xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Wei Liu <wei.liu2@citrix.com>
To: Xen-devel <xen-devel@lists.xenproject.org>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>,
	Wei Liu <wei.liu2@citrix.com>, Jan Beulich <jbeulich@suse.com>
Subject: [PATCH for-next v2 05/10] x86/domain: factor out pv_vcpu_initialise
Date: Tue, 25 Apr 2017 14:52:06 +0100	[thread overview]
Message-ID: <20170425135211.4696-6-wei.liu2@citrix.com> (raw)
In-Reply-To: <20170425135211.4696-1-wei.liu2@citrix.com>

Move PV specific vcpu initialisation code to said function, but leave
the only line needed by idle domain in vcpu_initialise.

Use pv_vcpu_destroy in error path to simplify code. It is safe to do so
because the destruction function accepts partially initialised vcpu
struct.

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
Cc: Jan Beulich <jbeulich@suse.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
---
 xen/arch/x86/domain.c | 99 ++++++++++++++++++++++++++-------------------------
 1 file changed, 50 insertions(+), 49 deletions(-)

diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c
index cde0917f5b..38fc4f5d8b 100644
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -398,6 +398,50 @@ static void pv_destroy_gdt_ldt_l1tab(struct domain *d, struct vcpu *v)
     destroy_perdomain_mapping(d, GDT_VIRT_START(v), 1 << GDT_LDT_VCPU_SHIFT);
 }
 
+static void pv_vcpu_destroy(struct vcpu *v);
+static int pv_vcpu_initialise(struct vcpu *v)
+{
+    struct domain *d = v->domain;
+    int rc;
+
+    ASSERT(!is_idle_domain(d));
+
+    spin_lock_init(&v->arch.pv_vcpu.shadow_ldt_lock);
+
+    rc = pv_create_gdt_ldt_l1tab(d, v);
+    if ( rc )
+        goto done;
+
+    BUILD_BUG_ON(NR_VECTORS * sizeof(*v->arch.pv_vcpu.trap_ctxt) >
+                 PAGE_SIZE);
+    v->arch.pv_vcpu.trap_ctxt = xzalloc_array(struct trap_info,
+                                              NR_VECTORS);
+    if ( !v->arch.pv_vcpu.trap_ctxt )
+    {
+        rc = -ENOMEM;
+        goto done;
+    }
+
+    /* PV guests by default have a 100Hz ticker. */
+    v->periodic_period = MILLISECS(10);
+
+    v->arch.pv_vcpu.ctrlreg[4] = real_cr4_to_pv_guest_cr4(mmu_cr4_features);
+
+    if ( is_pv_32bit_domain(d) )
+    {
+        if ( (rc = setup_compat_arg_xlat(v)) )
+            goto done;
+
+        if ( (rc = setup_compat_l4(v)) )
+            goto done;
+    }
+
+ done:
+    if ( rc )
+        pv_vcpu_destroy(v);
+    return rc;
+}
+
 int vcpu_initialise(struct vcpu *v)
 {
     struct domain *d = v->domain;
@@ -424,61 +468,18 @@ int vcpu_initialise(struct vcpu *v)
     spin_lock_init(&v->arch.vpmu.vpmu_lock);
 
     if ( is_hvm_domain(d) )
-    {
         rc = hvm_vcpu_initialise(v);
-        goto done;
-    }
-
-
-    spin_lock_init(&v->arch.pv_vcpu.shadow_ldt_lock);
-
-    if ( !is_idle_domain(d) )
-    {
-        rc = pv_create_gdt_ldt_l1tab(d, v);
-        if ( rc )
-            goto done;
-
-        BUILD_BUG_ON(NR_VECTORS * sizeof(*v->arch.pv_vcpu.trap_ctxt) >
-                     PAGE_SIZE);
-        v->arch.pv_vcpu.trap_ctxt = xzalloc_array(struct trap_info,
-                                                  NR_VECTORS);
-        if ( !v->arch.pv_vcpu.trap_ctxt )
-        {
-            pv_destroy_gdt_ldt_l1tab(d, v);
-            rc = -ENOMEM;
-            goto done;
-        }
-
-        /* PV guests by default have a 100Hz ticker. */
-        v->periodic_period = MILLISECS(10);
-    }
+    else if ( is_pv_domain(d) && !is_idle_domain(d) )
+        rc = pv_vcpu_initialise(v);
     else
-        v->arch.cr3 = __pa(idle_pg_table);
-
-    v->arch.pv_vcpu.ctrlreg[4] = real_cr4_to_pv_guest_cr4(mmu_cr4_features);
-
-    if ( is_pv_32bit_domain(d) )
     {
-        if ( (rc = setup_compat_arg_xlat(v)) )
-            goto done;
-
-        if ( (rc = setup_compat_l4(v)) )
-        {
-            free_compat_arg_xlat(v);
-            goto done;
-        }
+        /* Idle domain */
+        v->arch.cr3 = __pa(idle_pg_table);
+        rc = 0;
     }
- done:
+
     if ( rc )
-    {
         vcpu_destroy_fpu(v);
-
-        if ( is_pv_domain(d) )
-        {
-            pv_destroy_gdt_ldt_l1tab(d, v);
-            xfree(v->arch.pv_vcpu.trap_ctxt);
-        }
-    }
     else if ( !is_idle_domain(v->domain) )
         vpmu_initialise(v);
 
-- 
2.11.0


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

  parent reply	other threads:[~2017-04-25 13:52 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-25 13:52 [PATCH for-next v2 00/10] x86: refactor x86/domain.c Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 01/10] x86/mm: make free_perdomain_mappings idempotent Wei Liu
2017-04-25 13:54   ` Andrew Cooper
2017-04-25 13:52 ` [PATCH for-next v2 02/10] x86/domain: provide pv_{create, destroy}_gdt_ldt_l1tab and use them Wei Liu
2017-04-25 13:56   ` Andrew Cooper
2017-04-25 14:02     ` Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 03/10] x86/domain: make release_compact_l4 NULL tolerant Wei Liu
2017-04-25 13:56   ` Andrew Cooper
2017-04-25 13:52 ` [PATCH for-next v2 04/10] x86/domain: factor out pv_vcpu_destroy Wei Liu
2017-04-25 13:58   ` Andrew Cooper
2017-04-26 12:21   ` Jan Beulich
2017-04-25 13:52 ` Wei Liu [this message]
2017-04-25 14:08   ` [PATCH for-next v2 05/10] x86/domain: factor out pv_vcpu_initialise Andrew Cooper
2017-04-25 14:27     ` Wei Liu
2017-04-26 12:25   ` Jan Beulich
2017-04-26 12:53     ` Wei Liu
2017-04-26 13:27       ` Jan Beulich
2017-04-26 13:35         ` Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 06/10] x86/domain: push some code down to hvm_domain_initialise Wei Liu
2017-04-25 14:15   ` Andrew Cooper
2017-04-25 14:29     ` Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 07/10] x86/domain: factor out pv_domain_destroy Wei Liu
2017-04-25 14:18   ` Andrew Cooper
2017-04-26 12:32   ` Jan Beulich
2017-04-26 12:56     ` Wei Liu
2017-04-26 13:29       ` Jan Beulich
2017-04-26 13:36         ` Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 08/10] x86/domain: factor out pv_domain_initialise Wei Liu
2017-04-25 14:30   ` Andrew Cooper
2017-04-25 13:52 ` [PATCH for-next v2 09/10] x86/domain: move PV specific code to pv/domain.c Wei Liu
2017-04-25 14:52   ` Andrew Cooper
2017-04-25 15:07     ` Wei Liu
2017-04-26  6:04     ` Jan Beulich
2017-04-26 12:38       ` Andrew Cooper
2017-04-26 12:39     ` Jan Beulich
2017-04-26 12:46       ` Andrew Cooper
2017-04-26 13:00         ` Wei Liu
2017-04-26 13:26         ` Jan Beulich
2017-04-26 13:32           ` Wei Liu
2017-04-26 14:12             ` Jan Beulich
2017-04-26 15:46     ` Wei Liu
2017-04-25 13:52 ` [PATCH for-next v2 10/10] x86/domain: move HVM specific code to hvm/domain.c Wei Liu
2017-04-25 14:56   ` Andrew Cooper

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=20170425135211.4696-6-wei.liu2@citrix.com \
    --to=wei.liu2@citrix.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=xen-devel@lists.xenproject.org \
    /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).