Re: FDF Spec question?

["Gao, Liming" <liming.gao@intel.com>]

Andrew:
  I am not clear your usage. Could you use source section and compressed section for the data before and after compression?  Here, section means PI EFI section format file. Take Lzma compression as the example,  source data is the section stream. It may be combined of more than one sections, such as two FV sections. OVMF FVMAIN_COMPACT is just a case. Compressed section is the guided section. It is GUIDED section header + compressed data. Here, there is only one GUIDED section header. 

  For the below case,  BaseTools will add RAW section header for the input newform.bin, then call the matched GUIDed tool to process this raw section to get the processed data, last append GUIDed section header before the processed data as the output GUIDed section. 
  SECTION GUIDED 4C2B8C75-C6F6-11E6-B483-B8E8562CBAFA  PROCESSING_REQUIRED = TRUE {
    SECTION RAW = $(WORKSPACE)/MyPackage/MyNewType/Binary/newform.bin
  }

Thanks
Liming
> -----Original Message-----
> From: edk2-devel [mailto:edk2-devel-bounces@lists.01.org] On Behalf Of
> Andrew Fish
> Sent: Thursday, December 22, 2016 9:38 AM
> To: Gao, Liming <liming.gao@intel.com>
> Cc: edk2-devel <edk2-devel@lists.01.org>
> Subject: Re: [edk2] FDF Spec question?
> 
> 
> > On Dec 21, 2016, at 5:18 PM, Gao, Liming <liming.gao@intel.com> wrote:
> >
> > Andrew:
> >  You mean newform.bin has section header. It is just a leaf section. You
> expect to directly add it into GUIDED section without specify section header
> again. Right?
> >
> 
> Liming,
> 
> Yes.
> 
> The issue is Leaf section header in the FDF is part of the thing that is getting
> compressed by the build tool. So if you skip that compression step you don't
> really have the section header, as all you really have is the compressed data.
> When you decompress the data (process the encapsulation section) it starts
> with the section header (it could be added pragmatically or just part of the
> original data). So if you skip the compression phase it seems you want to just
> point directly to the encapsulated data (the result).
> 
> Basically my current syntax is pre-pending an unused leaf section header at
> the beginning of the compressed data. The decompressor has to index
> sizeof(EFI_COMMON_SECTION_HEADER) into the Source buffer to find the
> header of the compressed data to process. The size of the compressed data
> is already represented in the encapsulation section, so this extra leaf section
> is an artifact of the FDF syntax.
> 
> Thanks,
> 
> Andrew Fish
> 
> 
> > Thanks
> > Liming
> > -----Original Message-----
> > From: edk2-devel [mailto:edk2-devel-bounces@lists.01.org] On Behalf Of
> Andrew Fish
> > Sent: Thursday, December 22, 2016 1:29 AM
> > To: Gao, Liming <liming.gao@intel.com>
> > Cc: edk2-devel <edk2-devel@lists.01.org>
> > Subject: Re: [edk2] FDF Spec question?
> >
> >
> >> On Dec 21, 2016, at 6:27 AM, Gao, Liming <liming.gao@intel.com> wrote:
> >>
> >> Andrew:
> >> Encapsulation section data is the section stream, not raw format. Then,
> Leaf section can be extracted from encapsulation section.
> >>
> >
> > Liming,
> >
> > I happen to have the section stream in binary form checked into the source
> tree. The section extraction code actually dynamically creates the section
> header(s) for the stream and this means the section header in the FV is not
> really used (the section extraction code actually just skips over it).
> >
> > Technically speaking only the post processed GUID'ed section needs to
> produce the leaf. My case is kind of like I compressed the data without the
> section header. The section header is redundant data from the section
> extraction codes point of view as the binary format already has that info.
> >
> > Thanks,
> >
> > Andrew Fish
> >
> >> Thanks
> >> Liming
> >> -----Original Message-----
> >> From: edk2-devel [mailto:edk2-devel-bounces@lists.01.org] On Behalf Of
> Andrew Fish
> >> Sent: Wednesday, December 21, 2016 8:27 AM
> >> To: edk2-devel <edk2-devel@lists.01.org>
> >> Subject: [edk2] FDF Spec question?
> >>
> >> Is it possible in the FDF syntax to make the payload of a
> EFI_SECTION_GUID_DEFINED (Encapsulating section) a raw binary file?
> >>
> >> I can make the contents of the GUID'ed encapsulation section a section.
> But when I'm processing the GUID'ed section I just end up skipping over the
> extra 4 byte section header so it is unused space.
> >>
> >> SECTION GUIDED 4C2B8C75-C6F6-11E6-B483-B8E8562CBAFA
> PROCESSING_REQUIRED = TRUE {
> >>   SECTION RAW =
> $(WORKSPACE)/MyPackage/MyNewType/Binary/newform.bin
> >> }
> >>
> >> There is a syntax to make a leaf section a binary file.
> >>
> >> SECTION SUBTYPE_GUID AFC13561-9A65-4754-9C93-E133B3B8767C =
> $(WORKSPACE)/MyPackage/MyNewType/Binary/newform.bin
> >>
> >>
> >> Thanks,
> >>
> >> Andrew Fish
> >> _______________________________________________
> >> edk2-devel mailing list
> >> edk2-devel@lists.01.org
> >> https://lists.01.org/mailman/listinfo/edk2-devel
> >> _______________________________________________
> >> edk2-devel mailing list
> >> edk2-devel@lists.01.org
> >> https://lists.01.org/mailman/listinfo/edk2-devel
> >
> > _______________________________________________
> > edk2-devel mailing list
> > edk2-devel@lists.01.org
> > https://lists.01.org/mailman/listinfo/edk2-devel
> 
> _______________________________________________
> edk2-devel mailing list
> edk2-devel@lists.01.org
> https://lists.01.org/mailman/listinfo/edk2-devel