Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
Programming

Journal tomhudson's Journal: What kind of person names an executable as a .ko file? 8

The compiler command (I've left most of it out:

gcc (bunch of stuff omitted) -o filename.ko

files ending in dot-ko are supposed to be loadable kernel modules. So I wasted a day wondering WTF because the source files certainly didn't look like loadable modules.

Turns out they're not - the guy just picked the .ko extension for the heck of it, or because it was eventually supposed to become a kernel module, or ... I don't know, maybe they thought that by naming it like a kernel module, the kernel would "load it into kernel space and it would run faster."

This discussion has been archived. No new comments can be posted.

What kind of person names an executable as a .ko file?

Comments Filter:
  • Ah, the joy in spending boatloads of time chasing down a convention problem. :-)

    I feel for ya...
    • It sucks because I was told we were going to be writing this as a bunch of kernel modules, so when I see .ko files, I figure that parts been done already as a module ... but the code doesn't make sense, so ... maybe the "kernel magic" is in another include file ... no, not there ... not that one ... not that one either ...

      Finally, "This will never compile." "Yes it will." "How? There are no functions for initializing the module." "It's not a kernel module." "Then why is it NAMED as a module?" "The name sh

      • by fcgreg ( 670777 )
        The details are a little worse than I imagined. It sure would have been nice if they told you about that up-front, eh?

        And yes, given your description I'd say the odds are low that they'll drop their code. Then again, stranger things happen every day. :-)
        • Well, now that there are two of us saying this code is crap (okay, we say stuff like "did you see THIS file?", "Check out this section here...", etc.)

          Maybe with a good alternative there'll be some movement. I sure hope so.

  • It's unfortunate there's no compiler command to output a few brain cells. This would be yet another case where that would be highly useful.

  • I'm trying to think of how I would have handled that situation, and I can see loosing a few of my hairs. With luck, only the greying ones.
    Please tell me you didn't lose more that 75 minutes on this... That counts as Decepticon Maneuver #43, "Just Keep Them Guessing"
    • I spent a couple of hours today going through all the files manually to check dependancies, then making a half-decent make file.

      When I say "half-decent", I mean it. It's only half what a decent makefile would be, but its better than no make file at all.

      "No make file" I hear you say ... shocking, isn't it?

      Spent another couple of hours getting rid of all the "warning: suchandsuch soandso in file whatever" messages so it compiles clean. It may not run proplerly, but at least a few of the bugs are scotched

All great ideas are controversial, or have been at one time.

Working...