Trace-driven simulation is often used in the design of computer systems, especially caches and translation lookaside buffers. Capturing address traces to drive such simulations has been problematic, often involving 1000:1 software overhead to trace a target workload, and/or mechanisms that cause significant distortions in the recorded data. A new technique for capturing address traces has been developed to use a processor's microcode to record addresses in a reserved part of main memory as a side effect of normal execution. An experimental implementation of this technique on a VAX1 8200 processor shows a number of advantages over previous techniques, including fewer distortions of the address trace and a hundred times faster recording. With this technique, it is possible to gather full operating-system traces of multi-tasking workloads.

ATUM: a new technique for capturing address traces using microcode

A computer system includes a memory for storing instructions executable by a processor and an on-chip trace memory having a plurality of locations for storing trace information that indicates execution flow in the processor. A trace access instruction provides for access to the on-chip trace memory on the processor. The trace access instruction can be a write instruction or a read instruction. Typically, both read and write to the trace memory is provided. The system also has the capability to trace on start or restart of an executable thread by providing to the processor an indication of which executable thread to trace via a debug port. That indicates to the processor to provide trace information when the executable thread starts. When execution of the executable thread starts, the processor places an identifier corresponding to the executable thread into the trace memory to indicate that subsequent entries placed into the trace memory are part of said executable thread. The processor may also provide an entry indicating when the thread stops executing.

Processor having a trace access instruction to access on-chip trace memory

A method is provided for scheduling execution of a plurality of threads executed in a multithreaded processor. Resource utilizations of each of the plurality of threads are measured while the plurality of threads are concurrently executing in the multithreaded processor. Each of the plurality of threads is scheduled according to the measured resource utilizations using a thread scheduler.

Method for scheduling threads in a multithreaded processor

Executing programs coded in an instruction set of a first computer on a computer of a second, different architecture. An operating system maintains an association between each one of a set of concurrent threads and a set of computer resources of the thread's context. Without modifying a pre-existing operating system of the computer, an entry exception is establishing to be raised on each entry to the operating system at a specified entry point or on a specified condition. The entry exception has an associated entry handler programmed to save a context of an interrupted thread and modify the thread context before delivering the modified context to the operating system. A resumption exception is established to be raised on each resumption from the operating system complementary to one of the specified entries. The resumption exception has an associated exit handler programmed to restore the context saved by a corresponding execution of the entry handler. The entry exception, exit exception, entry handler, and exit handler are cooperatively designed to maintain an association between a one of the threads and an extended context of the thread through a context change induced by the operating system, the extended context including resources of the computer associated with the thread beyond those resources whose association with the thread is maintained by the operating system.

Apparatus for executing programs for a first computer architechture on a computer of a second architechture

In-circuit emulation (ICE) and software debug facilities are included in a processor via a debug interface that interfaces a target processor to a host system. The debug interface, located on the target processor, includes a trace controller that monitors signals produced by the target processor to detect specified conditions and to produce a trace record of the specified conditions including a notification of the conditions and selected information relating to the conditions. The trace controller formats a trace information record and stores the trace information record in a trace buffer in a plurality of trace data storage elements. The trace data storage elements have a format that includes a trace code (TCODE) field indicative of a type of trace information and a trace data (TDATA) field indicative of a type of trace information data.

Debug interface including a compact trace record storage

An operation of a processor is traced while fetching instructions from a memory to operate the processor. The tracing involves detecting an unpredictable fetching of instructions on the assumption that a predictable fetching can be reconstructed without any further input. The unpredictable fetching is identified as being due to either computable, conditional, or unanticipated events. Upon detecting the events, process control information, such as the next instruction to be fetched is recorded in a queue, and from the queue the information can be stored in a trace buffer. During reconstruction of the operation, the trace buffer, and the image including the instructions can be examined to analyze the real-time operation of the processor.

Method and apparatus for tracing unpredictable execution flows in a trace buffer of a high-speed computer system

A data flow of a processor is traced while accessing data stored in a memory and in a plurality of registers during operation of the processor. The tracing involves detecting an unpredictable accessing of data on the assumption that a predictable accessing can be reconstructed without any further input. The unpredictable accessing is identified by setting and clearing a trace bit associated with each of the registers according to identifying the accessing as direct memory-to-register, register-to-register, constant-to-register, and indirect memory. If a trace bit is set on a register storing data being used as a base address during the indirect memory acceding, data flow control information, such as the base address stored in the register being used during the indirect acceding is recorded in a queue, and from the queue the information can be stored in a trace buffer. During reconstruction of the operation, the trace buffer, and a copy of the data having an initial state can be examined to analyze the data flows during the real-time operation of the processor.

Apparatus and method for tracing data flows in high-speed computer systems

A system for evaluating the performance of a computer system having a processor that passes through a plurality of processor states during operation and an associated system memory includes an operating unit for receiving a request to monitor specific process states from a user. Firmware causes the processor to enter the desired processor state requested by the user. The hardware identifies the occurrence of the desired processor state. Information relating to the occurrence of the desired process state is accumulated the memory. The accumulated information is read from memory and a report is provided to the user.

Computer system performance evaluation system and method

Fast virtual-address translation and instruction parsing are two hard problems of implementing many modern architectures, including the VAX computer architecture. Meeting the additional constraint of fitting a VAX 11/780-computer-speed implementation into a small number of chips requires a careful coupling of microcode and chip hardware. This paper describes the chip set microarchitecture and the microcode strategies that achieve 11/780 performance. The key features are a VAX instruction prefetch unit occupying one-fourth of a custom NMOS chip, and a memory subsystem occupying another one-fourth of the chip.

Richard L. Sites

Papers

Method and apparatus for tracing unpredictable execution flows in a trace buffer of a high-speed computer system

Apparatus and method for tracing data flows in high-speed computer systems

Computer system performance evaluation system and method

A chip set microarchitecture for a high-performance VAX implementation