ELF loading
in [ASM]
|
From: rhyde on 27 Aug 2007 14:14 On Aug 27, 8:12 am, Betov <be...(a)free.fr> wrote: > santosh <santosh....(a)gmail.com> écrivaitnews:faumf5$3b7$1(a)aioe.org: > > > > > > > Betov wrote: > > >> santosh <santosh....(a)gmail.com> écrivaitnews:faufng$at2$1(a)aioe.org: > > >>> BTW, do you compress the sources in your final PE? > > >> What for? Do you compress your Sources, on your disk? > > > Yes. > > > For example I recently downloaded the source distribution for Emacs > > 22.1 and compiled and installed it. To compile I have to unzip and > > untar the compressed archive which is ~37 Mb. After unzipping it > > occupies ~140 Mb. > > > So after compilation and install, I recompress the distribution. A > > saving of ~100 Mb is not insignificant. > > :]]]]]]]]]]]]]] > > Keep sure than, when you will have written 100 Megas of Assembly > Sources, you will be *VERY* *VERY* *VERY* *VERY* *VERY* old. > > :]]]]]]]]]]]]]] This is your problem understanding why things like object files and static linking are necessary. You only think "in the small". Though it is unlikely a single individual would ever create 100 megabytes of assembly source code, it's perfectly reasonable for 100 programmers working together on the same project to create 100 megabytes of assembly source code. And that's where RosAsm falls flat on it's face with the "team-hostile" monofile source format -- it isn't easy for a large team to use such a product. Someday, when you can think beyond small projects (and "4 megas" or otherwise, RosAsm is a small project), you'll understand the limitations of the scheme you force on RosAsm users. hLater, Randy Hyde
From: rhyde on 27 Aug 2007 14:16 On Aug 27, 8:42 am, santosh <santosh....(a)gmail.com> wrote: > Betov wrote: > > santosh <santosh....(a)gmail.com> écrivaitnews:faumf5$3b7$1(a)aioe.org: > > >> Betov wrote: > > >>> santosh <santosh....(a)gmail.com> écrivaitnews:faufng$at2$1(a)aioe.org: > > >>>> BTW, do you compress the sources in your final PE? > > >>> What for? Do you compress your Sources, on your disk? > > >> Yes. > > >> For example I recently downloaded the source distribution for Emacs > >> 22.1 and compiled and installed it. To compile I have to unzip and > >> untar the compressed archive which is ~37 Mb. After unzipping it > >> occupies ~140 Mb. > > >> So after compilation and install, I recompress the distribution. A > >> saving of ~100 Mb is not insignificant. > > > :]]]]]]]]]]]]]] > > > Keep sure than, when you will have written 100 Megas of Assembly > > Sources, you will be *VERY* *VERY* *VERY* *VERY* *VERY* old. > > > :]]]]]]]]]]]]]] > > Yes, that's why people say that HLLs are more "productive" than > assembler. :) > > PS. Wannabee might actually achieve 100 Mb one of these days.- Hide quoted text - Easy enough to do with cut and paste programming :-) 1 byte, cut and pasted is two bytes. two bytes, cut and pasted is four bytes. four bytes, cut and pasted is eight bytes. ....repeat a couple dozen times... 100s of MBs of source code! hLater, Randy Hyde
From: rhyde on 27 Aug 2007 14:28 On Aug 27, 7:38 am, Betov <be...(a)free.fr> wrote: > CodeMonk <jas...(a)yahoo.com> écrivait news:kGAAi.36284$t9.18582 > > Fact is that embeeding the sources inside the PEs has been the > second best decision i ever took (the first one was the Right- > Click features). It was certainly the most interesting feature you created. However, having gone done the path of a proprietary source file format myself (The LISA assembler, some 30 years ago), I can say without hesitation, from experience, that this path is a mistake. You give up the ability to use a *huge* variety of tools that work with plain text files when you go with a proprietary source format. And like RosAsm, LISA had "export and import" text file commands. It was a pain. > > And fact is that a decision can be told accurate, when it *works* > at a practical level. Sure it *works*. Doesn't mean it's a good design, though. I've "been there, done that" and I can tell you from experience that it's *not* as good as using plain text files for the source code. > For a Programming Tool, there is no other > final question but the users' *productivity*. And this was the major complaint with proprietary source formats 30 years ago. It's the same problem today -- productivity. You get locked into the editing tools and filters provided by the program; or forced to take the pain to explicitly save and reload the text format. Extra work. Outside the normal workflow. All that produces less productivity. > All guys who have > given a try to RosAsm and to another Assembler, have always > complained about how the other Assembler was way more difficult > and making them less productive than RosAsm. Oh, you mean Wannabee? Come on Rene, you have so few users that you don't have a reasonable statistical sample. And anecdotal evidence doesn't cut it. Have you actually *measured* people's productivities with the different products? And when these "RosAsm users" were using these other products, was the lack of productivity *really* due to embedding the source code in the executable? Or was it because the RosAsm user wasn't very experienced with the other assembler? Give us a break. > I sometimes heard > some saying "I am going to NASM or to FASM, because it can do > more output types", but i never heard any saying "I go to NASM > or FASM, because it is easier to use and more productive". It makes them a hell of a lot more productive if they want to produce ELF or Mach-O output types :-) > > How is this, if the basic design is wrong? Why do you ask? You've been told this many times in the past. By forcing a proprietary text format on people, you force them to use your tools, or you force them to go through extra steps if they want to use other tools. Despite your constant "chest-bombing" about your "sources editor" and "right-click", the bottom line is that many people find other source code editors to be more to their liking (making them more productive, as they already know those editors and don't have to learn, and remember how to use, your's). Other people have lots of tools (particularly under Linux, now that you're exploring there) that they like to use. Tools that can be automated via scripts, like grep, sed, awk, and so on. Tools that work on *text* files, not your proprietary source format. Your source format scheme is like everything else about RosAsm: a completely non-standard way of doing things that forces the user to learn to do things the "Rene way" rather than the way they've grown accustomed to after many years of using existing programming tools. And that *kills* productivity. Sure, someone who has *only* used RosAsm will find it more productive than some other system. But someone who has used many different programming tools will find RosAsm a bit less productive than you claim. hLater, Randy Hyde
From: Betov on 27 Aug 2007 14:51 "rhyde(a)cs.ucr.edu" <rhyde(a)cs.ucr.edu> �crivait news:1188238220.551093.6820(a)q4g2000prc.googlegroups.com: > On Aug 27, 9:23 am, Betov <be...(a)free.fr> wrote: >> >> No idea. The biggest source ever written with RosAsm is RosAsm >> itself, and RosAsm is one of biggest Assembly Source ever written. > > Far from it, dude. > You do realize that OS\360 was written in assembly? It was just a > *little* bigger than <name stolen from ReactOS>:RosAsm. Ooooppssss... i forgot that the universal expert in number of lines, sizes, dates, was around! :) Hey! Clown! I told you something wrong: Before going and learn Power Basic, first try to make an idea of the difference in between a File and an OS. Take care, clown, you are walking on your lace! :)) Betov. < http://rosasm.org >
From: Betov on 27 Aug 2007 14:53
"rhyde(a)cs.ucr.edu" <rhyde(a)cs.ucr.edu> �crivait news:1188238466.264740.294160(a)q5g2000prf.googlegroups.com: > This is your problem understanding why things like object files and > static linking are necessary Right, clown. You are right, i will never understand this. :) Betov. < http://rosasm.org > |