From: k11stan <k1...@gm...> - 2009-01-20 23:16:18
|
hey every1 - occasionally if I do not boot to osx for an extended period of time (roughly a month), the fan behavior (and likely also throttling) seems to get very weird ( the fans randomly rev up , the temps seem to spike , and the whole setup gets really unstable) .. after i boot to osx the problem seems to go away .. is there any way to preserve the nvram values that osx likes to set ? I'm using 2.6.25 with mactel patches and version 8.51.3 of fglrx .. I throttle the card using radeontool ...thanx for any suggestions stan |
From: Jonas P. <cat...@gm...> - 2009-01-21 08:54:13
|
Hi all, On Wed, Jan 21, 2009 at 00:16, k11stan <k1...@gm...> wrote: > hey every1 - occasionally if I do not boot to osx for an extended period of > time (roughly a month), the fan behavior (and likely also throttling) seems > to get very weird ( the fans randomly rev up , the temps seem to spike , and > the whole setup gets really unstable) .. after i boot to osx the problem > seems to go away .. is there any way to preserve the nvram values that osx > likes to set ? I'm using 2.6.25 with mactel patches and version 8.51.3 of > fglrx .. I throttle the card using radeontool ...thanx for any suggestions I guess this is no real help, but for reference I can say that the fans on my ATI-based MBP (2,2) were also acting up now and then. I didn't boot into OSX a lot, so where that is related I have no idea. However, I can say that my current MacBookPro4,1 (nVidia based) is a LOT more sane when it comes to fans - and even when they spin up (due to occasional high load) they don't make nearly as much noise. On the downside, I never had any "applesmc: wait status failed" messages on the old system, whereas it seems that polling this info (fans, temps etc) fail - and then retry successfully - quite often now. I don't know whether this is actual HW differences or if the driver just works differently in newer kernels. BTW: I have not booted OSX on this box for at least 3 months. Hope this was of some use for someone / Jonas |
From: Justin P. M. <jus...@gm...> - 2009-01-21 09:23:50
|
Jonas Petersson wrote: > Hi all, > > On Wed, Jan 21, 2009 at 00:16, k11stan <k1...@gm...> wrote: > >> hey every1 - occasionally if I do not boot to osx for an extended period of >> time (roughly a month), the fan behavior (and likely also throttling) seems >> to get very weird ( the fans randomly rev up , the temps seem to spike , and >> the whole setup gets really unstable) .. after i boot to osx the problem >> seems to go away .. is there any way to preserve the nvram values that osx >> likes to set ? I'm using 2.6.25 with mactel patches and version 8.51.3 of >> fglrx .. I throttle the card using radeontool ...thanx for any suggestions >> > > I guess this is no real help, but for reference I can say that the > fans on my ATI-based MBP (2,2) were also acting up now and then. I > didn't boot into OSX a lot, so where that is related I have no idea. > However, I can say that my current MacBookPro4,1 (nVidia based) is a > LOT more sane when it comes to fans - and even when they spin up (due > to occasional high load) they don't make nearly as much noise. On the > downside, I never had any "applesmc: wait status failed" messages on > the old system, whereas it seems that polling this info (fans, temps > etc) fail - and then retry successfully - quite often now. I don't > know whether this is actual HW differences or if the driver just works > differently in newer kernels. BTW: I have not booted OSX on this box > for at least 3 months. > > Hope this was of some use for someone / Jonas > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by: > SourcForge Community > SourceForge wants to tell your story. > http://p.sf.net/sfu/sf-spreadtheword > _______________________________________________ > Mactel-linux-users mailing list > Mac...@li... > https://lists.sourceforge.net/lists/listinfo/mactel-linux-users > > From here I don't experience any anomaly's accept when configured wrong. (on a clean system(debootstrap)); I usually just apt-get powersaved then adduser "yourename" powerdev; then all is good. (issues: when running anything with flashplayer do a powersave -e Powersave; powersave -l); then you should have the fans under-control. regards; Justin P. Mattock |
From: Justin P. M. <jus...@gm...> - 2009-01-21 09:29:03
|
Jonas Petersson wrote: > Hi all, > > On Wed, Jan 21, 2009 at 00:16, k11stan <k1...@gm...> wrote: > >> hey every1 - occasionally if I do not boot to osx for an extended period of >> time (roughly a month), the fan behavior (and likely also throttling) seems >> to get very weird ( the fans randomly rev up , the temps seem to spike , and >> the whole setup gets really unstable) .. after i boot to osx the problem >> seems to go away .. is there any way to preserve the nvram values that osx >> likes to set ? I'm using 2.6.25 with mactel patches and version 8.51.3 of >> fglrx .. I throttle the card using radeontool ...thanx for any suggestions >> > > I guess this is no real help, but for reference I can say that the > fans on my ATI-based MBP (2,2) were also acting up now and then. I > didn't boot into OSX a lot, so where that is related I have no idea. > However, I can say that my current MacBookPro4,1 (nVidia based) is a > LOT more sane when it comes to fans - and even when they spin up (due > to occasional high load) they don't make nearly as much noise. On the > downside, I never had any "applesmc: wait status failed" messages on > the old system, whereas it seems that polling this info (fans, temps > etc) fail - and then retry successfully - quite often now. I don't > know whether this is actual HW differences or if the driver just works > differently in newer kernels. BTW: I have not booted OSX on this box > for at least 3 months. > > Hope this was of some use for someone / Jonas > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by: > SourcForge Community > SourceForge wants to tell your story. > http://p.sf.net/sfu/sf-spreadtheword > _______________________________________________ > Mactel-linux-users mailing list > Mac...@li... > https://lists.sourceforge.net/lists/listinfo/mactel-linux-users > > that is(sorry accidentally got drunk(presidential inauguration); having acpi; acpid installed first.(before powersaved, or you'll run into issues); regards; Justin P. Mattock |
From: Jonas P. <cat...@gm...> - 2009-01-22 06:40:04
|
[Added list back to cc] On Thu, Jan 22, 2009 at 02:46, k11stan <k1...@gm...> wrote: > I think i figured out what is up - i turned on live previews and mipmapping > for the task switcher in compiz which would cause odd behaviour in the ati > driver .. as soon as i killed it the temps went down (which is weird since > the nvidia macbook never had that issue) .. compiz/ati driver bug it seems > .. thanx for your input so far Yes, it sounds familiar that I had to turn of a lot of the frills to make fans sane. > ..btw - did anyone get nix going reliably > on those new fangled macbookpros yet ? (the ones with the dual video cards) I have not personally had the opportunity to try it, but I recall that one of the issues was that you needed some fiddling due to the fact that the (single) button is more or less UNDER the trackpad. I believe the stuff you need is here: http://bitmath.org/code/ Best / Jonas > On Wed, Jan 21, 2009 at 10:06 AM, Jonas Petersson <jon...@xm...> > wrote: >> k11stan wrote: >>> ow i also have an nvidia macbook at home (3,1 i believe) and all works >>> fine with a patched kernel and nvidia drivers (i actually get more battery >>> time in linux than in osx ).. >> >> That's a nice one - had it before (Pro), but when it got stolen I had no >> choice but to upgrade. The 4,1 hirez screen is really nice and the >> multitouch tablet is "interesting" - was iffy to start with. >> >>> the ati macbook is a work laptop - i think 99%of the problem stems from >>> the crap drivers ati puts out ... >> >> 101% agreed. I noticed that ATI often hung after about a weeks worth of >> suspending to ram. I traced it to actually not being a suspend issue at all: >> Just switching to/from text console would always hang solid on the 16th (or >> was it 17?) round - even freshly booted. No workaround I know of. Was very >> happy to get rid of that one. If you suspect ATI - try that simple >> experiment. |