💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › freenode-kisslinux-2020-01-08.txt captured on 2024-06-16 at 13:52:22.
⬅️ Previous capture (2021-12-17)
-=-=-=-=-=-=-
2020-01-08T00:53:26 #kisslinux <konimex> so far all of my compiled programs (incl. python, etc. not tested with llvm yet but I doubt it requires them) actually doesn't require yacc/m4 at compile time
2020-01-08T00:53:38 #kisslinux <konimex> neat
2020-01-08T00:54:19 #kisslinux <E5ten> the kernel and mesa need bison
2020-01-08T00:55:15 #kisslinux <konimex> for now I only need CLI for the test system, GUI is a different beast to tackle on
2020-01-08T00:58:36 #kisslinux <micro_O> oh fuck
2020-01-08T00:58:48 #kisslinux <micro_O> welp, i hope these logs don't get scraped
2020-01-08T00:58:53 #kisslinux <micro_O> boo middle-click-as-paste default
2020-01-08T00:59:10 #kisslinux <konimex> this channel does get logged
2020-01-08T00:59:50 #kisslinux <micro_O> yeah im emailing the person who runs it now
2020-01-08T01:08:58 #kisslinux <micro_O> baleeted yay
2020-01-08T01:17:42 #kisslinux <konimex> unsurprisingly, some programs do need GNU make (such as busybox), but almost all programs I tested with ./configure so far works well with bmake
2020-01-08T01:39:26 #kisslinux <konimex> well, nevermind
2020-01-08T01:53:55 #kisslinux <E5ten> ?
2020-01-08T02:01:39 #kisslinux <konimex> some programs (such as curl and mandoc) can't seem to find the docs/manpages with bmake install (gmake, however, can find them fine)
2020-01-08T02:01:53 #kisslinux <konimex> https://termbin.com/2dh2 (bmake and bmake install)
2020-01-08T02:02:03 #kisslinux <konimex> https://termbin.com/edvr (bmake and gmake install)
2020-01-08T03:15:12 #kisslinux <TwistedFate> how easy is to have gnu on kiss distro?
2020-01-08T03:58:50 #kisslinux <konimex> TwistedFate: iirc someone created "kissD" aka. kiss package manager+glibc+systemd+all that
2020-01-08T03:58:54 #kisslinux <konimex> two secs
2020-01-08T03:59:20 #kisslinux <konimex> https://github.com/fanboimsft/kissD
2020-01-08T04:11:28 #kisslinux <konimex> man efibootmgr is a pain
2020-01-08T04:11:54 #kisslinux <E5ten> the build system is soooo fucked
2020-01-08T07:44:47 #kisslinux <kingk22> o/
2020-01-08T08:05:14 #kisslinux <icyphox> konimex: +1 for efibootmgr being a PITA
2020-01-08T08:05:41 #kisslinux <icyphox> but once you get the hang of it, it's alright
2020-01-08T08:06:12 #kisslinux <icyphox> my wireless is still dysfunctional
2020-01-08T08:06:24 #kisslinux <icyphox> all networking, actually
2020-01-08T08:06:46 #kisslinux <icyphox> only avenue that i have yet to try -- older kernel version
2020-01-08T08:08:23 #kisslinux <konimex> *using* efibootmgr is actually straightforward, building it is a different story, especially if one is at a "non-standard" build environment
2020-01-08T08:09:17 #kisslinux <icyphox> ah
2020-01-08T08:09:23 #kisslinux <icyphox> right
2020-01-08T08:32:51 #kisslinux <lieuxnoir> `efibootmgr --verbose` still doesn't work on my machine after the patch
2020-01-08T08:33:12 #kisslinux <lieuxnoir> i'll send the logs later, im on the phone
2020-01-08T09:18:04 #kisslinux <dylanaraps> lieuxnoir: Thanks
2020-01-08T09:18:20 #kisslinux <dylanaraps> The efibootmgr code is gross
2020-01-08T09:25:37 #kisslinux <dzove855> dylanaraps: sorry that i didn't have given you yesterday the information about my kiss linux problems. I had try'd myself, and have solved the issue that i got. So finally i have a clean installation :D
2020-01-08T09:25:48 #kisslinux <dylanaraps> Nice :)
2020-01-08T09:26:39 #kisslinux <dylanaraps> I want to write a Wiki page listing kernel configuration options which are required for a working install.
2020-01-08T09:29:48 #kisslinux <dzove855> yeaah would be nice :D also i think we can write some wiki for specific hardware use, like which option to enable for atheros wifi cards ans so on
2020-01-08T09:29:58 #kisslinux <dylanaraps> Yup
2020-01-08T09:30:20 #kisslinux <dylanaraps> ooo
2020-01-08T09:30:31 #kisslinux <dylanaraps> It seems like the efibootmgr issue is actually an issue in efivar.
2020-01-08T09:31:05 #kisslinux <konimex> the efivar.h not found problem?
2020-01-08T09:31:57 #kisslinux <dylanaraps> No no
2020-01-08T09:32:02 #kisslinux <dylanaraps> efibootmgr -v shows errors
2020-01-08T09:32:20 #kisslinux <dylanaraps> The verbose option doesn't actually work.
2020-01-08T09:32:28 #kisslinux <dylanaraps> It's a GCC 9 issue with efivar.
2020-01-08T09:32:42 #kisslinux <dylanaraps> Ubuntu, Debian and Arch pull in 4~ patches from upstream.
2020-01-08T09:32:56 #kisslinux <dylanaraps> b98ba8921010d03f46704a476c69861515deb1ca 2020-01-08T09:32:56 #kisslinux <dylanaraps> c3c553db85ff10890209d0fe48fb4856ad68e4e0 2020-01-08T09:32:56 #kisslinux <dylanaraps> 0dad6d78a7fb5f6c5fb4a1d646040539db6cf865 2020-01-08T09:32:56 #kisslinux <dylanaraps> fdb803402fb32fa6d020bac57a40c7efe4aabb7d 2020-01-08T09:32:57 #kisslinux <dylanaraps> 4e04afc2df9bbc26e5ab524b53a6f4f1e61d7c9e
2020-01-08T09:35:22 #kisslinux <dylanaraps> I might just swap the package to use a git snapshot of master keeping the version the same.
2020-01-08T09:35:24 #kisslinux <dylanaraps> Dunno yet.
2020-01-08T09:36:05 #kisslinux <dylanaraps> Upstream fo efivar and efibootmgr are seeing commits semi-regularly though no releases have been made in a long while.
2020-01-08T09:43:49 #kisslinux <dylanaraps> OK
2020-01-08T09:43:50 #kisslinux <dylanaraps> Pushed.
2020-01-08T09:45:46 #kisslinux <adamantium> o/
2020-01-08T09:45:50 #kisslinux <adamantium> sup fam
2020-01-08T09:48:35 #kisslinux <dylanaraps> Hey
2020-01-08T09:56:25 #kisslinux <dylanaraps> E5ten: https://github.com/systemd/systemd/issues/14495
2020-01-08T09:56:30 #kisslinux <dylanaraps> Lord
2020-01-08T10:41:38 #kisslinux <lieuxnoir> dylanaraps: forgot to change curl's sources
2020-01-08T10:43:23 #kisslinux <dylanaraps> Oops.
2020-01-08T10:43:25 #kisslinux <dylanaraps> Fixing now.
2020-01-08T10:43:26 #kisslinux <dylanaraps> Thanks.
2020-01-08T10:44:35 #kisslinux <dylanaraps> Fixed.
2020-01-08T10:45:13 #kisslinux <lieuxnoir> :D
2020-01-08T11:11:12 #kisslinux <TwistedFate> howdy
2020-01-08T11:14:07 #kisslinux <TwistedFate> just to clarify, if i run kiss, poetterware crap won't be able to sneak in as an update or something like that?
2020-01-08T11:23:12 #kisslinux <konimex> TwistedFate: as long as you're using official (main and community) repos only, yes, no systemd et al.
2020-01-08T11:25:06 #kisslinux <TwistedFate> nice, good to see another distro that shares the hatred towards poetterware crap
2020-01-08T11:27:44 #kisslinux <lieuxnoir> does anyone else still has issues with kernel configuration?
2020-01-08T11:32:27 #kisslinux <konimex> I just do a "localyesconfig" and be done with it
2020-01-08T11:32:54 #kisslinux <konimex> ^ from another system
2020-01-08T11:33:02 #kisslinux <lieuxnoir> yeah, was about to suggest lol
2020-01-08T11:33:13 #kisslinux <TwistedFate> weird, i can't seem to find videos about kiss on youtube
2020-01-08T11:33:28 #kisslinux <TwistedFate> is kiss new?
2020-01-08T11:33:37 #kisslinux <konimex> relatively new, yes
2020-01-08T11:33:39 #kisslinux <lieuxnoir> yes-ish
2020-01-08T11:35:38 #kisslinux <TwistedFate> would having glibc feel like tilting at windmills on kiss?
2020-01-08T11:37:42 #kisslinux <konimex> from kiss itself? you'll need a cross-compiler
2020-01-08T11:38:03 #kisslinux <konimex> you could probably bootstrap kiss-repo (replacing musl with glibc) from another distro, you'll need a relatively minimal distro though
2020-01-08T11:38:10 #kisslinux <konimex> (i.e. not arch)
2020-01-08T11:38:39 #kisslinux <TwistedFate> adamantium: adam!!
2020-01-08T11:38:42 #kisslinux <TwistedFate> hi!
2020-01-08T11:46:12 #kisslinux <adamantium> https://github.com/a-schaefers/kiss-overlay/blob/master/boot.el/files/boot.el
2020-01-08T11:46:18 #kisslinux <adamantium> TwistedFate: !!!!! You joining us?
2020-01-08T11:46:40 #kisslinux <TwistedFate> i-ii don't know yet :3
2020-01-08T11:46:43 #kisslinux <adamantium> cool
2020-01-08T11:46:51 #kisslinux <TwistedFate> because muh games :/
2020-01-08T11:46:51 #kisslinux <adamantium> give it a whirl see what you think
2020-01-08T11:47:04 #kisslinux <adamantium> i am still rewriting init in emacs lisp
2020-01-08T11:47:09 #kisslinux <adamantium> brb
2020-01-08T12:12:25 #kisslinux <icyphox> i did an Alpine install -- drivers seem to load okay
2020-01-08T12:12:45 #kisslinux <icyphox> i'll copy over that config and try KISS again
2020-01-08T12:36:07 #kisslinux <lieuxnoir> icyphox: might also help
2020-01-08T12:36:08 #kisslinux <lieuxnoir> https://wiki.gentoo.org/wiki/Linux_firmware
2020-01-08T12:55:29 #kisslinux <icyphox> thanks lieuxnoir i've tried all of that :)
2020-01-08T12:55:43 #kisslinux <icyphox> funnily enough, the same config works fine on Alpine
2020-01-08T12:55:59 #kisslinux <icyphox> it's on KISS that shit hits the fan
2020-01-08T13:03:56 #kisslinux <lieuxnoir> how'd you copy the config on Alpine?
2020-01-08T13:05:41 #kisslinux <adamantium> You might ask in #alpine, but IIRC there is a way to get a kconfig directly from inspecting the file in /boot if the kernel is built with that feature
2020-01-08T13:06:40 #kisslinux <adamantium> The thing is we don't know what patches they may have applied to that kernel too. Not sure if it will work to just take all of the options they have and try to build a vanilla kernel
2020-01-08T13:06:44 #kisslinux <adamantium> But you could try
2020-01-08T13:06:52 #kisslinux <lieuxnoir> mine was taken from Arch through localyesconfig
2020-01-08T13:07:55 #kisslinux <lieuxnoir> then i just strip it from time to time
2020-01-08T13:07:58 #kisslinux <adamantium> icyphox: you did try simply make defconfig before right
2020-01-08T13:08:32 #kisslinux <icyphox> lieuxnoir: the config-lts file was in /boot
2020-01-08T13:09:26 #kisslinux <icyphox> oh my bad, it's not the "same" config, but froma quick glance it looked about the same as my KISS config
2020-01-08T13:09:37 #kisslinux <icyphox> adamantium: yes
2020-01-08T13:09:48 #kisslinux <lieuxnoir> icyphox: did you modify it before compiling it on KISS?
2020-01-08T13:09:58 #kisslinux <icyphox> modify the config?
2020-01-08T13:09:58 #kisslinux <icyphox> ofc
2020-01-08T13:11:06 #kisslinux <lieuxnoir> what was the problem?
2020-01-08T13:11:26 #kisslinux <lieuxnoir> mine had everything baked in to save me from trouble
2020-01-08T13:11:49 #kisslinux <lieuxnoir> including the extra firmware files which wasn't in the kernel
2020-01-08T13:13:44 #kisslinux <icyphox> yeah i tried that
2020-01-08T13:14:04 #kisslinux <icyphox> it still complains about "no valid firmware found!" or something to that effect
2020-01-08T13:14:52 #kisslinux <lieuxnoir> maybe try another config from another distro, perhaps?
2020-01-08T13:23:23 #kisslinux <icyphox> lieuxnoir: yeah i'm going to do that
2020-01-08T13:23:36 #kisslinux <icyphox> btw, how do i uninstall grub
2020-01-08T13:23:53 #kisslinux <lieuxnoir> `kiss r grub`?
2020-01-08T13:23:58 #kisslinux <icyphox> i want to just use efistub
2020-01-08T13:24:16 #kisslinux <icyphox> no, but the /boot/grub folder remains
2020-01-08T13:24:22 #kisslinux <icyphox> and it still boots into grub
2020-01-08T13:24:25 #kisslinux <lieuxnoir> yeah, i did that days ago
2020-01-08T13:24:28 #kisslinux <icyphox> also, i'm on alpine
2020-01-08T13:24:46 #kisslinux <lieuxnoir> remove the boot entry manually
2020-01-08T13:24:53 #kisslinux <icyphox> using?
2020-01-08T13:25:05 #kisslinux <lieuxnoir> `efibootmgr -Bb 00something`
2020-01-08T13:25:11 #kisslinux <icyphox> ah
2020-01-08T13:25:45 #kisslinux <icyphox> hmm
2020-01-08T13:25:55 #kisslinux <icyphox> err
2020-01-08T13:26:08 #kisslinux <icyphox> i've got 3 entries that sound the same
2020-01-08T13:26:30 #kisslinux <icyphox> EFI Hard Drive, Notebook Hard Drive and Internal Hard Disk
2020-01-08T13:26:31 #kisslinux <icyphox> lol
2020-01-08T13:27:30 #kisslinux <lieuxnoir> no entry named GRUB or something?
2020-01-08T13:27:35 #kisslinux <icyphox> nothing
2020-01-08T13:27:39 #kisslinux <lieuxnoir> what's the verbose output looks like?
2020-01-08T13:27:51 #kisslinux <icyphox> verbose is bugged
2020-01-08T13:28:13 #kisslinux <lieuxnoir> oh
2020-01-08T13:28:28 #kisslinux <lieuxnoir> i can do nothing about that
2020-01-08T13:28:35 #kisslinux <icyphox> yeah
2020-01-08T13:28:36 #kisslinux <icyphox> lol
2020-01-08T13:29:01 #kisslinux <lieuxnoir> arch still uses an older efibootmgr which still works fine
2020-01-08T13:31:31 #kisslinux <Aarg> those entries are added automatically, your bios/efi does it icyphox
2020-01-08T13:31:48 #kisslinux <lieuxnoir> though you can patch and build it yourself manually
2020-01-08T13:33:08 #kisslinux <Aarg> you could delete them but there is no point really
2020-01-08T13:34:45 #kisslinux <lieuxnoir> you could also check if your device has good UEFI implementation
2020-01-08T13:35:17 #kisslinux <lieuxnoir> i've encountered laptops before which doesn't respect boot entries made with `efibootmgr`
2020-01-08T13:35:25 #kisslinux <lieuxnoir> they won't appear on the BIOS lol
2020-01-08T14:06:51 #kisslinux <icyphox> lieuxnoir: no my system's UEFI behaves alright haha
2020-01-08T14:07:16 #kisslinux <icyphox> Aarg: yeah, i thought as much. so deleting them should be alright?
2020-01-08T14:08:43 #kisslinux <lieuxnoir> icyphox: then the system booting GRUB questions me
2020-01-08T14:10:32 #kisslinux <lieuxnoir> probably, it just auto-detects it inside your Internal Hard Disk
2020-01-08T14:12:03 #kisslinux <icyphox> hm
2020-01-08T14:12:05 #kisslinux <icyphox> btw
2020-01-08T14:12:13 #kisslinux <icyphox> blkid isn't showing me a PARTUUIF
2020-01-08T14:12:18 #kisslinux <icyphox> PARTUUID*
2020-01-08T14:12:45 #kisslinux <icyphox> what do i use in efibootmgr?
2020-01-08T14:14:28 #kisslinux <adamantium> if blkid isn't showing you a PARTUUID then you need to create a partition and install a filesystem on to it
2020-01-08T14:14:48 #kisslinux <adamantium> cfdisk /dev/sda if using bios
2020-01-08T14:14:58 #kisslinux <adamantium> cgdisk /dev/sda if using uefi
2020-01-08T14:15:10 #kisslinux <adamantium> MBR if bios, GPT if uefi
2020-01-08T14:15:22 #kisslinux <adamantium> create one partition write quit
2020-01-08T14:15:32 #kisslinux <icyphox> that's mighty weird. i dl have a partition with an fs
2020-01-08T14:15:35 #kisslinux <icyphox> do*
2020-01-08T14:15:36 #kisslinux <adamantium> then mkfs.ext4 /dev/sda1
2020-01-08T14:15:47 #kisslinux <adamantium> then mount /dev/sda1 /mnt
2020-01-08T14:15:50 #kisslinux <adamantium> and get to work
2020-01-08T14:15:51 #kisslinux <icyphox> alpine's setup script created these
2020-01-08T14:16:04 #kisslinux <adamantium> i dont know alpine
2020-01-08T14:16:14 #kisslinux <adamantium> this is #kisslinux dude lol
2020-01-08T14:16:50 #kisslinux <icyphox> well yeah okay
2020-01-08T14:16:54 #kisslinux <adamantium> Okay, I'm sorry, I know you're trying to get kiss to just work lol
2020-01-08T14:17:03 #kisslinux <adamantium> So what are you trying to do exactly at this point
2020-01-08T14:18:41 #kisslinux <icyphox> nvm i don't need to bother with efibootmgr right now
2020-01-08T14:19:07 #kisslinux <icyphox> i'll try copying over the kernel config and doing KISS again
2020-01-08T14:22:34 #kisslinux <konimex> if GPT, the ESP should be fat32, not ext4
2020-01-08T14:22:56 #kisslinux <adamantium> Yeah
2020-01-08T14:23:06 #kisslinux <adamantium> Avoid uefi if at all possible man
2020-01-08T14:23:15 #kisslinux <adamantium> most bios have a toggle back to legacy option
2020-01-08T14:23:23 #kisslinux <adamantium> Keeps your system so much simpler
2020-01-08T14:25:03 #kisslinux <adamantium> hey konimex wanna try my systemE overlay
2020-01-08T14:25:28 #kisslinux <adamantium> I got a working elisp rc.boot script lmao
2020-01-08T14:25:30 #kisslinux <konimex> I already have an llvm overlay on test so for now, nope
2020-01-08T14:25:55 #kisslinux <konimex> also not really a fan of emacs so there's that
2020-01-08T14:25:59 #kisslinux <adamantium> lol
2020-01-08T14:26:32 #kisslinux <adamantium> I understand. It's just hillarious to me, the whole thing
2020-01-08T14:27:48 #kisslinux <adamantium> I use Emacs, not a fan, I want a better Emacs, but can't find one.
2020-01-08T14:27:54 #kisslinux <adamantium> There are many Emacses, in theory.
2020-01-08T14:28:04 #kisslinux <adamantium> So I'm stuck with gnu Emacs
2020-01-08T14:28:55 #kisslinux <adamantium> I learned it on evil mode coming from vim, but eventually found I don't like modal editing as much I thought, and use default now, though I swap control and alt on the left side of the keyboard, makes Emacs much more ergonomic that way.
2020-01-08T14:35:00 #kisslinux <Aarg[m]> icyphox: you probably can delete all entries but it's your risk
2020-01-08T14:35:22 #kisslinux <polymer> omg this irc client birch do be kinda ontop
2020-01-08T14:35:29 #kisslinux <Aarg[m]> I just deleted all the entries I had and some of them got added again
2020-01-08T14:35:31 #kisslinux <polymer> as in very ontop
2020-01-08T14:37:20 #kisslinux <icyphox> adamantium: oh, i thought legacy was trash and UEFI was the new cool
2020-01-08T14:37:24 #kisslinux <icyphox> ok i'll revert
2020-01-08T14:37:37 #kisslinux <Aarg[m]> Regarding UUID, since it's just the commandline handed to the kernel, maybe other properties work too
2020-01-08T14:37:42 #kisslinux <Aarg[m]> Maybe even labels
2020-01-08T14:37:52 #kisslinux <icyphox> Aarg[m]: yeah they keep popping back
2020-01-08T14:38:31 #kisslinux <Aarg[m]> Ah
2020-01-08T14:46:35 #kisslinux <lieuxnoir> you can just use root=/dev/sdX
2020-01-08T14:46:47 #kisslinux <lieuxnoir> icyphox
2020-01-08T14:47:48 #kisslinux <lieuxnoir> root=LABEL=<partition label> works too
2020-01-08T15:04:42 #kisslinux <dylanaraps> TwistedFate: No poettering ever. KISS today is what it'll always be (No moving to Wayland etc as an example)
2020-01-08T15:04:53 #kisslinux <dylanaraps> It's one of the reasons why I made KISS. ;)
2020-01-08T15:05:08 #kisslinux <dylanaraps> icyphox: Update efivar, I should have fixed the bugged output.
2020-01-08T15:05:23 #kisslinux <dylanaraps> It was compiler issues in efivar causing issues in efibootmgr.
2020-01-08T15:07:14 #kisslinux <dylanaraps> polymer: ?!?!?!?
2020-01-08T15:10:25 #kisslinux <dylanaraps> https://www.phoronix.com/scan.php?page=news_item&px=Systemd-Facebook-OOMD
2020-01-08T15:10:25 #kisslinux <dylanaraps> Oh boy. More Facebook
2020-01-08T15:11:31 #kisslinux <icyphox> dylanaraps: cool! i'm starting a KISS install fresh. going to use Alpine's kernel config.
2020-01-08T15:11:36 #kisslinux <icyphox> because that worked.
2020-01-08T15:11:57 #kisslinux <dylanaraps> Sounds good :)
2020-01-08T15:12:55 #kisslinux <dylanaraps> Once you get a working config you can go back and trim the fat too.
2020-01-08T15:13:23 #kisslinux <dylanaraps> (Disabling anything you don't need per se)
2020-01-08T15:13:51 #kisslinux <icyphox> yep
2020-01-08T15:15:53 #kisslinux <dzove855> No one asked lennart to do all this :(
2020-01-08T15:16:24 #kisslinux <dylanaraps> His employer did. ;)
2020-01-08T15:17:28 #kisslinux <dzove855> his working at redhat right?
2020-01-08T15:19:46 #kisslinux <icyphox> is this just a "systemd sucks" discussion or did lennart do something
2020-01-08T15:19:51 #kisslinux <icyphox> i'm ootl
2020-01-08T15:24:37 #kisslinux <Aarg> Boot0000* KISS HD(1,GPT,7be111b7-e41a-452d-9c58-c042c54e85d0,0x800,0xfa000)/File(vmlinuz)r.o.o.t.=.P.A.R.T.U.U.I.D.=.6.7.2.c.7.7.9.1.-.f.3...r.w.4.2.9.b.-.a.1.3.c.-.1.6.3.b.0.8.b.e.c.5.4.7
2020-01-08T15:24:49 #kisslinux <Aarg> see the rw just in the middle of the UUID?
2020-01-08T15:25:11 #kisslinux <Aarg> my /proc/cmdline is also truncated where the rw starts
2020-01-08T15:25:14 #kisslinux <Aarg> so weird
2020-01-08T16:19:16 #kisslinux <pulga> Hi guys
2020-01-08T16:20:05 #kisslinux <pulga> I have a issue when compiling the Linux kernel on my amd machine
2020-01-08T16:22:20 #kisslinux <adamantium> hey guys, i know Emacs is old bloat but if you want to follow my work of rolling my own kiss-init using Emacs lisp and emacs I now have a repo for that
2020-01-08T16:22:30 #kisslinux <adamantium> https://github.com/a-schaefers/systemE
2020-01-08T16:25:08 #kisslinux <pulga> When i run the make i get this error
2020-01-08T16:25:12 #kisslinux <pulga> kernel/exit.o: warning: objtool: __x64_sys_exit_group()+0x14: unreachable instruction
2020-01-08T16:26:15 #kisslinux <dylanaraps> pulga: Weird.
2020-01-08T16:28:27 #kisslinux <dylanaraps> Any settings (gcc, binutils, etc) changed?
2020-01-08T16:28:43 #kisslinux <dylanaraps> Did you change your CFLAGS, LDFLAGS etc too?
2020-01-08T16:29:35 #kisslinux <dylanaraps> Do you have: "[ ] Compile-time stack metadata validation" enabled?
2020-01-08T16:29:53 #kisslinux <dylanaraps> Where did you get your .config from? Did you create it yourself?
2020-01-08T16:31:20 #kisslinux <pulga> Is the Slackware huge config file
2020-01-08T16:31:52 #kisslinux <pulga> I not changed the cflags
2020-01-08T16:32:23 #kisslinux <dylanaraps> It's an issue with the config.
2020-01-08T16:32:31 #kisslinux <dylanaraps> Specifically something it enables.
2020-01-08T16:33:15 #kisslinux <pulga> I see
2020-01-08T16:33:31 #kisslinux <pulga> I Will check até nigth for the compile time option
2020-01-08T16:33:37 #kisslinux <pulga> At*
2020-01-08T16:33:43 #kisslinux <dylanaraps> I'd do 'make defconfig' and 'make menuconfig', enabling anything needed for your hardware.
2020-01-08T16:33:51 #kisslinux <dylanaraps> (rm the .config first)
2020-01-08T16:34:10 #kisslinux <dylanaraps> You can find the required options for your hardware online too.
2020-01-08T16:34:31 #kisslinux <dylanaraps> Remember: The drivers needed to solely to *boot* the machine need to be
2020-01-08T16:34:38 #kisslinux <dylanaraps> '*' and not 'm'
2020-01-08T16:34:55 #kisslinux <dylanaraps> This includes filesystem drivers, disk controller drivers etc.
2020-01-08T16:35:08 #kisslinux <dylanaraps> s/needed to/needed/
2020-01-08T16:35:34 #kisslinux <dylanaraps> '*' translates to 'y' in the .config file itself. In 'menuconfig' they're '*'.
2020-01-08T16:36:40 #kisslinux <dylanaraps> Also see: https://github.com/torvalds/linux/blob/master/tools/objtool/Documentation/stack-validation.txt#L205
2020-01-08T16:37:23 #kisslinux <dylanaraps> The error happens due to stack validation being enabled in Slackware's .config.
2020-01-08T16:40:52 #kisslinux <pulga> Thanks for the gelo dylanaraps
2020-01-08T16:41:17 #kisslinux <dylanaraps> np
2020-01-08T16:46:42 #kisslinux <Aarg> dylanaraps: https://gitlab.alpinelinux.org/alpine/aports/issues/7374
2020-01-08T16:46:56 #kisslinux <Aarg> maybe you want to consider this for the kiss baselayout (/etc/profile)
2020-01-08T16:52:27 #kisslinux <dylanaraps> Aarg: I don't know if this will make a difference with any software.
2020-01-08T16:53:53 #kisslinux <dylanaraps> I don't know if anything is locale aware in KISS as we don't use intltool/gettext/icu/etc.
2020-01-08T16:55:35 #kisslinux <dylanaraps> Do you have any use for it yourself or cases where setting '$LANG' fixed an issue?
2020-01-08T16:56:06 #kisslinux <Aarg> yeah, tmux needs LANG
2020-01-08T16:56:12 #kisslinux <Aarg> that's why I discovered it
2020-01-08T16:57:10 #kisslinux <Aarg> without it, tmux will only display ascii(?)
2020-01-08T17:10:30 #kisslinux <pulgovisk> Musl support some kind of locale defs?
2020-01-08T17:27:34 #kisslinux <dylanaraps> 'tmux -u' and there's a config option as well for utf8.
2020-01-08T17:27:52 #kisslinux <dylanaraps> pulgovisk: Yes, partially.
2020-01-08T17:28:11 #kisslinux <dylanaraps> Aarg: There's no harm in setting it regardless though. :P
2020-01-08T17:28:18 #kisslinux <dylanaraps> Will fix it later today.
2020-01-08T17:28:59 #kisslinux <Aarg> yeah, that was the idea, might as well add it
2020-01-08T18:53:39 #kisslinux <adamantium> dylanaraps: I heard the way we "seed random" is not actually doing anything
2020-01-08T18:54:24 #kisslinux <adamantium> cat /var/random.seed > /dev/urandom
2020-01-08T18:54:35 #kisslinux <adamantium> Anyone know if this actually works?/
2020-01-08T19:34:33 #kisslinux <dylanaraps> You can try 'cat /proc/sys/kernel/random/entropy_avail' right after boot.
2020-01-08T19:34:51 #kisslinux <dylanaraps> > It is also possible to write to /dev/random. This allows any user to mix random data into the pool.
2020-01-08T19:34:56 #kisslinux <dylanaraps> https://en.wikipedia.org/wiki//dev/random
2020-01-08T19:35:36 #kisslinux <dylanaraps> I don't know if this also applies to /dev/urandom.
2020-01-08T19:35:41 #kisslinux <dylanaraps> I'll look for a source on this.
2020-01-08T19:37:39 #kisslinux <dylanaraps> > Writing to /dev/random or /dev/urandom will update the entropy pool with the data written, but this will not result in a higher entropy count. This means that it will impact the contents read from both files, but it will not make reads from /dev/random faster.
2020-01-08T19:37:53 #kisslinux <dylanaraps> > ... This differs from writing to /dev/random or /dev/urandom, which only adds some data but does not increment the entropy count.
2020-01-08T19:37:57 #kisslinux <dylanaraps> https://linux.die.net/man/4/urandom
2020-01-08T19:38:46 #kisslinux <dylanaraps> Also, this page shows how to seed the generator and it is exactly the same method we use.
2020-01-08T19:40:02 #kisslinux <dylanaraps> (Though we hardcode the pool size)
2020-01-08T19:40:45 #kisslinux <dylanaraps> adamantium:
2020-01-08T19:42:10 #kisslinux <dylanaraps> brb
2020-01-08T19:44:51 #kisslinux <dylanaraps> If you're wondering why the pool size is hardcoded: https://wiki.archlinux.org/index.php/Random_number_generation
2020-01-08T19:45:10 #kisslinux <dylanaraps> > While Linux kernel 2.4 did have writable /proc entries for controlling the entropy pool size, in newer kernels only read_wakeup_threshold and write_wakeup_threshold are writable. The pool size is now hardcoded in kernel line 275 of /drivers/char/random.c
2020-01-08T19:45:42 #kisslinux <dylanaraps> > The kernel's pool size is given by INPUT_POOL_WORDS * OUTPUT_POOL_WORDS which makes, as already stated, 4096 bits.
2020-01-08T19:45:55 #kisslinux <dylanaraps> 4096 bits = 512 bytes (What we set it to)
2020-01-08T19:48:32 #kisslinux <dylanaraps> We use /dev/urandom over /dev/random so as to not block during boots. It was announced today that /dev/random in upcoming kernels won't block either so we'll eventually swap to it.
2020-01-08T19:49:01 #kisslinux <dylanaraps> The kernel for 4 years has actually seeded /dev/urandom itself in addition to our seeding: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e192be9d9a30555aae2ca1dc3aad37cba484cd4a
2020-01-08T19:49:49 #kisslinux <dylanaraps> This whole process is a fickle mess of conflicting and unclear information so if we are doing this wrongly I'd like to fix it.
2020-01-08T19:50:11 #kisslinux <dylanaraps> If you have any experience here it'd be appreciated. :)
2020-01-08T19:51:49 #kisslinux <dylanaraps> The kernel also warns about reading /dev/urandom without entropy. https://patchwork.kernel.org/patch/9173499/
2020-01-08T19:56:14 #kisslinux <adamantium> dylanaraps: I don't have a bouncer I missed your message when testing stuff
2020-01-08T19:56:23 #kisslinux <dylanaraps> https://freenode.logbot.info/kisslinux/20200108
2020-01-08T19:56:30 #kisslinux <dylanaraps> Logs are public :)
2020-01-08T19:56:38 #kisslinux <dylanaraps> Scroll down to the bottom.
2020-01-08T19:57:22 #kisslinux <adamantium> thanks
2020-01-08T20:00:18 #kisslinux <dylanaraps> Regarding: https://github.com/kisslinux/init/pull/5
2020-01-08T20:00:19 #kisslinux <adamantium> Got it. dylanaraps I no nothing about this subject. fickle mess it is, then.
2020-01-08T20:00:30 #kisslinux <adamantium> know*
2020-01-08T20:00:36 #kisslinux <dylanaraps> I'll boot my system with your changes for a little to see how it goes.
2020-01-08T20:00:49 #kisslinux <dylanaraps> Yup...
2020-01-08T20:01:06 #kisslinux <adamantium> k
2020-01-08T20:01:12 #kisslinux <adamantium> good luck and don't blame me if it explodes
2020-01-08T20:01:28 #kisslinux <dylanaraps> Where did you hear about our seeding process possibly not working?
2020-01-08T20:01:51 #kisslinux <adamantium> Just a person in #emacs who was looking at my elisp rewrite of it
2020-01-08T20:02:15 #kisslinux <adamantium> I think he is a knowledgable person, though, but from what you shared, I think he's probably wrong
2020-01-08T20:02:18 #kisslinux <dylanaraps> Systemd works the same more or less.
2020-01-08T20:02:20 #kisslinux <dylanaraps> https://github.com/systemd/systemd/blob/master/src/random-seed/random-seed.c
2020-01-08T20:02:25 #kisslinux <dylanaraps> It also writes to /dev/urandom.
2020-01-08T20:02:39 #kisslinux <adamantium> cool
2020-01-08T20:02:44 #kisslinux <dylanaraps> All non-systemd distributions do it using our method or similar.
2020-01-08T20:02:45 #kisslinux <adamantium> So does systemE
2020-01-08T20:02:52 #kisslinux <adamantium> =)
2020-01-08T20:03:13 #kisslinux <dylanaraps> heh
2020-01-08T20:03:24 #kisslinux <adamantium> BTW I think my Emacs init is faster than shell and i'm not sure how thats possible.
2020-01-08T20:03:31 #kisslinux <dylanaraps> lol
2020-01-08T20:03:35 #kisslinux <dylanaraps> Interesting
2020-01-08T20:03:47 #kisslinux <dylanaraps> It might be worth seeing if you do anything slightly differently.
2020-01-08T20:03:48 #kisslinux <adamantium> Could be in my head.
2020-01-08T20:04:04 #kisslinux <adamantium> I suppose i took a few shortcuts. That's what it is.
2020-01-08T20:04:13 #kisslinux <dylanaraps> I also run with a modified rc.boot :P
2020-01-08T20:04:14 #kisslinux <adamantium> Like I didn't mess with luks because I dont use that stuff
2020-01-08T20:04:21 #kisslinux <dylanaraps> That's a mess too.
2020-01-08T20:04:30 #kisslinux <dylanaraps> I had to write a parser for crypttab in POSIX shell lol
2020-01-08T20:06:19 #kisslinux <dylanaraps> OK. Modified my scripts.
2020-01-08T20:06:27 #kisslinux <dylanaraps> Will run with it for a few days.
2020-01-08T20:07:19 #kisslinux <E5ten> https://marc.info/?l=linux-netdev&m=157851343312263&w=2 sent a patch to make yacc usage in iproute2 POSIX compatible
2020-01-08T20:07:42 #kisslinux <adamantium> cool
2020-01-08T20:07:55 #kisslinux <dylanaraps> Nice
2020-01-08T20:07:59 #kisslinux <dylanaraps> Will they accept it though :(
2020-01-08T20:08:02 #kisslinux <dylanaraps> ?(
2020-01-08T20:08:03 #kisslinux <dylanaraps> ?*
2020-01-08T20:08:04 #kisslinux <E5ten> no idea
2020-01-08T20:08:19 #kisslinux <dylanaraps> It seems hard to get these specific fixes merged.
2020-01-08T20:08:28 #kisslinux <E5ten> if they don't I'll try with a less intrusive byacc compatibility patch
2020-01-08T20:09:44 #kisslinux <E5ten> I'm also not exactly 100% sure it's POSIX compatible, but it does only use POSIX flags and openbsd yacc handled it and I didn't notice anything left that seemed non-POSIX, but I also don't know yacc so idk
2020-01-08T20:10:15 #kisslinux <E5ten> not sure how much flex adds over POSIX lex but maybe it'd be worth attempting a followup POSIX lex compatibility patch if this one is accepted
2020-01-08T20:10:22 #kisslinux <dylanaraps> Yeah
2020-01-08T20:10:34 #kisslinux <dylanaraps> adamantium: Does suckless killall5 accept '-s'?
2020-01-08T20:10:59 #kisslinux <adamantium> let me see
2020-01-08T20:10:59 #kisslinux <E5ten> if suckless was actually suckless they wouldn't provide a killall5 ;)
2020-01-08T20:11:12 #kisslinux <dylanaraps> It's so handy for boot scripts though
2020-01-08T20:11:54 #kisslinux <dylanaraps> > goto do_it_now;
2020-01-08T20:11:59 #kisslinux <dylanaraps> 10/10 code in busybox
2020-01-08T20:12:01 #kisslinux <adamantium> dylanaraps: usage: killall5 [-o pid1,pid2,..,pidN] [-s signal]
2020-01-08T20:12:12 #kisslinux <E5ten> god busybox code is so gross
2020-01-08T20:12:13 #kisslinux <dylanaraps> Also: do_it_sooner:
2020-01-08T20:12:29 #kisslinux <dylanaraps> do_it_sooner is before do_it_now
2020-01-08T20:12:33 #kisslinux <dylanaraps> 10/10 code here
2020-01-08T20:12:41 #kisslinux <dylanaraps> adamantium: Thanks
2020-01-08T20:12:51 #kisslinux <dylanaraps> We need to swap to '-s TERM' and '-s KILL'.
2020-01-08T20:12:52 #kisslinux <E5ten> looking at toybox code when I basically don't even know C and it still all makes sense and then I try looking at busybox code and my mind completely melts
2020-01-08T20:13:00 #kisslinux <adamantium> dylanaraps: k
2020-01-08T20:13:04 #kisslinux <dylanaraps> Busybox needs '-TERM' and '-KILL'.
2020-01-08T20:13:13 #kisslinux <dylanaraps> They both accept '-s TERM' etc though.
2020-01-08T20:13:19 #kisslinux <adamantium> Nice catch
2020-01-08T20:14:01 #kisslinux <dylanaraps> busybox kill.c handles kill, killall and killall5.
2020-01-08T20:14:03 #kisslinux <dylanaraps> Hence the mess.
2020-01-08T20:14:20 #kisslinux <dylanaraps> kill is a shell builtin as well.
2020-01-08T20:14:22 #kisslinux <dylanaraps> heh
2020-01-08T20:14:25 #kisslinux <adamantium> it's done
2020-01-08T20:14:54 #kisslinux <dylanaraps> Thanks
2020-01-08T20:15:57 #kisslinux <dylanaraps> I might port killall5 to pure shell fyi.
2020-01-08T20:16:03 #kisslinux <adamantium> cool
2020-01-08T20:16:06 #kisslinux <dylanaraps> It's simply a loop over /proc.
2020-01-08T20:16:11 #kisslinux <adamantium> I can't stand shell you're a mad man
2020-01-08T20:16:31 #kisslinux <dylanaraps> It's my favourite language for some reason
2020-01-08T20:16:34 #kisslinux <E5ten> shell is the bexst
2020-01-08T20:17:02 #kisslinux <adamantium> you wrote wal so it better be your favorite. why did you make pywal after that? Then you decided you like shell better I take it?
2020-01-08T20:17:06 #kisslinux <E5ten> nothing provides a thrill quite like doing a hacky bullshit thing in shell that works
2020-01-08T20:17:47 #kisslinux <adamantium> I think shell with outsourcing expensive things to awk can be very powerful
2020-01-08T20:17:57 #kisslinux <adamantium> But I hate debugging it and there's just too many quirks
2020-01-08T20:19:14 #kisslinux <dylanaraps> I wanted to learn Python I guess.
2020-01-08T20:19:22 #kisslinux <dylanaraps> Now I know that I hate Python.
2020-01-08T20:20:08 #kisslinux <dylanaraps> You've probably never seen shell like this: https://termbin.com/o2mcx
2020-01-08T20:20:13 #kisslinux <dylanaraps> Heh. Was fun to write.
2020-01-08T20:24:37 #kisslinux <adamantium> mystifying
2020-01-08T20:25:03 #kisslinux <dylanaraps> Basically you can chain expressions inside (( )) using a comma.
2020-01-08T20:25:12 #kisslinux <dylanaraps> You can also use (( )) as the function body.
2020-01-08T20:25:42 #kisslinux <dylanaraps> Then, 'expression ? true : false' can be chained together for conditions.
2020-01-08T20:27:06 #kisslinux <dylanaraps> No loops though :(
2020-01-08T20:27:33 #kisslinux <dylanaraps> You *could* loop by embedding one with $() though you'd have to print an integer (and only an integer))
2020-01-08T20:28:51 #kisslinux <adamantium> I suppose your love for shell script could be like my love for emacs lisp
2020-01-08T20:28:58 #kisslinux <adamantium> both people say are terrible languages
2020-01-08T20:29:12 #kisslinux <adamantium> but so much fun
2020-01-08T20:29:15 #kisslinux <dylanaraps> Yup
2020-01-08T20:29:24 #kisslinux <dylanaraps> People just suck at both langs :^)
2020-01-08T20:33:01 #kisslinux <adamantium> dadayyyyum
2020-01-08T20:38:31 #kisslinux <adamantium> I would like to see another boot loader other than grub, what do you guys recommend. Is lilo still a thing
2020-01-08T20:38:37 #kisslinux <E5ten> ok after very brief looking into this
2020-01-08T20:38:41 #kisslinux <E5ten> a follow up POSIX lex patch will not be occuring
2020-01-08T20:38:45 #kisslinux <adamantium> or alpine uses syslinux ...
2020-01-08T20:39:23 #kisslinux <adamantium> or maybe the funtoo's "boot-update" + grub legacy
2020-01-08T20:39:31 #kisslinux <dylanaraps> lilo isn't in development but works.
2020-01-08T20:39:37 #kisslinux <dylanaraps> elilo exists also for EFI
2020-01-08T20:39:42 #kisslinux <adamantium> I'd like to see a "simple" alternative to grub is all ......
2020-01-08T20:39:56 #kisslinux <adamantium> grub generated code is nasty too, and bugs me
2020-01-08T20:40:14 #kisslinux <dylanaraps> Grub is just the default. There's endless documentation for it online and the single package supports both BIOS and UEFI.
2020-01-08T20:40:18 #kisslinux <adamantium> agreed
2020-01-08T20:40:18 #kisslinux <dylanaraps> All bootloaders suck tbh.
2020-01-08T20:40:21 #kisslinux <adamantium> Yep
2020-01-08T20:40:39 #kisslinux <adamantium> Emacs bootloader perhaps is due
2020-01-08T20:40:41 #kisslinux <dylanaraps> efistubs are the best option for those using EFI though.
2020-01-08T20:40:53 #kisslinux <dylanaraps> Only if they work with your motherboard though.
2020-01-08T20:40:53 #kisslinux <adamantium> Right
2020-01-08T20:40:59 #kisslinux <adamantium> I just hate uefi for so many reasons
2020-01-08T20:41:02 #kisslinux <dylanaraps> As do I
2020-01-08T20:41:02 #kisslinux <adamantium> 1. it's not robust
2020-01-08T20:41:07 #kisslinux <adamantium> You cannot raid your esp
2020-01-08T20:41:16 #kisslinux <adamantium> that is enough. it sucks then
2020-01-08T20:41:25 #kisslinux <adamantium> that's why it's not used in server space
2020-01-08T20:41:40 #kisslinux <dylanaraps> I use BIOS with a single partition for the whole system.
2020-01-08T20:41:41 #kisslinux <adamantium> 2. shitty uefi bios implementations.
2020-01-08T20:41:44 #kisslinux <dylanaraps> Literally just /
2020-01-08T20:42:44 #kisslinux <adamantium> 3. secure boot is complex as hell and worthless except to prevent evil maid??????? get a dog
2020-01-08T20:43:03 #kisslinux <dylanaraps> Yup
2020-01-08T20:43:04 #kisslinux <adamantium> dylanaraps: I have the same disk layout as you.
2020-01-08T20:43:09 #kisslinux <adamantium> lol
2020-01-08T20:43:14 #kisslinux <dylanaraps> It's simple :P
2020-01-08T20:43:17 #kisslinux <E5ten> I wasn't aware that there were motherboards that efistub doesn't work with?
2020-01-08T20:43:20 #kisslinux <adamantium> my security policy is encrypt files that should be encrypted
2020-01-08T20:43:27 #kisslinux <adamantium> gnupg is your friend
2020-01-08T20:43:33 #kisslinux <dylanaraps> A lot of motherboards have super shitty firmware.
2020-01-08T20:44:07 #kisslinux <dylanaraps> My old PC's wiped my added EFI entries every boot for example.
2020-01-08T20:44:09 #kisslinux <E5ten> but what about efistub would make it not work where other EFI bootloaders do on those motherboards with shitty firmware?
2020-01-08T20:44:30 #kisslinux <dylanaraps> Windows is their only friend
2020-01-08T20:44:37 #kisslinux <dylanaraps> Beats me really.
2020-01-08T20:44:49 #kisslinux <adamantium> bioses suck worse than boot loaders
2020-01-08T20:45:08 #kisslinux <dylanaraps> if not windows then rm
2020-01-08T20:45:13 #kisslinux <dylanaraps> Wouldn't surprise me lol
2020-01-08T20:45:58 #kisslinux <dylanaraps> We're talking shitty firmware which is unlikely to every see updates. So long as it boots Windows it's fine.
2020-01-08T20:46:01 #kisslinux <dylanaraps> s/every/ever
2020-01-08T20:52:07 #kisslinux <dylanaraps> https://termbin.com/7vtm
2020-01-08T20:52:21 #kisslinux <dylanaraps> This is killall5 in POSIX shell minus '-o'.
2020-01-08T20:52:28 #kisslinux <E5ten> about the byacc changes being hard to merge from before, I do have more hope for this one because it makes the usage fully POSIX compatible instead of going from supporting bison to bison and byacc but still nothing else outside of that
2020-01-08T20:52:41 #kisslinux <dylanaraps> Yeah
2020-01-08T20:53:00 #kisslinux <E5ten> since when can you set a positional parameter like that?
2020-01-08T20:53:11 #kisslinux <E5ten> I don't think it works
2020-01-08T20:53:22 #kisslinux <dylanaraps> You're right
2020-01-08T20:53:34 #kisslinux <dylanaraps> I haven't tested this yet for obvious reasons.
2020-01-08T20:53:55 #kisslinux <adamantium> test it inside of systemd-nspawn
2020-01-08T20:54:08 #kisslinux <E5ten> case "$1" in '') set -- TERM; esac
2020-01-08T20:54:08 #kisslinux <E5ten> ;)
2020-01-08T20:54:12 #kisslinux <dylanaraps> https://termbin.com/gc1l4
2020-01-08T20:54:23 #kisslinux <dylanaraps> Or I could use a variable. :P
2020-01-08T20:54:35 #kisslinux <dylanaraps> I'll test it to shutdown my system of course.
2020-01-08T20:54:41 #kisslinux <dylanaraps> This is how we do things here. :^)
2020-01-08T20:54:51 #kisslinux <E5ten> never, they're a sin 😏
2020-01-08T20:55:07 #kisslinux <dylanaraps> oops
2020-01-08T20:55:12 #kisslinux <dylanaraps> This is broken
2020-01-08T20:55:13 #kisslinux <dylanaraps> Found a bug
2020-01-08T20:55:24 #kisslinux <dylanaraps> '$pid' will equal '/proc/XXX' and not 'XXX.
2020-01-08T20:55:27 #kisslinux <dylanaraps> HEH
2020-01-08T20:56:50 #kisslinux <E5ten> is the -d check necessary? can't that glob only match directories?
2020-01-08T20:57:02 #kisslinux <dylanaraps> It's to check if the glob fails.
2020-01-08T20:57:12 #kisslinux <E5ten> ah right, forgot about that
2020-01-08T20:57:29 #kisslinux <dylanaraps> It's also needed now for other reasons.
2020-01-08T20:57:30 #kisslinux <dylanaraps> Hang on
2020-01-08T20:57:37 #kisslinux <dylanaraps> https://termbin.com/tbo8
2020-01-08T20:57:47 #kisslinux <dylanaraps> The glob is now simply '*' as I cd to /proc.
2020-01-08T20:57:55 #kisslinux <E5ten> to avoid having to run it each time the loop occurs maybe you could just check [ -d /proc/1 ] before the loop?
2020-01-08T20:58:30 #kisslinux <dylanaraps> Even simpler would be 'cd /proc || exit 0' at the start.
2020-01-08T20:58:55 #kisslinux <dylanaraps> If proc exists it must have contents no?
2020-01-08T20:59:21 #kisslinux <E5ten> I mean I wouldn't think that those checks are meant for systems where everything is correct?
2020-01-08T20:59:36 #kisslinux <E5ten> so if you can't assume /proc you probably can't assume it has subdirectories if it does exist
2020-01-08T20:59:43 #kisslinux <dylanaraps> Yu
2020-01-08T20:59:44 #kisslinux <dylanaraps> p
2020-01-08T21:00:16 #kisslinux <dylanaraps> https://termbin.com/s97h
2020-01-08T21:00:19 #kisslinux <dylanaraps> Now it's good.
2020-01-08T21:00:23 #kisslinux <dylanaraps> brb testing it
2020-01-08T21:00:37 #kisslinux <E5ten> also in current form someone can pass just the signal without a -, maybe remove the ##- from the reference to sig and assign sig like sig=${1:--TERM}
2020-01-08T21:00:56 #kisslinux <dylanaraps> That's the idea.
2020-01-08T21:01:08 #kisslinux <dylanaraps> You assign it without '-' though it has a check to strip it if provided.
2020-01-08T21:01:29 #kisslinux <E5ten> I thought the goal was to pass it using a - like you would with killall5?
2020-01-08T21:01:59 #kisslinux <dylanaraps> suckless killall5 doesn't support -SIG actually.
2020-01-08T21:02:05 #kisslinux <E5ten> oh
2020-01-08T21:02:06 #kisslinux <dylanaraps> '-s SIG' is the portable way apparently.
2020-01-08T21:02:20 #kisslinux <dylanaraps> busybox doesn't support SIG without '-'.
2020-01-08T21:02:39 #kisslinux <E5ten> but if either -s SIG or -SIG is requires isn't allowing just SIG wrong either way?
2020-01-08T21:02:42 #kisslinux <dylanaraps> This function is solely for internal use in our shutdown script so whatever is fine.
2020-01-08T21:02:51 #kisslinux <E5ten> fair enough
2020-01-08T21:04:37 #kisslinux <E5ten> I feel like the cd failing is more of an exit 1 kinda situation than an exit 0?
2020-01-08T21:06:29 #kisslinux <E5ten> according to POSIX only -s SIG is mandatory for kill, -SIG is XSI, so you should switch it to use -s SIG for kill
2020-01-08T21:06:44 #kisslinux <dylanaraps> I'm not even sure if shutdown should be aborted.
2020-01-08T21:06:45 #kisslinux <dylanaraps> Will do.
2020-01-08T21:07:39 #kisslinux <E5ten> oh I didn't even mean exit like quit the script, I assumed you'd change that to a return when you made it a function
2020-01-08T21:07:58 #kisslinux <dylanaraps> Yup
2020-01-08T21:08:01 #kisslinux <dylanaraps> My bad
2020-01-08T21:08:13 #kisslinux <E5ten> I just mean that for killall5 I'd expect not being able to cd to /proc would be a failure not a successful exit
2020-01-08T21:08:29 #kisslinux <dylanaraps> Problem is that the next command will just overwrite the exit status.
2020-01-08T21:08:41 #kisslinux <dylanaraps> I'll make it output a warning and continue.
2020-01-08T21:09:39 #kisslinux <E5ten> also, maybe for the loop you can do for pid in / and then pid=${pid%/} and then when you do the case add an option to check if it's ''
2020-01-08T21:09:48 #kisslinux <E5ten> * also, maybe for the loop you can do for pid in */ and then pid=${pid%/} and then when you do the case add an option to check if it's '*'
2020-01-08T21:09:55 #kisslinux <E5ten> * also, maybe for the loop you can do for pid in */ and then pid=${pid%/} and then when you do the case add an option to check if it's '*'
2020-01-08T21:10:07 #kisslinux <E5ten> I can't figure out how to make this formatting not fuck itslef
2020-01-08T21:10:11 #kisslinux <E5ten> itself*
2020-01-08T21:10:12 #kisslinux <adamantium> dylanaraps: speaking of killall5 ... in my shutdown.el script, I can't get past killall5 -s TERM , like the script just stops running
2020-01-08T21:10:18 #kisslinux <adamantium> am i killing my shutdown script
2020-01-08T21:10:19 #kisslinux <adamantium> lol
2020-01-08T21:10:25 #kisslinux <dylanaraps> Possibly lol
2020-01-08T21:10:27 #kisslinux <adamantium> wtf
2020-01-08T21:10:35 #kisslinux <dylanaraps> Try making it print after killall5
2020-01-08T21:10:36 #kisslinux <E5ten> point is add a / to the glob and then strip it, and use the case to make sure the glob isn't a *, and then you can remove the -d check
2020-01-08T21:10:43 #kisslinux <adamantium> dylanaraps: I did try that
2020-01-08T21:10:45 #kisslinux <adamantium> It wont print
2020-01-08T21:10:58 #kisslinux <adamantium> See emacs spawns proccesses
2020-01-08T21:11:00 #kisslinux <dylanaraps> It's being killed then
2020-01-08T21:11:01 #kisslinux <dylanaraps> lol
2020-01-08T21:11:18 #kisslinux <adamantium> grrrr
2020-01-08T21:11:22 #kisslinux <adamantium> I will find a way
2020-01-08T21:11:35 #kisslinux <dylanaraps> You can use '-o PID' to tell it to ignore Emacs' process group.
2020-01-08T21:12:00 #kisslinux <dylanaraps> E5ten: Will do
2020-01-08T21:12:10 #kisslinux <E5ten> also what's -1?
2020-01-08T21:12:26 #kisslinux <dylanaraps> All processes which aren't 1.
2020-01-08T21:12:27 #kisslinux <adamantium> I should be able to call killall5 in a way that will not kill Emacs
2020-01-08T21:12:32 #kisslinux <dylanaraps> Yes
2020-01-08T21:12:34 #kisslinux <E5ten> ah, cool
2020-01-08T21:12:37 #kisslinux <adamantium> if shell can do it
2020-01-08T21:12:46 #kisslinux <dylanaraps> See my message above adamantium
2020-01-08T21:12:54 #kisslinux <dylanaraps> > You can use '-o PID' to tell it to ignore Emacs' process group.
2020-01-08T21:13:10 #kisslinux <adamantium> killall5 has an -o ?
2020-01-08T21:13:16 #kisslinux <adamantium> oh ok i see it
2020-01-08T21:13:37 #kisslinux <adamantium> thx
2020-01-08T21:13:40 #kisslinux <adamantium> might have to do that
2020-01-08T21:13:43 #kisslinux <E5ten> http://ix.io/26SO
2020-01-08T21:13:55 #kisslinux <dylanaraps> oof
2020-01-08T21:14:17 #kisslinux <dylanaraps> E5ten: I need to also ensure that the glob matches only integers.
2020-01-08T21:14:20 #kisslinux <dylanaraps> That was close
2020-01-08T21:14:26 #kisslinux <dylanaraps> ie /proc/tty exists.
2020-01-08T21:14:49 #kisslinux <dylanaraps> Heh
2020-01-08T21:14:50 #kisslinux <dylanaraps> Fixed
2020-01-08T21:14:52 #kisslinux <E5ten> [!0-9] in case
2020-01-08T21:15:17 #kisslinux <dylanaraps> I'm doing
2020-01-08T21:15:24 #kisslinux <dylanaraps> [0-9]*/
2020-01-08T21:15:48 #kisslinux <E5ten> but then a hypothetical dir like 1tty could exist lol
2020-01-08T21:15:57 #kisslinux <dylanaraps> I'm aware :^)
2020-01-08T21:15:58 #kisslinux <dylanaraps> OK
2020-01-08T21:16:06 #kisslinux <dylanaraps> Needs to be more r o b u s t
2020-01-08T21:16:17 #kisslinux <E5ten> http://ix.io/26SP
2020-01-08T21:16:42 #kisslinux <E5ten> oh I guess that can have the check for specifically * removed
2020-01-08T21:16:46 #kisslinux <E5ten> it'd be matched by the last thing
2020-01-08T21:16:54 #kisslinux <dylanaraps> Nice
2020-01-08T21:17:08 #kisslinux <dylanaraps> OK
2020-01-08T21:17:10 #kisslinux <dylanaraps> Now it's done.
2020-01-08T21:17:12 #kisslinux <dylanaraps> brb
2020-01-08T21:18:39 #kisslinux <E5ten> it'd be cool if POSIX guaranteed kill as a builtin so this could be 100% pure shell in any POSIX compliant shell lol
2020-01-08T21:18:48 #kisslinux <dylanaraps> It does doesn't it?
2020-01-08T21:18:51 #kisslinux <dylanaraps> OK
2020-01-08T21:19:00 #kisslinux <dylanaraps> I think it kills the script itself.
2020-01-08T21:19:05 #kisslinux <E5ten> I don't think it's guaranteed to be built in
2020-01-08T21:20:09 #kisslinux <dylanaraps> https://termbin.com/ofm2
2020-01-08T21:20:25 #kisslinux <adamantium> (emacs-pid)
2020-01-08T21:20:27 #kisslinux <adamantium> built in!
2020-01-08T21:20:29 #kisslinux <dylanaraps> :
2020-01-08T21:20:30 #kisslinux <dylanaraps> :D
2020-01-08T21:20:37 #kisslinux <dylanaraps> It kills my script as well.
2020-01-08T21:20:38 #kisslinux <dylanaraps> HM
2020-01-08T21:20:47 #kisslinux <adamantium> lmao
2020-01-08T21:20:59 #kisslinux <adamantium> We literally have the same issue doing two diff things
2020-01-08T21:22:07 #kisslinux <adamantium> (shell-command (concat "ubase-box killall5 -o " (emacs-pid)" -s TERM"))
2020-01-08T21:22:10 #kisslinux <adamantium> wish me luck
2020-01-08T21:22:16 #kisslinux <dylanaraps> heh
2020-01-08T21:23:28 #kisslinux <dylanaraps> brb
2020-01-08T21:23:32 #kisslinux <dylanaraps> added a bunch of echos
2020-01-08T21:23:46 #kisslinux <adamantium> hmm i am still here.
2020-01-08T21:23:52 #kisslinux <adamantium> "wrong type argument"
2020-01-08T21:23:56 #kisslinux <dylanaraps> lol
2020-01-08T21:23:58 #kisslinux <adamantium> lol
2020-01-08T21:24:12 #kisslinux <dylanaraps> You have " ()"
2020-01-08T21:24:18 #kisslinux <dylanaraps> Space in argument?
2020-01-08T21:24:19 #kisslinux <dylanaraps> Oh
2020-01-08T21:24:24 #kisslinux <dylanaraps> nvm
2020-01-08T21:24:33 #kisslinux <adamantium> (shell-command (concat "ubase-box killall5 -o " (format "%s" (emacs-pid)) " -s TERM"))
2020-01-08T21:24:37 #kisslinux <dylanaraps> Ah
2020-01-08T21:24:38 #kisslinux <adamantium> emacs-pid is an int
2020-01-08T21:24:41 #kisslinux <adamantium> needs to be a string
2020-01-08T21:24:45 #kisslinux <adamantium> this should work now
2020-01-08T21:24:50 #kisslinux <dylanaraps> brb again
2020-01-08T21:28:16 #kisslinux <dylanaraps> https://termbin.com/8dvb
2020-01-08T21:28:23 #kisslinux <adamantium> Alright. -o was the ticket
2020-01-08T21:28:26 #kisslinux <dylanaraps> It runs
2020-01-08T21:28:51 #kisslinux <adamantium> heres mines https://termbin.com/lyjo
2020-01-08T21:29:03 #kisslinux <adamantium> longest shebang in history btw
2020-01-08T21:29:34 #kisslinux <dylanaraps> nice
2020-01-08T21:30:25 #kisslinux <adamantium> it only powers off, needs to reboot and take an argument now
2020-01-08T21:30:59 #kisslinux <dylanaraps> My killall5 runs fine but I see no output in the tty on shutdown.
2020-01-08T21:31:19 #kisslinux <dylanaraps> What's weird is that all the pids still exist in the second run of killall5.
2020-01-08T21:31:30 #kisslinux <dylanaraps> https://termbin.com/8dvb
2020-01-08T21:31:38 #kisslinux <dylanaraps> Notice how the list is duplicated.
2020-01-08T21:31:46 #kisslinux <dylanaraps> Second iteration is with KILL.
2020-01-08T21:32:01 #kisslinux <dylanaraps> First iteration should have killed things though.
2020-01-08T21:32:32 #kisslinux <dylanaraps> The glob is in a shitty order too.
2020-01-08T21:32:35 #kisslinux <dylanaraps> I need to fix that.
2020-01-08T21:32:48 #kisslinux <dylanaraps> It doesn't go in order of value.
2020-01-08T21:33:00 #kisslinux <dylanaraps> Goes in order of first digit.
2020-01-08T21:33:06 #kisslinux <E5ten> not sure that's really fixable?
2020-01-08T21:33:14 #kisslinux <dylanaraps> 200 > 99
2020-01-08T21:33:18 #kisslinux <E5ten> isn't that just like how globs work
2020-01-08T21:33:21 #kisslinux <dylanaraps> Yes
2020-01-08T21:33:40 #kisslinux <dylanaraps> Is it an issue though?
2020-01-08T21:33:41 #kisslinux <dylanaraps> Hm
2020-01-08T21:34:05 #kisslinux <dylanaraps> I'm still wondering why everything still exists in /proc after kill TERM.
2020-01-08T21:36:02 #kisslinux <dylanaraps> OH
2020-01-08T21:36:06 #kisslinux <dylanaraps> I need to skip 0 as well.
2020-01-08T21:36:26 #kisslinux <dylanaraps> 0 is the kernel
2020-01-08T21:36:56 #kisslinux <dylanaraps> Though 0 doesn't exist in /proc.
2020-01-08T21:37:12 #kisslinux <E5ten> so it's not an issue then is it?
2020-01-08T21:37:23 #kisslinux <dylanaraps> No
2020-01-08T21:37:38 #kisslinux <E5ten> but either way since the dirs in /proc don't have leading 0s the glob can be changed to 1-9
2020-01-08T21:37:54 #kisslinux <dylanaraps> Yeah
2020-01-08T21:38:00 #kisslinux <dylanaraps> I wonder if $PPID is needed too.
2020-01-08T21:38:07 #kisslinux <dylanaraps> $ is the shell itself.
2020-01-08T21:38:37 #kisslinux <E5ten> $PPID is only different from $ in a subshell right?
2020-01-08T21:39:21 #kisslinux <adamantium> Anyone use the wpa_supplicant hook feature of dhcpcd ?
2020-01-08T21:39:23 #kisslinux <E5ten> so if killall5_sh isn't running in a subshell $PPID shouldn't be necessary?
2020-01-08T21:39:26 #kisslinux <dylanaraps> It's needed.
2020-01-08T21:39:33 #kisslinux <dylanaraps> $PPID = the shell
2020-01-08T21:39:36 #kisslinux <dylanaraps> $ = the script
2020-01-08T21:39:52 #kisslinux <adamantium> I ask because ... I do, but wpa_supplicant is very noisy on my getty login / tty1 and it drives me crazy
2020-01-08T21:39:58 #kisslinux <adamantium> I want to make it shut up
2020-01-08T21:40:30 #kisslinux <dylanaraps> I use iwd personally
2020-01-08T21:41:00 #kisslinux <dylanaraps> When I was using wpa_supplicant/dhcpcd I spawned them both and they figured it out.
2020-01-08T21:41:11 #kisslinux <dylanaraps> Try searching around online?
2020-01-08T21:41:12 #kisslinux <adamantium> well it runs great
2020-01-08T21:41:15 #kisslinux <adamantium> i have no issues
2020-01-08T21:41:19 #kisslinux <dylanaraps> Yeah
2020-01-08T21:41:25 #kisslinux <adamantium> it just spams its' connecting information on tty1
2020-01-08T21:41:27 #kisslinux <adamantium> and i hate that
2020-01-08T21:41:32 #kisslinux <dylanaraps> You could put the first getty on tty2
2020-01-08T21:42:00 #kisslinux <adamantium> it spams the wifi informationn whereever getty you are looking at
2020-01-08T21:42:07 #kisslinux <adamantium> "authenticated"
2020-01-08T21:42:09 #kisslinux <adamantium> stuff like that
2020-01-08T21:42:18 #kisslinux <dylanaraps> Ah
2020-01-08T21:42:30 #kisslinux <adamantium> ""link becomes ready""
2020-01-08T21:42:33 #kisslinux <adamantium> "associated"
2020-01-08T21:42:38 #kisslinux <adamantium> like 10-20 lines every startup
2020-01-08T21:42:43 #kisslinux <dylanaraps> Try: tty1::respawn:/bin/login
2020-01-08T21:42:48 #kisslinux <dylanaraps> That's what I use.
2020-01-08T21:42:50 #kisslinux <adamantium> over the top of what you are reading
2020-01-08T21:43:07 #kisslinux <dylanaraps> This skips running a getty.
2020-01-08T21:43:16 #kisslinux <adamantium> okay, interesting!
2020-01-08T21:43:21 #kisslinux <adamantium> Something I have not tried.
2020-01-08T21:43:24 #kisslinux <dylanaraps> I run no gettys :P
2020-01-08T21:43:26 #kisslinux <adamantium> What are the differences
2020-01-08T21:43:58 #kisslinux <adamantium> it still asks your password i assume
2020-01-08T21:44:11 #kisslinux <adamantium> So what is the difference? I never did learn too much about a getty vs a login
2020-01-08T21:44:51 #kisslinux <dylanaraps> The getty runs login iirc.
2020-01-08T21:46:35 #kisslinux <dylanaraps> Basically, you skip opening a TTY.
2020-01-08T21:46:52 #kisslinux <dylanaraps> busybox getty does the following: Open TTY, prompt for login name, then invoke /bin/login
2020-01-08T21:47:07 #kisslinux <dylanaraps> (Though /bin/login prompts for login name anyway)
2020-01-08T21:48:28 #kisslinux <dylanaraps> brb
2020-01-08T21:55:12 #kisslinux <adamantium> i guess we dont need tty?
2020-01-08T21:55:15 #kisslinux <adamantium> weird
2020-01-08T21:58:57 #kisslinux <dylanaraps> OK
2020-01-08T21:58:59 #kisslinux <dylanaraps> Pushed.
2020-01-08T21:59:05 #kisslinux <dylanaraps> Turns out it actually works fine.
2020-01-08T21:59:39 #kisslinux <dylanaraps> Did 'ps >> file' for both invocations.
2020-01-08T21:59:47 #kisslinux <dylanaraps> s/ps/ps aux/
2020-01-08T22:00:00 #kisslinux <dylanaraps> https://github.com/kisslinux/init/commit/9a8a16ef6fcb253fa2fcb36e8a5c9e60b0213e28
2020-01-08T22:03:32 #kisslinux <dylanaraps> adamantium: One issue with the umount change is that it spits an error out for /dev.
2020-01-08T22:03:57 #kisslinux <dylanaraps> Basically, the previous command would skip /dev, /sys, /proc and /tmp.
2020-01-08T22:04:19 #kisslinux <dylanaraps> It's fine, the only issue is the message which could mislead.
2020-01-08T22:04:22 #kisslinux <adamantium> Right
2020-01-08T22:04:26 #kisslinux <adamantium> I believe it is fine
2020-01-08T22:04:55 #kisslinux <adamantium> For example we also try to mount dev on boot and that errs 1 because kernel is automounting it
2020-01-08T22:05:18 #kisslinux <adamantium> Well I suppose you are using a test though. now that I remember
2020-01-08T22:05:49 #kisslinux <dylanaraps> What umount flags are supported in sbase/ubase?
2020-01-08T22:06:09 #kisslinux <adamantium> usage: umount [-lfn] target...
2020-01-08T22:06:13 #kisslinux <adamantium> usage: umount -a [-lfn]
2020-01-08T22:06:34 #kisslinux <dylanaraps> Ah
2020-01-08T22:07:02 #kisslinux <dylanaraps> OK
2020-01-08T22:07:09 #kisslinux <dylanaraps> We'll leave the message for now.
2020-01-08T22:07:30 #kisslinux <adamantium> Lol
2020-01-08T22:09:25 #kisslinux <dylanaraps> I made the changes in #5 manually. Thanks :)
2020-01-08T22:12:03 #kisslinux <TwistedFate> howdy ho
2020-01-08T22:12:55 #kisslinux <dylanaraps> Hello
2020-01-08T22:19:06 #kisslinux <dylanaraps> I'm off to bed.
2020-01-08T22:19:08 #kisslinux <dylanaraps> Ciao.
2020-01-08T22:21:48 #kisslinux <TwistedFate> good night