Flexible datapath offload chaining

A computer-implemented method for processing a chain of operations through an input output (IO) subsystem without central processing unit (CPU) involvement, the method comprising:

• providing, via the CPU, the chain of operations to the IO subsystem, wherein the IO subsystem is coupled to the CPU over Peripheral Component Interconnect Express (PCIe);
• processing, with the IO subsystem, the chain of operations by:
• retrieving, from a memory coupled to the CPU, data associated with the chain of operations, wherein the data comprises a size descriptor for the data;
• and
• executing each of the operations in the chain to determine an output for each of the operations based on the data and any output priorly determined for each of the operations, wherein the output determined for each of the executed operations is provided, as a current output, for execution of the respective next operation in the chain by:
• performing a transformation of the data according to the current output;
• updating the size descriptor for the data based on the current output when the current output changes the size of the data;
• persisting the current output to a local memory;
• and
• retrieving the current output stored to the local memory when executing the respective next operation in the chain;
• and
• providing, via the IO subsystem, an output for the chain of operations to the CPU.

One or more non-transitory computer-readable storage media coupled to one or more processors and an input output (IO) subsystem, the IO subsystem coupled to the one or more processors over Peripheral Component Interconnect Express (PCIe);

• the one or more non-transitory computer-readable storage media having instructions stored thereon which, when executed by the one or more processors and the IO subsystem, cause the one or more processors and the IO subsystem to perform a method comprising:
• providing, via the one or more processors, a chain of operations to the IO subsystem;
• processing, with the IO subsystem, the chain of operations by:
• retrieving, from a memory coupled to the CPU, data associated with the chain of operations, wherein the data comprises a size descriptor for the data;
• and
• executing each of the operations in the chain to determine an output for each of the operations based on the data and any output priorly determined for each of the operations, wherein the output determined for each of the executed operations is provided, as a current output, for execution of the respective next operation in the chain by:
• performing a transformation of the data according to the current output;
• updating the size descriptor for the data based on the current output when the current output changes the size of the data;
• persisting the current output to a local memory;
• and
• retrieving the current output stored to the local memory when executing the respective next operation in the chain;
• and
• providing, via the TO subsystem, an output for the chain of operations to the one or more processors.