Metadata predictor

1. A computer-implemented method comprising:

operating an index pipeline to generate indices in an index buffer in which the indices are used for reading out a memory device, the indices to the index buffer being generated using an accelerator, wherein writing to the accelerator is based, at least in part, on whether a target address for a branch taken is intraline;

populating a prediction cache with metadata of instructions, the metadata of instructions being read from the memory device; and

operating a prediction pipeline to generate a prediction using the metadata of the instructions from the prediction cache, the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline.

2. The computer-implemented method of claim 1 , wherein the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline comprises continually indexing the memory device to read out the metadata for the prediction cache independent of processes in the prediction pipeline.

3. The computer-implemented method of claim 1 , wherein the index buffer is configured to read out the memory device having the metadata for the prediction cache without reliance on and in advance of generating the prediction using the metadata from the prediction cache.

4. The computer-implemented method of claim 1 , wherein the prediction pipeline is configured to output a line of the metadata of the instructions from the prediction cache.

5. The computer-implemented method of claim 1 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions from the prediction cache without having to re-access the memory device.

6. The computer-implemented method of claim 1 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions that has been output from the prediction cache in order to generate a new prediction, responsive to a previous prediction being predicted to the line of the metadata.

7. The computer-implemented method of claim 1 , wherein the prediction is provided to an out-of-order instruction execution pipeline, the out-of-order instruction execution pipeline operating in parallel to the index pipeline and the prediction pipeline.

8. A system comprising:

a memory having computer readable instructions; and

one or more processors for executing the computer readable instructions, the computer readable instructions controlling the one or more processors to perform processes comprising:

operating an index pipeline to generate indices in an index buffer in which the indices are used for reading out a memory device, the indices to the index buffer being generated using an accelerator, wherein writing to the accelerator is based, at least in part, on whether a target address for a branch taken is intraline;

populating a prediction cache with metadata of instructions from the memory device; and

operating a prediction pipeline to generate a prediction using the metadata of the instructions from the prediction cache, the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline.

9. The system of claim 8 , wherein the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline comprises continually indexing the memory device to read out the metadata for the prediction cache independent of processes in the prediction pipeline.

10. The system of claim 8 , wherein the index buffer is configured to read out the memory device having the metadata for the prediction cache without reliance on and in advance of generating the prediction using the metadata from the prediction cache.

11. The system of claim 8 , wherein the prediction pipeline is configured to output a line of the metadata of the instructions from the prediction cache.

12. The system of claim 8 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions from the prediction cache without having to re-access the memory device.

13. The system of claim 8 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions that has been output from the prediction cache in order to generate a new prediction, responsive to a previous prediction being predicted to the line of the metadata.

14. The system of claim 8 , wherein the prediction is provided to an out-of-order instruction execution pipeline, the out-of-order instruction execution pipeline operating in parallel to the index pipeline and the prediction pipeline.

15. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors to cause the one or more processors to perform processes comprising:

operating an index pipeline to generate indices in an index buffer in which the indices are used for reading out a memory device, the indices to the index buffer being generated using an accelerator, wherein writing to the accelerator is based, at least in part, on whether a target address for a branch taken is intraline;

populating a prediction cache with metadata of instructions from the memory device; and

operating a prediction pipeline to generate a prediction using the metadata of the instructions from the prediction cache, the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline.

16. The computer program product of claim 15 , wherein the populating of the prediction cache with the metadata of the instructions being performed asynchronously to the operating of the prediction pipeline comprises continually indexing the memory device to read out the metadata for the prediction cache independent of processes in the prediction pipeline.

17. The computer program product of claim 15 , wherein the index buffer is configured to read out the memory device having the metadata for the prediction cache without reliance on and in advance of generating the prediction using the metadata from the prediction cache.

18. The computer program product of claim 15 , wherein the prediction pipeline is configured to output a line of the metadata of the instructions from the prediction cache.

19. The computer program product of claim 15 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions from the prediction cache without having to re-access the memory device.

20. The computer program product of claim 15 , wherein the prediction pipeline is configured to continuously reuse a line of the metadata of the instructions that has been output from the prediction cache in order to generate a new prediction, responsive to a previous prediction being predicted to the line of the metadata.