> On Feb 4, 2016, at 2:35 PM, Estelle Yeh <[email protected]> wrote: > > Thanks for the tips, Andrew. > > - It seems to be using DuetPkg/Library/DuetTimerLib/X86TimerLib.c, and I get > this from the report file: > c:\edk2\DuetPkg\Library\DuetTimerLib\DuetTimerLib.inf > {TimerLib} >
That seems a bit strange. A Duet driver assumes it was boot on a system that already had firmware run. So that might be your problem. https://github.com/tianocore/edk2/blob/master/DuetPkg/ReadMe.txt > - I added the "INF MdeModulePkg/Universal/TimestampDxe/TimestampDxe.inf" > entry to the .fdf but I still get "Not found" returned from LocateProtocol > (using %r in a Print). Does it matter where that TimestampDxe.inf entry is > placed in the .dsc and .fdf files? You have to do both. DSC builds, FDF puts it in FV/FD. If you remove the DSC entry it may seem to work, but after you do your 1st clean it will no longer work. > Who should be calling TimestampDriverInitialize()? Better question: How can I > make sure that my platform is producing that protocol? I don't see any listed > in the report file. > If it is in a FV that the DXE Core knows about it will get dispatched. That driver has a Dependency Expression (depex) of TRUE so it should dispatch early in DXE. You should see DEBUG prints from the DXE core about what is getting dispatched. It will either show the driver name or the FILE_GUID from the INF C10194E7-DEB2-4AF4-9EEE-BFFDE4D7D4C7 Thanks, Andrew Fish > Thanks, > > Estelle _______________________________________________ edk2-devel mailing list [email protected] https://lists.01.org/mailman/listinfo/edk2-devel
