1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
|
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>Mu - 302stack_allocate.subx</title>
<meta name="Generator" content="Vim/8.2">
<meta name="plugin-version" content="vim8.1_v2">
<meta name="syntax" content="none">
<meta name="settings" content="number_lines,use_css,pre_wrap,no_foldcolumn,expand_tabs,line_ids,prevent_copy=,use_input_for_pc=fallback">
<meta name="colorscheme" content="minimal-light">
<style>
<!--
pre { white-space: pre-wrap; font-family: monospace; color: #000000; background-color: #ffffd7; }
body { font-size:12pt; font-family: monospace; color: #000000; background-color: #ffffd7; }
a { color:inherit; }
* { font-size:12pt; font-size: 1em; }
.subxComment { color: #005faf; }
.SpecialChar { color: #d70000; }
.LineNr { }
.subxFunction { color: #af5f00; text-decoration: underline; }
.Constant { color: #008787; }
.CommentedCode { color: #8a8a8a; }
-->
</style>
<script>
<!--
/* function to open any folds containing a jumped-to line before jumping to it */
function JumpToLine()
{
var lineNum;
lineNum = window.location.hash;
lineNum = lineNum.substr(1); /* strip off '#' */
if (lineNum.indexOf('L') == -1) {
lineNum = 'L'+lineNum;
}
var lineElem = document.getElementById(lineNum);
/* Always jump to new location even if the line was hidden inside a fold, or
* we corrected the raw number to a line ID.
*/
if (lineElem) {
lineElem.scrollIntoView(true);
}
return true;
}
if ('onhashchange' in window) {
window.onhashchange = JumpToLine;
}
-->
</script>
</head>
<body onload='JumpToLine();'>
<a href='https://github.com/akkartik/mu/blob/main/302stack_allocate.subx'>https://github.com/akkartik/mu/blob/main/302stack_allocate.subx</a>
<pre id='vimCodeElement'>
<span id="L1" class="LineNr"> 1 </span><span class="subxComment"># A function which pushes n zeros on the stack.</span>
<span id="L2" class="LineNr"> 2 </span><span class="subxComment"># Really only intended to be called from code generated by mu.subx (for array</span>
<span id="L3" class="LineNr"> 3 </span><span class="subxComment"># vars on the stack).</span>
<span id="L4" class="LineNr"> 4 </span>
<span id="L5" class="LineNr"> 5 </span>== code
<span id="L6" class="LineNr"> 6 </span>
<span id="L7" class="LineNr"> 7 </span><span class="CommentedCode">#? Entry:</span>
<span id="L8" class="LineNr"> 8 </span><span class="CommentedCode">#? # . prologue</span>
<span id="L9" class="LineNr"> 9 </span><span class="CommentedCode">#? 89/<- %ebp 4/r32/esp</span>
<span id="L10" class="LineNr">10 </span><span class="CommentedCode">#? #</span>
<span id="L11" class="LineNr">11 </span><span class="CommentedCode">#? 68/push 0xfcfdfeff/imm32</span>
<span id="L12" class="LineNr">12 </span><span class="CommentedCode">#? b8/copy-to-eax 0x34353637/imm32</span>
<span id="L13" class="LineNr">13 </span><span class="CommentedCode">#? $dump-stack0:</span>
<span id="L14" class="LineNr">14 </span><span class="CommentedCode">#? (push-n-zero-bytes 4)</span>
<span id="L15" class="LineNr">15 </span><span class="CommentedCode">#? 68/push 0x20/imm32</span>
<span id="L16" class="LineNr">16 </span><span class="CommentedCode">#? $dump-stack9:</span>
<span id="L17" class="LineNr">17 </span><span class="CommentedCode">#? b8/copy-to-eax 1/imm32/exit</span>
<span id="L18" class="LineNr">18 </span><span class="CommentedCode">#? cd/syscall 0x80/imm8</span>
<span id="L19" class="LineNr">19 </span>
<span id="L20" class="LineNr">20 </span><span class="subxComment"># This is not a regular function, so it won't be idiomatic.</span>
<span id="L21" class="LineNr">21 </span><span class="subxComment"># Registers must be properly restored.</span>
<span id="L22" class="LineNr">22 </span><span class="subxComment"># Registers can be spilled, but that modifies the stack and needs to be</span>
<span id="L23" class="LineNr">23 </span><span class="subxComment"># cleaned up.</span>
<span id="L24" class="LineNr">24 </span>
<span id="L25" class="LineNr">25 </span><span class="subxComment"># Overhead:</span>
<span id="L26" class="LineNr">26 </span><span class="subxComment"># 62 + n*6 instructions to push n bytes.</span>
<span id="L27" class="LineNr">27 </span><span class="subxComment"># If we just emitted code to push n zeroes, it would be:</span>
<span id="L28" class="LineNr">28 </span><span class="subxComment"># 5 bytes for 4 zero bytes, so 1.25 bytes per zero. And that's not even</span>
<span id="L29" class="LineNr">29 </span><span class="subxComment"># instructions.</span>
<span id="L30" class="LineNr">30 </span><span class="subxComment"># But on the other hand it would destroy the instruction cache, where this</span>
<span id="L31" class="LineNr">31 </span><span class="subxComment"># approach requires 15 instructions, fixed.</span>
<span id="L32" class="LineNr">32 </span>
<span id="L33" class="LineNr">33 </span><span class="subxComment"># n must be positive</span>
<span id="L34" class="LineNr">34 </span><span class="subxFunction">push-n-zero-bytes</span>: <span class="subxComment"># n: int</span>
<span id="L35" class="LineNr">35 </span><span class="Constant">$push-n-zero-bytes:prologue</span>:
<span id="L36" class="LineNr">36 </span> 89/<- *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L56'>Push-n-zero-bytes-ebp</a></span> 5/r32/ebp <span class="subxComment"># spill ebp without affecting stack</span>
<span id="L37" class="LineNr">37 </span> 89/<- %ebp 4/r32/esp
<span id="L38" class="LineNr">38 </span><span class="Constant">$push-n-zero-bytes:copy-ra</span>:
<span id="L39" class="LineNr">39 </span> <span class="subxComment"># -- esp = ebp</span>
<span id="L40" class="LineNr">40 </span> 89/<- *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L60'>Push-n-zero-bytes-eax</a></span> 0/r32/eax
<span id="L41" class="LineNr">41 </span> 8b/-> *esp 0/r32/eax
<span id="L42" class="LineNr">42 </span> 2b/subtract *(ebp+4) 4/r32/esp
<span id="L43" class="LineNr">43 </span> <span class="subxComment"># -- esp+n = ebp</span>
<span id="L44" class="LineNr">44 </span> 89/<- *esp 0/r32/eax
<span id="L45" class="LineNr">45 </span> 8b/-> *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L60'>Push-n-zero-bytes-eax</a></span> 0/r32/eax
<span id="L46" class="LineNr">46 </span><span class="Constant">$push-n-zero-bytes:bulk-cleaning</span>:
<span id="L47" class="LineNr">47 </span> 89/<- *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L58'>Push-n-zero-bytes-esp</a></span> 4/r32/esp
<span id="L48" class="LineNr">48 </span> 81 0/subop/add *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L58'>Push-n-zero-bytes-esp</a></span> 4/imm32
<span id="L49" class="LineNr">49 </span> 81 0/subop/add *(ebp+4) 4/imm32
<span id="L50" class="LineNr">50 </span> (<a href='120allocate.subx.html#L869'>zero-out</a> *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L58'>Push-n-zero-bytes-esp</a></span> *(ebp+4)) <span class="subxComment"># n+4</span>
<span id="L51" class="LineNr">51 </span><span class="Constant">$push-n-zero-bytes:epilogue</span>:
<span id="L52" class="LineNr">52 </span> 8b/-> *<span class="SpecialChar"><a href='302stack_allocate.subx.html#L56'>Push-n-zero-bytes-ebp</a></span> 5/r32/ebp <span class="subxComment"># restore spill</span>
<span id="L53" class="LineNr">53 </span> c3/return
<span id="L54" class="LineNr">54 </span>
<span id="L55" class="LineNr">55 </span>== data
<span id="L56" class="LineNr">56 </span><span class="SpecialChar">Push-n-zero-bytes-ebp</span>: <span class="subxComment"># (addr int)</span>
<span id="L57" class="LineNr">57 </span> 0/imm32
<span id="L58" class="LineNr">58 </span><span class="SpecialChar">Push-n-zero-bytes-esp</span>: <span class="subxComment"># (addr int)</span>
<span id="L59" class="LineNr">59 </span> 0/imm32
<span id="L60" class="LineNr">60 </span><span class="SpecialChar">Push-n-zero-bytes-eax</span>:
<span id="L61" class="LineNr">61 </span> 0/imm32
</pre>
</body>
</html>
<!-- vim: set foldmethod=manual : -->
|