void)
+{
+ /* Reset the count on a new slice */
+ if (time_after(jiffies, cur_slice)) {
+ cur_slice = jiffies;
+ forward_cnt = diag9c_forwarding_hz / HZ;
+ }
+ return forward_cnt-- ? 1 : 0;
/o\
seems a "<= 0 " has been forgotten here
I send the update asap.
regards,
Pierre
--
Pierre Morel
IBM Lab Boeblingen
