D600: S1 and S3 resume

D600: S1 and S3 resume

[Petr Olivka]

Hi !

  I have more question and informations about D600 with linux:

  I tested the patch for xfree radeon_driver.c, and now video wake up
after resume from S3.  But system is terrible slow. CPU is half second
working and half second is stopped. I have kernel 2.6-test10, bios A05.


  I have problem with resume from S1 and S3, when
uhci-hcd and sound card is loaded. When D600 resume from S1, display "back
to C!", "set pci latency 00:1d.0, 00:1d.1 and 00:1d.2 to 64" and hang.
When resume from S3 I have black LCD and do not see anything.
(I have to say, that 00:1d.0-1-2-7 are usb devices on pci.)


  What kernel is at this moment the best choice for D600? 2.4 or 2.6? What
you are using and your experiences with suspend?


  And back to patch for X. I debug the code and ax, bx, cx, dx items in
Int10 struct are zero. The "->num-0xe6" is little mystery for me.

  I did try call Int 10h using "lrmi" - linux real mode interface, but
it failed (SIGSEGV), when try outw to some port.


And last information: when I suspend to S3 in "Dock", after resume is
display ok, even I remove the notebook from dock before wakeup. But CPU is
slow.


I did ask dell about more technical informations about D600, but they
did say me, that all technical delatils are marked internally
"confidential" and nobody can share them. I hope Dell will change your
restriction and improve support for linux.

poli








-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Re: D600: S1 and S3 resume

[Emmanuel Thomé]

>   I tested the patch for xfree radeon_driver.c, and now video wake up
> after resume from S3.  But system is terrible slow. CPU is half second
> working and half second is stopped. I have kernel 2.6-test10, bios A05.
 
>   I have problem with resume from S1 and S3, when uhci-hcd and sound
>   card is loaded. When D600 resume from S1, display "back to C!", "set
>   pci latency 00:1d.0, 00:1d.1 and 00:1d.2 to 64" and hang.  When
>   resume from S3 I have black LCD and do not see anything.  (I have to
>   say, that 00:1d.0-1-2-7 are usb devices on pci.)
 
I now have a patch for this (it's one single problem : your cpu is very
busy on irq11 dealing w/ usb panic messages. Look up your syslog, you'll
get an idea). You'll have to wait a few jiffees for the patch though, as
I'm on the road right now.

The patch in question is more exactly a kludge than a real cure of the
problem. It seems that on S3 resume, the ehci hcd is fine, but two of
the three uhci hcd's wake up in a messy state. Disabling them when such
a messy state is detected does the trick (hunt for USBCMD_HCRESET in
uhci-hcd.c, IIRC).

 
>   What kernel is at this moment the best choice for D600? 2.4 or 2.6? What
> you are using and your experiences with suspend?

2.6, definitely. Unless you're happy with S1.

A short data: My laptop draws ~16W when running, and this goes down to
0.4W on S3.

 
>   And back to patch for X. I debug the code and ax, bx, cx, dx items in
> Int10 struct are zero. The "->num-0xe6" is little mystery for me.


Err, ole ? (I bloody don't know the answer. The last time I did real
mode vesa programming was a looooong time ago).


>   I did try call Int 10h using "lrmi" - linux real mode interface, but
> it failed (SIGSEGV), when try outw to some port.

You probably have to resort to a couple of hacks in order to have your
real mode thingy (a virtual machine, I presume) access IO ports.

> And last information: when I suspend to S3 in "Dock", after resume is
> display ok, even I remove the notebook from dock before wakeup. But CPU is
> slow.

See above.

> I did ask dell about more technical informations about D600, but they
> did say me, that all technical delatils are marked internally
> "confidential" and nobody can share them. I hope Dell will change your
> restriction and improve support for linux.

Very expectable from canned_answers_daemon-QXIuMnnTXwEAsOHDdpmTTuG/Ez6ZCGd0@public.gmane.org

E.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Re: D600: S1 and S3 resume

[ole.rohne-vJEk5272eHo]

Emmanuel Thomé <Emmanuel.Thome-/zGXu1G9BXs@public.gmane.org> writes:

> >   And back to patch for X. I debug the code and ax, bx, cx, dx items in
> > Int10 struct are zero. The "->num-0xe6" is little mystery for me.
> 
> Err, ole ? (I bloody don't know the answer. The last time I did real
> mode vesa programming was a looooong time ago).

If the "0xe6" is the question, I don't really remember where it came
from; I might have done something that I don't really want to
remember. Disa**embling BIOS comes to mind as something not very
pretty.






-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Re: D600: S1 and S3 resume

[Petr Olivka]

> A short data: My laptop draws ~16W when running, and this goes down to
> 0.4W on S3.
>
hm, I am not measure yet, in S1 I have about 7-8W. S3 with 0.4W, it is
maybe four days sleeping, it is betr than I expect.

> > And last information: when I suspend to S3 in "Dock", after resume is
> > display ok, even I remove the notebook from dock before wakeup. But CPU is
> > slow.
>
> See above.
yes, but I want to say, that D600 with Dock  resume from S3 without any
video patch.


> Very expectable from canned_answers_daemon-QXIuMnnTXwEAsOHDdpmTTuG/Ez6ZCGd0@public.gmane.org
>
yes, it is true {:-). But dell promise me, when I will want buy more then
100 piecies of D600, I can require technical documentation {:-)

poli





-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Re: D600: S1 and S3 resume

[Emmanuel Thomé]

[-- Attachment #1: Type: text/plain, Size: 2555 bytes --]


Regarding your usb problem, you might try the silly patch attached.
First make sure in any case that you shut down the usb subsystem before
suspending ! The patch does nothing intelligent, but it helps. On _my_
dell d600, running 2.6.0-test11, this prevents the uhci controllers #2
and #3 from waking up, as they are bogus on wakeup (and put unnecessary
pressure on syslog). Having only one uhci controller active means that
only the bottom usb plug is functional on S3 resume.

Patch applies to 2.6.0-test9, test10, test11 afaik.

Now note that there seems to be power management patches floating around
in linux-usb-devel. You might want to try some of them. There has to be
a proper way of resuming these usb controllers.

If you have some sort of better success with usb, do tell me. I'm
currently trying to make something decent out of the proprietary
winmodem driver from linuxant (not sure I'll be succesful: it's such a
messy crap...). Currently, the winmodem causes a lot of instability with
S3.

Regards,

E.

On Tue, Nov 25, 2003 at 01:27:01PM +0100, Petr Olivka wrote:
> Hi !
> 
>   I have more question and informations about D600 with linux:
> 
>   I tested the patch for xfree radeon_driver.c, and now video wake up
> after resume from S3.  But system is terrible slow. CPU is half second
> working and half second is stopped. I have kernel 2.6-test10, bios A05.
> 
> 
>   I have problem with resume from S1 and S3, when
> uhci-hcd and sound card is loaded. When D600 resume from S1, display "back
> to C!", "set pci latency 00:1d.0, 00:1d.1 and 00:1d.2 to 64" and hang.
> When resume from S3 I have black LCD and do not see anything.
> (I have to say, that 00:1d.0-1-2-7 are usb devices on pci.)
> 
> 
>   What kernel is at this moment the best choice for D600? 2.4 or 2.6? What
> you are using and your experiences with suspend?
> 
> 
>   And back to patch for X. I debug the code and ax, bx, cx, dx items in
> Int10 struct are zero. The "->num-0xe6" is little mystery for me.
> 
>   I did try call Int 10h using "lrmi" - linux real mode interface, but
> it failed (SIGSEGV), when try outw to some port.
> 
> 
> And last information: when I suspend to S3 in "Dock", after resume is
> display ok, even I remove the notebook from dock before wakeup. But CPU is
> slow.
> 
> 
> I did ask dell about more technical informations about D600, but they
> did say me, that all technical delatils are marked internally
> "confidential" and nobody can share them. I hope Dell will change your
> restriction and improve support for linux.
> 
> poli

[-- Attachment #2: usb.patch_manu --]
[-- Type: text/plain, Size: 980 bytes --]

--- linux-2.6.x-custom/drivers/usb/host/uhci-hcd.c.orig	2003-12-01 19:27:55.473667736 +0100
+++ linux-2.6.x-custom/drivers/usb/host/uhci-hcd.c	2003-12-01 19:30:05.675873992 +0100
@@ -2128,7 +2128,7 @@
 	}
 }
 
-static void start_hc(struct uhci_hcd *uhci)
+static int start_hc(struct uhci_hcd *uhci)
 {
 	unsigned int io_addr = uhci->io_addr;
 	int timeout = 1000;
@@ -2143,7 +2143,7 @@
 	while (inw(io_addr + USBCMD) & USBCMD_HCRESET) {
 		if (!--timeout) {
 			printk(KERN_ERR "uhci: USBCMD_HCRESET timed out!\n");
-			break;
+			return -ETIMEDOUT;
 		}
 	}
 
@@ -2161,6 +2161,8 @@
 	outw(USBCMD_RS | USBCMD_CF | USBCMD_MAXP, io_addr + USBCMD);
 
         uhci->hcd.state = USB_STATE_RUNNING;
+
+	return 0;
 }
 
 /*
@@ -2414,7 +2416,11 @@
 		uhci->fl->frame[i] = cpu_to_le32(uhci->skelqh[7 - irq]->dma_handle);
 	}
 
-	start_hc(uhci);
+	if (start_hc(uhci) == -ETIMEDOUT) {
+		init_stall_timer(hcd);
+		retval = -ENODEV;
+		goto err_alloc_skelqh;
+	}
 
 	init_stall_timer(hcd);
 

Re: D600: S1 and S3 resume

[Petr Olivka]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: TEXT/PLAIN; charset=X-UNKNOWN, Size: 4214 bytes --]


I did test this patch yesterday, and I have still the same behaviour.
After resume from S1 notebook hang. To not hang I everytime stop usbmbr
and remove all usb modules, but uhci-hcd fail when unloading:

uhci_hcd 0000:00:1d.0: remove, state 1
usb usb1: USB disconnect, address 1
uhci_hcd 0000:00:1d.0: USB bus 1 deregistered
uhci_hcd 0000:00:1d.1: remove, state 1
usb usb2: USB disconnect, address 1
uhci_hcd 0000:00:1d.1: USB bus 2 deregistered
uhci_hcd 0000:00:1d.2: remove, state 1
usb usb3: USB disconnect, address 1
usb 3-2: USB disconnect, address 2
pci_pool_destroy 0000:00:1d.2/uhci_td, def25000 busy
uhci_hcd 0000:00:1d.2: USB bus 3 deregistered
slab error in kmem_cache_destroy(): cache `uhci_urb_priv': Can't free all
objects
Call Trace:
 [<c013866d>] __slab_error+0x21/0x28
 [<c0138fe4>] kmem_cache_destroy+0x8c/0x114
 [<e28e86d0>] uhci_hcd_cleanup+0x18/0x50 [uhci_hcd]
 [<c012e54f>] sys_delete_module+0x17b/0x19c
 [<c0141510>] sys_munmap+0x44/0x64
 [<c0108faf>] syscall_call+0x7/0xb

uhci: not all urb_priv's were freed

I will forward this problem to usb-devel.

thanks
  poli


On Wed, 3 Dec 2003, Emmanuel [iso-8859-1] Thomé wrote:

>
> Regarding your usb problem, you might try the silly patch attached.
> First make sure in any case that you shut down the usb subsystem before
> suspending ! The patch does nothing intelligent, but it helps. On _my_
> dell d600, running 2.6.0-test11, this prevents the uhci controllers #2
> and #3 from waking up, as they are bogus on wakeup (and put unnecessary
> pressure on syslog). Having only one uhci controller active means that
> only the bottom usb plug is functional on S3 resume.
>
> Patch applies to 2.6.0-test9, test10, test11 afaik.
>
> Now note that there seems to be power management patches floating around
> in linux-usb-devel. You might want to try some of them. There has to be
> a proper way of resuming these usb controllers.
>
> If you have some sort of better success with usb, do tell me. I'm
> currently trying to make something decent out of the proprietary
> winmodem driver from linuxant (not sure I'll be succesful: it's such a
> messy crap...). Currently, the winmodem causes a lot of instability with
> S3.
>
> Regards,
>
> E.
>
> On Tue, Nov 25, 2003 at 01:27:01PM +0100, Petr Olivka wrote:
> > Hi !
> >
> >   I have more question and informations about D600 with linux:
> >
> >   I tested the patch for xfree radeon_driver.c, and now video wake up
> > after resume from S3.  But system is terrible slow. CPU is half second
> > working and half second is stopped. I have kernel 2.6-test10, bios A05.
> >
> >
> >   I have problem with resume from S1 and S3, when
> > uhci-hcd and sound card is loaded. When D600 resume from S1, display "back
> > to C!", "set pci latency 00:1d.0, 00:1d.1 and 00:1d.2 to 64" and hang.
> > When resume from S3 I have black LCD and do not see anything.
> > (I have to say, that 00:1d.0-1-2-7 are usb devices on pci.)
> >
> >
> >   What kernel is at this moment the best choice for D600? 2.4 or 2.6? What
> > you are using and your experiences with suspend?
> >
> >
> >   And back to patch for X. I debug the code and ax, bx, cx, dx items in
> > Int10 struct are zero. The "->num-0xe6" is little mystery for me.
> >
> >   I did try call Int 10h using "lrmi" - linux real mode interface, but
> > it failed (SIGSEGV), when try outw to some port.
> >
> >
> > And last information: when I suspend to S3 in "Dock", after resume is
> > display ok, even I remove the notebook from dock before wakeup. But CPU is
> > slow.
> >
> >
> > I did ask dell about more technical informations about D600, but they
> > did say me, that all technical delatils are marked internally
> > "confidential" and nobody can share them. I hope Dell will change your
> > restriction and improve support for linux.
> >
> > poli
>



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Re: D600: S1 and S3 resume

[Petr Olivka]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: TEXT/PLAIN; charset=X-UNKNOWN, Size: 2464 bytes --]

Sorry my late. I was decompressing and decoding DELL BIOS image
to obtain asm source of smm bios. Now I try to understand it. It is
terrible job.

In attach is my lspci and standby script.
Now I am using "stanby" script with rmmod, in attached script is modules
removing commented out.

poli


On Mon, 8 Dec 2003, Emmanuel [iso-8859-1] Thomé wrote:

>
> This seems pretty odd ; we should have very similar hardware, if not
> identical. Would you mind sending me your lspci -vn ? I'd be interested
> also in having a look to your suspend script.
>
> Note that the only situation in which I've had an oops triggered from
> the usb subsystem was with the original kernel setup: re-enabling all
> usb controllers on resume, regardless of timeouts. My patch explicitely
> fails on controller timeout, which avoids this sort of nasty problems
> (oopsing from rmmod).
>
> Also, do you have any usb device actually plugged into your system when
> suspending ?
>
> You're right poking usb-devel about this sort of woes. There definitely
> is some work to do in order to have usb behave better wrt suspend
> issues.
>
> E.
>
> On Fri, Dec 05, 2003 at 11:05:13AM +0100, Petr Olivka wrote:
> >
> > I did test this patch yesterday, and I have still the same behaviour.
> > After resume from S1 notebook hang. To not hang I everytime stop usbmbr
> > and remove all usb modules, but uhci-hcd fail when unloading:
> >
> > uhci_hcd 0000:00:1d.0: remove, state 1
> > usb usb1: USB disconnect, address 1
> > uhci_hcd 0000:00:1d.0: USB bus 1 deregistered
> > uhci_hcd 0000:00:1d.1: remove, state 1
> > usb usb2: USB disconnect, address 1
> > uhci_hcd 0000:00:1d.1: USB bus 2 deregistered
> > uhci_hcd 0000:00:1d.2: remove, state 1
> > usb usb3: USB disconnect, address 1
> > usb 3-2: USB disconnect, address 2
> > pci_pool_destroy 0000:00:1d.2/uhci_td, def25000 busy
> > uhci_hcd 0000:00:1d.2: USB bus 3 deregistered
> > slab error in kmem_cache_destroy(): cache `uhci_urb_priv': Can't free all
> > objects
> > Call Trace:
> >  [<c013866d>] __slab_error+0x21/0x28
> >  [<c0138fe4>] kmem_cache_destroy+0x8c/0x114
> >  [<e28e86d0>] uhci_hcd_cleanup+0x18/0x50 [uhci_hcd]
> >  [<c012e54f>] sys_delete_module+0x17b/0x19c
> >  [<c0141510>] sys_munmap+0x44/0x64
> >  [<c0108faf>] syscall_call+0x7/0xb
> >
> > uhci: not all urb_priv's were freed
> >
> > I will forward this problem to usb-devel.
> >
> > thanks
> >   poli
>
>

[-- Attachment #2: Type: TEXT/PLAIN, Size: 4154 bytes --]

00:00.0 Class 0600: 8086:3340 (rev 03)
	Flags: bus master, fast devsel, latency 0
	Memory at e0000000 (32-bit, prefetchable) [size=128M]
	Capabilities: [e4] #09 [f104]
	Capabilities: [a0] AGP version 2.0

00:01.0 Class 0604: 8086:3341 (rev 03)
	Flags: bus master, 66Mhz, fast devsel, latency 32
	Bus: primary=00, secondary=01, subordinate=01, sec-latency=32
	I/O behind bridge: 0000c000-0000cfff
	Memory behind bridge: fc000000-fdffffff
	Prefetchable memory behind bridge: e8000000-efffffff

00:1d.0 Class 0c03: 8086:24c2 (rev 01)
	Subsystem: 8086:4541
	Flags: bus master, medium devsel, latency 0, IRQ 11
	I/O ports at bf80 [size=32]

00:1d.1 Class 0c03: 8086:24c4 (rev 01)
	Subsystem: 8086:4541
	Flags: bus master, medium devsel, latency 0, IRQ 11
	I/O ports at bf40 [size=32]

00:1d.2 Class 0c03: 8086:24c7 (rev 01)
	Subsystem: 8086:4541
	Flags: bus master, medium devsel, latency 0, IRQ 11
	I/O ports at bf20 [size=32]

00:1d.7 Class 0c03: 8086:24cd (rev 01) (prog-if 20)
	Subsystem: 1028:011d
	Flags: bus master, medium devsel, latency 0, IRQ 11
	Memory at f4fffc00 (32-bit, non-prefetchable) [size=1K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] #0a [2080]

00:1e.0 Class 0604: 8086:2448 (rev 81)
	Flags: bus master, fast devsel, latency 0
	Bus: primary=00, secondary=02, subordinate=02, sec-latency=32
	I/O behind bridge: 0000d000-0000efff
	Memory behind bridge: f6000000-fbffffff

00:1f.0 Class 0601: 8086:24cc (rev 01)
	Flags: bus master, medium devsel, latency 0

00:1f.1 Class 0101: 8086:24ca (rev 01) (prog-if 8a [Master SecP PriP])
	Subsystem: 8086:4541
	Flags: bus master, medium devsel, latency 0, IRQ 11
	I/O ports at <ignored>
	I/O ports at <ignored>
	I/O ports at <ignored>
	I/O ports at <ignored>
	I/O ports at bfa0 [size=16]
	Memory at 20000000 (32-bit, non-prefetchable) [size=1K]

00:1f.5 Class 0401: 8086:24c5 (rev 01)
	Subsystem: 1028:011d
	Flags: bus master, medium devsel, latency 0, IRQ 11
	I/O ports at b800 [size=256]
	I/O ports at bc40 [size=64]
	Memory at f4fff800 (32-bit, non-prefetchable) [size=512]
	Memory at f4fff400 (32-bit, non-prefetchable) [size=256]
	Capabilities: [50] Power Management version 2

00:1f.6 Class 0703: 8086:24c6 (rev 01)
	Subsystem: 14f1:5422
	Flags: medium devsel, IRQ 11
	I/O ports at b400 [size=256]
	I/O ports at b080 [size=128]
	Capabilities: [50] Power Management version 2

01:00.0 Class 0300: 1002:4c66 (rev 01)
	Subsystem: 1028:011d
	Flags: bus master, VGA palette snoop, stepping, 66Mhz, medium devsel, latency 32, IRQ 11
	Memory at e8000000 (32-bit, prefetchable) [size=128M]
	I/O ports at c000 [size=256]
	Memory at fcff0000 (32-bit, non-prefetchable) [size=64K]
	Expansion ROM at <unassigned> [disabled] [size=128K]
	Capabilities: [58] AGP version 2.0
	Capabilities: [50] Power Management version 2

02:00.0 Class 0200: 14e4:16a6 (rev 02)
	Subsystem: 1028:8126
	Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 11
	Memory at faff0000 (64-bit, non-prefetchable) [size=64K]
	Capabilities: [40] #07 [0000]
	Capabilities: [48] Power Management version 2
	Capabilities: [50] Vital Product Data
	Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/3 Enable-

02:01.0 Class 0607: 1217:7113 (rev 20)
	Subsystem: 1028:011d
	Flags: bus master, stepping, slow devsel, latency 168, IRQ 11
	Memory at 20001000 (32-bit, non-prefetchable) [size=4K]
	Bus: primary=02, secondary=03, subordinate=06, sec-latency=176
	Memory window 0: 20400000-207ff000 (prefetchable)
	Memory window 1: 20800000-20bff000
	I/O window 0: 00004000-000040ff
	I/O window 1: 00004400-000044ff
	16-bit legacy interface ports at 0001

02:01.1 Class 0607: 1217:7113 (rev 20)
	Subsystem: 1028:011d
	Flags: bus master, stepping, slow devsel, latency 168, IRQ 11
	Memory at 20002000 (32-bit, non-prefetchable) [size=4K]
	Bus: primary=02, secondary=07, subordinate=0a, sec-latency=176
	Memory window 0: 20c00000-20fff000 (prefetchable)
	Memory window 1: 21000000-213ff000
	I/O window 0: 00004800-000048ff
	I/O window 1: 00004c00-00004cff
	16-bit legacy interface ports at 0001


[-- Attachment #3: Type: TEXT/PLAIN, Size: 568 bytes --]

#!/bin/sh

. /etc/profile 

USBMGR=/etc/init.d/usbmgr
AUMIX=/etc/init.d/aumix
PCMCIA=/etc/init.d/pcmcia

LCD="/root/radeon/radeontool light"

if [ -f /tmp/.X0-lock ]
  then
    LCD="xset -display :0.0 dpms force"
fi


$USBMGR stop
$PCMCIA stop
$AUMIX stop
ifdown eth0 

#for i in `lsmod | awk '{print $1;}'`; do rmmod $i; sleep 0.1; done

sync
$LCD off
hdparm -y /dev/hda

echo 1 > /proc/acpi/sleep
#echo -n "standby" > /sys/power/state

$LCD on

/etc/init.d/modutils start

ifup eth0

$USBMGR start
$PCMCIA start
$AUMIX start

RE: D600: S1 and S3 resume

[Yu, Luming]

>Sorry my late. I was decompressing and decoding DELL BIOS image
>to obtain asm source of smm bios. Now I try to understand it. It is
>terrible job.

If you think of it as ACPI bug, please provide information.



-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op=click