|
I'm trying to build a USB hub driver whose behavior is like windows' built in one.
My implementation is based on the ddk's example: toaster. Mass storage devices and scanners
works fine on my driver, however usb cameras can't be started.
I analysed the irps when the camera is plugined. After the start device irp, the driver received
a remove device irp. The ddk document says if anyone in the driver stack failed in start device irp,
pnp manager will send a remove device irp. I think there must be something wrong in the implementation before the remove device irp is sent.
There is a query interface irp which query TRANSLATOR_INTERFACE before the remove device irp is sent and I return STATUS_NO_SUPPORTED.
My response to start device irp is just like the toaster example does.
My question is:
1. Do I have to implement the TRANSLATOR_INTERFACE to fix the problem?
2. Or is there something else I have to do in my start device response?
|
|
|
|
|
If you're trying to build something that functions like the standard one, my first suggestion is to get an IRP tracker and monitor what the standard one does. If it processeses a given IRP, your replacement should as well.
Judy
|
|
|
|
|
Thanks Judy! I tried to process query interface irp for translator interface(which the standard one does), but I couldn't find any information about how to do it.Any suggestions?
|
|
|
|
|
Sorry, no. Never done USB or even looked at it out of curiousity.
Judy
|
|
|
|
|
Did you try returning a success status?
Have a google for it, there is some info out there on this message.
Morality is indistinguishable from social proscription
|
|
|
|
|
I'm stuck up with two processors T7250 (Intel Core 2 Duo 2 GHz, 2 MB cache) and T7500 (Intel Core 2 Duo 2.2 GHz, 4 MB cache)
Actually +0.2 GHz speed and +2B cache can make any significant difference in performance?
Between these two, there's nearly 150$ difference.
Please share your opinion.
-Sarath.
"Great hopes make everything great possible" - Benjamin Franklin
|
|
|
|
|
Hi,
I would not pay much for 10% higher frequency, assuming the FSB frequency is the same
(I rhink it is 800MHz for both).
I would not pay much for twice the cache size unless my main application would fit
in the larger one while not fitting the smaller one, which is very unlikely.
I would rather spend some money in getting:
- faster memory if my app is not disk bandwidth limited;
- faster disk, or a second disk, if it is (such as Visual Studio);
- more memory (unless you run only one or a few apps at a time).
As a software developer (with mainly small apps) the T7250 IMO is the ideal CPU right now,
it is the cheapest one with 800MHz FSB (so was the 1.8GHz T7100 earlier).
|
|
|
|
|
Thanx for you suggestions. What you said is true, just 10% increased cpu wont make big difference.
-Sarath.
"Great hopes make everything great possible" - Benjamin Franklin
|
|
|
|
|
Hi
well it all depends upon the number following T????.
Check out intel.com for refernce.
Spread wat u Know!
|
|
|
|
|
Hi,
when i used to Driver Installation/Uninstallation for multiple times i.e.10 times, Driver Installation/Uninstallation fails.
if any one can help me then please let me know
Regards,
Pavan
pavan
|
|
|
|
|
Look in the setup*.log files for details on the failure.
Judy
|
|
|
|
|
Why are you doing it ten times? Are you trying to kill the system? If so then why? You dont need proof that windows in unstable and buggy.
Morality is indistinguishable from social proscription
|
|
|
|
|
Two things I can think of. One each install/uninstall cycle represents testing a new build. The second is that the OP is testing the installs success under multiple somewhat different environments using the same OS install and just fiddling with the other stuff.
Otherwise [Microsoft is] toast in the long term no matter how much money they've got. They would be already if the Linux community didn't have it's head so firmly up it's own command line buffer that it looks like taking 15 years to find the desktop.
-- Matthew Faithfull
|
|
|
|
|
dan neely wrote: One each install/uninstall cycle represents testing a new build
I doubt it, unless the guy likes pain. Its easier to either copy over the sys file or use .kdfiles off of WinDbg.
dan neely wrote: The second is that the OP is testing the installs success
More likely. However, if the test doesnt reflect a valid use case then why do it?
Evetyone knows windows is fragile and easy to kill, whats the point proving it?
Morality is indistinguishable from social proscription
|
|
|
|
|
I have added USB wireless adapter to my Windoze (2000 Advanced Server - please do not laugh!). Its status is Enabled but the OS still uses the wired PCI card.
1. How does OS determine which network adapter to use?
2. Why does the "status" of wireless show "cable unplugged" at times?
3. It seems that the LAN device cannot be enabled / disabled in Control Panel with any confidence. It is a hit and wait and try again affair.
I am just asking because I have seen similar behavior in plain old vanilla serial RS232 interface. And NO, I'll not run out to "upgrade" to Vista Exceptionally Super SP12 XXL version.
Cheers
Vaclav
|
|
|
|
|
Vaclav_Sal wrote: How does OS determine which network adapter to use?
If you have multiple network adapters connected, the system is multi-homed. The decision as to which route packets take is dependent on:
1. The routing table.
2. The 'weight' given to each network interface.
Strictly the weight is part of the routing table, but takes effect only if identical routes are discovered.
The IP routing table can be made incredibly complex, but for most systems not otherwise configured, you basically get a default route (to 0.0.0.0 masked 0.0.0.0, i.e. anywhere) configured to use your default gateway via each interface, with a weight. The wired network card will typically get given a lower weight and be preferred to the wireless card. Likewise each card will have a route for the IP and mask it's configured for. If the cards are on the same network, again the wired card will be preferred.
If the cards are on different networks (IP ranges), packets for the wired card's network should go to the wired card and vice versa. Packets for other networks will go to whichever card the default gateway is nearest to. If there are multiple 'default gateways' the weight again controls which is used.
The routes for the wired card will be (or should be) removed if the OS detects that the cable has been unplugged. Only then will the wireless card be used.
You can view the routing table by issuing the route print command.
Example: I have a wired network card and a wireless card. I'm running Windows Vista. My IPv4 routing table is, with both enabled and in good coverage:
Network Destination Netmask Gateway Interface Metric
0.0.0.0 0.0.0.0 192.168.1.1 192.168.1.2 20
0.0.0.0 0.0.0.0 192.168.1.1 192.168.1.4 25
127.0.0.0 255.0.0.0 On-link 127.0.0.1 306
127.0.0.1 255.255.255.255 On-link 127.0.0.1 306
127.255.255.255 255.255.255.255 On-link 127.0.0.1 306
192.168.1.0 255.255.255.0 On-link 192.168.1.2 276
192.168.1.0 255.255.255.0 On-link 192.168.1.4 281
192.168.1.2 255.255.255.255 On-link 192.168.1.2 276
192.168.1.4 255.255.255.255 On-link 192.168.1.4 281
192.168.1.255 255.255.255.255 On-link 192.168.1.2 276
192.168.1.255 255.255.255.255 On-link 192.168.1.4 281
224.0.0.0 240.0.0.0 On-link 127.0.0.1 306
224.0.0.0 240.0.0.0 On-link 192.168.1.2 276
224.0.0.0 240.0.0.0 On-link 192.168.1.4 281
255.255.255.255 255.255.255.255 On-link 127.0.0.1 306
255.255.255.255 255.255.255.255 On-link 192.168.1.2 276
255.255.255.255 255.255.255.255 On-link 192.168.1.4 281 The wired card has address 192.168.1.2 while the wireless card is 192.168.1.4. You can see I have two routes for the 192.168.1.x network, with the wired card having weight 276 compared to wireless 281, so the wired card is preferred. Multicast is on 224.x.x.x and again wired is preferred to wireless and loopback (weight 306). Broadcasts (all-ones and 192.168.1.255) are all weighted to wired as well. Packets addressed to the loopback network (127.0.0.0) all go to the loopback interface. Finally, there are the two default routes and both state to send to 192.168.1.1, but again the wired network takes priority.
Why does the "status" of wireless show "cable unplugged" at times?
Most likely you're out of range or on the borders of coverage. Windows 2000 does not understand wireless cards, so the card reports poor network performance as 'cable unplugged', so that the OS can make better routing decisions. If you care, get an OS (e.g. XP SP2) which understands wireless cards.
Vaclav_Sal wrote: It seems that the LAN device cannot be enabled / disabled in Control Panel with any confidence. It is a hit and wait and try again affair.
I've not seen this behaviour, but it can take a while for wireless devices to associate with a network. They are also supposed to disassociate from the base station when being disabled, which should be acknowledged, which could take a while to time out if out of coverage.
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
Take out the LAN cable. This wil kill the IP address for it and remove it from therouting table. Give the system about 5 seconds and your wireless card will become the default IP route.
Morality is indistinguishable from social proscription
|
|
|
|
|
Hi all
m having Seagate Barracuda 80 GB HDD some days before i use all memory of it, but currently it display only 32 GB i checked it`s Jumper setting and all possible tricks but it still shows 32GB
Plz Tell solution........
The Stifler
--
Bugs can neither be created nor be removed from software by a developer. They can only be converted from one form to another. The total number of bugs in the software always remain constant.
|
|
|
|
|
Which O/S are you using?? What's the drive partitioned as?? What was the partition formatted with?? FAT, FAT32, NTFS??
|
|
|
|
|
M using
Windows XP, Drive Partitioned as Promary & Logical within Extended and partition formated as FAT32
but when i check HDD in Partition Magic 8 it display 32GB
before some days i used 76GB out of 80Gb
The Stifler
--
Bugs can neither be created nor be removed from software by a developer. They can only be converted from one form to another. The total number of bugs in the software always remain constant.
|
|
|
|
|
Windows XP will not create FAT32 partitions bigger than 32GB. You have to use a different tool to do so. The reason is that Microsoft believe that NTFS is a better choice on larger drives.
If you really need to use FAT32 on this drive you could try this FAT32 Formatter[^] which supports >32GB partitions.
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
Is it possible to get the data back from a partition that was replaced by a ghost image.I have a partition that got mistakenly replace by a image that was supposed to be of another partition.Anybody have any experience
_________________________
"When the superior man refrains from acting, his force is felt for a thousand li." Sun Tzu
|
|
|
|
|
It depends on a few factors:
-Was the image applied at a sector level? (eg, the image contains 100mb of an OS and 10GB of blank data (00), does it even overwite with blank data?)
-Did the image overwrite the sectors of data which you want to recover?
Im sure you wont be able to use any simple 1-click program to revert the whole partition, but what you could start with is using some hex editor to open the drive (as a physical medium, and or the partition itself) and then look at the sectors for data which may be recoverable, and of interest, an example of such a program is WinHex.
Thats a start...
However how to go from there depends on alot of factors, such as type of data, file system of the original and new partition, OSes and so on...
Hope you can atleast get something back...
//Johannes
|
|
|
|
|
Nope. You overwrote the data in the partition beyond any hope of getting it back.
|
|
|
|
|
IF the data you need to recover was actually overwritten (as opposed to being marked as empty) you're almost certainly going to have to contact a professional firm and pay $big bucks. Even if it wasn't overwritten trying to examine it via a hex editor to find missing files isn't viable unless you only need to recover a handful of non binary files. Unfortunately you're more or less SOL.
Otherwise [Microsoft is] toast in the long term no matter how much money they've got. They would be already if the Linux community didn't have it's head so firmly up it's own command line buffer that it looks like taking 15 years to find the desktop.
-- Matthew Faithfull
|
|
|
|