Switch to DuckDuckGo Search
   September 1, 2017  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >

bottom
[00:55:19] *** andy_js <andy_js!~andy@90.193.62.109> has quit IRC (Quit: andy_js)
[01:05:13] *** freakazoid0223 <freakazoid0223!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has joined #oi-dev
[01:06:07] *** freakazoid0223_ <freakazoid0223_!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has quit IRC (Ping timeout: 248 seconds)
[01:17:10] *** freakazoid0223_ <freakazoid0223_!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has joined #oi-dev
[01:19:35] *** freakazoid0223 <freakazoid0223!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has quit IRC (Ping timeout: 240 seconds)
[02:06:56] *** v_a_b1 <v_a_b1!~volker@p5DE8F672.dip0.t-ipconnect.de> has joined #oi-dev
[02:09:11] *** v_a_b <v_a_b!~volker@p5DE8F43F.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 240 seconds)
[02:47:05] *** csb88 <csb88!~Mutter@187.65.52.24> has joined #oi-dev
[02:50:26] *** csb88 <csb88!~Mutter@187.65.52.24> has quit IRC (Quit: Mutter: www.mutterirc.com)
[02:52:56] *** csb88 <csb88!~Mutter@187.65.52.24> has joined #oi-dev
[02:58:50] *** csb88 <csb88!~Mutter@187.65.52.24> has quit IRC (Remote host closed the connection)
[03:51:45] <oibot> Project illumos-gate build #4978: SUCCESS in 1 hr 46 min: https://hipster.openindiana.org/jenkins/job/illumos-gate/4978/
[04:31:21] *** alanc <alanc!~alanc@hqdc-proxy-mwg03-o.oracle.com> has quit IRC (Remote host closed the connection)
[04:31:38] *** alanc <alanc!~alanc@hqdc-proxy-mwg03-o.oracle.com> has joined #oi-dev
[04:31:38] *** ChanServ sets mode: +o alanc
[07:35:47] *** an_ <an_!uid25724@gateway/web/irccloud.com/x-bylvikpriayissja> has quit IRC (Quit: Connection closed for inactivity)
[07:39:37] *** freakazoid0223_ <freakazoid0223_!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has quit IRC (Quit: Why is the alphabet in that order? Is it because of that song?)
[07:50:13] *** tsoome <tsoome!~tsoome@85.253.208.213.cable.starman.ee> has quit IRC (Quit: tsoome)
[08:32:40] *** tsoome <tsoome!~tsoome@178-108-157-37.dyn.estpak.ee> has joined #oi-dev
[09:13:52] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #oi-dev
[09:17:17] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[09:22:53] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[09:23:13] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #oi-dev
[09:32:51] *** mnowak_ <mnowak_!mnowak_@nat/suse/x-okqhkgcrqbrhmekp> has joined #oi-dev
[09:53:03] *** psychicist__ <psychicist__!~psychicis@ip127-8-212-87.adsl2.static.versatel.nl> has quit IRC (Ping timeout: 248 seconds)
[10:04:17] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Quit: Be back later ...)
[10:50:41] *** andy_js <andy_js!~andy@90.193.62.109> has joined #oi-dev
[11:04:46] *** an_ <an_!uid25724@gateway/web/irccloud.com/x-ikccibchwkuvpmuy> has joined #oi-dev
[11:05:12] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[11:11:13] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 246 seconds)
[11:24:26] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[12:33:13] *** csb88 <csb88!~Mutter@187.65.52.24> has joined #oi-dev
[12:41:44] *** csb88 <csb88!~Mutter@187.65.52.24> has quit IRC (Quit: Mutter: www.mutterirc.com)
[12:49:39] <jimklimov> Hi all, I'm thinking about how the nvidia-* variant packages might be mediated so as to not conflict. It is relatively easy for directories, but I'm less certain about kernel modules now - specifically, if we have co-installed /kernel/drv/nvidia-(304|340|385) and a mediated "/kernel/drv/nvdia" symlink to either one of those, will `devfsadm`, module probing, maybe bootadm and other automatics go bonkers still trying to load all 3 binaries?
[12:50:24] <jimklimov> would it be safe to store them in some subdirectory near (or under) the modules tree, and mediate a symlink to that?
[12:50:55] <jimklimov> e.g. /kernel/drv/$(MACH)/.nvidia-implementations/nvidia-(304|340|385)
[12:51:03] <jimklimov> for a rough example
[13:20:51] *** tsoome <tsoome!~tsoome@178-108-157-37.dyn.estpak.ee> has quit IRC (Quit: tsoome)
[13:58:55] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Quit: Be back later ...)
[14:46:46] <Agnar> jimklimov: uhm, not sure if the modules can be loaded via a symlink
[14:47:21] <jimklimov> can a hardlink be mediated? =D
[14:47:32] <Agnar> no idea. needs to be tested
[15:09:24] <Woodstock> there already are symlinks in /kernel/drv
[15:09:53] <Woodstock> that is, /kernel/misc really, but that doesn't matter
[15:10:46] <Woodstock> i wouldn't bother with a special subdirectory, just call them nvidia.{304,340,385} or such
[15:11:59] <Agnar> ok
[15:24:28] <jimklimov> And won't they get loaded "by themselves" somehow, assuming driver_aliases etc. do not point to these suffixed names directly, but use the mediated name?
[15:26:40] <Woodstock> i don't think the kernel will load random stuff just because it's there
[15:26:56] *** tsoome <tsoome!~tsoome@85.253.208.213.cable.starman.ee> has joined #oi-dev
[15:28:05] <jimklimov> Well, maybe there is an equivalent of "modprobe -a" lurking in some scripts? Maybe `devfsadm` (or reconfig-reboots in general) might get confused this way?
[15:28:12] <jimklimov> too many maybes in one phrase ;)
[15:57:31] *** mnowak_ <mnowak_!mnowak_@nat/suse/x-okqhkgcrqbrhmekp> has quit IRC (Ping timeout: 246 seconds)
[15:59:44] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[16:07:40] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 240 seconds)
[16:08:45] *** freakazoid0223 <freakazoid0223!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has joined #oi-dev
[16:11:09] *** alp <alp!~alp@pyhalov.cc.rsu.ru> has quit IRC (Ping timeout: 268 seconds)
[16:59:49] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[17:03:59] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 248 seconds)
[17:21:55] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has joined #oi-dev
[19:21:41] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Ping timeout: 240 seconds)
[20:00:27] *** andy_js <andy_js!~andy@90.193.62.109> has quit IRC (Read error: No route to host)
[20:00:47] *** andy_js <andy_js!~andy@90.193.62.109> has joined #oi-dev
[20:05:13] *** wacki <wacki!~wacki@p5B2C49A4.dip0.t-ipconnect.de> has quit IRC (Quit: Lingo: www.lingoirc.com)
[20:29:29] *** FuntDobra <FuntDobra!~FuntDobra@213.251.226.143> has quit IRC (Quit: Leaving)
[20:30:27] *** FuntDobra <FuntDobra!~FuntDobra@213.251.226.143> has joined #oi-dev
[20:33:03] *** FuntDobraX1Gen5 <FuntDobraX1Gen5!~FuntDobra@213.251.226.143> has joined #oi-dev
[20:40:43] *** FuntDobraX1Gen5 <FuntDobraX1Gen5!~FuntDobra@213.251.226.143> has quit IRC (Ping timeout: 255 seconds)
[22:16:28] *** richlowe <richlowe!~richlowe@2607:fcc8:bcd4:b100::4> has joined #oi-dev
[22:23:17] *** freakazoid0223_ <freakazoid0223_!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has joined #oi-dev
[22:25:05] *** freakazoid0223 <freakazoid0223!~IceChat9@pool-72-94-95-33.phlapa.fios.verizon.net> has quit IRC (Ping timeout: 248 seconds)
[23:38:21] *** rk4n3 <rk4n3!~rk4n3@24-159-210-156.static.roch.mn.charter.com> has joined #oi-dev
[23:56:26] <pjama> @jimklimov: I looked into capability of having different nvidia drivers some time back but backed off because it's a bit out of my league.
[23:56:26] <pjama> I to, have a bunch of old but quite functional nvidia cards I don't want to throw out.
[23:56:26] <pjama> From memory, I think Oracle just have separate versions of nvidia packages that cannot be installed together.
[23:56:26] <pjama> which makes sense because you're not going to be changing your video card that often and if you do it shouldn't be that much more trouble to uninstall one package and install another.
[23:57:06] <pjama> also, I think there are some dependencies on "nvidia" that might need to be rationalised
top

   September 1, 2017  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >