Aquileo | Recent changes to feature-requestshttps://sourceforge.net/p/nasm/feature-requests/2018-07-07T08:56:00.307000ZRecent changes to feature-requestsAquileo | New(?) opcodes2018-07-07T08:56:00.307000Z2018-07-07T08:56:00.307000ZIouri Kharonhttps://sourceforge.net/u/yjh-styx/https://sourceforge.net934ae38abc004579c573a7a95a2fe429ff87a1e0<div class="markdown_content"><p>Please add opcodes for endbr32/endbr64 (see <a href="https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf" rel="nofollow">https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf</a>).</p></div>Aquileo | Ability to enable deprecated/experimental features via command line flags2016-06-28T15:18:23.678000Z2016-06-28T15:18:23.678000ZOleh Lukyrychhttps://sourceforge.net/u/olukyrich/https://sourceforge.net2c4972550fb83f9d5bfc42f84b66c87db0729f54<div class="markdown_content"><p>For the sake of backward compatibility, it seems reasonable to hide the deprecated functionality behind the cmd flag instead of completely removing it.<br/>
The same may be applied to experimental features in order to beta-test them, for instance.</p>
<p>The reason I raise it -- is the removal of the context-through lookup (http://www.nasm.us/doc/nasmdoc4.html#section-4.7.4).<br/>
I agree that this feature is dangerous and can lead to quite tricky problems, but when the programmer uses it consciously, it may appear very useful or even irreplaceable. <br/>
I would take an older version, but that means I shall get no further fixes, and features.</p>
<p>Thank you.</p></div>Aquileo | #196 Support self-relative addresses to data2013-07-27T07:48:37.389000Z2013-07-27T07:48:37.389000ZA Foghttps://sourceforge.net/u/agnerfog/https://sourceforge.net1a6664e854569f5e5048ff24f3606564348d0245<div class="markdown_content"><p>Sorry. I have moved it to <br />
<a href="http://bugzilla.nasm.us/show_bug.cgi?id=3392263" rel="nofollow">http://bugzilla.nasm.us/show_bug.cgi?id=3392263</a></p></div>Aquileo | #176 Negative relocations (bin)2013-07-26T20:08:43.523000Z2013-07-26T20:08:43.523000ZC. Maslochhttps://sourceforge.net/u/estron/https://sourceforge.netc821fc95b6df2e80b80461c1b80b46e8bc462f27<div class="markdown_content"><blockquote>
<p>Negative relocations are not possible in any of the object file format I have seen.</p>
</blockquote>
<p>Yes. And?</p>
<p>Besides, as you likely know, the bin output format of NASM isn't exactly an object format. It is more akin to a linker's output format.</p>
<blockquote>
<p>You probably need a self-relative relocation. See feature request 196</p>
</blockquote>
<p>I do not. I now and then find cases where relocations with multipliers other than one would be useful, even beyond what can be covered by the specific case of self-relativ relocations.</p></div>Aquileo | #196 Support self-relative addresses to data2013-07-26T11:37:40.518000Z2013-07-26T11:37:40.518000ZCyrill Gorcunovhttps://sourceforge.net/u/cyrillos/https://sourceforge.net5509e53c6ada388f105023470bea280ad1d6f22c<div class="markdown_content"><p>Could you please use bugzilla.nasm.us for future requests, we're trying to minimize use of SF buckracking.</p></div>Aquileo | #176 Negative relocations (bin)2013-07-26T11:35:08.856000Z2013-07-26T11:35:08.856000ZA Foghttps://sourceforge.net/u/agnerfog/https://sourceforge.net8f1a326750cdcd9a27fb4174c6a59c9469541417<div class="markdown_content"><p>Negative relocations are not possible in any of the object file format I have seen.<br />
You probably need a self-relative relocation. See feature request 196</p>
<p><a href="https://sourceforge.net/p/nasm/feature-requests/196/">https://sourceforge.net/p/nasm/feature-requests/196/</a></p></div>Aquileo | #176 Negative relocations (bin)2013-07-26T11:33:19.966000Z2013-07-26T11:33:19.966000ZA Foghttps://sourceforge.net/u/agnerfog/https://sourceforge.netcd7bdf11d1647a77e99a7c63af2841a916e74644<div class="markdown_content"><p>Negative relocations are not supported in any of the object file formats I have seen. You probably need a self-relative relocation.<br />
See feature request # 196</p>
<p><a href="https://sourceforge.net/p/nasm/feature-requests/196/">https://sourceforge.net/p/nasm/feature-requests/196/</a></p></div>Aquileo | Support self-relative addresses to data2013-07-26T11:31:03.135000Z2013-07-26T11:31:03.135000ZA Foghttps://sourceforge.net/u/agnerfog/https://sourceforge.net45b68001418d007e7b68b9e4871909d3475b6894<div class="markdown_content"><p>NASM does not support self-relative addresses from code to data, but YASM does. As YASM is no longer maintained, I have to port my code to NASM. Therefore this request.</p>
<p>I am making self-relative code in 32 bit mode without a GOT in this way:</p>
<div class="codehilite"><pre><span class="na">.data</span>
<span class="nf">mydata</span> <span class="no">dd</span> <span class="mi">0</span>
<span class="na">.code</span>
<span class="nf">call</span> <span class="no">get_esp_into_ecx</span>
<span class="nl">ecx_points_here:</span>
<span class="nf">mov</span> <span class="no">eax</span><span class="p">,</span><span class="err">[</span><span class="no">ecx</span> <span class="err">+</span> <span class="p">(</span><span class="no">mydata</span> <span class="p">-</span> <span class="no">ecx_points_here</span><span class="err">]</span> <span class="err">;</span> <span class="no">read</span> <span class="no">mydata</span>
<span class="nf">ret</span>
<span class="nl">get_esp_into_ecx:</span>
<span class="nf">mov</span> <span class="no">ecx</span><span class="p">,</span><span class="err">[</span><span class="no">esp</span><span class="err">]</span> <span class="err">;</span> <span class="no">this</span> <span class="no">function</span> <span class="no">reads</span> <span class="no">its</span> <span class="no">own</span> <span class="no">return</span> <span class="no">address</span>
<span class="nf">ret</span>
</pre></div>
<p>This is the same method that is used for making position-independent code in a Unix shared object, except I want to get rid of the slow GOT lookup.</p>
<p>As I said, this works in YASM, while NASM says error: beroset-p-637-invalid effective address<br />
What is needed is a self-relative address, similar to the rip-relative address in 64-bit mode.</p></div>Aquileo | Support for MASM and TASM ptr directive2012-04-23T10:10:14Z2012-04-23T10:10:14ZAnonymoushttps://sourceforge.net/u/userid-None/https://sourceforge.net6dc9d274f25d31630b86bd9c7c735169e1b2e766<div class="markdown_content"><p>I suggest that NASM should have an additional feature to support the 'ptr' directive making previous TASM and MASM users adapt the NASM syntax easily with less difficulty.</p></div>Aquileo | add support for VMFUNC instruction2011-12-21T06:43:14Z2011-12-21T06:43:14Znasm64developerhttps://sourceforge.net/u/nasm64developer/https://sourceforge.net421044d4fd416c574c111a0754652744fad4ee9f<div class="markdown_content"><p>It's documented in version 041 of the Intel manuals, in the section about VMX.</p>
<p>VMFUNC = 0Fh,01h,D4h</p>
<p>EAX is implied as its operand</p></div>