'mhlo' Dialecte

Opérations

mhlo.abs (mhlo :: AbsOp)

Opération abdominaux

Syntaxe:

operation ::= `mhlo.abs` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Effectue une opération abs par élément sur le tenseur operand et produit un tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#abs

Exemple:

%result = mhlo.abs %operand : tensor<3xi32>

Traits : AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.add (mhlo :: AddOp)

Ajouter une opération

Syntaxe:

operation ::= `mhlo.add` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Effectue l'addition élément par élément de deux tenseurs lhs et rhs et produit un tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#add

Exemple:

%result = mhlo.add %lhs, %rhs : tensor<2x2xi32>

Traits : AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.add_dependency (mhlo :: AddDependencyOp)

Opération AddDependency

Syntaxe:

operation ::= `mhlo.add_dependency` operands attr-dict `:` functional-type(operands, results)

Cette opération est privée au compilateur XLA, elle n'a donc pas encore de spécification.

De manière informelle, cette opération comporte deux opérandes : un opérande de données et un jeton. Le résultat de l’opération est l’opérande de données. Lorsqu'elle est utilisée avec AfterAll, cette opération permet de commander des opérations sans effets secondaires (celles qui ne produisent pas de valeurs de jeton).

Exemple:

%1 = mhlo.add_dependency %arg0, %0 : (tensor<3x4xf32>, !mhlo.token) -> tensor<3x4xf32>

Traits : AlwaysSpeculatableImplTrait

Interfaces : ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.after_all (mhlo::AfterAllOp)

Opération AfterAll

Syntaxe:

operation ::= `mhlo.after_all` $inputs attr-dict
              `:` custom<VariadicSameOperandsAndResultType>(ref($inputs), type($inputs), type($result))

Garantit que les opérations produisant les inputs sont exécutées avant toute opération qui dépend du result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#after_all

Exemple:

%result = mhlo.after_all %input0, %input1 : !mhlo.token

Traits : AlwaysSpeculatableImplTrait

Interfaces : ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.all_gather (mhlo::AllGatherOp)

Opération AllGather

Au sein de chaque groupe de processus de la grille de processus, concatène les valeurs du tenseur d'opérande de chaque processus le long de all_gather_dim et produit un tenseur de résultat. Le computation est appliqué séparément pour chaque opérande dans operands , produisant un résultat par opérande.

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#all_gather

Exemple:

%result = "mhlo.all_gather"(%operand) {
  all_gather_dim = 1 : i64,
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>
  // channel_id = 0
  channel_handle = #mhlo.channel_handle<handle = 0, type = 0>,
  // use_global_device_ids = false
} : (tensor<2x2xf32>) -> tensor<2x4xf32>

Traits : SameOperandsAndResultElementType

Les attributs:

Opérandes :

Résultats:

mhlo.all_reduce (mhlo :: AllReduceOp)

Opération AllReduce

Au sein de chaque groupe de processus dans la grille de processus, applique un computation de fonction de réduction aux valeurs d'un tenseur d'opérande de chaque processus et produit un tenseur de résultat. Le computation est appliqué séparément pour chaque opérande dans operands , produisant un résultat par opérande.

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#all_reduce

Exemple:

%result = "mhlo.all_reduce"(%operand) ({
  ^bb0(%arg0: tensor<f32>, %arg1: tensor<f32>):
    %0 = mhlo.add %arg1, %arg2 : tensor<f32>
    mhlo.return %0 : tensor<f32>
}) {
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>
  // channel_id = 0
  channel_handle = #mhlo.channel_handle<handle = 0, type = 0>
  // use_global_device_ids = false
} : (tensor<4xf32>) -> tensor<4xf32>

Traits : InferTensorType , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces : InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Opérandes :

Résultats:

mhlo.all_to_all (mhlo::AllToAllOp)

Opération AllToAll

Au sein de chaque groupe de processus dans la grille de processus, divise les valeurs du tenseur operand le long split_dimension en parties, disperse les parties divisées entre les processus, concatène les parties dispersées le long concat_dimension et produit un tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#all_to_all

Exemple:

%result = "mhlo.all_to_all"(%operand) {
  split_dimension = 1 : i64,
  concat_dimension = 0 : i64,
  split_count = 2 : i64,
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>
} : (tensor<2x4xf32>) -> tensor<4x2xf32>

Traits : AlwaysSpeculatableImplTrait , InferTensorType , SameOperandsElementType , SameOperandsShape , SameVariadicOperandSize

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Les attributs:

Opérandes :

Résultats:

mhlo.and (mhlo::AndOp)

Et le fonctionnement

Syntaxe:

operation ::= `mhlo.and` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Effectue un ET par élément de deux tenseurs lhs et rhs et produit un tenseur result

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#and

Exemple:

%result = mhlo.and %lhs, %rhs : tensor<2x2xi32>

Traits : AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.async_done (mhlo :: AsyncDoneOp)

Opération AsyncDone

Cette opération est privée au compilateur XLA, elle n'a donc pas encore de spécification.

De manière informelle, cette opération bloque jusqu'à la fin d'un calcul asynchrone. Il renvoie le résultat final du calcul asynchrone.

Consultez la documentation d'AsyncStart pour plus d'informations.

Interfaces : InferTypeOpInterface

Les attributs:

Opérandes :

Résultats:

mhlo.async_start (mhlo :: AsyncStartOp)

Opération AsyncStart

Cette opération est privée au compilateur XLA, elle n'a donc pas encore de spécification.

De manière informelle, cette opération lance un calcul asynchrone.

Ceci est utilisé lorsqu'il existe des fonctions qui contiennent à la fois des attentes asynchrones (telles que les DMA) et des calculs sur thread. Par exemple, une fonction peut consister en un calcul, un DMA, un autre calcul, un deuxième DMA et un calcul final. Cela serait représenté par un async_start suivi d'un async_update et d'un async_done. Le async_start effectuerait le premier calcul sur le thread, puis démarrerait le DMA. async_update attendrait la fin du DMA si ce n'était pas encore fait, puis exécuterait le deuxième calcul dans la fonction et démarrerait le deuxième DMA. Enfin, async_done attendrait ce dernier DMA, puis exécuterait le dernier calcul qui doit être exécuté sur le thread et renverrait le résultat de ce calcul final.

operands sont passés au calcul directement called_computation est la fonction qui sera exécutée de manière asynchrone execution_thread est le nom du thread dans lequel elle sera exécutée. Le thread principal est appelé « principal ». Tous les sujets ont des noms.

Cela renvoie tout l'état nécessaire entre les opérations asynchrones. Après l'affectation du tampon, les valeurs de retour représentent l'espace nécessaire pour contenir l'entrée, les résultats et tous les blocs-notes nécessaires ou modifiés par l'opération asynchrone.

Les attributs:

Opérandes :

Résultats:

mhlo.async_update (mhlo :: AsyncUpdateOp)

Opération AsyncUpdate

Cette opération est privée au compilateur XLA, elle n'a donc pas encore de spécification.

De manière informelle, cette opération bloque un calcul asynchrone jusqu'à une barrière de synchronisation. Cela renvoie bundle après avoir fonctionné dessus.

Consultez la documentation d'AsyncStart pour plus d'informations.

Interfaces : InferTypeOpInterface

Les attributs:

Opérandes :

Résultats:

mhlo.atan2 (mhlo::Atan2Op)

Opération Atan2

Syntaxe:

operation ::= `mhlo.atan2` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Effectue une opération atan2 par élément sur les tenseurs lhs et rhs et produit un tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#atan2

Exemple:

%result = mhlo.atan2 %lhs, %rhs : tensor<3xf32>

Traits : AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.batch_norm_grad (mhlo :: BatchNormGradOp)

Opération BatchNormGrad

Calcule les gradients de plusieurs entrées de BatchNormTrainingOp rétropropagant à partir de grad_output et produit les tenseurs grad_operand , grad_scale et grad_offset .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#batch_norm_grad

Exemple:

%grad_operand, %grad_scale, %grad_offset =
"mhlo.batch_norm_grad"(%operand, %scale, %mean, %variance, %grad_output) {
  epsilon = 0.0 : f32,
  feature_index = 2 : i64
} : (tensor<2x2x2xf32>, tensor<2xf32>, tensor<2xf32>, tensor<2xf32>,
    tensor<2x2x2xf32>) -> (tensor<2x2x2xf32>, tensor<2xf32>, tensor<2xf32>)

Traits : AlwaysSpeculatableImplTrait , InferTensorType

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Les attributs:

Opérandes :

Résultats:

mhlo.batch_norm_inference (mhlo :: BatchNormInferenceOp)

Opération BatchNormInference

Normalise le tenseur operand sur toutes les dimensions à l'exception de la dimension feature_index et produit un tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#batch_norm_inference

Exemple:

%result = "mhlo.batch_norm_inference"(%operand, %scale, %offset, %mean, %variance) {
  epsilon = 0.0 : f32,
  feature_index = 2 : i64
} : (tensor<2x2x2xf32>, tensor<2xf32>, tensor<2xf32>, tensor<2xf32>, tensor<2xf32>) -> tensor<2x2x2xf32>

Traits : AlwaysSpeculatableImplTrait , InferTensorType

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Les attributs:

Opérandes :

Résultats:

mhlo.batch_norm_training (mhlo :: BatchNormTrainingOp)

Opération BatchNormTraining

Calcule la moyenne et la variance entre les dimensions de lot et spatiales et normalise le tenseur operand , pour chaque caractéristique de la dimension feature_index et produit les tenseurs output , batch_mean et batch_var .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#batch_norm_training

Exemple:

%output, %batch_mean, %batch_var = "mhlo.batch_norm_training"(%operand, %scale, %offset) {
  epsilon = 0.0 : f32,
  feature_index = 2 : i64
} : (tensor<2x2x2xf32>, tensor<2xf32>, tensor<2xf32>) -> (tensor<2x2x2xf32>, tensor<2xf32>, tensor<2xf32>)

Traits : AlwaysSpeculatableImplTrait , InferTensorType

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Les attributs:

Opérandes :

Résultats:

mhlo.bitcast (mhlo :: BitcastOp)

Opération de diffusion de bits

Syntaxe:

operation ::= `mhlo.bitcast` operands attr-dict `:` functional-type(operands, results)

Cette opération est privée au compilateur XLA, elle n'a donc pas encore de spécification.

De manière informelle, cette opération modifie la forme de l'entrée de manière à ce que la disposition physique des éléments reste inchangée.

Cette opération nécessite des informations de mise en page pour donner un sens à « la disposition physique des éléments », et la prise en charge de la mise en page dans MHLO est actuellement en cours de développement.

Exemple:

%0 = mhlo.bitcast %arg0 : (tensor<3x4xf32>) -> tensor<3x4x1xf32>

Traits : AlwaysSpeculatableImplTrait

Interfaces : ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.bitcast_convert (mhlo :: BitcastConvertOp)

Opération BitcastConvert

Syntaxe:

operation ::= `mhlo.bitcast_convert` operands attr-dict `:` functional-type(operands, results)

Effectue une opération bitcast sur le tenseur operand et produit un tenseur result dans lequel les bits de l'ensemble du tenseur operand sont réinterprétés en utilisant le type du tenseur result .

Voir : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#bitcast_convert

Exemple:

%result = mhlo.bitcast_convert %operand : (tensor<2xf32>) -> tensor<2x4xi8>

Traits : AlwaysSpeculatableImplTrait

Interfaces : ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets : MemoryEffects::Effect{}

Opérandes :

Résultats:

mhlo.broadcast (Mhlo :: Broadcastop)

Opération de diffusion

Cette opération est en train de sortir de Stablehlo, donc elle n'est pas incluse dans la spécification: https://github.com/openxla/stablehlo/issues/3

De manière informelle, cette opération fait la même chose que la diffusion de XLA: https://www.tensorflow.org/xla/operation_semantics#broadcast

Exemple:

%result = mhlo.broadcast %operand, sizes = [1, 2] : (tensor<3xi32>) -> tensor<1x2x3xi32>

TRAITS: AlwaysSpeculatableImplTrait , InferTensorType , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.broadcast_in_dim (Mhlo :: BroadcastIndImop)

Opération de BroadcastIndim

Étend les dimensions et / ou le rang d'un tenseur d'entrée en dupliquant les données dans le tenseur operand et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#broadcast_in_dim

Exemple:

%result = mhlo.broadcast_in_dim %operand, dims = [2, 1] : (tensor<1x3xi32>) -> tensor<2x3x2xi32>

TRAITS: AlwaysSpeculatableImplTrait , HLO_CompatibleOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.case (Mhlo :: Caseop)

Opération de cas

Produit la sortie de l'exécution exactement une function à partir branches en fonction de la valeur de index .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#case

Exemple:

%result0, %result1 = "mhlo.case"(%index) ({
  mhlo.return %result_branch0, %result_branch0 : tensor<2xi64>, tensor<2xi64>
}, {
  mhlo.return %result_branch1, %result_branch1 : tensor<2xi64>, tensor<2xi64>
}) : (tensor<i32>) -> (tensor<2xi64>, tensor<2xi64>)

Traits: RecursiveMemoryEffects , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferTypeOpInterface

Opérandes:

Résultats:

mhlo.cbrt (mhlo :: cbrtop)

Opération CBRT

Syntaxe:

operation ::= `mhlo.cbrt` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Effectue le fonctionnement de la racine cubique sur l'élément sur le tenseur operand et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#cbrt

Exemple:

%result = mhlo.cbrt %operand : tensor<4xf32>

TRAITS: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.ceil (Mhlo :: Ceilop)

Opération de plate-forme

Syntaxe:

operation ::= `mhlo.ceil` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Effectue le plafond du tenseur operand sur le plan des éléments et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#ceil

Exemple:

%result = mhlo.ceil %operand : tensor<5xf32>

TRAITS: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.cholesky (mhlo :: choleskyop)

Fonctionnement de Cholesky

Calcule la décomposition de Cholesky d'un lot de matrices.

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#cholesky

Exemple:

%result = mhlo.cholesky %a, lower = true : tensor<3x3xf32>

TRAITS: AlwaysSpeculatableImplTrait , InferTensorType , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.clamp (Mhlo :: Clampop)

Opération de pince

Syntaxe:

operation ::= `mhlo.clamp` $min `,` $operand `,` $max attr-dict
              `:` custom<SameOperandsAndResultType>(type($min), type($operand), type($max), type($result))

Graque chaque élément du tenseur operand entre une valeur minimale et maximale et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#clamp

Exemple:

%result = mhlo.clamp %min, %operand, %max : tensor<3xi32>

TRAITS: AlwaysSpeculatableImplTrait , HLO_BroadcastingElementwise , InferTensorType , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.collective_broadcast (Mhlo :: CollectiveBroadcastop)

Opération CollectiveBroadcast

Dans chaque groupe de processus de la grille de processus, envoyez la valeur du tenseur operand du processus source aux processus cibles et produisez un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#Collective_broadcast

Exemple:

%result = "mhlo.collective_broadcast"(%operand) {
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>,
  channel_handle = #mhlo.channel_handle<handle = 0, type = 0>
} : (tensor<1x2xi64>) -> tensor<1x2xi64>

Traits: CompatibleOperandsAndResultType

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Opérandes:

Résultats:

mhlo.collective_permute (Mhlo :: CollectivePermuteop)

Opération collectivePermute

Dans chaque groupe de processus de la grille de processus, envoie la valeur du tenseur operand du processus source au processus cible et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#collective_permute

Exemple:

%result = "mhlo.collective_permute"(%operand) {
  source_target_pairs = dense<[[0, 1], [1, 2]]> : tensor<2x2xi64>,
  // channel_id = 0
  channel_handle = #mhlo.channel_handle<handle = 0, type = 0>
} : (tensor<4x2xf32>) -> tensor<4x2xf32>

TRAITS: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.compare (mhlo :: compareop)

Comparez le fonctionnement

Syntaxe:

operation ::= `mhlo.compare` $comparison_direction `,` $lhs `,` $rhs (`,` $compare_type^)?
              attr-dict `:` functional-type(operands, results)

Effectue une comparaison par élément des tenseurs lhs et rhs en fonction de comparison_direction et compare_type , et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#compare

Exemple:

%result = mhlo.compare LT, %lhs, %rhs, FLOAT : (tensor<2xf32>, tensor<2xf32>) -> tensor<2xi1>

TRAITS: AlwaysSpeculatableImplTrait , Elementwise , InferTensorType , SameOperandsAndResultShape , SameOperandsElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.complex (MHLO :: Complexop)

Opération complexe

Syntaxe:

operation ::= `mhlo.complex` operands attr-dict
              `:` custom<ComplexOpType>(type($lhs), type($rhs), type($result))

Effectue une conversion d'élément en une valeur complexe à partir d'une paire de valeurs réelles et imaginaires, lhs et rhs , et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#complex

Exemple:

%result = mhlo.complex %lhs, %rhs : tensor<2xcomplex<f32>>

TRAITS: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape , SameOperandsElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.composite (mhlo :: compositeop)

Opération composite

Syntaxe:

operation ::= `mhlo.composite` $name $inputs attr-dict `:` functional-type(operands, results)

Résume une opération constituée (composée) d'autres opérations stablehlo, en prenant inputs et composite_attributes et en produisant results . La sémantique de l'OP est mise en œuvre par l'attribut decomposition . L'OP composite peut être remplacé par sa décomposition sans changer la sémantique de programme. Dans les cas où l'inclinaison de la décomposition ne fournit pas la même sémantique OP, préférez utiliser custom_call .

Le champ version (par défaut à 0 ) est utilisé pour indiquer le changement de sémantique d'un composite.

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#composite

Exemple:

%results = mhlo.composite "my.op" %arg0, %arg1 {
  decomposition = @my_op,
  composite_attributes = { my_attribute = "my_value" },
  version = 1 : i32
} : (tensor<f32>, tensor<f32>) -> tensor<f32>

Interfaces: SymbolUserOpInterface

Les attributs:

Opérandes:

Résultats:

mhlo.compute_reshape_shape (mhlo :: ComputerShapeshapeop)

Opération de contrebande informatique

Syntaxe:

operation ::= `mhlo.compute_reshape_shape` operands attr-dict `:` functional-type(operands, results)

Cette opération est un travail en cours, il n'est donc pas encore inclus dans la spécification: https://github.com/openxla/stablehlo/issues/8

De manière informelle, cette opération calcule un Output_Shape pour DynamicReshapeop à partir du nombre de num_elements d'éléments dans un opérande de DynamicReshapeop et de la forme dynamic_shape fournies à TF Reshape: https://www.tensorflow.org/api_docs/python/tf/reshape

Par exemple, pour num_elements = 12 et dynamic_shape = [2, -1] , le result va être [2, 6] . Si les opérandes ne sont pas valides (par exemple, si les dimensions ne divisent pas uniformément le nombre d'éléments, ou s'il y a plusieurs valeurs -1 dans les dimensions), cela conduit à un comportement non défini.

Exemple:

%result = mhlo.compute_reshape_shape %num_elements, %dynamic_shape
       : (index, tensor<2xi32>) -> tensor<2xi32>

TRAITS: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.concatenate (mhlo :: concatenateop)

Opération de concaténate

COMPATENNES UN NOMBRE VARIADIQUE DE TENSEURS DANS inputs dimension DIMENDENCE Dans le même ordre que les arguments donnés et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#concatenate

Exemple:

%result = mhlo.concatenate %input0, %input1, dim = 0 : (tensor<3x2xi64>, tensor<1x2xi64>) -> tensor<4x2xi64>

TRAITS: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.constant (Mhlo :: Constantop)

Opération constante

Produit un tenseur output à partir d'une value constante.

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#constant

Exemple:

%output = mhlo.constant dense<[[0.0, 1.0], [2.0, 3.0]]> : tensor<2x2xf32>

Traits: AlwaysSpeculatableImplTrait , ConstantLike

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Résultats:

mhlo.convert (mhlo :: convertop)

Convertir

Syntaxe:

operation ::= `mhlo.convert` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Effectue une conversion d'élément d'un type d'élément à un autre sur le tenseur operand et produit un tenseur result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#convert

Exemple:

%result = mhlo.convert %operand : (tensor<3xi32>) -> tensor<3xcomplex<f32>>

TRAITS: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Opérandes:

Résultats:

mhlo.convolution (Mhlo :: Convolutionop)

Opération de convolution

Syntaxe:

operation ::= `mhlo.convolution` `(`operands`)`
              `dim_numbers` `=` custom<ConvolutionDimensions>($dimension_numbers) `,`
              `window` `=` `{` custom<WindowAttributes>($window_strides, $padding,
              $lhs_dilation, $rhs_dilation,
              $window_reversal) `}`
              attr-dict `:` functional-type(operands, results)

Calcule des produits DOT entre Windows de lhs et des tranches de rhs et produit result .

Voir: https://github.com/openxla/stablehlo/blob/main/docs/spe.md#convolution

Exemple:

%result = "mhlo.convolution"(%lhs, %rhs) {
  window_strides = dense<4> : tensor<2xi64>,
  padding = dense<0> : tensor<2x2xi64>,
  lhs_dilation = dense<2> : tensor<2xi64>,
  rhs_dilation = dense<1> : tensor<2xi64>,
  window_reversal = dense<false> : tensor<2xi1>,
  dimension_numbers = #mhlo.conv<[b, 0, 1, f]x[0, 1, i, o]->[b, 0, 1, f]>,
  feature_group_count = 1 : i64,
  batch_group_count = 1 : i64,
  precision_config = [#stablehlo<precision DEFAULT>, #stablehlo<precision DEFAULT>]
} : (tensor<1x4x4x1xi32>, tensor<3x3x1x1xi32>) -> tensor<1x2x2x1xi32>

TRAITS: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.copy (mhlo :: copyop)

Opération de copie

Syntaxe:

operation ::= `mhlo.copy` operands attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Cette opération est privée dans le compilateur XLA, il n'a donc pas encore de spécification.

De manière informelle, cette opération une copie de operand . Selon les métadonnées attachées à l'opération, il peut se comporter très différemment d'un no-opér.

Exemple:

%0 = mhlo.copy %arg0 : tensor<f32>

TRAITS: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effets: MemoryEffects::Effect{}

Les attributs:

Opérandes:

Résultats:

mhlo.cosine (mhlo::CosineOp)

Cosine operation

Syntaxe:

operation ::= `mhlo.cosine` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise cosine operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#cosine

Exemple:

%result = mhlo.cosine %operand : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.count_leading_zeros (mhlo::ClzOp)

Clz operation

Syntaxe:

operation ::= `mhlo.count_leading_zeros` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise count of the number of leading zero bits in the operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#count_leading_zeros

Exemple:

%result = mhlo.count_leading_zeros %operand : tensor<2x2xi8>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.create_token (mhlo::CreateTokenOp)

CreateToken operation

Syntaxe:

operation ::= `mhlo.create_token` attr-dict `:` type(results)

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as AfterAllOp with 0 inputs: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#after_all

Exemple:

%output = mhlo.create_token : !mhlo.token

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Résultats:

mhlo.cross-replica-sum (mhlo::CrossReplicaSumOp)

CrossReplicaSum operation

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as AllReduceOp with channel_id = 0 , use_global_device_ids = false and computation implementing addition: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#all_reduce

Exemple:

%result = "mhlo.cross-replica-sum"(%operand) {
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>
} : (tensor<4xf32>) -> tensor<4xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.cstr_reshapable (mhlo::CstrReshapableOp)

CstrReshapable operation

Syntaxe:

operation ::= `mhlo.cstr_reshapable` operands attr-dict `:` functional-type(operands, results)

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation creates a witness on the constraint that ComputeReshapeShape would succeed with the provided operands.

Exemple:

%result = mhlo.cstr_reshapable %num_elements, %dynamic_shape
       : (index, tensor<3xi32>) -> !shape.witness

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.custom_call (mhlo::CustomCallOp)

CustomCall operation

Syntaxe:

operation ::= `mhlo.custom_call` custom<CustomCallTarget>($call_target_name) `(` $inputs `)`
              attr-dict `:` functional-type(operands, results)

Encapsulates an implementation-defined operation call_target_name that takes inputs and called_computations and produces results .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#custom_call

Exemple:

%results = "mhlo.custom_call"(%input0) {
  call_target_name = "foo",
  has_side_effect = false,
  backend_config = "bar",
  api_version = 1 : i32,
  called_computations = [@foo]
} : (tensor<f32>) -> tensor<f32>

A custom call invokes code external to XLA. The `inputs` are passed to the
external code, and the external code is expected to produce a result of the
given type. The exact mechanism is backend-specific. For example, in the CPU
backend, a call instruction is emitted which targets a symbol with the name
`call_target_name`.

If XLA runtime is enabled for a backend, then custom calls use the runtime
custom call calling convention to call into the external functions. This
calling convention defines an ABI for encoding arguments, attributes and
results.

Depending on the API version there are two ways to pass extra bits of static
information to the external function:

1. For `API_VERSION_TYPED_FFI` custom calls `backend_config` must be a
   dictionary attribute, that will be encoded according to the custom call
   calling convention and passed to the external function as the attributes
   argument. External code is expected to use declarative bindings (see
   `xla/runtime/custom_call.h`) to decode them at run time. These custom
   calls are only supported if XLA uses XLA runtime.

2. For previous API versions it is the user responsibility to encode extra
   bits of static information as a string `backend_config` attribute, and
   decode it at run time.

Interfaces: MemoryEffectOpInterface

Les attributs:

Operands:

Résultats:

mhlo.divide (mhlo::DivOp)

Div operation

Syntaxe:

operation ::= `mhlo.divide` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise division of dividend lhs and divisor rhs tensors and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#divide

Exemple:

%result = mhlo.divide %lhs, %rhs : tensor<4xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.domain (mhlo::DomainOp)

Domain operation

This operation is private to the XLA compiler, so it is does not yet have a specification.

Informally, these operations are used to group instructions with the same DomainMetadata property. ShardingMetadata is the main use case today to group instructions on the same device. Domain instructions provide two major benefits:

• Prevent unintentionally optimizing instructions across domains.
• Automatically assign the metadata of the instructions created in the domain. Without domain instructions, each HLO optimization pass would have to check and propagate the metadata, which would be easy to miss and also adds complexity to the compiler. Since domain instructions connect two different domains, each domain instruction is associated with two DomainMetadata -- one on the operand side and one on the user side of the domain.

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dot (mhlo::DotOp)

Dot operation

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as XLA's Dot: https://www.tensorflow.org/xla/operation_semantics#dot

Exemple:

%0 = mhlo.dot %arg0, %arg1 : (tensor<1x2xi32>, tensor<2x1xi32>) -> tensor<1x1xi32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dot_general (mhlo::DotGeneralOp)

DotGeneral operation

Computes dot products between slices of lhs and slices of rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#dot_general

Exemple:

%result = "mhlo.dot_general"(%lhs, %rhs) {
  dot_dimension_numbers = #mhlo.dot<
    lhs_batching_dimensions = [0],
    rhs_batching_dimensions = [0],
    lhs_contracting_dimensions = [2],
    rhs_contracting_dimensions = [1]
  >,
  precision_config = [#stablehlo<precision DEFAULT>, #stablehlo<precision DEFAULT>]
} : (tensor<2x2x2xi32>, tensor<2x2x2xi32>) -> tensor<2x2x2xi32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_broadcast_in_dim (mhlo::DynamicBroadcastInDimOp)

DynamicBroadcastInDim operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as BroadcastInDimOp except that the result shape is specified dynamically via output_dimensions : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#broadcast_in_dim

It also accepts optional attributes to express static knowledge about the expanding behavior of dimensions. If not specified, all dimensions are assumed to be possibly expanding. The sets of dimensions that are known to be expanding and the set of dimensions that are known to be non-expanding must be disjoint and they must be a subset of the operand's dimensions.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_conv (mhlo::DynamicConvOp)

DynamicConv operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as ConvolutionOp except that padding is specified dynamically via d_padding : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#convolution

Exemple:

%result = "mhlo.dynamic_conv"(%lhs, %rhs, %d_padding) {
  window_strides = dense<4> : tensor<2xi64>,
  lhs_dilation = dense<2> : tensor<2xi64>,
  rhs_dilation = dense<1> : tensor<2xi64>,
  window_reversal = dense<false> : tensor<2xi1>,
  dimension_numbers = #mhlo.conv<[b, 0, 1, f]x[0, 1, i, o]->[b, 0, 1, f]>,
  feature_group_count = 1 : i64,
  batch_group_count = 1 : i64,
  precision_config = [#stablehlo<precision DEFAULT>, #stablehlo<precision DEFAULT>]
} : (tensor<1x4x4x1xi32>, tensor<3x3x1x1xi32>, tensor<2x2xi64>) -> tensor<1x2x2x1xi32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_gather (mhlo::DynamicGatherOp)

DynamicGather operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as GatherOp except that slice_sizes are specified dynamically: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#gather

Exemple:

%result = "mhlo.dynamic_gather"(%operand, %start_indices, %slice_sizes) {
  dimension_numbers = #mhlo.gather<
    offset_dims = [2, 3],
    collapsed_slice_dims = [0],
    start_index_map = [0, 2],
    index_vector_dim = 2>,
  indices_are_sorted = false
} : (tensor<3x4x2xi32>, tensor<2x3x2xi64>, tensor<3xi64>) -> tensor<2x3x2x2xi32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_iota (mhlo::DynamicIotaOp)

DynamicIota operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as IotaOp except that the result shape is specified dynamically via output_shape : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#iota

Exemple:

%0 = mhlo.dynamic_iota %arg0, dim = 0 : (tensor<1xindex>) -> tensor<4xi32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_pad (mhlo::DynamicPadOp)

DynamicPad operation

Syntaxe:

operation ::= `mhlo.dynamic_pad` operands attr-dict `:` functional-type(operands, results)

Dynamically Pads the operand , with amount of padding added at low-end/high-end/interior is passed through input tensors.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.dynamic_reshape (mhlo::DynamicReshapeOp)

DynamicReshape operation

Syntaxe:

operation ::= `mhlo.dynamic_reshape` operands attr-dict `:` functional-type(operands, results)

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as ReshapeOp except that the result shape is specified dynamically via output_shape : https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reshape

Exemple:

%0 = mhlo.dynamic_reshape %arg0, %shape : (tensor<?xf32>, tensor<2xindex>) -> tensor<?x?xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.dynamic_slice (mhlo::DynamicSliceOp)

DynamicSlice operation

Extracts a slice from the operand using dynamically-computed starting indices and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#dynamic_slice

Exemple:

%result = mhlo.dynamic_slice %operand, %start_indices0, %start_indices1, sizes = [2, 2]
  : (tensor<4x4xi32>, tensor<i64>, tensor<i64>) -> tensor<2x2xi32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.dynamic_update_slice (mhlo::DynamicUpdateSliceOp)

DynamicUpdateSlice operation

Syntaxe:

operation ::= `mhlo.dynamic_update_slice` operands attr-dict `:` functional-type(operands, results)

Produces a result tensor which is equal to the operand tensor except that the slice starting at start_indices is updated with the values in update .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#dynamic_update_slice

Exemple:

%result = mhlo.dynamic_update_slice %operand, %update, %start_indices0, %start_indices1
  : (tensor<4x4xi32>, tensor<2x2xi32>, tensor<i64>, tensor<i64>) -> tensor<4x4xi32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.einsum (mhlo::EinsumOp)

Einsum operation

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as TF's einsum: https://www.tensorflow.org/api_docs/python/tf/einsum

Exemple:

%result = "mhlo.einsum"(%lhs, %rhs) {
  einsum_config = "ab,bc->ac"
} : (tensor<4x16xf32>, tensor<16x4xf32>) -> tensor<4x4xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.erf (mhlo::ErfOp)

Erf operation

Syntaxe:

operation ::= `mhlo.erf` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise erf operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#erf

Exemple:

%result = mhlo.erf %operand : tensor<2x2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.exponential (mhlo::ExpOp)

Exp operation

Syntaxe:

operation ::= `mhlo.exponential` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise exponential operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#exponential

Exemple:

%result = mhlo.exponential %operand : tensor<2x2xf64>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.exponential_minus_one (mhlo::Expm1Op)

Expm1 operation

Syntaxe:

operation ::= `mhlo.exponential_minus_one` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise exponential minus one operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#exponential_minus_one

Exemple:

%result = mhlo.exponential_minus_one %operand : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.fft (mhlo::FftOp)

Fft operation

Performs the forward and inverse Fourier transforms for real and complex inputs/outputs.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#fft

Exemple:

%result = mhlo.fft %operand, type = FFT, length = [4] : (tensor<4xcomplex<f32>>) -> tensor<4xcomplex<f32>>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.floor (mhlo::FloorOp)

Floor operation

Syntaxe:

operation ::= `mhlo.floor` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise floor of operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#floor

Exemple:

%result = mhlo.floor %operand : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.fusion (mhlo::FusionOp)

Fusion operation

This operation is private to the XLA compiler, so it is does not yet have a specification.

Informally, this operation consists of a group of basic ops (represented as a region attached to it). It serves as a hint to the backend that it is beneficial to emit the contained ops into a single loop nest or kernel.

Les attributs:

Operands:

Résultats:

mhlo.gather (mhlo::GatherOp)

Gather operation

Gathers slices from operand tensor from offsets specified in start_indices and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#gather

Exemple:

%result = "mhlo.gather"(%operand, %start_indices) {
  dimension_numbers = #mhlo.gather<
    offset_dims = [2, 3],
    collapsed_slice_dims = [0],
    start_index_map = [0, 2],
    index_vector_dim = 2>,
  slice_sizes = dense<[0, 2, 2]> : tensor<3xi64>,
  indices_are_sorted = false
} : (tensor<3x4x2xi32>, tensor<2x3x2xi64>) -> tensor<2x3x2x2xi32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.get_dimension_size (mhlo::GetDimensionSizeOp)

GetDimensionSize operation

Produces the size of the given dimension of the operand .

See https://github.com/openxla/stablehlo/blob/main/docs/spec.md#get_dimension_size

Exemple:

%result = mhlo.get_dimension_size %operand, dim = 1 : (tensor<2x3xf32>) -> tensor<i32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.get_tuple_element (mhlo::GetTupleElementOp)

GetTupleElement operation

Syntaxe:

operation ::= `mhlo.get_tuple_element` $operand `[` $index `]` attr-dict `:` functional-type(operands, results)

Extracts element at index position of the operand tuple and produces a result .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#get_tuple_element

Exemple:

%result = mhlo.get_tuple_element %operand[0] : (tuple<tensor<2xf32>, tuple<tensor<i32>>>) -> tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.if (mhlo::IfOp)

If operation

Produces the output from executing exactly one branch from true_branch or false_branch depending on the value of pred .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#if

Example: %result = "mhlo.if"(%pred) ({ "mhlo.return"(%result_true_branch) : (tensor ) -> () }, { "mhlo.return"(%result_false_branch) : (tensor ) -> () }) : (tensor ) -> tensor

Traits: RecursiveMemoryEffects , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferTypeOpInterface

Operands:

Résultats:

mhlo.imag (mhlo::ImagOp)

Imag operation

Syntaxe:

operation ::= `mhlo.imag` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Extracts the imaginary part, element-wise, from the operand and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#imag

Exemple:

%result = mhlo.imag %operand : (tensor<2xcomplex<f32>>) -> tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.infeed (mhlo::InfeedOp)

Infeed operation

Reads data from the infeed and produces results .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#infeed

Exemple:

%results:2 = "mhlo.infeed"(%token) {
  infeed_config = ""
} : (!mhlo.token) -> (tensor<3x3x3xi32>, !mhlo.token)

Les attributs:

Operands:

Résultats:

mhlo.iota (mhlo::IotaOp)

Iota operation

Fills an output tensor with values in increasing order starting from zero along the iota_dimension dimension.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#iota

Exemple:

%output = mhlo.iota dim = 0 : tensor<4x5xi32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Résultats:

mhlo.is_finite (mhlo::IsFiniteOp)

IsFinite operation

Syntaxe:

operation ::= `mhlo.is_finite` $x attr-dict `:` functional-type(operands, results)

Performs element-wise check whether the value in x is finite (ie is neither +Inf, -Inf, nor NaN) and produces a y tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#is_finite

Exemple:

%y = mhlo.is_finite %x : (tensor<7xf32>) -> tensor<7xi1>

Traits: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.log (mhlo::LogOp)

Log operation

Syntaxe:

operation ::= `mhlo.log` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise logarithm operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#log

Exemple:

%result = mhlo.log %operand : tensor<2x2xf64>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.log_plus_one (mhlo::Log1pOp)

Log1p operation

Syntaxe:

operation ::= `mhlo.log_plus_one` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise logarithm plus one operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#log_plus_one

Exemple:

%result = mhlo.log_plus_one %operand : tensor<6xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.logistic (mhlo::LogisticOp)

Logistic operation

Syntaxe:

operation ::= `mhlo.logistic` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise logistic operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#logistic

Exemple:

%result = mhlo.logistic %operand : tensor<2x2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.map (mhlo::MapOp)

Map operation

Applies a map function computation to inputs along the dimensions and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#map

Exemple:

%result = "mhlo.map"(%input0, %input1) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = mhlo.multiply %arg0, %arg1 : tensor<i32>
    mhlo.return %0 : tensor<i32>
}) {
  dimensions = dense<[0, 1]> : tensor<2xi64>
} : (tensor<2x2xi32>, tensor<2x2xi32>) -> tensor<2x2xi32>

Traits: InferTensorType , RecursiveMemoryEffects , SameOperandsAndResultShape , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.maximum (mhlo::MaxOp)

Max operation

Syntaxe:

operation ::= `mhlo.maximum` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise max operation on tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#maximum

Exemple:

%result = mhlo.maximum %lhs, %rhs : tensor<4xf32>

Traits: AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.minimum (mhlo::MinOp)

Min operation

Syntaxe:

operation ::= `mhlo.minimum` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise min operation on tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#minimum

Exemple:

%result = mhlo.minimum %lhs, %rhs : tensor<4xf32>

Traits: AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.multiply (mhlo::MulOp)

Mul operation

Syntaxe:

operation ::= `mhlo.multiply` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise product of two tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#multiply

Exemple:

%result = mhlo.multiply %lhs, %rhs : tensor<2xi32>

Traits: AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.negate (mhlo::NegOp)

Neg operation

Syntaxe:

operation ::= `mhlo.negate` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise negation of operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#negate

Exemple:

%result = mhlo.negate %operand : tensor<2x3xi32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.not (mhlo::NotOp)

Not operation

Syntaxe:

operation ::= `mhlo.not` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise NOT of tensor operand of type integer and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#not

Exemple:

%result = mhlo.not %operand : tensor<5x3x1xi1>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.optimization_barrier (mhlo::OptimizationBarrierOp)

OptimizationBarrier operation

Syntaxe:

operation ::= `mhlo.optimization_barrier` attr-dict ($operand^ `:` custom<PairwiseOpType>(type($operand), type($result))):(`(` `)`)?

Ensures that the operations that produce the operand are executed before any operations that depend on the result and prevents compiler transformations from moving operations across the barrier. Other than that, the operation is an identity, ie result = operand .

See https://github.com/openxla/stablehlo/blob/main/docs/spec.md#optimization_barrier

Exemple:

%result0, %result1 = mhlo.optimization_barrier %operand0, %operand1 : tensor<f32>, tensor<f32>

Traits: AlwaysSpeculatableImplTrait , HLO_PairwiseSameOperandAndResultType

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.or (mhlo::OrOp)

Or operation

Syntaxe:

operation ::= `mhlo.or` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise OR of two tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#or

Exemple:

%result = mhlo.or %lhs, %rhs : tensor<2xi1>

Traits: AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.outfeed (mhlo::OutfeedOp)

Outfeed operation

Writes inputs to the outfeed and produces a result token.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#outfeed

Exemple:

%result = "mhlo.outfeed"(%input0, %token) {
  outfeed_config = ""
} : (tensor<3x3x3xi32>, !mhlo.token) -> !mhlo.token

Interfaces: InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.pad (mhlo::PadOp)

Pad operation

Expands operand by padding around the tensor as well as between the elements of the tensor with the given padding_value .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#pad

Exemple:

%0 = mhlo.pad %arg0, %arg1, low = [0, 1], high = [2, 1], interior = [1, 2]
  : (tensor<2x3xi32>, tensor<i32>) -> tensor<5x9xi32>

Traits: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.partition_id (mhlo::PartitionIdOp)

PartitionId operation

Syntaxe:

operation ::= `mhlo.partition_id` attr-dict `:` type(results)

Produces partition_id of the current process.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#partition_id

Exemple:

%result = mhlo.partition_id : tensor<ui32>

Interfaces: InferTypeOpInterface

Résultats:

mhlo.popcnt (mhlo::PopulationCountOp)

PopulationCount operation

Syntaxe:

operation ::= `mhlo.popcnt` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise count of the number of bits set in the operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#popcnt

Exemple:

%result = mhlo.popcnt %operand : tensor<4xi8>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.power (mhlo::PowOp)

Pow operation

Syntaxe:

operation ::= `mhlo.power` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise exponentiation of lhs tensor by rhs tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#power

Exemple:

%result = mhlo.power %lhs, %rhs : tensor<6xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.real (mhlo::RealOp)

Real operation

Syntaxe:

operation ::= `mhlo.real` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Extracts the real part, element-wise, from the operand and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#real

Exemple:

%result = mhlo.real %operand : (tensor<2xcomplex<f32>>) -> tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.real_dynamic_slice (mhlo::RealDynamicSliceOp)

RealDynamicSlice operation

Syntaxe:

operation ::= `mhlo.real_dynamic_slice` operands attr-dict `:` functional-type(operands, results)

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as SliceOp except that start_indices , limit_indices and strides are specified dynamically: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#slice

Exemple:

%result = mhlo.real_dynamic_slice %operand,
            %start_indices, %limit_indices, %strides
       : (tensor<256x?xf32>, tensor<2xindex>, tensor<2xindex>, tensor<2xindex>) -> tensor<256x?xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.recv (mhlo::RecvOp)

Recv operation

Receives data from a channel with channel_id and produces results .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#recv

Exemple:

%results:2 = "mhlo.recv"(%token) {
  // channel_id = 5 : i64,
  // channel_type = #stablehlo<channel_type HOST_TO_DEVICE>,
  channel_handle = #mhlo.channel_handle<handle = 5, type = 3>,
  is_host_transfer = true
} : (!mhlo.token) -> (tensor<3x4xi32>, !mhlo.token)

Les attributs:

Operands:

Résultats:

mhlo.reduce (mhlo::ReduceOp)

Reduce operation

Applies a reduction function body to inputs and init_values along the dimensions and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reduce

Exemple:

%result = "mhlo.reduce"(%input, %init_value) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = "mhlo.add"(%arg0, %arg1) : (tensor<i32>, tensor<i32>) -> tensor<i32>
    "mhlo.return"(%0) : (tensor<i32>) -> ()
}) {
  dimensions = dense<1> : tensor<1xi64>
} : (tensor<1x6xi32>, tensor<i32>) -> tensor<1xi32>

Traits: InferTensorType , RecursiveMemoryEffects , SameVariadicOperandSize , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.reduce_precision (mhlo::ReducePrecisionOp)

ReducePrecision operation

Syntaxe:

operation ::= `mhlo.reduce_precision` $operand `,` `format` `=` custom<ExponentMantissa>($exponent_bits, $mantissa_bits)
              attr-dict `:` custom<SameOperandsAndResultType>(type($operand), type($output))

Performs element-wise conversion of operand to another floating-point type that uses exponent_bits and mantissa_bits and back to the original floating-point type and produces an output tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reduce_precision

Exemple:

%output = mhlo.reduce_precision %operand, format = e5m2 : tensor<6xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.reduce_scatter (mhlo::ReduceScatterOp)

ReduceScatter operation

Within each process group in the process grid, performs reduction, using computations , over the values of the operand tensor from each process, splits the reduction result along scatter_dimension into parts, and scatters the split parts between the processes to produce the result .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reduce_scatter

Exemple:

%result = "mhlo.reduce_scatter"(%operand) ({
  ^bb0(%arg0: tensor<f32>, %arg1: tensor<f32>):
  %0 = mhlo.add %arg0, %arg1 : tensor<f32>
  mhlo.return %0 : tensor<f32>
}) {
  scatter_dimension = 1 : i64,
  replica_groups = dense<[[0, 1]]> : tensor<1x2xi64>,
  // channel_id = 0
  channel_handle = #mhlo.channel_handle<handle = 0, type = 0>
  // use_global_device_ids = false
} : (tensor<2x4xf32>) -> tensor<2x2xf32>

Les attributs:

Operands:

Résultats:

mhlo.reduce_window (mhlo::ReduceWindowOp)

ReduceWindow operation

Applies a reduction function body to windows of inputs and init_values and produces results .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reduce_window

Exemple:

%result = "mhlo.reduce_window"(%input, %init_value) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = mhlo.add %arg0, %arg1 : tensor<i32>
    mhlo.return %0 : tensor<i32>
}) {
  window_dimensions = dense<[2, 1]> : tensor<2xi64>,
  window_strides = dense<[4, 1]> : tensor<2xi64>,
  base_dilations = dense<[2, 1]> : tensor<2xi64>,
  window_dilations = dense<[3, 1]> : tensor<2xi64>,
  padding = dense<[[2, 1], [0, 0]]> : tensor<2x2xi64>
} : (tensor<3x2xi32>, tensor<i32>) -> tensor<2x2xi32>

Traits: InferTensorType , RecursiveMemoryEffects , SameVariadicOperandSize , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.remainder (mhlo::RemOp)

Rem operation

Syntaxe:

operation ::= `mhlo.remainder` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise remainder of dividend lhs and divisor rhs tensors and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#remainder

Exemple:

%result = mhlo.remainder %lhs, %rhs : tensor<4xi64>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.replica_id (mhlo::ReplicaIdOp)

ReplicaId operation

Syntaxe:

operation ::= `mhlo.replica_id` attr-dict `:` type(results)

Produces replica_id of the current process.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#replica_id

Exemple:

%result = mhlo.replica_id : tensor<ui32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Résultats:

mhlo.reshape (mhlo::ReshapeOp)

Reshape operation

Syntaxe:

operation ::= `mhlo.reshape` operands attr-dict `:` functional-type(operands, results)

Performs reshape of operand tensor to a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reshape

Exemple:

%result = mhlo.reshape %operand : (tensor<2xf32>) -> tensor<1x2xf32>

Traits: AlwaysSpeculatableImplTrait , HLO_CompatibleOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.return (mhlo::ReturnOp)

_This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/425

Informally, this operation serves as a terminator for regions defined by
the StableHLO ops. Non-StableHLO ops, e.g. `func.func`, have their own
terminators, e.g. `func.return`.

Example:

    ```mlir
    %result = "mhlo.reduce"(%input, %init_value) ({
      ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
        %0 = "mhlo.add"(%arg0, %arg1) : (tensor<i32>, tensor<i32>) -> tensor<i32>
        "mhlo.return"(%0) : (tensor<i32>) -> ()
    }) {
      dimensions = dense<1> : tensor<1xi64>
    } : (tensor<1x6xi32>, tensor<i32>) -> tensor<1xi32>
    ```_


Syntax:

```

operation ::= mhlo.return $results attr-dict ( : type($results)^)?

Traits: `AlwaysSpeculatableImplTrait`, `Terminator`

Interfaces: `ConditionallySpeculatable`, `NoMemoryEffect (MemoryEffectOpInterface)`

Effects: `MemoryEffects::Effect{}`

#### Operands:

| Operand | Description |
| :-----: | ----------- |
| `results` | variadic of ranked tensor of f8E4M3B11FNUZ type or f8E4M3FN type or f8E4M3FNUZ type or f8E5M2 type or f8E5M2FNUZ type or 16-bit float or 32-bit float or 64-bit float or bfloat16 type or pred (AKA boolean or 1-bit integer) or 4/8/16/32/64-bit signless integer or 4/8/16/32/64-bit unsigned integer or complex type with 32-bit float or 64-bit float elements or 4/8/16/32-bit uniform quantized signed integer or 4/8/16/32-bit uniform quantized unsigned integer or 4/8/16/32-bit uniform quantized per axis signed integer or 4/8/16/32-bit uniform quantized per axis unsigned integer values or token or nested tuple with any combination of ranked tensor of f8E4M3B11FNUZ type or f8E4M3FN type or f8E4M3FNUZ type or f8E5M2 type or f8E5M2FNUZ type or 16-bit float or 32-bit float or 64-bit float or bfloat16 type or pred (AKA boolean or 1-bit integer) or 4/8/16/32/64-bit signless integer or 4/8/16/32/64-bit unsigned integer or complex type with 32-bit float or 64-bit float elements or 4/8/16/32-bit uniform quantized signed integer or 4/8/16/32-bit uniform quantized unsigned integer values or ranked tensor of 4/8/16/32-bit uniform quantized per axis signed integer or 4/8/16/32-bit uniform quantized per axis unsigned integer values or token values


### `mhlo.reverse` (mhlo::ReverseOp)

_Reverse operation_

Reverses the order of elements in the `operand` along the specified
`dimensions` and produces a `result` tensor.

See:
<a href="https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reverse">https://github.com/openxla/stablehlo/blob/main/docs/spec.md#reverse</a>

Example:
```mlir
%result = mhlo.reverse %operand, dims = [1] : tensor<3x2xi32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.rng (mhlo::RngOp)

Rng operation

Generates random numbers using the rng_distribution algorithm and produces a result tensor of a given shape shape .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#rng

Exemple:

%result = mhlo.rng %a, %b, %shape, distribution = NORMAL : (tensor<i32>, tensor<i32>, tensor<2xi64>) -> tensor<3x3xi32>

Traits: InferTensorType

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.rng_bit_generator (mhlo::RngBitGeneratorOp)

RngBitGenerator operation

Returns an output filled with uniform random data and an updated output state output_state given an initial state initial_state using the pseudorandom number generator algorithm rng_algorithm .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#rng_bit_generator

Exemple:

%output_state, %output = mhlo.rng_bit_generator %initial_state, algorithm = THREE_FRY : (tensor<2xui64>) -> (tensor<2xui64>, tensor<2x2xui64>)

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.round_nearest_afz (mhlo::RoundOp)

Round operation

Syntaxe:

operation ::= `mhlo.round_nearest_afz` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise rounding towards the nearest integer, breaking ties away from zero, on the operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#round_nearest_afz

Exemple:

%result = mhlo.round_nearest_afz %operand : tensor<5xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.round_nearest_even (mhlo::RoundNearestEvenOp)

RoundNearestEven operation

Syntaxe:

operation ::= `mhlo.round_nearest_even` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise rounding towards the nearest integer, breaking ties towards the even integer, on the operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#round_nearest_even

Exemple:

%result = mhlo.round_nearest_even %operand : tensor<5xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.rsqrt (mhlo::RsqrtOp)

Rsqrt operation

Syntaxe:

operation ::= `mhlo.rsqrt` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise reciprocal square root operation on operand tensor and produces a result tensor, implementing the rSqrt operation from the IEEE-754 specification.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#rsqrt

Exemple:

%result = mhlo.rsqrt %operand : tensor<2x2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.scatter (mhlo::ScatterOp)

Scatter operation

Produces results tensors which are equal to inputs tensors except that several slices specified by scatter_indices are updated with the values updates using update_computation .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#scatter

Exemple:

%result = "mhlo.scatter"(%input, %scatter_indices, %update) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = mhlo.add %arg0, %arg1 : tensor<i32>
    mhlo.return %0 : tensor<i32>
}) {
  scatter_dimension_numbers = #mhlo.scatter<
    update_window_dims = [2,3],
    inserted_window_dims = [0],
    scatter_dims_to_operand_dims = [1, 0],
    index_vector_dim = 2>,
  indices_are_sorted = false,
  unique_indices = false
} : (tensor<3x4x2xi32>, tensor<2x3x2xi64>, tensor<2x3x2x2xi32>) -> tensor<3x4x2xi32>

Traits: RecursiveMemoryEffects , SameVariadicOperandSize

Interfaces: InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.select (mhlo::SelectOp)

Sélectionnez l'opération

Syntaxe:

operation ::= `mhlo.select` operands attr-dict `:`
              custom<SelectOpType>(type($pred), type($on_true), type($on_false), type($result))

Produces a result tensor where each element is selected from on_true or on_false tensor based on the value of the corresponding element of pred .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#select

Exemple:

%result = mhlo.select %pred, %on_true, %on_false : tensor<2x2xi1>, tensor<2x2xi32>

Traits: AlwaysSpeculatableImplTrait , HLO_BroadcastingElementwise , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.select_and_scatter (mhlo::SelectAndScatterOp)

SelectAndScatter operation

Scatters the values from the source tensor using scatter based on the outcome of reduce_window of the input tensor using select and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#select_and_scatter

Exemple:

%result = "mhlo.select_and_scatter"(%operand, %source, %init_value) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = "mhlo.compare"(%arg0, %arg1) {
      comparison_direction = #stablehlo<comparison_direction GE>
    } : (tensor<i32>, tensor<i32>) -> tensor<i1>
    "mhlo.return"(%0) : (tensor<i1>) -> ()
}, {
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = "mhlo.add"(%arg0, %arg1) : (tensor<i32>, tensor<i32>) -> tensor<i32>
    "mhlo.return"(%0) : (tensor<i32>) -> ()
}) {
  window_dimensions = dense<[3, 1]> : tensor<2xi64>,
  window_strides = dense<[2, 1]> : tensor<2xi64>,
  padding = dense<[[0, 1], [0, 0]]> : tensor<2x2xi64>
} : (tensor<4x2xi32>, tensor<2x2xi32>, tensor<i32>) -> tensor<4x2xi32>

Traits: RecursiveMemoryEffects

Interfaces: InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.send (mhlo::SendOp)

Send operation

Sends inputs to a channel channel_id and produces a result token.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#send

Exemple:

%result = "mhlo.send"(%operand, %token) {
  // channel_id = 5 : i64,
  // channel_type = #stablehlo<channel_type DEVICE_TO_HOST>,
  channel_handle = #mhlo.channel_handle<handle = 5, type = 2>,
  is_host_transfer = true
} : (tensor<3x4xi32>, !mhlo.token) -> !mhlo.token

Interfaces: InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.set_dimension_size (mhlo::SetDimensionSizeOp)

SetDimensionSize operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/8

Informally, this operation does the same thing as XLA's SetDimensionSize: https://www.tensorflow.org/xla/operation_semantics#setdimensionsize

Exemple:

%0 = mhlo.set_dimension_size %arg0, %arg1, dim = 1 : (tensor<4x2xf32>, tensor<i32>) -> tensor<4x2xf32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.shift_left (mhlo::ShiftLeftOp)

ShiftLeft operation

Syntaxe:

operation ::= `mhlo.shift_left` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise left-shift operation on the lhs tensor by rhs number of bits and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#shift_left

Exemple:

%result = mhlo.shift_left %lhs, %rhs : tensor<6xi8>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.shift_right_arithmetic (mhlo::ShiftRightArithmeticOp)

ShiftRightArithmetic operation

Syntaxe:

operation ::= `mhlo.shift_right_arithmetic` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise arithmetic right-shift operation on the lhs tensor by rhs number of bits and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#shift_right_arithmetic

Exemple:

%result = mhlo.shift_right_arithmetic %lhs, %rhs : tensor<6xi8>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.shift_right_logical (mhlo::ShiftRightLogicalOp)

ShiftRightLogical operation

Syntaxe:

operation ::= `mhlo.shift_right_logical` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise logical right-shift operation on the lhs tensor by rhs number of bits and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#shift_right_logical

Exemple:

%result = mhlo.shift_right_logical %lhs, %rhs : tensor<6xi8>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.sign (mhlo::SignOp)

Sign operation

Syntaxe:

operation ::= `mhlo.sign` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Returns the sign of the operand element-wise and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#sign

Exemple:

%result = mhlo.sign %operand : tensor<7xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.sine (mhlo::SineOp)

Sine operation

Syntaxe:

operation ::= `mhlo.sine` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise sine operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#sine

Exemple:

%result = mhlo.sine %operand : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.slice (mhlo::SliceOp)

Slice operation

Extracts a slice from the operand using statically-computed starting indices and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#slice

Exemple:

%result = "mhlo.slice" (%operand) {
  start_indices = dense<[1, 2]> : tensor<2xi64>,
  limit_indices = dense<[3, 4]> : tensor<2xi64>,
  strides = dense<1> : tensor<2xi64>
} : (tensor<3x4xi64>) -> tensor<2x2xi64>

Traits: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.sort (mhlo::SortOp)

Sort operation

Sorts a variadic number of tensors in inputs together, according to a custom comparator , along the given dimension and produces a variadic number of tensors as results .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#sort

Exemple:

%result0, %result1 = "mhlo.sort"(%input0, %input1) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>, %arg2: tensor<i32>, %arg3: tensor<i32>):
    %predicate = "mhlo.compare"(%arg0, %arg1) {
      comparison_direction = #stablehlo<comparison_direction GT>
      } : (tensor<i32>, tensor<i32>) -> tensor<i1>
    "mhlo.return"(%predicate) : (tensor<i1>) -> ()
}) {
  dimension = 0 : i64,
  is_stable = true
} : (tensor<2x3xi32>, tensor<2x3xi32>) -> (tensor<2x3xi32>, tensor<2x3xi32>)

Traits: InferTensorType , RecursiveMemoryEffects , SameOperandsAndResultShape

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.sparse_dot (mhlo::SparseDotOp)

Sparse dot operation

Similar to dot_general operation, with one or both of the operands being sparse. An additional argument provides sparsity meta information. Disclaimer: this op is experimental / a work in progress.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.sqrt (mhlo::SqrtOp)

Sqrt operation

Syntaxe:

operation ::= `mhlo.sqrt` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise square root operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#sqrt

Exemple:

%result = mhlo.sqrt %operand : tensor<2x2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.stochastic_convert (mhlo::StochasticConvertOp)

StochasticConvert operation

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/295

Informally, this operation performs element-wise conversion of values from a bigger type to a smaller one with stochastic rounding using the random number passed in.

Traits: AlwaysSpeculatableImplTrait , Elementwise

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.subtract (mhlo::SubtractOp)

Subtract operation

Syntaxe:

operation ::= `mhlo.subtract` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise subtraction of two tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#subtract

Exemple:

%result = mhlo.subtract %lhs, %rhs : tensor<2xi32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.tan (mhlo::TanOp)

Tan operation

Syntaxe:

operation ::= `mhlo.tan` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

This operation is a work in progress, so it is not yet included in the specification: https://github.com/openxla/stablehlo/issues/954

Informally, this operation returns Tan(operand) element-wise.

Exemple:

%0 = mhlo.tan %arg0 : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.tanh (mhlo::TanhOp)

Tanh operation

Syntaxe:

operation ::= `mhlo.tanh` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise hyperbolic tangent operation on operand tensor and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#tanh

Exemple:

%result = mhlo.tanh %operand : tensor<2xf32>

Traits: AlwaysSpeculatableImplTrait , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.topk (mhlo::TopKOp)

TopK operation

Syntaxe:

operation ::= `mhlo.topk` `(`$operand `,` `k` `=` $k (`,` `largest` `=` $largest^)? `)` attr-dict `:`
              type($operand) `->` `(`type($values)`,` type($indices)`)`

Returns top k values and their indices, along the last dimension of the operand if largest=true or the bottom k values if largest=false .

See: https://www.tensorflow.org/xla/operation_semantics#top-k

Exemple:

%values, %indices = mhlo.topk(%operand, k=5, largest=true)
  : tensor<100xf32> -> (tensor<5xf32>, tensor<5xi32>)

Traits: InferTensorType , RecursiveMemoryEffects

Interfaces: InferShapedTypeOpInterface , InferTypeOpInterface

Les attributs:

Operands:

Résultats:

mhlo.torch_index_select (mhlo::TorchIndexSelectOp)

TorchIndexSelect operation

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as PyTorch's index_select, augmented with support for batch dimensions: https://pytorch.org/docs/stable/generated/torch.index_select.html

The batch_dims attribute specifies the number of major batch dimensions (0 or more) that act like a multidimensional loop over both the operand and the index.

Exemple:

%result = "mhlo.torch_index_select"(%operand, %index) {
  dim = 2 : i64,
  batch_dims = 1 : i64
} : (tensor<8x128x3072x64xf32>, tensor<8x16x1024xi32>) -> tensor<8x128x16x1024x64xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.trace (mhlo::TraceOp)

Trace operation

Syntaxe:

operation ::= `mhlo.trace` $operand `,` $tag attr-dict `:` type($operand)

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/604

It is not used by JAX, PyTorch or TensorFlow, so it looks like we should've classified it as "Private to XLA" and not included it in StableHLO in the first place. With that in mind, its semantics will not be documented here.

Exemple:

mhlo.trace %arg0, "In test code." : tensor<5x1x5xi32>

Les attributs:

Operands:

mhlo.transpose (mhlo::TransposeOp)

Transpose operation

Permutes the dimensions of operand tensor using permutation and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#transpose

Exemple:

%0 = mhlo.transpose %arg0, dims = [2, 1, 0] : (tensor<1x2x3xi32>) -> tensor<3x2x1xi32>

Traits: AlwaysSpeculatableImplTrait , HLO_CompatibleOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.triangular_solve (mhlo::TriangularSolveOp)

TriangularSolve operation

Solves batches of systems of linear equations with lower or upper triangular coefficient matrices.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#triangular_solve

Exemple:

%result = "mhlo.triangular_solve"(%a, %b) {
  left_side = true,
  lower = true,
  unit_diagonal = false,
  transpose_a = #stablehlo<transpose NO_TRANSPOSE>
} : (tensor<3x3xf32>, tensor<3x3xf32>) -> tensor<3x3xf32>

Traits: AlwaysSpeculatableImplTrait , InferTensorType , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.tuple (mhlo::TupleOp)

Tuple operation

Syntaxe:

operation ::= `mhlo.tuple` $val attr-dict `:` custom<TupleOpType>(type($val), type($result))

Produces a result tuple from values val .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#tuple

Exemple:

%result = mhlo.tuple %val0, %val1 : tuple<tensor<2xf32>, tuple<tensor<i32>>>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.unary_einsum (mhlo::UnaryEinsumOp)

UnaryEinsum operation

This operation is on its way out of StableHLO, so it is not included in the specification: https://github.com/openxla/stablehlo/issues/3

Informally, this operation does the same thing as TF's einsum: https://www.tensorflow.org/api_docs/python/tf/einsum

Exemple:

%result = "mhlo.unary_einsum"(%operand) {
  einsum_config = "ab->a"
} : (tensor<4x16xf32>) -> tensor<4xf32>

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Les attributs:

Operands:

Résultats:

mhlo.uniform_dequantize (mhlo::UniformDequantizeOp)

UniformDequantize operation

Syntaxe:

operation ::= `mhlo.uniform_dequantize` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise conversion of quantized tensor operand to a floating-point tensor result according to the quantization parameters defined by the operand type.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#uniform_dequantize

Exemple:

%result = mhlo.uniform_dequantize %operand : (tensor<16x16x!quant.uniform<i8:f32, 34.0:16>>) -> tensor<16x16xf32>

Traits: AlwaysSpeculatableImplTrait , Elementwise , InferTensorType , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.uniform_quantize (mhlo::UniformQuantizeOp)

UniformQuantize operation

Syntaxe:

operation ::= `mhlo.uniform_quantize` $operand attr-dict
              `:` custom<SameOperandsAndResultType>(type($operand), type($result))

Performs element-wise conversion of floating-point tensor or quantized tensor operand to a quantized tensor result according to the quantization parameters defined by the result type.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#uniform_quantize

Exemple:

%result = mhlo.uniform_quantize %operand : (tensor<16x16xf32>) -> tensor<16x16x!quant.uniform<ui8:f32, 34.0:16>>

Traits: AlwaysSpeculatableImplTrait , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

mhlo.while (mhlo::WhileOp)

While operation

Produces the output from executing body function 0 or more times while the cond function outputs true .

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#while

Exemple:

%results0, %results1 = "mhlo.while"(%operand0, %operand1) ({
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = "mhlo.compare"(%arg0, %arg1) {
      comparison_direction = #stablehlo<comparison_direction LT>
    } : (tensor<i32>, tensor<i32>) -> tensor<i1>
    "mhlo.return"(%0) : (tensor<i1>) -> ()
}, {
  ^bb0(%arg0: tensor<i32>, %arg1: tensor<i32>):
    %0 = "mhlo.add"(%arg0, %constant0) : (tensor<i32>, tensor<i32>) -> tensor<i32>
    "mhlo.return"(%0, %arg1) : (tensor<i32>, tensor<i32>) -> ()
}) : (tensor<i32>, tensor<i32>) -> (tensor<i32>, tensor<i32>)

Traits: RecursiveMemoryEffects , SingleBlockImplicitTerminator<ReturnOp> , SingleBlock

Interfaces: InferTypeOpInterface , OpAsmOpInterface

Operands:

Résultats:

mhlo.xla.rng_get_and_update_state (mhlo::XlaRngGetAndUpdateStateOp)

XlaRngGetAndUpdateState operation

Syntaxe:

operation ::= `mhlo.xla.rng_get_and_update_state` attr-dict

This operation is private to the XLA compiler, so it is does not yet have a specification.

Informally, this operation represents the change of the global random number generator state for rng instructions. The global state is incremented by delta and the old state is returned.

The output is currently defined for a single output type. If this changes in the future to support multiple types, lowering to use of a global memref must ensure that a single memref is still used and updated appropriately.

Interfaces: InferTypeOpInterface

Les attributs:

Résultats:

mhlo.xor (mhlo::XorOp)

Xor operation

Syntaxe:

operation ::= `mhlo.xor` $lhs `,` $rhs attr-dict
              `:` custom<SameOperandsAndResultType>(type($lhs), type($rhs), type($result))

Performs element-wise XOR of two tensors lhs and rhs and produces a result tensor.

See: https://github.com/openxla/stablehlo/blob/main/docs/spec.md#xor

Exemple:

%result = mhlo.xor %lhs, %rhs : tensor<2xi32>

Traits: AlwaysSpeculatableImplTrait , Commutative , CompatibleOperandsAndResultType , Elementwise , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Operands:

Résultats:

Les attributs

ArgResultAliasAttr

Attribute that models the alias relationship of entry function argument

This attribute captures the alias relationship of an MHLO main function argument to one of the results, denoted by resultIndex . The argTupleIndices and resultTupleIndices are used to index into nested tuples in operand and result respectively. If isMustAlias is true then the operand-result pair must alias.

This is meant to be used as an attribute on a function argument in MHLO. For example, in the following code it expresses that %arg1 may alias 0-th result.

func @main(%arg0: tensor<2xf32>, %arg1: tensor<3xf32> {mhlo.result_alias =
    mhlo.result_alias<result_index = [2], ...>}
  ) -> tensor<2xf32>, tensor<3xf32> {
  // function body ...
}

Paramètres:

ChannelHandleAttr

two 64-bit integers 'handle' and 'type'

Syntaxe:

#mhlo.channel_handle<
  int64_t,   # handle
  int64_t   # type
>

Paramètres:

ComparisonDirectionAttr

Which comparison operation to perform.

Syntaxe:

#mhlo.comparison_direction<
  ::mlir::mhlo::ComparisonDirection   # value
>

Enum cases:

• EQ ( EQ )
• NE ( NE )
• GE ( GE )
• GT ( GT )
• LE ( LE )
• LT ( LT ) #### Parameters:

ComparisonTypeAttr

Which comparison type to use.

Syntaxe:

#mhlo.comparison_type<
  ::mlir::mhlo::ComparisonType   # value
>

Enum cases:

• NOTYPE ( NOTYPE )
• FLOAT ( FLOAT )
• TOTALORDER ( TOTALORDER )
• SIGNED ( SIGNED )
• UNSIGNED ( UNSIGNED ) #### Parameters:

ConvDimensionNumbersAttr

Structure of dimension information for conv op

Paramètres:

CrossProgramPrefetchAttr

Argument that is prefetched from another program

Syntaxe:

#mhlo.cross_program_prefetch<
  int64_t,   # parameter
  ::llvm::ArrayRef<int64_t>,   # indices
  std::optional<int64_t>   # offset
>

This attribute captures an argument that is prefetched from another program. For a given CrossProgramPrefetchAttr , parameter tells us which argument of the main function of the module is prefetched, and indices is a shape index telling us what subshape of that argument is prefetched.

A shape has a subshape iff it is a tuple. In that case, the subshape of the tuple by indices is the shape achieved after indexing by each element of indices in turn. For example, the [1,0] subshape of tuple<tuple<token, token>, tuple<tensor<i32>, token>> is tensor<i32> .

An empty value for indices means the whole shape is prefetched.

Par exemple,

module attributes { mhlo.cross_program_prefetch = [ #mhlo.cross_program_prefetch< parameter = 0, indices = [0]> ]} {
  func.func @copy(%arg0 : tuple<tensor<2x3xi32>, tensor<i32>>) -> tuple<tensor<2x3xi32>, tensor<i32>> {
    %0 = "mhlo.copy"(%arg0) {is_cross_program_prefetch}
    return %0 : tuple<tensor<2x3xi32>, tensor<i32>>
  }
  func.func @main(%arg0 : tuple<tensor<2x3xi32>, tensor<i32>>) -> tuple<tensor<2x3xi32>, tensor<i32>> {
    %1 = "mhlo.async_start"(%arg0) {called_computation=@copy}
    %2 = "mhlo.async_done"(%1) {called_computation=@copy}
    return %2 : tuple<tensor<2x3xi32>, tensor<i32>>
  }
}

The parameter = 0 tells us that the async copy of the 0 th parameter is a cross_program_prefetch , while the index of [0] tells us that the 0 th element of the tuple is prefetched while the other element of the tuple is not.

Paramètres:

CustomCallScheduleAttr

Specifies the desired schedule for the custom-call.

Syntaxe:

#mhlo.custom_call_schedule<
  ::mlir::mhlo::CustomCallSchedule   # value
>

Enum cases:

• NONE ( NONE )
• LATEST ( LATEST )
• EARLIEST ( EARLIEST ) #### Parameters:

DequantizeModeAttr

Dequantization mode. Only MIN_COMBINED is supported.

Syntaxe:

#mhlo.dequantize_mode<
  ::mlir::mhlo::DequantizeMode   # value
>

Enum cases:

• MIN_COMBINED ( MIN_COMBINED ) #### Parameters:

DomainKindAttr

Kind of domain metatdata attached to an HLO domain.

Syntaxe:

#mhlo.kind<
  ::mlir::mhlo::DomainKind   # value
>

Enum cases:

• sharding ( sharding ) #### Parameters:

DotDimensionNumbersAttr

Attribute that models the dimension information for dot.

Paramètres:

FftTypeAttr

XLA fast fourier transform type.

Syntaxe:

#mhlo.fft_type<
  ::mlir::mhlo::FftType   # value
>

Enum cases:

• FFT ( FFT )
• IFFT ( IFFT )
• RFFT ( RFFT )
• IRFFT ( IRFFT ) #### Parameters:

FusionKindAttr

fusion kind

Syntaxe:

#mhlo.fusion_kind<
  ::mlir::mhlo::FusionKind   # value
>

Enum cases:

• kLoop ( kLoop )
• kInput ( kInput )
• kOutput ( kOutput )
• kCustom ( kCustom ) #### Parameters:

GatherDimensionNumbersAttr

Attribute that models the dimension information for gather

Paramètres:

OutputOperandAliasAttr

Attribute that models the alias relationship of output and operand of a CustomCall op

Syntaxe:

#mhlo.output_operand_alias<
  ::llvm::ArrayRef<int64_t>,   # outputTupleIndices
  int64_t,   # operandIndex
  ::llvm::ArrayRef<int64_t>   # operandTupleIndices
>

This attribute captures the alias relationship of the output to one of the operands for a CustomCall op, denoted by operand_index . The output_tuple_indices and operand_tuple_indices are used to index into output and operand types. These indices lists are empty if the corresponding types are not tuple types, and can be arbitrarily long in case of arbitrarily nested tuple types.

See https://www.tensorflow.org/xla/aliasing

Example when used as array with in mhlo.custom-call:

%0 = "mhlo.custom_call"(%arg0, %arg1) {
  // other attributes
  output_operand_alias = [
    #mhlo.output_operand_alias<output_tuple_indices = [0],
                               operand_index = 0,
                               operand_tuple_indices = [1]>
  ]
} : (tuple<tensor<1x1xf32>, tensor<2x3xf32>>, tensor<5x5xf32>) -> tuple<tensor<2x3xf32>>

The output and the 0th operand are both tuples. The aliasing shows the
relationship between the 0th element in output tuple with the 1st element in
the 0th operand. And both of them are of the same type: tensor<2x3xf32>.

Paramètres:

PrecisionAttr

XLA precision for an operand. Has backend specific meaning.

Syntaxe:

#mhlo.precision<
  ::mlir::mhlo::Precision   # value
>

Enum cases:

• DEFAULT ( DEFAULT )
• HIGH ( HIGH )
• HIGHEST ( HIGHEST )
• PACKED_NIBBLE ( PACKED_NIBBLE ) #### Parameters:

RngAlgorithmAttr

XLA PRNG algorithm to be used.

Syntaxe:

#mhlo.rng_algorithm<
  ::mlir::mhlo::RngAlgorithm   # value
>

Enum cases:

• DEFAULT ( DEFAULT )
• THREE_FRY ( THREE_FRY )
• PHILOX ( PHILOX ) #### Parameters:

RngDistributionAttr

XLA PRNG distribution to be used.

Syntaxe:

#mhlo.rng_distribution<
  ::mlir::mhlo::RngDistribution   # value
>

Enum cases:

• UNIFORM ( UNIFORM )
• NORMAL ( NORMAL ) #### Parameters:

ScatterDimensionNumbersAttr

Attribute that models the dimension information for scatter

Paramètres:

SparsityDescriptorAttr

Describes structured (N:M) sparsity configuration

Syntaxe:

#mhlo.sparsity<
  int64_t,   # dimension
  int64_t,   # n
  int64_t   # m
>

This attribute is defined for a sparse dot operation with a structured sparse input tensor. With (N=2,M=4), every 4 consecutive logical elements have exactly 2 non-zero physical elements in the input tensor.

$dimension defines the index of the contracting dimension that is sparse (it has to be the most minor dimension). The additional metadata operand in the sparse dot operation defines which logical elements are zeroed out.

Paramètres:

TransposeAttr

Transpose options

Syntaxe:

#mhlo.transpose<
  ::mlir::mhlo::Transpose   # value
>

Enum cases:

• TRANSPOSE_INVALID ( TRANSPOSE_INVALID )
• NO_TRANSPOSE ( NO_TRANSPOSE )
• TRANSPOSE ( TRANSPOSE )
• ADJOINT ( ADJOINT ) #### Parameters:

TypeExtensionsAttr

Attribute that extends tensor type with MHLO type properties.

Syntaxe:

#mhlo.type_extensions<
  ::llvm::ArrayRef<int64_t>   # bounds
>

This attribute is used to extend MLIR tensor type with MHLO tensor specific properties. These properties aren't modeled in the MLIR type. This attribute is set in the encoding field of the tensor type.

See HLO_BoundedAttrInterface for documentation for bounds .

Paramètres:

Les types

AsyncBundleType

Opaque collection of other types

Syntaxe:

!mhlo.async_bundle<
  ::llvm::ArrayRef<Type>   # types
>

Paramètres: