ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801826
S. Harbison
{"title":"An architectural alternative to optimizing compilers","authors":"S. Harbison","doi":"10.1145/800050.801826","DOIUrl":"https://doi.org/10.1145/800050.801826","url":null,"abstract":"Programming languages are designed to make programming productive. Computer architectures are designed to make program execution efficient. Although architectures should be designed with programming languages in mind, it may be as inappropriate to make the computer execute the programming language directly it is to make the programmer use machine language. It is the compiler's job to match the programming language and the computer architectures, and therefore making compiler's efficient and easy to write are important design goals of a complete hardware/software system. This paper summerizes research completed in 1980 [5] on a computer architecture, TM, that takes over some of the more burdensome tasks of optimizing compilers for high-level-languages (HLL's), performing these tasks dynamically during the execution of the object program. This is a different approach to making compilers efficient than is commonly taken; more common approaches include devising more efficient optimization algorithms[I], being clever about when to do optimizations [4], and building the compilers semiautomatically [6].","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115424362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801840
Gene McDaniel
{"title":"An analysis of a mesa instruction set using dynamic instruction frequencies","authors":"Gene McDaniel","doi":"10.1145/800050.801840","DOIUrl":"https://doi.org/10.1145/800050.801840","url":null,"abstract":"The Mesa architecture is implemented on a variety of processors, and dynamic instruction frequency data for two programs is used to analyze the architecture in an implementation independent fashion.\u0000 The Mesa compiler allocates variables in an order based upon their static frequency of use, and measurements are provided that show that these static predictions predict run time usage as well. We provide an evaluation of the advantages and costs of Mesa's compact byte encoding, its reliance upon an evaluation stack, and its use of memory.\u0000 The Mesa language has evolved over time in a hardware environment oriented around 16-bit quantities with growing use of and accommodations to 32-bit quantities. The cost of emulating 32-bit data paths on a 16-bit machine is identified for a program that heavily exploits longer values.\u0000 Several potential areas for improving the execution speed of a Mesa processor with special purpose hardware are identified.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128961148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801832
B. R. Rau, C. D. Glaeser, E. M. Greenawalt
{"title":"Architectural support for the efficient generation of code for horizontal architectures","authors":"B. R. Rau, C. D. Glaeser, E. M. Greenawalt","doi":"10.1145/800050.801832","DOIUrl":"https://doi.org/10.1145/800050.801832","url":null,"abstract":"Horizontal architectures, such as the CDC Advanced Flexible Processor [I] and the FPS APi20-B [2}, consist of a number of resources that can operate in parallel, each of which is controlled by a field in the wide instruction word. Such architectures have been developed to perform high speed scientific computations at a modest cost: Figure 1 displays those characteristics of horizontal architectures that are germane to the issues discussed in this paper. The simultaneous requirements of high performance and low cost lead to an architecture consisting of multiple pipelined processing elements (PEs) such as adders and multipliers, a memory (which for scheduling purposes may be viewed as yet another PE with two operations: a READ and a WRITE), and an interconnect which ties them all together. The interconnect allows the result of one operation to be directly routed to another PE as one of the inputs for an operation that is to be performed there. The required memory bandwidth is reduced since temporary values need not be written to and read from the memory. The final aspect of horizontal processors that is of interest is that their program memories emit wide instructions which synchronously specify the actions of the multiple and possibly dissimilar PEs. The program memory is sequenced by a conventional sequencer that assumes sequential flow of control unless a branch is explicitly specified.\u0000 As a consequence of the simplicity of such an architecture, it is inexpensive relative to the potential performance of the multiple pipelined PEs. However, if this potential performance is to be realized, the multiple resources of a horizontal processor must be scheduled effectively. The scheduling task for conventional horizontal processors is quite complex and the construction of highly optimizing compilers for them is a difficult and expensiw3 project. The polycyclic architecture [3- 6] is a horizontal architecture with architectural support for the scheduling task. The cause of the complexity involved in scheduling conventional horizontal processors and the manner in which the polycyclic architecture addresses this issue are outlined in this paper.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132151089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801828
Douglas W. Jones
{"title":"Systematic protection mechanism design","authors":"Douglas W. Jones","doi":"10.1145/800050.801828","DOIUrl":"https://doi.org/10.1145/800050.801828","url":null,"abstract":"This work describes an attempt to systematically design a hardware resource protection mechanism when given the requirements of a particular language as a target. The design process is formalized as a structured walk through the multidimensional computer design space towards a hypothetical class of optimal machines. Each step in this walk involves a change in the distribution of work between the compiler and run-time system but no change in the source language semantics. The starting point for this walk is the result of a semantic analysis of the language to be implemented; typically, this produces a very high level machine where the compiler, if any, is trivial. The walk ends when no changes result in a net improvement. This does not guarantee that the result is even locally optimal, since the changes tried depend on the ingenuity and persistence of the designer.\u0000 This design approach has been used to arrive at a practical, general purpose protection mechanism oriented towards the needs of the Ada language (preliminary version). This architecture was evaluated by comparing it with the PDP-11/45. For the purpose of this comparison, the protection mechanism was incorporated into a partially specified PDP-11 like instruction set. The number of bits making up the processor state and the number of operations involved in address computation were evaluated. On this basis, the result appears to be competitive and worth further investigation.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"368 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122149270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801833
Robert E. McLear, D. M. Scheibelhut, E. Tammaru
{"title":"Guidelines for creating a debuggable processor","authors":"Robert E. McLear, D. M. Scheibelhut, E. Tammaru","doi":"10.1145/800050.801833","DOIUrl":"https://doi.org/10.1145/800050.801833","url":null,"abstract":"Hardware without software is of little use. Systems that ease the task of debugging software reduce cost and speed development. This paper presents guidelines for designing processors that ease debugging for real-time computer systems. Special hardware can aid the debugging process by tracing execution and accesses to memory. Such hardware requires access to signals that may not be readily available. Other, less exotic hardware provides an interface to the programmer and other processors. The hardware and software of the debugging system should not alter the real-time characteristics of the system under test and should be able to operate on a field-grade processor. It is undesirable to require special versions of processor hardware and software for the debugging system.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130180346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801820
J. Hennessy, N. Jouppi, F. Baskett, T. Gross, John T. Gill
{"title":"Hardware/software tradeoffs for increased performance","authors":"J. Hennessy, N. Jouppi, F. Baskett, T. Gross, John T. Gill","doi":"10.1145/800050.801820","DOIUrl":"https://doi.org/10.1145/800050.801820","url":null,"abstract":"Most new computer architectures are concerned with maximizing performance by providing suitable instruction sets for compiled code and providing support for systems functions. We argue that the most effective design methodology must make simultaneous tradeoffs across all three areas: hardware, software support, and systems support. Recent trends lean towards extensive hardware support for both the compiler and operating systems software. However, consideration of all possible design tradeoffs may often lead to less hardware support. Several examples of this approach are presented, including: omission of condition codes, word-addressed machines, and imposing pipeline interlocks in software. The specifics and performance of these approaches are examined with respect to the MIPS processor.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129049602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801843
N. Kamibayashi, H. Ogawana, K. Nagayama, H. Aiso
{"title":"Heart: An operating system nucleus machine implemented by firmware","authors":"N. Kamibayashi, H. Ogawana, K. Nagayama, H. Aiso","doi":"10.1145/800050.801843","DOIUrl":"https://doi.org/10.1145/800050.801843","url":null,"abstract":"This paper discusses the role of microprogramming in operating system design and shows several things: (1) advantages of the efficiency which may be gained from microcoded operating system primitives, (2) selecting the most appropriste primitives for implementation, and (3) an analysis of the tradeoffs among software, firmware, and hardware. The authors propose a practical approach of enhancing computer architecture level, from a view point of functional hierarchy of operating systems. In order to prove the advantages of this approach, we have designed and implemented an experimental abstract machine for an operating system nucleus. This research is an experimental design, and evaluation on its operating system nucleus machine,called HEART. HEART is a set of primitive and universal functions, and works as a nucleus of a multiprogrammed operating system. The research results of our approach are the followings: First, to clarify the properties of operating system nucleus, taking functional hierarchy of operating system into consideration. Second, to show the design of operating system nucleus based on novel concepts. Third, to confirm the possibility of implimenting operating system nucleus machine. Finally, we give a performance evaluation on microcoded HEART and the effectiveness of enhancing computer architecture level based on the properties of operating systems.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"21 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116780119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801844
S. Ahuja, A. Asthana
{"title":"A multi-microprocessor architecture with hardware support for communication and scheduling","authors":"S. Ahuja, A. Asthana","doi":"10.1145/800050.801844","DOIUrl":"https://doi.org/10.1145/800050.801844","url":null,"abstract":"We describe a multiprocessor system that attempts to enhance the system performance by incorporating into its architecture a number of key operating system concepts. In particular:\u0000 — the scheduling and synchronization of concurrent activities are built in at the hardware level,\u0000 — the interprocess communication functions are performed in hardware, and,\u0000 — a coupling between the scheduling and communication functions is provided which allows efficient implementation of parallel systems that is precluded when the scheduling and communication functions are realized in software.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"374 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116540838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801834
M. Wilkes
{"title":"Hardware support for memory protection: Capability implementations","authors":"M. Wilkes","doi":"10.1145/800050.801834","DOIUrl":"https://doi.org/10.1145/800050.801834","url":null,"abstract":"This paper is intended to stimulate discussion on the present state of hardware supported capability systems. Interest in such systems grew up in the mid-1960's and since that time information has been published on several different versions. In the opinion of some observers, the software complexity of these systems outweighs the advantage gained. The paper surveys the situation, and endeavors to set out the general features that a hardware supported capability system should possess. An attempt is made to identify the causes of the complexity and to make recommendations for removing them. The arguments for and against the tagging of capabilities are discussed and attention is drawn to a system of semi-tagging previously proposed by the author.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121437163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
ASPLOS IPub Date : 1982-03-01DOI: 10.1145/800050.801829
K. Reed
{"title":"On a general property of memory mapping tables","authors":"K. Reed","doi":"10.1145/800050.801829","DOIUrl":"https://doi.org/10.1145/800050.801829","url":null,"abstract":"The paper shows that memory mapping tables can be used to implement the display registers used in providing architectural support for block-structured languages such as Algol 60. This allows full lexical level addressing to be implemented on so-called von-Neuman machines.\u0000 The problems of fragmentation of the paged address space are explored, and machines with memory mapping schemes capable of supporting the proposals identified.\u0000 Attention is drawn to the similarity between segmented and paged schemes, and it is suggested that the latter may be used to support the former.","PeriodicalId":140161,"journal":{"name":"ASPLOS I","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1982-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127644228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}