ilo: add ilo_render_emit_launch_grid()

ilo_render_emit_launch_grid() emits all the hardware states needed for a
launch_grid() call.

Signed-off-by: Chia-I Wu <[email protected]>

8 files changed, 621 insertions, 2 deletions

diff --git a/src/gallium/drivers/ilo/Makefile.sources b/src/gallium/drivers/ilo/Makefile.sources
index 52f4ff25987..85ce3df931d 100644
--- a/src/gallium/drivers/ilo/Makefile.sources
+++ b/src/gallium/drivers/ilo/Makefile.sources
@@ -40,6 +40,7 @@ C_SOURCES := \
 	ilo_render_dynamic.c \
 	ilo_render_gen6.c \
 	ilo_render_gen7.c \
+	ilo_render_media.c \
 	ilo_render_surface.c \
 	ilo_screen.c \
 	ilo_screen.h \
diff --git a/src/gallium/drivers/ilo/ilo_builder_render.h b/src/gallium/drivers/ilo/ilo_builder_render.h
index f8a983883ac..aa09c877f62 100644
--- a/src/gallium/drivers/ilo/ilo_builder_render.h
+++ b/src/gallium/drivers/ilo/ilo_builder_render.h
@@ -57,8 +57,17 @@ gen6_PIPELINE_SELECT(struct ilo_builder *builder, int pipeline)
 
    ILO_DEV_ASSERT(builder->dev, 6, 7.5);
 
-   /* 3D or media */
-   assert(pipeline == 0x0 || pipeline == 0x1);
+   switch (pipeline) {
+   case GEN6_PIPELINE_SELECT_DW0_SELECT_3D:
+   case GEN6_PIPELINE_SELECT_DW0_SELECT_MEDIA:
+      break;
+   case GEN7_PIPELINE_SELECT_DW0_SELECT_GPGPU:
+      assert(ilo_dev_gen(builder->dev) >= ILO_GEN(7));
+      break;
+   default:
+      assert(!"unknown pipeline");
+      break;
+   }
 
    ilo_builder_batch_write(builder, cmd_len, &dw0);
 }
diff --git a/src/gallium/drivers/ilo/ilo_render.c b/src/gallium/drivers/ilo/ilo_render.c
index 945d4cdb5aa..00b57b808ad 100644
--- a/src/gallium/drivers/ilo/ilo_render.c
+++ b/src/gallium/drivers/ilo/ilo_render.c
@@ -456,3 +456,42 @@ ilo_render_emit_draw(struct ilo_render *render,
 
    draw_session_end(render, vec, &session);
 }
+
+int
+ilo_render_get_launch_grid_len(const struct ilo_render *render,
+                               const struct ilo_state_vector *vec)
+{
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   return ilo_render_get_launch_grid_surface_states_len(render, vec) +
+          ilo_render_get_launch_grid_dynamic_states_len(render, vec) +
+          ilo_render_get_launch_grid_commands_len(render, vec);
+}
+
+void
+ilo_render_emit_launch_grid(struct ilo_render *render,
+                            const struct ilo_state_vector *vec,
+                            const unsigned thread_group_offset[3],
+                            const unsigned thread_group_dim[3],
+                            unsigned thread_group_size,
+                            const struct pipe_constant_buffer *input,
+                            uint32_t pc)
+{
+   struct ilo_render_launch_grid_session session;
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   assert(input->buffer);
+
+   memset(&session, 0, sizeof(session));
+
+   session.thread_group_offset = thread_group_offset;
+   session.thread_group_dim = thread_group_dim;
+   session.thread_group_size = thread_group_size;
+   session.input = input;
+   session.pc = pc;
+
+   ilo_render_emit_launch_grid_surface_states(render, vec, &session);
+   ilo_render_emit_launch_grid_dynamic_states(render, vec, &session);
+   ilo_render_emit_launch_grid_commands(render, vec, &session);
+}
diff --git a/src/gallium/drivers/ilo/ilo_render.h b/src/gallium/drivers/ilo/ilo_render.h
index 0cf1d0311b0..a85b2800fb1 100644
--- a/src/gallium/drivers/ilo/ilo_render.h
+++ b/src/gallium/drivers/ilo/ilo_render.h
@@ -30,6 +30,7 @@
 
 #include "ilo_common.h"
 
+struct pipe_constant_buffer;
 struct ilo_blitter;
 struct ilo_builder;
 struct ilo_query;
@@ -87,4 +88,17 @@ void
 ilo_render_emit_draw(struct ilo_render *render,
                      const struct ilo_state_vector *vec);
 
+int
+ilo_render_get_launch_grid_len(const struct ilo_render *render,
+                               const struct ilo_state_vector *vec);
+
+void
+ilo_render_emit_launch_grid(struct ilo_render *render,
+                            const struct ilo_state_vector *vec,
+                            const unsigned thread_group_offset[3],
+                            const unsigned thread_group_dim[3],
+                            unsigned thread_group_size,
+                            const struct pipe_constant_buffer *input,
+                            uint32_t pc);
+
 #endif /* ILO_RENDER_H */
diff --git a/src/gallium/drivers/ilo/ilo_render_dynamic.c b/src/gallium/drivers/ilo/ilo_render_dynamic.c
index 5c3687318b8..07d5664fe39 100644
--- a/src/gallium/drivers/ilo/ilo_render_dynamic.c
+++ b/src/gallium/drivers/ilo/ilo_render_dynamic.c
@@ -28,6 +28,7 @@
 #include "ilo_common.h"
 #include "ilo_blitter.h"
 #include "ilo_builder_3d.h"
+#include "ilo_builder_media.h"
 #include "ilo_state.h"
 #include "ilo_render_gen.h"
 
@@ -440,3 +441,108 @@ ilo_render_emit_rectlist_dynamic_states(struct ilo_render *render,
    assert(ilo_builder_dynamic_used(render->builder) <= dynamic_used +
          ilo_render_get_rectlist_dynamic_states_len(render, blitter));
 }
+
+static void
+gen6_emit_launch_grid_dynamic_samplers(struct ilo_render *r,
+                                       const struct ilo_state_vector *vec,
+                                       struct ilo_render_launch_grid_session *session)
+{
+   const unsigned shader_type = PIPE_SHADER_COMPUTE;
+   const struct ilo_shader_state *cs = vec->cs;
+   const struct ilo_sampler_cso * const *samplers =
+      vec->sampler[shader_type].cso;
+   const struct pipe_sampler_view * const *views =
+      (const struct pipe_sampler_view **) vec->view[shader_type].states;
+   int sampler_count, i;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   sampler_count = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SAMPLER_COUNT);
+
+   assert(sampler_count <= Elements(vec->view[shader_type].states) &&
+          sampler_count <= Elements(vec->sampler[shader_type].cso));
+
+   for (i = 0; i < sampler_count; i++) {
+      r->state.cs.SAMPLER_BORDER_COLOR_STATE[i] = (samplers[i]) ?
+         gen6_SAMPLER_BORDER_COLOR_STATE(r->builder, samplers[i]) : 0;
+   }
+
+   r->state.cs.SAMPLER_STATE = gen6_SAMPLER_STATE(r->builder, samplers, views,
+         r->state.cs.SAMPLER_BORDER_COLOR_STATE, sampler_count);
+}
+
+static void
+gen6_emit_launch_grid_dynamic_pcb(struct ilo_render *r,
+                                  const struct ilo_state_vector *vec,
+                                  struct ilo_render_launch_grid_session *session)
+{
+   r->state.cs.PUSH_CONSTANT_BUFFER = 0;
+   r->state.cs.PUSH_CONSTANT_BUFFER_size = 0;
+}
+
+static void
+gen6_emit_launch_grid_dynamic_idrt(struct ilo_render *r,
+                                   const struct ilo_state_vector *vec,
+                                   struct ilo_render_launch_grid_session *session)
+{
+   const struct ilo_shader_state *cs = vec->cs;
+   struct gen6_idrt_data data;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   memset(&data, 0, sizeof(data));
+
+   data.cs = cs;
+   data.sampler_offset = r->state.cs.SAMPLER_STATE;
+   data.binding_table_offset = r->state.cs.BINDING_TABLE_STATE;
+
+   data.curbe_size = r->state.cs.PUSH_CONSTANT_BUFFER_size;
+   data.thread_group_size = session->thread_group_size;
+
+   session->idrt = gen6_INTERFACE_DESCRIPTOR_DATA(r->builder, &data, 1);
+   session->idrt_size = 32;
+}
+
+int
+ilo_render_get_launch_grid_dynamic_states_len(const struct ilo_render *render,
+                                              const struct ilo_state_vector *vec)
+{
+   const int alignment = 32 / 4;
+   int num_samplers;
+   int len = 0;
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   num_samplers = ilo_shader_get_kernel_param(vec->cs,
+         ILO_KERNEL_SAMPLER_COUNT);
+
+   /* SAMPLER_STATE array and SAMPLER_BORDER_COLORs */
+   if (num_samplers) {
+      /* prefetches are done in multiples of 4 */
+      num_samplers = align(num_samplers, 4);
+
+      len += align(GEN6_SAMPLER_STATE__SIZE * num_samplers, alignment) +
+         align(GEN6_SAMPLER_BORDER_COLOR__SIZE, alignment) * num_samplers;
+   }
+
+   len += GEN6_INTERFACE_DESCRIPTOR_DATA__SIZE;
+
+   return len;
+}
+
+void
+ilo_render_emit_launch_grid_dynamic_states(struct ilo_render *render,
+                                           const struct ilo_state_vector *vec,
+                                           struct ilo_render_launch_grid_session *session)
+{
+   const unsigned dynamic_used = ilo_builder_dynamic_used(render->builder);
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   gen6_emit_launch_grid_dynamic_samplers(render, vec, session);
+   gen6_emit_launch_grid_dynamic_pcb(render, vec, session);
+   gen6_emit_launch_grid_dynamic_idrt(render, vec, session);
+
+   assert(ilo_builder_dynamic_used(render->builder) <= dynamic_used +
+         ilo_render_get_launch_grid_dynamic_states_len(render, vec));
+}
diff --git a/src/gallium/drivers/ilo/ilo_render_gen.h b/src/gallium/drivers/ilo/ilo_render_gen.h
index 32d1237c18e..28d5030b9ae 100644
--- a/src/gallium/drivers/ilo/ilo_render_gen.h
+++ b/src/gallium/drivers/ilo/ilo_render_gen.h
@@ -118,6 +118,15 @@ struct ilo_render {
          uint32_t PUSH_CONSTANT_BUFFER;
          int PUSH_CONSTANT_BUFFER_size;
       } wm;
+
+      struct {
+         uint32_t BINDING_TABLE_STATE;
+         uint32_t SURFACE_STATE[ILO_MAX_SURFACES];
+         uint32_t SAMPLER_STATE;
+         uint32_t SAMPLER_BORDER_COLOR_STATE[ILO_MAX_SAMPLERS];
+         uint32_t PUSH_CONSTANT_BUFFER;
+         int PUSH_CONSTANT_BUFFER_size;
+      } cs;
    } state;
 };
 
@@ -157,6 +166,17 @@ struct ilo_render_rectlist_session {
    uint32_t vb_end;
 };
 
+struct ilo_render_launch_grid_session {
+   const unsigned *thread_group_offset;
+   const unsigned *thread_group_dim;
+   unsigned thread_group_size;
+   const struct pipe_constant_buffer *input;
+   uint32_t pc;
+
+   uint32_t idrt;
+   int idrt_size;
+};
+
 int
 ilo_render_get_draw_commands_len_gen6(const struct ilo_render *render,
                                       const struct ilo_state_vector *vec);
@@ -239,6 +259,15 @@ ilo_render_emit_rectlist_commands(struct ilo_render *render,
 }
 
 int
+ilo_render_get_launch_grid_commands_len(const struct ilo_render *render,
+                                        const struct ilo_state_vector *vec);
+
+void
+ilo_render_emit_launch_grid_commands(struct ilo_render *render,
+                                     const struct ilo_state_vector *vec,
+                                     const struct ilo_render_launch_grid_session *session);
+
+int
 ilo_render_get_draw_dynamic_states_len(const struct ilo_render *render,
                                        const struct ilo_state_vector *vec);
 
@@ -257,6 +286,15 @@ ilo_render_emit_rectlist_dynamic_states(struct ilo_render *render,
                                         struct ilo_render_rectlist_session *session);
 
 int
+ilo_render_get_launch_grid_dynamic_states_len(const struct ilo_render *render,
+                                              const struct ilo_state_vector *vec);
+
+void
+ilo_render_emit_launch_grid_dynamic_states(struct ilo_render *render,
+                                           const struct ilo_state_vector *vec,
+                                           struct ilo_render_launch_grid_session *session);
+
+int
 ilo_render_get_draw_surface_states_len(const struct ilo_render *render,
                                        const struct ilo_state_vector *vec);
 
@@ -265,6 +303,15 @@ ilo_render_emit_draw_surface_states(struct ilo_render *render,
                                     const struct ilo_state_vector *vec,
                                     struct ilo_render_draw_session *session);
 
+int
+ilo_render_get_launch_grid_surface_states_len(const struct ilo_render *render,
+                                              const struct ilo_state_vector *vec);
+
+void
+ilo_render_emit_launch_grid_surface_states(struct ilo_render *render,
+                                           const struct ilo_state_vector *vec,
+                                           struct ilo_render_launch_grid_session *session);
+
 void
 gen6_wa_pre_pipe_control(struct ilo_render *r, uint32_t dw1);
 
diff --git a/src/gallium/drivers/ilo/ilo_render_media.c b/src/gallium/drivers/ilo/ilo_render_media.c
new file mode 100644
index 00000000000..f6b466c2996
--- /dev/null
+++ b/src/gallium/drivers/ilo/ilo_render_media.c
@@ -0,0 +1,229 @@
+/*
+ * Mesa 3-D graphics library
+ *
+ * Copyright (C) 2014 LunarG, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Chia-I Wu <[email protected]>
+ */
+
+#include "genhw/genhw.h"
+
+#include "ilo_builder_media.h"
+#include "ilo_builder_mi.h"
+#include "ilo_builder_render.h"
+#include "ilo_state.h"
+#include "ilo_render_gen.h"
+
+struct gen7_l3_config {
+   int slm;
+   int urb;
+   int rest;
+   int dc;
+   int ro;
+   int is;
+   int c;
+   int t;
+};
+
+/*
+ * From the Ivy Bridge PRM, volume 1 part 7, page 10:
+ *
+ *     "Normal L3/URB mode (non-SLM mode), uses all 4 banks of L3 equally to
+ *      distribute cycles. The following allocation is a suggested programming
+ *      model. Note all numbers below are given in KBytes."
+ *
+ * From the Haswell PRM, volume 7, page 662:
+ *
+ *     "The configuration for {SLM = 0,URB = 224,DC = 32,RO = 256,IS = 0,C =
+ *      0,T =0, SUM 512} was validated as a later supported configuration and
+ *      can be utilized if desired."
+ */
+static const struct gen7_l3_config gen7_l3_non_slm_configs[] = {
+   /*       SLM   URB  Rest    DC    RO   I/S     C     T */
+   [0] = {    0,  256,    0,    0,  256,    0,    0,    0, },
+   [1] = {    0,  256,    0,  128,  128,    0,    0,    0, },
+   [2] = {    0,  256,    0,   32,    0,   64,   32,  128, },
+   [3] = {    0,  224,    0,   64,    0,   64,   32,  128, },
+   [4] = {    0,  224,    0,  128,    0,   64,   32,   64, },
+   [5] = {    0,  224,    0,   64,    0,  128,   32,   64, },
+   [6] = {    0,  224,    0,    0,    0,  128,   32,  128, },
+   [7] = {    0,  256,    0,    0,    0,  128,    0,  128, },
+
+   [8] = {    0,  224,    0,   32,  256,    0,    0,    0, },
+};
+
+/*
+ * From the Ivy Bridge PRM, volume 1 part 7, page 11:
+ *
+ *     "With the existence of Shared Local Memory, a 64KB chunk from each of
+ *      the 2 L3 banks will be reserved for SLM usage. The remaining cache
+ *      space is divided between the remaining clients. SLM allocation is done
+ *      via reducing the number of ways on the two banks from 64 to 32."
+ *
+ * From the Haswell PRM, volume 7, page 662:
+ *
+ *     "The configuration for {SLM = 128,URB = 128,DC = 0,RO = 256,IS = 0,C =
+ *      0,T =0, SUM 512} was validated as a later supported configuration and
+ *      can be utilized if desired. For this configuration, global atomics
+ *      must be programmed to be in GTI."
+ */
+static const struct gen7_l3_config gen7_l3_slm_configs[] = {
+   /*       SLM   URB  Rest    DC    RO   I/S     C     T */
+   [0] = {  128,  128,    0,  128,  128,    0,    0,    0, },
+   [1] = {  128,  128,    0,   64,    0,   64,   64,   64, },
+   [2] = {  128,  128,    0,   32,    0,   64,   32,  128, },
+   [3] = {  128,  128,    0,   32,    0,  128,   32,   64, },
+
+   [4] = {  128,  128,    0,    0,  256,    0,    0,    0, },
+};
+
+static void
+gen7_launch_grid_l3(struct ilo_render *r, bool use_slm)
+{
+   uint32_t l3sqcreg1, l3cntlreg2, l3cntlreg3;
+   const struct gen7_l3_config *conf;
+
+   /*
+    * This function mostly follows what beignet does.  I do not know why, for
+    * example, CON4DCUNC should be reset.  I do not know if it should be set
+    * again after launch_grid().
+    */
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   if (use_slm)
+      conf = &gen7_l3_slm_configs[1];
+   else
+      conf = &gen7_l3_non_slm_configs[4];
+
+   /* unset GEN7_REG_L3SQCREG1_CON4DCUNC (without readback first) */
+   if (ilo_dev_gen(r->dev) >= ILO_GEN(7.5)) {
+      l3sqcreg1 = GEN75_REG_L3SQCREG1_SQGPCI_24 |
+                  GEN75_REG_L3SQCREG1_SQHPCI_8;
+   } else {
+      l3sqcreg1 = GEN7_REG_L3SQCREG1_SQGHPCI_18_6;
+   }
+
+   l3cntlreg2 = (conf->dc / 8) << GEN7_REG_L3CNTLREG2_DCWASS__SHIFT |
+                (conf->ro / 8) << GEN7_REG_L3CNTLREG2_RDOCPL__SHIFT |
+                (conf->urb / 8) << GEN7_REG_L3CNTLREG2_URBALL__SHIFT;
+
+   l3cntlreg3 = (conf->t / 8) << GEN7_REG_L3CNTLREG3_TXWYALL__SHIFT |
+                (conf->c / 8) << GEN7_REG_L3CNTLREG3_CTWYALL__SHIFT |
+                (conf->is / 8) << GEN7_REG_L3CNTLREG3_ISWYALL__SHIFT;
+
+   if (conf->slm) {
+      /*
+       * From the Ivy Bridge PRM, volume 1 part 7, page 11:
+       *
+       *     "Note that URB needs to be set as low b/w client in SLM mode,
+       *      else the hash will fail. This is a required s/w model."
+       */
+      l3cntlreg2 |= GEN7_REG_L3CNTLREG2_URBSLMB |
+                    GEN7_REG_L3CNTLREG2_SLMMENB;
+   }
+
+   gen6_MI_LOAD_REGISTER_IMM(r->builder, GEN7_REG_L3SQCREG1, l3sqcreg1);
+   gen6_MI_LOAD_REGISTER_IMM(r->builder, GEN7_REG_L3CNTLREG2, l3cntlreg2);
+   gen6_MI_LOAD_REGISTER_IMM(r->builder, GEN7_REG_L3CNTLREG3, l3cntlreg3);
+}
+
+int
+ilo_render_get_launch_grid_commands_len(const struct ilo_render *render,
+                                        const struct ilo_state_vector *vec)
+{
+   static int len;
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   if (!len) {
+      len +=
+         GEN6_PIPELINE_SELECT__SIZE +
+         GEN6_STATE_BASE_ADDRESS__SIZE +
+         GEN6_MEDIA_VFE_STATE__SIZE +
+         GEN6_MEDIA_CURBE_LOAD__SIZE +
+         GEN6_MEDIA_INTERFACE_DESCRIPTOR_LOAD__SIZE +
+         GEN6_MEDIA_STATE_FLUSH__SIZE;
+
+      len += ilo_render_get_flush_len(render) * 3;
+
+      if (ilo_dev_gen(render->dev) >= ILO_GEN(7)) {
+         len += GEN6_MI_LOAD_REGISTER_IMM__SIZE * 3 * 2;
+         len += GEN7_GPGPU_WALKER__SIZE;
+      }
+   }
+
+   return len;
+}
+
+void
+ilo_render_emit_launch_grid_commands(struct ilo_render *render,
+                                     const struct ilo_state_vector *vec,
+                                     const struct ilo_render_launch_grid_session *session)
+{
+   const unsigned batch_used = ilo_builder_batch_used(render->builder);
+   const uint32_t pcb = render->state.cs.PUSH_CONSTANT_BUFFER;
+   const int pcb_size = render->state.cs.PUSH_CONSTANT_BUFFER_size;
+   int simd_size;
+   bool use_slm;
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   simd_size = ilo_shader_get_kernel_param(vec->cs, ILO_KERNEL_CS_SIMD_SIZE);
+   use_slm = ilo_shader_get_kernel_param(vec->cs, ILO_KERNEL_CS_LOCAL_SIZE);
+
+   ilo_render_emit_flush(render);
+
+   if (ilo_dev_gen(render->dev) >= ILO_GEN(7)) {
+      gen7_launch_grid_l3(render, use_slm);
+      ilo_render_emit_flush(render);
+
+      gen6_PIPELINE_SELECT(render->builder,
+            GEN7_PIPELINE_SELECT_DW0_SELECT_GPGPU);
+   } else {
+      gen6_PIPELINE_SELECT(render->builder,
+            GEN6_PIPELINE_SELECT_DW0_SELECT_MEDIA);
+   }
+
+   gen6_state_base_address(render->builder, true);
+
+   gen6_MEDIA_VFE_STATE(render->builder, pcb_size, use_slm);
+
+   if (pcb_size)
+      gen6_MEDIA_CURBE_LOAD(render->builder, pcb, pcb_size);
+
+   gen6_MEDIA_INTERFACE_DESCRIPTOR_LOAD(render->builder,
+         session->idrt, session->idrt_size);
+
+   gen7_GPGPU_WALKER(render->builder, session->thread_group_offset,
+         session->thread_group_dim, session->thread_group_size, simd_size);
+
+   gen6_MEDIA_STATE_FLUSH(render->builder);
+
+   if (ilo_dev_gen(render->dev) >= ILO_GEN(7) && use_slm) {
+      ilo_render_emit_flush(render);
+      gen7_launch_grid_l3(render, false);
+   }
+
+   assert(ilo_builder_batch_used(render->builder) <= batch_used +
+         ilo_render_get_launch_grid_commands_len(render, vec));
+}
diff --git a/src/gallium/drivers/ilo/ilo_render_surface.c b/src/gallium/drivers/ilo/ilo_render_surface.c
index 22a7e488edf..657fbfdabec 100644
--- a/src/gallium/drivers/ilo/ilo_render_surface.c
+++ b/src/gallium/drivers/ilo/ilo_render_surface.c
@@ -381,3 +381,177 @@ ilo_render_emit_draw_surface_states(struct ilo_render *render,
    assert(ilo_builder_surface_used(render->builder) <= surface_used +
          ilo_render_get_draw_surface_states_len(render, vec));
 }
+
+static void
+gen6_emit_launch_grid_surface_view(struct ilo_render *r,
+                                   const struct ilo_state_vector *vec,
+                                   struct ilo_render_launch_grid_session *session)
+{
+   const struct ilo_shader_state *cs = vec->cs;
+   const struct ilo_view_state *view = &vec->view[PIPE_SHADER_COMPUTE];
+   uint32_t *surface_state = r->state.cs.SURFACE_STATE;
+   int base, count, i;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   base = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SURFACE_TEX_BASE);
+   count = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SURFACE_TEX_COUNT);
+
+   /* SURFACE_STATEs for sampler views */
+   surface_state += base;
+   for (i = 0; i < count; i++) {
+      if (i < view->count && view->states[i]) {
+         const struct ilo_view_cso *cso =
+            (const struct ilo_view_cso *) view->states[i];
+
+         surface_state[i] =
+            gen6_SURFACE_STATE(r->builder, &cso->surface, false);
+      } else {
+         surface_state[i] = 0;
+      }
+   }
+}
+
+static void
+gen6_emit_launch_grid_surface_const(struct ilo_render *r,
+                                    const struct ilo_state_vector *vec,
+                                    struct ilo_render_launch_grid_session *session)
+{
+   const struct ilo_shader_state *cs = vec->cs;
+   uint32_t *surface_state = r->state.cs.SURFACE_STATE;
+   struct ilo_view_surface view;
+   int base, count;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   base = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SURFACE_CONST_BASE);
+   count = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SURFACE_CONST_COUNT);
+
+   if (!count)
+      return;
+
+   ilo_gpe_init_view_surface_for_buffer(r->dev,
+         ilo_buffer(session->input->buffer),
+         session->input->buffer_offset,
+         session->input->buffer_size,
+         1, PIPE_FORMAT_NONE,
+         false, false, &view);
+
+   assert(count == 1 && session->input->buffer);
+   surface_state[base] = gen6_SURFACE_STATE(r->builder, &view, false);
+}
+
+static void
+gen6_emit_launch_grid_surface_cs_resource(struct ilo_render *r,
+                                          const struct ilo_state_vector *vec,
+                                          struct ilo_render_launch_grid_session *session)
+{
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   /* TODO */
+   assert(!vec->cs_resource.count);
+}
+
+static void
+gen6_emit_launch_grid_surface_global(struct ilo_render *r,
+                                          const struct ilo_state_vector *vec,
+                                          struct ilo_render_launch_grid_session *session)
+{
+   const struct ilo_shader_state *cs = vec->cs;
+   const struct ilo_global_binding_cso *bindings =
+      util_dynarray_begin(&vec->global_binding.bindings);
+   uint32_t *surface_state = r->state.cs.SURFACE_STATE;
+   int base, count, i;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   base = ilo_shader_get_kernel_param(cs, ILO_KERNEL_CS_SURFACE_GLOBAL_BASE);
+   count = ilo_shader_get_kernel_param(cs, ILO_KERNEL_CS_SURFACE_GLOBAL_COUNT);
+
+   if (!count)
+      return;
+
+   if (base + count > Elements(r->state.cs.SURFACE_STATE)) {
+      ilo_warn("too many global bindings\n");
+      count = Elements(r->state.cs.SURFACE_STATE) - base;
+   }
+
+   /* SURFACE_STATEs for global bindings */
+   surface_state += base;
+   for (i = 0; i < count; i++) {
+      if (i < vec->global_binding.count && bindings[i].resource) {
+         const struct ilo_buffer *buf = ilo_buffer(bindings[i].resource);
+         struct ilo_view_surface view;
+
+         assert(bindings[i].resource->target == PIPE_BUFFER);
+
+         ilo_gpe_init_view_surface_for_buffer(r->dev, buf, 0, buf->bo_size,
+               1, PIPE_FORMAT_NONE, true, true, &view);
+         surface_state[i] =
+            gen6_SURFACE_STATE(r->builder, &view, true);
+      } else {
+         surface_state[i] = 0;
+      }
+   }
+}
+
+static void
+gen6_emit_launch_grid_surface_binding_table(struct ilo_render *r,
+                                            const struct ilo_state_vector *vec,
+                                            struct ilo_render_launch_grid_session *session)
+{
+   const struct ilo_shader_state *cs = vec->cs;
+   int count;
+
+   ILO_DEV_ASSERT(r->dev, 7, 7.5);
+
+   count = ilo_shader_get_kernel_param(cs, ILO_KERNEL_SURFACE_TOTAL_COUNT);
+   if (count) {
+      r->state.cs.BINDING_TABLE_STATE = gen6_BINDING_TABLE_STATE(r->builder,
+            r->state.cs.SURFACE_STATE, count);
+   }
+}
+
+int
+ilo_render_get_launch_grid_surface_states_len(const struct ilo_render *render,
+                                              const struct ilo_state_vector *vec)
+{
+   const int alignment = 32 / 4;
+   int num_surfaces;
+   int len = 0;
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   num_surfaces = ilo_shader_get_kernel_param(vec->cs,
+         ILO_KERNEL_SURFACE_TOTAL_COUNT);
+
+   /* BINDING_TABLE_STATE and SURFACE_STATEs */
+   if (num_surfaces) {
+      len += align(num_surfaces, alignment) +
+         align(GEN6_SURFACE_STATE__SIZE, alignment) * num_surfaces;
+   }
+
+   return len;
+}
+
+void
+ilo_render_emit_launch_grid_surface_states(struct ilo_render *render,
+                                           const struct ilo_state_vector *vec,
+                                           struct ilo_render_launch_grid_session *session)
+{
+   const unsigned surface_used = ilo_builder_surface_used(render->builder);
+
+   ILO_DEV_ASSERT(render->dev, 7, 7.5);
+
+   /* idrt depends on the binding table */
+   assert(!session->idrt_size);
+
+   gen6_emit_launch_grid_surface_view(render, vec, session);
+   gen6_emit_launch_grid_surface_const(render, vec, session);
+   gen6_emit_launch_grid_surface_cs_resource(render, vec, session);
+   gen6_emit_launch_grid_surface_global(render, vec, session);
+   gen6_emit_launch_grid_surface_binding_table(render, vec, session);
+
+   assert(ilo_builder_surface_used(render->builder) <= surface_used +
+         ilo_render_get_launch_grid_surface_states_len(render, vec));
+}