Replicating test case data into a cache with non-naturally aligned data boundaries

Patent metadata

Field	Value
Publication number	US-10489259-B2
Application number	US-201815887968-A
Country	US
Kind code	B2
Filing date	Feb 2, 2018
Priority date	Jan 29, 2016
Publication date	Nov 26, 2019
Grant date	Nov 26, 2019

Abstract

Data is replicated into a memory cache with non-naturally aligned data boundaries to reduce the time needed to generate test cases for testing a processor. Placing data in the non-naturally aligned data boundaries as described herein allows replicated testing of the memory cache while preserving double word and quad word boundaries in segments of the replicated test data. This allows test cases to be generated for a section of memory and then replicated throughout the memory and tested by a single test branching back and using the next strand of the replicated test data in the memory cache.

First claim

The invention claimed is:

1. A computer-implemented method executed by at least one processor for testing a computer processor device comprising:
providing test data comprising a plurality of segments of test data that when placed consecutively in a memory together make a slice of test data, wherein each of the plurality of segments making up the slice of test data fill a portion of the memory comprising a plurality of cache lines of the memory; wherein the plurality of test segments are non-naturally aligned where the beginning and ending of each of the plurality of segments when placed consecutively in a memory do not line up with a cache line boundary, wherein the plurality of segments comprise an odd number of words where the odd number is chosen from 5, 7, 9, and 11;
placing multiple instances of the slice of test data in consecutive locations in a cache memory; and
running a test code on the consecutive slices of test data with non-naturally aligned boundaries by branching back to rerun the test code on each of the slices of test data.

2. The method of claim 1 wherein the step of running a test code on the consecutive slices of test data with non-naturally aligned boundaries further comprises:
executing test code with one or more test cases on a first slice of test data of the plurality of test data slices using a base offset;
determining if there are additional slices of test data; and
where there are additional slices of test data, modifying the base offset to point to a next test data slice and branching back to execute the test code with the modified base offset.

3. The method of claim 1 wherein the plurality of segments of test data comprises sub-segments of test data that include word, double word and quad word sub-segments.

4. The method of claim 3 wherein sub-segments in subsequent segments of the test data are arranged in different orders.

5. The method of claim 1 wherein each segment includes one single word, one double word and one quad word sub-segment.

6. The method of claim 1 wherein the slice of test data further comprises a plurality of strands of test cases that each comprise a plurality of segments.

7. The method of claim 6 wherein the plurality of strands comprises five strands with four segments in each strand with each segment having seven words of test data with four bytes in each word.

8. The apparatus of claim 1 wherein the plurality of segments are non-naturally aligned where the beginning and ending of each of the segments does not line up with a page crossing boundary.

9. The apparatus of claim 1 wherein the plurality of segments are non-naturally aligned where the beginning and ending of each of the segments does not line up with a 32 byte boundary.

10. A computer-implemented method executed by at least one processor for testing a computer processor device comprising:
providing test data comprising a plurality of segments of test data that when placed consecutively in a memory together make a slice of test data, wherein each of the plurality of segments making up the slice of test data fill a portion of the memory comprising a plurality of cache lines of the memory; wherein the plurality of test segments are non-naturally aligned where the beginning and ending of each of the plurality of segments when placed consecutively in a memory do not line up with a cache line boundary, wherein the plurality of segments comprise one single word, one double word and one quad word sub-segment, wherein sub-segments in subsequent segments of the test data are arranged in different orders;
placing multiple instances of the slice of test data in consecutive locations in a cache memory; and
running a test code on the consecutive slices of test data with non-naturally aligned boundaries by branching back to rerun the test code on each of the slices of test data.

Assignees

• IBM

Inventors

• Dusanapudi Manoj
• Kapoor Shakti

Classifications

• G06F11/263 Primary

  Generation of test inputs, e.g. test vectors, patterns or sequences {; with adaptation of the tested hardware for testability with external testers} · CPC title

• G11C29/36

  Data generation devices, e.g. data inverters · CPC title

• G06F11/2635

  using a storage for the test inputs, e.g. test ROM, script files · CPC title

• G06F3/065

  Replication mechanisms · CPC title

• G06F12/0806

  Multiuser, multiprocessor or multiprocessing cache systems · CPC title