intel/fs: Skip registers faster when setting spill costs

This might be slightly faster since we're doing one read rather than
two before we decide to skip.  The more important reason, however, is
because no_spill prevents us from re-spilling spill registers.  In the
new world in which we don't re-calculate liveness every spill, we may
not have valid liveness for spill registers so we shouldn't even look
their live ranges up.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110825
Fixes: e99081e76d4 "intel/fs/ra: Spill without destroying the..."
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Tested-by: Tapani Pälli <tapani.palli@intel.com>

diff --git a/src/intel/compiler/brw_fs_reg_allocate.cpp b/src/intel/compiler/brw_fs_reg_allocate.cpp
index 33631ddcd0ca3c7041af49c4acc3bc69f31c5c50..6a594a4f520e7afe3ba1c031adb6cc9590c264e3 100644 (file)
--- a/src/intel/compiler/brw_fs_reg_allocate.cpp
+++ b/src/intel/compiler/brw_fs_reg_allocate.cpp
@@ -922,6 +922,15 @@ fs_reg_alloc::set_spill_costs()
    }
 
    for (unsigned i = 0; i < fs->alloc.count; i++) {
+      /* Do the no_spill check first.  Registers that are used as spill
+       * temporaries may have been allocated after we calculated liveness so
+       * we shouldn't look their liveness up.  Fortunately, they're always
+       * used in SCRATCH_READ/WRITE instructions so they'll always be flagged
+       * no_spill.
+       */
+      if (no_spill[i])
+         continue;
+
       int live_length = fs->virtual_grf_end[i] - fs->virtual_grf_start[i];
       if (live_length <= 0)
          continue;
@@ -934,8 +943,7 @@ fs_reg_alloc::set_spill_costs()
        * to spill medium length registers with more uses.
        */
       float adjusted_cost = spill_costs[i] / logf(live_length);
-      if (!no_spill[i])
-        ra_set_node_spill_cost(g, first_vgrf_node + i, adjusted_cost);
+      ra_set_node_spill_cost(g, first_vgrf_node + i, adjusted_cost);
    }
 
    have_spill_costs = true;