package tlog

import (
	"fmt"
	"strconv"
	"strings"
)

// A Tile is a description of a transparency log tile.
// A tile of height H at level L offset N lists W consecutive hashes
// at level H*L of the tree starting at offset N*(2**H).
// A complete tile lists 2**H hashes; a partial tile lists fewer.
// Note that a tile represents the entire subtree of height H
// with those hashes as the leaves. The levels above H*L

	return "/" + tile.Path()
}

// readTile reads a single tile, either from the on-disk cache or the server.
func (c *Client) readTile(tile tlog.Tile) ([]byte, error) {
	type cached struct {
		data []byte
		err  error
	}

	result := c.tileCache.Do(tile, func() interface{} {
		// Try the requested tile in on-disk cache.
		data, err := c.ops.ReadCache(c.tileCacheKey(tile))

    }
  } else {
    output_type = input_type;
  }

  return builder->create<mlir::TF::ConcatOp>(
      location, output_type, concat_dimension_op.getOutput(), inputs);
}

// For tile sharded inputs to TPU computation, inject split op between the
// input values and TPU computation so that tiled input values are passed in
// as inputs to TPU computations. If more than one dimension is sharded, then

  %177 = "tf.Reshape"(%176, %9) {device = ""} : (tensor<i32>, tensor<0xi32>) -> tensor<i32>
  %178 = "tf.Pack"(%7, %177) {axis = 0 : i64, device = ""} : (tensor<i32>, tensor<i32>) -> tensor<2xi32>
  %179 = "tf.Tile"(%115, %178) {device = ""} : (tensor<?x1xi64>, tensor<2xi32>) -> tensor<?x?xi64>
  %180 = "tf.Mul"(%177, %118) {device = ""} : (tensor<i32>, tensor<i32>) -> tensor<i32>

  %multiples = "tf.Const"() { device = "/device:CPU:0", value = dense<[7,3]> : tensor<2xi32> } : () -> tensor<2xi32>
  // CHECK: T = !corert.string
  %output = "tf.Tile"(%arg, %multiples) { device = "/device:CPU:0" } : (tensor<1x2x!tf_type.string>, tensor<2xi32>) -> tensor<?x6x!tf_type.string>
  func.return %output : tensor<?x6x!tf_type.string>
}

// CHECK-LABEL: func @shape

// A sharding with OTHER type may be REPLICATED if:
// 'replicate_on_last_tile_dim' is true OR
// 'last_tile_dims' is not empty
// AND
// other than replicated last tile dims, all other dims are not sharded.
bool IsOtherReplicatedSharding(const xla::OpSharding& xla_sharding);

// Returns whether the sharding is split sharding. i.e. A sharding with OTHER
// type but not replicated.

  %2 = "tf.ExpandDims"(%1, %const_0) : (tensor<10000xi32>, tensor<i32>) -> tensor<1x10000xi32>
  %3 = "tf.Tile"(%2, %const_64_1) : (tensor<1x10000xi32>, tensor<2xi32>) -> tensor<64x10000xi32>
  // CHECK-NOT: tf.Range
  // CHECK-NOT: tf.ExpandDims
  // CHECK: [[TILE:%.*]] = "tf.Tile"
  // CHECK: return [[TILE]]
  func.return %3 : tensor<64x10000xi32>
}

	HasAVX512BITALG     bool // Advanced vector extension 512 Bit Algorithms
	HasAVX512BF16       bool // Advanced vector extension 512 BFloat16 Instructions
	HasAMXTile          bool // Advanced Matrix Extension Tile instructions
	HasAMXInt8          bool // Advanced Matrix Extension Int8 instructions
	HasAMXBF16          bool // Advanced Matrix Extension BFloat16 instructions
	HasBMI1             bool // Bit manipulation instruction set 1

  "TFL::HasOneUseOrUsedByOnlyBinaryOps($0)">>;

// Pattern for skipping Tile if it is mainly for broadcasting and the
// Op is already supporting broadcasting.
multiclass FuseTileBroadcastIntoFollowingBinary<Op BinaryOp> {
  def FuseTileBroadcastToBinaryOp1#BinaryOp : Pat<
    (BinaryOp:$result (TFL_TileOp $input, (Arith_ConstantOp $tile)),
     $operand, $act_func),
    (BinaryOp $input, $operand, $act_func),

		if vsphereVersion != "" {
			uaAppend(" vsphere-plugin-", vsphereVersion)
		}
	}

	if IsPCFTile() {
		pcfTileVersion := env.Get("MINIO_PCF_TILE_VERSION", "")
		if pcfTileVersion != "" {
			uaAppend(" pcf-tile-", pcfTileVersion)
		}
	}
	uaAppend("; ", "")

	if cpus, err := gopsutilcpu.Info(); err == nil && len(cpus) > 0 {
		cpuMap := make(map[string]struct{}, len(cpus))