[Mesa-dev] V2 Step towards dropping more GLSL IR opts

[Timothy Arceri]

This is a collection of patches I've written and collected that
aim to reduce the pain when dropping brw_do_channel_expressions()
and brw_do_vector_splitting() two passes that combined take about
14% of the time it takes to run shader-db. This series includes
Ken's move comparisions series with a couple of small changes.

V2:
- first two clean-up patches added
- added support for conditional helpers to expressions patch 3
 (used by  pathes 8 & 9)
- small update to patch 4
- patch 7 no londer converts to ffma directly
- some patches from last version pushed 

[PATCH 1/9] nir: tidy up swizzle handling in nir_search
[PATCH 2/9] nir: stop passing swizzle to search condition
[PATCH 3/9] nir/algebraic: add support for conditional helper
[PATCH 4/9] nir: Introduce a nir_opt_move_comparisons() pass.
[PATCH 5/9] i965: Move nir_lower_locals_to_regs a bit later.
[PATCH 6/9] i965: Use the nir_move_comparisons pass.
[PATCH 7/9] nir: add imprecise flrp optimisation
[PATCH 8/9] nir: add late opt to turn inot/b2f combos back to bcsel
[PATCH 9/9] nir: don't turn ieq/ine into inot if used by an if