[PATCH]0/5 Patch to split x86 specific code from kvm_main.c

[PATCH]0/5 Patch to split x86 specific code from kvm_main.c

[Zhang, Xiantao]

Hi Avi,
This series of patches are intended to further split x86 specific code
from kvm_main.c, and make kvm_main.c arch-independent.  
For easy review, I splitted them into small patches. With these patches,
we almost finish the first stage work for code split. 

[1/5] Remove unused header files include from kvm_main.c since the
corresponding functions have been moved to x86.c
[2/5] Move kvm_x86_ops definition to x86, and prepare for next step's
work.
[3/5] Using kvm_arch prefix to redefine kvm_x86_ops->ops in x86.c. and
make kvm_x86_ops as x86_specific part.
[4/5] Combine kvm_init and kvm_init_x86 into one kvm_init function, and
meanwhile, remove module_init and module_exit function since kvm_init
will called by svm and vmx.
[5/5] Make kvm_init as arch-indepenent, through defining a void pointer
type parameter . All archs will register their corresponding functions
into kvm module. For example, x86 will use this parameter to register
vmx/svm's ops. 

CC to Hollis and Carste to check whether they block your sides. :) 

Thanks 
Xiantao

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [PATCH]0/5 Patch to split x86 specific code from kvm_main.c

[Carsten Otte]

Zhang, Xiantao wrote:
> This series of patches are intended to further split x86 specific code
> from kvm_main.c, and make kvm_main.c arch-independent.  
> For easy review, I splitted them into small patches. With these patches,
> we almost finish the first stage work for code split. 

These patches don't apply on top of my git tree:
Wende Patch 01-xiantao.patch an
patching file drivers/kvm/kvm_main.c
Hunk #1 succeeded at 17 with fuzz 2.
Hunk #2 succeeded at 43 with fuzz 2.
patching file drivers/kvm/x86.c
Hunk #1 succeeded at 16 with fuzz 1.
Hunk #2 FAILED at 26.
1 out of 2 hunks FAILED -- rejects in file drivers/kvm/x86.c

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [PATCH]0/5 Patch to split x86 specific code from kvm_main.c

[Zhang, Xiantao]

[-- Attachment #1: Type: text/plain, Size: 1153 bytes --]

 Hi,Carste 
	Maybe some new commits  block it.  I will rebase them to latest commit, and resend :)
Thanks for you try.
Xiantao

-----Original Message-----
From: Carsten Otte [mailto:cotte-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org] 
Sent: 2007年11月8日 20:06
To: Zhang, Xiantao
Cc: Avi Kivity; kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org; Hollis Blanchard; carsteno-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org
Subject: Re: [PATCH]0/5 Patch to split x86 specific code from kvm_main.c

Zhang, Xiantao wrote:
> This series of patches are intended to further split x86 specific code
> from kvm_main.c, and make kvm_main.c arch-independent.  
> For easy review, I splitted them into small patches. With these patches,
> we almost finish the first stage work for code split. 

These patches don't apply on top of my git tree:
Wende Patch 01-xiantao.patch an
patching file drivers/kvm/kvm_main.c
Hunk #1 succeeded at 17 with fuzz 2.
Hunk #2 succeeded at 43 with fuzz 2.
patching file drivers/kvm/x86.c
Hunk #1 succeeded at 16 with fuzz 1.
Hunk #2 FAILED at 26.
1 out of 2 hunks FAILED -- rejects in file drivers/kvm/x86.c


[-- Attachment #2: Type: text/plain, Size: 314 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

[-- Attachment #3: Type: text/plain, Size: 186 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/kvm-devel

[Resend ] [PATCH]0/5 Patch to split x86 specific code fromkvm_main.c, rebase to latest commits

[Zhang, Xiantao]

[-- Attachment #1: Type: text/plain, Size: 1673 bytes --]

Rebased to latest commits. 
Thanks 
Xiantao

-----Original Message-----
From: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Zhang, Xiantao
Sent: 2007年11月8日 21:00
To: carsteno-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org
Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org; Hollis Blanchard; Avi Kivity
Subject: Re: [kvm-devel] [PATCH]0/5 Patch to split x86 specific code fromkvm_main.c

 Hi,Carste 
	Maybe some new commits  block it.  I will rebase them to latest commit, and resend :)
Thanks for you try.
Xiantao

-----Original Message-----
From: Carsten Otte [mailto:cotte-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org] 
Sent: 2007年11月8日 20:06
To: Zhang, Xiantao
Cc: Avi Kivity; kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org; Hollis Blanchard; carsteno-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org
Subject: Re: [PATCH]0/5 Patch to split x86 specific code from kvm_main.c

Zhang, Xiantao wrote:
> This series of patches are intended to further split x86 specific code
> from kvm_main.c, and make kvm_main.c arch-independent.  
> For easy review, I splitted them into small patches. With these patches,
> we almost finish the first stage work for code split. 

These patches don't apply on top of my git tree:
Wende Patch 01-xiantao.patch an
patching file drivers/kvm/kvm_main.c
Hunk #1 succeeded at 17 with fuzz 2.
Hunk #2 succeeded at 43 with fuzz 2.
patching file drivers/kvm/x86.c
Hunk #1 succeeded at 16 with fuzz 1.
Hunk #2 FAILED at 26.
1 out of 2 hunks FAILED -- rejects in file drivers/kvm/x86.c


[-- Attachment #2: 0001-Move-some-includes-to-x86.c-from-kvm_main.c-since-t.patch --]
[-- Type: application/octet-stream, Size: 1320 bytes --]

From e119cb6a06632e4ac0b2e0fc33054d6b07efdda3 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Thu, 8 Nov 2007 21:09:17 +0800
Subject: [PATCH] Move some includes to x86.c from kvm_main.c, since the related functions have been moved to x86.c
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
---
 drivers/kvm/kvm_main.c |    2 --
 drivers/kvm/x86.c      |    2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index d095002..8357fbb 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -17,7 +17,6 @@
 
 #include "kvm.h"
 #include "x86.h"
-#include "x86_emulate.h"
 #include "irq.h"
 
 #include <linux/kvm.h>
@@ -44,7 +43,6 @@
 #include <linux/mman.h>
 
 #include <asm/processor.h>
-#include <asm/msr.h>
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/desc.h>
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index e905d46..7ea470e 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -16,6 +16,7 @@
 
 #include "kvm.h"
 #include "x86.h"
+#include "x86_emulate.h"
 #include "segment_descriptor.h"
 #include "irq.h"
 
@@ -25,6 +26,7 @@
 #include <linux/module.h>
 
 #include <asm/uaccess.h>
+#include <asm/msr.h>
 
 #define MAX_IO_MSRS 256
 #define CR0_RESERVED_BITS						\
-- 
1.5.1.2


[-- Attachment #3: 0002-move-kvm_x86_ops-to-x86.c.patch --]
[-- Type: application/octet-stream, Size: 1546 bytes --]

From 5f2d0f4ba1570db707797018dc7cc73f81e15b77 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Thu, 8 Nov 2007 21:11:03 +0800
Subject: [PATCH] move kvm_x86_ops to x86.c
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
---
 drivers/kvm/kvm_main.c |    1 -
 drivers/kvm/x86.c      |    2 ++
 drivers/kvm/x86.h      |    2 ++
 3 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 8357fbb..de5bf1f 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -55,7 +55,6 @@ static LIST_HEAD(vm_list);
 
 static cpumask_t cpus_hardware_enabled;
 
-struct kvm_x86_ops *kvm_x86_ops;
 struct kmem_cache *kvm_vcpu_cache;
 EXPORT_SYMBOL_GPL(kvm_vcpu_cache);
 
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 7ea470e..e26e46a 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -44,6 +44,8 @@
 
 #define STAT_OFFSET(x) offsetof(struct kvm_vcpu, stat.x)
 
+struct kvm_x86_ops *kvm_x86_ops;
+
 struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "pf_fixed", STAT_OFFSET(pf_fixed) },
 	{ "pf_guest", STAT_OFFSET(pf_guest) },
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 663b822..ec32c26 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -85,6 +85,8 @@ struct kvm_vcpu {
 	struct x86_emulate_ctxt emulate_ctxt;
 };
 
+extern struct kvm_x86_ops *kvm_x86_ops;
+
 int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code);
 
 static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
-- 
1.5.1.2


[-- Attachment #4: 0003-Using-kvm_arch-prefix-to-redefine-kvm_x86_ops-ops-i.patch --]
[-- Type: application/octet-stream, Size: 6086 bytes --]

From b4ac5a637f879695bb1e149e7572248327c46ba7 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Thu, 8 Nov 2007 21:17:25 +0800
Subject: [PATCH] Using kvm_arch prefix to redefine kvm_x86_ops->ops in x86.c. and
make kvm_x86_ops as x86_specific part.
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
---
 drivers/kvm/kvm.h      |   15 +++++++++++++++
 drivers/kvm/kvm_main.c |   26 +++++++++++++-------------
 drivers/kvm/x86.c      |   47 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 75 insertions(+), 13 deletions(-)

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 091f9b7..e8f6ff1 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -649,6 +649,21 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
 
 __init void kvm_arch_init(void);
 
+
+void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
+void kvm_arch_vcpu_decache(struct kvm_vcpu *vcpu);
+void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
+void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
+struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id);
+
+int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
+void kvm_arch_hardware_enable(void *garbage);
+void kvm_arch_hardware_disable(void *garbage);
+int kvm_arch_hardware_setup(void);
+void kvm_arch_hardware_unsetup(void);
+void kvm_arch_check_processor_compat(void *rtn);
+
+
 static inline void kvm_guest_enter(void)
 {
 	account_system_vtime(current);
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index de5bf1f..629ab54 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -240,7 +240,7 @@ static void kvm_free_vcpus(struct kvm *kvm)
 			kvm_unload_vcpu_mmu(kvm->vcpus[i]);
 	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
 		if (kvm->vcpus[i]) {
-			kvm_x86_ops->vcpu_free(kvm->vcpus[i]);
+			kvm_arch_vcpu_free(kvm->vcpus[i]);
 			kvm->vcpus[i] = NULL;
 		}
 	}
@@ -891,7 +891,7 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, int n)
 	if (!valid_vcpu(n))
 		return -EINVAL;
 
-	vcpu = kvm_x86_ops->vcpu_create(kvm, n);
+	vcpu = kvm_arch_vcpu_create(kvm, n);
 	if (IS_ERR(vcpu))
 		return PTR_ERR(vcpu);
 
@@ -901,7 +901,7 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, int n)
 	BUG_ON((unsigned long)&vcpu->host_fx_image & 0xF);
 
 	vcpu_load(vcpu);
-	r = kvm_x86_ops->vcpu_reset(vcpu);
+	r = kvm_arch_vcpu_reset(vcpu);
 	if (r == 0)
 		r = kvm_mmu_setup(vcpu);
 	vcpu_put(vcpu);
@@ -934,7 +934,7 @@ mmu_unload:
 	vcpu_put(vcpu);
 
 free_vcpu:
-	kvm_x86_ops->vcpu_free(vcpu);
+	kvm_arch_vcpu_free(vcpu);
 	return r;
 }
 
@@ -1298,7 +1298,7 @@ static void decache_vcpus_on_cpu(int cpu)
 			 */
 			if (mutex_trylock(&vcpu->mutex)) {
 				if (vcpu->cpu == cpu) {
-					kvm_x86_ops->vcpu_decache(vcpu);
+					kvm_arch_vcpu_decache(vcpu);
 					vcpu->cpu = -1;
 				}
 				mutex_unlock(&vcpu->mutex);
@@ -1314,7 +1314,7 @@ static void hardware_enable(void *junk)
 	if (cpu_isset(cpu, cpus_hardware_enabled))
 		return;
 	cpu_set(cpu, cpus_hardware_enabled);
-	kvm_x86_ops->hardware_enable(NULL);
+	kvm_arch_hardware_enable(NULL);
 }
 
 static void hardware_disable(void *junk)
@@ -1325,7 +1325,7 @@ static void hardware_disable(void *junk)
 		return;
 	cpu_clear(cpu, cpus_hardware_enabled);
 	decache_vcpus_on_cpu(cpu);
-	kvm_x86_ops->hardware_disable(NULL);
+	kvm_arch_hardware_disable(NULL);
 }
 
 static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val,
@@ -1493,7 +1493,7 @@ static void kvm_sched_in(struct preempt_notifier *pn, int cpu)
 {
 	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);
 
-	kvm_x86_ops->vcpu_load(vcpu, cpu);
+	kvm_arch_vcpu_load(vcpu, cpu);
 }
 
 static void kvm_sched_out(struct preempt_notifier *pn,
@@ -1501,7 +1501,7 @@ static void kvm_sched_out(struct preempt_notifier *pn,
 {
 	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);
 
-	kvm_x86_ops->vcpu_put(vcpu);
+	kvm_arch_vcpu_put(vcpu);
 }
 
 int kvm_init_x86(struct kvm_x86_ops *ops, unsigned int vcpu_size,
@@ -1526,13 +1526,13 @@ int kvm_init_x86(struct kvm_x86_ops *ops, unsigned int vcpu_size,
 
 	kvm_x86_ops = ops;
 
-	r = kvm_x86_ops->hardware_setup();
+	r = kvm_arch_hardware_setup();
 	if (r < 0)
 		goto out;
 
 	for_each_online_cpu(cpu) {
 		smp_call_function_single(cpu,
-				kvm_x86_ops->check_processor_compatibility,
+				kvm_arch_check_processor_compat,
 				&r, 0, 1);
 		if (r < 0)
 			goto out_free_0;
@@ -1587,7 +1587,7 @@ out_free_2:
 out_free_1:
 	on_each_cpu(hardware_disable, NULL, 0, 1);
 out_free_0:
-	kvm_x86_ops->hardware_unsetup();
+	kvm_arch_hardware_unsetup();
 out:
 	kvm_x86_ops = NULL;
 	return r;
@@ -1603,7 +1603,7 @@ void kvm_exit_x86(void)
 	unregister_reboot_notifier(&kvm_reboot_notifier);
 	unregister_cpu_notifier(&kvm_cpu_notifier);
 	on_each_cpu(hardware_disable, NULL, 0, 1);
-	kvm_x86_ops->hardware_unsetup();
+	kvm_arch_hardware_unsetup();
 	kvm_x86_ops = NULL;
 }
 EXPORT_SYMBOL_GPL(kvm_exit_x86);
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index e26e46a..20bb469 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -2327,3 +2327,50 @@ void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
 	fx_restore(&vcpu->host_fx_image);
 }
 EXPORT_SYMBOL_GPL(kvm_put_guest_fpu);
+
+void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->vcpu_free(vcpu);
+}
+
+void kvm_arch_vcpu_decache(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->vcpu_decache(vcpu);
+}
+
+struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
+						unsigned int id)
+{
+	return kvm_x86_ops->vcpu_create(kvm, id);
+}
+
+int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	return kvm_x86_ops->vcpu_reset(vcpu);
+}
+
+void kvm_arch_hardware_enable(void *garbage)
+{
+	kvm_x86_ops->hardware_enable(garbage);
+}
+
+void kvm_arch_hardware_disable(void *garbage)
+{
+	kvm_x86_ops->hardware_disable(garbage);
+}
+
+int kvm_arch_hardware_setup(void)
+{
+	return kvm_x86_ops->hardware_setup();
+}
+
+void kvm_arch_hardware_unsetup(void)
+{
+	kvm_x86_ops->hardware_unsetup();
+}
+
+void kvm_arch_check_processor_compat(void *rtn)
+{
+	kvm_x86_ops->check_processor_compatibility(rtn);
+}
+
-- 
1.5.1.2


[-- Attachment #5: 0004-Combine-kvm_init-and-kvm_init_x86-into-one-kvm_init.patch --]
[-- Type: application/octet-stream, Size: 4037 bytes --]

From 0a12f34f02ca4d2060c230b5ea65cb1f269ab3fc Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Thu, 8 Nov 2007 21:18:48 +0800
Subject: [PATCH] Combine kvm_init and kvm_init_x86 into one kvm_init function, and
meanwhile, remove module_init and module_exit function since kvm_init
will called by svm and vmx.
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
---
 drivers/kvm/kvm.h      |    4 +-
 drivers/kvm/kvm_main.c |   60 +++++++++++++++++------------------------------
 drivers/kvm/svm.c      |    4 +-
 drivers/kvm/vmx.c      |    4 +-
 4 files changed, 28 insertions(+), 44 deletions(-)

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index e8f6ff1..4da3615 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -494,9 +494,9 @@ void vcpu_load(struct kvm_vcpu *vcpu);
 void vcpu_put(struct kvm_vcpu *vcpu);
 
 
-int kvm_init_x86(struct kvm_x86_ops *ops, unsigned int vcpu_size,
+int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
 		  struct module *module);
-void kvm_exit_x86(void);
+void kvm_exit(void);
 
 int kvm_mmu_module_init(void);
 void kvm_mmu_module_exit(void);
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 629ab54..5c40d3d 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -1504,12 +1504,27 @@ static void kvm_sched_out(struct preempt_notifier *pn,
 	kvm_arch_vcpu_put(vcpu);
 }
 
-int kvm_init_x86(struct kvm_x86_ops *ops, unsigned int vcpu_size,
+int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
 		  struct module *module)
 {
 	int r;
 	int cpu;
 
+	r = kvm_mmu_module_init();
+	if (r)
+		goto out4;
+
+	kvm_init_debug();
+
+	kvm_arch_init();
+
+	bad_page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+
+	if (bad_page == NULL) {
+		r = -ENOMEM;
+		goto out;
+	}
+
 	if (kvm_x86_ops) {
 		printk(KERN_ERR "kvm: already loaded the other module\n");
 		return -EEXIST;
@@ -1590,11 +1605,14 @@ out_free_0:
 	kvm_arch_hardware_unsetup();
 out:
 	kvm_x86_ops = NULL;
+	kvm_exit_debug();
+	kvm_mmu_module_exit();
+out4:
 	return r;
 }
-EXPORT_SYMBOL_GPL(kvm_init_x86);
+EXPORT_SYMBOL_GPL(kvm_init);
 
-void kvm_exit_x86(void)
+void kvm_exit(void)
 {
 	misc_deregister(&kvm_dev);
 	kmem_cache_destroy(kvm_vcpu_cache);
@@ -1605,43 +1623,9 @@ void kvm_exit_x86(void)
 	on_each_cpu(hardware_disable, NULL, 0, 1);
 	kvm_arch_hardware_unsetup();
 	kvm_x86_ops = NULL;
-}
-EXPORT_SYMBOL_GPL(kvm_exit_x86);
-
-static __init int kvm_init(void)
-{
-	int r;
-
-	r = kvm_mmu_module_init();
-	if (r)
-		goto out4;
-
-	kvm_init_debug();
-
-	kvm_arch_init();
-
-	bad_page = alloc_page(GFP_KERNEL | __GFP_ZERO);
-
-	if (bad_page == NULL) {
-		r = -ENOMEM;
-		goto out;
-	}
-
-	return 0;
-
-out:
-	kvm_exit_debug();
-	kvm_mmu_module_exit();
-out4:
-	return r;
-}
-
-static __exit void kvm_exit(void)
-{
 	kvm_exit_debug();
 	__free_page(bad_page);
 	kvm_mmu_module_exit();
 }
+EXPORT_SYMBOL_GPL(kvm_exit);
 
-module_init(kvm_init)
-module_exit(kvm_exit)
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index f54b2ea..11a0010 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -1722,13 +1722,13 @@ static struct kvm_x86_ops svm_x86_ops = {
 
 static int __init svm_init(void)
 {
-	return kvm_init_x86(&svm_x86_ops, sizeof(struct vcpu_svm),
+	return kvm_init(&svm_x86_ops, sizeof(struct vcpu_svm),
 			      THIS_MODULE);
 }
 
 static void __exit svm_exit(void)
 {
-	kvm_exit_x86();
+	kvm_exit();
 }
 
 module_init(svm_init)
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 6fc981d..2ba31b4 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -2659,7 +2659,7 @@ static int __init vmx_init(void)
 	memset(iova, 0xff, PAGE_SIZE);
 	kunmap(vmx_io_bitmap_b);
 
-	r = kvm_init_x86(&vmx_x86_ops, sizeof(struct vcpu_vmx), THIS_MODULE);
+	r = kvm_init(&vmx_x86_ops, sizeof(struct vcpu_vmx), THIS_MODULE);
 	if (r)
 		goto out1;
 
@@ -2680,7 +2680,7 @@ static void __exit vmx_exit(void)
 	__free_page(vmx_io_bitmap_b);
 	__free_page(vmx_io_bitmap_a);
 
-	kvm_exit_x86();
+	kvm_exit();
 }
 
 module_init(vmx_init)
-- 
1.5.1.2


[-- Attachment #6: 0005-Make-kvm_init-as-arch-indepenent-through-defining-a.patch --]
[-- Type: application/octet-stream, Size: 3938 bytes --]

From 60ac429810118effc9189504661a8c968b9117a6 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Thu, 8 Nov 2007 21:19:46 +0800
Subject: [PATCH] Make kvm_init as arch-indepenent, through defining a void pointer
type parameter . All archs will register their corresponding functions
into kvm module. For example, x86 will use this parameter to register
vmx/svm's ops.
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
---
 drivers/kvm/kvm.h      |    5 +++--
 drivers/kvm/kvm_main.c |   26 ++++++--------------------
 drivers/kvm/x86.c      |   27 ++++++++++++++++++++++++++-
 3 files changed, 35 insertions(+), 23 deletions(-)

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 4da3615..ce24fa1 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -494,7 +494,7 @@ void vcpu_load(struct kvm_vcpu *vcpu);
 void vcpu_put(struct kvm_vcpu *vcpu);
 
 
-int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
+int kvm_init(void *opaque, unsigned int vcpu_size,
 		  struct module *module);
 void kvm_exit(void);
 
@@ -647,7 +647,8 @@ int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
 				    struct kvm_debug_guest *dbg);
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
 
-__init void kvm_arch_init(void);
+int kvm_arch_init(void *opaque);
+void kvm_arch_exit(void);
 
 
 void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 5c40d3d..5a3ce76 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -1504,7 +1504,7 @@ static void kvm_sched_out(struct preempt_notifier *pn,
 	kvm_arch_vcpu_put(vcpu);
 }
 
-int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
+int kvm_init(void *opaque, unsigned int vcpu_size,
 		  struct module *module)
 {
 	int r;
@@ -1516,7 +1516,9 @@ int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
 
 	kvm_init_debug();
 
-	kvm_arch_init();
+	r = kvm_arch_init(opaque);
+	if (r)
+		goto out4;
 
 	bad_page = alloc_page(GFP_KERNEL | __GFP_ZERO);
 
@@ -1525,22 +1527,6 @@ int kvm_init(struct kvm_x86_ops *ops, unsigned int vcpu_size,
 		goto out;
 	}
 
-	if (kvm_x86_ops) {
-		printk(KERN_ERR "kvm: already loaded the other module\n");
-		return -EEXIST;
-	}
-
-	if (!ops->cpu_has_kvm_support()) {
-		printk(KERN_ERR "kvm: no hardware support\n");
-		return -EOPNOTSUPP;
-	}
-	if (ops->disabled_by_bios()) {
-		printk(KERN_ERR "kvm: disabled by bios\n");
-		return -EOPNOTSUPP;
-	}
-
-	kvm_x86_ops = ops;
-
 	r = kvm_arch_hardware_setup();
 	if (r < 0)
 		goto out;
@@ -1604,7 +1590,7 @@ out_free_1:
 out_free_0:
 	kvm_arch_hardware_unsetup();
 out:
-	kvm_x86_ops = NULL;
+	kvm_arch_exit();
 	kvm_exit_debug();
 	kvm_mmu_module_exit();
 out4:
@@ -1622,7 +1608,7 @@ void kvm_exit(void)
 	unregister_cpu_notifier(&kvm_cpu_notifier);
 	on_each_cpu(hardware_disable, NULL, 0, 1);
 	kvm_arch_hardware_unsetup();
-	kvm_x86_ops = NULL;
+	kvm_arch_exit();
 	kvm_exit_debug();
 	__free_page(bad_page);
 	kvm_mmu_module_exit();
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 20bb469..e2e8e63 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -1611,9 +1611,34 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 }
 EXPORT_SYMBOL_GPL(kvm_emulate_pio_string);
 
-__init void kvm_arch_init(void)
+int kvm_arch_init(void *opaque)
 {
+	struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque;
+
 	kvm_init_msr_list();
+
+	if (kvm_x86_ops) {
+		printk(KERN_ERR "kvm: already loaded the other module\n");
+		return -EEXIST;
+	}
+
+	if (!ops->cpu_has_kvm_support()) {
+		printk(KERN_ERR "kvm: no hardware support\n");
+		return -EOPNOTSUPP;
+	}
+	if (ops->disabled_by_bios()) {
+		printk(KERN_ERR "kvm: disabled by bios\n");
+		return -EOPNOTSUPP;
+	}
+
+	kvm_x86_ops = ops;
+
+	return 0;
+}
+
+void kvm_arch_exit(void)
+{
+	kvm_x86_ops = NULL;
 }
 
 int kvm_emulate_halt(struct kvm_vcpu *vcpu)
-- 
1.5.1.2


[-- Attachment #7: Type: text/plain, Size: 314 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

[-- Attachment #8: Type: text/plain, Size: 186 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/kvm-devel