Age | Commit message (Collapse) | Author |
|
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
structure
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Be consistent with exit status and printing errors to stderr.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Make the output nicer. Do not print the header if a specific block is
requested.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Marius Vlad <marius.c.vlad@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Since we no longer store pointers to previous blocks, we can free them.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Only dump the one matching panel_type by default. Cleans up the output,
and allows the users to get more verbose output if he so chooses.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Now that the parsers for blocks don't depend on the dump ordering, we
can also choose to dump specific sections only.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
The hex dump is useful, just not by default.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
All parsers fetch the information they need without ordering
constraints, so dump all in numerical order. This also makes it
unnecessary to track already dumped blocks.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
On some systems the VBT panel type may be overridden in the opregion,
and we can't necessarily get at that. Let the user specify it on the
command line.
As a byproduct, the section parsing order no longer matters.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Don't mandate a specific ordering on the parsing of the blocks.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
It's just good hygiene.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Allow putting stuff in there instead of a global. A bit like passing
dev_priv in the kernel.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
There's no point in dumping a fixed value.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Not sure it's a great idea to do platform specific parsing of the BIOS,
but at least make it possible to pass in the devid on the command line
and not just the environment.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Keep positional parameter support for entering filename for backwards
compatibility.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
No need for it to be global.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Fix typo on intel_bios_reader.c
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Similar to what's done in kernel. It's a bit artificial that the parsing
and dumping are two separate steps in the userspace tool, but it's
easier to follow and debug the code when both the kernel and userspace
are similar.
v2: don't segfault so much on dumping null pointers
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Reuse the same code as kernel. Also parses v3, although does not
actually dump that stuff yet.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
No need to reinvent wheels, reuse the code from kernel.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
In const we trust.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
The MIPI DSI sequence block v3+ has a separate block size field.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Due to the clever way the whole sequence block is specified without
forward compatibility, it's not possible to dump most blocks without
this.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Try to print something useful and helpful for the user.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Simplify things a lot, make it correct, don't pass random pointers to
free() on errors, etc.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Bail out on v3+, we don't support that just yet.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
Decode the MIPI [sic] device type.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Each bit in the device type is supposed to mean something. Decode their
meaning.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Check that the offset where expect to find the device id is withing the
BIOS image, instead of accessing whatever (if anything) happens to be
there.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
New stuff has been added to the end of the child device block at various
times, so using a hardcoded size for the block is a bad idea.
Fortunately the size of the block is listed in the VBT just before the
blocks themselves, so grab it from there.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
On BXT the SSC reference frequency is fixed 100MHz.
Signed-off-by: Imre Deak <imre.deak@intel.com>
|
|
VLV/CHV has a fixed 100MHz SSC reference frequency.
Signed-off-by: Imre Deak <imre.deak@intel.com>
|
|
Signed-off-by: Gaurav K Singh <gaurav.k.singh@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
|
|
Signed-off-by: Gaurav K Singh <gaurav.k.singh@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
|
|
With the header cleanup we can now give this header a suitable name,
since it now really only contains register access and other I/O
functions and assorted definitions.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
|
|
Only include what the header itself needs. The big fish here is
intel-gpu-tools.h. More will follow.
One ugly thing removed here is the duplicated GEN6_TD_CTL #define, one
of which was broken.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
|
|
We still need to dump some of the known sections explicitly due to
dependencies on information extracted, such as LFP data pointers and
panel_type.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Dump sections through a table based on the section id.
Hex dump the section. This works also for unknown sections.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Running intel_bios_reader upon itself causes the reader to crash and
burn. It obviously finds a VBT signature inside the binary, but then
does not rigorously check that all data blocks are valid before
dereferencing them.
Reported-by: Emanuel Bronshtein
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45205
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|