DLPack 1.1 update

include/matx/core/dlpack.h

@@ -15,11 +15,11 @@
 #define DLPACK_EXTERN_C
 #endif
 
-/*! \brief The current version of dlpack */
-#define DLPACK_VERSION 80
+/*! \brief The current major version of dlpack */
+#define DLPACK_MAJOR_VERSION 1
 
-/*! \brief The current ABI version of dlpack */
-#define DLPACK_ABI_VERSION 1
+/*! \brief The current minor version of dlpack */
+#define DLPACK_MINOR_VERSION 1
 
 /*! \brief DLPACK_DLL prefix for windows */
 #ifdef _WIN32
@@ -38,6 +38,33 @@
 #ifdef __cplusplus
 extern "C" {
 #endif
+
+/*!
+ * \brief The DLPack version.
+ *
+ * A change in major version indicates that we have changed the
+ * data layout of the ABI - DLManagedTensorVersioned.
+ *
+ * A change in minor version indicates that we have added new
+ * code, such as a new device type, but the ABI is kept the same.
+ *
+ * If an obtained DLPack tensor has a major version that disagrees
+ * with the version number specified in this header file
+ * (i.e. major != DLPACK_MAJOR_VERSION), the consumer must call the deleter
+ * (and it is safe to do so). It is not safe to access any other fields
+ * as the memory layout will have changed.
+ *
+ * In the case of a minor version mismatch, the tensor can be safely used as
+ * long as the consumer knows how to interpret all fields. Minor version
+ * updates indicate the addition of enumeration values.
+ */
+typedef struct {
+  /*! \brief DLPack major version. */
+  uint32_t major;
+  /*! \brief DLPack minor version. */
+  uint32_t minor;
+} DLPackVersion;
+
 /*!
  * \brief The device type in DLDevice.
  */
@@ -89,6 +116,8 @@ typedef enum {
   kDLWebGPU = 15,
   /*! \brief Qualcomm Hexagon DSP */
   kDLHexagon = 16,
+  /*! \brief Microsoft MAIA devices */
+  kDLMAIA = 17,
 } DLDeviceType;
 
 /*!
@@ -128,6 +157,26 @@ typedef enum {
   kDLComplex = 5U,
   /*! \brief boolean */
   kDLBool = 6U,
+  /*! \brief FP8 data types */
+  kDLFloat8_e3m4 = 7U,
+  kDLFloat8_e4m3 = 8U,
+  kDLFloat8_e4m3b11fnuz = 9U,
+  kDLFloat8_e4m3fn = 10U,
+  kDLFloat8_e4m3fnuz = 11U,
+  kDLFloat8_e5m2 = 12U,
+  kDLFloat8_e5m2fnuz = 13U,
+  kDLFloat8_e8m0fnu = 14U,
+  /*! \brief FP6 data types
+   * Setting bits != 6 is currently unspecified, and the producer must ensure it is set
+   * while the consumer must stop importing if the value is unexpected.
+   */
+  kDLFloat6_e2m3fn = 15U,
+  kDLFloat6_e3m2fn = 16U,
+  /*! \brief FP4 data types
+   * Setting bits != 4 is currently unspecified, and the producer must ensure it is set
+   * while the consumer must stop importing if the value is unexpected.
+   */
+  kDLFloat4_e2m1fn = 17U,
 } DLDataTypeCode;
 
 /*!
@@ -141,6 +190,12 @@ typedef enum {
  *   - int8: type_code = 0, bits = 8, lanes = 1
  *   - std::complex<float>: type_code = 5, bits = 64, lanes = 1
  *   - bool: type_code = 6, bits = 8, lanes = 1 (as per common array library convention, the underlying storage size of bool is 8 bits)
+ *   - float8_e4m3: type_code = 8, bits = 8, lanes = 1 (packed in memory)
+ *   - float6_e3m2fn: type_code = 16, bits = 6, lanes = 1 (packed in memory)
+ *   - float4_e2m1fn: type_code = 17, bits = 4, lanes = 1 (packed in memory)
+ *
+ *  When a sub-byte type is packed, DLPack requires the data to be in little bit-endian, i.e.,
+ *  for a packed data set D ((D >> (i * bits)) && bit_mask) stores the i-th element.
  */
 typedef struct {
   /*!
@@ -186,6 +241,9 @@ typedef struct {
    *   return size;
    * }
    * \endcode
+   *
+   * Note that if the tensor is of size zero, then the data pointer should be
+   * set to `NULL`.
    */
   void* data;
   /*! \brief The device of the tensor */
@@ -211,6 +269,13 @@ typedef struct {
  *  not meant to transfer the tensor. When the borrowing framework doesn't need
  *  the tensor, it should call the deleter to notify the host that the resource
  *  is no longer needed.
+ *
+ * \note This data structure is used as Legacy DLManagedTensor
+ *       in DLPack exchange and is deprecated after DLPack v0.8
+ *       Use DLManagedTensorVersioned instead.
+ *       This data structure may get renamed or deleted in future versions.
+ *
+ * \sa DLManagedTensorVersioned
  */
 typedef struct DLManagedTensor {
   /*! \brief DLTensor which is being memory managed */
@@ -219,14 +284,83 @@ typedef struct DLManagedTensor {
    *   which DLManagedTensor is used in the framework. It can also be NULL.
    */
   void * manager_ctx;
-  /*! \brief Destructor signature void (*)(void*) - this should be called
-   *   to destruct manager_ctx which holds the DLManagedTensor. It can be NULL
-   *   if there is no way for the caller to provide a reasonable destructor.
-   *   The destructors deletes the argument self as well.
+  /*!
+   * \brief Destructor - this should be called
+   * to destruct the manager_ctx  which backs the DLManagedTensor. It can be
+   * NULL if there is no way for the caller to provide a reasonable destructor.
+   * The destructor deletes the argument self as well.
    */
   void (*deleter)(struct DLManagedTensor * self);
 } DLManagedTensor;
+
+// bit masks used in in the DLManagedTensorVersioned
+
+/*! \brief bit mask to indicate that the tensor is read only. */
+#define DLPACK_FLAG_BITMASK_READ_ONLY (1UL << 0UL)
+
+/*!
+ * \brief bit mask to indicate that the tensor is a copy made by the producer.
+ *
+ * If set, the tensor is considered solely owned throughout its lifetime by the
+ * consumer, until the producer-provided deleter is invoked.
+ */
+#define DLPACK_FLAG_BITMASK_IS_COPIED (1UL << 1UL)
+
+/*
+ * \brief bit mask to indicate that whether a sub-byte type is packed or padded.
+ *
+ * The default for sub-byte types (ex: fp4/fp6) is assumed packed. This flag can
+ * be set by the producer to signal that a tensor of sub-byte type is padded.
+ */
+#define DLPACK_FLAG_BITMASK_IS_SUBBYTE_TYPE_PADDED (1UL << 2UL)
+
+/*!
+ * \brief A versioned and managed C Tensor object, manage memory of DLTensor.
+ *
+ * This data structure is intended to facilitate the borrowing of DLTensor by
+ * another framework. It is not meant to transfer the tensor. When the borrowing
+ * framework doesn't need the tensor, it should call the deleter to notify the
+ * host that the resource is no longer needed.
+ *
+ * \note This is the current standard DLPack exchange data structure.
+ */
+struct DLManagedTensorVersioned {
+  /*!
+   * \brief The API and ABI version of the current managed Tensor
+   */
+  DLPackVersion version;
+  /*!
+   * \brief the context of the original host framework.
+   *
+   * Stores DLManagedTensorVersioned is used in the
+   * framework. It can also be NULL.
+   */
+  void *manager_ctx;
+  /*!
+   * \brief Destructor.
+   *
+   * This should be called to destruct manager_ctx which holds the DLManagedTensorVersioned.
+   * It can be NULL if there is no way for the caller to provide a reasonable
+   * destructor. The destructor deletes the argument self as well.
+   */
+  void (*deleter)(struct DLManagedTensorVersioned *self);
+  /*!
+   * \brief Additional bitmask flags information about the tensor.
+   *
+   * By default the flags should be set to 0.
+   *
+   * \note Future ABI changes should keep everything until this field
+   *       stable, to ensure that deleter can be correctly called.
+   *
+   * \sa DLPACK_FLAG_BITMASK_READ_ONLY
+   * \sa DLPACK_FLAG_BITMASK_IS_COPIED
+   */
+  uint64_t flags;
+  /*! \brief DLTensor which is being memory managed */
+  DLTensor dl_tensor;
+};
+
 #ifdef __cplusplus
 }  // DLPACK_EXTERN_C
 #endif
-#endif  // DLPACK_DLPACK_H_
+#endif  // DLPACK_DLPACK_H_