At present binman requires that the Intel descriptor has an explicit offset. Generally this is 0 since the descriptor is at the start of the image. Add a default to handle this, so users don't need to specify the offset.
Signed-off-by: Simon Glass sjg@chromium.org ---
tools/binman/bsection.py | 8 ++++++++ tools/binman/etype/intel_descriptor.py | 2 ++ 2 files changed, 10 insertions(+)
diff --git a/tools/binman/bsection.py b/tools/binman/bsection.py index e0e3707db41..49b8ef3e3e0 100644 --- a/tools/binman/bsection.py +++ b/tools/binman/bsection.py @@ -477,6 +477,14 @@ class Section(object): return self._parent_section.GetRootSkipAtStart() return self._skip_at_start
+ def GetStartOffset(self): + """Get the start offset for this section + + Returns: + The first available offset in this section (typically 0) + """ + return self._skip_at_start + def GetImageSize(self): """Get the size of the image containing this section
diff --git a/tools/binman/etype/intel_descriptor.py b/tools/binman/etype/intel_descriptor.py index 9deb8dcf42c..661063457ed 100644 --- a/tools/binman/etype/intel_descriptor.py +++ b/tools/binman/etype/intel_descriptor.py @@ -47,6 +47,8 @@ class Entry_intel_descriptor(Entry_blob): def __init__(self, section, etype, node): Entry_blob.__init__(self, section, etype, node) self._regions = [] + if self.offset is None: + self.offset = self.section.GetStartOffset()
def GetOffsets(self): offset = self.data.find(FD_SIGNATURE)