With alternative_call() capable of handling compound types, the three
remaining hooks can be optimised at boot time too.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Daniel De Graaf <dgdegra@tycho.nsa.gov>
CC: Daniel Smith <dpsmith@apertussolutions.com>

I'm on the fence as to whether to declare this as fixing "xsm: convert xsm_ops
hook calls to alternative call"
---
 xen/include/xsm/xsm.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/xen/include/xsm/xsm.h b/xen/include/xsm/xsm.h
index 0b360e1a3553..82458066f625 100644
--- a/xen/include/xsm/xsm.h
+++ b/xen/include/xsm/xsm.h
@@ -579,13 +579,13 @@ static inline int xsm_hypfs_op(xsm_default_t def)
 
 static inline long xsm_do_xsm_op(XEN_GUEST_HANDLE_PARAM(void) op)
 {
-    return xsm_ops.do_xsm_op(op);
+    return alternative_call(xsm_ops.do_xsm_op, op);
 }
 
 #ifdef CONFIG_COMPAT
 static inline int xsm_do_compat_op(XEN_GUEST_HANDLE_PARAM(void) op)
 {
-    return xsm_ops.do_compat_op(op);
+    return alternative_call(xsm_ops.do_compat_op, op);
 }
 #endif
 
@@ -698,7 +698,7 @@ static inline int xsm_mmuext_op(
 static inline int xsm_update_va_mapping(
     xsm_default_t def, struct domain *d, struct domain *f, l1_pgentry_t pte)
 {
-    return xsm_ops.update_va_mapping(d, f, pte);
+    return alternative_call(xsm_ops.update_va_mapping, d, f, pte);
 }
 
 static inline int xsm_priv_mapping(
-- 
2.11.0

On 05.11.2021 14:55, Andrew Cooper wrote:
> With alternative_call() capable of handling compound types, the three
> remaining hooks can be optimised at boot time too.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>

Reviewed-by: Jan Beulich <jbeulich@suse.com>

On 05.11.2021 14:55, Andrew Cooper wrote:
> With alternative_call() capable of handling compound types, the three
> remaining hooks can be optimised at boot time too.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> ---
> CC: Daniel De Graaf <dgdegra@tycho.nsa.gov>
> CC: Daniel Smith <dpsmith@apertussolutions.com>
> 
> I'm on the fence as to whether to declare this as fixing "xsm: convert xsm_ops
> hook calls to alternative call"

Forgot to say a word on this: I'd consider Fixes: appropriate, as the
commit's description says nothing about these having been left out,
nor why.

Jan

On 08/11/2021 09:11, Jan Beulich wrote:
> On 05.11.2021 14:55, Andrew Cooper wrote:
>> With alternative_call() capable of handling compound types, the three
>> remaining hooks can be optimised at boot time too.
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
>> ---
>> CC: Daniel De Graaf <dgdegra@tycho.nsa.gov>
>> CC: Daniel Smith <dpsmith@apertussolutions.com>
>>
>> I'm on the fence as to whether to declare this as fixing "xsm: convert xsm_ops
>> hook calls to alternative call"
> Forgot to say a word on this: I'd consider Fixes: appropriate, as the
> commit's description says nothing about these having been left out,
> nor why.

Ok.  I'll insert one.

~Andrew

On 11/5/21 9:55 AM, Andrew Cooper wrote:
> With alternative_call() capable of handling compound types, the three
> remaining hooks can be optimised at boot time too.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> ---
> CC: Daniel De Graaf <dgdegra@tycho.nsa.gov>
> CC: Daniel Smith <dpsmith@apertussolutions.com>
> 
> I'm on the fence as to whether to declare this as fixing "xsm: convert xsm_ops
> hook calls to alternative call"

I don't know about fixing but completing was accurate. I am fairly 
certain in one of the incarnations of the patch set there was a comment 
identifying that these were left remaining since compound types were not 
supported by altcall.

Regardless,

Reviewed-by: Daniel P. Smith <dpsmith@apertussolutions.com>

v/r,
dps