LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH 0/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer
@ 2008-02-25 13:00 Markus Armbruster
  2008-02-25 13:02 ` [PATCH 1/3] xen: Make xen-blkfront write its protocol ABI to xenstore Markus Armbruster
                   ` (2 more replies)
  0 siblings, 3 replies; 13+ messages in thread
From: Markus Armbruster @ 2008-02-25 13:00 UTC (permalink / raw)
  To: linux-kernel
  Cc: virtualization, xen-devel, linux-fbdev-devel, adaplas,
	linux-input, dmitry.torokhov, akpm, jayakumar.lkml

This is a pair of Xen para-virtual frontend device drivers:
drivers/video/xen-fbfront.c provides a framebuffer, and
drivers/input/xen-kbdfront provides keyboard and mouse.

The backends run in dom0 user space.

Differences since last post:

* Required patch fixing 32-on-64 xen-blkfront included.

* Cleanup when xenkbd_probe() fails fixed.

* Don't store event channel in device info.

I started with the Xen version at
http://xenbits.xensource.com/linux-2.6.18-xen.hg?rev/ca05cf1a9bdc

Differences to that Xen version, for those who care:

* Rewritten on top of fb deferred I/O

* IRQ handler names visible in /proc and /sys match the driver names.

* Use framebuffer helper functions appropriate for framebuffer in
  system RAM.

* write() refreshes the framebuffer properly.

* off-by-one height of some screen refreshs fixed.

* Crash when register_framebuffer() fails fixed.

* Test for empty ring in input_handler() fixed.

* Deadlock in xen-kbdfront resume fixed.

* Cleanup when xenkbd_probe() fails fixed.

* General clean up.

I have a step-by-step patch series from that Xen version to my
version, if anybody is interested.  Might be useful for reviewers
familiar with the Xen version.

The patch consists of three parts:

1. xen: Make xen-blkfront write its protocol ABI to xenstore

   Could do without, but then I'd have to put the same bug in
   xen-fbfront and xen-kbdfront.

2. fbdev: Make deferred I/O work as advertized

   I need fb deferred I/O, but is utterly broken.  A fix has been
   floating around on linux-fbdev-devel as part of a larger patch,
   which as far as I know has not been merged anywhere, yet.  This is
   just the fix.

2. xen pvfb: Para-virtual framebuffer, keyboard and pointer driver

   The actual drivers.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [PATCH 1/3] xen: Make xen-blkfront write its protocol ABI to xenstore
  2008-02-25 13:00 [PATCH 0/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer Markus Armbruster
@ 2008-02-25 13:02 ` Markus Armbruster
  2008-02-25 13:03 ` [PATCH 2/3] fbdev: Make deferred I/O work as advertized Markus Armbruster
  2008-02-25 13:06 ` [PATCH 3/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer driver Markus Armbruster
  2 siblings, 0 replies; 13+ messages in thread
From: Markus Armbruster @ 2008-02-25 13:02 UTC (permalink / raw)
  To: linux-kernel
  Cc: virtualization, xen-devel, linux-fbdev-devel, adaplas,
	linux-input, dmitry.torokhov, akpm, jayakumar.lkml

Frontends are expected to write their protocol ABI to xenstore.  Since
the protocol ABI defaults to the backend's native ABI, things work
fine without that as long as the frontend's native ABI is identical to
the backend's native ABI.  This is not the case for xen-blkfront
running 32-on-64, because its ABI differs between 32 and 64 bit, and
thus needs this fix.

Based on http://xenbits.xensource.com/xen-unstable.hg?rev/c545932a18f3
and http://xenbits.xensource.com/xen-unstable.hg?rev/ffe52263b430 by
Gerd Hoffmann <kraxel@suse.de>

Signed-off-by: Markus Armbruster <armbru@redhat.com>

---
 drivers/block/xen-blkfront.c         |    7 +++++++
 include/xen/interface/io/protocols.h |   21 +++++++++++++++++++++
 2 files changed, 28 insertions(+), 0 deletions(-)
 create mode 100644 include/xen/interface/io/protocols.h

diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 8afce67..e69164a 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -46,6 +46,7 @@
 
 #include <xen/interface/grant_table.h>
 #include <xen/interface/io/blkif.h>
+#include <xen/interface/io/protocols.h>
 
 #include <asm/xen/hypervisor.h>
 
@@ -597,6 +598,12 @@ again:
 		message = "writing event-channel";
 		goto abort_transaction;
 	}
+	err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
+			    XEN_IO_PROTO_ABI_NATIVE);
+	if (err) {
+		message = "writing protocol";
+		goto abort_transaction;
+	}
 
 	err = xenbus_transaction_end(xbt, 0);
 	if (err) {
diff --git a/include/xen/interface/io/protocols.h b/include/xen/interface/io/protocols.h
new file mode 100644
index 0000000..01fc8ae
--- /dev/null
+++ b/include/xen/interface/io/protocols.h
@@ -0,0 +1,21 @@
+#ifndef __XEN_PROTOCOLS_H__
+#define __XEN_PROTOCOLS_H__
+
+#define XEN_IO_PROTO_ABI_X86_32     "x86_32-abi"
+#define XEN_IO_PROTO_ABI_X86_64     "x86_64-abi"
+#define XEN_IO_PROTO_ABI_IA64       "ia64-abi"
+#define XEN_IO_PROTO_ABI_POWERPC64  "powerpc64-abi"
+
+#if defined(__i386__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_32
+#elif defined(__x86_64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_64
+#elif defined(__ia64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_IA64
+#elif defined(__powerpc64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_POWERPC64
+#else
+# error arch fixup needed here
+#endif
+
+#endif
-- 
1.5.3.3


^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-02-25 13:00 [PATCH 0/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer Markus Armbruster
  2008-02-25 13:02 ` [PATCH 1/3] xen: Make xen-blkfront write its protocol ABI to xenstore Markus Armbruster
@ 2008-02-25 13:03 ` Markus Armbruster
  2008-02-26 14:11   ` Jaya Kumar
  2008-02-25 13:06 ` [PATCH 3/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer driver Markus Armbruster
  2 siblings, 1 reply; 13+ messages in thread
From: Markus Armbruster @ 2008-02-25 13:03 UTC (permalink / raw)
  To: linux-kernel
  Cc: virtualization, xen-devel, linux-fbdev-devel, adaplas,
	linux-input, dmitry.torokhov, akpm, jayakumar.lkml

Deferred I/O was utterly broken.  Reading the mmap()ed framebuffer
worked, but writing it made the VM endlessly invoke
vm_ops.page_mkwrite().  That happened because we failed to set
page->mapping and page->index.

The fix is to set them, and clean up properly before the framebuffer
gets released.

Fix extracted from this linux-fbdev-devel message:

    Subject: [PATCH 1/1 2.6.24] fbdev: defio and Metronomefb
    From: Jaya Kumar <jayakumar.lkml@gmail.com>
    Date: 2008-02-18 13:41:26

Signed-off-by: Jaya Kumar <jayakumar.lkml@gmail.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>

---

 drivers/video/fb_defio.c |   22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/drivers/video/fb_defio.c b/drivers/video/fb_defio.c
index 0f8cfb9..24843fd 100644
--- a/drivers/video/fb_defio.c
+++ b/drivers/video/fb_defio.c
@@ -4,7 +4,7 @@
  *  Copyright (C) 2006 Jaya Kumar
  *
  * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file COPYING in the main directory of this archive
+ * License. See the file COPYING in the main directory of this archive
  * for more details.
  */
 
@@ -31,7 +31,7 @@ static int fb_deferred_io_fault(struct vm_area_struct *vma,
 	unsigned long offset;
 	struct page *page;
 	struct fb_info *info = vma->vm_private_data;
-	/* info->screen_base is in System RAM */
+	/* info->screen_base is virtual memory */
 	void *screen_base = (void __force *) info->screen_base;
 
 	offset = vmf->pgoff << PAGE_SHIFT;
@@ -43,6 +43,15 @@ static int fb_deferred_io_fault(struct vm_area_struct *vma,
 		return VM_FAULT_SIGBUS;
 
 	get_page(page);
+
+	if (vma->vm_file)
+		page->mapping = vma->vm_file->f_mapping;
+	else
+		printk(KERN_ERR "no mapping available\n");
+
+	BUG_ON(!page->mapping);
+	page->index = vmf->pgoff;
+
 	vmf->page = page;
 	return 0;
 }
@@ -138,11 +147,20 @@ EXPORT_SYMBOL_GPL(fb_deferred_io_init);
 
 void fb_deferred_io_cleanup(struct fb_info *info)
 {
+	void *screen_base = (void __force *) info->screen_base;
 	struct fb_deferred_io *fbdefio = info->fbdefio;
+	struct page *page;
+	int i;
 
 	BUG_ON(!fbdefio);
 	cancel_delayed_work(&info->deferred_work);
 	flush_scheduled_work();
+
+	/* clear out the mapping that we setup */
+	for (i = 0 ; i < info->fix.smem_len; i += PAGE_SIZE) {
+		page = vmalloc_to_page(screen_base + i);
+		page->mapping = NULL;
+	}
 }
 EXPORT_SYMBOL_GPL(fb_deferred_io_cleanup);
 

^ permalink raw reply related	[flat|nested] 13+ messages in thread

* [PATCH 3/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer driver
  2008-02-25 13:00 [PATCH 0/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer Markus Armbruster
  2008-02-25 13:02 ` [PATCH 1/3] xen: Make xen-blkfront write its protocol ABI to xenstore Markus Armbruster
  2008-02-25 13:03 ` [PATCH 2/3] fbdev: Make deferred I/O work as advertized Markus Armbruster
@ 2008-02-25 13:06 ` Markus Armbruster
  2 siblings, 0 replies; 13+ messages in thread
From: Markus Armbruster @ 2008-02-25 13:06 UTC (permalink / raw)
  To: linux-kernel
  Cc: virtualization, xen-devel, linux-fbdev-devel, adaplas,
	linux-input, dmitry.torokhov, akpm, jayakumar.lkml

This is a pair of Xen para-virtual frontend device drivers:
drivers/video/xen-fbfront.c provides a framebuffer, and
drivers/input/xen-kbdfront provides keyboard and mouse.

The backends run in dom0 user space.

The two drivers are not in two separate patches, because the
intermediate step (one driver, not the other) is somewhat problematic:
the backend in dom0 needs both drivers, and will refuse to complete
device initialization unless they're both present.

Signed-off-by: Markus Armbruster <armbru@redhat.com>

---

 drivers/input/Kconfig            |    9 
 drivers/input/Makefile           |    2 
 drivers/input/xen-kbdfront.c     |  340 ++++++++++++++++++++++++
 drivers/video/Kconfig            |   14 
 drivers/video/Makefile           |    1 
 drivers/video/xen-fbfront.c      |  550 +++++++++++++++++++++++++++++++++++++++
 include/xen/interface/io/fbif.h  |  124 ++++++++
 include/xen/interface/io/kbdif.h |  114 ++++++++
 8 files changed, 1154 insertions(+)

diff --git a/drivers/input/Kconfig b/drivers/input/Kconfig
index 9dea14d..5f9d860 100644
--- a/drivers/input/Kconfig
+++ b/drivers/input/Kconfig
@@ -149,6 +149,15 @@ config INPUT_APMPOWER
 	  To compile this driver as a module, choose M here: the
 	  module will be called apm-power.
 
+config XEN_KBDDEV_FRONTEND
+	tristate "Xen virtual keyboard and mouse support"
+	depends on XEN_FBDEV_FRONTEND
+	default y
+	help
+	  This driver implements the front-end of the Xen virtual
+	  keyboard and mouse device driver.  It communicates with a back-end
+	  in another domain.
+
 comment "Input Device Drivers"
 
 source "drivers/input/keyboard/Kconfig"
diff --git a/drivers/input/Makefile b/drivers/input/Makefile
index 2ae87b1..98c4f9a 100644
--- a/drivers/input/Makefile
+++ b/drivers/input/Makefile
@@ -23,3 +23,5 @@ obj-$(CONFIG_INPUT_TOUCHSCREEN)	+= touchscreen/
 obj-$(CONFIG_INPUT_MISC)	+= misc/
 
 obj-$(CONFIG_INPUT_APMPOWER)	+= apm-power.o
+
+obj-$(CONFIG_XEN_KBDDEV_FRONTEND)	+= xen-kbdfront.o
diff --git a/drivers/input/xen-kbdfront.c b/drivers/input/xen-kbdfront.c
new file mode 100644
index 0000000..0f47f46
--- /dev/null
+++ b/drivers/input/xen-kbdfront.c
@@ -0,0 +1,340 @@
+/*
+ * Xen para-virtual input device
+ *
+ * Copyright (C) 2005 Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2006-2008 Red Hat, Inc., Markus Armbruster <armbru@redhat.com>
+ *
+ *  Based on linux/drivers/input/mouse/sermouse.c
+ *
+ *  This file is subject to the terms and conditions of the GNU General Public
+ *  License. See the file COPYING in the main directory of this archive for
+ *  more details.
+ */
+
+/*
+ * TODO:
+ *
+ * Switch to grant tables together with xen-fbfront.c.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/input.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/events.h>
+#include <xen/page.h>
+#include <xen/interface/io/fbif.h>
+#include <xen/interface/io/kbdif.h>
+#include <xen/xenbus.h>
+
+struct xenkbd_info {
+	struct input_dev *kbd;
+	struct input_dev *ptr;
+	struct xenkbd_page *page;
+	int irq;
+	struct xenbus_device *xbdev;
+	char phys[32];
+};
+
+static int xenkbd_remove(struct xenbus_device *);
+static int xenkbd_connect_backend(struct xenbus_device *, struct xenkbd_info *);
+static void xenkbd_disconnect_backend(struct xenkbd_info *);
+
+/*
+ * Note: if you need to send out events, see xenfb_do_update() for how
+ * to do that.
+ */
+
+static irqreturn_t input_handler(int rq, void *dev_id)
+{
+	struct xenkbd_info *info = dev_id;
+	struct xenkbd_page *page = info->page;
+	__u32 cons, prod;
+
+	prod = page->in_prod;
+	if (prod == page->in_cons)
+		return IRQ_HANDLED;
+	rmb();			/* ensure we see ring contents up to prod */
+	for (cons = page->in_cons; cons != prod; cons++) {
+		union xenkbd_in_event *event;
+		struct input_dev *dev;
+		event = &XENKBD_IN_RING_REF(page, cons);
+
+		dev = info->ptr;
+		switch (event->type) {
+		case XENKBD_TYPE_MOTION:
+			input_report_rel(dev, REL_X, event->motion.rel_x);
+			input_report_rel(dev, REL_Y, event->motion.rel_y);
+			break;
+		case XENKBD_TYPE_KEY:
+			dev = NULL;
+			if (test_bit(event->key.keycode, info->kbd->keybit))
+				dev = info->kbd;
+			if (test_bit(event->key.keycode, info->ptr->keybit))
+				dev = info->ptr;
+			if (dev)
+				input_report_key(dev, event->key.keycode,
+						 event->key.pressed);
+			else
+				printk(KERN_WARNING
+				       "xenkbd: unhandled keycode 0x%x\n",
+				       event->key.keycode);
+			break;
+		case XENKBD_TYPE_POS:
+			input_report_abs(dev, ABS_X, event->pos.abs_x);
+			input_report_abs(dev, ABS_Y, event->pos.abs_y);
+			break;
+		}
+		if (dev)
+			input_sync(dev);
+	}
+	mb();			/* ensure we got ring contents */
+	page->in_cons = cons;
+	notify_remote_via_irq(info->irq);
+
+	return IRQ_HANDLED;
+}
+
+static int __devinit xenkbd_probe(struct xenbus_device *dev,
+				  const struct xenbus_device_id *id)
+{
+	int ret, i;
+	struct xenkbd_info *info;
+	struct input_dev *kbd, *ptr;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info) {
+		xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
+		return -ENOMEM;
+	}
+	dev->dev.driver_data = info;
+	info->xbdev = dev;
+	info->irq = -1;
+	snprintf(info->phys, sizeof(info->phys), "xenbus/%s", dev->nodename);
+
+	info->page = (void *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+	if (!info->page)
+		goto error_nomem;
+
+	/* keyboard */
+	kbd = input_allocate_device();
+	if (!kbd)
+		goto error_nomem;
+	kbd->name = "Xen Virtual Keyboard";
+	kbd->phys = info->phys;
+	kbd->id.bustype = BUS_PCI;
+	kbd->id.vendor = 0x5853;
+	kbd->id.product = 0xffff;
+	kbd->evbit[0] = BIT(EV_KEY);
+	for (i = KEY_ESC; i < KEY_UNKNOWN; i++)
+		set_bit(i, kbd->keybit);
+	for (i = KEY_OK; i < KEY_MAX; i++)
+		set_bit(i, kbd->keybit);
+
+	ret = input_register_device(kbd);
+	if (ret) {
+		input_free_device(kbd);
+		xenbus_dev_fatal(dev, ret, "input_register_device(kbd)");
+		goto error;
+	}
+	info->kbd = kbd;
+
+	/* pointing device */
+	ptr = input_allocate_device();
+	if (!ptr)
+		goto error_nomem;
+	ptr->name = "Xen Virtual Pointer";
+	ptr->phys = info->phys;
+	ptr->id.bustype = BUS_PCI;
+	ptr->id.vendor = 0x5853;
+	ptr->id.product = 0xfffe;
+	ptr->evbit[0] = BIT(EV_KEY) | BIT(EV_REL) | BIT(EV_ABS);
+	for (i = BTN_LEFT; i <= BTN_TASK; i++)
+		set_bit(i, ptr->keybit);
+	ptr->relbit[0] = BIT(REL_X) | BIT(REL_Y);
+	input_set_abs_params(ptr, ABS_X, 0, XENFB_WIDTH, 0, 0);
+	input_set_abs_params(ptr, ABS_Y, 0, XENFB_HEIGHT, 0, 0);
+
+	ret = input_register_device(ptr);
+	if (ret) {
+		input_free_device(ptr);
+		xenbus_dev_fatal(dev, ret, "input_register_device(ptr)");
+		goto error;
+	}
+	info->ptr = ptr;
+
+	ret = xenkbd_connect_backend(dev, info);
+	if (ret < 0)
+		goto error;
+
+	return 0;
+
+ error_nomem:
+	ret = -ENOMEM;
+	xenbus_dev_fatal(dev, ret, "allocating device memory");
+ error:
+	xenkbd_remove(dev);
+	return ret;
+}
+
+static int xenkbd_resume(struct xenbus_device *dev)
+{
+	struct xenkbd_info *info = dev->dev.driver_data;
+
+	xenkbd_disconnect_backend(info);
+	memset(info->page, 0, PAGE_SIZE);
+	return xenkbd_connect_backend(dev, info);
+}
+
+static int xenkbd_remove(struct xenbus_device *dev)
+{
+	struct xenkbd_info *info = dev->dev.driver_data;
+
+	xenkbd_disconnect_backend(info);
+	if (info->kbd)
+		input_unregister_device(info->kbd);
+	if (info->ptr)
+		input_unregister_device(info->ptr);
+	free_page((unsigned long)info->page);
+	kfree(info);
+	return 0;
+}
+
+static int xenkbd_connect_backend(struct xenbus_device *dev,
+				  struct xenkbd_info *info)
+{
+	int ret, evtchn;
+	struct xenbus_transaction xbt;
+
+	ret = xenbus_alloc_evtchn(dev, &evtchn);
+	if (ret)
+		return ret;
+	ret = bind_evtchn_to_irqhandler(evtchn, input_handler,
+					0, dev->devicetype, info);
+	if (ret < 0) {
+		xenbus_free_evtchn(dev, evtchn);
+		xenbus_dev_fatal(dev, ret, "bind_evtchn_to_irqhandler");
+		return ret;
+	}
+	info->irq = ret;
+
+ again:
+	ret = xenbus_transaction_start(&xbt);
+	if (ret) {
+		xenbus_dev_fatal(dev, ret, "starting transaction");
+		return ret;
+	}
+	ret = xenbus_printf(xbt, dev->nodename, "page-ref", "%lu",
+			    virt_to_mfn(info->page));
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_printf(xbt, dev->nodename, "event-channel", "%u",
+			    evtchn);
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_transaction_end(xbt, 0);
+	if (ret) {
+		if (ret == -EAGAIN)
+			goto again;
+		xenbus_dev_fatal(dev, ret, "completing transaction");
+		return ret;
+	}
+
+	xenbus_switch_state(dev, XenbusStateInitialised);
+	return 0;
+
+ error_xenbus:
+	xenbus_transaction_end(xbt, 1);
+	xenbus_dev_fatal(dev, ret, "writing xenstore");
+	return ret;
+}
+
+static void xenkbd_disconnect_backend(struct xenkbd_info *info)
+{
+	if (info->irq >= 0)
+		unbind_from_irqhandler(info->irq, info);
+	info->irq = -1;
+}
+
+static void xenkbd_backend_changed(struct xenbus_device *dev,
+				   enum xenbus_state backend_state)
+{
+	struct xenkbd_info *info = dev->dev.driver_data;
+	int ret, val;
+
+	switch (backend_state) {
+	case XenbusStateInitialising:
+	case XenbusStateInitialised:
+	case XenbusStateUnknown:
+	case XenbusStateClosed:
+		break;
+
+	case XenbusStateInitWait:
+InitWait:
+		ret = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+				   "feature-abs-pointer", "%d", &val);
+		if (ret < 0)
+			val = 0;
+		if (val) {
+			ret = xenbus_printf(XBT_NIL, info->xbdev->nodename,
+					    "request-abs-pointer", "1");
+			if (ret)
+				printk(KERN_WARNING
+				       "xenkbd: can't request abs-pointer");
+		}
+		xenbus_switch_state(dev, XenbusStateConnected);
+		break;
+
+	case XenbusStateConnected:
+		/*
+		 * Work around xenbus race condition: If backend goes
+		 * through InitWait to Connected fast enough, we can
+		 * get Connected twice here.
+		 */
+		if (dev->state != XenbusStateConnected)
+			goto InitWait; /* no InitWait seen yet, fudge it */
+		break;
+
+	case XenbusStateClosing:
+		xenbus_frontend_closed(dev);
+		break;
+	}
+}
+
+static struct xenbus_device_id xenkbd_ids[] = {
+	{ "vkbd" },
+	{ "" }
+};
+
+static struct xenbus_driver xenkbd = {
+	.name = "vkbd",
+	.owner = THIS_MODULE,
+	.ids = xenkbd_ids,
+	.probe = xenkbd_probe,
+	.remove = xenkbd_remove,
+	.resume = xenkbd_resume,
+	.otherend_changed = xenkbd_backend_changed,
+};
+
+static int __init xenkbd_init(void)
+{
+	if (!is_running_on_xen())
+		return -ENODEV;
+
+	/* Nothing to do if running in dom0. */
+	if (is_initial_xendomain())
+		return -ENODEV;
+
+	return xenbus_register_frontend(&xenkbd);
+}
+
+static void __exit xenkbd_cleanup(void)
+{
+	xenbus_unregister_driver(&xenkbd);
+}
+
+module_init(xenkbd_init);
+module_exit(xenkbd_cleanup);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 758435f..67de177 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1903,6 +1903,20 @@ config FB_VIRTUAL
 
 	  If unsure, say N.
 
+config XEN_FBDEV_FRONTEND
+	tristate "Xen virtual frame buffer support"
+	depends on FB && XEN
+	select FB_SYS_FILLRECT
+	select FB_SYS_COPYAREA
+	select FB_SYS_IMAGEBLIT
+	select FB_SYS_FOPS
+	select FB_DEFERRED_IO
+	default y
+	help
+	  This driver implements the front-end of the Xen virtual
+	  frame buffer driver.  It communicates with a back-end
+	  in another domain.
+
 source "drivers/video/omap/Kconfig"
 
 source "drivers/video/backlight/Kconfig"
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 83e02b3..d3ab9b2 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -113,6 +113,7 @@ obj-$(CONFIG_FB_PS3)		  += ps3fb.o
 obj-$(CONFIG_FB_SM501)            += sm501fb.o
 obj-$(CONFIG_FB_XILINX)           += xilinxfb.o
 obj-$(CONFIG_FB_OMAP)             += omap/
+obj-$(CONFIG_XEN_FBDEV_FRONTEND)  += xen-fbfront.o
 
 # Platform or fallback drivers go here
 obj-$(CONFIG_FB_UVESA)            += uvesafb.o
diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c
new file mode 100644
index 0000000..619a6f8
--- /dev/null
+++ b/drivers/video/xen-fbfront.c
@@ -0,0 +1,550 @@
+/*
+ * Xen para-virtual frame buffer device
+ *
+ * Copyright (C) 2005-2006 Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2006-2008 Red Hat, Inc., Markus Armbruster <armbru@redhat.com>
+ *
+ *  Based on linux/drivers/video/q40fb.c
+ *
+ *  This file is subject to the terms and conditions of the GNU General Public
+ *  License. See the file COPYING in the main directory of this archive for
+ *  more details.
+ */
+
+/*
+ * TODO:
+ *
+ * Switch to grant tables when they become capable of dealing with the
+ * frame buffer.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fb.h>
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/events.h>
+#include <xen/page.h>
+#include <xen/interface/io/fbif.h>
+#include <xen/interface/io/protocols.h>
+#include <xen/xenbus.h>
+
+struct xenfb_info {
+	unsigned char		*fb;
+	struct fb_info		*fb_info;
+	int			x1, y1, x2, y2;	/* dirty rectangle,
+						   protected by dirty_lock */
+	spinlock_t		dirty_lock;
+	int			nr_pages;
+	int			irq;
+	struct xenfb_page	*page;
+	unsigned long 		*mfns;
+	int			update_wanted; /* XENFB_TYPE_UPDATE wanted */
+
+	struct xenbus_device	*xbdev;
+};
+
+static u32 xenfb_mem_len = XENFB_WIDTH * XENFB_HEIGHT * XENFB_DEPTH / 8;
+
+static int xenfb_remove(struct xenbus_device *);
+static void xenfb_init_shared_page(struct xenfb_info *);
+static int xenfb_connect_backend(struct xenbus_device *, struct xenfb_info *);
+static void xenfb_disconnect_backend(struct xenfb_info *);
+
+static void xenfb_do_update(struct xenfb_info *info,
+			    int x, int y, int w, int h)
+{
+	union xenfb_out_event event;
+	u32 prod;
+
+	event.type = XENFB_TYPE_UPDATE;
+	event.update.x = x;
+	event.update.y = y;
+	event.update.width = w;
+	event.update.height = h;
+
+	prod = info->page->out_prod;
+	/* caller ensures !xenfb_queue_full() */
+	mb();			/* ensure ring space available */
+	XENFB_OUT_RING_REF(info->page, prod) = event;
+	wmb();			/* ensure ring contents visible */
+	info->page->out_prod = prod + 1;
+
+	notify_remote_via_irq(info->irq);
+}
+
+static int xenfb_queue_full(struct xenfb_info *info)
+{
+	u32 cons, prod;
+
+	prod = info->page->out_prod;
+	cons = info->page->out_cons;
+	return prod - cons == XENFB_OUT_RING_LEN;
+}
+
+static void xenfb_refresh(struct xenfb_info *info,
+			  int x1, int y1, int w, int h)
+{
+	unsigned long flags;
+	int y2 = y1 + h - 1;
+	int x2 = x1 + w - 1;
+
+	if (!info->update_wanted)
+		return;
+
+	spin_lock_irqsave(&info->dirty_lock, flags);
+
+	/* Combine with dirty rectangle: */
+	if (info->y1 < y1)
+		y1 = info->y1;
+	if (info->y2 > y2)
+		y2 = info->y2;
+	if (info->x1 < x1)
+		x1 = info->x1;
+	if (info->x2 > x2)
+		x2 = info->x2;
+
+	if (xenfb_queue_full(info)) {
+		/* Can't send right now, stash it in the dirty rectangle */
+		info->x1 = x1;
+		info->x2 = x2;
+		info->y1 = y1;
+		info->y2 = y2;
+		spin_unlock_irqrestore(&info->dirty_lock, flags);
+		return;
+	}
+
+	/* Clear dirty rectangle: */
+	info->x1 = info->y1 = INT_MAX;
+	info->x2 = info->y2 = 0;
+
+	spin_unlock_irqrestore(&info->dirty_lock, flags);
+
+	if (x1 <= x2 && y1 <= y2)
+		xenfb_do_update(info, x1, y1, x2 - x1 + 1, y2 - y1 + 1);
+}
+
+static void xenfb_deferred_io(struct fb_info *fb_info,
+			      struct list_head *pagelist)
+{
+	struct xenfb_info *info = fb_info->par;
+	struct page *page;
+	unsigned long beg, end;
+	int y1, y2, miny, maxy;
+
+	miny = INT_MAX;
+	maxy = 0;
+	list_for_each_entry(page, pagelist, lru) {
+		beg = page->index << PAGE_SHIFT;
+		end = beg + PAGE_SIZE - 1;
+		y1 = beg / fb_info->fix.line_length;
+		y2 = end / fb_info->fix.line_length;
+		if (y2 >= fb_info->var.yres)
+			y2 = fb_info->var.yres - 1;
+		if (miny > y1)
+			miny = y1;
+		if (maxy < y2)
+			maxy = y2;
+	}
+	xenfb_refresh(info, 0, miny, fb_info->var.xres, maxy - miny + 1);
+}
+
+static struct fb_deferred_io xenfb_defio = {
+	.delay		= HZ / 20,
+	.deferred_io	= xenfb_deferred_io,
+};
+
+static int xenfb_setcolreg(unsigned regno, unsigned red, unsigned green,
+			   unsigned blue, unsigned transp,
+			   struct fb_info *info)
+{
+	u32 v;
+
+	if (regno > info->cmap.len)
+		return 1;
+
+#define CNVT_TOHW(val, width) ((((val)<<(width))+0x7FFF-(val))>>16)
+	red = CNVT_TOHW(red, info->var.red.length);
+	green = CNVT_TOHW(green, info->var.green.length);
+	blue = CNVT_TOHW(blue, info->var.blue.length);
+	transp = CNVT_TOHW(transp, info->var.transp.length);
+#undef CNVT_TOHW
+
+	v = (red << info->var.red.offset) |
+	    (green << info->var.green.offset) |
+	    (blue << info->var.blue.offset);
+
+	switch (info->var.bits_per_pixel) {
+	case 16:
+	case 24:
+	case 32:
+		((u32 *)info->pseudo_palette)[regno] = v;
+		break;
+	}
+
+	return 0;
+}
+
+static void xenfb_fillrect(struct fb_info *p, const struct fb_fillrect *rect)
+{
+	struct xenfb_info *info = p->par;
+
+	sys_fillrect(p, rect);
+	xenfb_refresh(info, rect->dx, rect->dy, rect->width, rect->height);
+}
+
+static void xenfb_imageblit(struct fb_info *p, const struct fb_image *image)
+{
+	struct xenfb_info *info = p->par;
+
+	sys_imageblit(p, image);
+	xenfb_refresh(info, image->dx, image->dy, image->width, image->height);
+}
+
+static void xenfb_copyarea(struct fb_info *p, const struct fb_copyarea *area)
+{
+	struct xenfb_info *info = p->par;
+
+	sys_copyarea(p, area);
+	xenfb_refresh(info, area->dx, area->dy, area->width, area->height);
+}
+
+static ssize_t xenfb_write(struct fb_info *p, const char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct xenfb_info *info = p->par;
+	ssize_t res;
+
+	res = fb_sys_write(p, buf, count, ppos);
+	xenfb_refresh(info, 0, 0, info->page->width, info->page->height);
+	return res;
+}
+
+static struct fb_ops xenfb_fb_ops = {
+	.owner		= THIS_MODULE,
+	.fb_read	= fb_sys_read,
+	.fb_write	= xenfb_write,
+	.fb_setcolreg	= xenfb_setcolreg,
+	.fb_fillrect	= xenfb_fillrect,
+	.fb_copyarea	= xenfb_copyarea,
+	.fb_imageblit	= xenfb_imageblit,
+};
+
+static irqreturn_t xenfb_event_handler(int rq, void *dev_id)
+{
+	/*
+	 * No in events recognized, simply ignore them all.
+	 * If you need to recognize some, see xen-kbdfront's
+	 * input_handler() for how to do that.
+	 */
+	struct xenfb_info *info = dev_id;
+	struct xenfb_page *page = info->page;
+
+	if (page->in_cons != page->in_prod) {
+		info->page->in_cons = info->page->in_prod;
+		notify_remote_via_irq(info->irq);
+	}
+
+	/* Flush dirty rectangle: */
+	xenfb_refresh(info, INT_MAX, INT_MAX, -INT_MAX, -INT_MAX);
+
+	return IRQ_HANDLED;
+}
+
+static int __devinit xenfb_probe(struct xenbus_device *dev,
+				 const struct xenbus_device_id *id)
+{
+	struct xenfb_info *info;
+	struct fb_info *fb_info;
+	int ret;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (info == NULL) {
+		xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
+		return -ENOMEM;
+	}
+	dev->dev.driver_data = info;
+	info->xbdev = dev;
+	info->irq = -1;
+	info->x1 = info->y1 = INT_MAX;
+	spin_lock_init(&info->dirty_lock);
+
+	info->fb = vmalloc(xenfb_mem_len);
+	if (info->fb == NULL)
+		goto error_nomem;
+	memset(info->fb, 0, xenfb_mem_len);
+
+	info->nr_pages = (xenfb_mem_len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+	info->mfns = vmalloc(sizeof(unsigned long) * info->nr_pages);
+	if (!info->mfns)
+		goto error_nomem;
+
+	/* set up shared page */
+	info->page = (void *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+	if (!info->page)
+		goto error_nomem;
+
+	xenfb_init_shared_page(info);
+
+	/* abusing framebuffer_alloc() to allocate pseudo_palette */
+	fb_info = framebuffer_alloc(sizeof(u32) * 256, NULL);
+	if (fb_info == NULL)
+		goto error_nomem;
+
+	/* complete the abuse: */
+	fb_info->pseudo_palette = fb_info->par;
+	fb_info->par = info;
+
+	fb_info->screen_base = info->fb;
+
+	fb_info->fbops = &xenfb_fb_ops;
+	fb_info->var.xres_virtual = fb_info->var.xres = info->page->width;
+	fb_info->var.yres_virtual = fb_info->var.yres = info->page->height;
+	fb_info->var.bits_per_pixel = info->page->depth;
+
+	fb_info->var.red = (struct fb_bitfield){16, 8, 0};
+	fb_info->var.green = (struct fb_bitfield){8, 8, 0};
+	fb_info->var.blue = (struct fb_bitfield){0, 8, 0};
+
+	fb_info->var.activate = FB_ACTIVATE_NOW;
+	fb_info->var.height = -1;
+	fb_info->var.width = -1;
+	fb_info->var.vmode = FB_VMODE_NONINTERLACED;
+
+	fb_info->fix.visual = FB_VISUAL_TRUECOLOR;
+	fb_info->fix.line_length = info->page->line_length;
+	fb_info->fix.smem_start = 0;
+	fb_info->fix.smem_len = xenfb_mem_len;
+	strcpy(fb_info->fix.id, "xen");
+	fb_info->fix.type = FB_TYPE_PACKED_PIXELS;
+	fb_info->fix.accel = FB_ACCEL_NONE;
+
+	fb_info->flags = FBINFO_FLAG_DEFAULT;
+
+	ret = fb_alloc_cmap(&fb_info->cmap, 256, 0);
+	if (ret < 0) {
+		framebuffer_release(fb_info);
+		xenbus_dev_fatal(dev, ret, "fb_alloc_cmap");
+		goto error;
+	}
+
+	fb_info->fbdefio = &xenfb_defio;
+	fb_deferred_io_init(fb_info);
+
+	ret = register_framebuffer(fb_info);
+	if (ret) {
+		fb_deferred_io_cleanup(fb_info);
+		fb_dealloc_cmap(&fb_info->cmap);
+		framebuffer_release(fb_info);
+		xenbus_dev_fatal(dev, ret, "register_framebuffer");
+		goto error;
+	}
+	info->fb_info = fb_info;
+
+	ret = xenfb_connect_backend(dev, info);
+	if (ret < 0)
+		goto error;
+
+	return 0;
+
+ error_nomem:
+	ret = -ENOMEM;
+	xenbus_dev_fatal(dev, ret, "allocating device memory");
+ error:
+	xenfb_remove(dev);
+	return ret;
+}
+
+static int xenfb_resume(struct xenbus_device *dev)
+{
+	struct xenfb_info *info = dev->dev.driver_data;
+
+	xenfb_disconnect_backend(info);
+	xenfb_init_shared_page(info);
+	return xenfb_connect_backend(dev, info);
+}
+
+static int xenfb_remove(struct xenbus_device *dev)
+{
+	struct xenfb_info *info = dev->dev.driver_data;
+
+	xenfb_disconnect_backend(info);
+	if (info->fb_info) {
+		fb_deferred_io_cleanup(info->fb_info);
+		unregister_framebuffer(info->fb_info);
+		fb_dealloc_cmap(&info->fb_info->cmap);
+		framebuffer_release(info->fb_info);
+	}
+	free_page((unsigned long)info->page);
+	vfree(info->mfns);
+	vfree(info->fb);
+	kfree(info);
+
+	return 0;
+}
+
+static unsigned long vmalloc_to_mfn(void *address)
+{
+	return pfn_to_mfn(vmalloc_to_pfn(address));
+}
+
+static void xenfb_init_shared_page(struct xenfb_info *info)
+{
+	int i;
+
+	for (i = 0; i < info->nr_pages; i++)
+		info->mfns[i] = vmalloc_to_mfn(info->fb + i * PAGE_SIZE);
+
+	info->page->pd[0] = vmalloc_to_mfn(info->mfns);
+	info->page->pd[1] = 0;
+	info->page->width = XENFB_WIDTH;
+	info->page->height = XENFB_HEIGHT;
+	info->page->depth = XENFB_DEPTH;
+	info->page->line_length = (info->page->depth / 8) * info->page->width;
+	info->page->mem_length = xenfb_mem_len;
+	info->page->in_cons = info->page->in_prod = 0;
+	info->page->out_cons = info->page->out_prod = 0;
+}
+
+static int xenfb_connect_backend(struct xenbus_device *dev,
+				 struct xenfb_info *info)
+{
+	int ret, evtchn;
+	struct xenbus_transaction xbt;
+
+	ret = xenbus_alloc_evtchn(dev, &evtchn);
+	if (ret)
+		return ret;
+	ret = bind_evtchn_to_irqhandler(evtchn, xenfb_event_handler,
+					0, dev->devicetype, info);
+	if (ret < 0) {
+		xenbus_free_evtchn(dev, evtchn);
+		xenbus_dev_fatal(dev, ret, "bind_evtchn_to_irqhandler");
+		return ret;
+	}
+	info->irq = ret;
+
+ again:
+	ret = xenbus_transaction_start(&xbt);
+	if (ret) {
+		xenbus_dev_fatal(dev, ret, "starting transaction");
+		return ret;
+	}
+	ret = xenbus_printf(xbt, dev->nodename, "page-ref", "%lu",
+			    virt_to_mfn(info->page));
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_printf(xbt, dev->nodename, "event-channel", "%u",
+			    evtchn);
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
+			    XEN_IO_PROTO_ABI_NATIVE);
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_printf(xbt, dev->nodename, "feature-update", "1");
+	if (ret)
+		goto error_xenbus;
+	ret = xenbus_transaction_end(xbt, 0);
+	if (ret) {
+		if (ret == -EAGAIN)
+			goto again;
+		xenbus_dev_fatal(dev, ret, "completing transaction");
+		return ret;
+	}
+
+	xenbus_switch_state(dev, XenbusStateInitialised);
+	return 0;
+
+ error_xenbus:
+	xenbus_transaction_end(xbt, 1);
+	xenbus_dev_fatal(dev, ret, "writing xenstore");
+	return ret;
+}
+
+static void xenfb_disconnect_backend(struct xenfb_info *info)
+{
+	if (info->irq >= 0)
+		unbind_from_irqhandler(info->irq, info);
+	info->irq = -1;
+}
+
+static void xenfb_backend_changed(struct xenbus_device *dev,
+				  enum xenbus_state backend_state)
+{
+	struct xenfb_info *info = dev->dev.driver_data;
+	int val;
+
+	switch (backend_state) {
+	case XenbusStateInitialising:
+	case XenbusStateInitialised:
+	case XenbusStateUnknown:
+	case XenbusStateClosed:
+		break;
+
+	case XenbusStateInitWait:
+InitWait:
+		xenbus_switch_state(dev, XenbusStateConnected);
+		break;
+
+	case XenbusStateConnected:
+		/*
+		 * Work around xenbus race condition: If backend goes
+		 * through InitWait to Connected fast enough, we can
+		 * get Connected twice here.
+		 */
+		if (dev->state != XenbusStateConnected)
+			goto InitWait; /* no InitWait seen yet, fudge it */
+
+		if (xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+				 "request-update", "%d", &val) < 0)
+			val = 0;
+		if (val)
+			info->update_wanted = 1;
+		break;
+
+	case XenbusStateClosing:
+		xenbus_frontend_closed(dev);
+		break;
+	}
+}
+
+static struct xenbus_device_id xenfb_ids[] = {
+	{ "vfb" },
+	{ "" }
+};
+
+static struct xenbus_driver xenfb = {
+	.name = "vfb",
+	.owner = THIS_MODULE,
+	.ids = xenfb_ids,
+	.probe = xenfb_probe,
+	.remove = xenfb_remove,
+	.resume = xenfb_resume,
+	.otherend_changed = xenfb_backend_changed,
+};
+
+static int __init xenfb_init(void)
+{
+	if (!is_running_on_xen())
+		return -ENODEV;
+
+	/* Nothing to do if running in dom0. */
+	if (is_initial_xendomain())
+		return -ENODEV;
+
+	return xenbus_register_frontend(&xenfb);
+}
+
+static void __exit xenfb_cleanup(void)
+{
+	xenbus_unregister_driver(&xenfb);
+}
+
+module_init(xenfb_init);
+module_exit(xenfb_cleanup);
+
+MODULE_LICENSE("GPL");
diff --git a/include/xen/interface/io/fbif.h b/include/xen/interface/io/fbif.h
new file mode 100644
index 0000000..5a934dd
--- /dev/null
+++ b/include/xen/interface/io/fbif.h
@@ -0,0 +1,124 @@
+/*
+ * fbif.h -- Xen virtual frame buffer device
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Copyright (C) 2005 Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2006 Red Hat, Inc., Markus Armbruster <armbru@redhat.com>
+ */
+
+#ifndef __XEN_PUBLIC_IO_FBIF_H__
+#define __XEN_PUBLIC_IO_FBIF_H__
+
+/* Out events (frontend -> backend) */
+
+/*
+ * Out events may be sent only when requested by backend, and receipt
+ * of an unknown out event is an error.
+ */
+
+/* Event type 1 currently not used */
+/*
+ * Framebuffer update notification event
+ * Capable frontend sets feature-update in xenstore.
+ * Backend requests it by setting request-update in xenstore.
+ */
+#define XENFB_TYPE_UPDATE 2
+
+struct xenfb_update {
+	uint8_t type;		/* XENFB_TYPE_UPDATE */
+	int32_t x;		/* source x */
+	int32_t y;		/* source y */
+	int32_t width;		/* rect width */
+	int32_t height;		/* rect height */
+};
+
+#define XENFB_OUT_EVENT_SIZE 40
+
+union xenfb_out_event {
+	uint8_t type;
+	struct xenfb_update update;
+	char pad[XENFB_OUT_EVENT_SIZE];
+};
+
+/* In events (backend -> frontend) */
+
+/*
+ * Frontends should ignore unknown in events.
+ * No in events currently defined.
+ */
+
+#define XENFB_IN_EVENT_SIZE 40
+
+union xenfb_in_event {
+	uint8_t type;
+	char pad[XENFB_IN_EVENT_SIZE];
+};
+
+/* shared page */
+
+#define XENFB_IN_RING_SIZE 1024
+#define XENFB_IN_RING_LEN (XENFB_IN_RING_SIZE / XENFB_IN_EVENT_SIZE)
+#define XENFB_IN_RING_OFFS 1024
+#define XENFB_IN_RING(page) \
+	((union xenfb_in_event *)((char *)(page) + XENFB_IN_RING_OFFS))
+#define XENFB_IN_RING_REF(page, idx) \
+	(XENFB_IN_RING((page))[(idx) % XENFB_IN_RING_LEN])
+
+#define XENFB_OUT_RING_SIZE 2048
+#define XENFB_OUT_RING_LEN (XENFB_OUT_RING_SIZE / XENFB_OUT_EVENT_SIZE)
+#define XENFB_OUT_RING_OFFS (XENFB_IN_RING_OFFS + XENFB_IN_RING_SIZE)
+#define XENFB_OUT_RING(page) \
+	((union xenfb_out_event *)((char *)(page) + XENFB_OUT_RING_OFFS))
+#define XENFB_OUT_RING_REF(page, idx) \
+	(XENFB_OUT_RING((page))[(idx) % XENFB_OUT_RING_LEN])
+
+struct xenfb_page {
+	uint32_t in_cons, in_prod;
+	uint32_t out_cons, out_prod;
+
+	int32_t width;          /* width of the framebuffer (in pixels) */
+	int32_t height;         /* height of the framebuffer (in pixels) */
+	uint32_t line_length;   /* length of a row of pixels (in bytes) */
+	uint32_t mem_length;    /* length of the framebuffer (in bytes) */
+	uint8_t depth;          /* depth of a pixel (in bits) */
+
+	/*
+	 * Framebuffer page directory
+	 *
+	 * Each directory page holds PAGE_SIZE / sizeof(*pd)
+	 * framebuffer pages, and can thus map up to PAGE_SIZE *
+	 * PAGE_SIZE / sizeof(*pd) bytes.  With PAGE_SIZE == 4096 and
+	 * sizeof(unsigned long) == 4, that's 4 Megs.  Two directory
+	 * pages should be enough for a while.
+	 */
+	unsigned long pd[2];
+};
+
+/*
+ * Wart: xenkbd needs to know resolution.  Put it here until a better
+ * solution is found, but don't leak it to the backend.
+ */
+#ifdef __KERNEL__
+#define XENFB_WIDTH 800
+#define XENFB_HEIGHT 600
+#define XENFB_DEPTH 32
+#endif
+
+#endif
diff --git a/include/xen/interface/io/kbdif.h b/include/xen/interface/io/kbdif.h
new file mode 100644
index 0000000..fb97f42
--- /dev/null
+++ b/include/xen/interface/io/kbdif.h
@@ -0,0 +1,114 @@
+/*
+ * kbdif.h -- Xen virtual keyboard/mouse
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Copyright (C) 2005 Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2006 Red Hat, Inc., Markus Armbruster <armbru@redhat.com>
+ */
+
+#ifndef __XEN_PUBLIC_IO_KBDIF_H__
+#define __XEN_PUBLIC_IO_KBDIF_H__
+
+/* In events (backend -> frontend) */
+
+/*
+ * Frontends should ignore unknown in events.
+ */
+
+/* Pointer movement event */
+#define XENKBD_TYPE_MOTION  1
+/* Event type 2 currently not used */
+/* Key event (includes pointer buttons) */
+#define XENKBD_TYPE_KEY     3
+/*
+ * Pointer position event
+ * Capable backend sets feature-abs-pointer in xenstore.
+ * Frontend requests ot instead of XENKBD_TYPE_MOTION by setting
+ * request-abs-update in xenstore.
+ */
+#define XENKBD_TYPE_POS     4
+
+struct xenkbd_motion {
+	uint8_t type;		/* XENKBD_TYPE_MOTION */
+	int32_t rel_x;		/* relative X motion */
+	int32_t rel_y;		/* relative Y motion */
+};
+
+struct xenkbd_key {
+	uint8_t type;		/* XENKBD_TYPE_KEY */
+	uint8_t pressed;	/* 1 if pressed; 0 otherwise */
+	uint32_t keycode;	/* KEY_* from linux/input.h */
+};
+
+struct xenkbd_position {
+	uint8_t type;		/* XENKBD_TYPE_POS */
+	int32_t abs_x;		/* absolute X position (in FB pixels) */
+	int32_t abs_y;		/* absolute Y position (in FB pixels) */
+};
+
+#define XENKBD_IN_EVENT_SIZE 40
+
+union xenkbd_in_event {
+	uint8_t type;
+	struct xenkbd_motion motion;
+	struct xenkbd_key key;
+	struct xenkbd_position pos;
+	char pad[XENKBD_IN_EVENT_SIZE];
+};
+
+/* Out events (frontend -> backend) */
+
+/*
+ * Out events may be sent only when requested by backend, and receipt
+ * of an unknown out event is an error.
+ * No out events currently defined.
+ */
+
+#define XENKBD_OUT_EVENT_SIZE 40
+
+union xenkbd_out_event {
+	uint8_t type;
+	char pad[XENKBD_OUT_EVENT_SIZE];
+};
+
+/* shared page */
+
+#define XENKBD_IN_RING_SIZE 2048
+#define XENKBD_IN_RING_LEN (XENKBD_IN_RING_SIZE / XENKBD_IN_EVENT_SIZE)
+#define XENKBD_IN_RING_OFFS 1024
+#define XENKBD_IN_RING(page) \
+	((union xenkbd_in_event *)((char *)(page) + XENKBD_IN_RING_OFFS))
+#define XENKBD_IN_RING_REF(page, idx) \
+	(XENKBD_IN_RING((page))[(idx) % XENKBD_IN_RING_LEN])
+
+#define XENKBD_OUT_RING_SIZE 1024
+#define XENKBD_OUT_RING_LEN (XENKBD_OUT_RING_SIZE / XENKBD_OUT_EVENT_SIZE)
+#define XENKBD_OUT_RING_OFFS (XENKBD_IN_RING_OFFS + XENKBD_IN_RING_SIZE)
+#define XENKBD_OUT_RING(page) \
+	((union xenkbd_out_event *)((char *)(page) + XENKBD_OUT_RING_OFFS))
+#define XENKBD_OUT_RING_REF(page, idx) \
+	(XENKBD_OUT_RING((page))[(idx) % XENKBD_OUT_RING_LEN])
+
+struct xenkbd_page {
+	uint32_t in_cons, in_prod;
+	uint32_t out_cons, out_prod;
+};
+
+#endif

^ permalink raw reply related	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-02-25 13:03 ` [PATCH 2/3] fbdev: Make deferred I/O work as advertized Markus Armbruster
@ 2008-02-26 14:11   ` Jaya Kumar
  2008-02-26 14:45     ` Markus Armbruster
  0 siblings, 1 reply; 13+ messages in thread
From: Jaya Kumar @ 2008-02-26 14:11 UTC (permalink / raw)
  To: Markus Armbruster
  Cc: linux-kernel, virtualization, xen-devel, linux-fbdev-devel,
	adaplas, linux-input, dmitry.torokhov, akpm

On Mon, Feb 25, 2008 at 8:03 AM, Markus Armbruster <armbru@redhat.com> wrote:
>
>     Subject: [PATCH 1/1 2.6.24] fbdev: defio and Metronomefb
>     From: Jaya Kumar <jayakumar.lkml@gmail.com>
>     Date: 2008-02-18 13:41:26

Hi Markus,

Andrew pointed out that there may be race conditions associated with
this patch. [ http://marc.info/?l=linux-fbdev-devel&m=120376473020396&w=2
] So I would not encourage anyone to merge it. I'll try to figure
things out this weekend.

Thanks,
jaya

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-02-26 14:11   ` Jaya Kumar
@ 2008-02-26 14:45     ` Markus Armbruster
  2008-02-26 15:01       ` Jaya Kumar
  0 siblings, 1 reply; 13+ messages in thread
From: Markus Armbruster @ 2008-02-26 14:45 UTC (permalink / raw)
  To: Jaya Kumar
  Cc: linux-kernel, virtualization, xen-devel, linux-fbdev-devel,
	adaplas, linux-input, dmitry.torokhov, akpm

"Jaya Kumar" <jayakumar.lkml@gmail.com> writes:

> On Mon, Feb 25, 2008 at 8:03 AM, Markus Armbruster <armbru@redhat.com> wrote:
>>
>>     Subject: [PATCH 1/1 2.6.24] fbdev: defio and Metronomefb
>>     From: Jaya Kumar <jayakumar.lkml@gmail.com>
>>     Date: 2008-02-18 13:41:26
>
> Hi Markus,
>
> Andrew pointed out that there may be race conditions associated with
> this patch. [ http://marc.info/?l=linux-fbdev-devel&m=120376473020396&w=2
> ] So I would not encourage anyone to merge it. I'll try to figure
> things out this weekend.
>
> Thanks,
> jaya

Thanks for the timely info.  I'm not in an undue hurry to get this
merged.  As long as we're moving forward, I'm happy.

What about pushing the fb_defio fixes independently of any new
fb_defio users?  If fb_defio was worth merging into Linus's tree, it
should be worth fixing there, whether new users are in shape already
or not.

If we must have a new user, well, I could easily whip up something
like FB_VIRTUAL on top of fb_defio.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-02-26 14:45     ` Markus Armbruster
@ 2008-02-26 15:01       ` Jaya Kumar
  2008-03-19 10:45         ` Markus Armbruster
  0 siblings, 1 reply; 13+ messages in thread
From: Jaya Kumar @ 2008-02-26 15:01 UTC (permalink / raw)
  To: Markus Armbruster
  Cc: linux-kernel, virtualization, xen-devel, linux-fbdev-devel,
	adaplas, linux-input, dmitry.torokhov, akpm

On Tue, Feb 26, 2008 at 9:45 AM, Markus Armbruster <armbru@redhat.com> wrote:
>
>  What about pushing the fb_defio fixes independently of any new
>  fb_defio users?  If fb_defio was worth merging into Linus's tree, it
>  should be worth fixing there, whether new users are in shape already
>  or not.

I think that Andrew's message is saying that there may be a race
condition in the defio patch itself as opposed to the defio user
patch.

If there is no race condition or other problems, then I think it would
make sense to merge the defio patch independent of metronomefb or
other new patches that use defio.

Thanks,
jaya

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-02-26 15:01       ` Jaya Kumar
@ 2008-03-19 10:45         ` Markus Armbruster
  2008-03-19 11:24           ` Jaya Kumar
  0 siblings, 1 reply; 13+ messages in thread
From: Markus Armbruster @ 2008-03-19 10:45 UTC (permalink / raw)
  To: Jaya Kumar
  Cc: linux-kernel, virtualization, xen-devel, linux-fbdev-devel,
	adaplas, linux-input, dmitry.torokhov, akpm

"Jaya Kumar" <jayakumar.lkml@gmail.com> writes:

> On Tue, Feb 26, 2008 at 9:45 AM, Markus Armbruster <armbru@redhat.com> wrote:
>>
>>  What about pushing the fb_defio fixes independently of any new
>>  fb_defio users?  If fb_defio was worth merging into Linus's tree, it
>>  should be worth fixing there, whether new users are in shape already
>>  or not.
>
> I think that Andrew's message is saying that there may be a race
> condition in the defio patch itself as opposed to the defio user
> patch.
>
> If there is no race condition or other problems, then I think it would
> make sense to merge the defio patch independent of metronomefb or
> other new patches that use defio.
>
> Thanks,
> jaya

Any progress on this?

Considering that fb_defio is utterly broken without the fix (writing
the frame buffer makes the VM endlessly invoke vm_ops.page_mkwrite()),
wouldn't it make sense to merge the fix even if it still has issues?

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-03-19 10:45         ` Markus Armbruster
@ 2008-03-19 11:24           ` Jaya Kumar
  2008-03-19 18:37             ` Andrew Morton
  0 siblings, 1 reply; 13+ messages in thread
From: Jaya Kumar @ 2008-03-19 11:24 UTC (permalink / raw)
  To: Markus Armbruster
  Cc: linux-kernel, virtualization, xen-devel, linux-fbdev-devel,
	adaplas, linux-input, dmitry.torokhov, akpm

On Wed, Mar 19, 2008 at 6:45 PM, Markus Armbruster <armbru@redhat.com> wrote:
>
>  Any progress on this?
>
>  Considering that fb_defio is utterly broken without the fix (writing
>  the frame buffer makes the VM endlessly invoke vm_ops.page_mkwrite()),
>  wouldn't it make sense to merge the fix even if it still has issues?
>

Yes, Andrew has merged the fix into -mm. I see it in mmotm

http://userweb.kernel.org/~akpm/mmotm/broken-out/fbdev-defio-and-metronomefb-v4.patch

Thanks,
jaya

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-03-19 11:24           ` Jaya Kumar
@ 2008-03-19 18:37             ` Andrew Morton
  2008-03-19 22:30               ` Jaya Kumar
  0 siblings, 1 reply; 13+ messages in thread
From: Andrew Morton @ 2008-03-19 18:37 UTC (permalink / raw)
  To: Jaya Kumar
  Cc: Markus Armbruster, linux-kernel, virtualization, xen-devel,
	linux-fbdev-devel, adaplas, linux-input, dmitry.torokhov

On Wed, 19 Mar 2008 19:24:25 +0800 "Jaya Kumar" <jayakumar.lkml@gmail.com> wrote:

> On Wed, Mar 19, 2008 at 6:45 PM, Markus Armbruster <armbru@redhat.com> wrote:
> >
> >  Any progress on this?
> >
> >  Considering that fb_defio is utterly broken without the fix (writing
> >  the frame buffer makes the VM endlessly invoke vm_ops.page_mkwrite()),
> >  wouldn't it make sense to merge the fix even if it still has issues?
> >
> 
> Yes, Andrew has merged the fix into -mm. I see it in mmotm
> 
> http://userweb.kernel.org/~akpm/mmotm/broken-out/fbdev-defio-and-metronomefb-v4.patch
> 

Do we need some of all of that patch in 2.6.25?  I wasn't aware of such a
need.

If the drivers/video/fb_defio.c hunk of that patch fixes something then I'd
be inclined to merge the whole thing - adding a new driver won't hurt
anyone.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-03-19 18:37             ` Andrew Morton
@ 2008-03-19 22:30               ` Jaya Kumar
  2008-03-19 22:39                 ` Andrew Morton
  0 siblings, 1 reply; 13+ messages in thread
From: Jaya Kumar @ 2008-03-19 22:30 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Markus Armbruster, linux-kernel, virtualization, xen-devel,
	linux-fbdev-devel, adaplas, linux-input, dmitry.torokhov

On Thu, Mar 20, 2008 at 2:37 AM, Andrew Morton
<akpm@linux-foundation.org> wrote:
>
>  Do we need some of all of that patch in 2.6.25?  I wasn't aware of such a
>  need.
>
>  If the drivers/video/fb_defio.c hunk of that patch fixes something then I'd
>  be inclined to merge the whole thing - adding a new driver won't hurt
>  anyone.
>

Yes, the defio hunk is a bugfix. I agree with pushing the whole thing
to 2.6.25 if there are people who need it in 2.6.25. I believe there
is no risk of impact on anyone else.

Thanks,
jaya

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-03-19 22:30               ` Jaya Kumar
@ 2008-03-19 22:39                 ` Andrew Morton
  2008-03-19 22:55                   ` Jaya Kumar
  0 siblings, 1 reply; 13+ messages in thread
From: Andrew Morton @ 2008-03-19 22:39 UTC (permalink / raw)
  To: Jaya Kumar
  Cc: armbru, linux-kernel, virtualization, xen-devel,
	linux-fbdev-devel, adaplas, linux-input, dmitry.torokhov

On Thu, 20 Mar 2008 06:30:37 +0800
"Jaya Kumar" <jayakumar.lkml@gmail.com> wrote:

> On Thu, Mar 20, 2008 at 2:37 AM, Andrew Morton
> <akpm@linux-foundation.org> wrote:
> >
> >  Do we need some of all of that patch in 2.6.25?  I wasn't aware of such a
> >  need.
> >
> >  If the drivers/video/fb_defio.c hunk of that patch fixes something then I'd
> >  be inclined to merge the whole thing - adding a new driver won't hurt
> >  anyone.
> >
> 
> Yes, the defio hunk is a bugfix. I agree with pushing the whole thing
> to 2.6.25 if there are people who need it in 2.6.25. I believe there
> is no risk of impact on anyone else.
> 

Well all we have in the changelog is

  Implement support for the E-Ink Metronome controller.  It provides an
  mmapable interface to the controller using defio support.  It was tested
  with a gumstix pxa255 with Vizplex media using Xfbdev and various X
  clients such as xeyes, xpdf, xloadimage.

grump.  This say nothing at all about the change to drivers/video/fb_defio.c.

a) the changelog _should_ document this change.  Please send a paragraph and
   I'll add it.

b) If this discussion hadn't happened, that bug (whatever it is) would
   have remained unfixed in 2.6.25.  As a direct consequence of poor
   changelogging.

c) This bugfix (whatever it is) should have been in a separate patch.

Thanks.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [PATCH 2/3] fbdev: Make deferred I/O work as advertized
  2008-03-19 22:39                 ` Andrew Morton
@ 2008-03-19 22:55                   ` Jaya Kumar
  0 siblings, 0 replies; 13+ messages in thread
From: Jaya Kumar @ 2008-03-19 22:55 UTC (permalink / raw)
  To: Andrew Morton
  Cc: armbru, linux-kernel, virtualization, xen-devel,
	linux-fbdev-devel, adaplas, linux-input, dmitry.torokhov

On Thu, Mar 20, 2008 at 6:39 AM, Andrew Morton
<akpm@linux-foundation.org> wrote:
>
>  Well all we have in the changelog is
>
>   Implement support for the E-Ink Metronome controller.  It provides an
>   mmapable interface to the controller using defio support.  It was tested
>   with a gumstix pxa255 with Vizplex media using Xfbdev and various X
>   clients such as xeyes, xpdf, xloadimage.
>
>  grump.  This say nothing at all about the change to drivers/video/fb_defio.c.

I apologize. I should have done better.

>
>  a) the changelog _should_ document this change.  Please send a paragraph and
>    I'll add it.

This patch fixes the following bug. Defio would cause a hang on write
access to the framebuffer as the page fault would be called
ad-infinitum. It fixes fb_defio by setting the mapping to be used by
page_mkclean.

>
>  b) If this discussion hadn't happened, that bug (whatever it is) would
>    have remained unfixed in 2.6.25.  As a direct consequence of poor
>    changelogging.

Understood. The root cause of my flaw was my assumption that other
than me, there weren't any users of defio so I was in no hurry to get
it fixed.

>
>  c) This bugfix (whatever it is) should have been in a separate patch.

Understood. It won't happen again.

Thanks,
jaya

^ permalink raw reply	[flat|nested] 13+ messages in thread

end of thread, other threads:[~2008-03-20  0:03 UTC | newest]

Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-02-25 13:00 [PATCH 0/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer Markus Armbruster
2008-02-25 13:02 ` [PATCH 1/3] xen: Make xen-blkfront write its protocol ABI to xenstore Markus Armbruster
2008-02-25 13:03 ` [PATCH 2/3] fbdev: Make deferred I/O work as advertized Markus Armbruster
2008-02-26 14:11   ` Jaya Kumar
2008-02-26 14:45     ` Markus Armbruster
2008-02-26 15:01       ` Jaya Kumar
2008-03-19 10:45         ` Markus Armbruster
2008-03-19 11:24           ` Jaya Kumar
2008-03-19 18:37             ` Andrew Morton
2008-03-19 22:30               ` Jaya Kumar
2008-03-19 22:39                 ` Andrew Morton
2008-03-19 22:55                   ` Jaya Kumar
2008-02-25 13:06 ` [PATCH 3/3] xen pvfb: Para-virtual framebuffer, keyboard and pointer driver Markus Armbruster

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).