David Härdeman wrote:

On Sun, Mar 28, 2010 at 09:51:17PM -0300, Mauro Carvalho Chehab wrote:

quoted

I spoke too soon... removing the index causes a problem at the read ioctl: there's no way
to retrieve just the non-sparsed values.

There's one solution that would allow both read/write and compat to work nicely,
but the API would become somewhat asymmetrical:

At get (EVIOCGKEYCODEBIG):
	use index/len as input and keycode/scancode as output;

At set (EVIOCSKEYCODEBIG):
	use scancode/keycode/len as input (and, optionally, index as output).

This was exactly the approach I had in mind when I suggested using 
indexes.

Doesn't work perfectly. The asymmetry has a side effect on the internal logic: 

EVIOCGKEYCODEBIG should be implemented with a pseudo-code like:
	kt_entry = getkeycodebig_from_index(index);

EVIOCSKEYCODEBIG should be implemented with a pseudo-code like:
	kt_entry = getkeycodebig_from_scan(scan, len);
	old_key = kt_entry->keycode;

	kt_entry->keycode = newkey;
	if (setkeycodebig(kt_entry) == 0)
		keyup(old_key);

As you see, the input parameters for the getkeycodebig*() are different.

So, this approach requires 3 ops instead of 2. Yet, as scancode->keycode is
needed anyway, this doesn't actually hurts.

I just added the patches that implement those two ioctls on my IR development tree.
I tested only the original EVIOCGKEYCODE/EVIOSGKEYCODE and calling a clear_table
function using EVIOCSKEYCODEBIG via emulation.

My next step is to test the remaining ir-keytable functions via emulation, and then
implement the *BIG ioctls at ir-core, for testing.

I haven't test yet the *keycode*default methods. 

After having it fully tested, I'll submit the complete input ioctl patch via ML.

-- 

Cheers,
Mauro