Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Fix IHEX firmware generation/loading | Marc Zyngier | 2008-08-02 | 1 | -1/+1 |
| | | | | | | | | | Fix both the IHEX firmware generation (len field always null, and EOF marker a byte too short) and loading (struct ihex_binrec needs to be packed to reflect the on-disk structure). Signed-off-by: Marc Zyngier <maz@misterjones.org> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com> | ||||
* | ihex: request_ihex_firmware() function to load and validate firmware | David Woodhouse | 2008-07-10 | 1 | -0/+24 |
| | | | | | | | Provide a helper to load the file and validate it in one call, to simplify error handling in the drivers which are going to use it. Signed-off-by: David Woodhouse <dwmw2@infradead.org> | ||||
* | ihex.h: binary representation of ihex records | David Woodhouse | 2008-07-10 | 1 | -0/+50 |
Some devices need their firmware as a set of {address, len, data...} records in some specific order rather than a simple blob. The normal way of doing this kind of thing is 'ihex', which is a text format and not entirely suitable for use in the kernel. This provides a binary representation which is very similar, but much more compact -- and a helper routine to skip to the next record, because the alignment constraints mean that everybody will screw it up for themselves otherwise. Also a helper function which can verify that a 'struct firmware' contains a valid set of ihex records, and that following them won't run off the end of the loaded data. Signed-off-by: David Woodhouse <dwmw2@infradead.org> |