Getting started with a wlan driver
in [Device Drivers]
|
From: Thomas F. Divine on 8 Feb 2010 10:08 Let me know if this works for you... Thomas F Divine http://www.pcausa.com "djMax" <djMax(a)discussions.microsoft.com> wrote in message news:8B59AEA3-BEF0-41FD-86BB-C839AA103D0E(a)microsoft.com... > Do the open source Linux drivers change this equation at all? I'm dealing > with constrained hardware - namely the RT73 line of chips, which I know on > the Linux side support all they need to. Wouldn't that code be the same > (conceptually) on the Windows side? > > "Thomas F. Divine" wrote: > >> The approach of treating an 802.11 adapter hardware as a non-network >> device >> is the solution for developing active network diagnostic tools on the >> Windows platform. >> >> Unfortunately, adapter vendors use development techniques and have >> business >> models that prevent this approach from being practical. On the Windows >> platform (as well as on others...) adapter vendors use firmware and >> "hardware abstraction" DLLs that 1.) make it easy to do a ordinary things >> and 2.) make it difficult to do anything else. This approach effectively >> hides much of the functionality that is needed for active diagnostics. >> >> To follow this approach you will need to work with adapter vendors and >> obtain their proprietary hardware and software interface descriptions. >> They >> don't provide the information that is needed for you to follow this >> approach. >> >> Good luck, >> >> Thomas F Divine >> http://www.pcausa.com >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> news:D82741B7-EBCD-4E8A-9940-DA2FB42295FF(a)microsoft.com... >> > Makes painful sense. In my case I don't even want the USB device to >> > work >> > for >> > "normal" wifi. Could I just treat it as an arbitrary piece of hardware >> > rather than a network card? Or as a non-wireless network card? >> > >> > "Thomas F. Divine" wrote: >> > >> >> This goal may not be achievable on the Windows platform. >> >> >> >> Generally the system 802.11 miniports and adapters are operating in >> >> the >> >> "Extensible Station" (ExSTA) mode. In this mode station operation is >> >> under >> >> control of the Microsoft Native Wi-Fi software. >> >> >> >> Microsoft has decided to provide a crippled capability for ISVs to >> >> control >> >> the Native 802.11 miniports. This is called "Network Monitor Operation >> >> Mode" >> >> (NetMon). Believe me, they have locked this down so that the term >> >> "monitor" >> >> has real meaning. In particular, from the WDK: >> >> >> >> "While in NetMon mode, the miniport driver can only receive packets >> >> based >> >> on >> >> the current packet filter settings. The driver cannot send packets >> >> either >> >> on >> >> its own or through a call to its MiniportSendNetBufferLists function" >> >> >> >> I don't see the practical possibility of sending raw 802.11 packets as >> >> being >> >> achievable because of Microsoft's decision in this regards. >> >> >> >> In fact, just getting NetMon mode to work properly is a chore - at >> >> least >> >> for >> >> me. I have some pre-Alpha executables at: >> >> >> >> http://www.pcausa.com/WlanExplorer >> >> >> >> Thomas F. Divine >> >> http://www.pcausa.com >> >> >> >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> >> news:462AD494-39C6-42F6-9152-06885B36C9DB(a)microsoft.com... >> >> > I'm an experienced programmer (15 yrs, C, Java, C++, C#, etc) but >> >> > I've >> >> > not >> >> > done any driver development. I'm looking to write a driver that >> >> > sends >> >> > raw >> >> > 802.11 packets. Might anyone be kind enough to point me to the best >> >> > way >> >> > to >> >> > get started? (Perhaps a related sample, or an open source driver of >> >> > some >> >> > sort) >> >> > >> >> > I have installed the WDK and taken a brief look at the network >> >> > samples, >> >> > but >> >> > they don't seem low level enough. >> >> > >> >> > Thanks! >> >>
From: djMax on 8 Feb 2010 12:51 So using winusb and libusbdotnet I've been able to connect to the card from userspace. Now the question is what to say to it. :) "Thomas F. Divine" wrote: > Let me know if this works for you... > > Thomas F Divine > http://www.pcausa.com > > "djMax" <djMax(a)discussions.microsoft.com> wrote in message > news:8B59AEA3-BEF0-41FD-86BB-C839AA103D0E(a)microsoft.com... > > Do the open source Linux drivers change this equation at all? I'm dealing > > with constrained hardware - namely the RT73 line of chips, which I know on > > the Linux side support all they need to. Wouldn't that code be the same > > (conceptually) on the Windows side? > > > > "Thomas F. Divine" wrote: > > > >> The approach of treating an 802.11 adapter hardware as a non-network > >> device > >> is the solution for developing active network diagnostic tools on the > >> Windows platform. > >> > >> Unfortunately, adapter vendors use development techniques and have > >> business > >> models that prevent this approach from being practical. On the Windows > >> platform (as well as on others...) adapter vendors use firmware and > >> "hardware abstraction" DLLs that 1.) make it easy to do a ordinary things > >> and 2.) make it difficult to do anything else. This approach effectively > >> hides much of the functionality that is needed for active diagnostics. > >> > >> To follow this approach you will need to work with adapter vendors and > >> obtain their proprietary hardware and software interface descriptions. > >> They > >> don't provide the information that is needed for you to follow this > >> approach. > >> > >> Good luck, > >> > >> Thomas F Divine > >> http://www.pcausa.com > >> > >> > >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message > >> news:D82741B7-EBCD-4E8A-9940-DA2FB42295FF(a)microsoft.com... > >> > Makes painful sense. In my case I don't even want the USB device to > >> > work > >> > for > >> > "normal" wifi. Could I just treat it as an arbitrary piece of hardware > >> > rather than a network card? Or as a non-wireless network card? > >> > > >> > "Thomas F. Divine" wrote: > >> > > >> >> This goal may not be achievable on the Windows platform. > >> >> > >> >> Generally the system 802.11 miniports and adapters are operating in > >> >> the > >> >> "Extensible Station" (ExSTA) mode. In this mode station operation is > >> >> under > >> >> control of the Microsoft Native Wi-Fi software. > >> >> > >> >> Microsoft has decided to provide a crippled capability for ISVs to > >> >> control > >> >> the Native 802.11 miniports. This is called "Network Monitor Operation > >> >> Mode" > >> >> (NetMon). Believe me, they have locked this down so that the term > >> >> "monitor" > >> >> has real meaning. In particular, from the WDK: > >> >> > >> >> "While in NetMon mode, the miniport driver can only receive packets > >> >> based > >> >> on > >> >> the current packet filter settings. The driver cannot send packets > >> >> either > >> >> on > >> >> its own or through a call to its MiniportSendNetBufferLists function" > >> >> > >> >> I don't see the practical possibility of sending raw 802.11 packets as > >> >> being > >> >> achievable because of Microsoft's decision in this regards. > >> >> > >> >> In fact, just getting NetMon mode to work properly is a chore - at > >> >> least > >> >> for > >> >> me. I have some pre-Alpha executables at: > >> >> > >> >> http://www.pcausa.com/WlanExplorer > >> >> > >> >> Thomas F. Divine > >> >> http://www.pcausa.com > >> >> > >> >> > >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message > >> >> news:462AD494-39C6-42F6-9152-06885B36C9DB(a)microsoft.com... > >> >> > I'm an experienced programmer (15 yrs, C, Java, C++, C#, etc) but > >> >> > I've > >> >> > not > >> >> > done any driver development. I'm looking to write a driver that > >> >> > sends > >> >> > raw > >> >> > 802.11 packets. Might anyone be kind enough to point me to the best > >> >> > way > >> >> > to > >> >> > get started? (Perhaps a related sample, or an open source driver of > >> >> > some > >> >> > sort) > >> >> > > >> >> > I have installed the WDK and taken a brief look at the network > >> >> > samples, > >> >> > but > >> >> > they don't seem low level enough. > >> >> > > >> >> > Thanks! > >> >>
From: Thomas F. Divine on 8 Feb 2010 13:24 Exactly. What to say... Thomas F Divine "djMax" <djMax(a)discussions.microsoft.com> wrote in message news:2504ACAA-30DA-4428-80E2-208661D1BE03(a)microsoft.com... > So using winusb and libusbdotnet I've been able to connect to the card > from > userspace. Now the question is what to say to it. :) > > "Thomas F. Divine" wrote: > >> Let me know if this works for you... >> >> Thomas F Divine >> http://www.pcausa.com >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> news:8B59AEA3-BEF0-41FD-86BB-C839AA103D0E(a)microsoft.com... >> > Do the open source Linux drivers change this equation at all? I'm >> > dealing >> > with constrained hardware - namely the RT73 line of chips, which I know >> > on >> > the Linux side support all they need to. Wouldn't that code be the >> > same >> > (conceptually) on the Windows side? >> > >> > "Thomas F. Divine" wrote: >> > >> >> The approach of treating an 802.11 adapter hardware as a non-network >> >> device >> >> is the solution for developing active network diagnostic tools on the >> >> Windows platform. >> >> >> >> Unfortunately, adapter vendors use development techniques and have >> >> business >> >> models that prevent this approach from being practical. On the Windows >> >> platform (as well as on others...) adapter vendors use firmware and >> >> "hardware abstraction" DLLs that 1.) make it easy to do a ordinary >> >> things >> >> and 2.) make it difficult to do anything else. This approach >> >> effectively >> >> hides much of the functionality that is needed for active diagnostics. >> >> >> >> To follow this approach you will need to work with adapter vendors and >> >> obtain their proprietary hardware and software interface descriptions. >> >> They >> >> don't provide the information that is needed for you to follow this >> >> approach. >> >> >> >> Good luck, >> >> >> >> Thomas F Divine >> >> http://www.pcausa.com >> >> >> >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> >> news:D82741B7-EBCD-4E8A-9940-DA2FB42295FF(a)microsoft.com... >> >> > Makes painful sense. In my case I don't even want the USB device to >> >> > work >> >> > for >> >> > "normal" wifi. Could I just treat it as an arbitrary piece of >> >> > hardware >> >> > rather than a network card? Or as a non-wireless network card? >> >> > >> >> > "Thomas F. Divine" wrote: >> >> > >> >> >> This goal may not be achievable on the Windows platform. >> >> >> >> >> >> Generally the system 802.11 miniports and adapters are operating in >> >> >> the >> >> >> "Extensible Station" (ExSTA) mode. In this mode station operation >> >> >> is >> >> >> under >> >> >> control of the Microsoft Native Wi-Fi software. >> >> >> >> >> >> Microsoft has decided to provide a crippled capability for ISVs to >> >> >> control >> >> >> the Native 802.11 miniports. This is called "Network Monitor >> >> >> Operation >> >> >> Mode" >> >> >> (NetMon). Believe me, they have locked this down so that the term >> >> >> "monitor" >> >> >> has real meaning. In particular, from the WDK: >> >> >> >> >> >> "While in NetMon mode, the miniport driver can only receive packets >> >> >> based >> >> >> on >> >> >> the current packet filter settings. The driver cannot send packets >> >> >> either >> >> >> on >> >> >> its own or through a call to its MiniportSendNetBufferLists >> >> >> function" >> >> >> >> >> >> I don't see the practical possibility of sending raw 802.11 packets >> >> >> as >> >> >> being >> >> >> achievable because of Microsoft's decision in this regards. >> >> >> >> >> >> In fact, just getting NetMon mode to work properly is a chore - at >> >> >> least >> >> >> for >> >> >> me. I have some pre-Alpha executables at: >> >> >> >> >> >> http://www.pcausa.com/WlanExplorer >> >> >> >> >> >> Thomas F. Divine >> >> >> http://www.pcausa.com >> >> >> >> >> >> >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> >> >> news:462AD494-39C6-42F6-9152-06885B36C9DB(a)microsoft.com... >> >> >> > I'm an experienced programmer (15 yrs, C, Java, C++, C#, etc) but >> >> >> > I've >> >> >> > not >> >> >> > done any driver development. I'm looking to write a driver that >> >> >> > sends >> >> >> > raw >> >> >> > 802.11 packets. Might anyone be kind enough to point me to the >> >> >> > best >> >> >> > way >> >> >> > to >> >> >> > get started? (Perhaps a related sample, or an open source driver >> >> >> > of >> >> >> > some >> >> >> > sort) >> >> >> > >> >> >> > I have installed the WDK and taken a brief look at the network >> >> >> > samples, >> >> >> > but >> >> >> > they don't seem low level enough. >> >> >> > >> >> >> > Thanks! >> >> >>
From: Thomas F. Divine on 8 Feb 2010 16:35 The WIndows 7 WDK (7600.16385.0) Includes sources to two Native 802.11 NDIS miniports. One is USB. See the usbnwifi sample. The other is a PCI-based 802.11 miniport. See the athwifi sample. Thomas F. Divine "djMax" <djMax(a)discussions.microsoft.com> wrote in message news:2504ACAA-30DA-4428-80E2-208661D1BE03(a)microsoft.com... > So using winusb and libusbdotnet I've been able to connect to the card > from > userspace. Now the question is what to say to it. :) > > "Thomas F. Divine" wrote: > >> Let me know if this works for you... >> >> Thomas F Divine >> http://www.pcausa.com >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> news:8B59AEA3-BEF0-41FD-86BB-C839AA103D0E(a)microsoft.com... >> > Do the open source Linux drivers change this equation at all? I'm >> > dealing >> > with constrained hardware - namely the RT73 line of chips, which I know >> > on >> > the Linux side support all they need to. Wouldn't that code be the >> > same >> > (conceptually) on the Windows side? >> > >> > "Thomas F. Divine" wrote: >> > >> >> The approach of treating an 802.11 adapter hardware as a non-network >> >> device >> >> is the solution for developing active network diagnostic tools on the >> >> Windows platform. >> >> >> >> Unfortunately, adapter vendors use development techniques and have >> >> business >> >> models that prevent this approach from being practical. On the Windows >> >> platform (as well as on others...) adapter vendors use firmware and >> >> "hardware abstraction" DLLs that 1.) make it easy to do a ordinary >> >> things >> >> and 2.) make it difficult to do anything else. This approach >> >> effectively >> >> hides much of the functionality that is needed for active diagnostics. >> >> >> >> To follow this approach you will need to work with adapter vendors and >> >> obtain their proprietary hardware and software interface descriptions. >> >> They >> >> don't provide the information that is needed for you to follow this >> >> approach. >> >> >> >> Good luck, >> >> >> >> Thomas F Divine >> >> http://www.pcausa.com >> >> >> >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> >> news:D82741B7-EBCD-4E8A-9940-DA2FB42295FF(a)microsoft.com... >> >> > Makes painful sense. In my case I don't even want the USB device to >> >> > work >> >> > for >> >> > "normal" wifi. Could I just treat it as an arbitrary piece of >> >> > hardware >> >> > rather than a network card? Or as a non-wireless network card? >> >> > >> >> > "Thomas F. Divine" wrote: >> >> > >> >> >> This goal may not be achievable on the Windows platform. >> >> >> >> >> >> Generally the system 802.11 miniports and adapters are operating in >> >> >> the >> >> >> "Extensible Station" (ExSTA) mode. In this mode station operation >> >> >> is >> >> >> under >> >> >> control of the Microsoft Native Wi-Fi software. >> >> >> >> >> >> Microsoft has decided to provide a crippled capability for ISVs to >> >> >> control >> >> >> the Native 802.11 miniports. This is called "Network Monitor >> >> >> Operation >> >> >> Mode" >> >> >> (NetMon). Believe me, they have locked this down so that the term >> >> >> "monitor" >> >> >> has real meaning. In particular, from the WDK: >> >> >> >> >> >> "While in NetMon mode, the miniport driver can only receive packets >> >> >> based >> >> >> on >> >> >> the current packet filter settings. The driver cannot send packets >> >> >> either >> >> >> on >> >> >> its own or through a call to its MiniportSendNetBufferLists >> >> >> function" >> >> >> >> >> >> I don't see the practical possibility of sending raw 802.11 packets >> >> >> as >> >> >> being >> >> >> achievable because of Microsoft's decision in this regards. >> >> >> >> >> >> In fact, just getting NetMon mode to work properly is a chore - at >> >> >> least >> >> >> for >> >> >> me. I have some pre-Alpha executables at: >> >> >> >> >> >> http://www.pcausa.com/WlanExplorer >> >> >> >> >> >> Thomas F. Divine >> >> >> http://www.pcausa.com >> >> >> >> >> >> >> >> >> "djMax" <djMax(a)discussions.microsoft.com> wrote in message >> >> >> news:462AD494-39C6-42F6-9152-06885B36C9DB(a)microsoft.com... >> >> >> > I'm an experienced programmer (15 yrs, C, Java, C++, C#, etc) but >> >> >> > I've >> >> >> > not >> >> >> > done any driver development. I'm looking to write a driver that >> >> >> > sends >> >> >> > raw >> >> >> > 802.11 packets. Might anyone be kind enough to point me to the >> >> >> > best >> >> >> > way >> >> >> > to >> >> >> > get started? (Perhaps a related sample, or an open source driver >> >> >> > of >> >> >> > some >> >> >> > sort) >> >> >> > >> >> >> > I have installed the WDK and taken a brief look at the network >> >> >> > samples, >> >> >> > but >> >> >> > they don't seem low level enough. >> >> >> > >> >> >> > Thanks! >> >> >>
From: Pavel A. on 8 Feb 2010 19:12
"djMax" <djMax(a)discussions.microsoft.com> wrote in message news:462AD494-39C6-42F6-9152-06885B36C9DB(a)microsoft.com... > I'm an experienced programmer (15 yrs, C, Java, C++, C#, etc) but I've not > done any driver development. I'm looking to write a driver that sends raw > 802.11 packets. Might anyone be kind enough to point me to the best way > to > get started? (Perhaps a related sample, or an open source driver of some > sort) The *best* way to get started? a good consultant. Good luck, --pa |