📄 nasmdoca.htm
字号:
<html><head><title>NASM Manual</title></head><body><h1 align=center>The Netwide Assembler: NASM</h1><p align=center><a href="nasmdocb.html">Next Chapter</a> |<a href="nasmdo10.html">Previous Chapter</a> |<a href="nasmdoc0.html">Contents</a> |<a href="nasmdoci.html">Index</a><h2><a name="appendix-A">Appendix A: Ndisasm</a></h2><p>The Netwide Disassembler, NDISASM<h3><a name="section-A.1">A.1 Introduction</a></h3><p>The Netwide Disassembler is a small companion program to the NetwideAssembler, NASM. It seemed a shame to have an x86 assembler, complete witha full instruction table, and not make as much use of it as possible, sohere's a disassembler which shares the instruction table (and some otherbits of code) with NASM.<p>The Netwide Disassembler does nothing except to produce disassemblies of<em>binary</em> source files. NDISASM does not have any understanding ofobject file formats, like <code><nobr>objdump</nobr></code>, and it willnot understand <code><nobr>DOS .EXE</nobr></code> files like<code><nobr>debug</nobr></code> will. It just disassembles.<h3><a name="section-A.2">A.2 Getting Started: Installation</a></h3><p>See <a href="nasmdoc1.html#section-1.3">section 1.3</a> for installationinstructions. NDISASM, like NASM, has a <code><nobr>man page</nobr></code>which you may want to put somewhere useful, if you are on a Unix system.<h3><a name="section-A.3">A.3 Running NDISASM</a></h3><p>To disassemble a file, you will typically use a command of the form<p><pre> ndisasm [-b16 | -b32] filename</pre><p>NDISASM can disassemble 16-bit code or 32-bit code equally easily,provided of course that you remember to specify which it is to work with.If no <code><nobr>-b</nobr></code> switch is present, NDISASM works in16-bit mode by default. The <code><nobr>-u</nobr></code> switch (for USE32)also invokes 32-bit mode.<p>Two more command line options are <code><nobr>-r</nobr></code> whichreports the version number of NDISASM you are running, and<code><nobr>-h</nobr></code> which gives a short summary of command lineoptions.<h4><a name="section-A.3.1">A.3.1 COM Files: Specifying an Origin</a></h4><p>To disassemble a <code><nobr>DOS .COM</nobr></code> file correctly, adisassembler must assume that the first instruction in the file is loadedat address <code><nobr>0x100</nobr></code>, rather than at zero. NDISASM,which assumes by default that any file you give it is loaded at zero, willtherefore need to be informed of this.<p>The <code><nobr>-o</nobr></code> option allows you to declare adifferent origin for the file you are disassembling. Its argument may beexpressed in any of the NASM numeric formats: decimal by default, if itbegins with `<code><nobr>$</nobr></code>' or `<code><nobr>0x</nobr></code>'or ends in `<code><nobr>H</nobr></code>' it's<code><nobr>hex</nobr></code>, if it ends in `<code><nobr>Q</nobr></code>'it's <code><nobr>octal</nobr></code>, and if it ends in`<code><nobr>B</nobr></code>' it's <code><nobr>binary</nobr></code>.<p>Hence, to disassemble a <code><nobr>.COM</nobr></code> file:<p><pre> ndisasm -o100h filename.com</pre><p>will do the trick.<h4><a name="section-A.3.2">A.3.2 Code Following Data: Synchronisation</a></h4><p>Suppose you are disassembling a file which contains some data whichisn't machine code, and <em>then</em> contains some machine code. NDISASMwill faithfully plough through the data section, producing machineinstructions wherever it can (although most of them will look bizarre, andsome may have unusual prefixes, e.g.`<code><nobr>FS OR AX,0x240A</nobr></code>'), and generating `DB'instructions ever so often if it's totally stumped. Then it will reach thecode section.<p>Supposing NDISASM has just finished generating a strange machineinstruction from part of the data section, and its file position is now onebyte <em>before</em> the beginning of the code section. It's entirelypossible that another spurious instruction will get generated, startingwith the final byte of the data section, and then the correct firstinstruction in the code section will not be seen because the starting pointskipped over it. This isn't really ideal.<p>To avoid this, you can specify a`<code><nobr>synchronisation</nobr></code>' point, or indeed as manysynchronisation points as you like (although NDISASM can only handle 8192sync points internally). The definition of a sync point is this: NDISASMguarantees to hit sync points exactly during disassembly. If it is thinkingabout generating an instruction which would cause it to jump over a syncpoint, it will discard that instruction and output a`<code><nobr>db</nobr></code>' instead. So it <em>will</em> startdisassembly exactly from the sync point, and so you <em>will</em> see allthe instructions in your code section.<p>Sync points are specified using the <code><nobr>-s</nobr></code> option:they are measured in terms of the program origin, not the file position. Soif you want to synchronise after 32 bytes of a<code><nobr>.COM</nobr></code> file, you would have to do<p><pre> ndisasm -o100h -s120h file.com</pre><p>rather than<p><pre> ndisasm -o100h -s20h file.com</pre><p>As stated above, you can specify multiple sync markers if you need to,just by repeating the <code><nobr>-s</nobr></code> option.<h4><a name="section-A.3.3">A.3.3 Mixed Code and Data: Automatic (Intelligent) Synchronisation </a></h4><p>Suppose you are disassembling the boot sector of a<code><nobr>DOS</nobr></code> floppy (maybe it has a virus, and you need tounderstand the virus so that you know what kinds of damage it might havedone you). Typically, this will contain a <code><nobr>JMP</nobr></code>instruction, then some data, then the rest of the code. So there is a verygood chance of NDISASM being <em>misaligned</em> when the data ends and thecode begins. Hence a sync point is needed.<p>On the other hand, why should you have to specify the sync pointmanually? What you'd do in order to find where the sync point would be,surely, would be to read the <code><nobr>JMP</nobr></code> instruction, andthen to use its target address as a sync point. So can NDISASM do that foryou?<p>The answer, of course, is yes: using either of the synonymous switches<code><nobr>-a</nobr></code> (for automatic sync) or<code><nobr>-i</nobr></code> (for intelligent sync) will enable<code><nobr>auto-sync</nobr></code> mode. Auto-sync mode automaticallygenerates a sync point for any forward-referring PC-relative jump or callinstruction that NDISASM encounters. (Since NDISASM is one-pass, if itencounters a PC-relative jump whose target has already been processed,there isn't much it can do about it...)<p>Only PC-relative jumps are processed, since an absolute jump is eitherthrough a register (in which case NDISASM doesn't know what the registercontains) or involves a segment address (in which case the target codeisn't in the same segment that NDISASM is working in, and so the sync pointcan't be placed anywhere useful).<p>For some kinds of file, this mechanism will automatically put syncpoints in all the right places, and save you from having to place any syncpoints manually. However, it should be stressed that auto-sync mode is<em>not</em> guaranteed to catch all the sync points, and you may stillhave to place some manually.<p>Auto-sync mode doesn't prevent you from declaring manual sync points: itjust adds automatically generated ones to the ones you provide. It'sperfectly feasible to specify <code><nobr>-i</nobr></code> <em>and</em>some <code><nobr>-s</nobr></code> options.<p>Another caveat with auto-sync mode is that if, by some unpleasant fluke,something in your data section should disassemble to a PC-relative call orjump instruction, NDISASM may obediently place a sync point in a totallyrandom place, for example in the middle of one of the instructions in yourcode section. So you may end up with a wrong disassembly even if you useauto-sync. Again, there isn't much I can do about this. If you haveproblems, you'll have to use manual sync points, or use the<code><nobr>-k</nobr></code> option (documented below) to suppressdisassembly of the data area.<h4><a name="section-A.3.4">A.3.4 Other Options</a></h4><p>The <code><nobr>-e</nobr></code> option skips a header on the file, byignoring the first N bytes. This means that the header is <em>not</em>counted towards the disassembly offset: if you give<code><nobr>-e10 -o10</nobr></code>, disassembly will start at byte 10 inthe file, and this will be given offset 10, not 20.<p>The <code><nobr>-k</nobr></code> option is provided with twocomma-separated numeric arguments, the first of which is an assembly offsetand the second is a number of bytes to skip. This <em>will</em> count theskipped bytes towards the assembly offset: its use is to suppressdisassembly of a data section which wouldn't contain anything you wanted tosee anyway.<h3><a name="section-A.4">A.4 Bugs and Improvements</a></h3><p>There are no known bugs. However, any you find, with patches ifpossible, should be sent to<a href="mailto:jules@dsf.org.uk"><code><nobr>jules@dsf.org.uk</nobr></code></a>or<a href="mailto:anakin@pobox.com"><code><nobr>anakin@pobox.com</nobr></code></a>,or to the developer's site at<a href="https://sourceforge.net/projects/nasm/"><code><nobr>https://sourceforge.net/projects/nasm/</nobr></code></a>and we'll try to fix them. Feel free to send contributions and new featuresas well.<p>Future plans include awareness of which processors certain instructionswill run on, and marking of instructions that are too advanced for someprocessor (or are <code><nobr>FPU</nobr></code> instructions, or areundocumented opcodes, or are privileged protected-mode instructions, orwhatever).<p>That's All Folks!<p>I hope NDISASM is of some use to somebody. Including me. :-)<p>I don't recommend taking NDISASM apart to see how an efficientdisassembler works, because as far as I know, it isn't an efficient oneanyway. You have been warned.<p align=center><a href="nasmdocb.html">Next Chapter</a> |<a href="nasmdo10.html">Previous Chapter</a> |<a href="nasmdoc0.html">Contents</a> |<a href="nasmdoci.html">Index</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
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -