IRC Logs for #crux Wednesday, 2018-01-10

*** john_cephalopoda has quit IRC00:32
*** elderK has quit IRC00:56
*** groovy2shoes has quit IRC00:57
*** groovy2shoes has joined #crux01:02
*** john_cephalopoda has joined #crux01:30
*** john_cephalopoda has left #crux ()01:30
*** kori has quit IRC02:44
*** kori has joined #crux02:44
*** mutin-s has joined #crux03:16
*** mutin-sa has quit IRC03:18
*** mutin-s has quit IRC03:21
*** abenz has quit IRC03:25
*** abenz has joined #crux03:26
*** _________mavric6 has quit IRC03:42
*** _________mavric6 has joined #crux03:43
*** tilman_ has joined #crux03:53
rmullHmm, seems like my cpu is not supported in that ucode release. family 6 model 77 (atom c2758)03:55
j_vi didn't think the atoms were at risk, but could be wrong03:56
j_vat least not for meltdown03:56
rmullI never checked, I suppose that's nice to hear03:58
rmullIt's got some other hardware issue from a while back that could brick the thing under certain conditions IIRC03:58
j_vdon't take my word for, though. while i'm not an amd fanboy, i have mostly amd processors (due to price)03:58
j_vouch, sounds touchy04:00
abenzamd fanboy spotted04:00
abenzAMD, always the smarter choice..04:01
abenz:)04:01
*** abenz has quit IRC04:17
j_vwhat's up with that patch for gcc7 in the core/gcc repo for 3.4? it doesn't apply and inspecting it, it looks like a merge conflict or something like it05:40
j_vi've tried building gcc7 with private copies of gmp,mpfr,mpc embedded in the gcc src directory and tbh, i don't think gcc ready for mpfr-4 yet, i saw the patch for mpc and it seems to help, but gcc chokes on the newer mpfr when trying to build against it05:44
*** emmett1 has joined #crux05:55
*** Workster has quit IRC06:02
*** emmett1_ has joined #crux06:11
*** emmett1 has quit IRC06:14
*** Workster has joined #crux06:14
*** Workster has quit IRC06:14
*** Workster has joined #crux06:14
*** nwe has quit IRC06:15
*** nwe has joined #crux06:15
*** Workster has quit IRC06:30
*** Workster has joined #crux06:42
*** Workster has quit IRC06:42
*** Workster has joined #crux06:42
j_vsorry about the comment about gcc7 patch, i was looking at the contents of the removed glibc CVE-2017-15670 patch; not sure how i managed that06:47
*** emmett1_ has quit IRC07:13
*** chinarulezzz has joined #crux07:42
*** elderK has joined #crux08:00
elderKHey guys08:00
elderKI've managed to get the new laptop mostly up and running, but for a few issues.08:00
elderK1. X11 keeps freezing randomly. I haven't found out what triggers it yet. It's strange - the WM seems to freeze entirely. If I have FIrefox open, I'm unable to switch tabs in it. But if I have urxvt running, I can still type commands, etc.08:01
elderK. /var/log/Xorg.0.log doesn't seem to list anything weird.08:01
j_vhave you updated to fairly recent kernel08:02
j_v?08:02
elderKAnd for the most part, dmesg is sane too. One issue is that my Trackpad doesn't seem to be supported.08:02
elderKj_v: I'm running latest stable.08:02
elderK4.14.1308:02
elderKAnnoyingly, SD Card reader isn't functioning reall either, sdpci.08:02
elderKIT just seems to crap its pants whenever you try to do anything with it. But, I can live with that for now.08:02
elderKI'm more concerned with X freezing.08:02
elderKOr, the WM freezing. Or whatever the hell is going on there :)08:03
elderKTBH, I'm unsure of how to nail it down, to figure out what is causing the problem.08:03
elderKI should note, too, I have IOMMU disabled in EFI08:03
j_vthat is odd... do you capture stdout/stderr of x session?08:03
*** workodera has joined #crux08:03
elderKNo, but I will this time.08:03
elderKHopefully we'll catch something of note.08:03
elderKWhen it does it's weird... shit.08:03
elderKI'll dump what I can - then upload it for you all.08:04
elderKdmesg, xorg log, etc.08:04
j_vcool, someone should be able to help... i hope08:04
elderKI also have some (probably questions) about some kernel features, like the Crypto API :)08:04
elderKMe too.08:04
elderKI mean, I'm mostly happy with this new machine. But I am a little disappointed with the hardware support so far.08:04
elderKThen again, it seems to be a very recent machine.08:04
elderKSo that's expected.08:04
j_vfor 4.14.13, did you applied the staged stable patches onto 4.14.12?08:05
elderKThat isn't Crux' fault. Just maybe hasn't been enough time for the drivers to be updated for new hardware :)08:05
elderKNope. I downloaded the 4.14.13 directly from kernel.org08:05
j_vhuh, my ck4up must be broke08:05
elderK`No - I might be wrong08:06
elderKMoment.08:06
elderKMy mistake. 4.12.1208:06
elderKSorry j_v.08:06
elderKI've been looking at so many damned version numbers today :)08:06
j_voh, no problem, just was trying to stay on top of spectre patches is all08:06
j_vthough for upcoming spectre patches, some will need patched gcc7 to compile the new reptoline stuff08:07
j_vi'm building gcc7 toolchain now, and prep'ing a patched gcc7 from gcc git08:09
elderKHow's it going08:10
elderK?08:10
elderK:P I'm feeling very rusty - it's been about two years since I did a reCrux.08:10
j_vissues with the way i'm building gcc, by embedding the gmp,mpfr,mpc lib sources in the gcc sources08:10
j_vit's getting linker errors when linking against recent built mpc/mpfr08:11
j_vi need to dig out my old toolchain build scripts08:11
j_vi have vague recollection of similar issues in past, so it's not a big deal... just rusty here too08:12
elderK:P I am super glad I had the foresight to keep my configs under VCS08:12
j_vnice08:13
elderKIt's easy to get rusty with Crux. Once you've got it all set up and working, it tends to stay working for the most part.08:13
elderKYou can easily go for a year or so without having to seriously mess withs hit again.08:13
elderKWhich is great.08:13
elderKUnlike Windows where there's a decent chance it dies on the next update P08:13
elderK:)08:13
j_vyeah, very true. good too, cause i get busy with work often enough08:13
elderKSame here.08:14
elderKYou kind of have to be in the right headspace for "configuring" and stuff.08:14
elderKIt's fun when you're in the right mood.08:14
elderK:D08:14
j_vi agree, it's like been 'in the groove'08:14
j_vs/been/being/08:15
elderKJust like when you're hacking.08:15
elderKYou get in the zone :)08:15
elderKBoth are awesome timesinks.08:15
elderK:D08:15
elderKAha. X just locked.08:16
*** jue has quit IRC08:16
elderKI can't prove it for sure yet - but it seems to be related to the Fn button.08:16
elderKNothing weird in startx log08:17
elderKstartx > log 2>&108:17
j_vyeah, should do it08:17
j_vi put it in a script in ~/bin/startx08:18
elderKSmart :)08:18
elderKIt's so weird. The system is still going.08:18
elderKAnd I can still get terminals and shit to open08:18
j_v'/usr/bin/startx -- -nolisten tcp > $XORG_ERRORS_LOG 2>&108:18
elderKOkay, I lie. DISPLAY=:0.0 urxvt isn't working. Might be wrong display address08:19
j_vwonder if it could be because urxvt uses client/server model08:19
elderKI'm not using it urxvtd / urxvtc style.08:19
elderKWell, I just killed xinit from VT.08:20
elderK:P Machine has locked completely.08:20
elderKYeah. It seems to be Lenovo's FN button that's k illing it08:23
j_vis it mapped in text console?08:25
elderKI'll check.08:25
elderKI haven't built the special 'thinkpad-acpi' module.08:26
elderKI didn't want to use the damn FN key for that stuff.08:26
elderKMaybe that module is necessary.08:26
elderKI'll do a showkey in VT.08:26
elderKKeycode 143. When you press it down, it signals PRESS/RELEASE simultaneously.08:27
elderKDoesn't seem to wait for you to release the key08:27
Anselmoha08:28
AnselmoACTION is trying to configure a new thinkpad too. .08:28
Anselmobut presently more working on wifi08:28
elderKAnselmo: I've bypassed that for now - using Ethernet.08:29
elderKGotta get damned firmware for iwl :|08:29
*** timcowchip has joined #crux08:29
elderKWhat Thinkpad are you using?08:29
elderKj_v: Any ideas / advice?08:29
Anselmouhm, its a t41008:30
elderKYup - that's definitely how to reproduce the "X death." Just press the FN key08:30
elderKMine's an L570.08:30
Anselmowait, pressing Fn kills X ?08:30
elderKYours is probably a lot more grunty than mine :08:30
AnselmoO_O08:30
elderKYup.08:30
j_vi haven't had to deal with anything like that really, i mostly use desktop, so kbds are pretty basic here08:30
Anselmoha, mine seems decently ok, though a little beat up08:31
elderKI just replaced an aging Toshiba. I was really disgusted by that machine's build quality. Connectors started prying off the motherboard :| AFter only two years and I wasn't that rough.08:31
Anselmodoes xev say something weird when you press Fn ?08:32
elderKAbsolute death to maintain, too, you couldn't remove the keyboard without removing the entire motherboard.08:32
elderKI'll try and capture that next.08:32
elderKThis time, I'll run XEV before I press FN08:32
elderKHopefully we get something interesting :)08:32
Anselmohahahahaha08:32
Anselmooooh, hmm08:32
Anselmocause will it go to xev or just kill x. . .08:32
*** jue has joined #crux08:32
elderKXF86Wakeup08:33
elderKIt still froze X08:33
Anselmothis is sort of replacing a desktop that I am having. . .mysterious problems with and need to sort out but was desperate . .08:33
elderKI mean, mostly.  I can still use URXVT08:33
elderKBut the WM is dead.08:33
elderKAnd FF no longer seems to respond08:33
Anselmooh, hrm08:33
Anselmowhat wm is it ?08:33
elderKi3wm08:33
elderKOkay. Iit's worse than I thought - I can't change focus or anthing :P I can't kill XEV now.08:34
elderKRight: Fn mapped in X to XF86Wakeup08:34
elderKThat should help :)08:34
j_vweird that you still have input to urxvt08:34
elderKhttps://www.reddit.com/r/i3wm/comments/683hhn/thinkpad_fn_keys/08:34
elderK:D08:34
Anselmomy Fn key is mapped to the same thing08:35
elderKIt seems to be an i3 thing.08:35
Anselmoah hm08:35
elderKSo strange that xf86wakeup kills i308:37
j_vweird... use bspwm here, and sxhkd for input handling, so a bit different. only used i3 for short while08:37
AnselmoACTION is the same but ratpoison  . .08:38
Anselmobut I should probably be on my merry way . . .08:42
Anselmoso good luck with your. .  .weird things . . .08:42
j_velderK: is XF86Wakeup in your i3wm config file?08:43
elderKNope.08:47
elderKAnselmo: Take care.08:47
elderKAnd good luck with your Thinkpad :D08:47
elderKj_v: It's just defaulting.08:48
j_vso it seems like it's X, not the wm causing the issue, but that might be wrong... if it's getting the right press/release codes in xev, then it seems wrong to blame the input driver08:50
j_vi wonder if i3-input is crashing/segfault, like null ptr deref or out of bounds read08:53
fun2j_v, 4.12.13 is out :)08:59
j_vnice08:59
j_valready building09:00
elderKI'm not seeing any segvs09:04
elderKMaybe I'll fire up TWM09:04
elderKand see if I can kill that09:04
elderKI'd be perfectly happy with a way to just... ignore the key09:05
elderKbut xmodma -e 'keycode 151 = NoSymbol' didn't help09:05
*** timcowchip has quit IRC09:07
elderKHEY GUYS!09:24
elderKKinda semi figured it out09:24
elderKIT's not i3.09:24
elderKIf I launch i3 with ck-launch-session...09:24
elderKFN will kill it09:24
elderKIf I don't, everything is happy09:24
elderKIt's something to do with ck-launch-session.09:26
elderKor dbus maybe.09:26
j_vwhere's the ck-launch-session come from?09:26
elderKbecause if I launch BlackBox the same way, it dies with FN09:26
elderKI added it manually for some reason I forget now09:26
elderK:LD09:26
elderK:D09:26
elderKProbably for bluetooth headset.09:26
j_vconsolekit09:26
elderKYup09:26
elderKI'll see what depends on COnsoleKit09:27
elderKThat should narrow it down09:27
j_vi looked through the i3-input source and couldn't see anything that look like it could be a problem09:27
elderKYup. My VCS says "Now launches i3 via ConsoleKit so that BlueZ works with Pusle"09:28
j_vhuh... i'd say screw bluez, but that's me09:28
j_vand pulse09:29
elderKUnfortunately I use BT Headsets a lot :(09:29
elderKOTherwise I'd agree.09:29
j_vwith pulse?09:29
elderKIt's most seamless using Pulse.09:29
elderKAt least, it was really annoying doin git all through ALSA.09:29
j_vah, perhaps there's a work around on the console kit side09:30
elderKHopefully09:30
elderKStill, we have more clues now!09:30
elderK:D09:30
elderKTHE HUNT IS ON!09:30
j_vseems your pretty close, i'd say09:30
elderKYeah. Now I just need to figure out WTF to do now >:D09:30
elderKI will test BT without it though09:31
elderKJust to see if it's necessary.09:31
elderKI can't remember WHY I needed it09:31
elderKWeird. Doesn't seem necessary09:32
j_vwait, i thought you needed ck for i3 so that bluez works w/pulse?09:33
j_vfor your bt headphones?09:33
elderKyeah09:34
elderKMe too.09:34
elderKIt doesn't seem necessary.09:34
elderKI just did a quick test.09:34
elderKBooted i3 without ck.09:34
elderKLoaded Firefox, audio.09:34
elderKConnected BT HS.09:34
elderKWorks.09:34
j_vwell, sounds to me that ck is maybe not so crucial09:35
elderKYeah. And FN doesn't kill i3 anymore.09:35
elderK:D09:35
j_vnice, must be a relief09:37
elderKIt might have been a hack in the past to get BT working without ACTUALLY configuring BT proper :P09:37
elderKThis time I added a special group so I could access it via DBus and stuff09:38
elderKAnd my user is in that group.09:38
elderKAnd I start dbus at boot, etc.09:38
elderKTBH, I have no @#!$ing clue why ck-launch-session was needed.09:38
elderK:D09:38
elderKI'm doing another test :)09:38
elderKAnd yes, big relief.09:38
elderKAt least the machine is usable now.09:38
elderKThere are probably tons of little things that I haven't migrated. But I'll hit them when I need them, I guess.09:38
elderKj_v: Thank you so much for your help.09:39
elderKI really appreciate it :)09:39
j_vhey, no problem, you did all the work09:39
j_vi was just the cheering squad09:39
j_vpretty cool, though, i usually track my issues down, but sometimes not the same day :09:40
j_vso you are ahead of me on this one09:40
elderKI think it was just luck.09:42
elderKI don't know why I tried it without ck-launch-session.09:42
elderKI just commented out syndaemon, etc.09:42
elderKFor some reason thinking maybe those were @!#$ing shit up09:42
elderKAnd boom, we foudn it.09:42
elderKTomorrow, I might do the ck-launch-session shit again.09:42
elderKAnd see if I can't get some logs as to what happens when XF86Wakeup is pressed.09:43
j_vgood idea... be cool to know just what the actual issue is so you don't get slammed unexpectedly later09:43
elderKYeah, exactly.09:44
elderKj_v: As for why I use Pulse, I like that I can:09:46
elderK1. Seamlessly disconnect various audio outputs and have stuff change to another.09:46
elderK2. I don't have to manually specify when I launch a program, which card I am using.09:46
elderK3. IIRC, BT gave me issues with multiple programs being unable to output audio at once.09:46
elderKThat and all the wikis I found effectively said "BlueZ with ALSA is no longer really supported."09:46
elderKI avoided Pulse for many, many years :P09:46
elderKBut that ultimately sold me on it09:46
elderKj_v: BTW, is there something I have to do so that prt-get dependent and stuff works properly?09:49
elderKif I go, say,09:49
j_vcool, i've had my eye on some new headphones... nice to know that bt option will work... i always knew i'd have to give in and use pulse, i just have this thing about the guy with initials LP09:49
elderKprt-get dependent foobar09:49
elderKand I know stuff uses foobar,09:49
elderKthat command shows nothing09:49
j_vadd --all09:49
j_vor/and --recursive09:49
elderKj_v: Same reason why I stayed away fo rso long.09:49
elderKTBH, It's not a GREAT experience.09:50
elderKYou need BlueZ5 for that.09:50
elderKI'm using BlueZ4. Since 5 seemed tied to systemd in various ways.09:50
elderKAnd I couldn't be fucked.09:50
elderKMaybe not systemd, but it was WAY MORE BULLSHIT.09:50
elderK:P09:50
*** BitPuffin has joined #crux09:50
elderKLike, sometimes you'll connect your headset. And try listening to something. Except it won't work for some reason, it won't correctly get into A2DP profile (which you need for high quality sound.)09:51
elderKInstead, you get stuck in "telephony mode" sometimes. Which sounds like shit.09:51
elderKAnd you have to disconnect/fuck with settings until it eventually just "Works" again.09:51
elderKSo, while it works, it's not without annoyances.09:51
j_vsounds like computers and gadgets in general, to me09:51
elderKAlso: I don't have "autoconnect" in range. I have to manually disconnect/connect the headset. This is   probably because I'm not using any kind of "BT Agent". I'm just using bluez-simple-agent, the stuff that comes with BlueZ09:52
j_vmakes sense, all that auto stuff has prices09:52
elderKYup.09:52
elderKI checked out Blueman and stuff. But the fact that it uses Python and requires a ton of other dependencies made me nope out of there.09:53
elderK:P09:53
elderKIt was more all the extra dependencies.09:53
elderKOh an dit needs BlueZ5.09:54
*** fun2 has quit IRC09:54
j_vpython is ok to me when it's for hacks for short run tools, i don't like it or perl for hardware interface stuff09:55
elderK:P Now all I have to do is get sd working :P And that's an upstream issue09:56
*** abenz has joined #crux09:56
elderKprt-get --all --recursive dependent consolekit still doesn't show me what needs it?09:58
*** Sigyn has left #crux ("Leaving the channel (no spam or action taken for 3 days.) /invite Sigyn #crux again if needed")09:58
j_vhave you tried booting with iommu enabled since you are on a newer kernel?09:59
elderKNo.10:00
elderKI've been afraid to, tbh10:00
elderKThe NVME read stuff didn't have reliably. It happened only under extreme load.10:00
elderK*happen10:00
elderKI'm not sure how to "safely" test it.10:01
elderKAny ideas>10:01
elderK?10:01
j_vdo you have pci-id for it? can you lspci and get info on it?10:01
j_vit would help to know if the nvme driver has quirks set for it so we could tell whether known issues might already have been addressed10:03
elderKYup, sure.10:04
*** fun2 has joined #crux10:05
elderKI'll dump lspci, pastebin it.10:05
fun2elderK, j_v I use bluez with alsa for my headset. In case someone wants to try I have bluez-alsa port in my repo and I use bluez from maraku's repo. no consolekit.10:10
*** afterK has joined #crux10:10
*** afterK has quit IRC10:10
*** afterK has joined #crux10:10
elderKThere's my alt :D10:10
elderKfun2: Turns out I don't need CKit if I build BlueZ with X11 support.10:10
elderKfun2: Does your bluez alsa handle disconnect, like, if the headset disconnects, does it transfer audio output to another device? And back again when the headset reconnects?10:11
j_vfun2: very cool... will have to give a role as soon as i pick some new head phones10:11
afterKj_v: https://nopaste.xyz/?77d737dc7ecb0b98#7YIZYe1M725a/tsJ8c1nbWJTcXejmh4WfEuz4hwGB2A=10:11
afterKHopefully this will give you enough information.10:11
*** afterK has quit IRC10:12
fun2it reconnects when I came back in range10:12
elderKfun2: What are you using as a Bluetooth agent, if anything?10:12
elderKAnd have you noticed that BT gets out of sync with what you're watching when you walk away then come back?10:13
elderKSay YouTube. You go to another room, and come back.10:13
fun2Bluetoothd10:13
fun2i think10:13
elderKHow do you make it so that devices connect automatically? I always get errors about "No agent for device found."10:13
j_velderK: that, 0xf1a5, there is a quirk and workaround added to drivers/nvme/host/pci.c10:13
*** BitPuffin has quit IRC10:13
elderKfun2: I'd be interested in your port. :)10:13
fun2arch linux wiki says something about autopower10:14
elderKj_v: There is? :D Is it in 4.14.12?10:14
j_vthat was the one that that ubuntu bug report was on about that you linked to much earlier10:14
j_vyes10:14
elderKj_v: Also, what did you search for when you were checking for quirks? :)10:14
elderK:D AWESOME!10:14
elderKSo IOMMU should be sane10:14
elderKGod damn you guys rule!10:14
elderK:D10:14
*** BitPuffin has joined #crux10:14
j_vwell, as far as that nvme driver, it shouldn't freak on you hardware anymore, at least going on what the commit for that addition says: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/drivers/nvme/host/pci.c?h=v4.14.13&id=50af47d04ca530544b27affffb0722f158e2bb9c10:17
j_vthat commit actually was added in 4.12.y branch, i think10:17
elderKWell, it might be prudent to continue setting everything up. Then, image the system.10:19
elderKTHEN turn IOMMU on.10:19
elderKI mean, it's taken around 13 hours to get it to where it is now :P10:19
elderKAnd that was just build time.10:19
elderK:D10:19
j_vright, think you are prudent10:19
elderKACTION shrugs10:20
elderKI'm more lazy than prudent.10:20
elderKACTION bows10:20
elderK:)10:20
j_vwell, i hate doing things twice when not necessary10:20
elderK:P We could flip a coin and decide that way10:20
j_valso hate breaking things when i rush10:20
elderKAny advice for cloning the system? I usually fire up "Clonezilla"10:21
elderKDo you have any recommendations? :)10:21
j_vme too, though it's been a while10:21
elderKRight. Tomorrow, I'll clone the system.10:22
elderKThen I'll enable IOMMU10:22
elderKAnd THEN10:22
elderKI'll rebuild every package.10:22
j_vnice10:22
elderKIf it doesn't give me errors after say, an hour or so. I'll consider it afe.10:22
*** ryuo has joined #crux10:22
elderKBTw, I remember why I used ConsoleKit now.10:22
j_voh?10:23
elderKAfter a bit of tinkering: It was needed only so that Pulse would start up on its own when other programs needed it.10:23
elderKTurns out that if I mess with the Pkgfile and build with X11 support, I can get a scirpt, start-x11-pulse or something.10:23
elderKThat will set up the necessary stuff manually.10:23
elderKThen I can get rid of polkit/consolekit10:23
elderK:D10:23
j_vsounds easier to me10:24
elderKI still have no why PolicyKit / ConsoleKit shat their pants when I pressed Fn.10:24
j_vi gotta step out for a smoke, brb10:24
elderKWell, I'm off to bed.10:24
elderKThanks again :)10:24
elderKEnjoy your cigarette.10:25
j_vthanks, should do the same after10:26
fun2elderk, btw, with alsa you can't switch the sound output to another device unless you use udev and restart the process mpv/browser, from what I've read10:28
fun2or change .asoundrc10:29
j_vah, i've got a patch for adding reptoline to gcc-7.2.0, now to try building with it10:51
j_vbig patch10:51
j_vs/reptoline/retpoline/11:10
*** g0relike-2 has joined #crux11:23
*** jue has quit IRC11:26
*** g0relike has quit IRC11:26
*** jue has joined #crux11:34
frinnstwhat cpu are amd fam17h? ryzen?11:51
*** abenz has quit IRC11:51
joacimnu-fx i think11:52
joacimno thats fam15h11:53
j_vzen, i think11:53
joacimyeah fam17h is zen11:53
frinnstx86/microcode/AMD: Add support for fam17h microcode loading11:53
*** leetspete1 has quit IRC11:55
*** workodera has quit IRC12:09
*** fun2 has quit IRC12:10
*** workodera has joined #crux12:12
ryuoI once saw a bumper sticker that read, "If you can read this, you're following too closely."12:21
*** elderK has quit IRC12:39
*** leah2 has quit IRC12:43
*** leah2 has joined #crux12:43
*** leah2 has quit IRC12:50
cruxbot[contrib.git/3.3]: monit: adopted12:51
cruxbot[contrib.git/3.3]: at: fix source URL, update to 3.1.20, set UNMAINTAINED (FS#1553)12:51
cruxbot[contrib.git/3.3]: pmwiki: update to 2.2.10612:51
j_vi found this copy of the new microcode: https://dev.gentoo.org/~whissi/dist/linux-firmware/microcode_amd_fam17h.tar.gz13:05
j_vthis page discusses some details, which might be relevant: https://bugs.archlinux.org/task/5695113:14
j_vhttps://www.suse.com/support/update/announcement/2018/suse-su-20180009-1/ claims that the amd microde update turns off branch prediction, though this page claims that is incorrect: https://www.phoronix.com/scan.php?page=news_item&px=AMD-Branch-Prediction-Still13:19
*** leah2 has joined #crux13:22
frinnstinstalling some random firmware from a /~<user> url doesnt seem like a good idea13:29
frinnstwhy are they trying to hide it?13:29
joacimunless it is hosted at crux.guru13:30
j_vi'm just trying to track down as many copies of it as i can. seems that suse and maybe redhat/fedora are main official recipiants of official microcode file from AMD, but it's hard to tell... it's a pile of shit13:33
j_vi'd like to compare cksums on various copies, not saying anyone should use, but no telling how long before amd does 'the right thing (TM)'13:34
frinnstit should be in https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/13:39
frinnstyeah im not blaming you - just observing that its all fucked up :-)13:40
j_vok, it's ok... i don't mind if you do go off on me (well, to point). and i guess part of me is so wound up over just how fucked up this is, that i needed to share it. i guess you could accuse me of spreading fodder and not be too far from the truth. ;)13:42
j_v*to a point13:42
j_vthough i just compared sha512sums on the microcode files from that gentoo user url and the one from suse's source rpm (https://download.opensuse.org/source/tumbleweed/repo/oss/suse/src/kernel-firmware-20180104-1.1.src.rpm) match13:45
j_v*they match13:46
cruxbot[opt.git/3.3]: dialog: update to 1.3-20171209, new maintainer13:46
*** amonn_ has quit IRC14:34
*** fun2 has joined #crux14:36
*** abenz has joined #crux14:39
frinnstwhy have all swedish ERP systems decided to shit the bed this year?15:21
frinnstnothing but erp problems all wee15:21
frinnstk15:21
*** john_cephalopoda has joined #crux15:25
joacimerotic role play?15:28
cruxbot[contrib.git/3.3]: pv: updated to version 1.6.615:30
workoderayeah these swedish erotic role play sytems right, unbelievable15:39
abenzfrinnst: ERP? as in odoo kinda stuff?16:04
abenzgood wallpaper: https://quotefancy.com/quote/22257/John-Wooden-The-true-test-of-a-man-s-character-is-what-he-does-when-no-one-is-watching16:24
*** workodera has quit IRC16:38
*** leetspete1 has joined #crux16:57
pedjaif a browser ships with a search extension for a site, does that imply that they approve of it, or is it just for a better user experience?17:21
pedjaI am curious what's the reasoning behind Brave browser apparently shipping with search for infogalactic, conservative 'answer' to 'too liberal and atheist' Wikipedia :)17:24
pedjathat question is curiously missing from Lunduke's interview with Eich17:25
*** JanC_ has joined #crux17:26
Anselmowell. .. it makes me think of mozilla including google and yahoosearches in firefox. . .17:27
*** john_cephalopoda has quit IRC17:28
pedjasame Lunduke that slammed Mozilla for donating to RiseUp because RiseUp provides e-mail for a group that American government consider terrorists17:29
Anselmothey obviously cannot include all the search things on the internet. . . and what they do include seems to imply some prefference17:29
Anselmoor at least that they beleive it to be legitimate17:29
pedjaEich opened an issue that led to adding them, so it would seem that he sees them as legitimate source17:33
pedjawhich is totally fine, his browser, he can do whatever he wants.17:36
Anselmohm17:36
*** john_cephalopoda has joined #crux17:40
pedjathe question is, if they are pushing their own agenda this way, can they be really trusted with a user privacy?17:45
john_cephalopodapedja: Google and yahoo paid mozilla to be in their browser.17:46
pedjaEich's claim in the interview is that they can't be evil because of some crypto voodoo17:47
*** BitPuffin has quit IRC17:47
Anselmoif the US government paid mozilla a lot of money to backdoor firefox. .. would we excuse them ?17:48
Anselmothough thats a sort of extreme comparison . .17:48
pedjajohn_cephalopoda, and that's public knowledge, so the reasoning behind it is pretty clear17:48
john_cephalopodaThe browser is open-source. If you don't like the search engine, take the code and remove those parts.17:49
pedjayou're missing the point. I am OK with them adding whoever they want, as long as there is a clear explanation about that decision17:52
john_cephalopodaWhy has there to be a clear explanation?17:53
Anselmoyou should be able to trust the developers of the software you use. . .17:53
john_cephalopodaYou don't have to trust the devs when the code is open.17:53
pedjaI trust Crux devs, and that's about it :)17:54
john_cephalopodaNow I'm reminded of all the crappy software again.17:55
Anselmoha17:55
frinnstpedja: lol17:55
frinnstI have a bridge for sale, interested?17:55
pedjanah, the costs of maintaining it are too high. thanks for the offer, thou17:57
pedjajohn_cephalopoda, I am sure you've already seen this :) https://www.ece.cmu.edu/~ganger/712.fall02/papers/p761-thompson.pdf18:02
*** leah2 has quit IRC18:06
*** leah2 has joined #crux18:09
john_cephalopodaAh, yeah, that one is well-known.18:11
*** onodera has joined #crux18:12
john_cephalopodaWith "crappy software", I was more targeting the badly-written and bloated software, as criticised in Wirth's paper "A plea for lean software" ( https://cr.yp.to/bib/1995/wirth.pdf ).18:13
*** abenz has quit IRC18:33
*** tsaop has joined #crux18:41
*** tsaop has quit IRC19:43
marakufun2, funny enough i was also testing bluez-alsa a week or two ago... however i got really bad quality audio from it. what's your setup like?19:51
*** elderK has joined #crux19:59
elderKGood morning peeps.19:59
elderKI was wondering how you use a GIt repository as a source for packages20:00
elderKI'm trying to create my own linux-firmware port, one that is more recent than in opt20:00
john_cephalopodaelderK: Yes, it's possible.20:01
elderK:) I'm not sure what format the SOURCE variable has to be in for it to retrieve from Git :)20:02
elderKAhhhh. Is the driver for getting a bunch of ports from some place20:06
elderKor for downloading some source file?20:06
elderKHmm20:06
elderKI think I understand now.20:06
jaegerIn the past I've left source empty and used a small script in the build function to check out or pull the code20:07
john_cephalopodaelderK: You just have to put it into a github repo and then write a driver like that one: https://github.com/jmf/jmf/wiki/Git-driver-for-the-jmf-repo20:13
elderKAre there any conventions I should use wrt to naming variables in a Pkgfile, nonstandard ones?20:15
elderKsay, one for a git repo I'm cloning from?20:15
elderKjaeger: I'm unfamiliar with Git - how do I 'export'? Like, clone something - but don't keep the .git metadata.20:15
elderKOr do I just nuke .git in what I've cloned? :D20:16
jaegerIf you're doing that, you might as well just make a tarball, right?20:16
jaegerbut yeah, nuking the .git directory will remove all git metadata20:16
tilman"git archive" is what other systems might call "export"20:17
tilmanwhether or not that works depends on the remote site though ;D20:17
elderKjaeger: I suppose so.20:17
john_cephalopodaelderK: There are some templates in the 6c37 project.20:22
john_cephalopodahttps://github.com/6c37/cross <-- In this repo, there is a script in utils/ called "mkport", which will auto-generate a skeleton, including git stuff.20:23
elderKI wonder why some things seem to be able to load firmware at runtime20:32
*** ryuo has quit IRC20:32
elderKand others cant20:32
elderKlike i915. It wants some firmware. It's in /lib/firmware, etc.20:33
elderKBut the kernel says "Hey, I can't find it. direct load failed."20:33
elderKSuppose it has to be compiled into the kernel20:33
frinnstjohn_cephalopoda, elderK: prtcreate in prt-utils20:33
jaegerIt probably does, yes20:33
john_cephalopodaelderK: I had a similar problem. There's a way to include firmware blobs into the kernel.20:34
elderKYeah. Drivers -> Generic :D20:35
john_cephalopodaDevice Drivers > Generic Driver Options20:35
elderKInclude External20:35
elderK:D20:35
elderKWe think alike!20:35
john_cephalopodaHeh :D20:35
john_cephalopodaI always have to use it on my laptop for the Intel wifi driver.20:36
*** ryu0 has joined #crux20:36
elderKSame here usually.20:37
elderKI'm surprised - I've never seen a GPU need firmware before.20:37
elderKkbl_dmc_ver1_0120:37
john_cephalopodaMy GPUs don't need firmware.20:38
elderKAnd BT.20:42
elderKAny tips on how to manage power nicely on Battery guys?20:42
elderKUsually I just ignore the problem :P20:42
jaegercheck out powertop, maybe20:42
elderKJust make some script put powertop's recommendations into effect when acpi detects loss of AC?20:43
jaegerThat sounds reasonable to me... but I don't have a laptop running linux currently20:44
elderKRight. Only one last problem to deal with, it seems:20:44
elderKMy SD card reader just... makes the system scream.20:44
elderKDriver is sdpci20:45
elderK"mmc0 unknown controller version (3). You may experience problems,"20:45
elderKAnd oh, how I do20:45
john_cephalopodaI usually just enable _everything_ in the SD card driver section.20:47
elderKIll module them al20:48
elderKAnd see what's loaded.20:49
elderK:D20:49
elderKYeah, same thing is loaded as I had built in20:52
elderKSees the card just fine. When you mount - all hell breaks lose.20:52
john_cephalopodaAre you sure that you got NTFS and FAT support in the kernel?20:54
elderKSD is formatted XFS :P21:08
elderKIt's hardware issue. :|21:08
elderKMounts fine on older laptop. New laptop, the hardware issues a ton of odd errors21:09
elderKERROR -84 transferring data, etc.21:09
elderKIt's one of thsoe "Micro SDs" inside an 'adapter" tha tmakes it fit SD slots.21:09
elderKFor all I know, maybe the issue is simply the adapter.21:09
elderKLike, the framing that hte SD Micro is being fit into21:10
elderKSeems unlikely. But who knows21:10
elderKThis is annoying but not a critical problem :)21:10
elderKWifi is up!21:10
elderK:)21:10
*** chinarulezzz has quit IRC21:13
*** fun3 has joined #crux21:15
elderKHey fun3!21:18
fun3Hi, elderK!21:19
elderKWhat do you know about laptop power management?21:19
fun3021:19
elderKIm taking a gander at ArchWiki now for ideas.21:19
elderKI generally just ignore the problem and run my battery for as long as possible.21:19
elderK:P21:19
fun3I think I've used pm-something21:20
fun3once21:20
*** Henschi has joined #crux21:21
elderKI've always just echo mem > /sys/power/state21:21
elderK:D21:21
fun3hehe21:22
fun3btw, most of the git servers have a link to get the tarballs, per commit hash or tag21:22
elderKI solved the firmware issue the brute force way.21:23
elderKI just have a tbz2 :D21:23
elderKI guess I need to figure out which modules don't like being suspended.21:23
fun3You can make all builtins :D21:23
fun3no module, no problem21:23
*** onodera has quit IRC21:25
fun3maraku: I'm not an audiophile, but I don't get poor quality with my bt setup: services(dbus,bluetoothd,bluealsa), simple .asoundrc, a minimal conf for bluetooth for autopower the PC controler21:25
fun3and everything else done in bluetoothctl (with my user in the lp group)21:26
elderKbluetoothctl?21:28
elderKWhat BlueZ version are you running?21:28
elderKBlueZ4 doesn't seem to have bluetoothctl :(21:28
fun35.121:28
elderKI wonder what advantages v5 brings21:28
fun3https://github.com/maraku/crux/blob/master/bluez/Pkgfile21:29
*** Henschi has quit IRC21:29
fun3And I've put:[Policy]21:29
fun3AutoEnable=true21:29
fun3in /usr/etc/bluetooth/main.conf21:29
fun3and my .asoundrc http://ix.io/DZS21:31
elderKAutoEnable, what exactly does that do?21:32
elderKTurn on the BT controller at boot?21:32
fun3yes21:32
elderKI thought that happened automatically :D21:33
fun3:))21:33
elderK"InitiallyPowered" setting.21:33
fun3maybe, it didn't worked on the first reboot21:33
elderKBut then again, I'm using BlueZ4.21:33
elderKI wonder how many new dependencies you need, vs. Bluez4.21:33
elderKIf any.21:34
elderK:D Otherwise I'll update.21:34
fun3http://ix.io/DZT21:35
*** john_cephalopoda has left #crux (""A tree works like your brain." ~ The Org Manual")21:35
elderKMuch appreciated fun3 :)21:35
fun3jagger seems to have fewer: http://ix.io/DZU21:36
elderKPerhaps he uses fewer profiles.21:37
fun3no, glib is "redundant" in maraku's version21:37
fun3:))21:37
fun3lol21:37
fun3still, alsa is painful when changing outputs21:37
elderKPulse is pretty sweet with that.21:39
elderKIt's pretty much the only reason I use Pulse at all.21:39
elderKOtherwise I'd kick it back to hell21:39
elderK:D21:39
elderKSometime I'm going to have to prune my personal ports place.21:40
elderKI've accumulated way too many ports that I don't use anymore21:40
fun3I was prepared to use pulse for bluez, but found bluez-alsa21:41
fun3I did configure pulse, first, as described here: https://wiki.archlinux.org/index.php/PulseAudio#ALSA.2Fdmix_without_grabbing_hardware_device21:42
fun3but in .config/21:43
fun3and uninstalled when everything worked :))21:43
elderKI think the extra des, btw, come partly from ical21:48
elderKlike db, gobject-introspection, etc.21:48
fun2elderK, i libical installed from df repository21:51
fun2i have21:51
*** pitillo has quit IRC21:52
*** fun3 has quit IRC21:53
elderKIt'll be cool to see if BlueZ5 does "range" detection21:54
elderKLike, with Windows, if you walk away from the machine playing a video21:54
elderKThen come back, it is always in sync.21:54
elderKBlueZ4, doing that will get audio out of sync with video.21:54
*** frinnst has quit IRC21:58
*** frinnst has joined #crux21:58
*** frinnst has quit IRC21:58
*** frinnst has joined #crux21:58
elderKGuys, is there a better way to get a package's deps than findde?22:00
fun2it looks like my bluez-alsa port has the same bug as slim FS#154622:09
elderK?22:18
*** abenz has joined #crux22:18
elderKfun2: Which conf files does your bluez install?22:21
elderKbluez5 only seems to have main.conf and bluetooth.conf22:21
fun2maraku's port install none22:23
elderKSo you had to write your own /etc/bluetooth/main.conf etc?22:25
cruxbot[contrib.git/3.3]: [notify] openexr: updated to version 2.2.122:29
cruxbot[contrib.git/3.3]: ilmbase: updated to version 2.2.122:29
fun2yes. two lines in /usr/etc...22:29
jaegerelderK: not really, I think... you could read the upstream package's install documentation or look at their website, or use ldd, etc... but that's all manual vs. finddeps22:30
fun2i'm testing range detection elderK22:30
elderKjaeger: How does start-stop-daemon work?22:32
elderKI'm using it to run some program22:32
elderKBut it just hangs.22:32
elderKI know bluetoothd detaches22:32
jaegerI've not looked closely at it, sorry22:33
elderKNP :)22:38
*** samathy has joined #crux22:43
elderKfun2: BlueZ5 seems awesome so far!22:46
elderKfun2: What are the results of your range experiments? :D22:46
fun2poor22:47
elderKYou're getting out of sync too?\22:47
fun2if the bt disconnects the mpv stops22:47
fun2if it doesn't but loose the sound for a couple of seconds becomes out of sync22:48
elderKMine still goes out of sync, if I don't d/c22:48
elderKSay, I walk from here to the kitchen and back.22:48
elderKThen I press pause on th elayer.22:49
elderKIt takes a few seconds for the audio to stop.22:49
elderKClick play, another few seconds.22:49
fun2if it disconnects what happens to your mpv?22:50
elderKPause. Which is okay with me.22:51
fun2here it stops22:51
elderKAll I need now, is for the Agent to run without me having to manually start bluetoothctl22:51
fun2probably alsa or bluez-alsa22:51
fun2i see the user docs ar still missing from bluez22:53
fun2only api is documented22:53
elderKfun2: Do you have an /etc/bluetooth/audio.conf?22:57
elderKno strings in bluetoothd for it22:58
elderKSeems it is no longer used?22:58
fun2no audio.conf elderK22:59
fun2i have main.conf in /usr/etc/bluetooth/23:00
fun2there is bt-agent in bluez-tools23:01
fun2alan's repo23:02
*** samathy has quit IRC23:04
*** samathy has joined #crux23:05
elderKIs it what you use, fun2?23:05
fun2i use only bluetoothctl Kung fu23:06
elderKACTION nods23:06
elderKI did something similar for Bluez4.23:06
elderKI just wrote some small scripts. I'd fire them off whenever I wanted audio.23:07
fun2but only for 3 days :)23:07
elderKbha, bhc, etc.23:07
elderK:P23:07
elderKNow I'm getting spoiled by how much better BlueZ5 stuff is23:07
elderK:D23:07
fun2when out of sync, pause and resume from headset seems to fix the sync, elderK23:10
*** Lukc has quit IRC23:22
*** Lukc has joined #crux23:22
*** samathy has quit IRC23:42
*** samathy has joined #crux23:42
*** samathy has quit IRC23:44
elderKI don't have a pause buttons on headset?23:44
elderKTHanks though fun :)23:44

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!