January 2004 - Darwin-kernel

Re: question about mach ports and fork
by Jim Magee 21 Jan '04

21 Jan '04

On Jan 21, 2004, at 12:47 PM, Nathan T. Hjelm wrote: I know when fork/vfork are called that all mach ports are invalidated in the child process. Actually, only fork() and exec() invalidate the Mach ports. The current vfork() implementation actually arranges for the Mach task associated with the parent to be shared with the child (resulting in two procs running in one Mach task and having equal access to Mach resources like ports and VM). But it's only an implementation artifact and you shouldn't depend upon it. You must stick to the list of vfork()-safe APIs and behaviors (exec(), _exit(), not popping the stack, etc...) to assure continued operations regardless of the implementation. Is there any way to re-validate these ports and have them shared between the child and parent processes? I am particularly interested in being able to use IOServices assigned to the parent in the child process. There is no [guaranteed safe] way to re-validate them under the same port name with the same number of references (thus making the port names sprinkled throughout the process memory valid again). But there are some common techniques used to pass send rights to these ports from the parent to the child under a [randomly chosen] new name. And many of the frameworks allow for instantiation of objects passing an explicit port to use - rather than looking them anew. I'm not sure about IOService, however. You can't just open a new IOKit service connection in the child? This is sometimes impossible because the parent was privileged and the child not (so I feel your pain). Solving the problem in a general way at the Mach port level (by inheriting ports across fork()) would be nice, but it violates general assumptions like only one receive right being in existence for a given Mach port. Changing all the code that depends on these assumption is almost as difficult as working around the fork() problem in the first place. --Jim _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

question about mach ports and fork
by Nathan T. Hjelm 21 Jan '04

21 Jan '04

I think I already know the answer to this (being no) but I thought I would ask. I hope this is the correct list to address this too... I know when fork/vfork are called that all mach ports are invalidated in the child process. Is there any way to re-validate these ports and have them shared between the child and parent processes? I am particularly interested in being able to use IOServices assigned to the parent in the child process. thanks, -Nathan _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: darwin-kernel digest, Vol 2 #459 - 10 msgs
by Jim Magee 21 Jan '04

21 Jan '04

On Jan 20, 2004, at 5:27 AM, ollie wrote: You could have your kext either launch the daemon using KUNCExecute or else have it send a kernel event when ever it requires action from your daemon. This will remove all polling from your current design. Please, do NOT use KUNCExecute(). It's a "wing and a prayer" mechanism (there are many situations where it will fail silently) only added for people who were too stubborn to switch to the "driver down from the daemon, not the out from the kernel" model of execution under Mac OS X/Darwin. Each release, the timing of system boot changes, and therefore the timing of when KUNCExecute() will succeed also changes. There are several mechanism available to register the daemon from user-space for launching only when needed. These are much more reliable than the KUNCExecute() mechanism, as the user-level dependencies are always satisfied before the daemon can be registered in the first place. --Jim _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: IOATAController device blocking bus
by RJ Auburn 21 Jan '04

21 Jan '04

*This message was transferred with a trial version of CommuniGate(tm) Pro* Just for closure on this issue it looks like the hard drive may be going bad. When I swapped to a new drive it started working ok and the clicks went away. I have not had a chance to do a lot of testing on the drive in my new firewire enclosure yet to see if it still happens there. RJ --- RJ Auburn CTO, Voxeo Corporation tel:+1-407-418-1800 On Jan 15, 2004, at 13:31, RJ Auburn wrote: Garth, Thanks for the info... It also looks like IOATADevConfig.cpp is missing the closing brace at the end of the file. Once I fixed that it all complies ok... RJ --- RJ Auburn CTO, Voxeo Corporation tel:+1-407-418-1800 On Jan 15, 2004, at 10:41, Garth Cummings wrote: *This message was transferred with a trial version of CommuniGate(tm) Pro* Hi RJ, On Jan 14, 2004, at 10:21 AM, RJ Auburn wrote: *This message was transferred with a trial version of CommuniGate(tm) Pro* Is anyone else having problems getting IOATAFamily to compile? When try to compile the 10.3.2. version I get the following errors that are below. Maybe I'm missing some magic bit of the build setup or something. RJ <snippage> In file included from IOATAController.h:34, from ATADeviceNub.cpp:36: ATATimerEventSource.h:34:1: unterminated #ifndef from ATADeviceNub.cpp:36: In file included from IOATAController.h:34, ATATimerEventSource.h:34:1: unterminated #ifndef No, the compiler message is correct. Somehow the version of ATATimerEventSource.h in Darwin lost the last line, which should read: #endif /*_ATATIMEREVENTSOURCE_H*/ I've passed this on to the folks internally to figure out how this happened. Thanks, --gc _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

SCSI user client
by Girish 21 Jan '04

21 Jan '04

Hello Darwin Devloper I am searching for sample code of scsi user client.But I am not able to find sample code. Please anyody help me to find the sample code Regards, Girish Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: sosend, soreceive, soconnect
by Jim Magee 20 Jan '04

20 Jan '04

On Jan 20, 2004, at 5:57 AM, Mike Smith wrote: On Jan 19, 2004, at 9:29 PM, Justin Walker wrote: On Monday, January 19, 2004, at 09:05 PM, Matt Jaffa wrote: Please advise me in this area, I am looking to finish my project quickly. You will need to adjust your goals. It doesn't appear that you have much experience with this kind of development, and this is tricky code. Amen. Tuning this will take some time, too. Look at the Darwin sources ... <snip>. Most everyone first choses a user<-->kernel communication mechanism based on the way data is supposed to be handled in the "typical" case. But the real trick is determining what you want the edge-case semantics to be. For instance, what happens when the daemon is not keeping up with the filter and more data keeps coming in? Do you drop the data? Do you queue it? For how long and how much? If data is dropped, how do you detect it and recover from it? The answers to these questions, matched against the behavioral characteristics of the user<-->kernel communications mechanism available to you, will go much further towards pointing out a selection than an arbitrary "it's networking code so it has to be a socket" kind of dogma (but obviously a socket tends to match the semantics _most_ networking code expects). --Jim _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: sosend, soreceive, soconnect
by Mike Smith 20 Jan '04

20 Jan '04

On Jan 19, 2004, at 9:29 PM, Justin Walker wrote: On Monday, January 19, 2004, at 09:05 PM, Matt Jaffa wrote: Amanda and others: I would love to implement what SimpleUserClient is doing but with my application, instead of the KERNEL socket AF_UNIX implementation, I don't know exactly how to do it. You can't use the UserClient interface for your project, since it is designed for IOKit usage, and yours is not an IOKit kext. It's relatively trivial to become enough of an IOKit kext to get a node in the IORegistry, at which point attaching a UserClient is straightforward as per the documentation. A UserClient isn't a bad way to go about doing this, but you could use a sysctl node and achieve similar functionality for much less overhead. Neither is going to be super-performant, but then this doesn't sound like a performance-intensive application. And what if the daemon is not running, it won't be polling for work to do, so the stuff that is sitting in semaphore_wait mode will always wait and nothing will work. If your user-mode component is not running, the kernel-mode component should exit. 'Exit' is perhaps not quite the right way to describe it. You would generally do this: - Start daemon. - Daemon loads filter module. - Daemon waits for filter module to initialise, then calls the sysctl to get work. - Loop... - When an exit request is received, the daemon unloads the filter module. Don't start the daemon using KUNC, do it the other way around. And in fact, the above's not actually good enough; you need a simple and foolproof wrapper that loads and unloads the module, since a crash in the daemon will leave the filter module loaded. Please advise me in this area, I am looking to finish my project quickly. You will need to adjust your goals. It doesn't appear that you have much experience with this kind of development, and this is tricky code. Amen. Tuning this will take some time, too. Look at the Darwin sources to see how to implement a SYSCTL_PROC node. Make use of the CTL_NOLOCK flag (to prevent you locking out other sysctl consumers). You can both return work to the kernel and pull work from it in a single call. In your daemon, use sysctlnametomib() to look your node up once and then invoke sysctl() directly against the returned mib to avoid the lookup overhead on every call that you'd get with sysctlbyname(). One final question; have you considered using a divert(4) socket? = Mike _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: darwin-kernel digest, Vol 2 #459 - 10 msgs
by ollie 20 Jan '04

20 Jan '04

On 20 Jan 2004, at 06:00, darwin-kernel-request(a)lists.apple.com wrote: From: Matt Jaffa <matt(a)knowligent.com> Subject: Re: sosend, soreceive, soconnect Date: Mon, 19 Jan 2004 22:05:56 -0700 To: Amanda Walker <amanda(a)alfar.com> Amanda and others: I would love to implement what SimpleUserClient is doing but with my application, instead of the KERNEL socket AF_UNIX implementation, I don't know exactly how to do it. So I have a KEXT that is a socket/Protocol filter. And I insert my code to do all the checking within the send function that is within my filter. How am i supposed to do this, my daemon will keep checking with the KEXT to see if it has work to do? And within my send function i block it with a semaphore??? Until the daemon picks it up, signals then does work and then the daemon goes back to checking for work to do???? And what if the daemon is not running, it won't be polling for work to do, so the stuff that is sitting in semaphore_wait mode will always wait and nothing will work. A number of points here Matt.. As pointed out already you can't use the IOKit user client as your not an IOKit kext, unless you want to do a lot of changing to you overall design. I'd recommend that you use the kern control mechanism for communicating with your kext in this instance. It's more than adequate for small simple control like communications and it quite easy to use. Check out the header /sys/sys_domain.h. You can see it in use somewhere ? in the darwin sources.. I remember finding it once upon a time. You could have your kext either launch the daemon using KUNCExecute or else have it send a kernel event when ever it requires action from your daemon. This will remove all polling from your current design. Lastly, you could create a new thread in your kext to handle the comms from user space. e.g. thread_t mattsCommThread = kernel_thread(kernel_task, comms_handler); In this way, you can block the main thread, have the daemon connect to your kext and give it's instructions etc., and then unblock your main thread. Be sure to put a timer on this block in case your daemon does not respond correctly and you block the main thread indefinitely. Your comms thread will really do very little, except register with the controller, then loop around a block. Any incoming comms will wake your thread, be processed by your write function, as entered in the kern_ctl_reg struct, and then it'll block again, after it has unblocked your main thread of course. Please advise me in this area, I am looking to finish my project quickly. Good Luck, -Ollie. Thanks, Matt _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: sosend, soreceive, soconnect
by Matt Jaffa 20 Jan '04

20 Jan '04

Justin, just so you know i am not asking anyone to program my program for me, I am asking for suggestions, I am afraid to open a DTS as i fear i will get the same negative response as i get here. Anyone who has no experience with Kernel code at the beggining of course is going to be asking so many questions, I just ask them since documentation on this stuff is lacking. Matt On Jan 19, 2004, at 10:29 PM, Justin Walker wrote: On Monday, January 19, 2004, at 09:05 PM, Matt Jaffa wrote: Amanda and others: I would love to implement what SimpleUserClient is doing but with my application, instead of the KERNEL socket AF_UNIX implementation, I don't know exactly how to do it. You can't use the UserClient interface for your project, since it is designed for IOKit usage, and yours is not an IOKit kext. So I have a KEXT that is a socket/Protocol filter. And I insert my code to do all the checking within the send function that is within my filter. How am i supposed to do this, You need to review kernel code. I don't think there is anyone else with enough time to design and program your project for you. my daemon will keep checking with the KEXT to see if it has work to do? And within my send function i block it with a semaphore??? Until the daemon picks it up, signals then does work and then the daemon goes back to checking for work to do???? As I mentioned earlier, there is code in the 'bsd/netat' directory that does something similar. NFS clients and servers do similar things, but the code is much more complex than the AURP 'daemon' that runs (or, more accurately, ran) in kernel mode. As long as you want to block awaiting incoming datagrams, you need a thread on which to sleep, and therefore you need to create a kernel process. And what if the daemon is not running, it won't be polling for work to do, so the stuff that is sitting in semaphore_wait mode will always wait and nothing will work. If your user-mode component is not running, the kernel-mode component should exit. The AURP code should give a hint regarding the way to do this. See also the NFS client and server code. Please advise me in this area, I am looking to finish my project quickly. You will need to adjust your goals. It doesn't appear that you have much experience with this kind of development, and this is tricky code. What do you have running so far? Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | It's not whether you win or lose... | It's whether *I* win or lose. *-------------------------------------- *-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: sosend, soreceive, soconnect
by Matt Jaffa 20 Jan '04

20 Jan '04

So far i have the daemon running and the KEXT is already written, I just need to be able to communicate between the two and also tidy up the way i handle when there are many connections put on hold while my daemon is doing some work to see if it will allow the connection. On Jan 19, 2004, at 10:29 PM, Justin Walker wrote: On Monday, January 19, 2004, at 09:05 PM, Matt Jaffa wrote: Amanda and others: I would love to implement what SimpleUserClient is doing but with my application, instead of the KERNEL socket AF_UNIX implementation, I don't know exactly how to do it. You can't use the UserClient interface for your project, since it is designed for IOKit usage, and yours is not an IOKit kext. So I have a KEXT that is a socket/Protocol filter. And I insert my code to do all the checking within the send function that is within my filter. How am i supposed to do this, You need to review kernel code. I don't think there is anyone else with enough time to design and program your project for you. my daemon will keep checking with the KEXT to see if it has work to do? And within my send function i block it with a semaphore??? Until the daemon picks it up, signals then does work and then the daemon goes back to checking for work to do???? As I mentioned earlier, there is code in the 'bsd/netat' directory that does something similar. NFS clients and servers do similar things, but the code is much more complex than the AURP 'daemon' that runs (or, more accurately, ran) in kernel mode. As long as you want to block awaiting incoming datagrams, you need a thread on which to sleep, and therefore you need to create a kernel process. And what if the daemon is not running, it won't be polling for work to do, so the stuff that is sitting in semaphore_wait mode will always wait and nothing will work. If your user-mode component is not running, the kernel-mode component should exit. The AURP code should give a hint regarding the way to do this. See also the NFS client and server code. Please advise me in this area, I am looking to finish my project quickly. You will need to adjust your goals. It doesn't appear that you have much experience with this kind of development, and this is tricky code. What do you have running so far? Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | It's not whether you win or lose... | It's whether *I* win or lose. *-------------------------------------- *-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0