Announcement

Collapse
No announcement yet.

OpenMandriva Is Also Making Plans To Move Away From 32-Bit Support

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • angrypie
    replied
    If we don't need backwards compatibility anymore we might as well drop x86 entirely, since we're stuck with it exactly because of the huge software library.

    Leave a comment:


  • Weasel
    replied
    Originally posted by starshipeleven View Post
    You don't know shit, john snow.

    At most you are wasting more space with the Long as now you fit what originally was 32bit in a 64bit variable, everything else stays the same.
    Except when recompilation introduces new security vulnerabilities: https://blog.acolyer.org/2016/11/17/...bit-platforms/

    We all know every crap "software developer" and their mom abuses 'int' everywhere like true retards.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by duby229 View Post
    You need to define variable types to fit the data, and variable types are not the same size between 32bit and 64bit x86. Almost every software written by anyone, good or bad, needs case modifications to get them to compile on both 32bit and 64bit.
    You don't know shit, john snow.

    At most you are wasting more space with the Long as now you fit what originally was 32bit in a 64bit variable, everything else stays the same.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by duby229 View Post
    At minimum, long longs need to be changed to longs. That's bare minimum. Almost -all- code needs to have some case modifications to get it to compile for both 32bit and 64bit x86. They -aren't- totally compatible.
    Yo that's peanuts, not "fixes requiring testing", unless you are doing bad shit with the code like manual bit swapping in the variable and such.

    Plus there is still -HUGE- amounts of x86 asm in production as we speak.
    I'm not sure why x86 ASM even matters in this discussion. That's again performance-critical code.

    You can't call something shit just because it uses the features of the architecture it was -designed- for you stupid idiot.
    I'm calling it shit because if doing cosmetic changes to variable declarations and such does cause issues it's indeed doing bad shit.

    The only place where that is ok is for performance-critical code, there everything goes, even bad shit, as you NEED performance.

    And now I'm convinced you've never written a single piece of software.
    That's not my job so I'm not really amazing but I can still hold my own against internet posers like you I guess.

    Variable sizes -DO- change. The worst of them is long long on 32bit is just a long on 64bit. Structures on 64bit x86 are -NOT- the same as structures on 32bit x86.
    https://stackoverflow.com/questions/...64-bit-windows

    tl;dr the only one that changes is the long from 32bit to 64, on Linux/Unix (and not on Windows). Long long still exists and is still 64bit in a 64bit system, be it Windows or Linux, so you don't need to touch that.

    Also, if you used platform-neutral integer type names as shown in the link you don't necessarily need to care about architecture or bitness changes.

    Also for the struct thing, see what other sane programmers actually do in another reply in this forum https://www.phoronix.com/forums/foru...29#post1108029
    Last edited by starshipeleven; 22 June 2019, 12:05 PM.

    Leave a comment:


  • Vistaus
    replied
    Originally posted by starshipeleven View Post
    Is your Google broken good sir?
    It is, which is why I use DuckDuckGo.

    Leave a comment:


  • duby229
    replied
    Originally posted by starshipeleven View Post
    That's the only one that I was talking about.

    And yes I remain convinced that if your software needs any significant fixing to be migrated to 64bit it's because it was coded badly.

    A lot of opensource software didn't have amazing code standards and still doesn't.
    You need to define variable types to fit the data, and variable types are not the same size between 32bit and 64bit x86. Almost every software written by anyone, good or bad, needs case modifications to get them to compile on both 32bit and 64bit.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by the_scx View Post
    Then all open source Linux software are a piece of sh*t. I can give you a lot of examples, where it was necessary to make some fixes to build something:
    - To support 64-bit arches.
    That's the only one that I was talking about.

    And yes I remain convinced that if your software needs any significant fixing to be migrated to 64bit it's because it was coded badly.

    A lot of opensource software didn't have amazing code standards and still doesn't.

    Leave a comment:


  • skeevy420
    replied
    Originally posted by the_scx View Post
    snip
    You're not wrong, I'm just more of a glass-half-full type of person.

    Especially with all the developments as of late -- I find it hard to be that negative towards Wine when in four or five years it went from "barely usable" to "a tool in my box". I just like how damn near all of my stuff just works, YMMV, on the latest builds of Wine/Proton without needing 75 custom prefixes made on a per-game basis (outside of edge-cases like FFVII, of course).

    Leave a comment:


  • duby229
    replied
    Originally posted by starshipeleven View Post
    what is a "structure" in this context?

    because variable sizes don't change and a "structure" in c/c++ is basically a list of variables.
    OMG, you really are a fucking retarded moron. And now I'm convinced you've never written a single piece of software. Variable sizes -DO- change. The worst of them is long long on 32bit is just a long on 64bit. Structures on 64bit x86 are -NOT- the same as structures on 32bit x86.

    Leave a comment:


  • duby229
    replied
    Originally posted by xfcemint View Post

    Most of my software is multi-platform, but switching to 64-bits changes sizes of structures. That change needs extensive testing.
    How can I be sure that some old program of mine doesn't depend on some structure having exactly X bytes? Or some library that my program is using?

    Very risky stuff.
    Think of all the man-hours required for testing, and for checking every structure definition in the source code.
    Thank you! Exactly this! ^^

    Leave a comment:

Working...
X