Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.os.linux.hardware > #743 > unrolled thread
| Started by | Richard Kimber <richardkimber@btinternet.com> |
|---|---|
| First post | 2011-10-31 08:54 -0500 |
| Last post | 2011-11-01 23:39 +0000 |
| Articles | 8 — 4 participants |
Back to article view | Back to comp.os.linux.hardware
cpu temperature Richard Kimber <richardkimber@btinternet.com> - 2011-10-31 08:54 -0500
Re: cpu temperature J G Miller <miller@yoyo.ORG> - 2011-10-31 14:10 +0000
Re: cpu temperature Bob Tennent <BobT@cs.queensu.ca> - 2011-10-31 14:46 +0000
Re: cpu temperature Clemens Ladisch <clemens@ladisch.de> - 2011-11-01 20:30 +0100
Re: cpu temperature Richard Kimber <richardkimber@btinternet.com> - 2011-11-01 17:34 -0500
Re: cpu temperature Clemens Ladisch <clemens@ladisch.de> - 2011-11-02 19:21 +0100
Re: cpu temperature Richard Kimber <richardkimber@btinternet.com> - 2011-11-04 09:46 -0500
Re: cpu temperature J G Miller <miller@yoyo.ORG> - 2011-11-01 23:39 +0000
| From | Richard Kimber <richardkimber@btinternet.com> |
|---|---|
| Date | 2011-10-31 08:54 -0500 |
| Subject | cpu temperature |
| Message-ID | <gZqdnXEyyOIsNTPTnZ2dnUVZ8midnZ2d@giganews.com> |
Since upgrading to Ubuntu 11.10 with kernel 3.0.0-12 I find that my cpu runs 5-10 degrees hotter with that than with 2.6.38-10 which I have moved back to. Is this a known 'feature' of the new kernel?
[toc] | [next] | [standalone]
| From | J G Miller <miller@yoyo.ORG> |
|---|---|
| Date | 2011-10-31 14:10 +0000 |
| Message-ID | <j8ma8d$7aa$1@dont-email.me> |
| In reply to | #743 |
On Monday, October 31st, 2011 at 08:54:57h -0500, Richard Kimber wrote: > Since upgrading to Ubuntu 11.10 with kernel 3.0.0-12 I find that my cpu > runs 5-10 degrees hotter with that than with 2.6.38-10 which I have > moved back to. > > Is this a known 'feature' of the new kernel? A web search would have revealed to you that others have also observed this phenomenon. <http://www.linuxquestions.ORG/questions/linux-kernel-70/linux-kernel-3-0-battery-life-power-consumption-896637/> The power regression occurred in one of the 2.6.38 versions, presumably after the version you were previously using. <http://www.phoronix.COM/scan.php?page=news_item&px=OTM3NQ> Why the kernel development team have not fixed the issue by now, is anybody's guess, unless it is fixed in 3.1.
[toc] | [prev] | [next] | [standalone]
| From | Bob Tennent <BobT@cs.queensu.ca> |
|---|---|
| Date | 2011-10-31 14:46 +0000 |
| Message-ID | <slrnjatd71.u6g.BobT@linus.cs.queensu.ca> |
| In reply to | #744 |
On Mon, 31 Oct 2011 14:10:22 +0000 (UTC), J G Miller wrote: > Why the kernel development team have not fixed the issue > by now, is anybody's guess, unless it is fixed in 3.1. It's possible that the "regression" might have *fixed* a bug, and that a side effect of this fix is increased power demand.
[toc] | [prev] | [next] | [standalone]
| From | Clemens Ladisch <clemens@ladisch.de> |
|---|---|
| Date | 2011-11-01 20:30 +0100 |
| Message-ID | <9havngF9hhU1@mid.individual.net> |
| In reply to | #744 |
J G Miller wrote: > On Monday, October 31st, 2011 at 08:54:57h -0500, Richard Kimber wrote: >> Since upgrading to Ubuntu 11.10 with kernel 3.0.0-12 I find that my cpu >> runs 5-10 degrees hotter with that than with 2.6.38-10 which I have >> moved back to. > > Why the kernel development team have not fixed the issue > by now, is anybody's guess, unless it is fixed in 3.1. That power management feature locks up the machine on some hardware configurations which cannot be detected automatically. Anyway, that would only affect PCIe, not the CPU. Richard, if your temperature is measured with the coretemp driver (any recent Intel CPU), it's possible that you see the effect of some bugfix that corrected the temperature measurement offset (i.e., the actual temperature did not change). Regards, Clemens
[toc] | [prev] | [next] | [standalone]
| From | Richard Kimber <richardkimber@btinternet.com> |
|---|---|
| Date | 2011-11-01 17:34 -0500 |
| Message-ID | <69qdnbbhSddB7i3TnZ2dnUVZ8nWdnZ2d@giganews.com> |
| In reply to | #750 |
On Tue, 01 Nov 2011 20:30:52 +0100, Clemens Ladisch wrote: > Richard, if your temperature is measured with the coretemp driver (any > recent Intel CPU), it's possible that you see the effect of some bugfix > that corrected the temperature measurement offset (i.e., the actual > temperature did not change). Actually, I'm using k10temp and it87 with AMD Phenom 2x4 3.0 - Richard.
[toc] | [prev] | [next] | [standalone]
| From | Clemens Ladisch <clemens@ladisch.de> |
|---|---|
| Date | 2011-11-02 19:21 +0100 |
| Message-ID | <9hdg14Ftj5U1@mid.individual.net> |
| In reply to | #752 |
Richard Kimber wrote: > I'm using k10temp and it87 with AMD Phenom 2x4 3.0 There have been no significant changes in k10temp. Did the values of _both_ drivers change by a similar amount? If yes, then your CPU indeed runs hotter. Regards, Clemens
[toc] | [prev] | [next] | [standalone]
| From | Richard Kimber <richardkimber@btinternet.com> |
|---|---|
| Date | 2011-11-04 09:46 -0500 |
| Message-ID | <v5WdnVD_sId_Zy7TnZ2dnUVZ8hudnZ2d@giganews.com> |
| In reply to | #756 |
On Wed, 02 Nov 2011 19:21:20 +0100, Clemens Ladisch wrote: > Did the values of _both_ drivers change by a similar amount? If yes, > then your CPU indeed runs hotter. Yes. - Richard.
[toc] | [prev] | [next] | [standalone]
| From | J G Miller <miller@yoyo.ORG> |
|---|---|
| Date | 2011-11-01 23:39 +0000 |
| Message-ID | <j8pvvt$gj$1@dont-email.me> |
| In reply to | #750 |
On Tuesday, November 1st, 2011 at 20:30:52h +0100, Clemens Ladisch wrote: > it's possible that you see the effect of some bugfix that corrected the > temperature measurement offset (i.e., the actual temperature did not change). Which could be a change to the sensors package (possible just an entry in the sensors3.conf file) rather than the actual kernel its-self.
[toc] | [prev] | [standalone]
Back to top | Article view | comp.os.linux.hardware
csiph-web