IP Library › Granted Patent US 9,053,030
Granted Patent B2
US 9,053,030 · App. 13/144,820 · Granted Jun 9, 2015

Cache memory and control method thereof with cache hit rate

Inventor: Yasushi Kanoh (Tokyo, JP)
Assignee: NEC CORPORATION
G06F12/0842
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 9,053,030
App. No.
13/144,820
Granted
Jun 9, 2015
Kind
B2
Abstract

A cache memory comprises a data array that stores a cashed block; a first address array that stores an address of the cached block; a second address array that stores an address of a first block to be removed from the data array when a cache miss occurs; and a control unit that transmits to a processor the first block stored in the data array as a cache hit block, when the address stored in the second address array results in a cache hit during a period before a second block which has caused the cache miss is read from a memory and written into the data array.

Claims (21)

1. A cache memory comprising:

a data array that stores a cached block;

a first address array that stores an address of the cached block;

a second address array that stores an address of a first block to be removed from the data array when a cache miss occurs; and

a control unit that transmits to a processor the first block stored in the data array as a cache hit block, in response to the address stored in the second address array resulting in a cache hit during a period before a second block that has caused the cache miss to be read from a memory and written into the data array, wherein the cache memory is configured to wait for the second block to be read from memory when it is determined that an entry in the second address array corresponding to the first block is not vacant.

2. The cache memory according to claim 1 , wherein the control unit invalidates the address of the first block stored in the second address array when writing the second block into the data array.

3. The cache memory according to claim 1 , wherein the second address array stores an identifier for a way in which the first block is stored.

4. The cache memory according to claim 1 , wherein the second address array stores a flag indicating whether or not the first block has resulted in the cache hit and has been rewritten.

5. The cache memory according to claim 4 , wherein the control unit, before the second block is read from the memory and written into the data array, refers to the flag stored in the second address array and, in response to the first block having been rewritten, writes back the first block to the memory.

6. A cache memory comprising:

an address array that stores an address of a block to be removed from a data array when a cache miss occurs; and

a control unit that transmits to a processor the block stored in the data array as a cache hit block, in response to the address stored in the address array resulting in a cache hit during a period before a block that has caused the cache miss being read from a memory and stored in the data array, wherein the cache memory is configured to wait for the second block to be read from memory when it is determined that entry in the second address array corresponding to the first block is not vacant.

7. An electronic computer comprising the cache memory according to claim 1 .

8. The electronic computer according to claim 7 , comprising a multi-core or multi-thread processor.

9. A control method of a cache memory comprising a data array that stores a cached block, a first address array that stores an address of the cached block, and a second address array, the method comprising:

storing in the second address array an address of a first block to be removed from the data array when a cache miss occurs; and

transmitting to a processor the first block stored in the data array as a cache hit block, in response to the address stored in the second address array resulting in a cache hit during a period before a second block which has caused the cache miss being read from a memory and written into the data array, wherein the cache memory is configured to wait for the second block to be read from memory when it is determined that an entry in the second address array ding to the first block is not vacant.

10. The cache memory control method according to claim 9 , further comprising: invalidating the address of the first address stored in the second address array when writing the second block into the data array.

11. The cache memory control method according to claim 9 , comprising storing in the second address array an identifier for a way in which the first block is stored.

12. The cache memory control method according to claim 9 , comprising storing in the second address array a flag indicating whether or not the first block has resulted in the cache hit and has been rewritten.

13. The cache memory control method according to claim 12 , comprising, before the second block is read from the memory and written into the data array, referring to the flag stored in the second address array and, if the first block has been rewritten, writing back the first block to the memory.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 19, 2011
From: KANOH, YASUSHI
To: NEC CORPORATION
Reel/Frame 026611/0654 →
Priority Claims (1)
JP 2009-016224 · Jan 28, 2009 · national
Continuity (1)
Related Publication 20110283041A1 · Nov 17, 2011