Menu
▾
▴
dynamicpower-devel — Dynamic Power Management development
You can subscribe to this list here.
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(1) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 |
Jan
|
Feb
|
Mar
(2) |
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
(1) |
Aug
(1) |
Sep
|
Oct
(1) |
Nov
|
Dec
(4) |
| 2005 |
Jan
(10) |
Feb
|
Mar
(14) |
Apr
(8) |
May
(2) |
Jun
|
Jul
(1) |
Aug
(7) |
Sep
|
Oct
(1) |
Nov
(1) |
Dec
|
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
(3) |
Dec
|
| 2007 |
Jan
(2) |
Feb
(2) |
Mar
(5) |
Apr
(12) |
May
(2) |
Jun
|
Jul
(10) |
Aug
(2) |
Sep
(2) |
Oct
|
Nov
(2) |
Dec
(4) |
| 2008 |
Jan
(2) |
Feb
(1) |
Mar
(1) |
Apr
|
May
|
Jun
(1) |
Jul
(3) |
Aug
|
Sep
(13) |
Oct
(6) |
Nov
(27) |
Dec
(37) |
| 2009 |
Jan
(13) |
Feb
(14) |
Mar
(11) |
Apr
(19) |
May
(41) |
Jun
(21) |
Jul
(29) |
Aug
(8) |
Sep
(2) |
Oct
|
Nov
|
Dec
(4) |
| 2010 |
Jan
(6) |
Feb
(5) |
Mar
(5) |
Apr
|
May
(2) |
Jun
(1) |
Jul
|
Aug
(2) |
Sep
(1) |
Oct
(2) |
Nov
|
Dec
|
|
From: Todd P. <tp...@mv...> - 2005-01-13 03:44:34
|
Amit Choudhary wrote: > Hi Todd, > > Please find the attached patch for checking if the max value of the > constraint is less than min. The new constraints are not stored and an error > message is printed. Hello Amit -- Thanks, I've got it in the queue for next time. -- Todd |
|
From: Amit C. <Ami...@mi...> - 2005-01-12 08:49:59
|
Hi Todd, Please find the attached patch for checking if the max value of the constraint is less than min. The new constraints are not stored and an = error message is printed. Regards, Amit -----Original Message----- From: Todd Poynor [mailto:tp...@mv...] Sent: Wednesday, January 12, 2005 1:48 AM To: Amit Choudhary Cc: dyn...@li... Subject: Re: [Dynamicpower-devel] DPM update for 2.6.10 Amit Choudhary wrote: > Hi Todd, >=20 > In function constraints_store(), shall we check if max is really >=3D = min. Hi Amit -- Sounds good, patch welcome or I'll put it on the list for=20 next time, thanks, --=20 Todd |
|
From: Todd P. <tp...@mv...> - 2005-01-11 20:18:15
|
Amit Choudhary wrote: > Hi Todd, > > In function constraints_store(), shall we check if max is really >= min. Hi Amit -- Sounds good, patch welcome or I'll put it on the list for next time, thanks, -- Todd |
|
From: Amit C. <Ami...@mi...> - 2005-01-11 14:43:42
|
Hi All, I was maintaining a small document about the DPM implementation. I am attaching that text document with this mail. It may help someone who is = new to the dpm patch. Because of new versions getting released, few things = may be a bit obsolete. It was started when I applied the 2.6.9 patch. But most = of the things are valid for 2.6.10 as well. If there are any errors, please = let me know. Regards, Amit |
|
From: Amit C. <Ami...@mi...> - 2005-01-11 13:27:47
|
Hi Todd, In function constraints_store(), shall we check if max is really >=3D = min. Regards, Amit -----Original Message----- From: Todd Poynor [mailto:tp...@mv...] Sent: Tuesday, January 11, 2005 8:24 AM To: dyn...@li... Subject: [Dynamicpower-devel] DPM update for 2.6.10 Dynamic Power Management dynamicpower.sf.net has been updated for = 2.6.10: Core: Remove IRQ disable/enable around generic operating point setting code, = since setting some operating points may require sleeping/scheduling, = especially in driver suspend/resume callbacks. NOTE: If interrupts must be disabled for a code sequence used to set an operating point in your machine-dependent code, the IRQ save/restore logic must be moved there. Fix oops if bad op/class name written to policy state mapping attribute, e.g.: # echo stewpid > /sys/dpm/policy/mypolicy/idle Modify /sys/dpm/state/<name> attributes to be kobjects. Add /sys/dpm/state/<name>/constraints for reading devices constraining a state from userspace. Restore force attributes for operating points, although forcing is not implemented in the kernel. Port stats interfaces to sysfs: /sys/dpm/state/stats, /sys/dpm/op/stats, /sys/dpm/policy/stats. All stats configured with CONFIG_DPM_STATS, no separate symbol for CONFIG_DPM_OPT_STATS, idle stats and MD (machine-dependent) stats = dropped. Check for and reject double disable/enable, which could cause lockups (due to pm_idle save/restore logic). Call machine-dependent cleanup routine on disable to be symmetric with enable processing, and avoid oopses due to improperly handled pm_idle = loop save/restore logic therein. Mark procfs interface deprecated, must be explicitly enabled. Restore ability to create classes via /proc. Fix compile if CONFIG_DPM=3Dn OMAP: OMAP1710 H3 support (kernel DVS from Vladimir Barinov and setup util from Amit Choudhary). Proper setup of tps65013 voltage scaling under discussion in community. Changes for mods in upstream timers code. Temporarily make private copy of MPU timer 0 code until public interfaces are straightened out. (Includes fixes from Vladimir Barinov). Fixup pm_idle loop save and restore logic. Add IRQ protection around frequency/voltage scaling code in operating = point set routine (since upper layer no longer does this). OMAP1610 H2 ULPD register access changes to match upstream. --=20 Todd ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ Dynamicpower-devel mailing list Dyn...@li... https://lists.sourceforge.net/lists/listinfo/dynamicpower-devel |
|
From: Todd P. <tp...@mv...> - 2005-01-11 02:54:11
|
Dynamic Power Management dynamicpower.sf.net has been updated for 2.6.10: Core: Remove IRQ disable/enable around generic operating point setting code, since setting some operating points may require sleeping/scheduling, especially in driver suspend/resume callbacks. NOTE: If interrupts must be disabled for a code sequence used to set an operating point in your machine-dependent code, the IRQ save/restore logic must be moved there. Fix oops if bad op/class name written to policy state mapping attribute, e.g.: # echo stewpid > /sys/dpm/policy/mypolicy/idle Modify /sys/dpm/state/<name> attributes to be kobjects. Add /sys/dpm/state/<name>/constraints for reading devices constraining a state from userspace. Restore force attributes for operating points, although forcing is not implemented in the kernel. Port stats interfaces to sysfs: /sys/dpm/state/stats, /sys/dpm/op/stats, /sys/dpm/policy/stats. All stats configured with CONFIG_DPM_STATS, no separate symbol for CONFIG_DPM_OPT_STATS, idle stats and MD (machine-dependent) stats dropped. Check for and reject double disable/enable, which could cause lockups (due to pm_idle save/restore logic). Call machine-dependent cleanup routine on disable to be symmetric with enable processing, and avoid oopses due to improperly handled pm_idle loop save/restore logic therein. Mark procfs interface deprecated, must be explicitly enabled. Restore ability to create classes via /proc. Fix compile if CONFIG_DPM=n OMAP: OMAP1710 H3 support (kernel DVS from Vladimir Barinov and setup util from Amit Choudhary). Proper setup of tps65013 voltage scaling under discussion in community. Changes for mods in upstream timers code. Temporarily make private copy of MPU timer 0 code until public interfaces are straightened out. (Includes fixes from Vladimir Barinov). Fixup pm_idle loop save and restore logic. Add IRQ protection around frequency/voltage scaling code in operating point set routine (since upper layer no longer does this). OMAP1610 H2 ULPD register access changes to match upstream. -- Todd |
|
From: Todd P. <tp...@mv...> - 2005-01-06 00:55:41
|
Amit Choudhary wrote: > Hi Todd, > > After applying the patch, I compile the kernel without enabling the DPM > option. I get the following error. > > CC drivers/base/power/power-dpm.o > drivers/base/power/power-dpm.c: In function `constraints_show': > drivers/base/power/power-dpm.c:237: error: `dpm_param_names' undeclared > (first u > se in this function) > drivers/base/power/power-dpm.c:237: error: (Each undeclared identifier is > report > ed only once > drivers/base/power/power-dpm.c:237: error: for each function it appears in.) > drivers/base/power/power-dpm.c: In function `constraints_store': > drivers/base/power/power-dpm.c:296: error: `DPM_PP_NBR' undeclared (first use > in > this function) > drivers/base/power/power-dpm.c:297: error: `dpm_param_names' undeclared > (first u > se in this function) > make[3]: *** [drivers/base/power/power-dpm.o] Error 1 > make[2]: *** [drivers/base/power] Error 2 > make[1]: *** [drivers/base] Error 2 > make: *** [drivers] Error 2 > > > There are some other places where code related to DPM is not within > #ifdef...#endif. Hello Amit -- I think that bug is fixed in the post-2.6.9 patch I sent, please let me know if there's any other places that need to be fixed up. Some of the stuff, such as power event notifiers, are intended to work independently of DPM (yes, I should probably split things up into separate patches, ugh). Thanks, -- Todd |
|
From: Amit C. <Ami...@mi...> - 2005-01-05 10:57:05
|
Hi Todd, After applying the patch, I compile the kernel without enabling the DPM option. I get the following error. CC drivers/base/power/power-dpm.o drivers/base/power/power-dpm.c: In function `constraints_show': drivers/base/power/power-dpm.c:237: error: `dpm_param_names' undeclared (first u se in this function) drivers/base/power/power-dpm.c:237: error: (Each undeclared identifier = is report ed only once drivers/base/power/power-dpm.c:237: error: for each function it appears = in.) drivers/base/power/power-dpm.c: In function `constraints_store': drivers/base/power/power-dpm.c:296: error: `DPM_PP_NBR' undeclared = (first use in this function) drivers/base/power/power-dpm.c:297: error: `dpm_param_names' undeclared (first u se in this function) make[3]: *** [drivers/base/power/power-dpm.o] Error 1 make[2]: *** [drivers/base/power] Error 2 make[1]: *** [drivers/base] Error 2 make: *** [drivers] Error 2 There are some other places where code related to DPM is not within #ifdef...#endif. Regards, Amit |
|
From: Todd P. <tp...@mv...> - 2004-12-20 21:52:30
|
Amit Choudhary wrote: > Hi, > > I have a suggestion for the DPM patch. All the changes should be covered > under #ifdef CONFIG_DPM ... #endif. So, that if I can use the same kernel > with and without DPM support. As of now, once I have included the DPM patch, > there is no way of compiling without the DPM support. It gives errors if I > try to compile after unchecking the DPM option. Hi, I've fixed one known problem with compilation when CONFIG_DPM=n using latest patches, have been waiting for 2.6.10 to release new patches. If there's a compilation problem with CONFIG_DPM=n then it's a bug, send the details and I'll make sure it's fixed. For better or worse, various power management related things for embedded systems are finding their way into the patches, including things not tied to the core DPM currently selected with CONFIG_DPM (such as kernel-to-userspace power event notifiers). Most of the non-DPM-specific stuff is selected with CONFIG_PM, which should be generally appropriate, although deviations from upstream PM code could be configured separately. The drivers/base/power/power-dpm.o file does contain things that are supposed to be present with use of the DPM core and is selected with CONFIG_PM, some of which is temporary while potentially community-acceptable designs are being formulated. For now I'm fixing the bugs as they arise. Thanks, -- Todd |
|
From: Todd P. <tp...@mv...> - 2004-12-20 21:07:38
|
Amit Choudhary wrote: > Hi, > > I have downloaded the 2.6.9 linux-omap dpm patch from sourceforge. I > downloaded the 2.6.9 linux-omap kernel from bkbits. There is no > include/linux/dpm.h file and so it gave compilation errors. Where can I find > these files and other related header files (if there are any). Hello Amit -- Been on vacation, you may have already found an answer in the meantime, but just in case: apply the generic DPM patch and then the OMAP-specific patch. A couple more tidbits at http://dynamicpower.sourceforge.net/dpm-omap.html may help as well. -- Todd |
|
From: Amit C. <Ami...@mi...> - 2004-12-15 07:45:01
|
Hi, I have a suggestion for the DPM patch. All the changes should be covered under #ifdef CONFIG_DPM ... #endif. So, that if I can use the same = kernel with and without DPM support. As of now, once I have included the DPM = patch, there is no way of compiling without the DPM support. It gives errors if = I try to compile after unchecking the DPM option. Even the makefiles should be like: obj-$(CONFIG_DPM) +=3D dpm.o "drivers/base/power/Makefile" doesn't have it. I will help if you = require any. Regards, Amit |
|
From: Amit C. <Ami...@mi...> - 2004-12-14 07:31:35
|
Hi, I have downloaded the 2.6.9 linux-omap dpm patch from sourceforge. I downloaded the 2.6.9 linux-omap kernel from bkbits. There is no include/linux/dpm.h file and so it gave compilation errors. Where can I = find these files and other related header files (if there are any). Regards, Amit |
|
From: Todd P. <tp...@mv...> - 2004-10-20 01:58:17
|
The Dynamic Power Management Project has released new versions of the Linux DPM kernel patches and utilities. See http://sourceforge.net/projects/dynamicpower/ Core: * Add x86 Centrino 1.7G model setup script (from Armin Kuster). * Power parameters are now named, names provided by the board-specific code. SysFS files and messages now use these names. * Add operating point sysfs attributes for power parameter values. Operating points may now be created without the (board-specific, defined by kernel headers) ordered list of power parameter values. The parameters may now be filled in individually by writing to the associated attributes. This also allows parameters for an existing operating point to be modified at runtime. * Add sysfs attributes for operating states (not yet used, for future stats). * Move read/write active operating state interfaces to new /sys/dpm/state/active attribute. * Add policy sysfs attributes for mapping operating states to operating points. Policies may now be created without the (board-specific, defined by kernel headers) ordered list of mappings (in operating state index oreder). Now the mappings may be filled in individually by writing to the associated attributes. This also allows mappings for an existing policy to be modified at runtime. * Move interface for read/write active policy to new /sys/dpm/policy/active attribute. * Add /proc/pid/dpmstate for reading and setting task states (based on contribution from Armin Kuster). * Disable forced operating points, these need work (and are better handled in userspace). * Move sysfs and portions of /proc interface going forward to new dpm-ui.c * Remove obsolete interface support routines. * Resync with kernel.org for 2.6.9. OMAP: * Add omap730 setup script (from Armin Kuster). * Changes due to new OMAP clocking code. * Allow -1 "no change" value for all power parameters. * Resync with linux-omap.bkbits.net for 2.6.9-final. -- Todd |
|
From: Todd P. <tp...@mv...> - 2004-08-26 01:52:54
|
DPM has been revved for 2.6.8.1. http://sourceforge.net/projects/dynamicpower/ * Fix compile error if CONFIG_HOTPLUG=n * Resync with kernel.org 2.6.8.1. * Resync with linux-omap-open-source 2.6.8.1 and yet more OMAP register accessor changes. -- Todd Poynor MontaVista Software |
|
From: Todd P. <tp...@mv...> - 2004-07-08 01:50:11
|
Version 2.6.7 07/07/2004
* Added "classes" of operating points: DPM selects an operating point
from the class according to device constraints. See new section
Operating Point Classes in the DPM Developer's Guide.
* Added general power event notifiers and change the existing
device-power notifiers:
* device_power_notify() -> device_power_event()
* use power.agent
* event device-event
* DEVPATH=sysfs-path replaced with DEVICE=bus-id
See new section Power Event Notifiers in the DPM Developer's Guide.
* Added OMAP1610 H2 Dynamic Voltage Scaling (experimental).
dpm-setup-omap1610 is updated to include two 1.1V operating points.
* Fixes for CONFIG_DPM/CONFIG_PM=n and ARM cleanups (thanks George
Davis) and CONFIG_HOTPLUG=n.
* Removed dpm_md_init() (just user __initcall() for machine-dependent
init, no ordering needed between generic and machine-dependent; the
ordered part is now handled via dpm_md_startup()).
* Sync with linux-2.6.7 and linux-omap tree as of 07/07/2004.
--
Todd Poynor
MontaVista Software
|
|
From: Todd P. <tp...@mv...> - 2004-06-11 02:04:02
|
Core DPM has been updated recently to match linux-2.6.7-rc2, to add device power state event userspace notifiers, and to start adding class interfaces back in (turns out they can be useful). OMAP platform support has been rev'ed 3 times recently to match recently-in-flux register definitions and accessors (and we're probably still relying on a few that will soon be changed again). -- Todd Poynor MontaVista Software |
|
From: <ben...@id...> - 2004-05-25 09:10:13
|
Dear Open Source developer I am doing a research project on "Fun and Software Development" in which I kindly invite you to participate. You will find the online survey under http://fasd.ethz.ch/qsf/. The questionnaire consists of 53 questions and you will need about 15 minutes to complete it. With the FASD project (Fun and Software Development) we want to define the motivational significance of fun when software developers decide to engage in Open Source projects. What is special about our research project is that a similar survey is planned with software developers in commercial firms. This procedure allows the immediate comparison between the involved individuals and the conditions of production of these two development models. Thus we hope to obtain substantial new insights to the phenomenon of Open Source Development. With many thanks for your participation, Benno Luthiger PS: The results of the survey will be published under http://www.isu.unizh.ch/fuehrung/blprojects/FASD/. We have set up the mailing list fa...@we... for this study. Please see http://fasd.ethz.ch/qsf/mailinglist_en.html for registration to this mailing list. _______________________________________________________________________ Benno Luthiger Swiss Federal Institute of Technology Zurich 8092 Zurich Mail: benno.luthiger(at)id.ethz.ch _______________________________________________________________________ |
|
From: Todd P. <tp...@mv...> - 2004-04-20 23:44:03
|
The SourceForge DPM project is updated to add device constraints on allowed operating points and support for TI OMAP1610 (voltage scaling is still TODO). Thanks to Armin Kuster for the DPM setup script entries for 192MHz speed on 1610. Release notes at http://dynamicpower.sourceforge.net/README Information on constraints at http://dynamicpower.sourceforge.net/dpm-constraints.html Info on OMAP support at http://dynamicpower.sourceforge.net/dpm-omap.html -- Todd Poynor MontaVista Software |
|
From: Todd P. <tp...@mv...> - 2004-03-25 03:42:59
|
The Dynamic Power Management project is pleased to announce availability of patches and utilities to enable the DPM frequency scaling core for the TI OMAP1510 in Linux 2.6, courtesy of MontaVista Software. DPM explores technologies to improve power management, with a focus on techniques applicable to running systems, adjusting power parameters on-the-fly in order to save power during brief idle periods or execution of tasks with lower power demands. See http://sourceforge.net/projects/dynamicpower Comments and development participation are welcomed. -- Todd Poynor MontaVista Software |
|
From: Todd P. <tod...@ya...> - 2004-03-23 21:12:35
|
The community DPM 2.6 patch is updated to match Linux 2.6.4, with many cleanups and a bug fix or two. Grab new centrino utilities dpm-utils-centrino-2.6.4 from the new dpm-utils-centrino package. Support for additional platforms will be added in the near future. ===== Todd __________________________________ Do you Yahoo!? Yahoo! Finance Tax Center - File online. File on time. http://taxes.yahoo.com/filing.html |
|
From: Todd P. <tod...@ya...> - 2003-12-12 23:34:34
|
The DPM for Linux 2.6/Centrino patch has been brought up to 2.6.0-test11. The primary interface is switching from /proc to sysfs. Grab new utilities from dpm-utils to use sysfs to set up DPM. If you've tried out DPM please do let us know if you have any comments or anything else to contribute, thanks, ===== Todd __________________________________ Do you Yahoo!? New Yahoo! Photos - easier uploading and sharing. http://photos.yahoo.com/ |
|
From: Napoo <na...@ds...> - 2003-11-08 03:55:10
|
When do you plan on updating the dpm patch to a newer test version? Napo'o |
236 messages has been excluded from this view by a project administrator.
