With the recent simplifications, it becomes obvious that smp_mb() isn't the
right barrier.  Strictly speaking, MONITOR is ordered as a load, but smp_rmb()
isn't correct either, as this only pertains to local ordering.  All we need is
a compiler barrier().

Merge the barier() into the monitor() itself, along with an explantion.

No functional change.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Roger Pau Monné <roger.pau@citrix.com>
CC: Anthony PERARD <anthony.perard@vates.tech>
CC: Michal Orzel <michal.orzel@amd.com>
CC: Julien Grall <julien@xen.org>
CC: Stefano Stabellini <sstabellini@kernel.org>

v2:
 * Move earlier in the series, not that it matters IMO.
 * Expand the commit message.
---
 xen/arch/x86/acpi/cpu_idle.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/xen/arch/x86/acpi/cpu_idle.c b/xen/arch/x86/acpi/cpu_idle.c
index 141f0f0facf6..68dd44be5bb0 100644
--- a/xen/arch/x86/acpi/cpu_idle.c
+++ b/xen/arch/x86/acpi/cpu_idle.c
@@ -66,8 +66,12 @@ static always_inline void monitor(
     alternative_input("", "clflush (%[addr])", X86_BUG_CLFLUSH_MONITOR,
                       [addr] "a" (addr));
 
+    /*
+     * The memory clobber is a compiler barrier.  Subseqeunt reads from the
+     * monitored cacheline must not be reordered over MONITOR.
+     */
     asm volatile ( "monitor"
-                   :: "a" (addr), "c" (ecx), "d" (edx) );
+                   :: "a" (addr), "c" (ecx), "d" (edx) : "memory" );
 }
 
 static always_inline void mwait(unsigned int eax, unsigned int ecx)
@@ -453,7 +457,6 @@ void mwait_idle_with_hints(unsigned int eax, unsigned int ecx)
     const unsigned int *this_softirq_pending = &softirq_pending(cpu);
 
     monitor(this_softirq_pending, 0, 0);
-    smp_mb();
 
     if ( !*this_softirq_pending )
     {
-- 
2.39.5

On Fri, Jul 04, 2025 at 05:34:06PM +0100, Andrew Cooper wrote:
> With the recent simplifications, it becomes obvious that smp_mb() isn't the
> right barrier.  Strictly speaking, MONITOR is ordered as a load, but smp_rmb()
> isn't correct either, as this only pertains to local ordering.  All we need is
> a compiler barrier().
> 
> Merge the barier() into the monitor() itself, along with an explantion.
> 
> No functional change.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>

Acked-by: Roger Pau Monné <roger.pau@citrix.com>

Thanks, Roger.