1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> 2<html xmlns="http://www.w3.org/1999/xhtml"> 3<head> 4<meta http-equiv="Content-Type" content="text/html; charset=windows-1252" /> 5<meta http-equiv="Content-Style-Type" content="text/css" /> 6<title>DSP_Scaling</title> 7<link rel="stylesheet" href="../../css/nitro.css" type="text/css" /> 8</head> 9<body> 10 11<h1>DSP_Scaling* <IMG src="../../image/TWL.gif" align="middle"></H1> 12<H2>Syntax</H2> 13 14<pre><code> 15BOOL DSP_ScalingAsyncEx( 16 const void* src, 17 void* dst, 18 u16 img_width, 19 u16 img_height, 20 f32 rx, 21 f32 ry, 22 DSPGraphicsScalingMode mode, 23 u16 x, 24 u16 y, 25 u16 width, 26 u16 height, 27 DSP_GraphicsCallback callback 28 ); 29 30BOOL DSP_Scaling(src, dst, img_width, img_height, rx, ry, mode); 31BOOL DSP_ScalingEx(src, dst, img_width, img_height, rx, ry, mode, x, y, width, height); 32BOOL DSP_ScalingAsync(src, dst, img_width, img_height, rx, ry, mode, callback); 33</code></pre> 34 35<h2>Arguments</h2> 36<table style="width:100%"> 37<tbody> 38<tr> 39<td style="width:13%"><CODE>src</CODE></td> 40<td style="width:87%">Buffer in main memory that holds the data to convert.</td> 41</tr> 42<tr> 43<td style="width:13%"><CODE>dst</CODE></td> 44<td style="width:87%">Buffer in main memory to hold the converted data.<br /><span style='color:Red'>You must allocate a buffer large enough to store the converted data.</span></td> 45</tr> 46<tr> 47<td style="width:13%"><CODE>img_width</CODE></td> 48<td style="width:87%">Width of the image to convert.</td> 49</tr> 50<tr> 51<td style="width:13%"><CODE>img_height</CODE></td> 52<td style="width:87%">Height of the image to convert.</td> 53</tr> 54<tr> 55<td style="width:13%"><CODE>rx</CODE></td> 56<td style="width:87%">Horizontal scaling factor.<br /><span style='color:Red'>Valid between 31 and 0.001. Decimals are truncated to three decimal places.</span></td> 57</tr> 58<tr> 59<td style="width:13%"><CODE>ry</CODE></td> 60<td style="width:87%">Vertical scaling factor.<br /><span style='color:Red'>Valid between 31 and 0.001. Decimals are truncated to three decimal places.</span> 61</td> 62</tr> 63<tr> 64<td style="width:13%"><CODE>mode</CODE></td> 65<td style="width:87%">Interpolation method for scaling up (expanding) or scaling down (shrinking). Specifies the <CODE>DSPGraphicsScalingMode</CODE> (described below).</td> 66</tr> 67<tr> 68<td style="width:13%"><CODE>x</CODE></td> 69<td style="width:87%">X-coordinate of the region to process, with the coordinate system origin (0,0) located at the upper-left of <CODE>src</CODE>.</td> 70</tr> 71<tr> 72<td style="width:13%"><CODE>y</CODE></td> 73<td style="width:87%">Y-coordinate of the region to process, with the coordinate system origin (0,0) located at the upper-left of <CODE>src</CODE>.</td> 74</tr> 75<tr> 76<td style="width:13%"><CODE>width</CODE></td> 77<td style="width:87%">Width of the region to process.</td> 78</tr> 79<tr> 80<td style="width:13%"><CODE>height</CODE></td> 81<td style="width:87%">Height of the region to process.</td> 82</tr> 83<tr> 84<td style="width:13%"><CODE>callback</CODE></td> 85<td style="width:87%">Callback function when processing ends.</td> 86</tr> 87</tbody> 88</table> 89 90<h2>Return Values</h2> 91<p>Returns <CODE>TRUE</CODE> when processing ends normally.<br>Returns <CODE>FALSE</CODE> when conversion did not end normally or the graphics component is already processing something.</p> 92 93<h2>Description</h2> 94<p>Uses the DSP to expand or shrink image data.</p> 95 96<p>The addresses indicated by <CODE>src</CODE> and <CODE>dst</CODE> must be 4-byte aligned.</p> 97<p>This function uses the AHB to transfer data between main memory and the DSP (WRAM-C). Data transfers between main memory and the DSP run at a lower priority than DMA data transfers, so AHB arbitration causes delays if another process uses frequent DMA transfers while scaling is being processed.</p> 98 99<p>The scaling is specified by a floating-point number (<CODE>f32</CODE>) that is converted internally to a fixed-point number and calculated. Due to loss of precision during floating-point calculations, the post-conversion size may differ from the intended value. However, you can use the <a href="DSP_Macro.html"><CODE>DSP_CALC_SCALING_SIZE</CODE></a> macro to calculate the post-conversion size that results from the <CODE>DSP_Scaling*</CODE> functions. We have also provided <a href="DSP_CalcScalingFactor.html"><code>DSP_CalcScalingFactor32</code></a>, a function that takes the post-conversion size as an argument and gets the scaling factors to pass to this function. 100</p> 101 102<p><span style='color:Red'>Note that there are restrictions in this function for values that can be specified as arguments because of design issues of the internal processes. For more information, see <B>Note</B>, below.</span></p> 103 104<p> 105The interpolation method used while processing is defined by <CODE>DSPGraphicsScalingMode</CODE> as follows. 106</p> 107<pre><code> 108typedef DSPWord DSPGraphicsScalingMode; 109#define DSP_GRAPHICS_SCALING_MODE_N_NEIGHBOR (DSPGraphicsScalingMode)0x0001 110#define DSP_GRAPHICS_SCALING_MODE_BILINEAR (DSPGraphicsScalingMode)0x0002 111#define DSP_GRAPHICS_SCALING_MODE_BICUBIC (DSPGraphicsScalingMode)0x0003 112</code></pre> 113<p>Starting from the top, these represent nearest-neighbor, bilinear, and bicubic interpolation. As a basic rule, processing time increases as you move down the list. 114</p> 115 116<p>When the asynchronous versions of this function (<CODE>*Async</CODE>) are run, the callback function registered as an argument is invoked to send a notification that conversion has finished. <span style='color:Red'>Callback functions are invoked by interrupts from the DSP. If interrupts have been prohibited by a function such as <A href="../../os/irq/OS_DisableIrq.html"><CODE>OS_DisableIrq</CODE></A>, completion notifications from the DSP are missed, and callbacks never return. Do not disable DSP interrupts from another process while asynchronous versions of this function are running.</span> 117</p> 118 119<p>With the <CODE>*Ex</CODE> versions of this function, you can specify an arbitrary region of the source image and process only that region. 120</p> 121 122<h2>Note</h2> 123<p>In the synchronous version, the <code>OS_Sleep</code> function waits for the process to complete. Therefore, call the <A href="../../os/thread/OS_InitThread.html"><CODE>OS_InitThread</CODE></A>, <A href="../../os/time/OS_InitTick.html"><CODE>OS_InitTick</CODE></A>, and <A href="../../os/alarm/OS_InitAlarm.html"><CODE>OS_InitAlarm</CODE></A> functions beforehand.</p> 124 125<p>Be careful because when the following conditions are met with this function, the internal DSP process does not end. If the conditions are met <a href="../../os/debug/SDK_ASSERT.html"><CODE>SDK_ASSERTMSG</CODE></a> is called. 126</p> 127 128<ul> 129<li>When the post-process data size is larger than the size of the original data 130<table style="width:80%"> 131<tbody> 132<tr> 133<td style="width:55%"><B>mode</B></td><td><B>Condition</B></td> 134</tr> 135<tr> 136<td><CODE>DSP_GRAPHICS_SCALING_MODE_N_NEIGHBOR<br>DSP_GRAPHICS_SCALING_MODE_BILINEAR</CODE></td> 137<td>(Width of output data) (vertical scaling factor) >=8192</td> 138</tr><tr> 139<td><CODE>DSP_GRAPHICS_SCALING_MODE_BICUBIC</CODE></td> 140<td>(Width of output data) (vertical scaling factor) >=4096</td> 141</tr> 142</tbody> 143</table> 144 145<li>When the post-process data size is smaller than the size of the original data 146<table style="width:80%"> 147<tbody> 148<tr> 149<td style="width:55%"><B>mode</B></td><td><B>Condition</B></td> 150</tr> 151<tr> 152<td><CODE>DSP_GRAPHICS_SCALING_MODE_N_NEIGHBOR<br>DSP_GRAPHICS_SCALING_MODE_BILINEAR</CODE></td> 153<td>(Width of input data) >=8192 </td> 154<tr> 155<td><CODE>DSP_GRAPHICS_SCALING_MODE_BICUBIC</CODE></td> 156<td>(Width of input data) >=4096 </td> 157</tr> 158</tbody> 159</table> 160</ul> 161 162<h2>See Also</h2> 163<p><a href="DSP_LoadGraphics.html"><CODE>DSP_LoadGraphics</CODE></a><BR> <a href="DSP_UnloadGraphics.html"><CODE>DSP_UnloadGraphics</CODE></a><BR> <a href="DSP_ScalingFx.html"><CODE>DSP_ScalingFx*</CODE></a><BR> <a href="../../os/irq/OS_DisableIrq.html"><CODE>OS_DisableIrq</CODE></a><BR> <a href="DSP_Macro.html"><CODE>DSP_CALC_SCALING_SIZE</CODE></a><BR> <a href="DSP_CalcScalingFactor.html"><CODE>DSP_CalcScalingFactor*</CODE></a></p> 164 165<h2>Revision History</h2> 166<p> 1672009/04/03 Added descriptions about the restrictions related to arguments in <B>Note</B>.<br> 2008/10/22 Added a note about changes in the synchronous function.<br> 2008/09/06 Added information on data transfer methods. <br>2008/08/28 Added information to <B>Description</B> about a function that calculates factors. Also added links and descriptions of the <CODE>fx32</CODE> version. <br>2008/08/18 Described how prohibiting DSP interrupts affect asynchronous versions. <br>2008/05/22 Initial version. 168</p> 169<hr><p>CONFIDENTIAL</p></body> 170</html> 171
