IP Library Granted Patent US 10,614,151
Granted Patent B2
US 10,614,151 · App. 16/528,826 · Granted Apr 7, 2020

Permuting in a matrix-vector processor

Inventors: Dong Hyuk Woo (San Jose, CA); Gregory Michael Thorson (Waunakee, WI); Andrew Everett Phelps (Middleton, WI); Olivier Temam (Antony, FR); Jonathan Ross (Mountain View, CA); Christopher Aaron Clark (Madison, WI)
Assignee: Google LLC
G06F17/16G06F7/76G06F9/30032G06F9/30036
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,614,151
App. No.
16/528,826
Granted
Apr 7, 2020
Kind
B2
Abstract

A circuit comprises an input register configured to receive an input vector of elements, a control register configured to receive a control vector of elements, wherein each element of the control vector corresponds to a respective element of the input vector, and wherein each element specifies a permutation of a corresponding element of the input vector, and a permute execution circuit configured to generate an output vector of elements corresponding to a permutation of the input vector. Generating each element of the output vector comprises accessing, at the input register, a particular element of the input vector, accessing, at the control register, a particular element of the control vector corresponding to the particular element of the input vector, and outputting the particular element of the input vector as an element at a particular position of the output vector that is selected based on the particular element of the control vector.

Claims (30)

1. A circuit comprising:

an input register configured to receive an input vector of input elements;

a control register configured to receive a control vector of control elements, wherein each control element of the control vector (i) corresponds to a respective output element of an output vector of output elements corresponding to a transformation of the input vector and (ii) defines which input element of the input vector is used to generate the corresponding output element of the output vector; and

an execution circuit configured to generate the output vector, wherein generating each output element of the output vector comprises:

accessing, at a particular position of the control register, a particular control element of the control vector corresponding to a particular position of the output vector;

selecting a particular position of the input register based on the particular control element of the control vector corresponding to the particular position of the output vector;

accessing, at the particular position of the input register, a particular input element of the input vector; and

outputting the particular input element of the input vector as an output element at the particular position of the output vector.

2. The circuit of claim 1 , comprising an output register configured to receive the output vector.

3. The circuit of claim 2 , wherein the output register is configured to output the output vector as a staggered output such that a single output element of the output vector is output at each cycle in an order beginning with a lowest-order bit position of the output vector.

4. The circuit of claim 1 , wherein the input vector is received as a staggered input such that a single input element of the input vector is received at each cycle in an order beginning with a lowest-order bit position of the input vector.

5. The circuit of claim 4 , wherein receiving the input vector as a staggered input comprises:

receiving, at a flattening register, each input element of the input vector in a separate cycle in an order beginning with a lowest-order bit position of the flattening register; and

popping all input elements of the input vector simultaneously from the flattening register to the input register.

6. The circuit of claim 5 , wherein popping all input elements of the input vector simultaneously from the flattening register to the input register comprises:

determining that a highest-order bit of the flattening register has received valid data corresponding to a highest-order bit of the input vector; and

popping all input elements of the input vector simultaneously from the flattening register to the input register in response to determining that the highest-order bit of the flattening register has received valid data corresponding to the highest-order bit of the input vector.

7. The circuit of claim 5 , wherein popping all input elements of the input vector simultaneously from the flattening register to the input register comprises:

determining that the flattening register has received a number of input elements of the input vector equal to a dimension of the input vector; and

popping all input elements of the input vector simultaneously from the flattening register to the input register in response to determining that the flattening register has received the number of input elements of the input vector equal to the dimension of the input vector.

8. The circuit of claim 1 , wherein the control vector is received from an off-chip processor that is separate from the circuit.

9. The circuit of claim 1 , wherein the execution circuit comprises a memory crossbar.

10. The circuit of claim 1 , wherein the execution circuit comprises multiple many-to-one multiplexors, and wherein an output of each many-to-one multiplexor of the execution circuit is controlled by a corresponding control element of the control vector.

11. The circuit of claim 1 , wherein the particular control element of the control vector corresponding to the particular position of the output vector specifies the particular position of the input register.

12. The circuit of claim 1 , wherein the particular control element of the control vector corresponding to the particular position of the output vector specifies a number of positions to rotate the particular input element of the input vector for the particular input element of the input vector to be output as an output element at the particular position of the output vector.

13. The circuit of claim 12 , wherein selecting the particular position of the input register based on the particular control element of the control vector corresponding to the particular position of the output vector comprises:

identifying a position of the input register that is the specified number of positions from the particular position of the output vector; and

selecting the identified position of the input register that is the specified number of positions from the particular position of the output vector as the particular position of the input register.

14. The circuit of claim 12 , wherein the number of positions to rotate the particular input element of the input vector is a number of positions to rotate the particular input element of the input vector in a particular direction.

15. The circuit of claim 1 , wherein the input vector of input elements corresponds to a row of an input matrix or a column of an input matrix.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 2, 2019
From: WOO, DONG HYUK; THORSON, GREGORY MICHAEL; PHELPS, ANDREW EVERETT; TEMAM, OLIVIER; ROSS, JONATHAN; CLARK, CHRISTOPHER AARON
To: GOOGLE INC.
Reel/Frame 049939/0049 →
ENTITY CONVERSION Recorded Aug 2, 2019
From: GOOGLE INC.
To: GOOGLE LLC
Reel/Frame 049945/0060 →
Continuity (5)
Continuation 16283913 · Feb 25, 2019
Continuation 15966275 · Apr 30, 2018
Continuation 15496418 · Apr 25, 2017
Provisional Application 62460394 · Feb 17, 2017
Related Publication 20190354570A1 · Nov 21, 2019