X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fmesa%2Fdrivers%2Fdri%2Fi965%2Fbrw_optimize.c;h=cbed2bd5cb168effaa621d884b65825d3bbcf006;hb=ed24e19070b7dff12670151b2d184f31c845ccae;hp=e79b3ddea3502d9b8e798a9955efa8d3d51ff76d;hpb=8f3bdeaad610d7d5a5c6e73e1e9c721219595754;p=mesa.git

diff --git a/src/mesa/drivers/dri/i965/brw_optimize.c b/src/mesa/drivers/dri/i965/brw_optimize.c
index e79b3ddea35..cbed2bd5cb1 100644
--- a/src/mesa/drivers/dri/i965/brw_optimize.c
+++ b/src/mesa/drivers/dri/i965/brw_optimize.c
@@ -26,12 +26,609 @@
  */
 
 #include "main/macros.h"
-#include "shader/program.h"
-#include "shader/prog_print.h"
+#include "program/program.h"
+#include "program/prog_print.h"
 #include "brw_context.h"
 #include "brw_defines.h"
 #include "brw_eu.h"
 
+const struct brw_instruction_info brw_opcodes[128] = {
+    [BRW_OPCODE_MOV] = { .name = "mov", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_FRC] = { .name = "frc", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_RNDU] = { .name = "rndu", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_RNDD] = { .name = "rndd", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_RNDE] = { .name = "rnde", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_RNDZ] = { .name = "rndz", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_NOT] = { .name = "not", .nsrc = 1, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_LZD] = { .name = "lzd", .nsrc = 1, .ndst = 1 },
+
+    [BRW_OPCODE_MUL] = { .name = "mul", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_MAC] = { .name = "mac", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_MACH] = { .name = "mach", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_LINE] = { .name = "line", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_PLN] = { .name = "pln", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_SAD2] = { .name = "sad2", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_SADA2] = { .name = "sada2", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_DP4] = { .name = "dp4", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_DPH] = { .name = "dph", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_DP3] = { .name = "dp3", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_DP2] = { .name = "dp2", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_MATH] = { .name = "math", .nsrc = 2, .ndst = 1 },
+
+    [BRW_OPCODE_AVG] = { .name = "avg", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_ADD] = { .name = "add", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_SEL] = { .name = "sel", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_AND] = { .name = "and", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_OR] = { .name = "or", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_XOR] = { .name = "xor", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_SHR] = { .name = "shr", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_SHL] = { .name = "shl", .nsrc = 2, .ndst = 1, .is_arith = 1 },
+    [BRW_OPCODE_ASR] = { .name = "asr", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_CMP] = { .name = "cmp", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_CMPN] = { .name = "cmpn", .nsrc = 2, .ndst = 1 },
+
+    [BRW_OPCODE_SEND] = { .name = "send", .nsrc = 1, .ndst = 1 },
+    [BRW_OPCODE_NOP] = { .name = "nop", .nsrc = 0, .ndst = 0 },
+    [BRW_OPCODE_JMPI] = { .name = "jmpi", .nsrc = 1, .ndst = 0 },
+    [BRW_OPCODE_IF] = { .name = "if", .nsrc = 2, .ndst = 0 },
+    [BRW_OPCODE_IFF] = { .name = "iff", .nsrc = 2, .ndst = 1 },
+    [BRW_OPCODE_WHILE] = { .name = "while", .nsrc = 2, .ndst = 0 },
+    [BRW_OPCODE_ELSE] = { .name = "else", .nsrc = 2, .ndst = 0 },
+    [BRW_OPCODE_BREAK] = { .name = "break", .nsrc = 2, .ndst = 0 },
+    [BRW_OPCODE_CONTINUE] = { .name = "cont", .nsrc = 1, .ndst = 0 },
+    [BRW_OPCODE_HALT] = { .name = "halt", .nsrc = 1, .ndst = 0 },
+    [BRW_OPCODE_MSAVE] = { .name = "msave", .nsrc = 1, .ndst = 1 },
+    [BRW_OPCODE_PUSH] = { .name = "push", .nsrc = 1, .ndst = 1 },
+    [BRW_OPCODE_MRESTORE] = { .name = "mrest", .nsrc = 1, .ndst = 1 },
+    [BRW_OPCODE_POP] = { .name = "pop", .nsrc = 2, .ndst = 0 },
+    [BRW_OPCODE_WAIT] = { .name = "wait", .nsrc = 1, .ndst = 0 },
+    [BRW_OPCODE_DO] = { .name = "do", .nsrc = 0, .ndst = 0 },
+    [BRW_OPCODE_ENDIF] = { .name = "endif", .nsrc = 2, .ndst = 0 },
+};
+
+static INLINE
+GLboolean brw_is_arithmetic_inst(const struct brw_instruction *inst)
+{
+   return brw_opcodes[inst->header.opcode].is_arith;
+}
+
+static const GLuint inst_stride[7] = {
+    [0] = 0,
+    [1] = 1,
+    [2] = 2,
+    [3] = 4,
+    [4] = 8,
+    [5] = 16,
+    [6] = 32
+};
+
+static const GLuint inst_type_size[8] = {
+    [BRW_REGISTER_TYPE_UD] = 4,
+    [BRW_REGISTER_TYPE_D] = 4,
+    [BRW_REGISTER_TYPE_UW] = 2,
+    [BRW_REGISTER_TYPE_W] = 2,
+    [BRW_REGISTER_TYPE_UB] = 1,
+    [BRW_REGISTER_TYPE_B] = 1,
+    [BRW_REGISTER_TYPE_F] = 4
+};
+
+static INLINE GLboolean
+brw_is_grf_written(const struct brw_instruction *inst,
+                   int reg_index, int size,
+                   int gen)
+{
+   if (brw_opcodes[inst->header.opcode].ndst == 0)
+      return GL_FALSE;
+
+   if (inst->bits1.da1.dest_address_mode != BRW_ADDRESS_DIRECT)
+      if (inst->bits1.ia1.dest_reg_file == BRW_GENERAL_REGISTER_FILE)
+         return GL_TRUE;
+
+   if (inst->bits1.da1.dest_reg_file != BRW_GENERAL_REGISTER_FILE)
+      return GL_FALSE;
+
+   const int reg_start = reg_index * REG_SIZE;
+   const int reg_end = reg_start + size;
+
+   const int type_size = inst_type_size[inst->bits1.da1.dest_reg_type];
+   const int write_start = inst->bits1.da1.dest_reg_nr*REG_SIZE
+                         + inst->bits1.da1.dest_subreg_nr;
+   int length, write_end;
+
+   /* SEND is specific */
+   if (inst->header.opcode == BRW_OPCODE_SEND) {
+      if (gen >= 5)
+         length = inst->bits3.generic_gen5.response_length*REG_SIZE;
+      else 
+         length = inst->bits3.generic.response_length*REG_SIZE;
+   }
+   else {
+      length = 1 << inst->header.execution_size;
+      length *= type_size;
+      length *= inst->bits1.da1.dest_horiz_stride;
+   }
+
+   /* If the two intervals intersect, we overwrite the register */
+   write_end = write_start + length;
+   const int left = MAX2(write_start, reg_start);
+   const int right = MIN2(write_end, reg_end);
+
+   return left < right;
+}
+
+static GLboolean
+brw_is_mrf_written_alu(const struct brw_instruction *inst,
+		       int reg_index, int size)
+{
+   if (brw_opcodes[inst->header.opcode].ndst == 0)
+      return GL_FALSE;
+
+   if (inst->bits1.da1.dest_reg_file != BRW_MESSAGE_REGISTER_FILE)
+      return GL_FALSE;
+
+   if (inst->bits1.da1.dest_address_mode != BRW_ADDRESS_DIRECT)
+      return GL_TRUE;
+
+   const int reg_start = reg_index * REG_SIZE;
+   const int reg_end = reg_start + size;
+
+   const int mrf_index = inst->bits1.da1.dest_reg_nr & 0x0f;
+   const int is_compr4 = inst->bits1.da1.dest_reg_nr & BRW_MRF_COMPR4;
+   const int type_size = inst_type_size[inst->bits1.da1.dest_reg_type];
+
+   /* We use compr4 with a size != 16 elements. Strange, we conservatively
+    * consider that we are writing the register.
+    */
+   if (is_compr4 && inst->header.execution_size != BRW_EXECUTE_16)
+      return GL_TRUE;
+
+   /* Here we write mrf_{i} and mrf_{i+4}. So we read two times 8 elements */
+   if (is_compr4) {
+      const int length = 8 * type_size * inst->bits1.da1.dest_horiz_stride;
+
+      /* First 8-way register */
+      const int write_start0 = mrf_index*REG_SIZE
+                             + inst->bits1.da1.dest_subreg_nr;
+      const int write_end0 = write_start0 + length;
+
+      /* Second 8-way register */
+      const int write_start1 = (mrf_index+4)*REG_SIZE
+                             + inst->bits1.da1.dest_subreg_nr;
+      const int write_end1 = write_start1 + length;
+
+      /* If the two intervals intersect, we overwrite the register */
+      const int left0 = MAX2(write_start0, reg_start);
+      const int right0 = MIN2(write_end0, reg_end);
+      const int left1 = MAX2(write_start1, reg_start);
+      const int right1 = MIN2(write_end1, reg_end);
+
+      if (left0 < right0 || left1 < right1)
+	 return GL_TRUE;
+   }
+   else {
+      int length;
+      length = 1 << inst->header.execution_size;
+      length *= type_size;
+      length *= inst->bits1.da1.dest_horiz_stride;
+
+      /* If the two intervals intersect, we write into the register */
+      const int write_start = inst->bits1.da1.dest_reg_nr*REG_SIZE
+                            + inst->bits1.da1.dest_subreg_nr;
+      const int write_end = write_start + length;
+      const int left = MAX2(write_start, reg_start);
+      const int right = MIN2(write_end, reg_end);
+
+      if (left < right)
+	 return GL_TRUE;
+   }
+
+   return GL_FALSE;
+}
+
+/* SEND may perform an implicit mov to a mrf register */
+static GLboolean brw_is_mrf_written_send(const struct brw_instruction *inst,
+					 int reg_index, int size)
+{
+
+   const int reg_start = reg_index * REG_SIZE;
+   const int reg_end = reg_start + size;
+   const int mrf_start = inst->header.destreg__conditionalmod;
+   const int write_start = mrf_start * REG_SIZE;
+   const int write_end = write_start + REG_SIZE;
+   const int left = MAX2(write_start, reg_start);
+   const int right = MIN2(write_end, reg_end);
+
+   if (inst->header.opcode != BRW_OPCODE_SEND ||
+       inst->bits1.da1.src0_reg_file == 0)
+      return GL_FALSE;
+
+   return left < right;
+}
+
+/* Specific path for message register since we need to handle the compr4 case */
+static INLINE GLboolean
+brw_is_mrf_written(const struct brw_instruction *inst, int reg_index, int size)
+{
+   return (brw_is_mrf_written_alu(inst, reg_index, size) ||
+	   brw_is_mrf_written_send(inst, reg_index, size));
+}
+
+static INLINE GLboolean
+brw_is_mrf_read(const struct brw_instruction *inst,
+                int reg_index, int size, int gen)
+{
+   if (inst->header.opcode != BRW_OPCODE_SEND)
+      return GL_FALSE;
+   if (inst->bits2.da1.src0_address_mode != BRW_ADDRESS_DIRECT)
+      return GL_TRUE;
+
+   const int reg_start = reg_index*REG_SIZE;
+   const int reg_end = reg_start + size;
+
+   int length, read_start, read_end;
+   if (gen >= 5)
+      length = inst->bits3.generic_gen5.msg_length*REG_SIZE;
+   else 
+      length = inst->bits3.generic.msg_length*REG_SIZE;
+
+   /* Look if SEND uses an implicit mov. In that case, we read one less register
+    * (but we write it)
+    */
+   if (inst->bits1.da1.src0_reg_file != 0)
+      read_start = inst->header.destreg__conditionalmod;
+   else {
+      length--;
+      read_start = inst->header.destreg__conditionalmod + 1;
+   }
+   read_start *= REG_SIZE;
+   read_end = read_start + length;
+
+   const int left = MAX2(read_start, reg_start);
+   const int right = MIN2(read_end, reg_end);
+
+   return left < right;
+}
+
+static INLINE GLboolean
+brw_is_grf_read(const struct brw_instruction *inst, int reg_index, int size)
+{
+   int i, j;
+   if (brw_opcodes[inst->header.opcode].nsrc == 0)
+      return GL_FALSE;
+
+   /* Look at first source. We must take into account register regions to
+    * monitor carefully the read. Note that we are a bit too conservative here
+    * since we do not take into account the fact that some complete registers
+    * may be skipped
+    */
+   if (brw_opcodes[inst->header.opcode].nsrc >= 1) {
+
+      if (inst->bits2.da1.src0_address_mode != BRW_ADDRESS_DIRECT)
+         if (inst->bits1.ia1.src0_reg_file == BRW_GENERAL_REGISTER_FILE)
+            return GL_TRUE;
+      if (inst->bits1.da1.src0_reg_file != BRW_GENERAL_REGISTER_FILE)
+         return GL_FALSE;
+
+      const int reg_start = reg_index*REG_SIZE;
+      const int reg_end = reg_start + size;
+
+      /* See if at least one of this element intersects the interval */
+      const int type_size = inst_type_size[inst->bits1.da1.src0_reg_type];
+      const int elem_num = 1 << inst->header.execution_size;
+      const int width = 1 << inst->bits2.da1.src0_width;
+      const int row_num = elem_num >> inst->bits2.da1.src0_width;
+      const int hs = type_size*inst_stride[inst->bits2.da1.src0_horiz_stride];
+      const int vs = type_size*inst_stride[inst->bits2.da1.src0_vert_stride];
+      int row_start = inst->bits2.da1.src0_reg_nr*REG_SIZE
+                    + inst->bits2.da1.src0_subreg_nr;
+      for (j = 0; j < row_num; ++j) {
+         int write_start = row_start;
+         for (i = 0; i < width; ++i) {
+            const int write_end = write_start + type_size;
+            const int left = write_start > reg_start ? write_start : reg_start;
+            const int right = write_end < reg_end ? write_end : reg_end;
+            if (left < right)
+               return GL_TRUE;
+            write_start += hs;
+         }
+         row_start += vs;
+      }
+   }
+
+   /* Second src register */
+   if (brw_opcodes[inst->header.opcode].nsrc >= 2) {
+
+      if (inst->bits3.da1.src1_address_mode != BRW_ADDRESS_DIRECT)
+         if (inst->bits1.ia1.src1_reg_file == BRW_GENERAL_REGISTER_FILE)
+            return GL_TRUE;
+      if (inst->bits1.da1.src1_reg_file != BRW_GENERAL_REGISTER_FILE)
+         return GL_FALSE;
+
+      const int reg_start = reg_index*REG_SIZE;
+      const int reg_end = reg_start + size;
+
+      /* See if at least one of this element intersects the interval */
+      const int type_size = inst_type_size[inst->bits1.da1.src1_reg_type];
+      const int elem_num = 1 << inst->header.execution_size;
+      const int width = 1 << inst->bits3.da1.src1_width;
+      const int row_num = elem_num >> inst->bits3.da1.src1_width;
+      const int hs = type_size*inst_stride[inst->bits3.da1.src1_horiz_stride];
+      const int vs = type_size*inst_stride[inst->bits3.da1.src1_vert_stride];
+      int row_start = inst->bits3.da1.src1_reg_nr*REG_SIZE
+                    + inst->bits3.da1.src1_subreg_nr;
+      for (j = 0; j < row_num; ++j) {
+         int write_start = row_start;
+         for (i = 0; i < width; ++i) {
+            const int write_end = write_start + type_size;
+            const int left = write_start > reg_start ? write_start : reg_start;
+            const int right = write_end < reg_end ? write_end : reg_end;
+            if (left < right)
+               return GL_TRUE;
+            write_start += hs;
+         }
+         row_start += vs;
+      }
+   }
+
+   return GL_FALSE;
+}
+
+static INLINE GLboolean
+brw_is_control_done(const struct brw_instruction *mov) {
+   return
+       mov->header.dependency_control != 0 ||
+       mov->header.thread_control != 0 ||
+       mov->header.mask_control != 0 ||
+       mov->header.saturate != 0 ||
+       mov->header.debug_control != 0;
+}
+
+static INLINE GLboolean
+brw_is_predicated(const struct brw_instruction *mov) {
+   return mov->header.predicate_control != 0;
+}
+
+static INLINE GLboolean
+brw_is_grf_to_mrf_mov(const struct brw_instruction *mov,
+                      int *mrf_index,
+                      int *grf_index,
+                      GLboolean *is_compr4)
+{
+   if (brw_is_predicated(mov) ||
+       brw_is_control_done(mov) ||
+       mov->header.debug_control != 0)
+      return GL_FALSE;
+
+   if (mov->bits1.da1.dest_address_mode != BRW_ADDRESS_DIRECT ||
+       mov->bits1.da1.dest_reg_file != BRW_MESSAGE_REGISTER_FILE ||
+       mov->bits1.da1.dest_reg_type != BRW_REGISTER_TYPE_F ||
+       mov->bits1.da1.dest_horiz_stride != BRW_HORIZONTAL_STRIDE_1 ||
+       mov->bits1.da1.dest_subreg_nr != 0)
+      return GL_FALSE;
+
+   if (mov->bits2.da1.src0_address_mode != BRW_ADDRESS_DIRECT ||
+       mov->bits1.da1.src0_reg_file != BRW_GENERAL_REGISTER_FILE ||
+       mov->bits1.da1.src0_reg_type != BRW_REGISTER_TYPE_F ||
+       mov->bits2.da1.src0_width != BRW_WIDTH_8 ||
+       mov->bits2.da1.src0_horiz_stride != BRW_HORIZONTAL_STRIDE_1 ||
+       mov->bits2.da1.src0_vert_stride != BRW_VERTICAL_STRIDE_8 ||
+       mov->bits2.da1.src0_subreg_nr != 0 ||
+       mov->bits2.da1.src0_abs != 0 ||
+       mov->bits2.da1.src0_negate != 0)
+      return GL_FALSE;
+
+   *grf_index = mov->bits2.da1.src0_reg_nr;
+   *mrf_index = mov->bits1.da1.dest_reg_nr & 0x0f;
+   *is_compr4 = (mov->bits1.da1.dest_reg_nr & BRW_MRF_COMPR4) != 0;
+   return GL_TRUE;
+}
+
+static INLINE GLboolean
+brw_is_grf_straight_write(const struct brw_instruction *inst, int grf_index)
+{
+   /* remark: no problem to predicate a SEL instruction */
+   if ((!brw_is_predicated(inst) || inst->header.opcode == BRW_OPCODE_SEL) &&
+       brw_is_control_done(inst) == GL_FALSE &&
+       inst->header.execution_size == 4 &&
+       inst->header.access_mode == BRW_ALIGN_1 &&
+       inst->bits1.da1.dest_address_mode == BRW_ADDRESS_DIRECT &&
+       inst->bits1.da1.dest_reg_file == BRW_GENERAL_REGISTER_FILE &&
+       inst->bits1.da1.dest_reg_type == BRW_REGISTER_TYPE_F &&
+       inst->bits1.da1.dest_horiz_stride == BRW_HORIZONTAL_STRIDE_1 &&
+       inst->bits1.da1.dest_reg_nr == grf_index &&
+       inst->bits1.da1.dest_subreg_nr == 0 &&
+       brw_is_arithmetic_inst(inst))
+      return GL_TRUE;
+
+   return GL_FALSE;
+}
+
+static INLINE GLboolean
+brw_inst_are_equal(const struct brw_instruction *src0,
+                   const struct brw_instruction *src1)
+{
+   const GLuint *field0 = (GLuint *) src0;
+   const GLuint *field1 = (GLuint *) src1;
+   return field0[0] == field1[0] &&
+          field0[1] == field1[1] &&
+          field0[2] == field1[2] &&
+          field0[3] == field1[3];
+}
+
+static INLINE void
+brw_inst_copy(struct brw_instruction *dst,
+              const struct brw_instruction *src)
+{
+   GLuint *field_dst = (GLuint *) dst;
+   const GLuint *field_src = (GLuint *) src;
+   field_dst[0] = field_src[0];
+   field_dst[1] = field_src[1];
+   field_dst[2] = field_src[2];
+   field_dst[3] = field_src[3];
+}
+
+static void brw_remove_inst(struct brw_compile *p, const GLboolean *removeInst)
+{
+   int i, nr_insn = 0, to = 0, from = 0;
+
+   for (from = 0; from < p->nr_insn; ++from) {
+      if (removeInst[from])
+         continue;
+      if(to != from)
+         brw_inst_copy(p->store + to, p->store + from);
+      to++;
+   }
+
+   for (i = 0; i < p->nr_insn; ++i)
+      if (removeInst[i] == GL_FALSE)
+         nr_insn++;
+   p->nr_insn = nr_insn;
+}
+
+/* The gen code emitter generates a lot of duplications in the
+ * grf-to-mrf moves, for example when texture sampling with the same
+ * coordinates from multiple textures..  Here, we monitor same mov
+ * grf-to-mrf instrutions and remove repeated ones where the operands
+ * and dst ahven't changed in between.
+ */
+void brw_remove_duplicate_mrf_moves(struct brw_compile *p)
+{
+   const int gen = p->brw->intel.gen;
+   int i, j;
+
+   GLboolean *removeInst = calloc(sizeof(GLboolean), p->nr_insn);
+   for (i = 0; i < p->nr_insn; i++) {
+      if (removeInst[i])
+         continue;
+
+      const struct brw_instruction *mov = p->store + i;
+      int mrf_index, grf_index;
+      GLboolean is_compr4;
+
+      /* Only consider _straight_ grf-to-mrf moves */
+      if (!brw_is_grf_to_mrf_mov(mov, &mrf_index, &grf_index, &is_compr4))
+         continue;
+
+      const int mrf_index0 = mrf_index;
+      const int mrf_index1 = is_compr4 ? mrf_index0+4 : mrf_index0+1;
+      const int simd16_size = 2 * REG_SIZE;
+
+      for (j = i + 1; j < p->nr_insn; j++) {
+         const struct brw_instruction *inst = p->store + j;
+
+         if (brw_inst_are_equal(mov, inst)) {
+            removeInst[j] = GL_TRUE;
+            continue;
+         }
+
+         if (brw_is_grf_written(inst, grf_index, simd16_size, gen) ||
+             brw_is_mrf_written(inst, mrf_index0, REG_SIZE) ||
+             brw_is_mrf_written(inst, mrf_index1, REG_SIZE))
+            break;
+      }
+   }
+
+   brw_remove_inst(p, removeInst);
+   free(removeInst);
+}
+
+/* Replace moves to MRFs where the value moved is the result of a
+ * normal arithmetic operation with computation right into the MRF.
+ */
+void brw_remove_grf_to_mrf_moves(struct brw_compile *p)
+{
+   int i, j, prev;
+   struct brw_context *brw = p->brw;
+   const int gen = brw->intel.gen;
+   const int simd16_size = 2*REG_SIZE;
+
+   GLboolean *removeInst = calloc(sizeof(GLboolean), p->nr_insn);
+   assert(removeInst);
+
+   for (i = 0; i < p->nr_insn; i++) {
+      if (removeInst[i])
+         continue;
+
+      struct brw_instruction *grf_inst = NULL;
+      const struct brw_instruction *mov = p->store + i;
+      int mrf_index, grf_index;
+      GLboolean is_compr4;
+
+      /* Only consider _straight_ grf-to-mrf moves */
+      if (!brw_is_grf_to_mrf_mov(mov, &mrf_index, &grf_index, &is_compr4))
+         continue;
+
+      /* Using comp4 enables a stride of 4 for this instruction */
+      const int mrf_index0 = mrf_index;
+      const int mrf_index1 = is_compr4 ? mrf_index+4 : mrf_index+1;
+
+      /* Look where the register has been set */
+      prev = i;
+      GLboolean potential_remove = GL_FALSE;
+      while (prev--) {
+
+         /* If _one_ instruction writes the grf, we try to remove the mov */
+         struct brw_instruction *inst = p->store + prev;
+         if (brw_is_grf_straight_write(inst, grf_index)) {
+            potential_remove = GL_TRUE;
+            grf_inst = inst;
+            break;
+         }
+
+      }
+
+      if (potential_remove == GL_FALSE)
+         continue;
+      removeInst[i] = GL_TRUE;
+
+      /* Monitor first the section of code between the grf computation and the
+       * mov. Here we cannot read or write both mrf and grf register
+       */
+      for (j = prev + 1; j < i; ++j) {
+         struct brw_instruction *inst = p->store + j;
+         if (removeInst[j])
+            continue;
+         if (brw_is_grf_written(inst, grf_index, simd16_size, gen)   ||
+             brw_is_grf_read(inst, grf_index, simd16_size)           ||
+             brw_is_mrf_written(inst, mrf_index0, REG_SIZE)   ||
+             brw_is_mrf_written(inst, mrf_index1, REG_SIZE)   ||
+             brw_is_mrf_read(inst, mrf_index0, REG_SIZE, gen) ||
+             brw_is_mrf_read(inst, mrf_index1, REG_SIZE, gen)) {
+            removeInst[i] = GL_FALSE;
+            break;
+         }
+      }
+
+      /* After the mov, we can read or write the mrf. If the grf is overwritten,
+       * we are done
+       */
+      for (j = i + 1; j < p->nr_insn; ++j) {
+         struct brw_instruction *inst = p->store + j;
+         if (removeInst[j])
+            continue;
+
+         if (brw_is_grf_read(inst, grf_index, simd16_size)) {
+            removeInst[i] = GL_FALSE;
+            break;
+         }
+
+         if (brw_is_grf_straight_write(inst, grf_index))
+            break;
+      }
+
+      /* Note that with the top down traversal, we can safely pacth the mov
+       * instruction
+       */
+      if (removeInst[i]) {
+         grf_inst->bits1.da1.dest_reg_file = mov->bits1.da1.dest_reg_file;
+         grf_inst->bits1.da1.dest_reg_nr = mov->bits1.da1.dest_reg_nr;
+      }
+   }
+
+   brw_remove_inst(p, removeInst);
+   free(removeInst);
+}
+
 static GLboolean
 is_single_channel_dp4(struct brw_instruction *insn)
 {