Abstract: Embodiments of the invention are directed to methods for handling cache prefetch requests. The method includes receiving a request to prefetch data from main memory to a cache. The method further includes based on a determination that the prefetch request is a speculative prefetch request, determining if the cache is being used for transactional memory. The method further includes based on a determination that the cache is not being used for transactional memory, processing the prefetch request. The method further includes based on a determination that the cache is being used for transactional memory, and a determination if the prefetch request can be processed without affecting transactional memory, processing the prefetch request. The method further includes based on a determination that the cache is being used for transactional memory, and a determination if the prefetch request can be processed without affecting transactional memory, rejecting the prefetch request.

Abstract: A processor memory is stress tested with a variable link stack depth using link stack test segments with non-naturally aligned data boundaries. Link stack test segments are interspersed into test code of a processor memory tests to change the link stack depth without changing results of the test code. The link stack test segments are the same structure as the segments of the test code and have non-naturally aligned boundaries. The link stack test segments include branch to target, push/pop, push and pop segments. The depth of the link stack is varied independent of the memory test code by changing the number to branches in the branch to target segment and varying the number of the push/pop segments.

Abstract: A processor memory is stress tested with a variable link stack depth using link stack test segments with non-naturally aligned data boundaries. Link stack test segments are interspersed into test code of a processor memory tests to change the link stack depth without changing results of the test code. The link stack test segments are the same structure as the segments of the test code and have non-naturally aligned boundaries. The link stack test segments include branch to target, push/pop, push and pop segments. The depth of the link stack is varied independent of the memory test code by changing the number to branches in the branch to target segment and varying the number of the push/pop segments.

Abstract: Data is replicated into a memory cache and cache inhibited memory in data segments with segment size that provides 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 allows replicated testing of the memory cache and cache inhibited memory while preserving double word and quad word boundaries in segments of the replicated test data. This allows test cases generated for cacheable memory to be replicated and used for cache inhibited memory. The processor can then use a single test replicated in this manner by branching back and using the next slice of the replicated test data in the memory cache and cache inhibited memory.

Abstract: Embodiments herein provide a testing apparatus (whether physical or simulated) for testing a non-core MMU in a processor chip. Unlike core MMUs, non-core MMUs may be located in a part of the processor chip outside of the processing cores in the chip. Instead of being used to perform address translation requests sent by the processing core, the non-core MMUs may be used by other hardware modules in the processor chip such as compression engines, crypto engines, accelerators, etc. In one embodiment, the testing apparatus includes a MMU testor that transmits the translation requests to the non-core MMU which tests its functionality. Using the data provided in the translation requests, the non-core MMU performs virtual to physical address translations. The non-core MMU transmits the results of these translations to the MMU testor which compares these results to expected results to identify any design flaws in the non-core MMU.

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.

Abstract: Test cases for testing speculative execution of instructions are replicated into a memory with non-naturally aligned data boundaries to create a non-contiguous instruction stream to efficiently test a processor. Placing test cases with test code and test data in the non-naturally aligned data boundaries as described herein allows test code to test speculative execution of branches. The test case includes a branch with a hint bit set to cause the hardware to mispredict the path of the branch to cause speculative execution of test code, bad code or erroneously execute data. The processor can then be tested to see if it properly flushes the speculatively executed code upon taking the opposite branch of the mispredicted path.

Abstract: A mapping may be changed in a table stored in memory. The table may map a first set of addresses, for a set of data, to a second set of addresses. The changing of the mapping may including mapping the first set of addresses to a third set of addresses. In response to the changing of the mapping, one or more flush operations may be executed to invalidate one or more entries within one or more address translation caches. The one or more entries may include the second set of addresses. In response to the executing of the one or more flush operations, a first test case may be run. The first test case may be to test whether any of the first set of addresses are mapping to the second set of addresses.

Abstract: An aspect includes driving a plurality of commands to an interface unit of a circuit design in a hardware-accelerated simulator to dynamically initialize the circuit design to run one or more test cases based on an initialization sequence with breakpoint support. A state of the circuit design is examined through the interface unit based on triggering of a breakpoint on the hardware-accelerated simulator. A next action to perform in the initialization sequence is determined based on the state of the circuit design as determined through the interface unit. Execution of one or more scripts select the initialization sequence from a plurality of test cases, set the breakpoint, modify a state of the circuit design as the next action to perform, and capture a plurality of test results based on execution of the initialization sequence through the interface unit.

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.

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.

Abstract: A system and method synchronizes heterogeneous agents in a computer system with a software synchronization mechanism. Agents of the computer system connected to a common memory, including agents lacking a hardware synchronization system, can be synchronized with the software synchronization mechanism. The synchronized agents can cause collisions on the same cache line in order to stress test the memory of the system. Each agent updates a first array to indicate it has arrived at the synchronization. After all the agents have arrived, each agent then updates a second array to announce its exit.

Abstract: Embodiments disclose techniques for executing a test case to test a processor by bypassing an instruction pipeline of the processor. In one embodiment, the processor receives a plurality of test cases to execute on the processor. Each test case includes one or more instructions. Once received, the processor loads a plurality of registers with one or more first register values for the test case by bypassing the instruction pipeline. Once loaded, the processor runs the test case using the one or more first register values. The processor then retrieves, from the plurality of registers, one or more second register values associated with results of the test case run, by bypassing the instruction pipeline.

Abstract: Embodiments herein provide a testing apparatus (whether physical or simulated) for testing a non-core MMU in a processor chip. Unlike core MMUs, non-core MMUs may be located in a part of the processor chip outside of the processing cores in the chip. Instead of being used to perform address translation requests sent by the processing core, the non-core MMUs may be used by other hardware modules in the processor chip such as compression engines, crypto engines, accelerators, etc. In one embodiment, the testing apparatus includes a MMU testor that transmits the translation requests to the non-core MMU which tests its functionality. Using the data provided in the translation requests, the non-core MMU performs virtual to physical address translations. The non-core MMU transmits the results of these translations to the MMU testor which compares these results to expected results to identify any design flaws in the non-core MMU.

Abstract: A transactional memory test tests a transactional memory system on a computer processor. A test case with a transactional memory test is constructed such that the transactional memory test proceeds further in the transaction mode if the transaction is successful. In contrast, if there is a failure of the transaction, then the transactional memory test is executed in the non-transaction state. The transactional memory test stresses both transaction success and transaction failure cases of the processor for more efficient validation of the computer processor. Also, when checking the correctness of the test case the correct result remains the same whether the transaction passes or fails to simplify verification.

Abstract: Embodiments herein pre-load memory translations used to perform virtual to physical memory translations in a computing system that switches between virtual machines (VMs). Before a processor switches from executing the current VM to the new VM, a hypervisor may retrieve previously saved memory translations for the new VM and load them into cache or main memory. Thus, when the new VM begins to execute, the corresponding memory translations are in cache rather than in storage. Thus, when these memory translations are needed to perform virtual to physical address translations, the processor does not have to wait to pull the memory translations for slow storage devices (e.g., a hard disk drive).

Abstract: Embodiments herein pre-load memory translations used to perform virtual to physical memory translations in a computing system that switches between virtual machines (VMs). Before a processor switches from executing the current VM to the new VM, a hypervisor may retrieve previously saved memory translations for the new VM and load them into cache or main memory. Thus, when the new VM begins to execute, the corresponding memory translations are in cache rather than in storage. Thus, when these memory translations are needed to perform virtual to physical address translations, the processor does not have to wait to pull the memory translations for slow storage devices (e.g., a hard disk drive).

Abstract: An apparatus and a method are disclosed for tracking use of a specialized hardware component in symmetric multiprocessing computing device. A device includes a memory and one or more multi-core processors attached to a coherent memory bus. A proxy for a specialized hardware component (SHC) such as an accelerator, FPGA, or ASIC is placed in communication with the coherent memory bus. The proxy may be attached to another bus such as a peripheral component interconnect express (PCIe) bus. A tracker updates an allocated counting register with counts of events related to use of the SHC. When requested, information from the counting register is provided such as to an external device or client communicating with the multiprocessing computing device. The tracker may follow calls or messages to the SHC. The tracker may accumulate a count of message size or mode of use or use of certain functions of the SHC.

Abstract: A mapping may be changed in a table stored in memory. The table may map a first set of addresses, for a set of data, to a second set of addresses. The changing of the mapping may including mapping the first set of addresses to a third set of addresses. In response to the changing of the mapping, one or more flush operations may be executed to invalidate one or more entries within one or more address translation caches. The one or more entries may include the second set of addresses. In response to the executing of the one or more flush operations, a first test case may be run. The first test case may be to test whether any of the first set of addresses are mapping to the second set of addresses.

Abstract: Efficiently generating effective address translations for memory management test cases including obtaining a first set of EAs, wherein each EA comprises an effective segment ID and a page, wherein each effective segment ID of each EA in the first set of EAs is mapped to a same first effective segment; obtaining a set of virtual address corresponding to the first set of EAs; translating the first set of EAs by applying a hash function to each virtual address in the set of virtual addresses to obtain a first set of PTEG addresses mapped to a first set of PTEGs; and generating a translation for a second set of EAs to obtain a second set of PTEG addresses mapped to the first set of PTEGs.