08:40 karolherbst: I was told to held a presentation about nouveau for the ccchh by ccchh members :)
08:41 karolherbst: gonna be fun I guess
08:43 mupuf: karolherbst: so, yhou don't have a ticket, but you can talk there :D
08:43 mupuf: hehe, well, congrats!
08:45 karolherbst: :D
08:45 karolherbst: yeah
08:45 karolherbst: it just the hamburg part, but still
08:50 karolherbst: I was there yesterday on their "meeting", it was fun
08:55 karolherbst: mupuf: ohh wait, you didn't understand, it will be in the club building, not the congress
08:55 karolherbst: and I guess also after the congress somewhen
09:05 mupuf: ah, i see
09:07 Yoshimo: hh for hamburg karol?
09:07 Yoshimo: ah slow connection, ignore that
09:20 slingamn: i'm using a GeForce GTX 560 to drive two monitors, under a 4.8.8 kernel and Xorg 1.19 (both from Fedora). i'm having issues with DPMS
09:21 slingamn: specifically, DPMS is enabled in X, but when the screens power off, the primary screen powers back on within a few seconds (the secondary stays off)
09:22 slingamn: this is the case whether DPMS turns on automatically or i use `xset dpms force off`
09:52 karolherbst: slingamn: I guess upon activity the second screens turns on again?
10:25 slingamn: karolherbst: correct
10:58 trusiad: Hi
10:59 trusiad: Since my last update on arch linux, I've got the weirdest bug on gnome-shell
10:59 trusiad: dmesg complains about a 'fail ttm_validate'
10:59 trusiad: then no fonts, no images, sometimes it freezes
11:00 trusiad: http://pastebin.com/D6JQ0ch9
11:00 trusiad: Could someone help me out to get gnome working...
11:00 trusiad: please
11:01 trusiad: It's only on gnome, every other DE is fine
19:24 gluttony1: karolherbst: huhu o/
19:24 karolherbst: hiho
19:26 karolherbst: this will be hell to learn all the nick names
19:27 karolherbst: by the way, what is the best way to check if a source of an instruction is an immediate value for real by not using getImmediate?
19:29 karolherbst: mhh, maybe I could just check for OP == MOV and file == IMMEDIATE, no idea if that would get them all
19:31 karolherbst: mupuf: also waiting for your review regarding that power budget stuff ;) any idea when you will find some time? or should I resend and CC you?
19:31 mupuf: oh, I didn't give you my Rb?
19:31 mupuf: I did review it on sunday
19:31 karolherbst: :D
19:31 karolherbst: mhh
19:31 karolherbst: let me check
19:32 karolherbst: there is one mail from you this month
19:32 karolherbst: so, no
19:32 mupuf: Well, I guess I forgot to tell you on IRC
19:32 karolherbst: ohh I see
19:32 mupuf: I was being too implicit when I said that the next version with the selection of the power budget could wait until we actually enforce it
19:33 karolherbst: :D
19:33 karolherbst: I understand
19:33 mupuf: fix the problem pierre found and I will consider it good-enough for me
19:33 karolherbst: ohh right, have to go over that again
19:33 mupuf: ;)
19:34 karolherbst: or did I already
19:36 karolherbst: ohh right
19:36 karolherbst: I completly reworked that first patch
19:39 karolherbst: is there a proper way to align if (x || y || z) by the way if I have to add a line break?
19:43 karolherbst: mupuf: I will resend the series, cause the changes are too big
20:17 mupuf: I do not have any prefere ce
20:59 karolherbst: mhh I need a gm20x gpu :(
21:00 Yoshimo: physically or just someone with access ?
21:00 karolherbst: pretty much physically
21:01 imirkin_: inside of his head :)
21:02 gluttony1: karolherbst: how long?
21:02 karolherbst: no clue
21:02 karolherbst: more than a month
21:03 karolherbst: I have another idea, let me think
21:04 Yoshimo: it should be rather cheap now to get a 950 or so, as pascal cards pretty much replaced them
21:05 Yoshimo: 120-130 on ebay, 144 on geizhals. Not sure if that is low enough
21:05 karolherbst: and it might break
21:06 gluttony1: karolherbst: ok, i could borrow u mine, but it shouldn't break :D
21:06 Yoshimo: what do you have in mind that might ruin the card karol? We have overheating protection in the card if i remember correctly
21:07 karolherbst: that isn't the problem
21:09 Yoshimo: i can only guess unless you enlighten me
21:26 karolherbst: mhh mhh mhh
21:40 karolherbst: is there a way to check what an application does within an mmaped area?
21:49 mwk: like valgrind-mmt?
21:49 karolherbst: yeah, basically
21:53 karolherbst: I want to intercept any read to a marked region mmaped . Should I just see how valgrind-mmt does it or is there something like this already?
21:53 karolherbst: *every
22:00 karolherbst: okay
22:00 karolherbst: I think I got something usefull
22:00 karolherbst: this will be nice
22:12 karolherbst: or not
22:13 mwk: uh, but valgrind-mmt already does exactly that
22:13 mwk: what's different about your needs?
22:13 karolherbst: I also need to know what addresses I have to filter out later
22:14 karolherbst: or is there a better way than --mmt-trace-all-mem ?
22:14 mwk: IIRC you can mark specific files as traced
22:14 mwk: failing that, just apply some hacks to valgrind-mmt
22:14 karolherbst: ohh wait
22:14 karolherbst: that application opens /dev/mem ;)
22:15 mwk: mmt basically does full cpu simulation, it's not exactly easy to duplicate that
22:15 karolherbst: but maybe that would work indeed
22:15 karolherbst: let me see
22:16 karolherbst: looks good though
22:18 karolherbst: mmt vrashes...
22:18 karolherbst: https://gist.github.com/karolherbst/df1c166dbff07c3f65e73fce162440bd
22:19 karolherbst: --2427-- VALGRIND INTERNAL ERROR: Valgrind received a signal 11 (SIGSEGV) - exiting
22:20 karolherbst: if (!FD_ISSET(fd, &trace_fds))
22:20 karolherbst: ...
22:29 karolherbst: " mmt_map_region (fd=0, start=67108864, end=50308784, offset=4278167320, prot=137132736, flags=1024)"
22:29 karolherbst: this looks wrong....
22:32 karolherbst: \o/
22:34 karolherbst: mhh, or maybe it doesn't work ?
22:34 karolherbst: odd
22:57 karolherbst: skeggsb: don't question this table, cause it is odd and crazy :p