Abstract

Many existing or potential programming tools require the program to be completely recompiled with a special compiler option. This is usually inconvenient for the program developer, and may reduce the usefulness of the tool or the frequency with which the tool is employed. It may also require the maintenance of different versions of standard libraries, each compiled with the appropriate options for a different tool. The difference between modules compiled with and without the special option is often simple and regular. If so, we can effect this difference by modifying the normally-compiled object code at link time, instead of recompiling. This reduces the overhead of using the tool by an order of magnitude, making it much more convenient. i 1. Introduction Recompiling an entire multi-module program from scratch is usually so expensive o that one does it only reluctantly. In spite of this, many useful tools for program ptimization or performance analysis require the recompilation of...

Citations

339	Limits of Instruction-Level Parallelism - Wall - 1991
285	Gprof: A call graph execution profiler - Graham, Kessler, et al. - 1982
192	Available Instruction-Level Parallelism for Superscalar and Superpipelined Machines - Jouppi, Wall - 1989
169	Global register allocation at link time - Wall - 1986
108	ATUM: A New Technique for Capturing Address Traces Using Microcode - Agarwal, Sites, et al. - 1986
94	Tradeoffs in two-level on-chip caching - Jouppi, Wilton - 1994
91	The case for the reduced instruction set computer - Patterson, Ditzel - 1980
90	Link-Time Optimization of Address Calculation on a 64-bit Architecture - Srivastava, Wall - 1994
74	Long Address Traces from RISC Machines: Generation and Analysis - Borg, Kessler, et al. - 1989
68	Language as the Machine Description - Wall, Powell - 1987
60	Simple and flexible datagram access controls for UNIX-based gateways - Mogul - 1989
53	Experience with a Software-defined Machine Architecture - Wall - 1991
47	Visa Protocols for Controlling Inter-Organizational Datagram Flow: Extended Description." "Cache Coherence in Distributed Systems - Anand - 1987
47	WRL Research Report 88/5 - Bartlett - 1987
47	WRL Research Report 89/2 - Hamburgen - 1987
44	Transparent Controls for Interactive Graphics - Bartlett - 1992
43	WRL Research Report 89/1 - Turrini - 1987
41	Michael ‘‘Measured Capacity of an Ethernet - Mogul, Rashid - 1987
38	A 20-MIPS Sustained 32-bit CMOS Microproces"1990 DECWRL/Livermore Magic Release." sor with High Ratio of Sustained to Peak PerforRobert - Jouppi, Tang
38	Pool Boiling Enhancement Techniques for Water at Low Pressure - McGillis, Fitch, et al. - 1990
36	Incremental, Generational Mostly-Copying Garbage Collection in Uncooperative Environ"A Smart Frame Buffer." ments - Yip - 1992
34	Windows vs. Register Allocation - Wall, “Register - 1988
32	Global Register Allocation at Link-time - Wall - 1986
32	Optimal Finned Heat Sinks." William R. Hamburgen. "Compacting Garbage Collection with Ambiguous WRL - Bartlett - 1986
32	Michael "Measured Capacity of an Ethernet - Mogul, Rashid - 1987
32	Fragmentation Considered Harmful." WRL Research Report 88/4 - Kent, Mogul - 1988
32	Pool Boiling on Small Heat Dissipating Elements in Water at Subatmospheric Pressure - McGillis, Fitch, et al. - 1992
32	Architectural and Organizational trade-offs in the design of the MultiTitan CPU - Jouppi - 1988
32	Experience: Using an Intermediate WRL Research Report 88/2 - Mahler - 1988
26	Fluoroelastomer Pressure Pad Design for Microelectronic Applications.’’ Alberto - Makino, Fitch - 1993
26	WRL Technical Notes "TCP/IP PrintServer: Print Server Protocol." "Predicting Program Behavior Using Real or EsBrian - Reid, Kent - 1988
26	Report 90/4 - Research - 1990
24	Compacting Garbage Collection with Ambiguous WRL - Hamburgen - 1986
24	Fragmentation Considered Harmful.’’ WRL Research Report 88/4 - Kent, Mogul - 1988
23	Register Allocation and Exhaustive Peephole Optimization - Davidson, Fraser - 1984
21	Efficient Generation of Test Patterns Using Boolean Satisfiablilty." "Available Instruction-Level Parallelism for SuperTracy Larrabee. scalar and - Jouppi, Wall - 1990
19	tion at Link-Time." WRL Research Report 91/7 - Wall - 1991
17	Boolean Matching for Full-Custom ECL Gates - Mayo, Touati - 1993
15	Reports "Titan System Manual." "MultiTitan: Four Architecture Papers - Research - 1986
15	Simple and Flexible Datagram Access Controls for "Link-Time Code Modification." Unix-based Gateways - Wall - 1987
14	WRL Technical Notes ‘‘TCP/IP PrintServer: Print Server Protocol.’’ ‘‘Boiling Binary Mixtures at Subatmospheric PresBrian - Reid, Kent - 1988
13	Electrical Evaluation Of The BIPS-0 Package" "MTOOL: A Method For Detecting Memory Bot - Boyle - 1990
13	tlenecks." WRL Technical Note TN-29 - Hennessy - 1992
12	Noise Issues in the ECL Circuit Family." "Spritely NFS: Implementation and Performance of Jeffrey Y.F. Tang and - Yang - 1990
11	Two Papers on Test Pattern Generation." Tracy Larrabee. "A Unified Vector/Scalar Floating-Point ArchitecWRL Research Report 90/3 - Jouppi, Bertoni, et al. - 1990
11	Architectural and Organizational Tradeoffs in the "Efficient Use of Workstations for Passive Monitor- Design of the MultiTitan CPU." ing of Local Area Networks - Jouppi - 1990
11	Integration and Packaging Plateaus of Processor "A One-Dimensional Thermal Model for the VAX Performance." 9000 Multi Chip Units - Jouppi - 1990
11	TCP/IP PrintServer: Server Architecture and Implementation. " "Cache Replacement with Dynamic Exclusion - McFarling - 1990
11	Mostly-Copying Garbage Collection Picks Up Generations and C++." "TurboChannel Versatec Adapter - Boggs - 1992
10	WRL Research Report 89/11 - McGillis, Fitch, et al. - 1990