intel/fs: Add support for copy-propagating a block of multiple FIXED_GRFs.

In cases where a LOAD_PAYLOAD instruction copies a single block of
sequential GRF registers into the destination (see
is_identity_payload()), splitting the block copy into a number of ACP
entries (one for each LOAD_PAYLOAD source) is undesirable, because
that prevents copy propagation into any instructions which read
multiple components at once with the same source (the barycentric
source of the LINTERP instruction is going to be the overwhelmingly
most common example).

Technically it would also be possible to do this for VGRF sources, but
there is little benefit from that since register coalesce already
covers many of those cases -- There is no way for a block of
FIXED_GRFs to be coalesced into a VGRF though.

This prevents the following shader-db regressions (including SIMD32
programs) in combination with the interpolation rework part of this
series.  On SKL:

   total instructions in shared programs: 18595160 -> 18828562 (1.26%)
   instructions in affected programs: 13374946 -> 13608348 (1.75%)
   helped: 7
   HURT: 108977

   total spills in shared programs: 9116 -> 9106 (-0.11%)
   spills in affected programs: 404 -> 394 (-2.48%)
   helped: 7
   HURT: 9

   total fills in shared programs: 8994 -> 9176 (2.02%)
   fills in affected programs: 898 -> 1080 (20.27%)
   helped: 7
   HURT: 9

   LOST:   469
   GAINED: 220

On SNB:

   total instructions in shared programs: 13996898 -> 14096222 (0.71%)
   instructions in affected programs: 8088546 -> 8187870 (1.23%)
   helped: 2
   HURT: 66520

   total spills in shared programs: 2985 -> 2961 (-0.80%)
   spills in affected programs: 632 -> 608 (-3.80%)
   helped: 2
   HURT: 0

   total fills in shared programs: 3144 -> 3128 (-0.51%)
   fills in affected programs: 1515 -> 1499 (-1.06%)
   helped: 2
   HURT: 0

   LOST:   0
   GAINED: 4

Reviewed-by: Kenneth Graunke <[email protected]>

1 files changed, 7 insertions, 5 deletions

diff --git a/src/intel/compiler/brw_fs_copy_propagation.cpp b/src/intel/compiler/brw_fs_copy_propagation.cpp
index adf0f2fefff..4f5493e4f43 100644
--- a/src/intel/compiler/brw_fs_copy_propagation.cpp
+++ b/src/intel/compiler/brw_fs_copy_propagation.cpp
@@ -48,8 +48,8 @@ struct acp_entry : public exec_node {
    fs_reg dst;
    fs_reg src;
    unsigned global_idx;
-   uint8_t size_written;
-   uint8_t size_read;
+   unsigned size_written;
+   unsigned size_read;
    enum opcode opcode;
    bool saturate;
 };
@@ -895,7 +895,8 @@ can_propagate_from(fs_inst *inst)
             (inst->src[0].file == FIXED_GRF &&
              inst->src[0].is_contiguous())) &&
            inst->src[0].type == inst->dst.type &&
-           !inst->is_partial_write());
+           !inst->is_partial_write()) ||
+          is_identity_payload(FIXED_GRF, inst);
 }
 
 /* Walks a basic block and does copy propagation on it using the acp
@@ -948,11 +949,12 @@ fs_visitor::opt_copy_propagation_local(void *copy_prop_ctx, bblock_t *block,
        * operand of another instruction, add it to the ACP.
        */
       if (can_propagate_from(inst)) {
-         acp_entry *entry = ralloc(copy_prop_ctx, acp_entry);
+         acp_entry *entry = rzalloc(copy_prop_ctx, acp_entry);
          entry->dst = inst->dst;
          entry->src = inst->src[0];
          entry->size_written = inst->size_written;
-         entry->size_read = inst->size_read(0);
+         for (unsigned i = 0; i < inst->sources; i++)
+            entry->size_read += inst->size_read(i);
          entry->opcode = inst->opcode;
          entry->saturate = inst->saturate;
          acp[entry->dst.nr % ACP_HASH_SIZE].push_tail(entry);