home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: gnu.emacs.help
- Path: sparky!uunet!cs.utexas.edu!torn!nott!bnrgate!bmerh85!bmerh85!hamish
- From: Hamish.Macdonald@x400gate.bnr.ca (Hamish Macdonald)
- Subject: Re: Byte-Compiling emacs load files
- In-Reply-To: mac@bnr.ca's message of Tue, 10 Nov 1992 16:26:24 GMT
- Message-ID: <1992Nov11.140516.2315@bmerh85.bnr.ca>
- Lines: 21
- Sender: news@bmerh85.bnr.ca (Usenet News)
- Organization: Bell Northern Research
- References: <1992Nov10.162624.13800@bnr.ca>
- Date: Wed, 11 Nov 92 14:05:16 GMT
-
- >>>>> On Tue, 10 Nov 1992 16:26:24 GMT,
- >>>>> In message <1992Nov10.162624.13800@bnr.ca>,
- >>>>> mac@bnr.ca (Michael Campbell) wrote:
-
- Michael> I have recently been compiling some "extensions" to my
- Michael> standard emacs environment, particularily the hexl hex
- Michael> editing mode, and the evi vi emulator. At one point, someone
- Michael> who sent me an email suggested that if I first loaded the .el
- Michael> file, and *then* compiled the function, the resultant .elc
- Michael> file was "faster", "more efficient", etc. [...]
-
- Michael> Is there any truth to this? If there is, why is the result
- Michael> better than just directly compiling the unloaded file? Any
- Michael> and all insight would be greatly appreciated. Please post
- Michael> replies to this newsgroup.
-
- I believe that it is Emacs Lisp macros which make a difference. If
- the macros are defined when the code is compiled, the resulting
- byte-code is more efficient. In some cases, the byte-compiled code
- won't even work if it the source file was not loaded before it was
- compiled.
-