I have digged a bit further... the second patch still needs a sign-off from Martin (or anybody else at Sage) and all of this should be tested on real hardware. I don't trust this code without it... I have tested most of the descriptor handling by using binary images from various vendors and generations, but the consequences on flashrom execution and the whole Silvermont code flow are untested.
Stefan Tauner (3): Refine Flash Component descriptor handling. Add support for Intel Silvermont: Bay Trail, Rangeley and Avoton. Add support for Intel Wildcat Point PCH.
chipset_enable.c | 244 ++++++++++++++++++----- ich_descriptors.c | 188 +++++++++++++---- ich_descriptors.h | 39 +++- ichspi.c | 38 ++-- programmer.h | 2 + util/ich_descriptors_tool/ich_descriptors_tool.c | 11 + 6 files changed, 407 insertions(+), 115 deletions(-)