Your message dated Tue, 04 Oct 2011 12:01:08 +0200 with message-id <[email protected]> and subject line Re: nvidia-settings: Overclocking Auto Detect never available has caused the Debian Bug report #447517, regarding nvidia-settings: Overclocking Auto Detect never available to be marked as done.
This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact [email protected] immediately.) -- 447517: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=447517 Debian Bug Tracking System Contact [email protected] with problems
--- Begin Message ---Package: nvidia-settings Version: 1.0+20070502-1 Severity: normal After some major system "upgrades" (switching RAM sticks, motherboard, PSU, graphic cards and xservers) I found that nvidia-settings no longer could auto-detect the optimal overclocking frequencies. The "Auto Detect" button would always be greyed out even though I had enabled "Coolbits" in Xorg.conf I downgraded nvidia-settings to the version in testing (1.0+20060516-3) and suddenly "Auto Detect" would be available. After some further digging I think I found the bug; The problem is that according NVCtrl.h (from nvidia-settings/testing) NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION is a write-only attribute so NvCtrlGetAttribute(..,NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION,..) will always fail. There should be a way to discover if the current card support optimal frequency detection, but trying to read a write-only attribute isn't the way. :) Removing the attempted read of NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION enables the "Auto Detect" button again. I haven't checked if this is fixed in the upstream version. I'll attach quick-and-dirty patch later - the friendly but rusty hacker Mikael Åkersund Quotes from relevent files below File: nvidia-settings-1.0+20070502/src/gtk+-2.x/ctkclocks.c Line 320: if ( overclocking_enabled ) { ret = NvCtrlGetAttribute(handle, NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION, &value); if ( ret == NvCtrlSuccess ) { # <---- Will always fail ret = NvCtrlGetAttribute(handle, NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE, &value); if ( ret != NvCtrlSuccess ) return NULL; probing_optimal = (value == NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE_BUSY); auto_detection_available = TRUE; } } Here's a snip from /usr/include/NVCtrl/NVCtrl.h [Line 1244]: /* * * NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION - set to _START to * * initiate testing for the optimal 3D clock frequencies. Once * * found, the optimal clock frequencies will be returned by the * * NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS attribute asynchronously * * (using an X event, see XNVCtrlSelectNotify). * * * * To cancel an ongoing test for the optimal clocks, set the * * NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION attribute to _CANCEL * * * * Note: unless NV_CTRL_GPU_OVERCLOCKING_STATE is set to _MANUAL, the * * optimal clock detection process is unavailable. * */ #define NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION 95 /* -W- */ # <--- No reading! #define NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_START 0 #define NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_CANCEL 1 -------------------------------------------- -- System Information: Debian Release: lenny/sid APT prefers testing APT policy: (990, 'testing'), (140, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.22-4-d3-ul8-1 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages nvidia-settings depends on: ii libatk1.0-0 1.20.0-1 The ATK accessibility toolkit ii libc6 2.6.1-1 GNU C Library: Shared libraries ii libcairo2 1.4.10-1+b2 The Cairo 2D vector graphics libra ii libfontconfig1 2.4.2-1.2 generic font configuration library ii libglib2.0-0 2.14.1-5 The GLib library of C routines ii libgtk2.0-0 2.10.13-1 The GTK+ graphical user interface ii libpango1.0-0 1.18.2-1 Layout and rendering of internatio ii libx11-6 2:1.0.3-7 X11 client-side library ii libxcursor1 1:1.1.9-1 X cursor management library ii libxext6 1:1.0.3-2 X11 miscellaneous extension librar ii libxfixes3 1:4.0.3-2 X11 miscellaneous 'fixes' extensio ii libxi6 2:1.1.3-1 X11 Input extension library ii libxinerama1 1:1.0.2-1 X11 Xinerama extension library ii libxrandr2 2:1.2.2-1 X11 RandR extension library ii libxrender1 1:0.9.4-1 X Rendering Extension client libra Versions of packages nvidia-settings recommends: ii nvidia-glx 100.14.19-1 NVIDIA binary Xorg driver -- no debconf information _________________________________________________________________ Explore the seven wonders of the world http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE--- ctkclocks.c.orig 2007-10-21 21:29:28.000000000 +0200 +++ ctkclocks.c 2007-10-21 21:38:34.000000000 +0200 @@ -319,18 +319,13 @@ if ( overclocking_enabled ) { ret = NvCtrlGetAttribute(handle, - NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION, - &value); - if ( ret == NvCtrlSuccess ) { - ret = NvCtrlGetAttribute(handle, - NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE, - &value); - if ( ret != NvCtrlSuccess ) - return NULL; - probing_optimal = - (value == NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE_BUSY); - auto_detection_available = TRUE; - } + NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE, + &value); + if ( ret != NvCtrlSuccess ) + return NULL; + probing_optimal = + (value == NV_CTRL_GPU_OPTIMAL_CLOCK_FREQS_DETECTION_STATE_BUSY); + auto_detection_available = TRUE; } /* Can we access the 2D clocks? */
--- End Message ---
--- Begin Message ---On 2011-05-17 20:36, Mikael Åkersund wrote: > Just week ago I updated to a new graphics card (Gtx 560Ti) which seems to > have overclocking disabled all-together in Nvidia's drivers (270.41.06-1) > so I can't verify 100% that the bug is gone right now. > IIRC the bug was fixed after a couple of months as I can clearly remember > trying out how high my previous graphics cards would overclock. In that case I'll close this report now as unreproducible. Andreas
--- End Message ---
