Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Submission Summary: 0 pending, 15 declined, 1 accepted (16 total, 6.25% accepted)

For the out-of-band Slashdot experience (mostly headlines), follow us on Twitter, or Facebook. ×

+ - Adventures in microchip repair

plcurechax writes: From Intel's own website, a "soft-news" or promotional pieces takes a high level look at technology behind fixing design mistakes in microprocessors, "Microscopic Adventures of a Chip Circuitry Repairman":

For nearly two decades, the pursuit of perfection has led Nikos Troullinos down minuscular rabbit holes to fix tiny design mistakes that can cause computer processor circuitry to malfunction..

While Slashdot regulars and IT veterans don't need to be reminded about well publicized follies of past processor flaws that have been discovered, from the infamous Pentium floating point division bug (FDIV) discovered in 1994, to the TSX flaw on Haswell to early Broadwell processors discovered in 2014. TSXTransactional Synchronization Extensions.

Given the complexity and vast number of processor models, few flaws are discovered outside of the manufacturer. I believe an average of less than 1.0 (flaw) per technology generation. While Intel's processor flaws are the best publicized, that is at least in part due to having the largest brand awareness amongst consumers. Non-Intel x86 and other non-x86 microprocessors have had flaws as well from classic 8-bit micros used in 1980s personal computers and game systems to the latest AMD and ARM offerings.

The Intel Pentium FDIV bug occurred at a time when the company had been spending considerable amounts of money and effort in mainstream advertising intended to build brand awareness, direct to average consumers, not just IT professors and computing enthusiasts. Bob Colwell, retired Intel engineer who worked on the Pentium Pro (P6) to the Pentium 4 (NetBurst / Willamette), discusses this in an appendix of his book, The Pentium Chronicles, Colwell discusses his own involvement in internal FDIV bug reporting, and Intel's surprise and poor handling of the public relations fiasco which perplexed top executives and engineers for quite some time.

You can do this in a number of ways. IBM chose to do all of them. Why do you find that funny? -- D. Taylor, Computer Science 350

Working...