Managing a divided load reorder queue

Abstract

Managing a divided load reorder queue including storing load instruction data for a load instruction in an expanded LRQ entry in the LRQ; launching the load instruction from the expanded LRQ entry; determining that the load instruction is in a finished state; moving a subset of the load instruction data from the expanded LRQ entry to a compact LRQ entry in the LRQ, wherein the compact LRQ entry is smaller than the expanded LRQ entry; and removing the load instruction data from the expanded LRQ entry.

1 . A method of managing a divided load reorder queue (LRQ), the method comprising:
 determining that a load instruction is in a finished state; and   moving a subset of load instruction data from an expanded LRQ entry in the LRQ to a compact LRQ entry in the LRQ, wherein the compact LRQ entry is smaller than the expanded LRQ entry.   
     
     
         2 . The method of  claim 1 , further comprising:
 determining that the load instruction is in a completed state; and   removing the subset of the load instruction data from the compact LRQ entry in the LRQ.   
     
     
         3 . The method of  claim 1 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 determining that at least one compact LRQ entry is available to store the subset of the load instruction data.   
     
     
         4 . The method of  claim 1 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 moving the load instruction data from the expanded LRQ entry to a staging area;   extracting the subset of the load instruction data from the load instruction data; and   storing the extracted subset of the load instruction data in the compact LRQ entry in the LRQ.   
     
     
         5 . The method of  claim 1 , wherein the expanded LRQ entry is one of a plurality of expanded LRQ entries in the LRQ, and wherein the compact LRQ entry is one of a plurality of compact LRQ entries in the LRQ. 
     
     
         6 . The method of  claim 1 , wherein the subset of the load instruction data comprises a hash of a portion of the load instruction data. 
     
     
         7 . The method of  claim 1 , wherein a size of the compact LRQ entry is less than half of a size of the expanded LRQ entry. 
     
     
         8 . A multi-slice processor configured to carry out:
 determining that a load instruction is in a finished state; and   moving a subset of load instruction data from an expanded LRQ entry in the LRQ to a compact LRQ entry in the LRQ, wherein the compact LRQ entry is smaller than the expanded LRQ entry.   
     
     
         9 . The multi-slice processor of  claim 8 , further configured to carry out:
 determining that the load instruction is in a completed state; and   removing the subset of the load instruction data from the compact LRQ entry in the LRQ.   
     
     
         10 . The multi-slice processor of  claim 8 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 determining that at least one compact LRQ entry is available to store the subset of the load instruction data.   
     
     
         11 . The multi-slice processor of  claim 8 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 moving the load instruction data from the expanded LRQ entry to a staging area;   extracting the subset of the load instruction data from the load instruction data; and   storing the extracted subset of the load instruction data in the compact LRQ entry in the LRQ.   
     
     
         12 . The multi-slice processor of  claim 8 , wherein the expanded LRQ entry is one of a plurality of expanded LRQ entries in the LRQ, and wherein the compact LRQ entry is one of a plurality of compact LRQ entries in the LRQ. 
     
     
         13 . The multi-slice processor of  claim 8 , wherein the subset of the load instruction data comprises a hash of a portion of the load instruction data. 
     
     
         14 . The multi-slice processor of  claim 8 , wherein a size of the compact LRQ entry is less than half of a size of the expanded LRQ entry. 
     
     
         15 . An apparatus comprising a multi-slice processor configured to carry out:
 determining that a load instruction is in a finished state; and   moving a subset of load instruction data from an expanded LRQ entry in the LRQ to a compact LRQ entry in the LRQ, wherein the compact LRQ entry is smaller than the expanded LRQ entry.   
     
     
         16 . The apparatus of  claim 15 , where in the multi-slice processor is further configured to carry out:
 determining that the load instruction is in a completed state; and   removing the subset of the load instruction data from the compact LRQ entry in the LRQ.   
     
     
         17 . The apparatus of  claim 15 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 determining that at least one compact LRQ entry is available to store the subset of the load instruction data.   
     
     
         18 . The apparatus of  claim 15 , wherein moving the subset of the load instruction data from the expanded LRQ entry to the compact LRQ entry in the LRQ comprises:
 moving the load instruction data from the expanded LRQ entry to a staging area;   extracting the subset of the load instruction data from the load instruction data; and   storing the extracted subset of the load instruction data in the compact LRQ entry in the LRQ.   
     
     
         19 . The apparatus of  claim 15 , wherein the expanded LRQ entry is one of a plurality of expanded LRQ entries in the LRQ, and wherein the compact LRQ entry is one of a plurality of compact LRQ entries in the LRQ. 
     
     
         20 . The apparatus of  claim 15 , wherein the subset of the load instruction data comprises a hash of a portion of the load instruction data.