bug#63258: GCC package lacks librt.so (and instead only provides versioned librt.so.1)

[Mekeor Melire]

The "gcc" package (as well as the "gcc-toolchain" package) (since 
version 12?) lacks the librt.so file which some stuff links to.

As a consequence, in bug#63012, Josselin Poiret (CC) had to patch 
Zig to not link to that file. Another consequence is that I had to 
install gcc-toolchain:static in order to make GHC compile XMonad.

On #guix IRC channel, Josselin Poiret stated that we should 
include an empty librt.a in the output of the "gcc" package. And 
Ludo (CC) suggested to temporarily include that file only in 
gcc-toolchain, in order to avoid world rebuild.

Here's the full relevant IRC log:

--8<---------------cut here---------------start------------->8---
2023-05-04 09:49:54 mekeor: bjc: for me, gcc-toolchain@12 is 
shipped with librt.so.1 which is present in my ~/.guix-profile/lib 
folder which is listed in my $LIBRARY_PATH variable. still, ghc 
can't find it :/
2023-05-04 09:56:15 jpoiret: mekeor: that's because ghc and others 
look for librt.so, not librt.so.1
2023-05-04 09:56:31 jpoiret: we forgot to include an empty librt.a 
in the out output of gcc unfortunately
2023-05-04 09:56:59 jpoiret: you can add gcc-toolchain:static for 
now
2023-05-04 09:57:04 mekeor: jpoiret: is there a workaround? :) 
also, is there a patch already
2023-05-04 09:57:08 mekeor: oh cool
2023-05-04 09:57:12 jpoiret: I don't think so
2023-05-04 09:57:50 jpoiret: basically librt.so.1 is empty now, 
since everything is provided by libc
2023-05-04 09:59:43 mekeor: the "static" output fixes my problem. 
thank you very much, jpoiret!
2023-05-04 09:59:47 jpoiret: the problem is that it'll require a 
world rebuild so it's not an easy change to make now. But we'll 
probably have to do it anyway
2023-05-04 10:00:01 jpoiret: might as well batch some other 
important changes like updating the glibc again
2023-05-04 10:00:08 jpoiret: just to break everything again :)
2023-05-04 10:00:41 civodul: jpoiret: we could add an empty 
librt.a in gcc-toolchain, rather than gcc?
2023-05-04 10:01:00 civodul: that would address most practical 
issues, no?
2023-05-04 10:01:26 jpoiret: ah, that's right :)
2023-05-04 10:01:55 jpoiret: but still, it's probably a good idea 
to have it in gcc, zig wanted it and I had to patch out -librt
2023-05-04 10:01:59 jpoiret: -lrt *
2023-05-04 10:02:54 civodul: yes, that'd be the right fix, but in 
the meantime we could have the easy fix :-)
--8<---------------cut here---------------end--------------->8---