/etc/passwd and symlinks

[Ralf Friedl]

Paul Fox wrote:
> ralf wrote:
>  > 
>  > Also, I don't see why insmod should need the realpath of the object. I 
>  > thought the basename of the object file is used as the module name, but 
>  > I have also seen few cases where the module name was not the file name, 
>  > so that seems to be only a convention and not mandatory.
>
> so you're saying the current code in insmod.c, and even a change to
> use the _follow() version of readlink, would be okay, right?  just making
> sure i understand you.  :-)
>   
No, I didn't at all look at the insmod code. Actually this seems to 
refer to the code for 2.4 kernels, and I haven't used 2.4 kernels for 
some time.
As realpath was used in add_ksymoops_symbols, I suppose ksymoops might 
also work with the unmodified filename, as the kernel would follow the 
same symlinks and eventually arrive at the same file.
But I don't know and I don't have a kernel 2.4 or the utilities for the 
older kernels to try it.

>  > I noticed that you test against MAXSYMLINKS only in the case of relative 
>  > symlinks. I consider this inconsistent.
>
> oops, you're right.  that's an oversight, which i'll fix.  (note
> that the test isn't really guaranteed -- there may be more
> undetected symlinks embedded in the middle of the path, since i'm
> only checking the tail.  but it will be caught eventually.)
>   
symlinks in the path may cause ELOOP in readlink. That should be no 
problem, the final path would case ELOOP anyway.

Regards
Ralf Friedl