00:53 karolherbst: Tom^: any any results?
01:51 hakzsam: skeggsb_, btw, I'll try to have a look at the nvif series this weekend (esp the mesa one)
03:04 Tom^: karolherbst: nope didnt test, i had a few beers in me. was tired and the sound of maybe running on 1.22V didnt sound healthy :p
03:19 karolherbst: :D
03:19 karolherbst: it is just +0.05V above what it runs on "normally"
03:21 karolherbst: but you can always decrease the value by a multiple of 0.0125
03:33 Tom^: (info.max - 0.0125) ?
03:39 karolherbst: yeah
03:39 karolherbst: or 0.0125 * 4
03:39 karolherbst: currently the voltage needs to be a value inside the voltage table
03:39 karolherbst: and yours goes like this: -- Mode GPIO, Base voltage 825000 µV, voltage step -12500 µV, acceptable range [825000, 1212500] µV --
03:40 karolherbst: don't know why the step is sometimes negative, but nouveau does the right thing
03:59 Tom^: hm SSE register return with SSE disabled , im not allowed todo float operations?
04:07 karolherbst: Tom^: not in the kernel
04:07 karolherbst: it is possible, but messy
04:08 karolherbst: Tom^: what did you try to do?
04:09 Tom^: i need breakfast before anything else, and mostly coffe.
04:10 karolherbst: Tom^: well the voltage is usually stored as uv
04:21 Tom^: oh right so info.max is essentially 1212500
04:28 karolherbst: yes
04:37 RSpliet: no, no floats in kernel, you don't need them even if you think you do
04:53 snacks: hey, apparently someone got Nouveau working with freesync, but I can't find a source or whether it actually happened.
04:53 snacks: did it actually happen?
04:53 snacks: if so, does anyone have a source?
05:11 agpu: hi. im trying to get nouveau to work with my 860m gpu (GK104), but it doesnt even seem to start correctly: https://paste.fedoraproject.org/295362/14487162/
05:12 agpu: is there something i can do about it?
05:12 karolherbst: agpu: yes
05:12 karolherbst: known issue
05:12 agpu: that was fast lol
05:12 karolherbst: there is a quirk only some (or maybe all?) mobile kepler cards need
05:13 agpu: which is?
05:13 karolherbst: agpu: try it out with nouveau.config=War00C800_0=1
05:18 chillfan: hey, i'm using the nouveau driver and getting a black screen at boot (the xserver starts without issue) and can no longer drop to a console, any instructions on how to fix this? I guess it has to do with the framebuffer or modesetting, just wondering if there's some link for explicit instructions about different modes or disabling that
05:18 karolherbst: chillfan: dmesg and xlog please
05:18 chillfan: one moment :)
05:21 RSpliet: snacks: doesn't ring a bell, sorry
05:21 chillfan: is there some place to pastebin that which will accept files as an input?
05:22 agpu: karolherbst: i can just put "options nouveau config=War00C800_0=1" into a file in /etc/modprobe.d/nouveau.conf right?
05:22 karolherbst: agpu: no, you forgot the dot
05:22 chillfan: nvm i'll copy/paste just a sec
05:22 karolherbst: but even then you need to update your initramfs
05:22 Tom^: and include that file into it
05:23 chillfan: this is my dmesg http://pastebin.com/xpXY3kfR
05:23 Tom^: which isnt done by default :p
05:23 chillfan: will just put the xorg.log in there too
05:23 karolherbst: agpu: ohhh no
05:23 karolherbst: it is fine
05:23 karolherbst: agpu: my fault
05:23 karolherbst: but update initramfs ,)
05:24 karolherbst: chillfan: dmesg looks fine
05:24 chillfan: and the xorg log http://pastebin.com/csCHQMsL
05:24 chillfan: yeah it's just a black screen for the terminal, xorg is working okay just not the console
05:24 chillfan: card is nvidia gtx 780 ti
05:25 karolherbst: chillfan: you mean you get a blakc screen after login?
05:25 karolherbst: chillfan: are you using plasma5 by any chance?
05:25 chillfan: karolherbst: around the time that modules start to load everything goes black, can't see the console coming up anymore, if i hit ctrl + alt + f1 i'll not see a console
05:25 karolherbst: chillfan: ohh okay
05:25 chillfan: no, nothing special, openbox and xorg from debian jessie (well devuan, but more or less same)
05:26 chillfan: if i try to drop down to console, i just see a frozen image of the xserver
05:27 chillfan: kernel is 4.1.13
05:27 Tom^: sounds like something that is fixed already because i dont see any such problems on my 780ti :o
05:27 karolherbst: chillfan: 780 ti?
05:27 chillfan: yeah
05:27 karolherbst: might be because of old kernel
05:27 chillfan: vendor is evga sc
05:27 karolherbst: and something fixed it later
05:27 chillfan: oh i see,
05:27 chillfan: Tom^: which kernel do you use? maybe I'll try and report back :)
05:28 Tom^: 4.3.0
05:28 karolherbst: chillfan: there wasn't mach chance to test on 780 ti cards
05:28 karolherbst: because for some reasons, most of the 780 ti owner just used the nvidia driver :D
05:28 chillfan:hides a guilty look
05:29 karolherbst: the 780 ti card is a bit different compared to other keplers
05:29 karolherbst: or maybe not
05:29 karolherbst: but it is something which isn't well tested
05:29 Tom^: as long as i get the card to 0f with proper volt i see no usage of the blob.
05:29 Tom^: =D
05:29 chillfan: yeah i thought it might be, as it didn't seem to have the same code name
05:29 karolherbst: Tom^: well
05:29 karolherbst: Tom^: you can only expect around 70% perf
05:30 chillfan: 70% might be enough for me depending what i test
05:30 chillfan: or use
05:30 Tom^: karolherbst: well now, nouveau on 0a ran games better in wine-nine then nvida blob did on same perf levels.
05:30 Tom^: karolherbst: so thats left to be tested. :p
05:30 chillfan: hm
05:31 chillfan: 0a, is this something to do with reclocking?
05:31 karolherbst: Tom^: yeah, but this is no fair comparison ;)
05:31 karolherbst: chillfan: yeah
05:31 Tom^: chillfan: mm
05:31 karolherbst: chillfan: mid perf level
05:31 chillfan: karolherbst: interesting, i'll have to check the docs for that
05:31 karolherbst: which docs?
05:31 Tom^: chillfan: its a bit work in progress on the 780ti
05:32 chillfan: like er, enabling instructions
05:32 chillfan: Tom^: which is okay if it does something for now, if i get enough fps in just a couple of games i'm happy
05:33 karolherbst: chillfan: well on the highest pstate you can expect 70% performance compared to the blob
05:33 chillfan: glad to see the xserver start for my card, thought it was a no go when i got black screen
05:33 karolherbst: but for this, you really need the 4.4 kernel or self compiled ;)
05:34 Tom^: and it wont quite yet run on highest pstate because of being to low volt :p
05:34 karolherbst: chillfan: ohh so it works with a newer kernel?
05:34 chillfan: karolherbst: the xserver starts
05:34 chillfan: in 4.1 kernel
05:34 chillfan: haven't built 4.3 yet
05:34 karolherbst: k
05:35 chillfan: guessing i should be on the edge to get the proper experience
05:36 agpu: it works! :D
05:37 karolherbst: agpu: nice
05:37 chillfan: honestly though, i expect console issues with anything to do with nvidia, it has always hated the frame buffer
05:37 agpu: thanks karolherbst
05:37 chillfan: for all cards i've tried
05:37 karolherbst: agpu: we need some info no
05:37 karolherbst: *now
05:37 karolherbst: agpu: so we can enable that quirk upstream
05:37 karolherbst: chillfan: lspci -s 01:00.0 -vv
05:37 agpu: ok np what do you need
05:37 karolherbst: I hope the gpu is at 01:00.0?
05:38 karolherbst: ohh wrong
05:38 chillfan: wrong name?
05:38 chillfan: my gpu is on that
05:38 karolherbst: lspci -s 01:00.0 -nnv
05:38 karolherbst: second line
05:38 karolherbst: there is something like [number:number]
05:38 karolherbst: this is what I need
05:39 chillfan: from me or agpu?
05:39 karolherbst: agpu:
05:39 chillfan: k got confused :D
05:39 karolherbst: yeah, my mistake
05:39 karolherbst: agpu: lspci -s 01:00.0 -nnv
05:39 karolherbst: :D
05:39 karolherbst: sorry
05:39 chillfan: no worries
05:40 chillfan: should my kernel maybe contain anything other than the standard options nouveau enables?
05:40 chillfan: i generally turn off all other framebuffer stuff
05:40 Tom^: nouveau.pstate=1 :p
05:40 chillfan: just to check i'm not misconfiguring it i mean
05:40 karolherbst: chillfan: well you need the nouveaudrmfb ;)
05:40 Tom^: oh you mean the kern configs
05:41 karolherbst: chillfan: but I doubt you can disable it
05:41 karolherbst: chillfan: anyway, dmesg output looks fine
05:41 karolherbst: chillfan: I just assume it is something messy
05:42 karolherbst: "(II) NOUVEAU(0): Setting screen physical size to 508 x 285" is this normal?
05:42 chillfan: hm no
05:43 karolherbst: yeah well I don't know if these are dots or inch or whatever :D
05:43 chillfan: xrandr says display size is 1920x1080
05:43 karolherbst: chillfan: end of "connected primary" line
05:43 karolherbst: I think this size is in mm
05:43 karolherbst: so it shall be fine
05:43 agpu: karolherbst: https://paste.fedoraproject.org/295365/87182151/
05:43 karolherbst: chillfan: right, maybe give me your xrandr output
05:44 karolherbst: agpu: ahh also a clevo
05:44 karolherbst: damn a 870m :/
05:44 agpu: why? ^^
05:44 chillfan: http://pastebin.com/HRmeG43L
05:44 karolherbst: agpu: no, it is just nearly the same I have most likely
05:44 chillfan: i think it should be using 144Hz too as per the xorg.conf
05:45 karolherbst: chillfan: ohhhh
05:45 chillfan: but i guess one step at a time
05:45 karolherbst: you have a xorg.conf?
05:45 karolherbst: please paste that too
05:45 chillfan: hah will do
05:46 chillfan: http://pastebin.com/N4xAwkx6
05:47 karolherbst: agpu: clevo P157sm?
05:48 karolherbst: chillfan: there seems to be nothing important in that xorg.conf file
05:48 karolherbst: chillfan: remove it then :p
05:48 karolherbst: I guess you have the nvidia stuff removed or disabled or something?
05:48 agpu: karolherbst: no, P150SM-A
05:48 karolherbst: agpu: ohh this one
05:49 chillfan: probably should, yeah nvidia driver made the file
05:49 chillfan: i just haven't tidied it up yet
05:49 chillfan: or removed it as i might be able to, was leaving it in case it sets my refresh for me
05:49 agpu: karolherbst: its a rebrand actually. but i think it doesnt make a difference
05:49 chillfan: anyway just built 4.3 so might boot to that in a sec
05:50 karolherbst: agpu: did you build the kernel or nouveau yourself?
05:50 karolherbst: agpu: no,not really
05:51 agpu: karolherbst: i installed the kernel + mesa from fedora rawhide Linux fedora 4.4.0-0.rc2.git2.1.fc24.x86_64
05:51 karolherbst: ahh
05:51 karolherbst: then it doesn't matter
05:52 agpu: i thought it would help to update because there was a lot of news around nouveau currently
05:52 karolherbst: ohh me
05:52 karolherbst: one line and I really did a mistake
05:53 chillfan: ok I can see a new option in linux 4.3
05:53 chillfan: CONFIG_DRM_FBDEV_EMULATION for xfree86 dri
05:53 agpu: karolherbst: should i test reclocking?
05:54 karolherbst: agpu: yes
05:54 chillfan: is that maybe something to do with it? I guess I want this disabled since I don't care about it, but maybe that's the feature i was missing?
05:54 karolherbst: agpu: http://lists.freedesktop.org/archives/nouveau/2015-November/023397.html
05:54 karolherbst: agpu: when this lands, you don't need the option anymore
05:54 karolherbst: chillfan: no clue
05:55 chillfan: well only one way to find out i guess, brb then :)
05:56 agpu: hehe nice
05:56 chillfan: well, same issue for now, will tinker with some options see what changes
05:57 karolherbst: chillfan: did you remove the xorg.conf?
05:57 chillfan: let me try that brb
05:59 agpu: i will reboot with reclocking enabled brb
06:00 chillfan: ok, with that removed still the same
06:00 chillfan: bear in mind, xserver totally working
06:00 karolherbst: so just the dispaly is black?
06:00 chillfan: no display isn't black, i talk to you from xserver now
06:01 karolherbst: ohh okay, so what is then the issue? :/
06:01 chillfan: but you know when you boot and you have the console, lines with [ ok ] or [ done ] etc
06:01 chillfan: that is what's black
06:01 chillfan: so if i stopped the xserver, i'd lose access to the display
06:02 karolherbst: not even a blinking cursor in the top left corner?
06:02 chillfan: nope
06:02 karolherbst: does the display go off or is there just nothing on it?
06:02 chillfan: on the 3.18 kernel i was using previously, the monitor reported no signal
06:02 chillfan: but on 4.1 monitor is clearly getting a signal but no display
06:03 chillfan: and same currently on 4.3
06:03 Tom^: are you using hdmi or displayport?
06:03 chillfan: i am using dvi-d
06:03 Tom^: then *shrug* because i use dvi too :<
06:03 Tom^: heck even two monitors and they both get mirrored as fast as kernel initramfs is loaded
06:04 chillfan: well my usual suspect for things like this is modesetting or fb driver
06:04 karolherbst: chillfan: you don't have any kernel command line or something?
06:05 Tom^: or does fedora try to load some bootsplash that isnt showing?
06:05 chillfan: karolherbst: what i see is the kernel output right up until the point the nouveau module loads
06:05 chillfan: Tom^: i'm using devuan (debian with sysvinit)
06:05 Tom^: *debian
06:05 Tom^: :p
06:05 chillfan: basically yes
06:06 Tom^: no i meant *debian to my fedora
06:06 chillfan: oh hah, no there's no graphical splashy thing
06:06 agpu: is AC the current power level?
06:07 chillfan: the AC?
06:07 karolherbst: agpu: current clock
06:07 chillfan: oh sorry my mistake, thought that was directed at me heh
06:07 karolherbst: agpu: but this line also display whether you are on battery or not
06:07 agpu: https://paste.fedoraproject.org/295366/19668144/
06:08 chillfan: ok just made some changes brb :)
06:08 karolherbst: when it is 0, the gpu is off
06:08 agpu: lol makes sense haah
06:09 agpu: AC: core 324 MHz memory 648 MHz (by default)
06:09 chillfan: ok so monitor reports no signal after nouveau driver loads, but signal comes back when xserver starts
06:10 karolherbst: chillfan: yeah, then there is something odd with where the kernel wants to display the stuff
06:10 karolherbst: chillfan: what cpu do you have?
06:10 chillfan: intel i5 4690k
06:10 chillfan: with the intel graphics disabled in bios
06:10 karolherbst: mhhh
06:11 karolherbst: the board doens't have dvi-d?
06:11 chillfan: the graphics board is dvi-d
06:11 karolherbst: nono, the motherboard I mean
06:11 karolherbst: not the nvidia gpu
06:11 chillfan: let me check
06:12 karolherbst: mhh, but when the intel gpu is disabled in the bios, strange
06:13 karolherbst: chillfan: usually for video accell and stuff you are far more better of with the intel gpu for general usage
06:13 agpu: looks like it worked but im not 100% sure
06:13 agpu: 0f: core 405-967 MHz memory 5000 MHz AC DC *
06:13 karolherbst: chillfan: prime offloading should also work on a desktop machine
06:14 karolherbst: chillfan: maybe the kernel does try to put the signal somewhere else
06:14 karolherbst: chillfan: you might want to just plug the monitor into different ports or something
06:14 chillfan: hm the motherboard supports dvi-d, the card is dual head, perhaps.. it mistakes?
06:15 karolherbst: maybe
06:15 Tom^: agpu: the line under that reports current. that * is just not to be trusted :p
06:15 agpu: AC: core 966 MHz memory 5000 MHz
06:15 agpu: so it did work
06:15 agpu: i guess
06:16 chillfan: well just checked now, no signal at all on the other dvi-d
06:16 Tom^: indeed, now it just depends on if you got high enough voltage for that clock or not. for me i didnt have it. :p
06:16 karolherbst: Tom^: ohh he will for sure
06:16 karolherbst: Tom^: on PWM you just get max voltage if you go beyond what is supported
06:17 karolherbst: :D
06:17 karolherbst: but those cards hardly ever use above 1V
06:17 chillfan: ok i'll try to work around it by disable modeset
06:17 karolherbst: chillfan: no
06:17 agpu: why doesnt my fan go to full rotation though?
06:17 Tom^: karolherbst: pwm?
06:17 karolherbst: chillfan: disabling modeset is not a fix, nor a workaround
06:17 karolherbst: it is: disable accell on my gpu
06:17 karolherbst: so no opengl at all ;)
06:18 chillfan: damn
06:18 karolherbst: I have no idea what could causes that, maybe imirkin knows more about this
06:19 karolherbst: Tom^: yeah well, later gpus aren't volting through GPIOs anymore, but a pwm
06:19 Agiofws: image link: https://i.imgur.com/7KuT4oa.png
06:19 Tom^: karolherbst: ah ok.
06:19 karolherbst: Tom^: some keplers and all maxwells
06:20 Tom^: chillfan: you dont have an optiom like this that sets primary GPU ? https://i.ytimg.com/vi/BtDkczrf5fI/maxresdefault.jpg
06:20 chillfan: Tom^: no i just disable the intel stuff and all works
06:24 karolherbst: chillfan: you should know that you only get very poort video acceleration on your card
06:24 karolherbst: with nouveau
06:24 Tom^: karolherbst: this is what im thinking https://gist.github.com/anonymous/4d727ec1e5bed7158254
06:25 chillfan: alright, i'll have a try and see what happens anyway
06:25 chillfan: more of a test to see what i can get out nouveau atm anyway
06:27 chillfan: https://wiki.freedesktop.org/nouveau/KernelModeSetting/ this seems to suggest i have a misconfigured kernel
06:28 chillfan: can i possibly do without fb?
06:28 chillfan: ah nvm i'll tinker with it, starting with fb on
06:33 chillfan: nouveau.pstate=1 should be in lilo/grub?
06:34 chillfan: bbs :)
06:37 agpu: i somehow managed to break it
06:38 agpu: https://paste.fedoraproject.org/295371/44872147/
06:38 agpu: no it looks like DRI_PRIME doesnt work anymore ^^
06:38 agpu: now
06:41 karolherbst: agpu: yeah well, that happens sometimes
06:41 karolherbst: reclocking isn't rock stable currently
06:42 karolherbst: but this looks like voltage too low
06:42 karolherbst: agpu: may I have your vbios?
07:16 karolherbst: Tom^: did you try it out with a higher voltage?
07:16 sdfgdgd: hello
07:20 sdfgdgd: someone just gave me a nv34 device but i tried using it and its very slow and i saw in the feature matrix that all power mangement its missing. Is there anything that could be done to help with it?
07:21 karolherbst: mupuf_: what do you think about a NvChipQual config options, which accepts values for 0-100 and determines which voltage nouveau will use. Default will be something like 25, which means nouveau will use (volt.max * 75 + volt.min * 25) / 100 as the voltage
07:21 pmoreau: mwk: Ping for #11, #15 (and why not #27 as well, but preferably the first two ones O:-) )
07:57 mupuf_: karolherbst: you want to add a workaround instead of spending the time to study voltage management?
08:04 chillfan: so i think i found what is wrong
08:04 chillfan: as it's now working
08:05 chillfan: i think the nouveafb driver expects an fbdriver already there and loaded
08:05 chillfan: is it not designed to work with just 'vga text console'?
08:06 chillfan: nvm working, i'll have a tinker with more options later :)
08:09 karolherbst: mupuf_: not really
08:09 karolherbst: mupuf_: I was thinking that if a gpu has a higher quality, then you need a lower voltage for a specific clock
08:09 karolherbst: it would be an option which would enable higher cstates
08:10 karolherbst: if you spend some time yourself figuring out how far you can go
08:15 karolherbst: mupuf_: by the way, do you think these unk0/unk1 values might be power consumption related? https://gist.github.com/karolherbst/28fcfc36013873249077
08:15 karolherbst: like in 0.01W
08:15 karolherbst: the dellm6800 has a power budget of 100W
08:15 karolherbst: mine has 80W
08:17 karolherbst: sadly these are the only two vbios I found with values there :/
08:18 mupuf_: karolherbst: yes, that would be a nice knob, but who would like to do the validation?
08:19 karolherbst: I have no idea, I have to work out why nvafakebios doesn't work for me :/
08:19 karolherbst: second: the blob always clocks to max for me (way beyond the turbo clock"
08:19 karolherbst: )
08:19 karolherbst: base 705, turbo 797, max 862 and blob always choose max
08:20 karolherbst: would be perfect to test with faking vbios without risking anything :/
08:20 karolherbst: but well
08:21 mupuf_: yeah
08:21 mupuf_: I would say try to debug it with nouveau
08:22 karolherbst: mupuf_: when I nvafakebios
08:22 karolherbst: should nvagetbios return the faked one?
08:23 mupuf_: if you use pramin
08:23 karolherbst: okay
08:25 karolherbst: mupuf_: "Invalid checksum. Broken vbios or broken retrieval method?"
08:25 karolherbst: when trying pramin after faking
08:25 mupuf_: hmm
08:25 mupuf_: what the heck!
08:25 karolherbst: "Invalid signature(0x55aa). You may want to try another retrieval method."
08:26 karolherbst: but there seems to be something
08:27 karolherbst: and my changes are thre too
08:29 karolherbst: mupuf_: after 0x10000 the uploaded vbios and the read out vbios differ
08:29 mupuf_: wonderful!
08:29 mupuf_: wait a sec, we may manage to make it work
08:30 karolherbst: and with differ I mean, they differ for real :D
08:30 mwk: pmoreau: I'll do that now; remind me, is there an ordering to these PRs?
08:30 mwk: as in, should I do 11 before 15 or the other way around?
08:33 mupuf_: karolherbst: something is writing on the vbios
08:33 mupuf_: we need to change the vga rom base address
08:33 mupuf_: oh my, the 2d acceleration on maxwell is funky
08:34 mupuf_: glamor does not seem perfect at all :D
08:34 karolherbst: mupuf_: okay, so what does that then mean?
08:34 imirkin: gnurou: ah ok. well it would save me some time if you could document the new descriptors... otherwise i'll have to find time to RE it
08:34 karolherbst: mupuf_: it doesn't look like random data though
08:35 karolherbst: mupuf_: it looks like this: https://gist.github.com/karolherbst/ed5051e63a3116456113
08:35 mupuf_: karolherbst: http://pastebin.com/4S5FbZ2F
08:35 mupuf_: this is on my nve6
08:35 imirkin: gnurou: which probably won't be until mupuf gets a GM20x into reator (or i happen to get one)
08:35 mupuf_: see, the blob reads from the ROM_WINDOW
08:35 mupuf_: in my case, it reads beef
08:35 mupuf_: and then it gives up
08:36 mupuf_: and tries reading from PROM instead ... and reads 55aa
08:36 mupuf_: or aa55
08:36 karolherbst: :O
08:36 karolherbst: somehow
08:36 karolherbst: I know that 55aa
08:36 karolherbst: thats a checksum
08:36 imirkin: no, it's the pci rom header
08:36 mupuf_: nope, that's not a checksum
08:36 karolherbst: no? mhh
08:36 mupuf_: exactly
08:37 karolherbst: mhh okay
08:37 karolherbst: I thought I already saw that number, oh well
08:37 gnurou: imirkin: my goal is to make things easy for you guys, so I'll try to document as well. no promises though as it is not my decision to do so.
08:38 mupuf_: so, I would like you to mmiotrace the blob after nvafakebios to make sure it picks up the right bios and that the checksum indeed fails
08:38 mupuf_: and then, we can be crazy and try to pick up another linear address to upload the bios to
08:38 karolherbst: mhhhh
08:38 imirkin: gnurou: ok, well as you can imagine, my priority is based around real use-cases, and for now, GM20x is not a real use-case for nouveau. when it becomes one, i'll spend some time on it :)
08:38 karolherbst: I hope mmiotracing works now again
08:38 mupuf_: karolherbst: does that sound crazy-enough for you?
08:39 karolherbst: no, I have no clue how that pramin stuff works or what it is, so everything sounds like a good idea to me anyway
08:39 karolherbst: :p
08:41 Yoshimo: imirkin: there are many cards that need work, you can't do them all
08:41 imirkin: Yoshimo: meh, it's not so bad -- they tend to be in (large) groups
08:42 imirkin: Yoshimo: it's just that there's no way to get GM20x working with nouveau, so i don't need to worry about it
08:42 Yoshimo: i am always in favor of doing one group of cards after the other
08:43 gnurou: imirkin: I understand and apologize for the inconveniences we are introducing. I also hope you understand that there are several nvidians fighting for your side. ;)
08:43 imirkin: gnurou: thanks :) and i'm in no way blaming you... merely explaining my position.
08:44 gnurou: imirkin: fully understand that. we have the same goal.
08:45 karolherbst: mupuf_: thing is, my laptop went a bit crazy after I tried to fake the vbios, maybe it was bbswtich related or something, because it always happens after I power on my gpu
08:45 pmoreau: mwk: Sorry, hadn't seen you had pinged me. The renaming should be done first, as the EVO relies on it (IIRC).
08:45 mwk: pmoreau: whoops, I already started going through the EVO ones
08:45 mwk: I'll review these commit-by-commit FWIW
08:45 gnurou: imirkin: and we probably share more frustrations than you think ;)
08:46 mupuf_: karolherbst: that's what happens when you write over memory of another process :p
08:46 pmoreau: mwk: Yeah, I saw your comment: this is what made me realise I should check IRC. :-D
08:46 pmoreau: mwk: I thought I had removed all those names using a '.' in them, but looks like some survived. :-/
08:49 karolherbst: mupuf_: I don't know, somehting really bad happens
08:49 karolherbst: mupuf_: display turns black, networks seems to be down and sysrq keys doesn't work anymore
08:49 karolherbst: but there is also no crash in pstore
08:50 mupuf_: karolherbst: do you understand what is happening?
08:50 karolherbst: no
08:50 mupuf_: imagine you are writing over some memory that is not yours
08:50 karolherbst: i see
08:50 mupuf_: now imagine if it was kernel memory
08:50 mupuf_: what could go right, right?
08:51 karolherbst: so my kernel just goes into a state worse than the worst
08:54 karolherbst: ohh yay, filesystem corruptions
08:55 mwk: pmoreau: I have a few minor comments to the EVO patchset, it's good to merge once you fix them
08:56 pmoreau: mwk: Cool! I'm working on fixing them
08:58 karolherbst: mupuf_: so, what could I try out?
09:03 mupuf_: karolherbst: well, you would have to allocate memory for it and get its linear address
09:03 mupuf_: which means you may have to write a kernel module that does a kmalloc
09:03 mupuf_: and then ask for its linear address
09:15 Tom^: karolherbst: nope was streaming hockey, SHL. but im gonna do it now.
09:16 pmoreau: mwk: 0x100 doesn't exist in the documentation of the EVO base channel, for none on the cards.
09:16 Tom^: karolherbst: so unless you find any issues with this change https://gist.github.com/anonymous/044bb05ead2a3bd1bd75 im going for it! i hope it wont burn up :p
09:18 mwk: pmoreau: then the documentation is incomplete
09:18 pmoreau: mwk: But there is one on the Overlay channel: `SET_COMPOSITION_CONTROL`
09:19 pmoreau: for all cards
09:19 mwk: I've scanned all the EVO channels
09:19 mwk: 0x100 is clearly something, it doesn't give the ILLEGAL_METHOD error like all the others
09:19 mwk: and there's an earlier comment saying that blob pokes something into it
09:20 mwk: it might be a method that nv doesn't want to document, eg. something dealing with HDCP
09:20 pmoreau: :-/
09:20 pmoreau: Could be
09:20 mwk: so leave it in, maybe with a comment that it's missing from nv docs
09:21 pmoreau: gnurou, tagr: Any idea what could be method 0x100 of the EVO base channel? It is not present in the header files you released earlier this year.
09:21 pmoreau: mwk: I'll do that
09:21 mwk: we're talking G80, for reference
09:24 Tom^: karolherbst: yea that didnt work, clk failed to raise voltage -50000
09:25 karolherbst: Tom^: ohh right
09:25 karolherbst: 0 - 50000
09:26 Tom^: you mean info.max can be 0 at times?
09:29 karolherbst: yeah
09:32 Tom^: karolherbst: easily fixed, if(info.max <= 0) return info.max; :p
09:32 Tom^: now im at 1.18 V
09:32 Tom^: time to see if it borks
09:35 Tom^: karolherbst: hm however it put me on 1307mhz now, so not sure i dare max it out. :p
09:40 chillfan: alright, looks like all working and just bad configured kernel before heh
09:42 pmoreau: mwk: :-( My push removed all your comments… For which method did you wanted me to add an 'r' access attribute? The `FREE` one? (In https://github.com/pierremoreau/envytools/commit/c7f61231db1b28237204d101ef922d58b94483ce)
09:44 chillfan: so erm, how can i set the pstate, check temps, change fan control, etc?
09:50 pmoreau: mwk: Thankfully, I can still see your comment by accessing the Github notification. \o/ Going to fix that one and repush
09:50 Tom^: karolherbst: hm any idea why the lowest cstate available on 0f is 40 which still is a bit to high mhz.
09:57 karolherbst: Tom^: yes, because you messed with the voltage calculation ;)
09:57 Tom^: ;_;
09:57 pmoreau: mwk: I should have fixed all of your comments for #11. :-)
09:57 pmoreau: Thank you for your review!
09:59 karolherbst: Tom^: but it should work fine
09:59 Tom^: karolherbst: well unigine heaven ran at stable 60 - 80 fps at high settings on 1920x1080 before freezing. CTSX_TIMEOUT or similiar
10:00 Tom^: so before i consider it a bug somewhere i rather be at max clocks i know works. :p
10:00 mwk: pmoreau: strictly speaking FREE is *not* a method, since it's read-only, as opposed to write-only :)
10:00 mwk: pmoreau: good job, merge it
10:01 mwk: I don't see any dependency on 15, so let's get it over with
10:01 pmoreau: mwk: Oh! Great!
10:01 Tom^: karolherbst: does the driver control the fan or is that internal? because i didnt notice it speedup any before it froze. perhaps it simply just overheated?
10:01 pmoreau: mwk: Well, there is an implicit dependency since it uses the BLOCKLINEAR notation from #15. ;-)
10:01 karolherbst: Tom^: mhh I guess the fan turns up, because the gpu crashed
10:02 Tom^: well no it didnt speedup at all
10:02 mwk: pmoreau: meh :)
10:02 Tom^: is what i meant
10:02 karolherbst: Tom^: I doubt it overheated though
10:02 Tom^: yea well perhaps misguided term, but more of to hot for these volts and mhz.
10:02 Tom^: at 100% usage :P
10:03 karolherbst: you could monitor it through hwmon, but I doubt that it was heat related
10:03 Tom^: because it ran for quite a bit, compared to when i didnt have volts raised where it simply froze instantly
10:05 karolherbst: it could be that something is still missing somewhere
10:05 karolherbst: Tom^: could you try to run the gpu at the same clock as under windows?
10:05 karolherbst: or the blob
10:05 Tom^: well i cant, because lowest cstate is 40 which in turn is 1164 mhz
10:05 Tom^: ~80mhz more then blob/stock at highest boost state
10:06 karolherbst: mhhh
10:06 karolherbst: why is 40 the lowest ?
10:06 chillfan: hm does nouveau use it87?
10:06 Tom^: karolherbst: you said it yourself, because i broke my volt calculations. xD
10:07 mwk: finally :)
10:07 karolherbst: Tom^: yeah, then unbreak it :p
10:07 Tom^: karolherbst: not sure, https://gist.github.com/anonymous/592e0117a8b34370c74c
10:07 mwk: now, #15...
10:07 Tom^: but then im back on to low volt, and it freezes because of that
10:07 pmoreau: mwk: Yup! I was wondering whether it was going to celebrate its first birthday or not. :-D
10:08 karolherbst: Tom^: you could just remove the volt adjustement
10:08 karolherbst: or check if the result is 0 before doing so
10:08 Tom^: which i am
10:09 karolherbst: something seems wrong then
10:09 Tom^: karolherbst: https://gist.github.com/anonymous/da379fc0f9b49879f7ad
10:10 Tom^: heh seems you use tabs meanwhile i expand tabs to spaces :>
10:10 karolherbst: Tom^: well kernel stuff says 8 spaces is a tab
10:11 karolherbst: mhhhh
10:11 karolherbst: something seems to be really odd
10:13 karolherbst: ohh maybe I did something wrong :o
10:13 Tom^: \o/ if so im glad my experiments made you find it.
10:19 karolherbst: Tom^: would you like to boot with config=volt=debug and give me the dmesg output?
10:20 Tom^: with my quirk?
10:20 karolherbst: doesn't matter
10:23 karolherbst: Tom^: I think I know what is wrong
10:23 karolherbst: but I want to confirm this with your dmesg
10:23 Tom^: uhm wasnt it nouveau.config=volt=debug
10:24 Tom^: because i dont see any changes in dmesg https://gist.github.com/anonymous/94da39948e75503b54ae besides my cmdline
10:24 karolherbst: ohhh right
10:24 karolherbst: mhhh
10:24 karolherbst: I am dumb
10:24 karolherbst: debug=volt=debug
10:24 karolherbst: :D
10:25 Tom^: ah
10:26 Tom^: now :p https://gist.github.com/anonymous/683066d2b455ce69e0da
10:29 karolherbst: this looks totally fine though: min: 825000uv max: 1212500uv
10:35 karolherbst: Tom^: could you open drm/nouveau/nvkm/subdev/clk/base.c
10:35 karolherbst: there should be a line with "if (volt && (voltage > volt->max_voltage || voltage < volt->min_voltage))"
10:36 karolherbst: remove that if for now and reboot with those changes
10:36 karolherbst: also the return after that if has to be removed
10:37 Tom^: mm
10:38 Tom^: i hope this wont put me in some insane cstate when i got to pstate 0f :p
10:38 karolherbst: no
10:42 Tom^: karolherbst: sure that gave me the cstates back just that it complains about failing to raise voltage -37500 and i suspect its because of my quirk. it cant reduce the voltage :p
10:43 karolherbst: then there is still something fishy then
10:44 Tom^: i mean im going to 0f which puts me in a high cstate with higher voltage, then i try to put it to cstate 35, which should reduce it a bit. but my quirk foobars the calculations
10:44 karolherbst: yeah, but your change messes something up
10:44 mwk: pmoreau: I've gone through #15, there are also a few fixes, mostly related to the "unit" word
10:44 mwk: which should be block
10:45 karolherbst: Tom^: ohhh I know
10:45 pmoreau: mwk: Thanks a lot! I'll have a look at your comments after dinner. :-)
10:46 karolherbst: Tom^: instead of if(info.max <= 0), do if(info.max <= 875000)
10:46 karolherbst: and then revert this change in clk/base.c
10:46 Tom^: ok
10:47 Tom^: or perhaps even info.max <= info.min ?
10:48 karolherbst: no
10:51 Tom^: karolherbst: sort of worked just that im on 1.21V now. you ruined my quirk ;_; =D
10:52 karolherbst: mhhh
10:52 karolherbst: select a proper cstate
10:52 karolherbst: Tom^: well this function calculates the voltage used for a voltage id, it doesn't effect the max voltage used at all
10:52 Tom^: yea i did, which is what put me on 1.21 too :p
10:52 karolherbst: select a lower one then :p
10:53 Tom^: but my mhz :<
10:54 Tom^: oh well there we go, 1088mhz and 1.18v
10:54 karolherbst: right
10:54 Tom^: time to see if it freezes or not
10:56 Tom^: karolherbst: http://i.imgur.com/hAG4ILA.jpg
10:57 Tom^: =D
10:57 karolherbst: mhhh 77
10:57 karolherbst: what settings?
10:57 Tom^: however im gonna have to see what temps this card usually is on because i aborted at 87C when it still didnt had increased fan speed
10:57 karolherbst: :O
10:57 karolherbst: ohhhh
10:57 karolherbst: yeah well
10:58 karolherbst: that doesn't sound right
10:58 Tom^: http://i.imgur.com/iUnHBZE.png settings
10:59 Tom^: so awesome fps tbh, :p
10:59 karolherbst: mehr
10:59 karolherbst: meh
11:00 karolherbst: I only get 21 fps there
11:00 Tom^: haha
11:00 Tom^: "Maximum temperature for a 780Ti regardless of model is 95c." so yea, fan control is borked. :p
11:03 Tom^:achievment earned: world first nouveau 80+ fps.
11:03 karolherbst: :p
11:03 Tom^: on unigine heaven.
11:04 Tom^: no idea where lm_sensors get the "high, crit, emerg" temps from but those are all way , way to high.
11:04 karolherbst: yeah I know
11:04 karolherbst: but the more important issue is, is that nouveau can't control your fan
11:04 Tom^: is nouveau perhaps just doing the same that it doesnt find my gpu hot so it doesnt increase the fan?
11:05 karolherbst: Tom^: no idea
11:05 karolherbst: I have to go now anyway, maybe mupuf_ wants to help you figure this one out
11:07 Tom^: yea temp_max etc are all way to high in /sys/devices/pci0000\:00/0000\:00\:01.0/0000\:01\:00.0/hwmon/hwmon2/t
11:07 karolherbst: yeah but that doesn't matter much
11:07 karolherbst: can you contrl the fan manually there?
11:07 Tom^: sure if i set pwm1_enabled to 0 and pwm1 to the percent it should run at
11:09 Tom^: hm nop, Invalid argument when trying to change pwm1
11:11 karolherbst: well we can look into this tomorrow maybe
11:11 Tom^: sure
11:12 karolherbst: at least I haven't touched anything fan related for now
13:02 pmoreau: mwk: I pushed two new commits taking care of your comments, to make it easier to review the additional changes than if I had rebased everything.
13:03 pmoreau: mwk: I'll squash them into the commits they fix (and rebase on master) before merging.
13:08 koz_: If my card will only reclock stably to 0a and no higher, is there anything much I can do about it?
13:08 RSpliet: koz: first thing you want to try is the freshest kernel you can possibly imagine
13:08 koz_: RSpliet: So 4.4rc-whatever?
13:09 RSpliet: I think that would be a good start yes
13:09 pmoreau: koz_: If it's a Fermi, complain to / ping RSpliet, if it's a Kepler, complain to / ping karolherbst. :-D
13:10 koz_: pmoreau: I guess karolherbst is getting a complaint when he next appears. :P
13:10 pmoreau:searches for popcorn
13:10 koz_: But even on 0a, the card is pretty solid (680).
13:10 Yoshimo: that guy is working his ass off, cut him some slack
13:10 RSpliet: Yoshimo: pmoreau wasn't entirely serious ;-)
13:11 RSpliet: pmoreau_: you can complain about Fermi all you want, that's not going to make a difference :-D
13:11 RSpliet: koz_: which kernel are you using then?
13:11 pmoreau: Yoshimo: I know how much Karol is involved!
13:12 RSpliet:throws some popcorn in pmoreau's clevage
13:12 pmoreau: RSpliet: :-/ It seemed to help to get reclocking working on my G96 ;-)
13:12 Yoshimo: i will join the phoronix club and complain about missing maxwell firmware then ;)
13:12 koz_: RSpliet: 4.2.something IIRC. Let me check.
13:12 RSpliet: hahaha, well, you get that privelege when you send me a card :-D
13:13 RSpliet: koz_: then fire up a 4.4-rcX kernel
13:13 koz_: RSpliet: Well, *compile*, then fire up.
13:13 RSpliet: koz_: well, if you're on Fedora you could grab one from Koji :-)
13:13 pmoreau: Yoshimo: Give gnurou and the other NVIDIA guys some rest. ;-)
13:13 koz_: RSpliet: Arch. :P
13:14 koz_: I guess we've heard nothing about the firmware recently, then?
13:14 RSpliet: fair enough, anyway, 4.4 has Karol's fix in, gets it going for a lot of cards
13:14 Yoshimo: eventually it will come, i have a gtx 280 and a 560ti to messwith until then
13:14 RSpliet: if not, you have a weird voltage regulator and some work is somewhere in the pipeline
13:14 pmoreau: RSpliet: Meh… I only have one Fermi, that you already have though. I should buy a few and send some to you, or set up a Reator. :-)
13:15 koz_: RSpliet: Will I need to update Mesa and libDRM as well to match?
13:15 RSpliet: pmoreau: I'm playing with my Fermi for now, once it's up and running I know there's a 85% solution
13:15 RSpliet: from then, I can look at other cards :-)
13:15 RSpliet: koz_: don't think so no
13:15 koz_: RSpliet: OK, that's good then.
13:16 pmoreau: koz_: There were some comments this morning hinting at some internal progress, but it'll probably take some more time before we see it.
13:16 koz_: pmoreau: Ah. Well, I'm merely curious.
13:16 koz_: I don't have any great personal investment in getting Maxwells up and going - kepler4lyf.
13:17 koz_:flashes gang sign.
13:17 pmoreau: :-)
13:17 pmoreau: I hope to make use of my 960 at some point, and of the Pascal I'll buy when they come out.
13:18 RSpliet:goes and gets some drink to help me RE that mem reclocking sequence
13:18 koz_: pmoreau: I have a Maxwell in the laptop I'm on, but I don't really need the power behind it.
13:18 koz_: My desktop is powered by an EVGA GeForce GTX 680.
13:19 pmoreau:is hoping to win in his fight against LLVM to get clover up and running again
13:21 Tom^: koz_: what happends when you try to run at 0f?
13:22 koz_: Tom^: It either straight-up doesn't run my monitor, or displays a hideous pixellated display with some kind of PGRAPH error.
13:22 Tom^: ok yes, get 4.4
13:22 Tom^: :p
13:23 koz_: OK, I'll try.
13:23 koz_: My main loss is the memory clock actually.
13:23 koz_: 0a gives me like, two-thirds max core clock.
13:23 koz_: But like, a *quarter* max memory clock.
13:25 Tom^: my only problem is the fan isnt ramping up so i cant run at 0f :<
13:25 Tom^: otherwise i had 77fps on unigine heaven on high at 1920x1080
13:26 koz_: Tom^: Nice! I'm trying to get Minetest on max settings, max view distance and 512 textures at 1920 x 1080 to give a consistent 60 frames.
13:27 koz_: Currently, my card can *sorta* manage consistent 30.
13:53 imirkin: Tom^: fan should scale up on its own, although perhaps you don't like the way it ramps up
13:53 imirkin: Tom^: if you flip nouveau into manual control mode, then you can set the percentages yourself
13:53 Tom^: imirkin: it doesnt, :/
13:53 Tom^: imirkin: i was on 87C when i aborted unigine heaven. and still only 1050rpm
13:54 imirkin: Tom^: hmmm... odd. 87C isn't so bad btw...
13:54 Tom^: 90ish or so is the topmost this card runs at 95 is when things go bad.
13:54 imirkin: Tom^: but i would have expected it to have spun up before then... perhaps fan speed adjustment just plain doesn't work? that'd def explain it...
13:55 Tom^: and yea i tried setting pwm1_enable to 0 , however changing pwm1 just gives me invalid argument
13:55 imirkin: mmmm... iirc you have to echo 1 into it? there's a wiki page.
13:55 imirkin: or rather... docs
13:55 imirkin: somewhere
13:57 Tom^: the thing is tho is that temp1_max , temp1_crit and temp1_emergency is way above what it should be
13:57 Tom^: so wouldnt surprise me if thats the only issue and it then thinks the card is cool enough while it isnt.
13:59 Tom^: but then again the only values i find is the ones in static void nvkm_therm_fan_set_defaults in nvkm/subdev/therm/fan.c and those seems reasonable.
14:02 imirkin: those are all read from vbios
14:02 imirkin: perhaps incorrectly
14:05 imirkin: gnurou: i know at least one frustration that i have and you don't -- docs :)
14:06 imirkin: gnurou: and perhaps even more importantly, notes on what's changed generation to generation
14:07 imirkin: mupuf_: i assume you don't have any GM20x?
14:08 Tom^: ok i left it running now, and once it hit 90C the fans went mad and did go 100% for a second or two until it dropped down. to ~84C
14:08 Tom^: :p
14:09 imirkin: there's supposed to be a ramp
14:09 imirkin: it's probably misunderstanding what's going on
14:10 Tom^: yea atleast it can control the fan. heh
14:10 imirkin: this is mupuf_'s favourite area... make the vbios available to him somewhere
14:10 Tom^: "[ 941.444992] nouveau 0000:01:00.0: therm: temperature (90 C) hit the 'fanboost' threshold"
14:10 imirkin: yeah, that's one of the emergency measures
14:10 imirkin: but normally fan speed is supposed to scale so you don't hit that
14:10 imirkin: at least... i think
14:11 Tom^: what timezone does he live in?
14:12 Tom^: so i know when to be online heh
14:15 imirkin: GMT+2 or so... helsinki
14:15 Tom^: oki
14:18 imirkin: koz_: there's an important patch in 4.4-rc1 which should improve your reclocking experience greatly
14:19 Tom^: imirkin: btw http://i.imgur.com/h8wN7Nv.jpg =D =D those fps.
14:21 imirkin: Tom^: that's 76 more than i get...
14:21 imirkin: btw, if you want to see an awesome effect, enable 8x msaa
14:21 Tom^: pstates make one hella difference, on 0a im only getting like 25
14:21 imirkin: yeah, your gpu also cost about 20x more than mine
14:21 Tom^: :P
14:22 Tom^: woah its like being on drugs
14:22 Tom^: O_o
14:23 imirkin: whoa wait, that's 77 fps at 1920x1080? i was talking about the fps i get with 640x480 :)
14:23 Tom^: haha
14:25 Tom^: i think its time to give in, the temperature table from the vbios aint telling me much.
14:35 koz_: imirkin: It did.
14:35 imirkin: koz_: great :)
14:35 koz_: I can now get 30fps consistently even when I turn everything up to 11 in Minetest.
14:35 koz_: Well, ~30fps.
14:35 imirkin: i guess that's good? i don't think i've ever run minetest
14:36 koz_: imirkin: It's pretty good.
14:36 koz_: Given that I have a 680, which is probably the most powerful card I can run that reclocks well.
14:36 Tom^: if only i get the fans working the 780ti is the one. :p
14:37 koz_: Tom^: Is that the thing stopping its reclock being good?
14:37 Tom^: anyways minetest looks like minecraft, how can it even run at 30fps?
14:37 Tom^: its like square blocks with 12x12 textures
14:37 koz_: Tom^: This is with 512 textures, max view distance, and all setting at max.
14:38 imirkin: i guess there are multiple square blocks?
14:38 Tom^: :P
14:39 koz_: Also, do any of you folks know what environment variable I have to set to show Open GL debug info when running (e.g.) Minetest?
14:39 koz_: I remember there being an application-agnostic way...
14:39 imirkin: what sort of debug info were you hoping for?
14:40 koz_: imirkin: Calls. I know Irrlicht has some inefficient calling issues, so I wanna patch mine.
14:40 koz_: I remember someone telling me this one, but I forgot.
14:40 imirkin: mmmmmm there is something
14:40 imirkin: MESA_DEBUG=api maybe?
14:40 koz_: It was something like that, yes.
14:40 koz_: I'll try it.
14:42 imirkin: MESA_VERBOSE=api
14:42 imirkin: only on debug builds though
14:49 koz_: imirkin: GL_DEBUG=verbose was what I sought. Thanks for directing me to it, though.
17:14 karolherbst: :O complaints?
17:15 karolherbst: koz_: if you want to help, vbios pls :p
17:26 karolherbst: koz_: ohh it works, nice :)