Stuck at faults in CPU registers
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2021-02-24 3:03 AM
Hi,
Can somebody help me to understand the below thing?
How to detect CPU registers stuck at faults? If CPU registers are stuck at fault, How can instruction send/stored in that register will be the actual one? I mean to test CPU register stuck at fault we will write code, Some of the data related to the test might store in CPU registers right, If it stuck at fault it will not store the data related to the test. In that case, how can we test those registers?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2021-02-24 3:25 AM
Well assuming they don't all fail, put the same value in all and then compare against each other, and then invert the pattern. Use multiple stress patterns, use different literal pools, and values you can load as immediate values.​
Up vote any posts that you find helpful, it shows what's working..
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2021-02-24 4:01 AM
In order to load patterns, we will write code, What if loading pattern data stored in CPU registers, have stuck at fault. In that case it may not store actual data right? Again it will be problem for testing faults ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
‎2021-02-24 4:09 AM
How much fail do you need to detect before determining the CPU is unusable?
Perhaps do some cascading failure testing first, and then move to word and bit level register testing.
Up vote any posts that you find helpful, it shows what's working..
