摘要:
A system and method for re-executing a test case and modifying the test case's effective addresses, effective segment identifiers (ESIDs), and virtual segment identifiers (VSIDs) in order to fully test a processor's SLB and TLB cells is presented. A test case generator generates a test case that includes an initial set of test case effective addresses, an initial set of ESIDs, and an initial set of VSIDs. The test case executor uses an effective address arithmetic function and a virtual address arithmetic function to modify the test case effective addresses, the ESIDs, and the VSIDs on each re-execution that, in turn, sets/unsets each bit within each SLB and TLB entry. In one embodiment, the invention described herein sequentially shifts segment lookaside buffer entries, whose ESIDs are in single bit increments, in order to fully test each ESID bit location within each SLB entry.
摘要:
A system and method for using resource pools and instruction pools for processor design verification and validation is presented. A test case generator organizes processor resources into resource pools using a resource pool mask. Next, the test case generator separates instructions into instruction pools based upon the resources that each instruction requires. The test case generator then creates a test case using one or more sub test cases by assigning a resource pool to each sub test case, identifying instruction pools that correspond the assigned test case, and building each sub test case using instructions included in the identified instruction pools.
摘要:
A system and method to reduce verification time by sharing memory between multiple test patterns and performing results checking after each test pattern executes one time is presented. A test pattern generator generates multiple test pattern sets, each of which including multiple test patterns. Each test pattern set is executed by a corresponding thread/processor until each test pattern included in the test pattern set has executed at least once. After all test patterns have executed at least once, a test pattern executor performs a memory error detection check to determine whether the system is functioning correctly. Since the invention described herein waits until all test patterns have executed before performing a memory error detection check, less time is spent on memory error detection checks, which allows more time to execute test patterns.
摘要:
A system and method processor testing using test pattern re-execution is presented. A processor re-executes test patterns using different timing scenarios in order to reduce test pattern build time and increase system test coverage. The invention described herein varies initial states of a processor's memory (cache, TLB, SLB, etc.) that, in turn, varies the timing scenarios when re-executing test patterns. By re-executing the test patterns instead of rebuilding new test patterns, verification quality is improved since more time is available for execution, verification and validation. In addition, since the test patterns result in the same final state, the invention described herein also simplifies error checking.
摘要:
A system and method for including independent instructions into a test case for intentionally provoking interrupts that may be used in conjunction with an instruction shuffling process is presented. A test case generator builds a test case that includes intentional interrupt instructions, which are constructed to intentionally provoke an interrupt, such as an instruction storage interrupt (ISI), a data storage interrupt (DSI), and alignment interrupt, and/or a program interrupt (PI). When a processor executes the test case and invokes an interrupt to an interrupt handler, the interrupt handler does not resolve the interrupt, but rather increments an instruction address register or a link register and resumes test case execution at an instruction subsequent to the instruction that caused the interrupt.
摘要:
During a test pattern build, a test pattern generator pseudo-randomly selects an address for a selected lwarx instruction and builds the lwarx instruction using the pseudo-random address into a test pattern. Subsequently, the test pattern generator builds a store instruction after the lwarx instruction using the pseudo-random address. The store instruction is adapted to store the pseudo-random address in a predetermined memory location. The test pattern generator also builds an interrupt service routine that services an interrupt associated with the interrupt request; checks the predetermined memory location; determines that the pseudo-random address is located in the predetermined memory location; and executes a subsequent lwarx instruction using the pseudo-random address.
摘要:
A system and method for verifying cache snoop logic and coherency between instruction cache and data cache using instruction stream “holes” that are created by branch instructions is presented. A test pattern generator includes instructions that load/store data into instruction stream holes. In turn, by executing the test pattern, a processor thread loads an L2 cache line into both instruction cache (icache) and data cache (dcache). The test pattern modifies the data in the dcache in response to a store instruction. In turn, the invention described herein identifies whether snoop logic detects the change and updates the icache's corresponding cache line accordingly.
摘要:
The present invention provides a method, computer program product, input/output device, and computer system for stress testing the I/O subsystem of a computer system. An input/output device capable of engaging in repetitive direct memory access (DMA) transfers with pseudo-randomized transfer parameters is allowed to execute multiple DMA transfers with varying parameters. In this way, a single type of device may be used to simulate the effects of multiple types of devices. Multiple copies of the same I/O device may be used concurrently in a single computer system along with processor software to access the same portions of memory. In this way, false sharing, true sharing may be effected.
摘要:
A method, apparatus and product for hardware verification using acceleration platform. The method comprising executing a first post-silicon testing program by a reference model, wherein during said executing the first post-silicon testing program one or more test-cases are generated; generating a second post-silicon testing program that is configured to execute the one or more test-cases; and executing the second post-silicon testing program on an acceleration platform.
摘要:
Testing real page number bits in a cache directory is provided. A specification of a cache to be tested is retrieved in order to test the real page number bits of the cache directory associated with the cache. A range within a real page number address of the cache directory is identified for performing page allocations using the specification of the cache. A random value x is generated that identifies a portion of the real page number bits to be tested. A first random value y is generated that identifies a first congruence class from a set of congruence classes within the portion of the cache to be tested. Responsive to the first congruence class failing to be allocated a predetermined number of times, one page size of memory for the first congruence class is allocated and a first allocation value is incremented by a value of 1.