[Xen-devel] [PATCH 3/3] x86/HVM: hvmemul_cmpxchg() should also use known_gla()

["Jan Beulich" <JBeulich@xxxxxxxx>]

To be consistent with the write and rmw cases the mapping approach
should not be used when the guest linear address translation is known.
This in particular excludes the discard-write case from bypassing the
emulation path. This also means that now EFLAGS should actually get
properly updated, despite the discarded write portion of the memory
access.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
Without "make hvmemul_map_linear_addr() honor p2m_ioreq_server" this
also helps the p2m_ioreq_server case.

TBD: While hvmemul_do_io() takes care of p2m_ioreq_server, I don't see
     it taking care of p2m_is_discard_write() types. Am I overlooking
     something?

--- a/xen/arch/x86/hvm/emulate.c
+++ b/xen/arch/x86/hvm/emulate.c
@@ -1472,9 +1472,12 @@ static int hvmemul_cmpxchg(
     else if ( hvmemul_ctxt->seg_reg[x86_seg_ss].dpl == 3 )
         pfec |= PFEC_user_mode;
 
-    mapping = hvmemul_map_linear_addr(addr, bytes, pfec, hvmemul_ctxt);
-    if ( IS_ERR(mapping) )
-        return ~PTR_ERR(mapping);
+    if ( !known_gla(addr, bytes, pfec) )
+    {
+        mapping = hvmemul_map_linear_addr(addr, bytes, pfec, hvmemul_ctxt);
+        if ( IS_ERR(mapping) )
+            return ~PTR_ERR(mapping);
+    }
 
     if ( !mapping )
     {





_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel