Conventional on-chip TLB hierarchies are unable to fully cover the growing application working-set sizes. To make things worse, Last-Level TLB (LLT) misses require multiple accesses to the page table even with the use of page walk caches. Consequently, LLT misses incur long address translation latency and hurt performance. This article proposes two low-overhead hardware mechanisms for reducing the frequency and penalty of on-die LLT misses. The first, Unified CAche and TLB (UCAT) , enables the conventional on-die Last-Level Cache to store cache lines and TLB entries in a single unified structure and increases on-die TLB capacity significantly. The second, DRAM-TLB , memoizes virtual to physical address translations in DRAM and reduces LLT miss penalty when UCAT is unable to fully cover total application working-set. DRAM-TLB serves as the next larger level in the TLB hierarchy that significantly increases TLB coverage relative to on-chip TLBs. The combination of these two mechanisms, DUCATI , is an address translation architecture that improves GPU performance by 81%; (up to 4.5×) while requiring minimal changes to the existing system design. We show that DUCATI is within 20%, 5%, and 2% the performance of a perfect LLT system when using 4KB, 64KB, and 2MB pages, respectively.

The author describes an experiment in which built-in self-test (BIST) was added to existing gate-array probe designs and simulation was used to easily identify controllability and visibility problems. He documents investigations of ECL (emitter coupled logic) gate arrays to study the advantages and costs of using BIST. The first goal of this experiment was to develop a set of rules or a design methodology that would make it possible to implement these BIST test cases without undue effort. The second was to measure the effectiveness of BIST by fault simulation. The third was to document the cost in power, pins, gates, and computer resources in the design environment. Three test case designs were chosen for this experiment such that the results would be as representative as possible.>