supersandro2000 has joined #nixos-infra
<
lukegb>
gchristensen: how's the configuration of the packet builders determined, btw
<
gchristensen>
how do you mean? (typing one handed)
<
lukegb>
I notice some of them are big-parallel
<
lukegb>
and there's a mix of things with different cores counts etc
<
lukegb>
nix.conf cores, that is
<
lukegb>
(presumably split by big-parallel vs not)
<
gchristensen>
rand() with a weighting towards not being big parallel
<
lukegb>
ah, makes sense
<
gchristensen>
oops
<
lukegb>
aha, and zfs
<
lukegb>
this also explains the other thing
<
gchristensen>
neat
<
gchristensen>
"neat"
<
lukegb>
du's tension between "tell me how large these files are" and "tell me how much disk space I'm using" strikes again
<
gchristensen>
bug in qemu they don't want to fix
<
gchristensen>
can't find it but qemu reports the wrong size of the file if it is on a fs with compression
<
gchristensen>
breaks the virt platform based around qemu, I forget which that is ...
<
gchristensen>
oh, duh, libvirt.
<
samueldr>
lukegb: fixing it
<
samueldr>
the du thing
<
samueldr>
but by verifying the correctness of my fix, I found another small issue
<
samueldr>
well, compounding issue
<
gchristensen>
I guess this is unrelate afterall :)
<
lukegb>
ooh nixops-libvirtd
<
gchristensen>
sounds like samueldr is bringing the goods
<
samueldr>
sunken costs fallacy is killing me
<
gchristensen>
how about a filesystem that always reports the size-on-disk as 1 byte
<
gchristensen>
oh, samueldr?
<
samueldr>
knowing I have written a much sturdier disk image infra
<
samueldr>
but patching things in this... ugly organic script
<
gchristensen>
ah :)
<
samueldr>
we already spoke about that previously :)
<
samueldr>
really, mixing the concerns is a terrible mistake
<
samueldr>
you should make the
*partition image* as a distinct step
<
samueldr>
then use that image's own data to make the partitions
<
gchristensen>
nice idea
<
samueldr>
so delay the final disk image until all partitions are built
<
samueldr>
this way you
**know** the actual sizes
<
samueldr>
no need to guesstimate them
<
gchristensen>
zfs makes that fairly annoying
<
samueldr>
ext4 does too
<
gchristensen>
yeah but....
<
lukegb>
my fix to 20.09 appears to have made the chromium test pass at least, so hooray, that'll hopefully advance
<
gchristensen>
if you give an entire disk to zfs, like this PR does, ZFS creates 2 partitions, #1 and #9 and uses all the disk space between the two
<
samueldr>
I don't follow
<
samueldr>
ZFS makes a GPT (or MBR) when given a raw block device?
<
samueldr>
this is actual from the implementation
<
samueldr>
it is trivial to back track the reserved size to make a
*snug* image!
<
samueldr>
what's the deal with ZFS doing partitions?
<
samueldr>
that's so wrong
<
lukegb>
iirc to do things like supporting bootloaders on the disk
<
samueldr>
is it
*always* adding a GPT to block devices?
<
samueldr>
then there's no issues
<
samueldr>
treat it as an opaque image
<
gchristensen>
ah, right
<
gchristensen>
yeah
<
samueldr>
I spent a
*lot* of time thinking about disk images and partitions
<
samueldr>
both with Mobile NixOS and [thing]
<
gchristensen>
maybe you can help me figure out how to clean up this PR
<
gchristensen>
b/c I deploy it to a lot of places and would love for it to be upstreamed :)
<
lukegb>
but it involves patching the BIOS that qemu is using, and that's just a binary blob in the qemu source tree
<
lukegb>
I think we can just build it separately and pass it as a command line param
* gchristensen
starts to wander off for the night
<
gchristensen>
thanks for the looking and fixups, y'all :) really wonderful
<
samueldr>
lukegb: I was patching the seabios for mobile nixos for a while
<
samueldr>
I swear I'm not trying to show off
<
samueldr>
btw, I'm testing the amazon image with 0 additionalSize
<
samueldr>
which is where things get "fun"
<
lukegb>
I'm not actually sure if patching the seabios would help here though tbf
<
samueldr>
well, you said you needed to patch the BIOS :)
<
lukegb>
right, yeah; I
_think_ it might help
zimbatm[m] has quit [Ping timeout: 245 seconds]
zimbatm[m] has joined #nixos-infra
roberth has joined #nixos-infra
cole-h has quit [Ping timeout: 252 seconds]
roberth has quit [Ping timeout: 245 seconds]
roberth has joined #nixos-infra
thefloweringash has joined #nixos-infra
cole-h has joined #nixos-infra
aristid has quit [*.net *.split]
domenkozar[m] has quit [*.net *.split]
garbas[m] has quit [*.net *.split]
gchristensen has quit [*.net *.split]
supersandro2000 has quit [*.net *.split]
MichaelRaskin has quit [*.net *.split]
aminechikhaoui has quit [*.net *.split]
asymmetric has quit [*.net *.split]
V has quit [*.net *.split]
andi- has quit [*.net *.split]
flokli has quit [*.net *.split]
eyJhb has quit [*.net *.split]
cole-h has quit [*.net *.split]
zimbatm[m] has quit [*.net *.split]
thefloweringash has quit [*.net *.split]
roberth has quit [*.net *.split]
mcint has quit [*.net *.split]
tazjin has quit [*.net *.split]
ikwildrpepper has quit [*.net *.split]
JJJollyjim has quit [*.net *.split]
Ox4A6F has quit [*.net *.split]
sterni has quit [*.net *.split]
ikwildrpepper has joined #nixos-infra
aminechikhaoui has joined #nixos-infra
V has joined #nixos-infra
asymmetric has joined #nixos-infra
garbas[m] has joined #nixos-infra
aristid has joined #nixos-infra
MichaelRaskin has joined #nixos-infra
supersandro2000 has joined #nixos-infra
andi- has joined #nixos-infra
cole-h has joined #nixos-infra
gchristensen has joined #nixos-infra
domenkozar[m] has joined #nixos-infra
Ericson2314 has joined #nixos-infra
flokli has joined #nixos-infra
eyJhb has joined #nixos-infra
JJJollyjim has joined #nixos-infra
ryantm has joined #nixos-infra
Ox4A6F has joined #nixos-infra
sterni has joined #nixos-infra
zimbatm[m] has joined #nixos-infra
hexa- has joined #nixos-infra
nh2[m] has joined #nixos-infra
Ericson2314 has quit [Max SendQ exceeded]
hexa- has quit [Max SendQ exceeded]
roberth has joined #nixos-infra
thefloweringash has joined #nixos-infra
JJJollyjim has quit [Ping timeout: 248 seconds]
ryantm has quit [Ping timeout: 248 seconds]
Ox4A6F has quit [Ping timeout: 248 seconds]
garbas[m] has quit [Ping timeout: 258 seconds]
domenkozar[m] has quit [Ping timeout: 258 seconds]
thefloweringash has quit [Ping timeout: 247 seconds]
roberth has quit [Ping timeout: 272 seconds]
qyliss has joined #nixos-infra
tazjin has joined #nixos-infra
endocrimes has joined #nixos-infra
mcint has joined #nixos-infra
NinjaTrappeur has joined #nixos-infra
zimbatm[m] has quit [Ping timeout: 276 seconds]
nh2[m] has quit [Ping timeout: 276 seconds]
hexa- has joined #nixos-infra
domenkozar[m] has joined #nixos-infra
JJJollyjim has joined #nixos-infra
Ericson2314 has joined #nixos-infra
zimbatm[m] has joined #nixos-infra
thefloweringash has joined #nixos-infra
colemickens has joined #nixos-infra
garbas[m] has joined #nixos-infra
roberth has joined #nixos-infra
nh2[m] has joined #nixos-infra
Ox4A6F has joined #nixos-infra
ryantm has joined #nixos-infra
JJJollyjim is now known as Guest97996
supersandro2000 has quit [Killed (hitchcock.freenode.net (Nickname regained by services))]
supersandro2000 has joined #nixos-infra