The present disclosure is related to cache memory, such as, for tracking defective cache lines.
As is well-known, a cache stores information for a computer or computing system in order to decrease data retrieval times for a processor. Some examples of computing systems are a personal digital assistant, internet tablet, and a cellular phone. The cache stores specific subsets of information in high-speed memory. A few examples of information are instructions, addresses, and data. When a processor requests a piece of information, the system checks the cache first to see if the information is stored within the cache. If so, the processor can retrieve the information much faster than if the data was stored in other computer readable media, such as, random access memory, a hard drive, compact disc read-only memory (CD ROM), or a floppy disk.
Modern integrated circuit (IC) devices include large numbers of gates on a single semiconductor chip, with these gates interconnected so as to perform multiple and complex functions. The fabrication of an IC incorporating such Very Large Scale Integration (VLSI) must be error free, as any manufacturing defect may prevent the IC from performing all of the functions that an IC or SoC is designed to perform. Such demands require verification of the design of the IC or SoC and also various types of electrical testing after the IC or SoC is manufactured.
As cache sizes increase, so does the impact to yield. Consequently, testing and manufacturing become more complex. Typically, one reduces the impact to yield by allowing a processor to have a predetermined number of defective lines. However, one would need to disable the defective lines during a manufacturing or normal execution. Regardless of the method of disabling the defective lines, one needs to track the defective lines. For example, one needs to track the defective lines across processor reset since the reset may have destroying the tags associated with the defective lines. One solution is placing a fixed number of defective cache lines in dedicated on-core resources. Consequently, this increases hardware requirements. Furthermore, this limits the number of disabled lines by the amount of dedicated resources and may only allow for a fixed number of disabled lines.
Claimed subject matter is particularly and distinctly pointed out in the concluding portion of the specification. The claimed subject matter, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings in which:
In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the claimed subject matter. However, it will be understood by those skilled in the art that the claimed subject matter may be practiced without these specific details. In other instances, well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the claimed subject matter.
An area of current technological development relates to improving the yield efficiency of cache memory. As previously described, one needs to track the disable lines across processor reset since the reset may have destroying the tags associated with the defective lines. One solution is placing the disabled cache lines in dedicated on-core resources, such as, on a processor. Consequently, this increases hardware requirements. Furthermore, this limits the number of disabled lines by the amount of dedicated resources. In contrast, the claimed subject matter teaches utilizing data bits in defective cache lines, such as, cache lines that have data bit errors, for creating a linked list or other data structure for storing relevant information regarding defective cache lines.
In one embodiment, an additional MESI (Modified Exclusive Shared Invalid) state could be utilized to disable cache lines in the data array. For example, in one embodiment, a P state is defined for a disabled cache line.
For example, seventeen bits for a 12 Mb cache (four for determining the way and thirteen for the index) is utilized to determine the next disabled cache line for a cache with a 1K bit line. Thus, sixty copies of the pointer (17 bits) to the next line can be used. In one embodiment, one could use an algorithm to analyze every pointer copy to determine which pointer pattern is represented the most times. Alternatively, one could also utilize only one pointer and use ECC. For the preceding example of 60 pointer copies in one disabled/defective cache line, twenty nine pointer copies could be destroyed due to defects. However, the data for the pointer could still be restored since thirty one copies are valid. Furthermore, the last disabled cache line could contain a repeated termination pattern for the pointer. Alternatively, one could utilize an invalid way encoding for the pointer in the last disabled cache line to indicate that this is the last disabled cache line.
The analysis continues with block 206 since the pointer is pointing to a cache line as indicated by the pointer in the previously described MSR. Since the pointer indicates that this line should be disabled, this particular cache line is placed in a disabled state, “P state”. Subsequently, a repetitive analysis is performed to determine whether the present cache line has a pointer to another cache line. If so, the single or multiple copies of data for the pointer is extracted to set the next pointer to the next cache line which will be disabled, as previously described in block 206. Otherwise, the processor continues with the reset operation.
Reference in the specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
In the foregoing specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes can be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Number | Name | Date | Kind |
---|---|---|---|
4357656 | Saltz et al. | Nov 1982 | A |
5200959 | Gross et al. | Apr 1993 | A |
5493667 | Huck et al. | Feb 1996 | A |
5958068 | Arimilli et al. | Sep 1999 | A |
5966734 | Mohamed et al. | Oct 1999 | A |
5974505 | Kuttanna et al. | Oct 1999 | A |
6108775 | Shiell et al. | Aug 2000 | A |
6263404 | Borkenhagen et al. | Jul 2001 | B1 |
6467048 | Olarig et al. | Oct 2002 | B1 |
6594785 | Gilbertson et al. | Jul 2003 | B1 |
6629201 | Dempsey et al. | Sep 2003 | B2 |
6922798 | Nemani et al. | Jul 2005 | B2 |
7073099 | Sutardja et al. | Jul 2006 | B1 |
7117312 | Cypher | Oct 2006 | B1 |
20020199142 | Gefen | Dec 2002 | A1 |
20030074601 | Schultz et al. | Apr 2003 | A1 |
20040080988 | Harari et al. | Apr 2004 | A1 |
20040153793 | Jarboe et al. | Aug 2004 | A1 |
20040170064 | Harari et al. | Sep 2004 | A1 |
20050286336 | Harari et al. | Dec 2005 | A1 |
Number | Date | Country | |
---|---|---|---|
20060015768 A1 | Jan 2006 | US |