[PATCH] xen/arm: Fix dom0 boot following c/s 580c45869

[PATCH] xen/arm: Fix dom0 boot following c/s 580c45869

[Andrew Cooper]

c/s 580c45869 "Call arch_domain_create() as early as possible in
domain_create()" overlooked the fact that ARM uses is_hardware_domain() in at
least two places during arch_domain_create().

The bug manifests as:

  (XEN) Freed 292kB init memory.
  (XEN) traps.c:2017:d0v0 HSR=0x938c0007 pc=0xc0639d08 gva=0xe0800004 gpa=0x00000010481004

when dom0 tries to use the vuart.  Judging by other uses of
is_hardware_domain(), I expect the x86 PVH dom0 boot is similarly broken.

Reposition the code which sets up hardware_domain so that the
is_hardware_domain() predicate works correctly all the way through domain
creation.

While moving it, leave a related comment explaining the positioning of the
is_priv assignment, which in hindsight should have been part of c/s ef765ec98
when exactly the same problem was discovered for the is_control_domain()
predicate.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Stefano Stabellini <sstabellini@kernel.org>
CC: Julien Grall <julien.grall@arm.com>
CC: Jan Beulich <JBeulich@suse.com>
CC: Wei Liu <wei.liu2@citrix.com>
CC: Roger Pau Monné <roger.pau@citrix.com>

Thanks to Julien for helping debug this after OSSTest gave me an error I
didn't understand, and sorry for breaking it!

I've got some other cleanup I'd like to do to the general hardware_domain
infrastrucutre, but I'll leave that to a later change so as to unduly block
staging.
---
 xen/common/domain.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/xen/common/domain.c b/xen/common/domain.c
index f64ad5f..256c59a 100644
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
@@ -273,8 +273,22 @@ struct domain *domain_create(domid_t domid,
         return ERR_PTR(-ENOMEM);
 
     d->domain_id = domid;
+
+    /* Sort out our idea of is_control_domain(). */
     d->is_privileged = is_priv;
 
+    /* Sort out our idea of is_hardware_domain(). */
+    if ( domid == 0 || domid == hardware_domid )
+    {
+        if ( hardware_domid < 0 || hardware_domid >= DOMID_FIRST_RESERVED )
+            panic("The value of hardware_dom must be a valid domain ID");
+
+        d->is_pinned = opt_dom0_vcpus_pin;
+        d->disable_migrate = 1;
+        old_hwdom = hardware_domain;
+        hardware_domain = d;
+    }
+
     /* Debug sanity. */
     ASSERT(is_system_domain(d) ? config == NULL : config != NULL);
 
@@ -354,16 +368,6 @@ struct domain *domain_create(domid_t domid,
         watchdog_domain_init(d);
         init_status |= INIT_watchdog;
 
-        if ( domid == 0 || domid == hardware_domid )
-        {
-            if ( hardware_domid < 0 || hardware_domid >= DOMID_FIRST_RESERVED )
-                panic("The value of hardware_dom must be a valid domain ID");
-            d->is_pinned = opt_dom0_vcpus_pin;
-            d->disable_migrate = 1;
-            old_hwdom = hardware_domain;
-            hardware_domain = d;
-        }
-
         if ( config->flags & XEN_DOMCTL_CDF_xs_domain )
         {
             d->is_xenstore = 1;
-- 
2.1.4


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

Re: [PATCH] xen/arm: Fix dom0 boot following c/s 580c45869

[Julien Grall]

Hi Andrew,

On 31/08/18 19:01, Andrew Cooper wrote:
> c/s 580c45869 "Call arch_domain_create() as early as possible in
> domain_create()" overlooked the fact that ARM uses is_hardware_domain() in at
> least two places during arch_domain_create().
> 
> The bug manifests as:
> 
>    (XEN) Freed 292kB init memory.
>    (XEN) traps.c:2017:d0v0 HSR=0x938c0007 pc=0xc0639d08 gva=0xe0800004 gpa=0x00000010481004
> 
> when dom0 tries to use the vuart.  Judging by other uses of
> is_hardware_domain(), I expect the x86 PVH dom0 boot is similarly broken.
> 
> Reposition the code which sets up hardware_domain so that the
> is_hardware_domain() predicate works correctly all the way through domain
> creation.
> 
> While moving it, leave a related comment explaining the positioning of the
> is_priv assignment, which in hindsight should have been part of c/s ef765ec98
> when exactly the same problem was discovered for the is_control_domain()
> predicate.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>

Reviewed-by: Julien Grall <julien.grall@arm.com>
Tested-by: Julien Grall <julien.grall@arm.com>

I will commit it now.

Cheers,

> ---
> CC: Stefano Stabellini <sstabellini@kernel.org>
> CC: Julien Grall <julien.grall@arm.com>
> CC: Jan Beulich <JBeulich@suse.com>
> CC: Wei Liu <wei.liu2@citrix.com>
> CC: Roger Pau Monné <roger.pau@citrix.com>
> 
> Thanks to Julien for helping debug this after OSSTest gave me an error I
> didn't understand, and sorry for breaking it!
> 
> I've got some other cleanup I'd like to do to the general hardware_domain
> infrastrucutre, but I'll leave that to a later change so as to unduly block
> staging.
> ---
>   xen/common/domain.c | 24 ++++++++++++++----------
>   1 file changed, 14 insertions(+), 10 deletions(-)
> 
> diff --git a/xen/common/domain.c b/xen/common/domain.c
> index f64ad5f..256c59a 100644
> --- a/xen/common/domain.c
> +++ b/xen/common/domain.c
> @@ -273,8 +273,22 @@ struct domain *domain_create(domid_t domid,
>           return ERR_PTR(-ENOMEM);
>   
>       d->domain_id = domid;
> +
> +    /* Sort out our idea of is_control_domain(). */
>       d->is_privileged = is_priv;
>   
> +    /* Sort out our idea of is_hardware_domain(). */
> +    if ( domid == 0 || domid == hardware_domid )
> +    {
> +        if ( hardware_domid < 0 || hardware_domid >= DOMID_FIRST_RESERVED )
> +            panic("The value of hardware_dom must be a valid domain ID");
> +
> +        d->is_pinned = opt_dom0_vcpus_pin;
> +        d->disable_migrate = 1;
> +        old_hwdom = hardware_domain;
> +        hardware_domain = d;
> +    }
> +
>       /* Debug sanity. */
>       ASSERT(is_system_domain(d) ? config == NULL : config != NULL);
>   
> @@ -354,16 +368,6 @@ struct domain *domain_create(domid_t domid,
>           watchdog_domain_init(d);
>           init_status |= INIT_watchdog;
>   
> -        if ( domid == 0 || domid == hardware_domid )
> -        {
> -            if ( hardware_domid < 0 || hardware_domid >= DOMID_FIRST_RESERVED )
> -                panic("The value of hardware_dom must be a valid domain ID");
> -            d->is_pinned = opt_dom0_vcpus_pin;
> -            d->disable_migrate = 1;
> -            old_hwdom = hardware_domain;
> -            hardware_domain = d;
> -        }
> -
>           if ( config->flags & XEN_DOMCTL_CDF_xs_domain )
>           {
>               d->is_xenstore = 1;
> 

-- 
Julien Grall

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