Re: [Xen-devel] [PATCH] xen/domctl: lower loglevel of XEN_DOMCTL_memory_mapping

["Chen, Tiejun" <tiejun.chen@xxxxxxxxx>]

> > If the 64 limit was arbitrary then I would suggest increasing it to at least
> > 1024 so that
> > at least 4M of BAR can be mapped in one go and it reduces the overhead by a
> > factor of 16.
> 1024 may be a little much, but 256 is certainly a possibility, plus
> Konrad's suggestion to allow this limit to be controlled via command
> line option.
Are you guys talking this way?

diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
index 3946e4c..a9671bb 100644
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -88,6 +88,10 @@ boolean_param("noapic", skip_ioapic_setup);
 s8 __read_mostly xen_cpuidle = -1;
 boolean_param("cpuidle", xen_cpuidle);

+/* once_mapping_mfns: memory mapping mfn bumbers once. */
+unsigned int xen_once_mapping_mfns;
+integer_param("once_mapping_mfns", xen_once_mapping_mfns);
+
 #ifndef NDEBUG
 unsigned long __initdata highmem_start;
 size_param("highmem-start", highmem_start);
diff --git a/xen/common/domctl.c b/xen/common/domctl.c
index 3bf39f1..82c85e3 100644
--- a/xen/common/domctl.c
+++ b/xen/common/domctl.c
@@ -33,6 +33,8 @@
 #include <public/domctl.h>
 #include <xsm/xsm.h>

+extern unsigned int xen_once_mapping_mfns;
+
 static DEFINE_SPINLOCK(domctl_lock);
 DEFINE_SPINLOCK(vcpu_alloc_lock);

@@ -1035,7 +1037,7 @@ long 
do_domctl(XEN_GUEST_HANDLE_PARAM(xen_domctl_t) u_domctl)
         ret = -E2BIG;
         /* Must break hypercall up as this could take a while. */
-        if ( nr_mfns > 64 )
+        if ( nr_mfns > xen_once_mapping_mfns )
             break;

         ret = -EPERM;

Thanks
Tiejun

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel