[Mesa-dev] GLSL swizzle lowering

[Ian Romanick]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Marek Ol?ák wrote:
> On Tue, Aug 31, 2010 at 2:14 AM, Ian Romanick <idr at freedesktop.org
> <mailto:idr at freedesktop.org>> wrote:
> 
>     While I was trying to get one of the Humus demos working today, it
>     occurred to me that we can possibly do better than
>     ir_vec_index_to_cond_assign to lower variable indexing of vectors.  In
>     addition to using conditional assignment, we can also use a dot-product
>     to pick a single element out of a vector.  The variable index operation
>     becomes:
> 
>     const vec4 gl_vec_selector[4] =
>        vec4[4](vec4(1.0, 0.0, 0.0, 0.0),
>                vec4(0.0, 1.0, 0.0, 0.0),
>                vec4(0.0, 0.0, 1.0, 0.0),
>                vec4(0.0, 0.0, 0.0, 1.0));
> 
>     ...
> 
>     float f = dot(v, gl_vec_selector[i]);

[snip]

> Neither r300 nor r500 supports the ARL opcode in fragment shaders (it's

Meh.  I always forget about that asymmetry.

> a D3D10 feature), which kind of makes this optimization a no-go. I
> suggest using SEQ instead:
> 
> bvec4 selector = equal(vec4(i), vec4(0,1,2,3));
> float f = dot(v, vec4(selector));
> 
> which should end up being just SEQ followed by DP4.

SEQ isn't part of the ARB_fragment_program / ARB_vertex_program
instruction set either.  Does R300 support that?  I won't be surprised
if i915 doesn't.  Of course, it doesn't support the ARL-based
optimization either.

SEQ is itself lowered to a sequence of instructions:

	SGE	t0, i.xxxx, { 0,  1,  2,  3};
	SGE	t1, i.xxxx, {-0, -1, -2, -3};
	MUL	selector, t0, t1;
	DP4	result, v, selector;

So, that probably still would be better than the mess that gets
generated today.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkx9OfMACgkQX1gOwKyEAw+HlQCfZCqRdGwn3x9k06/qReehqMSf
nCoAnjKMPoUlb/4QXeHW0EYOozlLpQn2
=X6R+
-----END PGP SIGNATURE-----