[PATCH] hvm/hpet: Alter hpet_init() to take a domain rather than vcpu

[PATCH] hvm/hpet: Alter hpet_init() to take a domain rather than vcpu

[Andrew Cooper]

There is nothing vcpu-specific about hpet_init(); all it does is follow the
vcpu's domain pointer to get at the domain vhpet state.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
CC: Jan Beulich <JBeulich@suse.com>

---

Despite the comment in scope in hvm_vcpu_initialise(), the call can't
currently be moved to hvm_domain_initialise() as the extra hpet_deinit() (in
an error path) needs an allocated vcpu which wouldn't be present at that
point.  I need more tuits to disentangle that mess.
---
 xen/arch/x86/hvm/hpet.c       |    6 +++---
 xen/arch/x86/hvm/hvm.c        |    2 +-
 xen/include/asm-x86/hvm/vpt.h |    2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/xen/arch/x86/hvm/hpet.c b/xen/arch/x86/hvm/hpet.c
index 167994e..81ad0d6 100644
--- a/xen/arch/x86/hvm/hpet.c
+++ b/xen/arch/x86/hvm/hpet.c
@@ -611,9 +611,9 @@ static int hpet_load(struct domain *d, hvm_domain_context_t *h)
 
 HVM_REGISTER_SAVE_RESTORE(HPET, hpet_save, hpet_load, 1, HVMSR_PER_DOM);
 
-void hpet_init(struct vcpu *v)
+void hpet_init(struct domain *d)
 {
-    HPETState *h = vcpu_vhpet(v);
+    HPETState *h = domain_vhpet(d);
     int i;
 
     memset(h, 0, sizeof(HPETState));
@@ -665,7 +665,7 @@ void hpet_deinit(struct domain *d)
 void hpet_reset(struct domain *d)
 {
     hpet_deinit(d);
-    hpet_init(d->vcpu[0]);
+    hpet_init(d);
 }
 
 /*
diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
index 51011ea..14db6dc 100644
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -2282,7 +2282,7 @@ int hvm_vcpu_initialise(struct vcpu *v)
     {
         /* NB. All these really belong in hvm_domain_initialise(). */
         pmtimer_init(v);
-        hpet_init(v);
+        hpet_init(d);
  
         /* Init guest TSC to start from zero. */
         hvm_set_guest_tsc(v, 0);
diff --git a/xen/include/asm-x86/hvm/vpt.h b/xen/include/asm-x86/hvm/vpt.h
index 41159d8..105fbdb 100644
--- a/xen/include/asm-x86/hvm/vpt.h
+++ b/xen/include/asm-x86/hvm/vpt.h
@@ -190,7 +190,7 @@ void pmtimer_deinit(struct domain *d);
 void pmtimer_reset(struct domain *d);
 int pmtimer_change_ioport(struct domain *d, unsigned int version);
 
-void hpet_init(struct vcpu *v);
+void hpet_init(struct domain *d);
 void hpet_deinit(struct domain *d);
 void hpet_reset(struct domain *d);
 
-- 
1.7.10.4

Re: [PATCH] hvm/hpet: Alter hpet_init() to take a domain rather than vcpu

[Jan Beulich]

>>> On 25.07.14 at 15:19, <andrew.cooper3@citrix.com> wrote:
> There is nothing vcpu-specific about hpet_init(); all it does is follow the
> vcpu's domain pointer to get at the domain vhpet state.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> CC: Jan Beulich <JBeulich@suse.com>
> 
> ---
> 
> Despite the comment in scope in hvm_vcpu_initialise(), the call can't
> currently be moved to hvm_domain_initialise() as the extra hpet_deinit() (in
> an error path) needs an allocated vcpu which wouldn't be present at that
> point.  I need more tuits to disentangle that mess.

And indeed I was about to ask that very question. Thanks for the
patch in any case!

Jan

Re: [PATCH] hvm/hpet: Alter hpet_init() to take a domain rather than vcpu

[Andrew Cooper]

On 25/07/14 16:19, Jan Beulich wrote:
>>>> On 25.07.14 at 15:19, <andrew.cooper3@citrix.com> wrote:
>> There is nothing vcpu-specific about hpet_init(); all it does is follow the
>> vcpu's domain pointer to get at the domain vhpet state.
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
>> CC: Jan Beulich <JBeulich@suse.com>
>>
>> ---
>>
>> Despite the comment in scope in hvm_vcpu_initialise(), the call can't
>> currently be moved to hvm_domain_initialise() as the extra hpet_deinit() (in
>> an error path) needs an allocated vcpu which wouldn't be present at that
>> point.  I need more tuits to disentangle that mess.
> And indeed I was about to ask that very question. Thanks for the
> patch in any case!
>
> Jan
>

I was wondering whether to extend the comment in the code, as it is not
very obvious that hpet_deinit(d) has an implicit requirement on
d->vcpu[0], but opted for the lazy route.

All of this is wrapped up in the pl_time which combines the vrtc, vhpet
and vpmt, where the vpmt has a hook onto vcpu[0], which all time pieces
share.

In an ideal world, I would expect there to be a vpmt per vcpu, and
vrtc/vhpet to hook onto the vioapic or vi8259 for the purpose of
delivering interrupts.  Part of the problem is that vcpu[0]'s guest time
is uses as a substitute for a domain wide time.

~Andrew