Commit Graph

4 Commits

Author SHA1 Message Date
Gaurav Shah
8dd20941cd Zero-out allocated buffer while creating kernel preamble header.
Since the kernel vblocks are always padded out to a fixed size, the unused (padded) memory was ending up containing random bytes, leading to vbutil_kernel generating vblocks that are not bit-identical when run with the same inputs.

BUG=none
TEST=see below
1) Use vbutil_kernel with the same set of inputs to generate two alternative vblocks.
2) Compare the 2 files - they must be bitwise identical space. The padding bytes must all be set to 0.
3) Generate a new signed image using the resign_image.sh script under scripts/image_signing. This signed image should be boot and install successfully on a maching running our custom firmware using the same set of root, firmware, and kernel keys.

Review URL: http://codereview.chromium.org/3076033
2010-08-04 16:45:42 -07:00
Randall Spangler
729b87258b Clean up of key block functions
No substantial new code, just making the old code consistent.

Review URL: http://codereview.chromium.org/2729021
2010-06-11 11:16:20 -07:00
Randall Spangler
d55c645372 Utility to pack public key, version, algorithm into a single file in VbPublicKey format
Review URL: http://codereview.chromium.org/2762009
2010-06-10 12:43:51 -07:00
Randall Spangler
d183644564 Major refactoring of structures, with unit tests. This matches the doc I sent out earlier.
Firmware-side code for LoadKernel() is in place now.  LoadFirmware() replacement coming soon.

The new functions are implemented in parallel to the existing ones (i.e., everything that used to work still does).

Review URL: http://codereview.chromium.org/2745007
2010-06-10 09:59:04 -07:00