http:^^www.cs.cornell.edu^info^courses^spring-95^cs314^hw^proj1.html
来自「This data set contains WWW-pages collect」· HTML 代码 · 共 269 行
HTML
269 行
MIME-Version: 1.0
Server: CERN/3.0
Date: Monday, 16-Dec-96 23:18:38 GMT
Content-Type: text/html
Content-Length: 8016
Last-Modified: Monday, 13-Mar-95 14:33:34 GMT
<html><head><title> Project 1: Martix Products </title></head><body>[<!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><a href="http://www.cs.cornell.edu/Info/Courses/Spring-95/CS314/home.html">Back to CS314 Home Page</a>]<h1> Project 1: Martix Products </h1><h3>Date Assigned: February 14, 1994 <br>Date Due: February 28, 1994</h3><h2> Introduction</h2>This project involves writing a small assembly language program for multiplyingmatrices. Your program will gets its input from and send its output to theBasePak runtime window.<H2>Specs</H2><H3> Overview</H3>Compute the matrix product <I>AB</I>---technically, <I>AB mod 2^16</I> ---where <I>A</I> and <I>B</I> are matrices of 16-bit unsigned integers. Theminimum matrix size will be <I>1x1</I>, and the maximum matrix size willbe <I>16x16</I>.<h3>Input parameters </h3><pre>-------------------------------------------------------------------------------------------------------------A m x n matrix of 16-bit unsigned integers A is the first matrix in the product. B n x p matrix of 16-bit unsigned integers B is the second matrix in the product. m unsigned 8-bit integer m is the number of rows in A n unsigned 8-bit integer n is the number of columns in A and the number of rows in B. p unsigned 8-bit integer p is the number of columns in B m, n, and p satisfy 1 <= m,n,p <= 16-------------------------------------------------------------------------------------------------------------</pre><h3>Output parameters</h3><pre>-------------------------------------------------------------------------C m x p matrix of 16-bit unsigned integers C = AB mod 2^16-------------------------------------------------------------------------</pre><H3>Arithmetic</H3>Do all arithmetic over integers mod 2^16, i.e., use unsigned 16-bit integermultiplication and addition and take only the low-order word of multiplicationresults.<H3>User interface</H3>The "user interface" of your program will be minimal---this is anassembly language program, after all. Your program should prompt the user forthe parameters <I>m</I>, <I>n</I>, and <I>p</I>, successively, then allow theuser to enter the values of the elements of the matrix, <I>A</I>, then thevalues of the matrix <I>B</I>. After these values have been entered, yourprogram should output the matrices <I>A</I> and <I>B</I> and then output theresult matrix <I>C</I>. (All numerical input and output should be in base 10.)<p>Matrices should be output in standard order, i.e., a row of elements in thematrix should correspond to a row of numbers on the screen. For example, if<I>A</I> were a <I>2 x 2</I> matrix with elements a11 = 1, a12 = 2, a21 = 3,and a22 = 4, then <I>A</I> would be output as: <pre> 1 2 3 4</pre>Matrices should be input by prompting the user for elements, one at a time, inthe same order in which they would be output. The text of the prompt shouldinclude the matrix name and the indices of the element currently being promptedfor. For example, the matrix <I>A</I> given above would be entered via thefollowing exchange (underlined text indicates the program's output,non-underlined text indicates the user's input):<PRE> <U>A[1][1]:</U> 1 <U>A[1][2]:</U> 2 <U>A[2][1]:</U> 3 <U>A[2][2]:</U> 4</PRE><H2>Sample Run</H2>A run of your program should look something like the following:<PRE><U>m</U>: 2<U>n</U>: 3<U>p</U>: 4<U>A[1][1]:</U> 0<U>A[1][2]:</U> 1<U>A[1][3]:</U> 2<U>A[2][1]:</U> 3<U>A[2][2]:</U> 4<p><U>A[2][3]:</U> 5<U>B[1][1]:</U> 10<U>B[1][2]:</U> 11<U>B[1][3]:</U> 12<U>B[1][4]:</U> 13<U>B[2][1]:</U> 14<U>B[2][2]:</U> 15<U>B[2][3]:</U> 16<U>B[2][4]:</U> 17<U>B[3][1]:</U> 18<U>B[3][2]:</U> 65510<U>B[3][3]:</U> 20<U>B[3][4]:</U> 21A: 0 1 23 4 5B:10 11 12 1314 15 16 1718 65510 20 21Result (A*B):50 65499 56 59176 65499 200 212</pre>which corresponds to:<br><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><img src="http://www.cs.cornell.edu/Info/Courses/Spring-95/CS314/hw/proj1.gif"><H2>Documentation</H2>The usual documentation guidelines (always explain how you're using registers,use symbolic constants whenever possible, state the input and output parametersof all subroutines, etc.) apply. Your program should have at least 3subroutines, one for inputting matrices, one for printing matrices, one formultiplying the matrices. Each subroutine should begin with a header that lookslike this:<PRE> ;------------------------------------------------------------------------- ; ; <Name of function> ; ; ; Synopsis ; <A brief description of what the function does, who calls it, ; why, and any exceptional conditions.> ; ; HLL description: ; ; <Put in high level language (e.g., C, pascal, pseudo-code) ; description of routine here.> ; ; Register usage: ; <Map out what registers are used for> ; ; Input Conditions: ; <What should the format of the registers and stack be on entry ; to the routine?> ; ; Output Conditions: ; <What will the format of the registers and stack be when the ; routine returns?> ; ; Stack frame: ; <What does the stack frame look like? For example:> ; ; Saved Registers ; A6 --> Old A6 0(A6) ; Return Address 4(A6) ; Param 1 8(A6) ; Param 2 12(A6) ; ;-------------------------------------------------------------------------</PRE>The subroutine code should immediately follow the subroutine header. Try tocomment each line and use block comments and blank spaces to separate logicalsections of the subroutine. For example, here's the code of a routine thatswaps the two registers D0 and D1 if D1 < D0.<PRE> SWAP: ; ; Prolog: Save registers, set up stack frame ; LINK A6,#0 ; set frame pointer -- no locals MOVEM.L D2,-(SP) ; Save registers we'll use. ; ; Compare D0 and D1 to see if we need to do anything. ; If not, then just jump to epilog and leave. ; CMP D1,D0 BLT DONE ; ; If we get here, D1 < D0. ; Swap the contents of D0 and D1, using D2 as a scratch register ; MOVE.L D0,D2 ; tmp:= D0 MOVE.L D1,D0 ; D0:= D1 MOVE.L D2,D0 ; D0:= tmp ; ; Epilog: Restore registers, clean up stack frame, return. ; DONE: MOVEM.L (SP)+,D2 ; Restore registers, etc. UNLK A6 RTS</PRE>Try to give a high level description in the function header, a more detaileddescription in the block comments, and a blow-by-blow in the individual linecomments. Don't just repeat what the code says, if possible.<p>In addition, for this project, be sure you document the following aspects ofyour program:<UL><LI>your method of laying out matrices in memory.<LI>your indexing conventions.</UL><H2> Misc. </H2>Your program should check that the values of <I>m</I>, <I>n</I>, and <I>p</I>supplied by the user fall within the allowable range.<p>You'll probably need to adapt the 32-bit <I>signed</I> decimal i/o routines inthe <B>BasePak</B> library for use as 16-bit <I>unsigned</I> routines. (Forthis project, simply taking the low-order word of the return from decin_longwould be acceptable for inputting 16-bit unsigned integers (though that wouldn'tbe a such a great idea in a <I>real</I> program), and using decout_long willwork for outputting 16-bit unsigned integers as long as you remember to clearthe high word of d0.)<H2>Submitting</H2>Hand in the following to the consulting office:<UL><LI>a hardcopy of your program.<LI>a hardcopy of a sample run of your program.<LI>an electronic copy of your program (on diskette).</UL>[<!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><a href="http://www.cs.cornell.edu/Info/Courses/Spring-95/CS314/home.html">Back to CS314 Home Page</a>]</body></html>⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?