diff mbox

[03/11] xen: defer call to xen_restrict until just before os_setup_post

Message ID 23202.30969.590920.764930@mariner.uk.xensource.com (mailing list archive)
State New, archived
Headers show

Commit Message

Ian Jackson March 9, 2018, 12:07 p.m. UTC
Ian Jackson writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> Eduardo Habkost writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> > I don't think we should have accelerator-specific code in main(),
> > if we already have accelerator classes that can abstract that
> > out.  I suggest adding a AccelClass;:setup_post() method that can
> > be called here.
> 
> I think I can do that.

How about this ?

From 61f11221afaa29e10021599420238e03836ba413 Mon Sep 17 00:00:00 2001
From: Ian Jackson <ian.jackson@eu.citrix.com>
Date: Fri, 9 Mar 2018 12:02:50 +0000
Subject: [PATCH v6.2 12/11] AccelClass: Introduce accel_setup_post

This is called just before os_setup_post.  Currently none of the
accelerators provide this hook, but the Xen one is going to provide
one in a moment.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
---
 accel/accel.c          | 9 +++++++++
 include/sysemu/accel.h | 3 +++
 vl.c                   | 1 +
 3 files changed, 13 insertions(+)

Comments

Eduardo Habkost March 9, 2018, 12:58 p.m. UTC | #1
On Fri, Mar 09, 2018 at 12:07:21PM +0000, Ian Jackson wrote:
> Ian Jackson writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> > Eduardo Habkost writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> > > I don't think we should have accelerator-specific code in main(),
> > > if we already have accelerator classes that can abstract that
> > > out.  I suggest adding a AccelClass;:setup_post() method that can
> > > be called here.
> > 
> > I think I can do that.
> 
> How about this ?
> 
> From 61f11221afaa29e10021599420238e03836ba413 Mon Sep 17 00:00:00 2001
> From: Ian Jackson <ian.jackson@eu.citrix.com>
> Date: Fri, 9 Mar 2018 12:02:50 +0000
> Subject: [PATCH v6.2 12/11] AccelClass: Introduce accel_setup_post
> 
> This is called just before os_setup_post.  Currently none of the
> accelerators provide this hook, but the Xen one is going to provide
> one in a moment.
> 
> Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>

Looks good to me.

Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>

That said, I don't think this should block the inclusion of the
previous patch in 2.12, if the Xen maintainers were already going
to merge it.
Stefano Stabellini March 11, 2018, 9:12 p.m. UTC | #2
On Fri, 9 Mar 2018, Eduardo Habkost wrote:
> On Fri, Mar 09, 2018 at 12:07:21PM +0000, Ian Jackson wrote:
> > Ian Jackson writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> > > Eduardo Habkost writes ("Re: [PATCH 03/11] xen: defer call to xen_restrict until just before os_setup_post"):
> > > > I don't think we should have accelerator-specific code in main(),
> > > > if we already have accelerator classes that can abstract that
> > > > out.  I suggest adding a AccelClass;:setup_post() method that can
> > > > be called here.
> > > 
> > > I think I can do that.
> > 
> > How about this ?
> > 
> > From 61f11221afaa29e10021599420238e03836ba413 Mon Sep 17 00:00:00 2001
> > From: Ian Jackson <ian.jackson@eu.citrix.com>
> > Date: Fri, 9 Mar 2018 12:02:50 +0000
> > Subject: [PATCH v6.2 12/11] AccelClass: Introduce accel_setup_post
> > 
> > This is called just before os_setup_post.  Currently none of the
> > accelerators provide this hook, but the Xen one is going to provide
> > one in a moment.
> > 
> > Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
> 
> Looks good to me.
> 
> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
> 
> That said, I don't think this should block the inclusion of the
> previous patch in 2.12, if the Xen maintainers were already going
> to merge it.

Ian,

The Xen side is almost entirely reviewed and should be OK, but you need
reviewed-bys on some non-Xen specific patches: #7 #8 #11.

Cheers,

Stefano
diff mbox

Patch

diff --git a/accel/accel.c b/accel/accel.c
index 93e2434..9cfab11 100644
--- a/accel/accel.c
+++ b/accel/accel.c
@@ -126,6 +126,15 @@  void accel_register_compat_props(AccelState *accel)
     register_compat_props_array(class->global_props);
 }
 
+void accel_setup_post(MachineState *ms)
+{
+    AccelState *accel = ms->accelerator;
+    AccelClass *acc = ACCEL_GET_CLASS(accel);
+    if (acc->setup_post) {
+        acc->setup_post(ms, accel);
+    }
+}
+
 static void register_accel_types(void)
 {
     type_register_static(&accel_type);
diff --git a/include/sysemu/accel.h b/include/sysemu/accel.h
index 5a632ce..637358f 100644
--- a/include/sysemu/accel.h
+++ b/include/sysemu/accel.h
@@ -40,6 +40,7 @@  typedef struct AccelClass {
     const char *name;
     int (*available)(void);
     int (*init_machine)(MachineState *ms);
+    void (*setup_post)(MachineState *ms, AccelState *accel);
     bool *allowed;
     /*
      * Array of global properties that would be applied when specific
@@ -68,5 +69,7 @@  extern unsigned long tcg_tb_size;
 void configure_accelerator(MachineState *ms);
 /* Register accelerator specific global properties */
 void accel_register_compat_props(AccelState *accel);
+/* Called just before os_setup_post (ie just before drop OS privs) */
+void accel_setup_post(MachineState *ms);
 
 #endif
diff --git a/vl.c b/vl.c
index e6e8e1e..3fd6401 100644
--- a/vl.c
+++ b/vl.c
@@ -4719,6 +4719,7 @@  int main(int argc, char **argv, char **envp)
         vm_start();
     }
 
+    accel_setup_post(current_machine);
     xen_setup_post();
     os_setup_post();