Bruce
Thu Jul 10 11:17:39 PDT 2008
Thanks for letting us know.
--
Bruce Eitman (eMVP)
Senior Engineer
Bruce.Eitman AT EuroTech DOT com
My BLOG
http://geekswithblogs.net/bruceeitman
EuroTech Inc.
www.EuroTech.com
"Skin Diver" <me@home.com> wrote in message
news:67844CB1-FD71-4158-BEE1-5780ECBF53DE@microsoft.com...
> Removing the API calls solved the problem.
>
>
>
> "Bruce Eitman [eMVP]" wrote:
>
>> Actually, it is a problem. We were surprised also, but Microsoft was
>> very
>> clear with us that we shouldn't use the bib file compress flag on
>> drivers.
>> Removing the C from the bib file resolved the problem.
>>
>> I assume that you mean the "Flags" value in the registry for the driver,
>> that was cleared to zero.
>>
>> Based on Skin Diver's responses though the more likely problem is calling
>> system APIs in power handler functions, as K.S. points out. We didn't
>> have
>> that problem.
>>
>> --
>> Bruce Eitman (eMVP)
>> Senior Engineer
>> Bruce.Eitman AT EuroTech DOT com
>> My BLOG
http://geekswithblogs.net/bruceeitman
>>
>> EuroTech Inc.
>> www.EuroTech.com
>>
>> "Dean Ramsier" <ramsiernospam@nospam.com> wrote in message
>> news:eXTGk0o4IHA.1192@TK2MSFTNGP05.phx.gbl...
>> >I don't think this (compression) should ever be a reason for getting a
>> >page
>> >fault in a driver. Drivers are loaded with LoadDriver() which keeps
>> >them
>> >completely in memory regardless of compression settings. This would
>> >only
>> >be an issue if the registry flag is set that causes the device manager
>> >to
>> >use LoadLibrary() instead of LoadDriver().
>> >
>> > --
>> > Dean Ramsier - eMVP
>> > BSQUARE Corporation
>> >
>> >
>> > "Bruce Eitman [eMVP]" <bruce.eitman.nospam@EuroTech.com.nospam> wrote
>> > in
>> > message news:OxA$Ztg4IHA.1952@TK2MSFTNGP03.phx.gbl...
>> >> Sorry, I thought that you had been working with Platform Builder long
>> >> enough to be familiar with bib files.
>> >>
>> >> I should ask this, are you targeting Windows CE 6.0? I think that you
>> >> are, based on the problem.
>> >>
>> >> The driver that is having problems, let's call it MyDriver.dll, must
>> >> be
>> >> in one of your bib files. But let's not worry about that too much
>> >> right
>> >> now. It ends up in ce.bib in your _FLATRELEASEDIR. So look in your
>> >> ce.bib file for the the driver, it will look something like:
>> >>
>> >> MyDriver.dll C:\WINCE600\OSDESIGNS\.....\MyDriver.dll NK
>> >> SHCK
>> >>
>> >> If it does in fact have the C within those last leters after NK, then
>> >> find MyDriver in one of your bib files where it will look like:
>> >>
>> >> MyDriver.dll $(_FLATRELEASEDIR)\MyDriver.dll NK SHCK
>> >>
>> >> Edit the bib file in your FILES folder and remove the C:
>> >>
>> >> MyDriver.dll $(_FLATRELEASEDIR)\MyDriver.dll NK SHK
>> >>
>> >> copy the file to your _FLATRELEASEDIR, and run makeimg. Now see if
>> >> the
>> >> problem goes away.
>> >>
>> >> --
>> >> Bruce Eitman (eMVP)
>> >> Senior Engineer
>> >> Bruce.Eitman AT EuroTech DOT com
>> >> My BLOG
http://geekswithblogs.net/bruceeitman
>> >>
>> >> EuroTech Inc.
>> >> www.EuroTech.com
>> >>
>> >> "Skin Diver" <me@home.com> wrote in message
>> >> news:eGmVqfg4IHA.4488@TK2MSFTNGP03.phx.gbl...
>> >>> Thanks for the info.
>> >>>
>> >>> my config.bib has the following ... which seems to be pretty much the
>> >>> default for all the example BSP's included with PB.
>> >>>
>> >>> ; @CESYSGEN IF !NK_NKNOCOMP
>> >>>
>> >>> COMPRESSION=ON
>> >>>
>> >>> ; @CESYSGEN ENDIF !NK_NKNOCOMP
>> >>>
>> >>> ; @CESYSGEN IF NK_NKNOCOMP
>> >>>
>> >>> COMPRESSION=OFF
>> >>>
>> >>> ; @CESYSGEN ENDIF NK_NKNOCOMP
>> >>>
>> >>>
>> >>>
>> >>> I see the "Compression" catalog item is solid square and has a whole
>> >>> bunch of dependent items.
>> >>>
>> >>> I assume the CESYSGEN from above comes from this???
>> >>>
>> >>>
>> >>>
>> >>> Should I just nix the conditionals and just use:
>> >>>
>> >>> COMPRESSION=OFF
>> >>>
>> >>> ???
>> >>>
>> >>> I would think maybe just the "device" process which loads all
>> >>> the"Builtin" drivers needs to be locked in memory?? or am I out in
>> >>> left
>> >>> field on my understanding?
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>> "Bruce Eitman [eMVP]" <bruce.eitman.nospam@EuroTech.com.nospam> wrote
>> >>> in
>> >>> message news:%23rcENwf4IHA.776@TK2MSFTNGP04.phx.gbl...
>> >>>
>> >>>> Is the driver compressed in your bib file? If it is, change that it
>> >>>> should go away.
>> >>>>
>> >>>> --
>> >>>> Bruce Eitman (eMVP)
>> >>>> Senior Engineer
>> >>>> Bruce.Eitman AT EuroTech DOT com
>> >>>> My BLOG
http://geekswithblogs.net/bruceeitman
>> >>>>
>> >>>> EuroTech Inc.
>> >>>> www.EuroTech.com
>> >>>>
>> >>>> "Skin Diver" <me@home.com> wrote in message
>> >>>> news:%23$U01lf4IHA.2060@TK2MSFTNGP02.phx.gbl...
>> >>>>>
>> >>>>>
>> >>>>> I occasionally get "VMProcessPageFault Error: Page Fault occured
>> >>>>> while
>> >>>>> in power handler! Address = 0xxxxxxx"
>> >>>>>
>> >>>>> and it never recovers. It only does it intermittently on a resume
>> >>>>> after a suspend operation.
>> >>>>>
>> >>>>> I narrowed it to my xxx_PowerOn routine in one of my drivers.
>> >>>>>
>> >>>>> It doesn't do much in the PowerOn routine...basically a lock, set a
>> >>>>> member variable, SetEvent(xx) and a unlock.
>> >>>>>
>> >>>>> Why does this happen and how can I prevent it??
>> >>>>>
>> >>>>> I undrstand you might be able to do some kind of page lock ??? How?
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>
>> >>>>
>> >>>
>> >>>
>> >>
>> >>
>> >
>> >
>>
>>
>>