RE: ntdev digest: March 20, 2002 Are nt4 video drivers compatable with win 2k ?

OSR_Community_User · March 21, 2002, 5:02am

Hi all,

Can anyone confirm if nt4 video drivers can be installed under win 2k and if
so what is the best way to install them ?

Robert Fernando
Anite Telecoms Ltd
110 Fleet Road
Fleet
Hampshire GU51 4BL
United Kingdom
Tel: +44 (0) 1252 775200
Fax: +44 (0) 1252 775 321
Email: xxxxx@anitetelecoms.com

Anite Telecoms Ltd, Registered in England No. 1721900 Registered Office:
100 Longwater Avenue, GreenPark, Reading, Berkshire RG2 6GP, United Kingdom

-----Original Message-----
From: NT Developers Interest List digest [mailto:xxxxx@lists.osr.com]
Sent: 21 March 2002 05:00
To: ntdev digest recipients
Subject: ntdev digest: March 20, 2002

NTDEV Digest for Wednesday, March 20, 2002.

RE: I am trying to intereact with the FireWire bridge
Re: USB power management
RE: XP SMP
Re: IM driver problem
Re: NDIS Binary-Compatibility on Windows 98 and Windows 2000 ?
Re: Please help me, here is my code
Re: Problems with Packet Driver
Looking for a driver for diamond multimedia Stealth video card 25
00 revision 1.03 to use it under win 2k sp2
Windows 2000 COM port driver blocking
RE: Windows 2000 COM port driver blocking
Power management in Windows ME
Net problems…
Re: Problems with Packet Driver
RE: Windows 2000 COM port driver blocking
Re: Problems with Packet Driver
Re: Looking for a driver for diamond multimedia Stealth video card 25
00 revision 1.03 to use it under win 2k sp2
local address in TDIEventConnect
Re: Problems with Packet Driver
VideoPortMapMemory call problem under WinXp
RE: XP SMP
RE: Windows 2000 COM port driver blocking
Checked build of SNMP?
RE: XP SMP
RE: Net problems…
Re: Problems with Packet Driver
Re: BUILD SOURCES from another directories
Re: Problems with Packet Driver

Subject: RE: I am trying to intereact with the FireWire bridge
From: “Udo Eberhardt”
Date: Wed, 20 Mar 2002 08:47:42 +0100
X-Message-Number: 1

This is a multi-part message in MIME format.

------=_NextPart_000_0011_01C1CFEB.E62408F0
Content-Type: text/plain;
charset=“US-ASCII”
Content-Transfer-Encoding: 7bit

In order to interact with a custom hardware you have to develop a WDM device
driver that talks to the 1394 bus driver (1394bus.sys) at its lower edge and
provides a custom API (IOCTLs, read, write) at its upper edge. You could use
the 1394diag sample driver that is included in the DDK. However, the
sample’s implementation is ugly. It is useful for simple testing purposes
only. It provides very bad performance because it behaves synchronously.
Alternatively, you could take a look at our product VHPD1394, see
http://www.thesycon.de/vhpd1394/eng/vhpd.htm

–
Udo Eberhardt
Thesycon GmbH, Germany
xxxxx@thesycon.de
www.thesycon.de

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of tarang garg
Sent: Tuesday, March 19, 2002 8:14 PM
To: NT Developers Interest List
Subject: [ntdev] I am trying to intereact with the FireWire bridge

I am trying to interact with the custom hardware connected to the FireWire
bridge through a FireWire PC Card in a PC running Win 2K. I am confused in
the sense do I really need a WDM driver on top of the Sbp2Port.sys to
communicate with the hardware or can I use certain IOCTL commands to bypass
the bridge and directly send commands to the custom hardware.

Can any one help me with this problem.

Thanks in advance.

Tarang

----------------------------------------------------------------------------
–
Do You Yahoo!?
Yahoo! Sports - live college hoops coverage — You are currently
subscribed to ntdev as: xxxxx@thesycon.de To unsubscribe send a blank email
to %%email.unsub%%

------=_NextPart_000_0011_01C1CFEB.E62408F0
Content-Type: text/html;
charset=“US-ASCII”
Content-Transfer-Encoding: quoted-printable

charset=3Dus-ascii">

size=3D2>In=20
order to interact with a custom hardware you have to develop a WDM =
device=20
driver that talks to the 1394 bus driver (1394bus.sys) at its lower edge =
and=20
provides a custom API (IOCTLs, read, write) at its upper edge. You could =
use the=20
1394diag sample driver that is included in the DDK. However, the =
sample’s=20
implementation is ugly. It is useful for simple testing purposes =
only.=20
It provides very bad performance because it behaves=20
synchronously.

size=3D2>Alternatively, you could take a look at our product VHPD1394, =
see href=3D"http://www.thesycon.de/vhpd1394/eng/vhpd.htm">http://www.thesycon=
.de/vhpd1394/eng/vhpd.htm

size=3D2>

–
Udo Eberhardt
Thesycon GmbH,=20
Germany
xxxxx@thesycon.de
www.thesycon.de

PAN>

style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px =
solid">

face=3DTahoma=20
size=3D2>-----Original Message-----
From:=20
xxxxx@lists.osr.com =
[mailto:xxxxx@lists.osr.com]On=20
Behalf Of tarang garg
Sent: Tuesday, March 19, 2002 8:14 =

PM
To: NT Developers Interest List
Subject: =
[ntdev] I am=20
trying to intereact with the FireWire bridge

I am trying to interact with the custom hardware connected to the =
FireWire=20
bridge through a FireWire PC Card in a PC running Win 2K. I am =
confused in the=20
sense do I really need a WDM driver on top of the Sbp2Port.sys to =
communicate=20
with the hardware or can I use certain IOCTL commands to bypass the =
bridge and=20
directly send commands to the custom hardware.

Can any one help me with this problem.

Thanks in advance.

Tarang

Do You Yahoo!?
href=3D"$rd_url/tag/http://sports.yahoo.com/">Yahoo!=20
Sports - live college hoops coverage --- You are currently =
subscribed to=20
ntdev as: xxxxx@thesycon.de To unsubscribe send a blank email to=20
%%email.unsub%%

------=_NextPart_000_0011_01C1CFEB.E62408F0--

----------------------------------------------------------------------

Subject: Re: USB power management
From: Corand van Strien
Date: Wed, 20 Mar 2002 10:28:02 +0100
X-Message-Number: 2

> Subject: Re: USB power management
> From: Mathias Ellinger
> Date: Tue, 19 Mar 2002 13:35:26 +0100
>
> Hello,
>
> >I am use the DriverWorks package of Numega to write a USB
> driver for a
> >Braille output device. The driver basically works but I have
> a problem with
> >some chipsets (Ali) under Windows XP. The device does not
> resume when the PC
> >is resuming from suspend.
> >From several sources I have determined that I have to
> implement some power
> >management functionality for suspension and resume. The
> default handling of
> >DriverWorks seems not te be enough.
>
> I mean this is enough. For devices with no power I have made my
> special hack and in OnPowerUp() I reset my USB devices with
> IOCTL_INTERNAL_USB_CYCLE_PORT.
>
> elli
>

Hello Mathias,
Thanks for your reply. Maybe I should have stated that I use a failry old,
acc. to Numega, version of DriverWorks (2.1). It appears that there is
simply no power management in this version of the library. So there is also
no OnPowerUp().
From where do you send the IOCTL_INTERNAL_USB_CYCLE_PORT, and where to?

Please include my e-mail address in your reply too because I receive the
digest version of this list.

Regards,
Corand

----------------------------------------------------------------------

Subject: RE: XP SMP
From: "Whitman, Steve"
Date: Wed, 20 Mar 2002 07:05:36 -0500
X-Message-Number: 3

I have very similar results on my dual CPU machine. However, if I bring up
task manager it displays two CPU graphs and when I bring up device manager
and select Processors it lists both processors.

But Windbg reports MP kernel with 1 proc and system properties just says
Pentium III, etc.

I'm sure that my system is using both processors. I believe that system
properties is behaving as it should (it reports the processor and since they
are both the same it doesn't need to list both). I don't know why Windbg
reports 1 processor.

- Steve -

> -----Original Message-----
> From: Jim Young [mailto:xxxxx@youngendeavors.com]
> Sent: Tuesday, March 19, 2002 10:51 PM
> To: NT Developers Interest List
> Subject: [ntdev] RE: XP SMP
>
>
> How many processors does the BIOS detect? If it only shows 1
> then I would
> suspect a defect in the CPU, socket, or motherboard.
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Hassan I Khan
> Sent: Tuesday, March 19, 2002 8:36 PM
> To: NT Developers Interest List
> Subject: [ntdev] XP SMP
>
>
> I recently assembled a dual-processor system for kernel
> debugging. Installed
> XP prof (checked version) came with msdn subscription (the
> very first XP
> CDs). During installation I didn't see any message or option
> that XP is
> going to run on 2 processors. System properties just says
> Pentium 1.0 GHz
> etc. Kernel debugger reports an MP kernel with (1 procs) etc.
>
> Whats going on here ? Has anyone done a dual-processor XP
> installation ? How
> do you do it ?
>
> Apologies for this question being slightly off-topic on this list.
>
> thanks for any comments,
> Hassan
>
> _________________________________________________________________
> Get your FREE download of MSN Explorer at
http://explorer.msn.com/intl.asp.

---
You are currently subscribed to ntdev as: xxxxx@youngendeavors.com
To unsubscribe send a blank email to %%email.unsub%%

---
You are currently subscribed to ntdev as: xxxxx@cognex.com
To unsubscribe send a blank email to %%email.unsub%%

----------------------------------------------------------------------

Subject: Re: IM driver problem
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 12:16:57 GMT
X-Message-Number: 4

I think you already know the answer because you said it yourself:
"Also i could not re-open the key since the
WrapperConfigurationContext is having the life time limited to
MiniportInitialize entry point."

Exactly! So what this means is you cannot "re-open" the miniport's
configuration after MiniportInitialize() has returned.

The PnP reconfiguration handler you mentioned is ProtocolPnPEvent()
with the 'NetPnPEvent' being 'NetEventReconfigure'. This is a
*protocol* function, not a miniport one.

Thus, if you need to re-read the miniport configuration, either:

1. stop and re-start the miniport instance using
NdisIMDeInitializeDeviceInstance() and
NdisIMInitializeDeviceInstance[Ex]().

2. or use NdisOpenProtocolConfiguration(). Note, however, that you
need to know the registry path to pass as 'ProtocolSection'. By
default, NdisOpenProtocolConfiguration() opens a subpath below
"HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services". Specify an
absolute path if you need to access any different location.

Stephan
---
On Tue, 19 Mar 2002 23:04:34 -0500, "Shashidhar"
wrote:

>
>Hi Stephan,
>I checked by going through the code and stepping through it and saw that
>all the keys are properly closed. And the configuration key i got during
>the MiniportInitialize using the function NdisOpenConfiguration() using
>WrapperConfigurationContext parameter passed to MiniportInitialize entry
>point. Also i could not re-open the key since the
>WrapperConfigurationContext is having the life time limited to
>MiniportInitialize entry point. Any suggestions are welcome.
>Thanks/Regards,
>Shashi

----------------------------------------------------------------------

Subject: Re: NDIS Binary-Compatibility on Windows 98 and Windows 2000 ?
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 12:25:09 GMT
X-Message-Number: 5

For NDIS drivers, do not specify BINARY_COMPATIBLE directly. Instead,
define NDIS_MINIPORT_DRIVER in your "sources" file.

Depending on the NDIS version your driver is built for, also define
one of

NDIS50_MINIPORT
NDIS40_MINIPORT

If none defined, NDIS 3.0 is assumed.

Note: W2K and W98 both support NDIS 5.0.

Stephan
---
On Mon, 18 Mar 2002 11:28:45 -0500, "Yee" wrote:

>
>Hi, people
> I write a miniport driver based on Win2K DDK,
>and it works well under Win2k. but It couldn't be
>loaded under Win98se though I define "BINARY_COMPATIBLE"
>in my driver.
> Is there anyone give me some suggestion ?
>Thanks.
>Yee

----------------------------------------------------------------------

Subject: Re: Please help me, here is my code
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 12:27:45 GMT
X-Message-Number: 6

If you change to ASCII instead of "Content-Transfer-Encoding: base64"
I might be able to read and answer your NDIS questions.

Stephan
---
On Tue, 19 Mar 2002 21:50:32 +0800, brucie wrote:

>dGhlIEJ1ZmZlckNvdW50IGFsd2F5IG5vdCBlcXVhbCB0byAxLCBCdXQgSSBjb3B5IGFsbCB0aGU
g
>ZGF0YSBmcm9tIHRoZSBvcmlnaW4gDQpwYWNrZXQgdG8gbXkgbmV3IHBhY2tldCB1c2luZyBOZGl
z
>Q29weWZyb21wYWNrZXR0b3BhY2tldC4NCg0KPkluIE5ld1RDUFNlbmQsIGlzIG5CdWZmZXJDb3V
u
>dCBhbHdheXMgPSAxPyBJIGhhdmVuJ3Qgd29ya2VkIHdpdGggdGhlIHRjcGlwIGhvb2tzLCBidXQ
g
>SSBkbyB3b3JrDQo+d2l0aCBhbiBORElTIElNIGRyaXZlci4gUGFja2V0cyB0cmF2ZWxsaW5nIGR
v
>d24gdGhlIHN0YWNrIG1vc3QgbGlrZWx5IGFyZSBzZWdtZW50ZWQgaW50byBtdWx0aXBsZQ0KPm5
k
[snip - another 150 lines]

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 12:35:10 GMT
X-Message-Number: 7

AFAIK, there is no VLAN support in NT4, ie. NDIS4. Thus, if the NIC
and its driver supports VLAN, it is probably some proprietary
solution.

Does your driver support VLAN tagging? Or is it just the NIC driver?
In the latter case, go and ask the NIC vendor (which one is it?).

BTW, what is 803.1q??? => 802.1q!

Stephan
---
On Mon, 18 Mar 2002 18:35:19 -0700, "Satrio Mualim"
wrote:

>I am looking for some suggestion/help in solving some packet driver
problems.
>
>We have a driver, roughly based on the NT Packet driver running on NT 4
SP6.
>It has been performing OK with no particular problems until recently we
turn
>on the 803.1q VLAN tagging on the NIC.
>
>Now it pauses after sending 3 packets. Looking into this problem I found
out
>that
>after each third packet my Protocol Characteristic StatusHandler and
>StatusCompleteHandler
>callbacks are executed with status NDIS_STATUS_RESET_START and
>NDIS_STATUS_RESET_END.
>
>Is there anything that has to be done differently/added when VLAN tagging
is
>enabled?
>Looking at the packets "on the wire" it seems that they are going out at
the
>normal speed until
>the third packet is out, after which there is a pause for about 4 to 6
seconds.
>The system has 2 NICs and if I turn off tagging on one of the NIC's,
packets are
>still going at the
>old fast speed on the NIC with no tagging, and slow as mentioned above for
the
>NIC with tagging.
>
>Any suggestions on where I should look to solve this problem?
>
>Thanks,
>
>Satrio

----------------------------------------------------------------------

Subject: Looking for a driver for diamond multimedia Stealth video card 25
00 revision 1.03 to use it under win 2k sp2
From: "Fernando, Robert"
Date: Wed, 20 Mar 2002 12:58:42 -0000
X-Message-Number: 8

Hi all,

Please could anyone tell me if there is a driver available for the following
card.
Diamond multimedia Stealth video 2500 revision 1.03

The chip set used on the card is Alliance Semiconductor promotion AT24

I have managed to find one for nt4 can this be used with w2k ?

Robert Fernando
Anite Telecoms Ltd
110 Fleet Road
Fleet
Hampshire GU51 4BL
United Kingdom
Tel: +44 (0) 1252 775200
Fax: +44 (0) 1252 775 321
Email: xxxxx@anitetelecoms.com

Anite Telecoms Ltd, Registered in England No. 1721900 Registered Office:
100 Longwater Avenue, GreenPark, Reading, Berkshire RG2 6GP, United Kingdom

----------------------------------------------------------------------

Subject: Windows 2000 COM port driver blocking
From: "Lee Rhodes"
Date: Wed, 20 Mar 2002 13:49:09 GMT
X-Message-Number: 9

Hi,
I have a driver which communicates with a modem over serial (V.24) or USB
ports.
The driver is a little like Unimodem. With the V.24 (COM) port
implementation,
I'm having problems reading after the modem goes into 'data mode' (i.e.
after
a call is connected). RAS is using this driver for dial-up networking so a
PPP
conversation ensues after the connection. A tx/rx sequence should occur (LCP
negotiation) but instead I see the RAS times-out the first tx request (after
2 seconds) and retransmits (so I get a tx/tx sequence) and then the serial
port
driver beneath me sends me a double response (2 LCP responses from the
remote
side of the connection). The double-response is usually in or around the 180
byte mark. I would expect to receive a 90 byte (-ish) response but it seems
that the device (or the serial port driver) is holding back until it has
accumulated
some data before delivering the incoming data to my async read complete.
This
would seem to be the case but after this initial blocking, I'm able to
receive
read buffers < 180 bytes. It just seems to be the initial read after going
into
data mode.

I'm not seeing any of this behaviour with the USB implementation which works
fine. I see the expected tx/rx sequence.

Can anyone give me a hint as to what would be causing the serial driver to
hold
back? I've set the following parameters after opening the COM port with
ZwCreateFile:

Queue size: 512 (read) 512 (write)
Handshake: RTS/CTS
XonLimit: 128
XoffLimit: 0
Read time out: 20 msecs
Baud: 115200
Line control: 1 stop bit, No parity, 8 data bits

Thanks,
Lee

***************************ADVERTISEMENT******************************
For ALL the latest Soccer news on your club, GAA sports results and the
latest on your F1 stars plus much more check out
http://sport.iol.ie/sport. Sport On-Line.... It's a passion

----------------------------------------------------------------------

Subject: RE: Windows 2000 COM port driver blocking
From: "Carlo Andreoli - Numerica Progetti"
Date: Wed, 20 Mar 2002 15:06:27 +0100
X-Message-Number: 10

> response but it seems
> that the device (or the serial port driver) is holding back=20
> until it has accumulated
> some data before delivering the incoming data to my async=20
> read complete.
>=20

Do you create your read IRP inside your unimodem-like driver?
If so, how many characters you require, and which read timeout
handling you set before sending the read Irp to the serial driver
(i.e. how exactly you set all members of the SERIAL_TIMEOUTS
structure)?

Carlo

----------------------------------------------------------------------

Subject: Power management in Windows ME
From: "Nagarajan Duraisamy - CTD, Chennai."
Date: Wed, 20 Mar 2002 20:19:05 +0530
X-Message-Number: 11

Hi,
We have a modem device for which we are writing a driver.
Our driver is the power policy owner of the device.
When we get a StartDevice PnP IRP, after doing all the processing
we power down the device to D3 state.

This is done to make sure that the device doesn't take any power when
not in use. We do a power up when somebody tries to open our device.Our
device doesn't support wakeup
and the only supported states are D0 and D3.

Subsequently, we get a IRP_MN_SET_POWER with D0 as the desired
device state. We do not initiate any such request as a power policy owner.
This happens after we complete IRP_MN_QUERY_PNP_DEVICE_STATE IRP with no
changes to the PNP_DEVICE_STATE.

When can such an IRP_MN_SET_POWER be sent by the config manager ? Are we
missing something ?

Thanks in advance.

With Regards
D.Nagarajan

----------------------------------------------------------------------

Subject: Net problems...
From: "Alexey Logachyov"
Date: Wed, 20 Mar 2002 16:57:49 +0200
X-Message-Number: 12

I've just installed Windows NT build 1381 checked. When I'm trying to
access it over network from my comp with WinXP installed I get ann
error. NT itself hits the following system event: The server received an
incorrectly formatted request from (?). When run under WinDBG I can see
another message: ValidatesmbHeader: Flags2 = c853, valid bits = f847,
invalid bit(s) = 10. I can access this computer from another one with
WinNT. Installing SP6 checked does not solve the problem.

Did anyone encounter this problem? Does anyone know how to solve it?

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: "Satrio Mualim"
Date: Wed, 20 Mar 2002 07:57:14 -0700
X-Message-Number: 13

Sorry, typo, 802.1q is what I meant
The NIC is the Intel Pro 100 (82558 chip).
I also found out that the packet driver will work properly if
I limit the packet size to about 200 bytes, i.e under the limit
it sends the packets the way it does before VLAN tagging is
enabled.
What I dont understand is the fact that it seems to be transparent
to TCP/IP which works the same way as before tagging was enabled.
I suspect that I must be missing some calls/settings that are required
to make it work with bigger packets

Thanks,

Satrio

Stephan Wolf wrote:

> AFAIK, there is no VLAN support in NT4, ie. NDIS4. Thus, if the NIC
> and its driver supports VLAN, it is probably some proprietary
> solution.
>
> Does your driver support VLAN tagging? Or is it just the NIC driver?
> In the latter case, go and ask the NIC vendor (which one is it?).
>
> BTW, what is 803.1q??? => 802.1q!
>
> Stephan
> ---
> On Mon, 18 Mar 2002 18:35:19 -0700, "Satrio Mualim"
> wrote:
>
> >I am looking for some suggestion/help in solving some packet driver
problems.
> >
> >We have a driver, roughly based on the NT Packet driver running on NT 4
SP6.
> >It has been performing OK with no particular problems until recently we
turn
> >on the 803.1q VLAN tagging on the NIC.
> >
> >Now it pauses after sending 3 packets. Looking into this problem I found
out
> >that
> >after each third packet my Protocol Characteristic StatusHandler and
> >StatusCompleteHandler
> >callbacks are executed with status NDIS_STATUS_RESET_START and
> >NDIS_STATUS_RESET_END.
> >
> >Is there anything that has to be done differently/added when VLAN tagging
is
> >enabled?
> >Looking at the packets "on the wire" it seems that they are going out at
the
> >normal speed until
> >the third packet is out, after which there is a pause for about 4 to 6
seconds.
> >The system has 2 NICs and if I turn off tagging on one of the NIC's,
packets are
> >still going at the
> >old fast speed on the NIC with no tagging, and slow as mentioned above
for the
> >NIC with tagging.
> >
> >Any suggestions on where I should look to solve this problem?
> >
> >Thanks,
> >
> >Satrio
>
> ---
> You are currently subscribed to ntdev as: xxxxx@agcs.com
> To unsubscribe send a blank email to %%email.unsub%%

----------------------------------------------------------------------

Subject: RE: Windows 2000 COM port driver blocking
From: "Lee Rhodes"
Date: Wed, 20 Mar 2002 15:14:29 GMT
X-Message-Number: 14

Thanks for the interest Carlo.

> Do you create your read IRP inside your unimodem-like driver?
Yes.

> If so, how many characters you require,
I send down 4 IRP_MJ_READ buffers with 512 byte buffers.

> and which read timeout handling you set before sending the read Irp
> to the serial driver (i.e. how exactly you set all members of
> the SERIAL_TIMEOUTS structure)?
SERIAL_TIMEOUTS serialTimeouts;
serialTimeouts.ReadIntervalTimeout = 0x14;
serialTimeouts.ReadTotalTimeoutMultiplier = 0;
serialTimeouts.ReadTotalTimeoutConstant = 0;
serialTimeouts.WriteTotalTimeoutMultiplier = 0;
serialTimeouts.WriteTotalTimeoutConstant = 0xffffffff;

Can you see anything wrong? It's just the initial read that it seems to
choke
on. Afterwards I can see smaller (80 and 40 byte) reads arriving in my
completion
routine.

Thanks again,
Lee

***************************ADVERTISEMENT******************************
For ALL the latest Soccer news on your club, GAA sports results and the
latest on your F1 stars plus much more check out
http://sport.iol.ie/sport. Sport On-Line.... It's a passion

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 16:01:59 GMT
X-Message-Number: 15

Hmm, actually, I was involved in a VLAN project for W2K. AFAIK,
built-in VLAN support is only in XP.

What needs to be done for VLAN tagging is that the VLAN header has to
be inserted for send packets and removed from receive packets at a
very low level, ie. "below" any protocol stack.

This can be done either in the NIC driver or in some intermediate
driver ("filter"). The VLAN header sits in (or after - depends on how
you look at it) the MAC header. Thus, a new *local* MAC header has to
be created in the NIC or IM driver.

I remember we also had problems with "large" frames.

I guess the problems go away if you set up the MAC header in a
separate NDIS_BUFFER in any frame that your protocol sends.

Stephan
---
On Wed, 20 Mar 2002 07:57:14 -0700, "Satrio Mualim"
wrote:

>Sorry, typo, 802.1q is what I meant
>The NIC is the Intel Pro 100 (82558 chip).
>I also found out that the packet driver will work properly if
>I limit the packet size to about 200 bytes, i.e under the limit
>it sends the packets the way it does before VLAN tagging is
>enabled.
>What I dont understand is the fact that it seems to be transparent
>to TCP/IP which works the same way as before tagging was enabled.
>I suspect that I must be missing some calls/settings that are required
>to make it work with bigger packets
>
>Thanks,
>
>Satrio
>
>Stephan Wolf wrote:
>
>> AFAIK, there is no VLAN support in NT4, ie. NDIS4. Thus, if the NIC
>> and its driver supports VLAN, it is probably some proprietary
>> solution.
>>
>> Does your driver support VLAN tagging? Or is it just the NIC driver?
>> In the latter case, go and ask the NIC vendor (which one is it?).
>>
>> BTW, what is 803.1q??? => 802.1q!
>>
>> Stephan
>> ---
>> On Mon, 18 Mar 2002 18:35:19 -0700, "Satrio Mualim"
>> wrote:
>>
>> >I am looking for some suggestion/help in solving some packet driver
problems.
>> >
>> >We have a driver, roughly based on the NT Packet driver running on NT 4
SP6.
>> >It has been performing OK with no particular problems until recently we
turn
>> >on the 803.1q VLAN tagging on the NIC.
>> >
>> >Now it pauses after sending 3 packets. Looking into this problem I found
out
>> >that
>> >after each third packet my Protocol Characteristic StatusHandler and
>> >StatusCompleteHandler
>> >callbacks are executed with status NDIS_STATUS_RESET_START and
>> >NDIS_STATUS_RESET_END.
>> >
>> >Is there anything that has to be done differently/added when VLAN
tagging is
>> >enabled?
>> >Looking at the packets "on the wire" it seems that they are going out at
the
>> >normal speed until
>> >the third packet is out, after which there is a pause for about 4 to 6
seconds.
>> >The system has 2 NICs and if I turn off tagging on one of the NIC's,
packets are
>> >still going at the
>> >old fast speed on the NIC with no tagging, and slow as mentioned above
for the
>> >NIC with tagging.
>> >
>> >Any suggestions on where I should look to solve this problem?
>> >
>> >Thanks,
>> >
>> >Satrio

----------------------------------------------------------------------

Subject: Re: Looking for a driver for diamond multimedia Stealth video card
25 00 revision 1.03 to use it under win 2k sp2
From: "Miguel Monteiro"
Date: Wed, 20 Mar 2002 16:13:32 -0000
X-Message-Number: 16

Hmmm... I'm afraid WinNT 4 is the best you can find - check:

http://www.diamondmm.com/default.asp?menu=support&submenu=Legacy_Graphic
s&item=drivers&product=Stealth_Video_2500

Maybe the NT 4 driver works well under Win2K (as legacy):
it's worth trying.

Also, search for it on:

http://pcdrivers.com

I've found some useful stuff there, in the past.

Hope it helps,

Miguel Monteiro
xxxxx@criticalsoftware.com
------------------------------------------------------------
Critical Software, S.A. - http://www.criticalsoftware.com
111 North Market Street, 6th floor, San Jose, CA, USA, 95113
Tel: +1.408.9711231, Fax +1.408.9383929
R. Pedro Nunes, IPN, 3030-199 Coimbra, Portugal
Tel: +351.239.700945 - Fax: +351.239.700905
------------------------------------------------------------
DISCLAIMER: This mail contents represent
my own personal opinions and do not, in any way,
represent the opinion or policy of Critical Software, S.A.
------------------------------------------------------------
"Humour and love are God's answers
to Human weaknesses" =:o)8
------------------------------------------------------------

----- Original Message -----
From: "Fernando, Robert"
To: "NT Developers Interest List"
Cc: "'NT Developers Interest List' (E-mail)"
Sent: Wednesday, March 20, 2002 12:58 PM
Subject: [ntdev] Looking for a driver for diamond multimedia Stealth
video card 25 00 revision 1.03 to use it under win 2k sp2

Hi all,

Please could anyone tell me if there is a driver available for the
following
card.
Diamond multimedia Stealth video 2500 revision 1.03

The chip set used on the card is Alliance Semiconductor promotion AT24

I have managed to find one for nt4 can this be used with w2k ?

Robert Fernando
Anite Telecoms Ltd
110 Fleet Road
Fleet
Hampshire GU51 4BL
United Kingdom
Tel: +44 (0) 1252 775200
Fax: +44 (0) 1252 775 321
Email: xxxxx@anitetelecoms.com

Anite Telecoms Ltd, Registered in England No. 1721900 Registered
Office:
100 Longwater Avenue, GreenPark, Reading, Berkshire RG2 6GP, United
Kingdom

---
You are currently subscribed to ntdev as: xxxxx@criticalsoftware.com
To unsubscribe send a blank email to %%email.unsub%%

----------------------------------------------------------------------

Subject: local address in TDIEventConnect
From: Richard Oak
Date: Wed, 20 Mar 2002 16:39:13 -0000
X-Message-Number: 17

Hi All,

I have a driver layered on top of \Device\Tcp. In my TDIEventConnect
handler, I want to know the local address of the connection. When I send a
TDI_QUERY_INFORMATION to get the address, it comes back with 0.0.0.0:4444,
which is what I supplied in the call to bind() in the user mode
application.

Is there a way to get the actual address of the local endpoint (e.g.
192.168.100.69:4444). If there are two network cards in the system, I want
to know which one the connection is coming in on so I can filter
differently.

Thanks in advance

Richie

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: "Satrio Mualim"
Date: Wed, 20 Mar 2002 10:28:22 -0700
X-Message-Number: 18

I was wondering about how the NIC driver inserts the VLAN tag and
found out the following:
- Creating the VLAN causes the original device name to be
hidden/unavailable.
Instead there is a new device name /device/iVLAN....
- Using the existing packet driver, just opening the new device, will cause
packets on the wire to be <0x8100> type> ....
where the 0x8100 and vlan tag are inserted by the NIC driver (as far as I
can tell)
and vlan tag is the value specified during the setup of the NIC
- If I cheat and modify the registry to make the original device available,
and send packets using it, I get packets on the wire that have everything
the way
they should be except the is 0x0000.
In either case I still have the problem where it resets the NIC every 3
packets.

I am interested in trying to do what you said in your last paragraph.
Why would the problem go away if I setup a different NDIS_BUFFER
for the MAC header and how do I make sure that the MAC header
buffer and the "payload" still go out as one packet on the wire?

Thanks,

Satrio

Stephan Wolf wrote:

> Hmm, actually, I was involved in a VLAN project for W2K. AFAIK,
> built-in VLAN support is only in XP.
>
> What needs to be done for VLAN tagging is that the VLAN header has to
> be inserted for send packets and removed from receive packets at a
> very low level, ie. "below" any protocol stack.
>
> This can be done either in the NIC driver or in some intermediate
> driver ("filter"). The VLAN header sits in (or after - depends on how
> you look at it) the MAC header. Thus, a new *local* MAC header has to
> be created in the NIC or IM driver.
>
> I remember we also had problems with "large" frames.
>
> I guess the problems go away if you set up the MAC header in a
> separate NDIS_BUFFER in any frame that your protocol sends.
>
> Stephan
> ---

----------------------------------------------------------------------

Subject: VideoPortMapMemory call problem under WinXp
From: "Joe Moriarty"
Date: Wed, 20 Mar 2002 12:58:10 -0500
X-Message-Number: 19

We're using VideoPortMapMemory to basically obtain a user-mode process
address mapping for a physical memory segment. This works fine under 2K,
but under XP, it fails with (essentially a) STATUS_INVALID_PARAMETER. One
of the "flags" associated with this routine is a specification of "how" to
map the memory. For 2K VIDEO_MEMORY_SPACE_MEMORY was used and the
"virtualAddress" variable was set to INVALID_HANDLE to mark this as "map to
the current process". For XP, we tried using NULL for virtualAddress (as
per the DDK), but that still failed. If VIDEO_MEMORY_SPACE_USER_MODE was
used, an address would get mapped, but it seems to be a kernel space
address, so when passed back up the chain, the application fails. Anyone
else run into this problem? Alternatively, how does one create a user-mode
virtual mapping from a physical address to the current process address
space.

Thanks In Advance,
Joe

----------------------------------------------------------------------

Subject: RE: XP SMP
From: "Hassan I Khan"
Date: Wed, 20 Mar 2002 18:03:31
X-Message-Number: 20

Thanks for sharing your experience. Though Bios detects 2 Pentium III I was
troubled to see windbg report MP kernel and 1 processor. Now I will go ahead

and verify through task manager.

Thanks much,
Hassan

>From: "Whitman, Steve"
>Reply-To: "NT Developers Interest List"
>To: "NT Developers Interest List"
>Subject: [ntdev] RE: XP SMP
>Date: Wed, 20 Mar 2002 07:05:36 -0500
>
>I have very similar results on my dual CPU machine. However, if I bring up
>task manager it displays two CPU graphs and when I bring up device manager
>and select Processors it lists both processors.
>
>But Windbg reports MP kernel with 1 proc and system properties just says
>Pentium III, etc.
>
>I'm sure that my system is using both processors. I believe that system
>properties is behaving as it should (it reports the processor and since
>they
>are both the same it doesn't need to list both). I don't know why Windbg
>reports 1 processor.
>
> - Steve -
>
>
> > -----Original Message-----
> > From: Jim Young [mailto:xxxxx@youngendeavors.com]
> > Sent: Tuesday, March 19, 2002 10:51 PM
> > To: NT Developers Interest List
> > Subject: [ntdev] RE: XP SMP
> >
> >
> > How many processors does the BIOS detect? If it only shows 1
> > then I would
> > suspect a defect in the CPU, socket, or motherboard.
> >
> > -----Original Message-----
> > From: xxxxx@lists.osr.com
> > [mailto:xxxxx@lists.osr.com]On Behalf Of Hassan I Khan
> > Sent: Tuesday, March 19, 2002 8:36 PM
> > To: NT Developers Interest List
> > Subject: [ntdev] XP SMP
> >
> >
> > I recently assembled a dual-processor system for kernel
> > debugging. Installed
> > XP prof (checked version) came with msdn subscription (the
> > very first XP
> > CDs). During installation I didn't see any message or option
> > that XP is
> > going to run on 2 processors. System properties just says
> > Pentium 1.0 GHz
> > etc. Kernel debugger reports an MP kernel with (1 procs) etc.
> >
> > Whats going on here ? Has anyone done a dual-processor XP
> > installation ? How
> > do you do it ?
> >
> > Apologies for this question being slightly off-topic on this list.
> >
> > thanks for any comments,
> > Hassan
> >
> >_________________________________________________________________
> > Get your FREE download of MSN Explorer at
>http://explorer.msn.com/intl.asp.
>
>
>---
>You are currently subscribed to ntdev as: xxxxx@youngendeavors.com
>To unsubscribe send a blank email to %%email.unsub%%
>
>
>
>
>---
>You are currently subscribed to ntdev as: xxxxx@cognex.com
>To unsubscribe send a blank email to %%email.unsub%%
>
>---
>You are currently subscribed to ntdev as: xxxxx@hotmail.com
>To unsubscribe send a blank email to %%email.unsub%%

_________________________________________________________________
MSN Photos is the easiest way to share and print your photos:
http://photos.msn.com/support/worldwide.aspx

----------------------------------------------------------------------

Subject: RE: Windows 2000 COM port driver blocking
From: "Carlo Andreoli - Numerica Progetti"
Date: Wed, 20 Mar 2002 19:11:54 +0100
X-Message-Number: 21

> serialTimeouts.ReadIntervalTimeout =3D 0x14;
> serialTimeouts.ReadTotalTimeoutMultiplier =3D 0;
> serialTimeouts.ReadTotalTimeoutConstant =3D 0;
> serialTimeouts.WriteTotalTimeoutMultiplier =3D 0;
> serialTimeouts.WriteTotalTimeoutConstant =3D 0xffffffff;
>=20

With this setting, the read should wait (without timeouts) until
the first char is received, then continue waiting until
- all requested chars are received
or
- 20 mSec elapse without receiving characters (after at least one
has been received)
Is this behavior what you want? (the setting of the timeouts is
quite tricky; one good reference is the SERIAL.SYS sample in the
Win2K DDK, check out how the SERIAL_TIMEOUTS are handled by the
SerialStartRead() in READ.C)

Also, be sure to set serialTimeouts *before* the answer has been
received (with a half-duplex protocol I would, in sequence, set
the Tx/Rx timeouts, send out the data and wait for the answer)

HTH

Carlo
=20

----------------------------------------------------------------------

Subject: Checked build of SNMP?
From: "S Elampooranan , Gurgaon"
Date: Thu, 21 Mar 2002 00:02:00 +0530
X-Message-Number: 22

Hi!

I am looking for the checked build of SNMP.exe ..though it's mentioned in MS
KB articles, I'm not able to locate it..

do U know how to locate/get one?

Elam

----------------------------------------------------------------------

Subject: RE: XP SMP
From: "Maxim S. Shatskih"
Date: Wed, 20 Mar 2002 21:44:55 +0300
X-Message-Number: 23

> are both the same it doesn't need to list both). I don't know why Windbg
> reports 1 processor.

IIRC when the OS is booted and the WinDbg connection is established, only 1
CPU is running, the second one is started later after
boot drivers initialization or such.
Do you have 0: or 1: prefixes in WinDbg's traces? If yes, then both CPUs are
working.

Max

----------------------------------------------------------------------

Subject: RE: Net problems...
From: "Eric Lee Steadle"
Date: Wed, 20 Mar 2002 14:28:02 -0500
X-Message-Number: 24

Install Netmon and grab a packet trace when you try and connect to your NT
build. I suspect that the NT box is trying a new SMB
operation that the NT box isn't expecting. The flags2 field is used to
specify options during SMB conversations. Every SMB packet
has one, so it's kind of tough to nail down the exact problem without
additional context.

MS is normally pretty good about remainging backward compatible with the SMB
protocol. If the XP box can't talk to your NT box the
'new' XP way, then it's supposed to fall back to the 'old' NT way. Odds are
there's a registry tweak that will get you working
again, but will sacrifice some whiz-bang functionality or, more probably,
security. You just need to figure out what the setting is.

ERX

P.S. Netmon can be a bitch to find in the MSDN subscription. Last time I
looked, it was on one of the SMS disks.

>-----Original Message-----
>From: xxxxx@lists.osr.com
>[mailto:xxxxx@lists.osr.com]On Behalf Of Alexey Logachyov
>Sent: Wednesday, March 20, 2002 9:58 AM
>To: NT Developers Interest List
>Subject: [ntdev] Net problems...
>
>
>I've just installed Windows NT build 1381 checked. When I'm trying to
>access it over network from my comp with WinXP installed I get ann
>error. NT itself hits the following system event: The server
>received an
>incorrectly formatted request from (?). When run under WinDBG I can see
>another message: ValidatesmbHeader: Flags2 = c853, valid bits = f847,
>invalid bit(s) = 10. I can access this computer from another one with
>WinNT. Installing SP6 checked does not solve the problem.
>
>Did anyone encounter this problem? Does anyone know how to solve it?
>
>
>---
>You are currently subscribed to ntdev as: xxxxx@spinnakernet.com
>To unsubscribe send a blank email to %%email.unsub%%

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: xxxxx@hotmail.com (Stephan Wolf)
Date: Wed, 20 Mar 2002 19:47:56 GMT
X-Message-Number: 25

AFAIK, "iVLAN" is an intermediate driver.

The driver has to make a local copy of the MAC header in order to be
able to insert the VLAN tag. If the complete packet is just one
NDIS_BUFFER then the driver *might* get into trouble. I am not saying
this is so but it could be.

The lookahead size also plays a major role here. NDIS requires that
the first NDIS_BUFFER in a received (!) packet is at least the
lookahead size (or the complete packet, whichever is shorter).

Do you specify any specific lookahead size for your protocol?

Anyway, if you say TCP/IP works ok then maybe because it also uses a
separate NDIS_BUFFER for the header(s).

You can also try and run the NDIS Tester against the NIC driver with
VLAN tagging enabled. If that passes, it's a problem of your protocol.
Otherwise, it's a bug in the NIC driver, which I expect.

Download "ndt391.exe" here:

ftp://ftp.microsoft.com/services/whql/ndis/

Stephan
---
On Wed, 20 Mar 2002 10:28:22 -0700, "Satrio Mualim"
wrote:

>
>
>I was wondering about how the NIC driver inserts the VLAN tag and
>found out the following:
>- Creating the VLAN causes the original device name to be
hidden/unavailable.
> Instead there is a new device name /device/iVLAN....
>- Using the existing packet driver, just opening the new device, will cause
> packets on the wire to be <0x8100> type> ....
> where the 0x8100 and vlan tag are inserted by the NIC driver (as far as
I can tell)
> and vlan tag is the value specified during the setup of the NIC
>- If I cheat and modify the registry to make the original device available,
> and send packets using it, I get packets on the wire that have everything
the way
> they should be except the is 0x0000.
>In either case I still have the problem where it resets the NIC every 3
packets.
>
>
>I am interested in trying to do what you said in your last paragraph.
>Why would the problem go away if I setup a different NDIS_BUFFER
>for the MAC header and how do I make sure that the MAC header
>buffer and the "payload" still go out as one packet on the wire?
>
>Thanks,
>
>Satrio
>
>Stephan Wolf wrote:
>
>> Hmm, actually, I was involved in a VLAN project for W2K. AFAIK,
>> built-in VLAN support is only in XP.
>>
>> What needs to be done for VLAN tagging is that the VLAN header has to
>> be inserted for send packets and removed from receive packets at a
>> very low level, ie. "below" any protocol stack.
>>
>> This can be done either in the NIC driver or in some intermediate
>> driver ("filter"). The VLAN header sits in (or after - depends on how
>> you look at it) the MAC header. Thus, a new *local* MAC header has to
>> be created in the NIC or IM driver.
>>
>> I remember we also had problems with "large" frames.
>>
>> I guess the problems go away if you set up the MAC header in a
>> separate NDIS_BUFFER in any frame that your protocol sends.
>>
>> Stephan
>> ---

----------------------------------------------------------------------

Subject: Re: BUILD SOURCES from another directories
From: Michal Vodicka
Date: Wed, 20 Mar 2002 21:15:47 +0100
X-Message-Number: 26

No workaround, it is standard solution. It is probably because build utility
is designed for building whole trees, not just one driver.

Maybe there is a solution (really workaround) which would allow to use
standard settings (i.e. makefile.def) and still allow sources in other
directories. Try

nmake MAKEDLL=1

in the directory with SOURCES instead of build. I believe the limitation to
current and parent directories is only for build utility and above could
work. Of course, there can be problem with dependencies. Note: I use it only
for quick builds and never tried with files in other directories.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]

> ----------
> From: xxxxx@yahoo.com[SMTP:xxxxx@yahoo.com]
> Reply To: xxxxx@lists.osr.com
> Sent: Monday, March 18, 2002 8:44 PM
> To: xxxxx@lists.osr.com
> Subject: [ntdev] Re: BUILD SOURCES from another directories
>
> This works but looks more like a workaround.
>
> I tried adding files from other directories to SOURCES a few years back.
> At that time DDK was
> quite shaky, so it didn't surprise me that it did not work.
> Since that time MS has done good job on DDK side, so I'm just curious as
> to why is it still not
> supported? Anyone has any idea?
> Unless I miss something obvious this should be trivial thing to do.
>
> -- Max.
>
>
>
> --- "Maxim S. Shatskih" wrote:
> > > If I specify source files from another directory, for example
> > >
> > > SOURCES= util\datapipe.c
> > >
> > > BUILD utility outputs error message. Maybe solution is to set some
> variable
> >
> > Put another SOURCES to that directory, which will build a .LIB.
> > Then link this .LIB to your driver binary.
> >
> > Max
> >
> >
> >
> > ---
> > You are currently subscribed to ntdev as: xxxxx@yahoo.com
> > To unsubscribe send a blank email to %%email.unsub%%
>
>
>__________________________________________________
> Do You Yahoo!?
> Yahoo! Sports - live college hoops coverage
> http://sports.yahoo.com/
>
> ---
> You are currently subscribed to ntdev as: michal.vodicka@st.com
> To unsubscribe send a blank email to %%email.unsub%%
>

----------------------------------------------------------------------

Subject: Re: Problems with Packet Driver
From: "Satrio Mualim"
Date: Wed, 20 Mar 2002 15:03:17 -0700
X-Message-Number: 27

Thanks for all the info and hints.
I tried to add another small NDIS_BUFFER and chained it to the packet
descriptor.
And, at least from my first test run, it seems to behave the way it should.
No more pause after each 3 packets.

I still have to figure out what this means to existing application code, but
at least it
can be made to work. And I agree that this maybe the same reason that TCP/IP
works with no problems.

I will check the lookahead size and see if this is another potential
problem.

Satrio

Stephan Wolf wrote:

> AFAIK, "iVLAN" is an intermediate driver.
>
> The driver has to make a local copy of the MAC header in order to be
> able to insert the VLAN tag. If the complete packet is just one
> NDIS_BUFFER then the driver *might* get into trouble. I am not saying
> this is so but it could be.
>
> The lookahead size also plays a major role here. NDIS requires that
> the first NDIS_BUFFER in a received (!) packet is at least the
> lookahead size (or the complete packet, whichever is shorter).
>
> Do you specify any specific lookahead size for your protocol?
>
> Anyway, if you say TCP/IP works ok then maybe because it also uses a
> separate NDIS_BUFFER for the header(s).
>
> You can also try and run the NDIS Tester against the NIC driver with
> VLAN tagging enabled. If that passes, it's a problem of your protocol.
> Otherwise, it's a bug in the NIC driver, which I expect.
>
> Download "ndt391.exe" here:
>
> ftp://ftp.microsoft.com/services/whql/ndis/
>
> Stephan
> ---

---

END OF DIGEST

---
You are currently subscribed to ntdev as: xxxxx@anitetelecoms.com
To unsubscribe send a blank email to %%email.unsub%%