January 2008 - Darwin-dev - lists.apple.com

Re: Cancel Sleep
by Michael Smith 02 Jan '08

02 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jan 1, 2008, at 4:28 PM, Andrew James wrote: = Mike _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am surprised that now you know the mission of the project that already exists and is used by thousands of users, you will not assist in finding a notification call that could be useful in other projects besides the Insomnia project as well. I am not terribly surprised that you haven't actually thought through the implications of what you're asking for, even after I tried to generalise them in order to illustrate why the mode of operation you want is not a supported thing, and not a good idea in general. However, you need to do this. Yes, I understand full well what it is that you are asking for, and why in the sort of applications that you personally have in mind, you think it would be a good idea. The problem is, this is not "a sleep hack for InsomniaX" that you are describing; this is "a fully supported mode of operation". I've tried to make it clear why this isn't likely to be a supported mode of operation - damaging the system, shutting down unexpectedly - these are class-action lawsuit material and generally not a good idea. So seriously maybe i should start a new topic for this problem since everyone has a mixed view on this, and no one will answer the current question that in reality has no effect on the sleep behavior of the system. so lets repeat the question just so everyone is clear..... What makes you think this would help? A new Subject: tag isn't going to change the situation; you need to stop repeating your question and start listening to the answer. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by Andrew James 02 Jan '08

02 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:cc:message-id:from:to:in-reply-to:content-type:content-transfer-encoding:mime-version:subject:date:references:x-mailer; bh=HxRdmhfyG4I5jkHiHD7KwYincmA9RWgYyDlx++V+IEM=; b=E9np6WH03pnDn72/PTAR/QWxtQZ9RQ2XFgrPAGy2upFDHNaA15GawQXYnqkQfYU9eR4tQ3AYozzdJ02T3bh41FFZKcv222MyKeuKjoPjSbKood2+MhsXtU2irlreO41snHfI6DIa8BIWi2GweUZPQmWtKZA+jwGF7bFUBJIkHz8= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=cc:message-id:from:to:in-reply-to:content-type:content-transfer-encoding:mime-version:subject:date:references:x-mailer; b=FAw07scFjkDlGO92kyzmwW4vxuiC7RtrOo7CV9H14GGc8HaUW5MXBI62eYOP/0CH1sqZ/6ogvZOi+rv/SN+GboVsTdlf+Xri8az+o301NhnBYW9ui1ke+soeUegvXBuH2uSkAnVjvxcmJB+rL8i+hiH4DJsk579U1+x2zxpvnws= On 02/01/2008, at 7:00 AM, Michael Smith wrote: On Jan 1, 2008, at 12:03 PM, Andrew James <semaja2(a)gmail.com> wrote: So if everybody can forget just for a second of the main goal, the mission at the moment is to find a way to receive the notification of the lid state changing that does not mix up with the power state changes. No, this is not what you are trying to do at all. Are you terribly surprised that nobody wants to help you? = Mike _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am surprised that now you know the mission of the project that already exists and is used by thousands of users, you will not assist in finding a notification call that could be useful in other projects besides the Insomnia project as well. You should also be aware that there is so many warnings strapped to the programs it is even more scary then trying to over clock a machine, and in the whole life of the project there has never been a reported case of someone causing damage because everyone has seen the warnings and understand the facts. As well as i am aware of the hotter it gets the smaller the life span is, but if you remove the point of people putting it into a bag the machine will be running in the same way Apple intended except for the need of extra hardware. So seriously maybe i should start a new topic for this problem since everyone has a mixed view on this, and no one will answer the current question that in reality has no effect on the sleep behavior of the system. so lets repeat the question just so everyone is clear..... the mission at the moment is to find a way to receive the notification of the lid state changing that does not mix up with the power state changes. Ill try those other mailing lists, and id like to remind everybody the solution exists to a extent as i am a developer of Insomnia and InsomniaX. However i would just like to take this back to basics and see if anyone can cover the biggest problem. What you are trying to do is establish a new mode of operation for the system; one that is currently explicitly precluded. Current Apple portables are *not* designed to run without cooling; the Intel chipset generates too much heat even in its lowest power mode for the system to survive operation without airflow. As a consequence, what you are trying to do is to define a mode of operation that is simply not safe; once the clamshell is closed, users will assume the system can be put into a bag or under a pile of stuff with no ventilation, and it will overheat. Certain Apple portables do support running with the clamshell closed, but with other constraints applied that make it clear that the user intends for the system to be awake and that more or less require that it have adequate access to cooling (keyboard attached, external monitor attached, explicit wakeup after the clamshell is closed). The consequences of overheating include emergency system shutdown (the most common and intended response to thermal distress), but can also include accelerated aging of components (backlight, fixed and optical disks, display, cosmetic surfaces) and more subtle effects. The system is designed to deal with being overheated occasionally and accidentally, but not as a routine matter of course, which is what your new mode of operation will facilitate. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by David Alger 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by David Elliott 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jan 1, 2008, at 3:30 PM, Michael Smith wrote: -Dave _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Certain Apple portables do support running with the clamshell closed, but with other constraints applied that make it clear that the user intends for the system to be awake and that more or less require that it have adequate access to cooling (keyboard attached, external monitor attached, explicit wakeup after the clamshell is closed). And that last one is the thing that draws the ire of many a PowerBook and MacBook Pro user. If I've already got AC power, and an external display, and an external keyboard then I ought to be able to open and close the lid at will and it should only cause the internal display to be made (un)available. As it stands now I might be sitting on my couch working and start something like a large download. I then decide to move to the office where I attach my keyboard and cinema display. Once that change settles I ought to be able to close the lid and have the internal display disabled but leave the system running. The system should not be forced in to sleep mode for me to then hit a key on the keyboard and immediately wake it up again. If you're simply trying to prevent people from stuffing the notebook into a bag with it still running, you ought to note the fact that it's pretty well impossible to do this while the power and the display and the keyboard are all connected. What exactly is the reason for requiring the system to go to sleep and then be woken up again? This email sent to site_archiver(a)lists.apple.com

1 0

Re: darwinbuild for gcc_os using 9A581.plist
by Michael Franz 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:references; bh=sv6a7IZiv9zYe+iYkO+OIf1iCv6BOyV/8pg0UTNmnJU=; b=nUk+2z5rWP0kgr/K8txhWVGUG08sn+quthfgvkK+Fn/B5Z4z//iVAm3XT1krsj0sMavvffRH5aJrSEJVjwwex30wYM1ZSr66YGUnzyCI27mRvY3wEK9PYUy8sE322PNrddtEXJ2DM45OysvnOSLB9hzO7jX5piPo4weLaCBLZ+k= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:references; b=cGWSDXEHhhQbO1xj/TSF/OCRfwSS+azsq9hS3XxGB11og6prsRR1qPi9CrZIa8tilP1aXn6uYSKUQoVf06fyLS/RdsF5Ht/krP8hN/62/uLbqNKjmDAo33g5iBgf7efnDjydZ6ExnENe7AZSpytRtTwMCOwIcUO+8fIXaRvHluE= _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by Michael Smith 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jan 1, 2008, at 12:03 PM, Andrew James <semaja2(a)gmail.com> wrote: Ill try those other mailing lists, and id like to remind everybody the solution exists to a extent as i am a developer of Insomnia and InsomniaX. However i would just like to take this back to basics and see if anyone can cover the biggest problem. So if everybody can forget just for a second of the main goal, the mission at the moment is to find a way to receive the notification of the lid state changing that does not mix up with the power state changes. No, this is not what you are trying to do at all. Are you terribly surprised that nobody wants to help you? = Mike _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… What you are trying to do is establish a new mode of operation for the system; one that is currently explicitly precluded. Current Apple portables are *not* designed to run without cooling; the Intel chipset generates too much heat even in its lowest power mode for the system to survive operation without airflow. As a consequence, what you are trying to do is to define a mode of operation that is simply not safe; once the clamshell is closed, users will assume the system can be put into a bag or under a pile of stuff with no ventilation, and it will overheat. Certain Apple portables do support running with the clamshell closed, but with other constraints applied that make it clear that the user intends for the system to be awake and that more or less require that it have adequate access to cooling (keyboard attached, external monitor attached, explicit wakeup after the clamshell is closed). The consequences of overheating include emergency system shutdown (the most common and intended response to thermal distress), but can also include accelerated aging of components (backlight, fixed and optical disks, display, cosmetic surfaces) and more subtle effects. The system is designed to deal with being overheated occasionally and accidentally, but not as a routine matter of course, which is what your new mode of operation will facilitate. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by Andre-John Mas 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 31-Dec-07, at 18:50 , Terry Lambert wrote: Andre _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… There's no contradiction here. The original question is about "how do I write code that would thwart power and thermal management for undisclosed reasons?". Running with the lid closed and an external monitor/projector and keyboard+mouse is a fully supported configuration, so long as he accepts that without adequate ventilation and cooling, he's subject to thermal throttling. "Adequate" here translates to "must be better with the lid closed than for the same performance with the lid open". Most of the "cooked" machines I have seen -- I believe the most spectacular one was Guido's IBM ThinkPad, before he learned how to throttle it down and turn off the backlight under FreeBSD; Jordan can correct me -- were not Macs. They happened either because the OS software, like Linux or FreeBSD, failed to implement thermal management at all, or the drivers that were supposed to talk to them were bogus or missing, or someone deliberately went out of their way to disable it (i.e.: someone answered the question that was asked here, and a programmer took the information and shot his foot off). I am surprised that this is not handled directly by the firmware, with the OS having an optional over-ride. The way I see it hardware should be protecting itself against software that doesn't fully recognise the hardware limitations, and only an explicite over-ride should change this behaviour. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by Andrew James 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:cc:message-id:from:to:in-reply-to:content-type:content-transfer-encoding:mime-version:subject:date:references:x-mailer; bh=tSkW6djZcxKA0f1v5Pqhq8YHa5V85SQVqW6rCgu24vs=; b=wlKpHneAzSkYk+beaXfMCMNNOPXA/+S6iKeFGXJTvqKmePCCSs+gsYyfij1bP21SiZvZzr8ADzvV6gt8z73g/WtcokzJymGR7T99b5e1Pchv58U4jh4DLyC3AGMAcZbCTeHMjQ8KsQOJE0rjm6ebvXTmLmoFojanEvug+1udH1M= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=cc:message-id:from:to:in-reply-to:content-type:content-transfer-encoding:mime-version:subject:date:references:x-mailer; b=BXrQ/BsvYMfOR8bBG8yEr7OPQiRIqZDL7mmnriR3EA2eTe1+W/OSO0fHp6qfT7zuYKyCMnr2iwSKpkOTDsBjo+jqSXMXeL6eeb6eB0PAG5fczIc0tOFvKQwFGslh+XilSeIG74b6kJRmL7munCz/TIw6AUfBU8ciHyQ7cfXxLE8= PS. Insomnia dev code is available at http://files.semaja2.net/Insomnia.zip On 01/01/2008, at 4:43 PM, Amanda Walker wrote: --Amanda _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Ill try those other mailing lists, and id like to remind everybody the solution exists to a extent as i am a developer of Insomnia and InsomniaX. However i would just like to take this back to basics and see if anyone can cover the biggest problem. So if everybody can forget just for a second of the main goal, the mission at the moment is to find a way to receive the notification of the lid state changing that does not mix up with the power state changes. My only idea that would of made sense would of been to use the "message" function and inside that check the power state and if the power state has changed then ignore the message (this would fix the problem of the power and lid state sharing the identical message signals) virtual IOReturn message(UInt32 type, IOService *provider, void *argument = 0); On Jan 1, 2008, at 12:03 AM, Andrew James wrote: Ok well the truth is im trying to create the over ride that will let you run the laptop in closed lid mode with out the extra hardware, it is a very needed feature. However i do plan on adding features to ensure the safety of the laptop (if temps get to high it will auto sleep etc) I'm still not clear on why it's "very needed", but I'll take your word for it. The software to fix this has been around for a long time (Insomnia.kext) but the current implementations used at the moment does not allow for some future features we wished to implement due to the message/notifications problems i mentioned earlier. I remember Insomnia from KisMac, though that use case is one where I'd be very worried about thermal problems (walking around with a laptop running in a backpack or bag). Something like a battery powered Mini (like <http://www.pamurray.com/manga/macbat/>) might be a better approach (or using a PDA to gather data, and upload to the laptop later). However, I do agree that there are other use cases where running a laptop while closed would be useful. PS. In a small survey i conducted the laptops in lid closed state did not show any major increases in temperatures over a duration of a few hours in normal operations (Not high CPU tasks) In an office environment with reasonable airflow, I am not too surprised--I've run my MacBook Pro for hours with an external monitor attached and the lid closed, for example, without any apparent ill effect. If I had to do this for a specific project, I'd probably end up building a custom kernel; it would surprise me if you could do everything you need to from a kext. You could always have your kext interpose itself by patching the running kernel, but this is (a) pretty risky, and (b) ties you to particular versions of the OS. If you're tied to particular OS revisions, you might as well just build a new kernel. And if we're in the realm of kexts and custom kernels, darwin- drivers or darwin-kernel might be a better place to ask this sort of question than darwin-dev. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by Amanda Walker 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com --Amanda _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Jan 1, 2008, at 12:03 AM, Andrew James wrote: Ok well the truth is im trying to create the over ride that will let you run the laptop in closed lid mode with out the extra hardware, it is a very needed feature. However i do plan on adding features to ensure the safety of the laptop (if temps get to high it will auto sleep etc) I'm still not clear on why it's "very needed", but I'll take your word for it. The software to fix this has been around for a long time (Insomnia.kext) but the current implementations used at the moment does not allow for some future features we wished to implement due to the message/notifications problems i mentioned earlier. I remember Insomnia from KisMac, though that use case is one where I'd be very worried about thermal problems (walking around with a laptop running in a backpack or bag). Something like a battery powered Mini (like <http://www.pamurray.com/manga/macbat/>) might be a better approach (or using a PDA to gather data, and upload to the laptop later). However, I do agree that there are other use cases where running a laptop while closed would be useful. PS. In a small survey i conducted the laptops in lid closed state did not show any major increases in temperatures over a duration of a few hours in normal operations (Not high CPU tasks) In an office environment with reasonable airflow, I am not too surprised--I've run my MacBook Pro for hours with an external monitor attached and the lid closed, for example, without any apparent ill effect. If I had to do this for a specific project, I'd probably end up building a custom kernel; it would surprise me if you could do everything you need to from a kext. You could always have your kext interpose itself by patching the running kernel, but this is (a) pretty risky, and (b) ties you to particular versions of the OS. If you're tied to particular OS revisions, you might as well just build a new kernel. And if we're in the realm of kexts and custom kernels, darwin-drivers or darwin-kernel might be a better place to ask this sort of question than darwin-dev. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Cancel Sleep
by David Alger 01 Jan '08

01 Jan '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0