| 1 |
On Sun, Nov 10, 2019 at 11:37 PM Caveman Al Toraboran |
| 2 |
<toraboracaveman@××××××××××.com> wrote: |
| 3 |
> |
| 4 |
> hi - is it possible to have some kind of fancy links that |
| 5 |
> know the name of the process that is trying to access |
| 6 |
> it, and based on its name, it links it to a file? |
| 7 |
|
| 8 |
Yes, and that's used pretty extensively in busybox. Symlinks named any |
| 9 |
of its tools that point to it directly call into *that* tool. |
| 10 |
|
| 11 |
> i think if we have this, we can solve slotting in a simpler |
| 12 |
> way. e.g. we install libs in their own non-conflicting |
| 13 |
> locations, and then install for them such fancy sym links |
| 14 |
> with access that routes accessing processes to the right |
| 15 |
> version of the lib. |
| 16 |
|
| 17 |
I'm not entirely sure about the library level. Your goal sounds like |
| 18 |
one link to many library versions instead of many links to one |
| 19 |
library. A symlink itself can only point at one thing, so the |
| 20 |
filesystem level can't do that. So you would have to have a shim |
| 21 |
library that gets loaded, then loads the version the calling process |
| 22 |
needs... based on either an exhaustive table of the mappings of every |
| 23 |
binary to the versions they need, or some other magic.You wouldn't be |
| 24 |
able to just pick any version of a library and hand it back to a |
| 25 |
process, since major versions on libraries tend to break the |
| 26 |
interfaces they present, so you would end up having a 1:1 mapping of |
| 27 |
any particular binary calling a singular library version. That already |
| 28 |
happens based on the linker, and even works for less specific |
| 29 |
versioning in many cases with symlinks from, say, libc.so.6 and |
| 30 |
libc.so.6.2 pointing to libc.so.6.2.13 (which is actually a many links |
| 31 |
to one library technique that's already used). To change what |
| 32 |
interface a calling process *can* work with will require rebuilding it |
| 33 |
against the new major version of the library, and if it's one of those |
| 34 |
circular situations where it's part of the build process, you have to |
| 35 |
have both the old and new library available (one of the reason slots |
| 36 |
exist). The binary itself should contain enough detail to load the |
| 37 |
version *it* needs already, and then should be requesting that from |
| 38 |
the dynamic linker already... |
| 39 |
|
| 40 |
-- |
| 41 |
Poison [BLX] |
| 42 |
Joshua M. Murphy |
Archives