Re: NdisMIndicateReceivePacket for large packets

Tech Tip: Click here to run a free scan for Windows Errors and optimize PC performance



Please be assured that miniport driver developers don't break rules
just of fun. They have reasons.

....... which is not going to make your life any easier if, as NDIS IM
developer, you have a misfortune to come across such miniport driver - out of
1000 clients 999 will be happy about your NDIS IM, and 1 (i.e. the one that
has this "masterpiece" on his machine) will complain. Therefore, it may take
you quite a while to figure out what is going on, and, after you do, I don't
really think that you will readily accept an argument like "I just had a
reason to do things this way" from the developer who wrote this
"masterpiece"....

In general, I believe that drivers should always expose a "proper" interface
on both upper and lower edges, no matter how they do things internally. If
you take this approach, you will never ever come into any conflict either
with the system or with the third-party software - even if internally your
driver is written "improperly". Unfortunately, most of AV/PF writers seem to
have a different opinion on the subject......


Anton Bassov

"Pavel A." wrote:

"Gianluca Varenni" <gianluca.varenni@xxxxxxxxxxxx> wrote in message news:eTzbFBaQHHA.4404@xxxxxxxxxxxxxxxxxxxxxxx

<soviet_bloke@xxxxxxxxxxx> wrote in message news:1169851453.832988.179860@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Eric,

Don't count on anything - even if your test show that everything works
fine......

Tricks like that are the main reason why the same NDIS IM may work
perfectly well with adapter X and fail with adapter Y for no obvious
reason. In other words, please don't do it - you will do a huge favour
to any NDIS-level developer who comes across your driver...

<rant>I'd really like to tell this to a lot of the Wi-Fi miniport driver developers...</rant>


Please be assured that miniport driver developers don't break rules just of fun. They have reasons.
And these reasons often are specific needs of customers, partners
and others we *must* appeace.

In the specific case of indicating up large packets - this may occur with some
drivers that have "wi-fi sniffer mode" - when they indicate raw .11 PDUs.

Regards,
--PA



.



Relevant Pages

  • Re: DropDownList advice needed
    ... Thanks, Peter and Kevin. ... reasons you both pointed out. ... because they say if me and another developer leave, ... >I hope someone in the Classic ASP newsgroup pointed out to you that 10,000 ...
    (microsoft.public.dotnet.framework.aspnet)
  • Re: Why Turbo Delphi for win32 need .NET SDK?
    ... disappointed because Turbo Delphi had .NET pre-requisites, ... An experienced developer might be thinking this way. ... I guess given that the reasons given for why it _cannot_ be removed are ... For the record - I currently cannot install Turbo Delphi.32 because I ...
    (borland.public.delphi.non-technical)
  • Re: Linux v2.6.27-rc1
    ... That may not sound like much, but it's enough to make me worry about ... developer that just has a few patches pending as it works for subsystem ... because for the above two reasons I do not think it can really work that ...
    (Linux-Kernel)
  • Re: Please Help
    ... community but it's far from "the most experienced and knowledgeable ... that don't have anything to do with CDP for various reasons. ... or a well-paid in-house developer with no incentive to rip ... In response to various other posts I reached out to my client today to ...
    (comp.databases.pick)
  • Re: TRICK: methods in ASPX pages with <%%> code blocks
    ... developer, adopt best practices. ... experiences, and by observing the experiences of others, over a long period ... >I don't have solid "evidence" of coding one way or the other. ... but your reasons for not doing just don't make sense to me. ...
    (microsoft.public.dotnet.framework.aspnet)