Search Results

Search found 141 results on 6 pages for 'invert'.

Page 2/6 | < Previous Page | 1 2 3 4 5 6  | Next Page >

  • How can you invert the colors of a PDF?

    - by legr3c
    I need to invert all the colors of a PDF document (background, text, graphics, and images). I want it persistent in the file so the inverted viewing options, that some viewers offer, won't help. Rasterizing the document and using image manipulation software is also not an option. I read somewhere that this can be done with the Enfocus PitStop plugin for Acrobat. However I didn't see a corresponding command anywhere. Am I missing something? Then I read that the ARTS PDF Crackerjack plugin for Acrobat offers negative printing so I tried that, too. The option is there but it simply doesn't work. I have been searching for very long for a way to do this. It seems like a common enough task but I just can't find out how to do it. Are there maybe any virtual printer drivers or something of the sort that support negative printing? Can anyone help?

    Read the article

  • How to invert scroll wheel in certain applications using AutoHotkey?

    - by endolith
    I want to be able to modify the scrolling/middle click behavior for individual apps on Windows 7, so that the scroll to zoom direction is always consistent across apps. This script makes the middle button act as a hand tool in Adobe Acrobat, for instance: ; Hand tool with middle button in Adobe Reader #IfWinActive ahk_class AdobeAcrobat Mbutton:: #IfWinActive ahk_class AcrobatSDIWindow Mbutton:: Send {Space down}{LButton down} ; Hold down the left mouse button. KeyWait Mbutton ; Wait for the user to release the middle button. Send {LButton up}{Space up} ; Release the left mouse button. return #IfWinActive (It would be great if this could be adapted to allow "throwing" the document, too, like in Android or iPhone interfaces, but I don't know if it's possible to control scrolling that precisely) How do I invert the scroll wheel--zoom direction?

    Read the article

  • Inverting matrix then decomposing gives different quaternion than decomposing then inverting the quat

    - by Fraser
    I'm getting different signs when I convert a matrix to quaternion and invert that, versus when I invert a matrix and then get the quaternion from it: Quaternion a = Quaternion.Invert(getRotation(m)); Quaternion b = getRotation(Matrix.Invert(m)); I would expect a and b to be identical (or inverses of each other). However, it looks like q1 = (x, y, -z, -w) while q2 = (-x, -y, w, z). In other words, the Z and W components have been switched for some reason. Note: getRotation() decomposes the transform matrix and returns just the rotation part of it (I've tried normalizing the result; it does nothing). The matrix m is a complete transform matrix and contains a translation (and possibly a scale) as well as a rotation. I'm using D3DXMatrixDecompose to do the actual decomposition.

    Read the article

  • How to invert arrow image placed before the first cell when tableview cells are swiped beyond first

    - by neha
    Hi all, In my application, I need to add this functionality that there should be an arrow image upside down placed before first cell and some text like "Pull down to refresh" and when user pulls the table beyond this then this arrow gets inverted and text changes to "Release to refresh" and when the user releases his finger, the data is refreshed and rows get added to that point So now this initial arrow and text moves upwards before the first cell again. Can anybody tell me wheather there's any event that does this? Or else which event I need to capture in order to add this functionality? Thanks in advance.

    Read the article

  • Specifying character

    - by danutenshu
    So below I have a code in C++ that is supposed to invert the arguments in a vector, but not the sequence. I have listed my problems as sidenotes in the code below. The invert function is supposed to invert each argument, and then the main function just outputs the inverted words in same order For instance, program("one two three four")=ruof eerth owt eno #include <iostream> #include <string> using namespace std; int invert(string normal) { string inverted; for (int num=normal.size()-1; num>=0; num--) { inverted.append(normal[num]); //I don't know how to get each character //I need another command for append } return **inverted**; <---- } int main(int argc, char* argv[]) { string text; for (int a=1; a<argc; a++) { text.append(invert(argv[a])); //Can't run the invert function text.append(" "); } cout << text << endl; return 0; }

    Read the article

  • How do I rename a mounted Truecrypt volume?

    - by invert
    When I mount the Truecrypt file on my USB drive it shows up as truecrypt1. The volume is FAT, using mtools to rename a volume label involves e2label /dev/sdbx, however truecrypt1 does not map to a physical partition. fdisk -l does not show the volume partition (only the physical USB device), and df -h lists the volume path as /dev/mapper/truecrypt1. Finally, using the Nautilus 'Rename' context action, gives the error: "Sorry, could not rename "truecrypt1" to "towel": Operation not supported by backend". Apparently this can be done in Win, but how can I rename this volume in Ubuntu? As Nicolas said, specifying the mount point names the partition the same. The truecrypt GUI does not remember the mount point I set, so I specify the mount points in a script which I placed in my main menu. #!/bin/bash gksudo truecrypt /media/usbdrive/encryptedfile /media/securedata/

    Read the article

  • Unit testing a controller in ASP.NET MVC 3

    - by Abdullah Al- Mansur
    public Double Invert(Double? id) { return (Double)(id / id); } I have done this for this test but fails please can anyone help with this cos just started with unit testing /* HINT: Remember that you are passing Invert an *integer* so * the value of 1 / input is calculated using integer arithmetic. * */ //Arrange var controller = new UrlParameterController(); int input = 7; Double expected = 0.143d; Double marginOfError = 0.001d; //Act var result = controller.Invert(input); //Assert Assert.AreEqual(expected, result, marginOfError); /* NOTE This time we use a different Assert.AreEqual() method, which * checks whether or not two Double values are within a specified * distance of one another. This is a good way to deal with rounding * errors from floating point arithmetic. Without the marginOfError * parameter the assertion fails. * */

    Read the article

  • `xcalib -i -a` controls only one of two screens - can it control both?

    - by drevicko
    I am using 2 screens, and wish to invert the colors on both of them without using compiz (I'm using gnome shell). I can use xcalib -invert -alter as suggested in this question, but it only inverts one of the screens. Is there a way to specify both of them, or even which of them, using xcalib? Is there another way? With xcalib, you can specify which screen to alter with the -d (-display) or -s (-screen) options, but alas, X seems to be aware of just one screen: when I use the w command I only see one (the usual ":0") ps: this question was originally posted by k0pernikus as a comment here.

    Read the article

  • How to negate current window in gnome shell?

    - by k0pernikus
    I dislike that most websites use a black font on white background for their sites, as it gets too tiresome for me to read. Back in the days of 11.04, using Gnome2 with compiz, there actually was a Negative feature that could negate the content of any window, making the background black and the font white. Much easier on the eyes for me. Yet since 11.10, using gnome shell with mutter, I have no idea if there is something alike out there. Hence my question: How do I negate the currently active window in gnome shell? I am not interested in alternative methods, e.g. user styles. I am aware of their existence but I find it much easier to just invert the screen by the hit of a key shortcut. I also want the solution to be application-agnostic. As I also from time to time would want to invert libre-office or some other glaringly white application.

    Read the article

  • Scheme Function to reverse elements of list of 2-list

    - by sudhirc
    This is an exercise from EOPL. Procedure (invert lst) takes lst which is a list of 2-lists and returns a list with each 2-list reversed. (define invert (lambda (lst) (cond((null? lst ) '()) ((= 2 (rtn-len (car lst))) ( cons(swap-elem (car lst)) (invert (cdr lst)))) ("List is not a 2-List")))) ;; Auxiliry Procedure swap-elements of 2 element list (define swap-elem (lambda (lst) (cons (car (cdr lst)) (car lst)))) ;; returns lengh of the list by calling (define rtn-len (lambda (lst) (calc-len lst 0))) ;; calculate length of the list (define calc-len (lambda (lst n) (if (null? lst) n (calc-len (cdr lst) (+ n 1))))) This seems to work however looks very verbose. Can this be shortened or written in more elegant way ? How I can halt the processing in any of the individual element is not a 2-list? At the moment execution proceed to next member and replacing current member with "List is not a 2-List" if current member is not a 2-list.

    Read the article

  • Triangle Picking Picking Back faces

    - by Tangeleno
    I'm having a bit of trouble with 3D picking, at first I thought my ray was inaccurate but it turns out that the picking is happening on faces facing the camera and faces facing away from the camera which I'm currently culling. Here's my ray creation code, I'm pretty sure the problem isn't here but I've been wrong before. private uint Pick() { Ray cursorRay = CalculateCursorRay(); Vector3? point = Control.Mesh.RayCast(cursorRay); if (point != null) { Tile hitTile = Control.TileMesh.GetTileAtPoint(point); return hitTile == null ? uint.MaxValue : (uint)(hitTile.X + hitTile.Y * Control.Generator.TilesWide); } return uint.MaxValue; } private Ray CalculateCursorRay() { Vector3 nearPoint = Control.Camera.Unproject(new Vector3(Cursor.Position.X, Control.ClientRectangle.Height - Cursor.Position.Y, 0f)); Vector3 farPoint = Control.Camera.Unproject(new Vector3(Cursor.Position.X, Control.ClientRectangle.Height - Cursor.Position.Y, 1f)); Vector3 direction = farPoint - nearPoint; direction.Normalize(); return new Ray(nearPoint, direction); } public Vector3 Camera.Unproject(Vector3 source) { Vector4 result; result.X = (source.X - _control.ClientRectangle.X) * 2 / _control.ClientRectangle.Width - 1; result.Y = (source.Y - _control.ClientRectangle.Y) * 2 / _control.ClientRectangle.Height - 1; result.Z = source.Z - 1; if (_farPlane - 1 == 0) result.Z = 0; else result.Z = result.Z / (_farPlane - 1); result.W = 1f; result = Vector4.Transform(result, Matrix4.Invert(ProjectionMatrix)); result = Vector4.Transform(result, Matrix4.Invert(ViewMatrix)); result = Vector4.Transform(result, Matrix4.Invert(_world)); result = Vector4.Divide(result, result.W); return new Vector3(result.X, result.Y, result.Z); } And my triangle intersection code. Ripped mainly from the XNA picking sample. public float? Intersects(Ray ray) { float? closestHit = Bounds.Intersects(ray); if (closestHit != null && Vertices.Length == 3) { Vector3 e1, e2; Vector3.Subtract(ref Vertices[1].Position, ref Vertices[0].Position, out e1); Vector3.Subtract(ref Vertices[2].Position, ref Vertices[0].Position, out e2); Vector3 directionCrossEdge2; Vector3.Cross(ref ray.Direction, ref e2, out directionCrossEdge2); float determinant; Vector3.Dot(ref e1, ref directionCrossEdge2, out determinant); if (determinant > -float.Epsilon && determinant < float.Epsilon) return null; float inverseDeterminant = 1.0f/determinant; Vector3 distanceVector; Vector3.Subtract(ref ray.Position, ref Vertices[0].Position, out distanceVector); float triangleU; Vector3.Dot(ref distanceVector, ref directionCrossEdge2, out triangleU); triangleU *= inverseDeterminant; if (triangleU < 0 || triangleU > 1) return null; Vector3 distanceCrossEdge1; Vector3.Cross(ref distanceVector, ref e1, out distanceCrossEdge1); float triangleV; Vector3.Dot(ref ray.Direction, ref distanceCrossEdge1, out triangleV); triangleV *= inverseDeterminant; if (triangleV < 0 || triangleU + triangleV > 1) return null; float rayDistance; Vector3.Dot(ref e2, ref distanceCrossEdge1, out rayDistance); rayDistance *= inverseDeterminant; if (rayDistance < 0) return null; return rayDistance; } return closestHit; } I'll admit I don't fully understand all of the math behind the intersection and that is something I'm working on, but my understanding was that if rayDistance was less than 0 the face was facing away from the camera, and shouldn't be counted as a hit. So my question is, is there an issue with my intersection or ray creation code, or is there another check I need to perform to tell if the face is facing away from the camera, and if so any hints on what that check might contain would be appreciated.

    Read the article

  • OpenNETCF Signature control question

    - by Vaccano
    I am using the Signature control in OpenNETCF. It works great for most everything I need. However, I need a way invert the signature and load it back in. It has a call to get the "bytes" for the signature (GetSignatureEx()). It returns a byte[] of the signature. This signature can then be loaded back in with LoadSignatureEx(). I can't seem to figure out the system for these bytes. I thought they may be coordinates, but it does not seem so now. If anyone out there knows a way to invert the signature and load it back in, I would be grateful to hear it.

    Read the article

  • byte[] operations in Java

    - by kape123
    Let's say I have array of bytes: byte[] arr = new byte[] { 0, 1, 2, 3, 4 }; Does platform has functions that I can use to play with this array - for example, how to invert it (get 4,3,2,1,0)? Or, how to invert part of it (2,1,0,3,4)? Get part of array (0,1,2,3)? I know I can manually write functions but I am curious if I'm missing useful util functions in platform that I should know about (and couldn't find any useful guide using google). Thanks!

    Read the article

  • Cant free memory.

    - by atch
    In code: int a[3][4] = {1,2,3,4, 5,6,7,8, 9,10,11,12}; template<class T, int row, int col> void invert(T a[row][col]) { T* columns = new T[col]; T* const free_me = columns; for (int i = 0; i < col; ++i) { for (int j = 0; j < row; ++j) { *columns = a[j][i]; ++columns;//SOMETIMES VALUE IS 0 } } delete[] free_me;//I'M GETTING ERROR OF HEAP ABUSE IN THIS LINE } int main(int argc, char* argv[]) { invert<int,3,4>(a); } I've observed that while iterating, value of variable columns equals zero and I think thats the problem. Thanks for your help.

    Read the article

  • O&rsquo;Reilly Deal of the Day 10/June/2014 - AngularJS Directives

    - by TATWORTH
    Originally posted on: http://geekswithblogs.net/TATWORTH/archive/2014/06/10/orsquoreilly-deal-of-the-day-10june2014---angularjs-directives.aspxToday’s half-price E-Book offer from O’Reilly at http://shop.oreilly.com/product/9781783280339.do is AngularJS Directives. “AngularJS, propelled by Google, is quickly becoming one of the most popular JavaScript MVC frameworks available, working to invert the development paradigm and bring data-driven modularity to the web frontend. Directives serve as the core building blocks in AngularJS and enable you to create reusable models that mold around your data structures and breathe new life into the intersection of HTML and JavaScript.”

    Read the article

  • Besxt Text-to-Speech Solution for my Website

    - by Tim Marshall
    I'm working on the 'Ease of Access' section of my website with the options to increase the font-size displayed on pages to a minimum, invert colours and whatnot. I wish to implement a plugin which, if enabled by the user, to read content on my website. Presumably my best option is a website plugin, however there might be some programming I've not come across which allows the likes of PHP to read content. I'm not entirely sure how this all works. Best Regards, Tim

    Read the article

  • Ray Picking Problems

    - by A Name I Haven't Decided On
    I've read so many answers on here about how to do Ray Picking, that I thought I had the idea of it down. But when I try to implement it in my game, I get garbage. I'm working with LWJGL. Here's the code: public static Ray getPick(int mouseX, int mouseY){ glPushMatrix(); //Setting up the Mouse Clip Vector4f mouseClip = new Vector4f((float)mouseX * 2 / 960f - 1, 1 - (float)mouseY * 2 / 640f ,0 ,1); //Loading Matrices FloatBuffer modMatrix = BufferUtils.createFloatBuffer(16); FloatBuffer projMatrix = BufferUtils.createFloatBuffer(16); glGetFloat(GL_MODELVIEW_MATRIX, modMatrix); glGetFloat(GL_PROJECTION_MATRIX, projMatrix); //Assigning Matrices Matrix4f proj = new Matrix4f(); Matrix4f model = new Matrix4f(); model.load(modMatrix); proj.load(projMatrix); //Multiplying the Projection Matrix by the Model View Matrix Matrix4f tempView = new Matrix4f(); Matrix4f.mul(proj, model, tempView); tempView.invert(); //Getting the Camera Position in World Space. The 4th Column of the Model View Matrix. model.invert(); Point cameraPos = new Point(model.m30, model.m31, model.m32); //Theoretically getting the vector the Picking Ray goes Vector4f rayVector = new Vector4f(); Matrix4f.transform(tempView, mouseClip, rayVector); rayVector.translate((float)-cameraPos.getX(),(float) -cameraPos.getY(),(float) -cameraPos.getZ(), 0f); rayVector.normalise(); glPopMatrix(); //This Basically Spits out a value that changes as the Camera moves. //When the Mouse moves, the values change around 0.001 points from screen edge to edge. System.out.format("Vector: %f %f %f%n", rayVector.x, rayVector.y, rayVector.z); //return new Ray(cameraPos, rayVector); return null; } I don't really know why this isn't working. I was hoping some more experienced eyes might be able to help me out. I can get the camera position like a champ, it's the vector the rays going in that I can't seem to get right. Thanks.

    Read the article

  • Per-pixel collision detection - why does XNA transform matrix return NaN when adding scaling?

    - by JasperS
    I looked at the TransformCollision sample on MSDN and added the Matrix.CreateTranslation part to a property in my collision detection code but I wanted to add scaling. The code works fine when I leave scaling commented out but when I add it and then do a Matrix.Invert() on the created translation matrix the result is NaN ({NaN,NaN,NaN},{NaN,NaN,NaN},...) Can anyone tell me why this is happening please? Here's the code from the sample: // Build the block's transform Matrix blockTransform = Matrix.CreateTranslation(new Vector3(-blockOrigin, 0.0f)) * // Matrix.CreateScale(block.Scale) * would go here Matrix.CreateRotationZ(blocks[i].Rotation) * Matrix.CreateTranslation(new Vector3(blocks[i].Position, 0.0f)); public static bool IntersectPixels( Matrix transformA, int widthA, int heightA, Color[] dataA, Matrix transformB, int widthB, int heightB, Color[] dataB) { // Calculate a matrix which transforms from A's local space into // world space and then into B's local space Matrix transformAToB = transformA * Matrix.Invert(transformB); // When a point moves in A's local space, it moves in B's local space with a // fixed direction and distance proportional to the movement in A. // This algorithm steps through A one pixel at a time along A's X and Y axes // Calculate the analogous steps in B: Vector2 stepX = Vector2.TransformNormal(Vector2.UnitX, transformAToB); Vector2 stepY = Vector2.TransformNormal(Vector2.UnitY, transformAToB); // Calculate the top left corner of A in B's local space // This variable will be reused to keep track of the start of each row Vector2 yPosInB = Vector2.Transform(Vector2.Zero, transformAToB); // For each row of pixels in A for (int yA = 0; yA < heightA; yA++) { // Start at the beginning of the row Vector2 posInB = yPosInB; // For each pixel in this row for (int xA = 0; xA < widthA; xA++) { // Round to the nearest pixel int xB = (int)Math.Round(posInB.X); int yB = (int)Math.Round(posInB.Y); // If the pixel lies within the bounds of B if (0 <= xB && xB < widthB && 0 <= yB && yB < heightB) { // Get the colors of the overlapping pixels Color colorA = dataA[xA + yA * widthA]; Color colorB = dataB[xB + yB * widthB]; // If both pixels are not completely transparent, if (colorA.A != 0 && colorB.A != 0) { // then an intersection has been found return true; } } // Move to the next pixel in the row posInB += stepX; } // Move to the next row yPosInB += stepY; } // No intersection found return false; }

    Read the article

  • Best Text-to-Speech Solution for my Website [on hold]

    - by Tim Marshall
    I'm working on the 'Ease of Access' section of my website with the options to increase the font-size displayed on pages to a minimum, invert colours and whatnot. I wish to implement a plugin which, if enabled by the user, to read content on my website. Presumably my best option is a website plugin, however there might be some programming I've not come across which allows the likes of PHP to read content. I'm not entirely sure how this all works.

    Read the article

  • How John Got 15x Improvement Without Really Trying

    - by rchrd
    The following article was published on a Sun Microsystems website a number of years ago by John Feo. It is still useful and worth preserving. So I'm republishing it here.  How I Got 15x Improvement Without Really Trying John Feo, Sun Microsystems Taking ten "personal" program codes used in scientific and engineering research, the author was able to get from 2 to 15 times performance improvement easily by applying some simple general optimization techniques. Introduction Scientific research based on computer simulation depends on the simulation for advancement. The research can advance only as fast as the computational codes can execute. The codes' efficiency determines both the rate and quality of results. In the same amount of time, a faster program can generate more results and can carry out a more detailed simulation of physical phenomena than a slower program. Highly optimized programs help science advance quickly and insure that monies supporting scientific research are used as effectively as possible. Scientific computer codes divide into three broad categories: ISV, community, and personal. ISV codes are large, mature production codes developed and sold commercially. The codes improve slowly over time both in methods and capabilities, and they are well tuned for most vendor platforms. Since the codes are mature and complex, there are few opportunities to improve their performance solely through code optimization. Improvements of 10% to 15% are typical. Examples of ISV codes are DYNA3D, Gaussian, and Nastran. Community codes are non-commercial production codes used by a particular research field. Generally, they are developed and distributed by a single academic or research institution with assistance from the community. Most users just run the codes, but some develop new methods and extensions that feed back into the general release. The codes are available on most vendor platforms. Since these codes are younger than ISV codes, there are more opportunities to optimize the source code. Improvements of 50% are not unusual. Examples of community codes are AMBER, CHARM, BLAST, and FASTA. Personal codes are those written by single users or small research groups for their own use. These codes are not distributed, but may be passed from professor-to-student or student-to-student over several years. They form the primordial ocean of applications from which community and ISV codes emerge. Government research grants pay for the development of most personal codes. This paper reports on the nature and performance of this class of codes. Over the last year, I have looked at over two dozen personal codes from more than a dozen research institutions. The codes cover a variety of scientific fields, including astronomy, atmospheric sciences, bioinformatics, biology, chemistry, geology, and physics. The sources range from a few hundred lines to more than ten thousand lines, and are written in Fortran, Fortran 90, C, and C++. For the most part, the codes are modular, documented, and written in a clear, straightforward manner. They do not use complex language features, advanced data structures, programming tricks, or libraries. I had little trouble understanding what the codes did or how data structures were used. Most came with a makefile. Surprisingly, only one of the applications is parallel. All developers have access to parallel machines, so availability is not an issue. Several tried to parallelize their applications, but stopped after encountering difficulties. Lack of education and a perception that parallelism is difficult prevented most from trying. I parallelized several of the codes using OpenMP, and did not judge any of the codes as difficult to parallelize. Even more surprising than the lack of parallelism is the inefficiency of the codes. I was able to get large improvements in performance in a matter of a few days applying simple optimization techniques. Table 1 lists ten representative codes [names and affiliation are omitted to preserve anonymity]. Improvements on one processor range from 2x to 15.5x with a simple average of 4.75x. I did not use sophisticated performance tools or drill deep into the program's execution character as one would do when tuning ISV or community codes. Using only a profiler and source line timers, I identified inefficient sections of code and improved their performance by inspection. The changes were at a high level. I am sure there is another factor of 2 or 3 in each code, and more if the codes are parallelized. The study’s results show that personal scientific codes are running many times slower than they should and that the problem is pervasive. Computational scientists are not sloppy programmers; however, few are trained in the art of computer programming or code optimization. I found that most have a working knowledge of some programming language and standard software engineering practices; but they do not know, or think about, how to make their programs run faster. They simply do not know the standard techniques used to make codes run faster. In fact, they do not even perceive that such techniques exist. The case studies described in this paper show that applying simple, well known techniques can significantly increase the performance of personal codes. It is important that the scientific community and the Government agencies that support scientific research find ways to better educate academic scientific programmers. The inefficiency of their codes is so bad that it is retarding both the quality and progress of scientific research. # cacheperformance redundantoperations loopstructures performanceimprovement 1 x x 15.5 2 x 2.8 3 x x 2.5 4 x 2.1 5 x x 2.0 6 x 5.0 7 x 5.8 8 x 6.3 9 2.2 10 x x 3.3 Table 1 — Area of improvement and performance gains of 10 codes The remainder of the paper is organized as follows: sections 2, 3, and 4 discuss the three most common sources of inefficiencies in the codes studied. These are cache performance, redundant operations, and loop structures. Each section includes several examples. The last section summaries the work and suggests a possible solution to the issues raised. Optimizing cache performance Commodity microprocessor systems use caches to increase memory bandwidth and reduce memory latencies. Typical latencies from processor to L1, L2, local, and remote memory are 3, 10, 50, and 200 cycles, respectively. Moreover, bandwidth falls off dramatically as memory distances increase. Programs that do not use cache effectively run many times slower than programs that do. When optimizing for cache, the biggest performance gains are achieved by accessing data in cache order and reusing data to amortize the overhead of cache misses. Secondary considerations are prefetching, associativity, and replacement; however, the understanding and analysis required to optimize for the latter are probably beyond the capabilities of the non-expert. Much can be gained simply by accessing data in the correct order and maximizing data reuse. 6 out of the 10 codes studied here benefited from such high level optimizations. Array Accesses The most important cache optimization is the most basic: accessing Fortran array elements in column order and C array elements in row order. Four of the ten codes—1, 2, 4, and 10—got it wrong. Compilers will restructure nested loops to optimize cache performance, but may not do so if the loop structure is too complex, or the loop body includes conditionals, complex addressing, or function calls. In code 1, the compiler failed to invert a key loop because of complex addressing do I = 0, 1010, delta_x IM = I - delta_x IP = I + delta_x do J = 5, 995, delta_x JM = J - delta_x JP = J + delta_x T1 = CA1(IP, J) + CA1(I, JP) T2 = CA1(IM, J) + CA1(I, JM) S1 = T1 + T2 - 4 * CA1(I, J) CA(I, J) = CA1(I, J) + D * S1 end do end do In code 2, the culprit is conditionals do I = 1, N do J = 1, N If (IFLAG(I,J) .EQ. 0) then T1 = Value(I, J-1) T2 = Value(I-1, J) T3 = Value(I, J) T4 = Value(I+1, J) T5 = Value(I, J+1) Value(I,J) = 0.25 * (T1 + T2 + T5 + T4) Delta = ABS(T3 - Value(I,J)) If (Delta .GT. MaxDelta) MaxDelta = Delta endif enddo enddo I fixed both programs by inverting the loops by hand. Code 10 has three-dimensional arrays and triply nested loops. The structure of the most computationally intensive loops is too complex to invert automatically or by hand. The only practical solution is to transpose the arrays so that the dimension accessed by the innermost loop is in cache order. The arrays can be transposed at construction or prior to entering a computationally intensive section of code. The former requires all array references to be modified, while the latter is cost effective only if the cost of the transpose is amortized over many accesses. I used the second approach to optimize code 10. Code 5 has four-dimensional arrays and loops are nested four deep. For all of the reasons cited above the compiler is not able to restructure three key loops. Assume C arrays and let the four dimensions of the arrays be i, j, k, and l. In the original code, the index structure of the three loops is L1: for i L2: for i L3: for i for l for l for j for k for j for k for j for k for l So only L3 accesses array elements in cache order. L1 is a very complex loop—much too complex to invert. I brought the loop into cache alignment by transposing the second and fourth dimensions of the arrays. Since the code uses a macro to compute all array indexes, I effected the transpose at construction and changed the macro appropriately. The dimensions of the new arrays are now: i, l, k, and j. L3 is a simple loop and easily inverted. L2 has a loop-carried scalar dependence in k. By promoting the scalar name that carries the dependence to an array, I was able to invert the third and fourth subloops aligning the loop with cache. Code 5 is by far the most difficult of the four codes to optimize for array accesses; but the knowledge required to fix the problems is no more than that required for the other codes. I would judge this code at the limits of, but not beyond, the capabilities of appropriately trained computational scientists. Array Strides When a cache miss occurs, a line (64 bytes) rather than just one word is loaded into the cache. If data is accessed stride 1, than the cost of the miss is amortized over 8 words. Any stride other than one reduces the cost savings. Two of the ten codes studied suffered from non-unit strides. The codes represent two important classes of "strided" codes. Code 1 employs a multi-grid algorithm to reduce time to convergence. The grids are every tenth, fifth, second, and unit element. Since time to convergence is inversely proportional to the distance between elements, coarse grids converge quickly providing good starting values for finer grids. The better starting values further reduce the time to convergence. The downside is that grids of every nth element, n > 1, introduce non-unit strides into the computation. In the original code, much of the savings of the multi-grid algorithm were lost due to this problem. I eliminated the problem by compressing (copying) coarse grids into continuous memory, and rewriting the computation as a function of the compressed grid. On convergence, I copied the final values of the compressed grid back to the original grid. The savings gained from unit stride access of the compressed grid more than paid for the cost of copying. Using compressed grids, the loop from code 1 included in the previous section becomes do j = 1, GZ do i = 1, GZ T1 = CA(i+0, j-1) + CA(i-1, j+0) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) S1 = T1 + T4 - 4 * CA1(i+0, j+0) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 enddo enddo where CA and CA1 are compressed arrays of size GZ. Code 7 traverses a list of objects selecting objects for later processing. The labels of the selected objects are stored in an array. The selection step has unit stride, but the processing steps have irregular stride. A fix is to save the parameters of the selected objects in temporary arrays as they are selected, and pass the temporary arrays to the processing functions. The fix is practical if the same parameters are used in selection as in processing, or if processing comprises a series of distinct steps which use overlapping subsets of the parameters. Both conditions are true for code 7, so I achieved significant improvement by copying parameters to temporary arrays during selection. Data reuse In the previous sections, we optimized for spatial locality. It is also important to optimize for temporal locality. Once read, a datum should be used as much as possible before it is forced from cache. Loop fusion and loop unrolling are two techniques that increase temporal locality. Unfortunately, both techniques increase register pressure—as loop bodies become larger, the number of registers required to hold temporary values grows. Once register spilling occurs, any gains evaporate quickly. For multiprocessors with small register sets or small caches, the sweet spot can be very small. In the ten codes presented here, I found no opportunities for loop fusion and only two opportunities for loop unrolling (codes 1 and 3). In code 1, unrolling the outer and inner loop one iteration increases the number of result values computed by the loop body from 1 to 4, do J = 1, GZ-2, 2 do I = 1, GZ-2, 2 T1 = CA1(i+0, j-1) + CA1(i-1, j+0) T2 = CA1(i+1, j-1) + CA1(i+0, j+0) T3 = CA1(i+0, j+0) + CA1(i-1, j+1) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) T5 = CA1(i+2, j+0) + CA1(i+1, j+1) T6 = CA1(i+1, j+1) + CA1(i+0, j+2) T7 = CA1(i+2, j+1) + CA1(i+1, j+2) S1 = T1 + T4 - 4 * CA1(i+0, j+0) S2 = T2 + T5 - 4 * CA1(i+1, j+0) S3 = T3 + T6 - 4 * CA1(i+0, j+1) S4 = T4 + T7 - 4 * CA1(i+1, j+1) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 CA(i+1, j+0) = CA1(i+1, j+0) + DD * S2 CA(i+0, j+1) = CA1(i+0, j+1) + DD * S3 CA(i+1, j+1) = CA1(i+1, j+1) + DD * S4 enddo enddo The loop body executes 12 reads, whereas as the rolled loop shown in the previous section executes 20 reads to compute the same four values. In code 3, two loops are unrolled 8 times and one loop is unrolled 4 times. Here is the before for (k = 0; k < NK[u]; k++) { sum = 0.0; for (y = 0; y < NY; y++) { sum += W[y][u][k] * delta[y]; } backprop[i++]=sum; } and after code for (k = 0; k < KK - 8; k+=8) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (y = 0; y < NY; y++) { sum0 += W[y][0][k+0] * delta[y]; sum1 += W[y][0][k+1] * delta[y]; sum2 += W[y][0][k+2] * delta[y]; sum3 += W[y][0][k+3] * delta[y]; sum4 += W[y][0][k+4] * delta[y]; sum5 += W[y][0][k+5] * delta[y]; sum6 += W[y][0][k+6] * delta[y]; sum7 += W[y][0][k+7] * delta[y]; } backprop[k+0] = sum0; backprop[k+1] = sum1; backprop[k+2] = sum2; backprop[k+3] = sum3; backprop[k+4] = sum4; backprop[k+5] = sum5; backprop[k+6] = sum6; backprop[k+7] = sum7; } for one of the loops unrolled 8 times. Optimizing for temporal locality is the most difficult optimization considered in this paper. The concepts are not difficult, but the sweet spot is small. Identifying where the program can benefit from loop unrolling or loop fusion is not trivial. Moreover, it takes some effort to get it right. Still, educating scientific programmers about temporal locality and teaching them how to optimize for it will pay dividends. Reducing instruction count Execution time is a function of instruction count. Reduce the count and you usually reduce the time. The best solution is to use a more efficient algorithm; that is, an algorithm whose order of complexity is smaller, that converges quicker, or is more accurate. Optimizing source code without changing the algorithm yields smaller, but still significant, gains. This paper considers only the latter because the intent is to study how much better codes can run if written by programmers schooled in basic code optimization techniques. The ten codes studied benefited from three types of "instruction reducing" optimizations. The two most prevalent were hoisting invariant memory and data operations out of inner loops. The third was eliminating unnecessary data copying. The nature of these inefficiencies is language dependent. Memory operations The semantics of C make it difficult for the compiler to determine all the invariant memory operations in a loop. The problem is particularly acute for loops in functions since the compiler may not know the values of the function's parameters at every call site when compiling the function. Most compilers support pragmas to help resolve ambiguities; however, these pragmas are not comprehensive and there is no standard syntax. To guarantee that invariant memory operations are not executed repetitively, the user has little choice but to hoist the operations by hand. The problem is not as severe in Fortran programs because in the absence of equivalence statements, it is a violation of the language's semantics for two names to share memory. Codes 3 and 5 are C programs. In both cases, the compiler did not hoist all invariant memory operations from inner loops. Consider the following loop from code 3 for (y = 0; y < NY; y++) { i = 0; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += delta[y] * I1[i++]; } } } Since dW[y][u] can point to the same memory space as delta for one or more values of y and u, assignment to dW[y][u][k] may change the value of delta[y]. In reality, dW and delta do not overlap in memory, so I rewrote the loop as for (y = 0; y < NY; y++) { i = 0; Dy = delta[y]; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += Dy * I1[i++]; } } } Failure to hoist invariant memory operations may be due to complex address calculations. If the compiler can not determine that the address calculation is invariant, then it can hoist neither the calculation nor the associated memory operations. As noted above, code 5 uses a macro to address four-dimensional arrays #define MAT4D(a,q,i,j,k) (double *)((a)->data + (q)*(a)->strides[0] + (i)*(a)->strides[3] + (j)*(a)->strides[2] + (k)*(a)->strides[1]) The macro is too complex for the compiler to understand and so, it does not identify any subexpressions as loop invariant. The simplest way to eliminate the address calculation from the innermost loop (over i) is to define a0 = MAT4D(a,q,0,j,k) before the loop and then replace all instances of *MAT4D(a,q,i,j,k) in the loop with a0[i] A similar problem appears in code 6, a Fortran program. The key loop in this program is do n1 = 1, nh nx1 = (n1 - 1) / nz + 1 nz1 = n1 - nz * (nx1 - 1) do n2 = 1, nh nx2 = (n2 - 1) / nz + 1 nz2 = n2 - nz * (nx2 - 1) ndx = nx2 - nx1 ndy = nz2 - nz1 gxx = grn(1,ndx,ndy) gyy = grn(2,ndx,ndy) gxy = grn(3,ndx,ndy) balance(n1,1) = balance(n1,1) + (force(n2,1) * gxx + force(n2,2) * gxy) * h1 balance(n1,2) = balance(n1,2) + (force(n2,1) * gxy + force(n2,2) * gyy)*h1 end do end do The programmer has written this loop well—there are no loop invariant operations with respect to n1 and n2. However, the loop resides within an iterative loop over time and the index calculations are independent with respect to time. Trading space for time, I precomputed the index values prior to the entering the time loop and stored the values in two arrays. I then replaced the index calculations with reads of the arrays. Data operations Ways to reduce data operations can appear in many forms. Implementing a more efficient algorithm produces the biggest gains. The closest I came to an algorithm change was in code 4. This code computes the inner product of K-vectors A(i) and B(j), 0 = i < N, 0 = j < M, for most values of i and j. Since the program computes most of the NM possible inner products, it is more efficient to compute all the inner products in one triply-nested loop rather than one at a time when needed. The savings accrue from reading A(i) once for all B(j) vectors and from loop unrolling. for (i = 0; i < N; i+=8) { for (j = 0; j < M; j++) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (k = 0; k < K; k++) { sum0 += A[i+0][k] * B[j][k]; sum1 += A[i+1][k] * B[j][k]; sum2 += A[i+2][k] * B[j][k]; sum3 += A[i+3][k] * B[j][k]; sum4 += A[i+4][k] * B[j][k]; sum5 += A[i+5][k] * B[j][k]; sum6 += A[i+6][k] * B[j][k]; sum7 += A[i+7][k] * B[j][k]; } C[i+0][j] = sum0; C[i+1][j] = sum1; C[i+2][j] = sum2; C[i+3][j] = sum3; C[i+4][j] = sum4; C[i+5][j] = sum5; C[i+6][j] = sum6; C[i+7][j] = sum7; }} This change requires knowledge of a typical run; i.e., that most inner products are computed. The reasons for the change, however, derive from basic optimization concepts. It is the type of change easily made at development time by a knowledgeable programmer. In code 5, we have the data version of the index optimization in code 6. Here a very expensive computation is a function of the loop indices and so cannot be hoisted out of the loop; however, the computation is invariant with respect to an outer iterative loop over time. We can compute its value for each iteration of the computation loop prior to entering the time loop and save the values in an array. The increase in memory required to store the values is small in comparison to the large savings in time. The main loop in Code 8 is doubly nested. The inner loop includes a series of guarded computations; some are a function of the inner loop index but not the outer loop index while others are a function of the outer loop index but not the inner loop index for (j = 0; j < N; j++) { for (i = 0; i < M; i++) { r = i * hrmax; R = A[j]; temp = (PRM[3] == 0.0) ? 1.0 : pow(r, PRM[3]); high = temp * kcoeff * B[j] * PRM[2] * PRM[4]; low = high * PRM[6] * PRM[6] / (1.0 + pow(PRM[4] * PRM[6], 2.0)); kap = (R > PRM[6]) ? high * R * R / (1.0 + pow(PRM[4]*r, 2.0) : low * pow(R/PRM[6], PRM[5]); < rest of loop omitted > }} Note that the value of temp is invariant to j. Thus, we can hoist the computation for temp out of the loop and save its values in an array. for (i = 0; i < M; i++) { r = i * hrmax; TEMP[i] = pow(r, PRM[3]); } [N.B. – the case for PRM[3] = 0 is omitted and will be reintroduced later.] We now hoist out of the inner loop the computations invariant to i. Since the conditional guarding the value of kap is invariant to i, it behooves us to hoist the computation out of the inner loop, thereby executing the guard once rather than M times. The final version of the code is for (j = 0; j < N; j++) { R = rig[j] / 1000.; tmp1 = kcoeff * par[2] * beta[j] * par[4]; tmp2 = 1.0 + (par[4] * par[4] * par[6] * par[6]); tmp3 = 1.0 + (par[4] * par[4] * R * R); tmp4 = par[6] * par[6] / tmp2; tmp5 = R * R / tmp3; tmp6 = pow(R / par[6], par[5]); if ((par[3] == 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp5; } else if ((par[3] == 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp4 * tmp6; } else if ((par[3] != 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp5; } else if ((par[3] != 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp4 * tmp6; } for (i = 0; i < M; i++) { kap = KAP[i]; r = i * hrmax; < rest of loop omitted > } } Maybe not the prettiest piece of code, but certainly much more efficient than the original loop, Copy operations Several programs unnecessarily copy data from one data structure to another. This problem occurs in both Fortran and C programs, although it manifests itself differently in the two languages. Code 1 declares two arrays—one for old values and one for new values. At the end of each iteration, the array of new values is copied to the array of old values to reset the data structures for the next iteration. This problem occurs in Fortran programs not included in this study and in both Fortran 77 and Fortran 90 code. Introducing pointers to the arrays and swapping pointer values is an obvious way to eliminate the copying; but pointers is not a feature that many Fortran programmers know well or are comfortable using. An easy solution not involving pointers is to extend the dimension of the value array by 1 and use the last dimension to differentiate between arrays at different times. For example, if the data space is N x N, declare the array (N, N, 2). Then store the problem’s initial values in (_, _, 2) and define the scalar names new = 2 and old = 1. At the start of each iteration, swap old and new to reset the arrays. The old–new copy problem did not appear in any C program. In programs that had new and old values, the code swapped pointers to reset data structures. Where unnecessary coping did occur is in structure assignment and parameter passing. Structures in C are handled much like scalars. Assignment causes the data space of the right-hand name to be copied to the data space of the left-hand name. Similarly, when a structure is passed to a function, the data space of the actual parameter is copied to the data space of the formal parameter. If the structure is large and the assignment or function call is in an inner loop, then copying costs can grow quite large. While none of the ten programs considered here manifested this problem, it did occur in programs not included in the study. A simple fix is always to refer to structures via pointers. Optimizing loop structures Since scientific programs spend almost all their time in loops, efficient loops are the key to good performance. Conditionals, function calls, little instruction level parallelism, and large numbers of temporary values make it difficult for the compiler to generate tightly packed, highly efficient code. Conditionals and function calls introduce jumps that disrupt code flow. Users should eliminate or isolate conditionls to their own loops as much as possible. Often logical expressions can be substituted for if-then-else statements. For example, code 2 includes the following snippet MaxDelta = 0.0 do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) if (Delta > MaxDelta) MaxDelta = Delta enddo enddo if (MaxDelta .gt. 0.001) goto 200 Since the only use of MaxDelta is to control the jump to 200 and all that matters is whether or not it is greater than 0.001, I made MaxDelta a boolean and rewrote the snippet as MaxDelta = .false. do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) MaxDelta = MaxDelta .or. (Delta .gt. 0.001) enddo enddo if (MaxDelta) goto 200 thereby, eliminating the conditional expression from the inner loop. A microprocessor can execute many instructions per instruction cycle. Typically, it can execute one or more memory, floating point, integer, and jump operations. To be executed simultaneously, the operations must be independent. Thick loops tend to have more instruction level parallelism than thin loops. Moreover, they reduce memory traffice by maximizing data reuse. Loop unrolling and loop fusion are two techniques to increase the size of loop bodies. Several of the codes studied benefitted from loop unrolling, but none benefitted from loop fusion. This observation is not too surpising since it is the general tendency of programmers to write thick loops. As loops become thicker, the number of temporary values grows, increasing register pressure. If registers spill, then memory traffic increases and code flow is disrupted. A thick loop with many temporary values may execute slower than an equivalent series of thin loops. The biggest gain will be achieved if the thick loop can be split into a series of independent loops eliminating the need to write and read temporary arrays. I found such an occasion in code 10 where I split the loop do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do into two disjoint loops do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) end do end do do i = 1, n do j = 1, m C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do Conclusions Over the course of the last year, I have had the opportunity to work with over two dozen academic scientific programmers at leading research universities. Their research interests span a broad range of scientific fields. Except for two programs that relied almost exclusively on library routines (matrix multiply and fast Fourier transform), I was able to improve significantly the single processor performance of all codes. Improvements range from 2x to 15.5x with a simple average of 4.75x. Changes to the source code were at a very high level. I did not use sophisticated techniques or programming tools to discover inefficiencies or effect the changes. Only one code was parallel despite the availability of parallel systems to all developers. Clearly, we have a problem—personal scientific research codes are highly inefficient and not running parallel. The developers are unaware of simple optimization techniques to make programs run faster. They lack education in the art of code optimization and parallel programming. I do not believe we can fix the problem by publishing additional books or training manuals. To date, the developers in questions have not studied the books or manual available, and are unlikely to do so in the future. Short courses are a possible solution, but I believe they are too concentrated to be much use. The general concepts can be taught in a three or four day course, but that is not enough time for students to practice what they learn and acquire the experience to apply and extend the concepts to their codes. Practice is the key to becoming proficient at optimization. I recommend that graduate students be required to take a semester length course in optimization and parallel programming. We would never give someone access to state-of-the-art scientific equipment costing hundreds of thousands of dollars without first requiring them to demonstrate that they know how to use the equipment. Yet the criterion for time on state-of-the-art supercomputers is at most an interesting project. Requestors are never asked to demonstrate that they know how to use the system, or can use the system effectively. A semester course would teach them the required skills. Government agencies that fund academic scientific research pay for most of the computer systems supporting scientific research as well as the development of most personal scientific codes. These agencies should require graduate schools to offer a course in optimization and parallel programming as a requirement for funding. About the Author John Feo received his Ph.D. in Computer Science from The University of Texas at Austin in 1986. After graduate school, Dr. Feo worked at Lawrence Livermore National Laboratory where he was the Group Leader of the Computer Research Group and principal investigator of the Sisal Language Project. In 1997, Dr. Feo joined Tera Computer Company where he was project manager for the MTA, and oversaw the programming and evaluation of the MTA at the San Diego Supercomputer Center. In 2000, Dr. Feo joined Sun Microsystems as an HPC application specialist. He works with university research groups to optimize and parallelize scientific codes. Dr. Feo has published over two dozen research articles in the areas of parallel parallel programming, parallel programming languages, and application performance.

    Read the article

< Previous Page | 1 2 3 4 5 6  | Next Page >