xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Mukesh Rathor <mukesh.rathor@oracle.com>
To: Xen-devel@lists.xensource.com
Subject: [PATCH 18/18] PVH xen: Add and remove foreign pages
Date: Fri, 24 May 2013 18:25:37 -0700	[thread overview]
Message-ID: <1369445137-19755-19-git-send-email-mukesh.rathor@oracle.com> (raw)
In-Reply-To: <1369445137-19755-1-git-send-email-mukesh.rathor@oracle.com>

In this patch, a new function, xenmem_add_foreign_to_pmap(), is added
to map pages from foreign guest into current dom0 for domU creation.
Also, allow XENMEM_remove_from_physmap to remove p2m_map_foreign
pages. Note, in this path, we must release the refcount that was taken
during the map phase.

Changes in V2:
  - Move the XENMEM_remove_from_physmap changes here instead of prev patch
  - Move grant changes from this to one of the next patches.
  - In xenmem_add_foreign_to_pmap(), do locked get_gfn
  - Fail the mappings for qemu mapping pages for memory not there.

Changes in V3:
  - remove mmio pages.
  - remove unrelated changes.
  - cleanup both add and remove.

Changes in V5:
  - add a comment in public/memory.h

Changes in V6:
  - replace ASSERTs with returning error code.

Signed-off-by: Mukesh Rathor <mukesh.rathor@oracle.com>
---
 xen/arch/x86/mm.c           |   77 +++++++++++++++++++++++++++++++++++++++++++
 xen/common/memory.c         |   44 ++++++++++++++++++++++---
 xen/include/public/memory.h |    2 +-
 3 files changed, 117 insertions(+), 6 deletions(-)

diff --git a/xen/arch/x86/mm.c b/xen/arch/x86/mm.c
index e992b4f..34137f9 100644
--- a/xen/arch/x86/mm.c
+++ b/xen/arch/x86/mm.c
@@ -4520,6 +4520,78 @@ static int handle_iomem_range(unsigned long s, unsigned long e, void *p)
     return 0;
 }
 
+/*
+ * Add frames from foreign domain to current domain's physmap. Similar to
+ * XENMAPSPACE_gmfn but the frame is foreign being mapped into current,
+ * and is not removed from foreign domain.
+ * Usage: libxl on pvh dom0 creating a guest and doing privcmd_ioctl_mmap.
+ * Side Effect: the mfn for fgfn will be refcounted so it is not lost
+ *              while mapped here. The refcnt is released in do_memory_op()
+ *              via XENMEM_remove_from_physmap.
+ * Returns: 0 ==> success
+ */
+static int xenmem_add_foreign_to_pmap(domid_t foreign_domid,
+                                      unsigned long fgfn, unsigned long gpfn)
+{
+    p2m_type_t p2mt, p2mt_prev;
+    int rc = 0;
+    unsigned long prev_mfn, mfn = 0;
+    struct domain *fdom, *currd = current->domain;
+    struct page_info *page = NULL;
+
+    if ( currd->domain_id == foreign_domid || foreign_domid == DOMID_SELF ||
+         !is_pvh_domain(currd) )
+        return -EINVAL;
+
+    if ( !is_hardware_domain(currd) || !(fdom = get_pg_owner(foreign_domid)) )
+        return -EPERM;
+
+    /* following will take a refcnt on the mfn */
+    page = get_page_from_gfn(fdom, fgfn, &p2mt, P2M_ALLOC);
+    if ( !page || !p2m_is_valid(p2mt) )
+    {
+        if ( page )
+            put_page(page);
+        put_pg_owner(fdom);
+        return -EINVAL;
+    }
+    mfn = page_to_mfn(page);
+
+    /* Remove previously mapped page if it is present. */
+    prev_mfn = mfn_x(get_gfn(currd, gpfn, &p2mt_prev));
+    if ( mfn_valid(prev_mfn) )
+    {
+        if ( is_xen_heap_mfn(prev_mfn) )
+            /* Xen heap frames are simply unhooked from this phys slot */
+            guest_physmap_remove_page(currd, gpfn, prev_mfn, 0);
+        else
+            /* Normal domain memory is freed, to avoid leaking memory. */
+            guest_remove_page(currd, gpfn);
+    }
+    /*
+     * Create the new mapping. Can't use guest_physmap_add_page() because it
+     * will update the m2p table which will result in  mfn -> gpfn of dom0
+     * and not fgfn of domU.
+     */
+    if ( set_foreign_p2m_entry(currd, gpfn, _mfn(mfn)) == 0 )
+    {
+        gdprintk(XENLOG_WARNING,
+                 "guest_physmap_add_page failed. gpfn:%lx mfn:%lx fgfn:%lx\n",
+                 gpfn, mfn, fgfn);
+        put_page(page);
+        rc = -EINVAL;
+    }
+
+    /*
+     * We must do this put_gfn after set_foreign_p2m_entry so another cpu
+     * doesn't populate the gpfn before us.
+     */
+    put_gfn(currd, gpfn);
+
+    put_pg_owner(fdom);
+    return rc;
+}
+
 static int xenmem_add_to_physmap_once(
     struct domain *d,
     const struct xen_add_to_physmap *xatp,
@@ -4582,6 +4654,11 @@ static int xenmem_add_to_physmap_once(
             page = mfn_to_page(mfn);
             break;
         }
+
+        case XENMAPSPACE_gmfn_foreign:
+            return xenmem_add_foreign_to_pmap(foreign_domid, xatp->idx,
+                                              xatp->gpfn);
+
         default:
             break;
     }
diff --git a/xen/common/memory.c b/xen/common/memory.c
index 06a0d0a..47f0cb0 100644
--- a/xen/common/memory.c
+++ b/xen/common/memory.c
@@ -675,9 +675,11 @@ long do_memory_op(unsigned long cmd, XEN_GUEST_HANDLE_PARAM(void) arg)
 
     case XENMEM_remove_from_physmap:
     {
+        unsigned long mfn;
         struct xen_remove_from_physmap xrfp;
         struct page_info *page;
-        struct domain *d;
+        struct domain *d, *foreign_dom = NULL;
+        p2m_type_t p2mt, tp;
 
         if ( copy_from_guest(&xrfp, arg, 1) )
             return -EFAULT;
@@ -695,11 +697,43 @@ long do_memory_op(unsigned long cmd, XEN_GUEST_HANDLE_PARAM(void) arg)
 
         domain_lock(d);
 
-        page = get_page_from_gfn(d, xrfp.gpfn, NULL, P2M_ALLOC);
-        if ( page )
+        /*
+         * if PVH, the gfn could be mapped to a mfn from foreign domain by the
+         * user space tool during domain creation. We need to check for that,
+         * free it up from the p2m, and release refcnt on it. In such a case,
+         * page would be NULL and the following call would not have refcnt'd
+         * the page. See also xenmem_add_foreign_to_pmap().
+         */
+        page = get_page_from_gfn(d, xrfp.gpfn, &p2mt, P2M_ALLOC);
+
+        if ( page || p2m_is_foreign(p2mt) )
         {
-            guest_physmap_remove_page(d, xrfp.gpfn, page_to_mfn(page), 0);
-            put_page(page);
+            if ( page )
+                mfn = page_to_mfn(page);
+            else
+            {
+                mfn = mfn_x(get_gfn_query(d, xrfp.gpfn, &tp));
+                foreign_dom = page_get_owner(mfn_to_page(mfn));
+
+                if ( !is_pvh_domain(d) || (d == foreign_dom) ||
+                     !p2m_is_foreign(tp) )
+                {
+                    rc = -EINVAL;
+                    domain_unlock(d);
+                    rcu_unlock_domain(d);
+                    break;
+                }
+            }
+
+            guest_physmap_remove_page(d, xrfp.gpfn, mfn, 0);
+            if ( page )
+                put_page(page);
+
+            if ( p2m_is_foreign(p2mt) )
+            {
+                put_page(mfn_to_page(mfn));
+                put_gfn(d, xrfp.gpfn);
+            }
         }
         else
             rc = -ENOENT;
diff --git a/xen/include/public/memory.h b/xen/include/public/memory.h
index 7a26dee..e319b5d 100644
--- a/xen/include/public/memory.h
+++ b/xen/include/public/memory.h
@@ -208,7 +208,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_machphys_mapping_t);
 #define XENMAPSPACE_gmfn_range   3 /* GMFN range, XENMEM_add_to_physmap only. */
 #define XENMAPSPACE_gmfn_foreign 4 /* GMFN from another dom,
                                     * XENMEM_add_to_physmap_range only.
-                                    */
+                                    * (PVH x86 only) */
 /* ` } */
 
 /*
-- 
1.7.2.3

  parent reply	other threads:[~2013-05-25  1:25 UTC|newest]

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-25  1:25 [PATCH 00/18][V6]: PVH xen: version 6 patches Mukesh Rathor
2013-05-25  1:25 ` [PATCH 01/18] PVH xen: turn gdb_frames/gdt_ents into union Mukesh Rathor
2013-05-31  9:13   ` Jan Beulich
2013-05-25  1:25 ` [PATCH 02/18] PVH xen: add XENMEM_add_to_physmap_range Mukesh Rathor
2013-05-31  9:28   ` Jan Beulich
2013-05-31  9:38     ` Ian Campbell
2013-05-31 10:14       ` Jan Beulich
2013-05-31 10:40         ` Ian Campbell
2013-06-05  0:24         ` Mukesh Rathor
2013-06-05  0:31     ` Mukesh Rathor
2013-06-05  7:32       ` Jan Beulich
2013-06-05 20:41         ` Mukesh Rathor
2013-06-06  6:43           ` Jan Beulich
2013-06-06 22:19             ` Mukesh Rathor
2013-06-07  6:13               ` Jan Beulich
2013-06-07 20:46                 ` Mukesh Rathor
2013-06-07 15:08             ` Konrad Rzeszutek Wilk
2013-06-07 15:48               ` Jan Beulich
2013-05-25  1:25 ` [PATCH 03/18] PVH xen: create domctl_memory_mapping() function Mukesh Rathor
2013-05-31  9:46   ` Jan Beulich
2013-06-05  0:47     ` Mukesh Rathor
2013-06-05  7:34       ` Jan Beulich
2013-05-25  1:25 ` [PATCH 04/18] PVH xen: add params to read_segment_register Mukesh Rathor
2013-05-31 10:00   ` Jan Beulich
2013-06-06  1:25     ` Mukesh Rathor
2013-06-06  6:48       ` Jan Beulich
2013-06-07  1:43         ` Mukesh Rathor
2013-06-07  6:29           ` Jan Beulich
2013-06-08  0:45             ` Mukesh Rathor
2013-06-10  8:01               ` Jan Beulich
2013-06-10 23:10                 ` Mukesh Rathor
2013-05-25  1:25 ` [PATCH 05/18] PVH xen: vmx realted preparatory changes for PVH Mukesh Rathor
2013-05-25  1:25 ` [PATCH 06/18] PVH xen: Move e820 fields out of pv_domain struct Mukesh Rathor
2013-06-05 15:33   ` Konrad Rzeszutek Wilk
2013-05-25  1:25 ` [PATCH 07/18] PVH xen: Introduce PVH guest type Mukesh Rathor
2013-05-25  1:25 ` [PATCH 08/18] PVH xen: tools changes to create PVH domain Mukesh Rathor
2013-06-12 14:58   ` Ian Campbell
2013-06-15  0:14     ` Mukesh Rathor
2013-06-17 11:11       ` Ian Campbell
2013-07-30 23:47         ` Mukesh Rathor
2013-07-31 12:00           ` Ian Campbell
2013-08-01  2:02             ` Mukesh Rathor
2013-08-01  8:01               ` Ian Campbell
2013-08-02  1:12                 ` Mukesh Rathor
2013-08-29  1:51                 ` Mukesh Rathor
2013-08-29  9:01                   ` Ian Campbell
2013-08-30  0:45                     ` Mukesh Rathor
2013-08-30  9:56                       ` Ian Campbell
2013-08-29 11:13               ` George Dunlap
2013-08-29 11:29                 ` Ian Campbell
2013-08-30  1:24                   ` Mukesh Rathor
2013-08-30  9:53                     ` Ian Campbell
2013-08-30 10:22                       ` George Dunlap
2013-08-30 10:27                     ` George Dunlap
2013-08-29  0:14         ` Mukesh Rathor
2013-07-31  1:06     ` Mukesh Rathor
2013-07-31 11:32       ` Ian Campbell
2013-05-25  1:25 ` [PATCH 09/18] PVH xen: domain creation code changes Mukesh Rathor
2013-05-25  1:25 ` [PATCH 10/18] PVH xen: create PVH vmcs, and also initialization Mukesh Rathor
2013-05-25  1:25 ` [PATCH 11/18] PVH xen: create read_descriptor_sel() Mukesh Rathor
2013-05-25  1:25 ` [PATCH 12/18] PVH xen: support hypercalls for PVH Mukesh Rathor
2013-06-05 15:27   ` Konrad Rzeszutek Wilk
2013-05-25  1:25 ` [PATCH 13/18] PVH xen: introduce vmx_pvh.c Mukesh Rathor
2013-05-25  1:25 ` [PATCH 14/18] PVH xen: some misc changes like mtrr, intr, msi Mukesh Rathor
2013-05-25  1:25 ` [PATCH 15/18] PVH xen: hcall page initialize, create PVH guest type, etc Mukesh Rathor
2013-05-25  1:25 ` [PATCH 16/18] PVH xen: Miscellaneous changes Mukesh Rathor
2013-06-05 15:39   ` Konrad Rzeszutek Wilk
2013-05-25  1:25 ` [PATCH 17/18] PVH xen: Introduce p2m_map_foreign Mukesh Rathor
2013-05-25  1:25 ` Mukesh Rathor [this message]
2013-06-05 15:23 ` [PATCH 00/18][V6]: PVH xen: version 6 patches Konrad Rzeszutek Wilk
2013-06-05 15:25   ` George Dunlap
2013-06-05 15:36   ` Ian Campbell
2013-06-05 18:34     ` Konrad Rzeszutek Wilk
2013-06-05 20:51       ` Ian Campbell
2013-06-05 22:01         ` Mukesh Rathor
2013-06-06  8:46           ` Ian Campbell
2013-06-07 13:56             ` Konrad Rzeszutek Wilk
2013-06-06 10:08     ` George Dunlap
2013-06-05 17:14   ` Tim Deegan
2013-06-06  7:29     ` Jan Beulich

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1369445137-19755-19-git-send-email-mukesh.rathor@oracle.com \
    --to=mukesh.rathor@oracle.com \
    --cc=Xen-devel@lists.xensource.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).