codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos-emacs
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos-emacs
asymptotically has joined #nixos-emacs
<bqv>
NobbZ[m]: i've just noticed that i have `(defun cc-bytecomp-is-compiling (&rest r) nil)` in my config. presumably i got around your issue by using deferred compilation and that line
<NobbZ[m]>
"bqv" (https://matrix.to/#/@freenode_bqv:matrix.org) its Not even reading the config for compiling Emacs I hope...
<bqv>
yes, hence deferred compilation
<bqv>
or are you already using that?
<NobbZ[m]>
I'm not aware of using it. Also I always assumed that for compiling Emacs, no config external to the build inputs are read.
<bqy>
The point of deferred compilation is to move the native-comp from build time to first-run time
<bqy>
That then means you can fiddle with how it compiles in your config, no?
<NobbZ[m]>
Maybe, but the error is when Emacs compiles it built in stuff.
bkv has joined #nixos-emacs
<NobbZ[m]>
It's not about building my config, but building emacs
<bqy>
I don't think deferred comp is limited to external modules
<bqy>
As ever, nothing in emacs *needs* to be compiled, it will run fine without
bqv has quit [Ping timeout: 272 seconds]
bkv is now known as bqv
<NobbZ[m]>
But isn't compiling that stuff part of the Emacs derivation?
<bqy>
Not if you enable deferred compilation!
<NobbZ[m]>
And where to enable it? And why does it work with one computer and not with another when both have the same config?
<adisbladis>
PSA:
<adisbladis>
Native-comp & AOT compilation infrastructure is merged in nixpkgs now
<adisbladis>
I've also merged the required changes in the overlay
<adisbladis>
And added a hydra jobset for aot packages
<bqy>
No clue why, but add NATIVE_FAST_BOOT=1 to the makeflags I think
<NobbZ[m]>
And what is the point of reproducibility within a derivation if the simple existence of a "~/emacs.d/init.el" can change everything
<bqy>
adisbladis: oh, cool
<bqy>
Link?
<NobbZ[m]>
"adisbladis" (https://matrix.to/#/@freenode_adisbladis:matrix.org) so the overlay will only work on unstable now?
<adisbladis>
NobbZ[m]: No, but the native-comp branch never worked on stable
<NobbZ[m]>
Ah, okay.
<adisbladis>
Tbh though I would kind of want to drop support for stable
<adisbladis>
I'm also splitting the hydra jobsets, each emacs drv get's their own jobset now
<adisbladis>
This will make package stats look a bit weird for a bit, but I think it's better over time
bqv has joined #nixos-emacs
bkv has joined #nixos-emacs
bqv has quit [Ping timeout: 260 seconds]
bkv is now known as bqv
bkv has joined #nixos-emacs
bqv has quit [Ping timeout: 246 seconds]
bkv is now known as bqv
bkv has joined #nixos-emacs
bqv has quit [Ping timeout: 260 seconds]
bkv is now known as bqv
<mog>
i just rebuilt nixos-unstable, 20.09pre236091.28fce082c8c (Nightingale), and have latest emacs overlay. when i start emacs i see Warning (comp): No write access for /nix/store/i9q9pvbbmmdl87jijnn3dhqdjn3kjbwp-emacs-packages-deps/share/emacs/site-lisp/elpa/use-package-20200721.2156/eln-x86_64-pc-linux-gnu-00d68a2f3b2f6094/use-package-ensure.eln skipping. [7 times]
<mog>
i thought the patches that just went in solved the compalation issue?
<mog>
my pacakges are installed like extraPackages = epkgs: (with epkgs.melpaPackages; [ magit direnv nix-mode use-package ]);
bkv has joined #nixos-emacs
bqv has quit [Ping timeout: 246 seconds]
bkv is now known as bqv
<mog>
i thought the patches that just went in solved the compalation issue?'
<mog>
nevermind i seem to have elns now
<mog>
dont know why use-package is being fussy though
dash has joined #nixos-emacs
<dash>
hi. the recent changes to emacs-overlay don't seem to play nice with my emacs-init.org - anybody worked that out yet? (It's trying to parse the org-mode file as elisp)
<talyz>
dash: hi! yes, you can tangle the init file in a separate derivation before reading it