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

Re: "Kicking" the screensaver/display timers
by Ethan Bold 21 Jul '08

21 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -Ethan On Jul 18, 2008, at 5:29 PM, Steve Checkoway wrote: UpdateSystemActivity( IdleActivity ); -- Steve Checkoway _______________________________________________ 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/ebold%40apple.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… Please don't use UpdateSystemActivity(). It will work, but IOPMAssertionCreateWithName()* is a better option. - UpdateSystemActivity() is in the Carbon framework; using it may cause you to unnecessarily link with this large & bulky framework, affecting app launch times and memory usage. - User's cant identify which process is keeping the display awake when you call UpdateSystemActivity. If you use IOPMAssertionCreateWithName(), OS X can keep track of which processes are responsible (visible by running "pmset -g"). - You only have to create an assertion once to keep the display awake; you need to call UpdateSystemActivity() once every 25-30 seconds to cause a fake keyboard press. It's cleaner code for you. Until now, UpdateSystemActivity() was the only game in town, but please don't use it anymore! We've got a better alternative. * Prefer IOPMAssertionCreateWithName(), which lets you pass a string describing the purpose of the assertion (for debugging & diagnostic use for Apple & yourself) over the vanilla IOPMASsertionCreate. If you invoke IOPMAssertionCreate from multiple points in your code; naming the assertions differently will help identify leaks. On Fri, Jul 18, 2008 at 05:24:45PM -0600, Shayne Wissler wrote: I need an API call/approach for "kicking" the the display-sleep/screensaver timers, as if the user had moved the mouse or pressed a key. I need a program to be able to wake up the display and/or stop the screensaver, without having the user interact with the mouse or keyboard. (E.g., for an app used in a classroom situation to broadcast a message to a Mac hanging on a wall that happened to be in sleep mode). This email sent to ebold(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: efficient driver/driver communication
by Michael Smith 21 Jul '08

21 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Jul 18, 2008, at 12:04 PM, darwin-dev-request(a)lists.apple.com wrote: Something like this: class foo .... IONotifier *notify; bar *partner; bool match_routine(void *ref, IOService *newService); ... }; bool foo::start(IOService *provider) { bool match_routine(void *ref, IOService *newService) { partner = OSDynamicCast(bar, newService); ... } = 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 is the most efficient way of moving data between two kexts?Are the IOService message/messageClient calls the recommended way to do this? Do they involve memcpy? The most efficient technique is to use IOService matching to locate the partner kext and then make direct method calls between the two. Since your kext isn't really discoverable until after your start routine has returned, you are best off using a notification to find the other kext. ... notify = addNotification(gIOPublishNotification, serviceMatching("bar"), match_routine, this, 0); registerService(); return(true); } Once foo has found bar, it can introduce itself with a registration call and you can make direct inter-class calls. Note that it can be risky doing this if the two have separate workloops, since the potential for deadlock exists. Depending on what you're actually doing, you may find it convenient to have one class defer most of its setup until it finds the other, and then call getWorkLoop() in order to share workloops. This email sent to site_archiver(a)lists.apple.com

1 0

Re: efficient driver/driver communication
by Godfrey van der Linden 21 Jul '08

21 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Godfrey On 2008-07-21, at 10:10 , Larry Emdur wrote: Yes, these are both IOKit drivers. Both drivers are mine and I can share headers. The KEXTs would be separate binaries. I assume it still safe to pass pointers around in this case, but what is the "fastest" way to do pass the pointers around? Laz. You are going to have to give us more context! Godfrey On 2008-07-19, at 4:53 , Larry Emdur wrote: _______________________________________________ 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/gvdl%40mac.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… Again it comes down to how the two drivers are related? Are they client/provider or superclass/subclass or just totally random drivers. How do you rendezvous the two drivers? On Sun, Jul 20, 2008 at 4:01 PM, Godfrey van der Linden <gvdl(a)mac.com> wrote: What is the driver environment you are running in? I assume IOKit. Are both drivers yours? Do you have a provider/client connection? Can you include headers between the two kexts? Do they live in the same binary? What is the memory going to be used for? The message/messageClient code is mainly used to call a subroutine in client driver from a provider. Finally within the kernel every thing lives in the same address space so you don't have to copy data you can just pass pointers. What is the most efficient way of moving data between two kexts? Are the IOService message/messageClient calls the recommended way to do this? Do they involve memcpy? This email sent to gvdl(a)mac.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: efficient driver/driver communication
by Larry Emdur 21 Jul '08

21 Jul '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:in-reply-to:mime-version:content-type :content-transfer-encoding:content-disposition:references; bh=7l/kTcrwpQvF1e1OuWYrh+x87Reo3W8v6v4MmxgWYgE=; b=xLzvKoa9gM7Hv33hDtoklxLRhehwPYOtwiwSBanEc2j75M5cyMPkq2pYtdPQpxBGtK KXjNiHrXG6rrEX8P0vpc6rEjX7RBa80DT3HdxfSNUmcSG+UBJJOild3Me652QyS322d3 lbx54A09U0qCsKe6L1IZvSWPC0TZBNn1aMUcI= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=lrCr7/dZI6xhRMbVCGlNU6NaQY4Hog4/3V4st/ixNIpy23x+hI60zFlL0ZQAkSFvKx xojFCi4zeTX3DttJYpv7XTNkgUeR/10SkEdKMuPAnF0gpev5s5dsX4NEJe40b6rU5OJG cDR8o91z+EL0d2CuWEAGm4tEDTHT5QCaTHlvw= Yes, these are both IOKit drivers. Both drivers are mine and I can share headers. The KEXTs would be separate binaries. I assume it still safe to pass pointers around in this case, but what is the "fastest" way to do pass the pointers around? Laz. On Sun, Jul 20, 2008 at 4:01 PM, Godfrey van der Linden <gvdl(a)mac.com> wrote: > You are going to have to give us more context! > > What is the driver environment you are running in? I assume IOKit. > Are both drivers yours? Do you have a provider/client connection? Can you > include headers between the two kexts? Do they live in the same binary? > What is the memory going to be used for? > > The message/messageClient code is mainly used to call a subroutine in client > driver from a provider. > Finally within the kernel every thing lives in the same address space so you > don't have to copy data you can just pass pointers. > > Godfrey > > On 2008-07-19, at 4:53 , Larry Emdur wrote: > >> What is the most efficient way of moving data between two kexts? >> Are the IOService message/messageClient calls the recommended way to do >> this? Do they involve memcpy? >> > _______________________________________________ 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: efficient driver/driver communication
by Godfrey van der Linden 20 Jul '08

20 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com You are going to have to give us more context! Godfrey On 2008-07-19, at 4:53 , Larry Emdur wrote: _______________________________________________ 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 is the driver environment you are running in? I assume IOKit. Are both drivers yours? Do you have a provider/client connection? Can you include headers between the two kexts? Do they live in the same binary? What is the memory going to be used for? The message/messageClient code is mainly used to call a subroutine in client driver from a provider. Finally within the kernel every thing lives in the same address space so you don't have to copy data you can just pass pointers. What is the most efficient way of moving data between two kexts? Are the IOService message/messageClient calls the recommended way to do this? Do they involve memcpy? This email sent to site_archiver(a)lists.apple.com

1 0

Mac OS 10.5.4, CF Lite, and iPhone 2.0 Source Posting
by William Siegrist 19 Jul '08

19 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com You can find it all at the usual location: http://www.opensource.apple.com/darwinsource/ Thank you. ---- William Siegrist Darwin Team _______________________________________________ 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 The Darwin Team is pleased to announce the release of the open source components for Mac OS 10.5.4 and iPhone 2.0 (SDK included). As usual, CF Lite for 10.5.4 has not been posted yet, but we _did_ post CF Lite for 10.5.3. We'll have the latest CF Lite up soon hopefully. smime.p7s

1 0

Re: "Kicking" the screensaver/display timers
by Steve Checkoway 19 Jul '08

19 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.5.17+20080114 (2008-01-14) On Fri, Jul 18, 2008 at 05:24:45PM -0600, Shayne Wissler wrote: > I need an API call/approach for "kicking" the the > display-sleep/screensaver timers, as if the user had moved the mouse > or pressed a key. I need a program to be able to wake up the display > and/or stop the screensaver, without having the user interact with the > mouse or keyboard. (E.g., for an app used in a classroom situation to > broadcast a message to a Mac hanging on a wall that happened to be in > sleep mode). UpdateSystemActivity( IdleActivity ); -- Steve Checkoway _______________________________________________ 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: "Kicking" the screensaver/display timers
by Ethan Bold 18 Jul '08

18 Jul '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com It's described in the 10.5 Darwin release notes here: http://developer.apple.com/releasenotes/Darwin/RN-IOKitPowerManagment/index… And discussed in practice here on a cocoa mailing list: http://www.cocoabuilder.com/archive/message/cocoa/2008/6/19/210697 -Ethan On Jul 18, 2008, at 4:24 PM, Shayne Wissler wrote: Hello, I need an API call/approach for "kicking" the the display-sleep/screensaver timers, as if the user had moved the mouse or pressed a key. I need a program to be able to wake up the display and/or stop the screensaver, without having the user interact with the mouse or keyboard. (E.g., for an app used in a classroom situation to broadcast a message to a Mac hanging on a wall that happened to be in sleep mode). Any pointers to a relevant API would be appreciated. Shayne _______________________________________________ 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/ebold%40apple.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… Check out IOPMAssertionCreate() with type kIOPMAssertionTypeNoDisplaySleep. This is new in 10.5. This email sent to ebold(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

"Kicking" the screensaver/display timers
by Shayne Wissler 18 Jul '08

18 Jul '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:mime-version:content-type:content-transfer-encoding :content-disposition; bh=I3SPbA8kQednsPJ4W3hm/MapHjnCyHHBO97xF/bLQTk=; b=gIFYUzHvYcFdoOz0lFSVqrWYi9bOJpHFVX7t8sCL05Kxy0cF6JaNLiXC0Su7lik+sz lisNxdRNJAPoxMfIHA9JwcnZW/MdBSYdi2YiK6ez2KgshsxnW27oS51JzYbfGBsibktX se5nXBpWkO0hbot23gPWnmaX0u3Ww09EbY0yo= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type :content-transfer-encoding:content-disposition; b=ZAxuHdlMaswcMMvcm8+DCpa03Hy5ptEKNCGoNkqy+yWF5kP1r2/5FBvc7jhi5of2LO 0ECErUgqje4lK7uH8cWdnBQJPMzBU6co8qU09iBk9ZmAEC1IjNbQ0F5BKtagDyoHA7+G dGZhRxgwdP4uplXbGNhes2rUv1+JK0LY5Nvjc= Hello, I need an API call/approach for "kicking" the the display-sleep/screensaver timers, as if the user had moved the mouse or pressed a key. I need a program to be able to wake up the display and/or stop the screensaver, without having the user interact with the mouse or keyboard. (E.g., for an app used in a classroom situation to broadcast a message to a Mac hanging on a wall that happened to be in sleep mode). Any pointers to a relevant API would be appreciated. Shayne _______________________________________________ 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

efficient driver/driver communication
by Larry Emdur 18 Jul '08

18 Jul '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:mime-version:content-type; bh=5pai6royN+R3evuQ8X4AILfEPoZiUkT06gZeZU50sLI=; b=g4DSwV/HbKQvyrSwuzmQMbbxpoeyjTltMY9VCep1v/ZENJaJuhctbsZvZEydRedwZm CiIWTosEvUdHkAgE0jbch4Zz7DqELad8LqM5AGsfGZgoHbLChkTtULoWcoVLCto2y8MI QrT3w103paIZERm4tsH1k8LPbPruB8zZPbBCc= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type; b=wgubjHxNVuj3RJ+CcXsoew+e1voLU2qGKqvhqFmhJ4KLHGzdJY5ecTPF5XHDhauRmu +GHOxONFffzPYnUQXFKbF6p2YclH3yEYW3yBjZz1HSxZ/u8dmq6mUWNnzwNJSePXjB+t +GsiEdgf0xy7a3MBVteAwACTx2vx01fkU2QfA= _______________________________________________ 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