xref: /xsrc/external/mit/MesaLib/dist/src/gallium/drivers/iris/iris_batch.c

/*
 * Copyright © 2017 Intel Corporation
 *
 * 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.
 */

/**
 * @file iris_batch.c
 *
 * Batchbuffer and command submission module.
 *
 * Every API draw call results in a number of GPU commands, which we
 * collect into a "batch buffer".  Typically, many draw calls are grouped
 * into a single batch to amortize command submission overhead.
 *
 * We submit batches to the kernel using the I915_GEM_EXECBUFFER2 ioctl.
 * One critical piece of data is the "validation list", which contains a
 * list of the buffer objects (BOs) which the commands in the GPU need.
 * The kernel will make sure these are resident and pinned at the correct
 * virtual memory address before executing our batch.  If a BO is not in
 * the validation list, it effectively does not exist, so take care.
 */

#include "iris_batch.h"
#include "iris_bufmgr.h"
#include "iris_context.h"
#include "iris_fence.h"

#include "drm-uapi/i915_drm.h"

#include "common/intel_aux_map.h"
#include "intel/common/intel_gem.h"
#include "util/hash_table.h"
#include "util/set.h"
#include "util/u_upload_mgr.h"
#include "main/macros.h"

#include <errno.h>
#include <xf86drm.h>

#if HAVE_VALGRIND
#include <valgrind.h>
#include <memcheck.h>
#define VG(x) x
#else
#define VG(x)
#endif

#define FILE_DEBUG_FLAG DEBUG_BUFMGR

static void
iris_batch_reset(struct iris_batch *batch);

static unsigned
num_fences(struct iris_batch *batch)
{
   return util_dynarray_num_elements(&batch->exec_fences,
                                     struct drm_i915_gem_exec_fence);
}

/**
 * Debugging code to dump the fence list, used by INTEL_DEBUG=submit.
 */
static void
dump_fence_list(struct iris_batch *batch)
{
   fprintf(stderr, "Fence list (length %u):      ", num_fences(batch));

   util_dynarray_foreach(&batch->exec_fences,
                         struct drm_i915_gem_exec_fence, f) {
      fprintf(stderr, "%s%u%s ",
              (f->flags & I915_EXEC_FENCE_WAIT) ? "..." : "",
              f->handle,
              (f->flags & I915_EXEC_FENCE_SIGNAL) ? "!" : "");
   }

   fprintf(stderr, "\n");
}

/**
 * Debugging code to dump the validation list, used by INTEL_DEBUG=submit.
 */
static void
dump_bo_list(struct iris_batch *batch)
{
   fprintf(stderr, "BO list (length %d):\n", batch->exec_count);

   for (int i = 0; i < batch->exec_count; i++) {
      struct iris_bo *bo = batch->exec_bos[i];
      struct iris_bo *backing = iris_get_backing_bo(bo);
      bool written = BITSET_TEST(batch->bos_written, i);

      fprintf(stderr, "[%2d]: %3d (%3d) %-14s @ 0x%016"PRIx64" (%-6s %8"PRIu64"B) %2d refs  %s\n",
              i,
              bo->gem_handle,
              backing->gem_handle,
              bo->name,
              bo->address,
              backing->real.local ? "local" : "system",
              bo->size,
              bo->refcount,
              written ? "(write)" : "");
   }
}

/**
 * Return BO information to the batch decoder (for debugging).
 */
static struct intel_batch_decode_bo
decode_get_bo(void *v_batch, bool ppgtt, uint64_t address)
{
   struct iris_batch *batch = v_batch;

   assert(ppgtt);

   for (int i = 0; i < batch->exec_count; i++) {
      struct iris_bo *bo = batch->exec_bos[i];
      /* The decoder zeroes out the top 16 bits, so we need to as well */
      uint64_t bo_address = bo->address & (~0ull >> 16);

      if (address >= bo_address && address < bo_address + bo->size) {
         return (struct intel_batch_decode_bo) {
            .addr = bo_address,
            .size = bo->size,
            .map = iris_bo_map(batch->dbg, bo, MAP_READ),
         };
      }
   }

   return (struct intel_batch_decode_bo) { };
}

static unsigned
decode_get_state_size(void *v_batch,
                      uint64_t address,
                      UNUSED uint64_t base_address)
{
   struct iris_batch *batch = v_batch;
   unsigned size = (uintptr_t)
      _mesa_hash_table_u64_search(batch->state_sizes, address);

   return size;
}

/**
 * Decode the current batch.
 */
static void
decode_batch(struct iris_batch *batch)
{
   void *map = iris_bo_map(batch->dbg, batch->exec_bos[0], MAP_READ);
   intel_print_batch(&batch->decoder, map, batch->primary_batch_size,
                     batch->exec_bos[0]->address, false);
}

void
iris_init_batch(struct iris_context *ice,
                enum iris_batch_name name,
                int priority)
{
   struct iris_batch *batch = &ice->batches[name];
   struct iris_screen *screen = (void *) ice->ctx.screen;

   batch->screen = screen;
   batch->dbg = &ice->dbg;
   batch->reset = &ice->reset;
   batch->state_sizes = ice->state.sizes;
   batch->name = name;
   batch->ice = ice;
   batch->contains_fence_signal = false;

   batch->fine_fences.uploader =
      u_upload_create(&ice->ctx, 4096, PIPE_BIND_CUSTOM,
                      PIPE_USAGE_STAGING, 0);
   iris_fine_fence_init(batch);

   batch->hw_ctx_id = iris_create_hw_context(screen->bufmgr);
   assert(batch->hw_ctx_id);

   iris_hw_context_set_priority(screen->bufmgr, batch->hw_ctx_id, priority);

   util_dynarray_init(&batch->exec_fences, ralloc_context(NULL));
   util_dynarray_init(&batch->syncobjs, ralloc_context(NULL));

   batch->exec_count = 0;
   batch->max_gem_handle = 0;
   batch->exec_array_size = 128;
   batch->exec_bos =
      malloc(batch->exec_array_size * sizeof(batch->exec_bos[0]));
   batch->bos_written =
      rzalloc_array(NULL, BITSET_WORD, BITSET_WORDS(batch->exec_array_size));

   batch->cache.render = _mesa_hash_table_create(NULL, _mesa_hash_pointer,
                                                 _mesa_key_pointer_equal);

   memset(batch->other_batches, 0, sizeof(batch->other_batches));

   for (int i = 0, j = 0; i < IRIS_BATCH_COUNT; i++) {
      if (i != name)
         batch->other_batches[j++] = &ice->batches[i];
   }

   if (INTEL_DEBUG(DEBUG_ANY)) {
      const unsigned decode_flags =
         INTEL_BATCH_DECODE_FULL |
         (INTEL_DEBUG(DEBUG_COLOR) ? INTEL_BATCH_DECODE_IN_COLOR : 0) |
         INTEL_BATCH_DECODE_OFFSETS |
         INTEL_BATCH_DECODE_FLOATS;

      intel_batch_decode_ctx_init(&batch->decoder, &screen->devinfo,
                                  stderr, decode_flags, NULL,
                                  decode_get_bo, decode_get_state_size, batch);
      batch->decoder.dynamic_base = IRIS_MEMZONE_DYNAMIC_START;
      batch->decoder.instruction_base = IRIS_MEMZONE_SHADER_START;
      batch->decoder.max_vbo_decoded_lines = 32;
   }

   iris_init_batch_measure(ice, batch);

   iris_batch_reset(batch);
}

static int
find_exec_index(struct iris_batch *batch, struct iris_bo *bo)
{
   unsigned index = READ_ONCE(bo->index);

   if (index < batch->exec_count && batch->exec_bos[index] == bo)
      return index;

   /* May have been shared between multiple active batches */
   for (index = 0; index < batch->exec_count; index++) {
      if (batch->exec_bos[index] == bo)
         return index;
   }

   return -1;
}

static void
ensure_exec_obj_space(struct iris_batch *batch, uint32_t count)
{
   while (batch->exec_count + count > batch->exec_array_size) {
      unsigned old_size = batch->exec_array_size;

      batch->exec_array_size *= 2;
      batch->exec_bos =
         realloc(batch->exec_bos,
                 batch->exec_array_size * sizeof(batch->exec_bos[0]));
      batch->bos_written =
         rerzalloc(NULL, batch->bos_written, BITSET_WORD,
                   BITSET_WORDS(old_size),
                   BITSET_WORDS(batch->exec_array_size));
   }
}

static void
add_bo_to_batch(struct iris_batch *batch, struct iris_bo *bo, bool writable)
{
   assert(batch->exec_array_size > batch->exec_count);

   iris_bo_reference(bo);

   batch->exec_bos[batch->exec_count] = bo;

   if (writable)
      BITSET_SET(batch->bos_written, batch->exec_count);

   bo->index = batch->exec_count;
   batch->exec_count++;
   batch->aperture_space += bo->size;

   batch->max_gem_handle =
      MAX2(batch->max_gem_handle, iris_get_backing_bo(bo)->gem_handle);
}

static void
flush_for_cross_batch_dependencies(struct iris_batch *batch,
                                   struct iris_bo *bo,
                                   bool writable)
{
   if (batch->measure && bo == batch->measure->bo)
      return;

   /* When a batch uses a buffer for the first time, or newly writes a buffer
    * it had already referenced, we may need to flush other batches in order
    * to correctly synchronize them.
    */
   for (int b = 0; b < ARRAY_SIZE(batch->other_batches); b++) {
      struct iris_batch *other_batch = batch->other_batches[b];
      int other_index = find_exec_index(other_batch, bo);

      /* If the buffer is referenced by another batch, and either batch
       * intends to write it, then flush the other batch and synchronize.
       *
       * Consider these cases:
       *
       * 1. They read, we read   =>  No synchronization required.
       * 2. They read, we write  =>  Synchronize (they need the old value)
       * 3. They write, we read  =>  Synchronize (we need their new value)
       * 4. They write, we write =>  Synchronize (order writes)
       *
       * The read/read case is very common, as multiple batches usually
       * share a streaming state buffer or shader assembly buffer, and
       * we want to avoid synchronizing in this case.
       */
      if (other_index != -1 &&
          (writable || BITSET_TEST(other_batch->bos_written, other_index)))
         iris_batch_flush(other_batch);
   }
}

/**
 * Add a buffer to the current batch's validation list.
 *
 * You must call this on any BO you wish to use in this batch, to ensure
 * that it's resident when the GPU commands execute.
 */
void
iris_use_pinned_bo(struct iris_batch *batch,
                   struct iris_bo *bo,
                   bool writable, enum iris_domain access)
{
   assert(iris_get_backing_bo(bo)->real.kflags & EXEC_OBJECT_PINNED);
   assert(bo != batch->bo);

   /* Never mark the workaround BO with EXEC_OBJECT_WRITE.  We don't care
    * about the order of any writes to that buffer, and marking it writable
    * would introduce data dependencies between multiple batches which share
    * the buffer. It is added directly to the batch using add_bo_to_batch()
    * during batch reset time.
    */
   if (bo == batch->screen->workaround_bo)
      return;

   if (access < NUM_IRIS_DOMAINS) {
      assert(batch->sync_region_depth);
      iris_bo_bump_seqno(bo, batch->next_seqno, access);
   }

   int existing_index = find_exec_index(batch, bo);

   if (existing_index == -1) {
      flush_for_cross_batch_dependencies(batch, bo, writable);

      ensure_exec_obj_space(batch, 1);
      add_bo_to_batch(batch, bo, writable);
   } else if (writable && !BITSET_TEST(batch->bos_written, existing_index)) {
      flush_for_cross_batch_dependencies(batch, bo, writable);

      /* The BO is already in the list; mark it writable */
      BITSET_SET(batch->bos_written, existing_index);
   }
}

static void
create_batch(struct iris_batch *batch)
{
   struct iris_screen *screen = batch->screen;
   struct iris_bufmgr *bufmgr = screen->bufmgr;

   /* TODO: We probably could suballocate batches... */
   batch->bo = iris_bo_alloc(bufmgr, "command buffer",
                             BATCH_SZ + BATCH_RESERVED, 1,
                             IRIS_MEMZONE_OTHER, BO_ALLOC_NO_SUBALLOC);
   iris_get_backing_bo(batch->bo)->real.kflags |= EXEC_OBJECT_CAPTURE;
   batch->map = iris_bo_map(NULL, batch->bo, MAP_READ | MAP_WRITE);
   batch->map_next = batch->map;

   ensure_exec_obj_space(batch, 1);
   add_bo_to_batch(batch, batch->bo, false);
}

static void
iris_batch_maybe_noop(struct iris_batch *batch)
{
   /* We only insert the NOOP at the beginning of the batch. */
   assert(iris_batch_bytes_used(batch) == 0);

   if (batch->noop_enabled) {
      /* Emit MI_BATCH_BUFFER_END to prevent any further command to be
       * executed.
       */
      uint32_t *map = batch->map_next;

      map[0] = (0xA << 23);

      batch->map_next += 4;
   }
}

static void
iris_batch_reset(struct iris_batch *batch)
{
   struct iris_screen *screen = batch->screen;
   struct iris_bufmgr *bufmgr = screen->bufmgr;

   iris_bo_unreference(batch->bo);
   batch->primary_batch_size = 0;
   batch->total_chained_batch_size = 0;
   batch->contains_draw = false;
   batch->contains_fence_signal = false;
   batch->decoder.surface_base = batch->last_surface_base_address;

   create_batch(batch);
   assert(batch->bo->index == 0);

   memset(batch->bos_written, 0,
          sizeof(BITSET_WORD) * BITSET_WORDS(batch->exec_array_size));

   struct iris_syncobj *syncobj = iris_create_syncobj(bufmgr);
   iris_batch_add_syncobj(batch, syncobj, I915_EXEC_FENCE_SIGNAL);
   iris_syncobj_reference(bufmgr, &syncobj, NULL);

   assert(!batch->sync_region_depth);
   iris_batch_sync_boundary(batch);
   iris_batch_mark_reset_sync(batch);

   /* Always add the workaround BO, it contains a driver identifier at the
    * beginning quite helpful to debug error states.
    */
   add_bo_to_batch(batch, screen->workaround_bo, false);

   iris_batch_maybe_noop(batch);
}

void
iris_batch_free(struct iris_batch *batch)
{
   struct iris_screen *screen = batch->screen;
   struct iris_bufmgr *bufmgr = screen->bufmgr;

   for (int i = 0; i < batch->exec_count; i++) {
      iris_bo_unreference(batch->exec_bos[i]);
   }
   free(batch->exec_bos);
   ralloc_free(batch->bos_written);

   ralloc_free(batch->exec_fences.mem_ctx);

   pipe_resource_reference(&batch->fine_fences.ref.res, NULL);

   util_dynarray_foreach(&batch->syncobjs, struct iris_syncobj *, s)
      iris_syncobj_reference(bufmgr, s, NULL);
   ralloc_free(batch->syncobjs.mem_ctx);

   iris_fine_fence_reference(batch->screen, &batch->last_fence, NULL);
   u_upload_destroy(batch->fine_fences.uploader);

   iris_bo_unreference(batch->bo);
   batch->bo = NULL;
   batch->map = NULL;
   batch->map_next = NULL;

   iris_destroy_hw_context(bufmgr, batch->hw_ctx_id);

   iris_destroy_batch_measure(batch->measure);
   batch->measure = NULL;

   _mesa_hash_table_destroy(batch->cache.render, NULL);

   if (INTEL_DEBUG(DEBUG_ANY))
      intel_batch_decode_ctx_finish(&batch->decoder);
}

/**
 * If we've chained to a secondary batch, or are getting near to the end,
 * then flush.  This should only be called between draws.
 */
void
iris_batch_maybe_flush(struct iris_batch *batch, unsigned estimate)
{
   if (batch->bo != batch->exec_bos[0] ||
       iris_batch_bytes_used(batch) + estimate >= BATCH_SZ) {
      iris_batch_flush(batch);
   }
}

static void
record_batch_sizes(struct iris_batch *batch)
{
   unsigned batch_size = iris_batch_bytes_used(batch);

   VG(VALGRIND_CHECK_MEM_IS_DEFINED(batch->map, batch_size));

   if (batch->bo == batch->exec_bos[0])
      batch->primary_batch_size = batch_size;

   batch->total_chained_batch_size += batch_size;
}

void
iris_chain_to_new_batch(struct iris_batch *batch)
{
   uint32_t *cmd = batch->map_next;
   uint64_t *addr = batch->map_next + 4;
   batch->map_next += 12;

   record_batch_sizes(batch);

   /* No longer held by batch->bo, still held by validation list */
   iris_bo_unreference(batch->bo);
   create_batch(batch);

   /* Emit MI_BATCH_BUFFER_START to chain to another batch. */
   *cmd = (0x31 << 23) | (1 << 8) | (3 - 2);
   *addr = batch->bo->address;
}

static void
add_aux_map_bos_to_batch(struct iris_batch *batch)
{
   void *aux_map_ctx = iris_bufmgr_get_aux_map_context(batch->screen->bufmgr);
   if (!aux_map_ctx)
      return;

   uint32_t count = intel_aux_map_get_num_buffers(aux_map_ctx);
   ensure_exec_obj_space(batch, count);
   intel_aux_map_fill_bos(aux_map_ctx,
                          (void**)&batch->exec_bos[batch->exec_count], count);
   for (uint32_t i = 0; i < count; i++) {
      struct iris_bo *bo = batch->exec_bos[batch->exec_count];
      add_bo_to_batch(batch, bo, false);
   }
}

static void
finish_seqno(struct iris_batch *batch)
{
   struct iris_fine_fence *sq = iris_fine_fence_new(batch, IRIS_FENCE_END);
   if (!sq)
      return;

   iris_fine_fence_reference(batch->screen, &batch->last_fence, sq);
   iris_fine_fence_reference(batch->screen, &sq, NULL);
}

/**
 * Terminate a batch with MI_BATCH_BUFFER_END.
 */
static void
iris_finish_batch(struct iris_batch *batch)
{
   const struct intel_device_info *devinfo = &batch->screen->devinfo;

   if (devinfo->ver == 12 && batch->name == IRIS_BATCH_RENDER) {
      /* We re-emit constants at the beginning of every batch as a hardware
       * bug workaround, so invalidate indirect state pointers in order to
       * save ourselves the overhead of restoring constants redundantly when
       * the next render batch is executed.
       */
      iris_emit_pipe_control_flush(batch, "ISP invalidate at batch end",
                                   PIPE_CONTROL_INDIRECT_STATE_POINTERS_DISABLE |
                                   PIPE_CONTROL_STALL_AT_SCOREBOARD |
                                   PIPE_CONTROL_CS_STALL);
   }

   add_aux_map_bos_to_batch(batch);

   finish_seqno(batch);

   /* Emit MI_BATCH_BUFFER_END to finish our batch. */
   uint32_t *map = batch->map_next;

   map[0] = (0xA << 23);

   batch->map_next += 4;

   record_batch_sizes(batch);
}

/**
 * Replace our current GEM context with a new one (in case it got banned).
 */
static bool
replace_hw_ctx(struct iris_batch *batch)
{
   struct iris_screen *screen = batch->screen;
   struct iris_bufmgr *bufmgr = screen->bufmgr;

   uint32_t new_ctx = iris_clone_hw_context(bufmgr, batch->hw_ctx_id);
   if (!new_ctx)
      return false;

   iris_destroy_hw_context(bufmgr, batch->hw_ctx_id);
   batch->hw_ctx_id = new_ctx;

   /* Notify the context that state must be re-initialized. */
   iris_lost_context_state(batch);

   return true;
}

enum pipe_reset_status
iris_batch_check_for_reset(struct iris_batch *batch)
{
   struct iris_screen *screen = batch->screen;
   enum pipe_reset_status status = PIPE_NO_RESET;
   struct drm_i915_reset_stats stats = { .ctx_id = batch->hw_ctx_id };

   if (intel_ioctl(screen->fd, DRM_IOCTL_I915_GET_RESET_STATS, &stats))
      DBG("DRM_IOCTL_I915_GET_RESET_STATS failed: %s\n", strerror(errno));

   if (stats.batch_active != 0) {
      /* A reset was observed while a batch from this hardware context was
       * executing.  Assume that this context was at fault.
       */
      status = PIPE_GUILTY_CONTEXT_RESET;
   } else if (stats.batch_pending != 0) {
      /* A reset was observed while a batch from this context was in progress,
       * but the batch was not executing.  In this case, assume that the
       * context was not at fault.
       */
      status = PIPE_INNOCENT_CONTEXT_RESET;
   }

   if (status != PIPE_NO_RESET) {
      /* Our context is likely banned, or at least in an unknown state.
       * Throw it away and start with a fresh context.  Ideally this may
       * catch the problem before our next execbuf fails with -EIO.
       */
      replace_hw_ctx(batch);
   }

   return status;
}

static void
move_syncobj_to_batch(struct iris_batch *batch,
                      struct iris_syncobj **p_syncobj,
                      unsigned flags)
{
   struct iris_bufmgr *bufmgr = batch->screen->bufmgr;

   if (!*p_syncobj)
      return;

   bool found = false;
   util_dynarray_foreach(&batch->syncobjs, struct iris_syncobj *, s) {
      if (*p_syncobj == *s) {
         found = true;
         break;
      }
   }

   if (!found)
      iris_batch_add_syncobj(batch, *p_syncobj, flags);

   iris_syncobj_reference(bufmgr, p_syncobj, NULL);
}

static void
update_bo_syncobjs(struct iris_batch *batch, struct iris_bo *bo, bool write)
{
   struct iris_screen *screen = batch->screen;
   struct iris_bufmgr *bufmgr = screen->bufmgr;

   /* Make sure bo->deps is big enough */
   if (screen->id >= bo->deps_size) {
      int new_size = screen->id + 1;
      bo->deps= realloc(bo->deps, new_size * sizeof(bo->deps[0]));
      memset(&bo->deps[bo->deps_size], 0,
             sizeof(bo->deps[0]) * (new_size - bo->deps_size));

      bo->deps_size = new_size;
   }

   /* When it comes to execbuf submission of non-shared buffers, we only need
    * to care about the reads and writes done by the other batches of our own
    * screen, and we also don't care about the reads and writes done by our
    * own batch, although we need to track them. Just note that other places of
    * our code may need to care about all the operations done by every batch
    * on every screen.
    */
   struct iris_bo_screen_deps *deps = &bo->deps[screen->id];
   int batch_idx = batch->name;

#if IRIS_BATCH_COUNT == 2
   /* Due to the above, we exploit the fact that IRIS_NUM_BATCHES is actually
    * 2, which means there's only one other batch we need to care about.
    */
   int other_batch_idx = 1 - batch_idx;
#else
   /* For IRIS_BATCH_COUNT == 3 we can do:
    *   int other_batch_idxs[IRIS_BATCH_COUNT - 1] = {
    *      (batch_idx ^ 1) & 1,
    *      (batch_idx ^ 2) & 2,
    *   };
    * For IRIS_BATCH_COUNT == 4 we can do:
    *   int other_batch_idxs[IRIS_BATCH_COUNT - 1] = {
    *      (batch_idx + 1) & 3,
    *      (batch_idx + 2) & 3,
    *      (batch_idx + 3) & 3,
    *   };
    */
#error "Implement me."
#endif

   /* If it is being written to by others, wait on it. */
   if (deps->write_syncobjs[other_batch_idx])
      move_syncobj_to_batch(batch, &deps->write_syncobjs[other_batch_idx],
                            I915_EXEC_FENCE_WAIT);

   /* If it's being written by our screen, wait on it too. This is relevant
    * when there are multiple contexts on the same screen. */
   if (deps->write_syncobjs[batch_idx])
      move_syncobj_to_batch(batch, &deps->write_syncobjs[batch_idx],
                            I915_EXEC_FENCE_WAIT);

   struct iris_syncobj *batch_syncobj = iris_batch_get_signal_syncobj(batch);

   if (write) {
      /* If we're writing to it, set our batch's syncobj as write_syncobj so
       * others can wait on us. Also wait every reader we care about before
       * writing.
       */
      iris_syncobj_reference(bufmgr, &deps->write_syncobjs[batch_idx],
                              batch_syncobj);

      move_syncobj_to_batch(batch, &deps->read_syncobjs[other_batch_idx],
                           I915_EXEC_FENCE_WAIT);
      move_syncobj_to_batch(batch, &deps->read_syncobjs[batch_idx],
                           I915_EXEC_FENCE_WAIT);

   } else {
      /* If we're reading, replace the other read from our batch index. */
      iris_syncobj_reference(bufmgr, &deps->read_syncobjs[batch_idx],
                             batch_syncobj);
   }
}

static void
update_batch_syncobjs(struct iris_batch *batch)
{
   struct iris_bufmgr *bufmgr = batch->screen->bufmgr;
   simple_mtx_t *bo_deps_lock = iris_bufmgr_get_bo_deps_lock(bufmgr);

   simple_mtx_lock(bo_deps_lock);

   for (int i = 0; i < batch->exec_count; i++) {
      struct iris_bo *bo = batch->exec_bos[i];
      bool write = BITSET_TEST(batch->bos_written, i);

      if (bo == batch->screen->workaround_bo)
         continue;

      update_bo_syncobjs(batch, bo, write);
   }
   simple_mtx_unlock(bo_deps_lock);
}

/**
 * Submit the batch to the GPU via execbuffer2.
 */
static int
submit_batch(struct iris_batch *batch)
{
   iris_bo_unmap(batch->bo);

   struct drm_i915_gem_exec_object2 *validation_list =
      malloc(batch->exec_count * sizeof(*validation_list));

   unsigned *index_for_handle =
      calloc(batch->max_gem_handle + 1, sizeof(unsigned));

   unsigned validation_count = 0;
   for (int i = 0; i < batch->exec_count; i++) {
      struct iris_bo *bo = iris_get_backing_bo(batch->exec_bos[i]);
      assert(bo->gem_handle != 0);

      bool written = BITSET_TEST(batch->bos_written, i);
      unsigned prev_index = index_for_handle[bo->gem_handle];
      if (prev_index > 0) {
         if (written)
            validation_list[prev_index].flags |= EXEC_OBJECT_WRITE;
      } else {
         index_for_handle[bo->gem_handle] = validation_count;
         validation_list[validation_count] =
            (struct drm_i915_gem_exec_object2) {
               .handle = bo->gem_handle,
               .offset = bo->address,
               .flags  = bo->real.kflags | (written ? EXEC_OBJECT_WRITE : 0) |
                         (iris_bo_is_external(bo) ? 0 : EXEC_OBJECT_ASYNC),
            };
         ++validation_count;
      }
   }

   free(index_for_handle);

   if (INTEL_DEBUG(DEBUG_BATCH | DEBUG_SUBMIT)) {
      dump_fence_list(batch);
      dump_bo_list(batch);
   }

   if (INTEL_DEBUG(DEBUG_BATCH)) {
      decode_batch(batch);
   }

   /* The requirement for using I915_EXEC_NO_RELOC are:
    *
    *   The addresses written in the objects must match the corresponding
    *   reloc.address which in turn must match the corresponding
    *   execobject.offset.
    *
    *   Any render targets written to in the batch must be flagged with
    *   EXEC_OBJECT_WRITE.
    *
    *   To avoid stalling, execobject.offset should match the current
    *   address of that object within the active context.
    */
   struct drm_i915_gem_execbuffer2 execbuf = {
      .buffers_ptr = (uintptr_t) validation_list,
      .buffer_count = validation_count,
      .batch_start_offset = 0,
      /* This must be QWord aligned. */
      .batch_len = ALIGN(batch->primary_batch_size, 8),
      .flags = I915_EXEC_RENDER |
               I915_EXEC_NO_RELOC |
               I915_EXEC_BATCH_FIRST |
               I915_EXEC_HANDLE_LUT,
      .rsvd1 = batch->hw_ctx_id, /* rsvd1 is actually the context ID */
   };

   if (num_fences(batch)) {
      execbuf.flags |= I915_EXEC_FENCE_ARRAY;
      execbuf.num_cliprects = num_fences(batch);
      execbuf.cliprects_ptr =
         (uintptr_t)util_dynarray_begin(&batch->exec_fences);
   }

   int ret = 0;
   if (!batch->screen->devinfo.no_hw &&
       intel_ioctl(batch->screen->fd, DRM_IOCTL_I915_GEM_EXECBUFFER2, &execbuf))
      ret = -errno;

   for (int i = 0; i < batch->exec_count; i++) {
      struct iris_bo *bo = batch->exec_bos[i];

      bo->idle = false;
      bo->index = -1;

      iris_get_backing_bo(bo)->idle = false;

      iris_bo_unreference(bo);
   }

   free(validation_list);

   return ret;
}

static const char *
batch_name_to_string(enum iris_batch_name name)
{
   const char *names[IRIS_BATCH_COUNT] = {
      [IRIS_BATCH_RENDER]  = "render",
      [IRIS_BATCH_COMPUTE] = "compute",
   };
   return names[name];
}

/**
 * Flush the batch buffer, submitting it to the GPU and resetting it so
 * we're ready to emit the next batch.
 */
void
_iris_batch_flush(struct iris_batch *batch, const char *file, int line)
{
   struct iris_screen *screen = batch->screen;

   /* If a fence signals we need to flush it. */
   if (iris_batch_bytes_used(batch) == 0 && !batch->contains_fence_signal)
      return;

   iris_measure_batch_end(batch->ice, batch);

   iris_finish_batch(batch);

   update_batch_syncobjs(batch);

   if (INTEL_DEBUG(DEBUG_BATCH | DEBUG_SUBMIT | DEBUG_PIPE_CONTROL)) {
      const char *basefile = strstr(file, "iris/");
      if (basefile)
         file = basefile + 5;

      fprintf(stderr, "%19s:%-3d: %s batch [%u] flush with %5db (%0.1f%%) "
              "(cmds), %4d BOs (%0.1fMb aperture)\n",
              file, line, batch_name_to_string(batch->name), batch->hw_ctx_id,
              batch->total_chained_batch_size,
              100.0f * batch->total_chained_batch_size / BATCH_SZ,
              batch->exec_count,
              (float) batch->aperture_space / (1024 * 1024));

   }

   int ret = submit_batch(batch);

   /* When batch submission fails, our end-of-batch syncobj remains
    * unsignalled, and in fact is not even considered submitted.
    *
    * In the hang recovery case (-EIO) or -ENOMEM, we recreate our context and
    * attempt to carry on.  In that case, we need to signal our syncobj,
    * dubiously claiming that this batch completed, because future batches may
    * depend on it.  If we don't, then execbuf would fail with -EINVAL for
    * those batches, because they depend on a syncobj that's considered to be
    * "never submitted".  This would lead to an abort().  So here, we signal
    * the failing batch's syncobj to try and allow further progress to be
    * made, knowing we may have broken our dependency tracking.
    */
   if (ret < 0)
      iris_syncobj_signal(screen->bufmgr, iris_batch_get_signal_syncobj(batch));

   batch->exec_count = 0;
   batch->max_gem_handle = 0;
   batch->aperture_space = 0;

   util_dynarray_foreach(&batch->syncobjs, struct iris_syncobj *, s)
      iris_syncobj_reference(screen->bufmgr, s, NULL);
   util_dynarray_clear(&batch->syncobjs);

   util_dynarray_clear(&batch->exec_fences);

   if (INTEL_DEBUG(DEBUG_SYNC)) {
      dbg_printf("waiting for idle\n");
      iris_bo_wait_rendering(batch->bo); /* if execbuf failed; this is a nop */
   }

   /* Start a new batch buffer. */
   iris_batch_reset(batch);

   /* EIO means our context is banned.  In this case, try and replace it
    * with a new logical context, and inform iris_context that all state
    * has been lost and needs to be re-initialized.  If this succeeds,
    * dubiously claim success...
    * Also handle ENOMEM here.
    */
   if ((ret == -EIO || ret == -ENOMEM) && replace_hw_ctx(batch)) {
      if (batch->reset->reset) {
         /* Tell gallium frontends the device is lost and it was our fault. */
         batch->reset->reset(batch->reset->data, PIPE_GUILTY_CONTEXT_RESET);
      }

      ret = 0;
   }

   if (ret < 0) {
#ifdef DEBUG
      const bool color = INTEL_DEBUG(DEBUG_COLOR);
      fprintf(stderr, "%siris: Failed to submit batchbuffer: %-80s%s\n",
              color ? "\e[1;41m" : "", strerror(-ret), color ? "\e[0m" : "");
#endif
      abort();
   }
}

/**
 * Does the current batch refer to the given BO?
 *
 * (In other words, is the BO in the current batch's validation list?)
 */
bool
iris_batch_references(struct iris_batch *batch, struct iris_bo *bo)
{
   return find_exec_index(batch, bo) != -1;
}

/**
 * Updates the state of the noop feature.  Returns true if there was a noop
 * transition that led to state invalidation.
 */
bool
iris_batch_prepare_noop(struct iris_batch *batch, bool noop_enable)
{
   if (batch->noop_enabled == noop_enable)
      return 0;

   batch->noop_enabled = noop_enable;

   iris_batch_flush(batch);

   /* If the batch was empty, flush had no effect, so insert our noop. */
   if (iris_batch_bytes_used(batch) == 0)
      iris_batch_maybe_noop(batch);

   /* We only need to update the entire state if we transition from noop ->
    * not-noop.
    */
   return !batch->noop_enabled;
}