Gathering and scattering multiple data elements

Patent metadata

Field	Value
Publication number	US-10114651-B2
Application number	US-201815862407-A
Country	US
Kind code	B2
Filing date	Jan 4, 2018
Priority date	Dec 22, 2009
Publication date	Oct 30, 2018
Grant date	Oct 30, 2018

Abstract

According to a first aspect, efficient data transfer operations can be achieved by: decoding by a processor device, a single instruction specifying a transfer operation for a plurality of data elements between a first storage location and a second storage location; issuing the single instruction for execution by an execution unit in the processor; detecting an occurrence of an exception during execution of the single instruction; and in response to the exception, delivering pending traps or interrupts to an exception handler prior to delivering the exception.

First claim

What is claimed is:

1. A processor comprising:
a destination register to store a plurality of data elements;
a source register to store a plurality of index values, each of which corresponds to one of the plurality of data elements;
a base address register to store a base address;
a mask register to store mask values, each mask value corresponding to one of the plurality of data elements;
a decoder to decode a vector gather instruction, the vector gather instruction having a scale field to specify a common scaling factor to be applied to the index values; and
execution circuitry coupled to the decoder, the execution circuitry to perform operations associated with the vector gather instruction, the operations comprising conditionally accessing, based on corresponding mask values, one or more of the plurality of data elements and storing the one or more of the plurality of data elements in the destination register;
wherein the execution circuitry is to scale the index values in accordance with the scale field to generate a corresponding plurality of scaled index values, and add the base address to each of the scaled index values to generate a corresponding plurality of non-contiguous system memory addresses for the data elements to be accessed and stored in the destination register.

2. The processor of claim 1 wherein system memory addresses are determined by adding a displacement value to the combination of the base address and the scaled index values.

3. The processor of claim 1 further comprising:
a plurality of cores, wherein the destination register, source register, base address register, mask register, decoder, and execution circuitry are integral to a first core of the plurality of cores.

4. The processor of claim 1 further comprising:
fetch circuitry and/or prefetch circuitry to fetch and/or prefetch the vector gather instruction from a system memory.

5. The processor of claim 3 further comprising:
a memory controller to couple the cores to a system memory.

6. The processor of claim 1 further comprising:
register rename circuitry to map one or more logical registers to physical registers of a register file.

7. The processor of claim 3 further comprising:
a level 1 cache integral to the first core to store instructions and data.

8. The processor of claim 1 further comprising:
a first interconnect to couple the processor to one or more system components.

9. The processor of claim 2 further comprising:
a second interconnect to couple the processor to one or more other processors.

10. The processor of claim 9 further comprising:
a third interconnect to couple the processor to a system memory.

Assignees

• Hughes Christopher J
• Chen Yen Kuang Y K
• Bomb Mayank
• Brandt Jason W
• Buxton Mark J
• Charney Mark J
• Chennupaty Srinivas
• Corbal Jesus
• Dixon Martin G
• Girkar Milind B
• Hall Jonathan C
• Ido Hideki Saito
• Lachner Peter
• Neiger Gilbert
• Newburn Chris J
• Parthasarathy Rajesh S
• Toll Bret L
• Valentine Robert
• Wiedemeier Jeffrey G
• Intel Corp

Inventors

• Hughes Christopher J
• Chen Yen-Kuang (Y K )
• Bomb Mayank
• Brandt Jason W
• Buxton Mark J
• Charney Mark J
• Chennupaty Srinivas
• Corbal Jesus
• Dixon Martin G
• Girkar Milind B
• Hall Jonathan C
• Ido Hideki (Saito)
• Lachner Peter
• Neiger Gilbert
• Newburn Chris J
• Parthasarathy Rajesh S
• Toll Bret L
• Valentine Robert
• Wiedemeier Jeffrey G

Classifications

• G06F9/30043

  LOAD or STORE instructions; Clear instruction · CPC title

• G06F9/3861 Primary

  Recovery, e.g. branch miss-prediction, exception handling (error detection or correction G06F11/00) · CPC title

• G06F9/30036

  Instructions to perform operations on packed data, e.g. vector, tile or matrix operations · CPC title

• G06F9/30038

  using a mask · CPC title

• G06F9/345

  of multiple operands or results {(addressing multiple banks G06F12/06)} · CPC title