February 2004 - Darwin-kernel

User Login
by Matt Jaffa 26 Feb '04

26 Feb '04

Hi, Is there a way to tell if a User is logged in from within the Kernel, like inside my KEXT? If so how do you get the name of that user that is logged in, and the current active user. If not is there a way to do it from within my daemon? 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: opening files in the kernel
by Quinn 25 Feb '04

25 Feb '04

At 21:48 -0800 24/2/04, Mike Smith wrote: IOW, do it "just like HFS" wherever you can. Agreed. Also, there's no reason why the lack of support for storing your file's extents in the volume header should slow you down. The only reason HFS Plus needs that information in the volume header is so that it can find the catalogue and extents overflow files. Once you have access to those files, you can store the information about your special files there. You should probably hide your special files from clients (like the "HFS+ Private Data" directory; look through the source for hfs_privdir_desc for hints on how to do this) and also mark it invisible in the Finder information so that, if your volume is mounted by a version of HFS other than yours, it isn't seen by the user. S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Technical Support * Networking, Communications, Hardware _______________________________________________ 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: opening files in the kernel
by Mike Smith 25 Feb '04

25 Feb '04

On Feb 24, 2004, at 2:20 PM, Chris Bergmann wrote: Sorry, i should have given a better description. I am using the normal (internal) hfs_* routines to open the file. However, it's the locking routines (which are system-provided routines which work on the vnode) which I'm having trouble with. Eg. I try something like this: GetFileInfo(vcb, kRootDirID, ".filename", &cnattr, &cfork); .. retval = hfs_getnewvnode(hfsmp, NULL, &cndesc, 0, &cnattr, &cfork, &vp); VREF(vp); VOP_UNLOCK(vp, 0, p); .. which I thought would have been a safe enough operation? this is done with the kernel funnel locked. Again, avoid using a vnode at all. Try to use the lower levels of HFS to manipulate the file directly. Having a vnode of any kind attached to your 'magic' files is just asking for trouble, unless you are going to treat them 100% like real files. IOW, do it "just like HFS" wherever you can. = 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: opening files in the kernel
by Chris Bergmann 24 Feb '04

24 Feb '04

Sorry, i should have given a better description. I am using the normal (internal) hfs_* routines to open the file. However, it's the locking routines (which are system-provided routines which work on the vnode) which I'm having trouble with. Eg. I try something like this: GetFileInfo(vcb, kRootDirID, ".filename", &cnattr, &cfork); .. retval = hfs_getnewvnode(hfsmp, NULL, &cndesc, 0, &cnattr, &cfork, &vp); VREF(vp); VOP_UNLOCK(vp, 0, p); .. which I thought would have been a safe enough operation? this is done with the kernel funnel locked. chris On 25/02/2004, at 2:34 AM, Mike Smith wrote: On Feb 24, 2004, at 1:11 AM, Chris Bergmann wrote: I've got my own HFS+ file system, it's a runtime loaded kext, it works. I'm now trying to make some modifications. I You're a filesystem; use your internal routines to manipulate these files. You don't need (or want) to use the system interfaces - that'd be a major layering violation. = 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: Swap partitions
by Shawn Erickson 24 Feb '04

24 Feb '04

On Feb 24, 2004, at 11:25 AM, Mickey S. wrote: Hello, I am new to the Darwin / MacOS world and I am looking for some guidance. For starters I have heard any number of pros and cons about the swap file system used in Darwin. I understand the basics of the Dynamic Pager system ,but I am wondering why there appears to be a lack if support to dedicated swap partitions. Is this a limitation of XNU? or is this to make Apple's end users happy I.E. when they buy a 150G disk they don't see 149G available. ;) . It is not a limitation of XNU, if fact no limitation exists, one can create swap partitions and target the pager at it. Of course doing so is not a normal configuration or support by Apple (at least I don't think Apple supports it). If you want to do it do it... for more information on the dynamic page try "man dynamic_pager". Just understand that is not normal and later updates make break your redirected swap. Having a swap partition will likely greatly limit how big the swap file(s) can grow and hence limit what the system can do before running out of virtual memory backing (which often leads to crashes...). So slicing off a percentage of the relatively huge disks one gets in systems today can be counter productive. Some folks will state that having a swap partition reduces the chance of file / swap file fragmentation. It does but the data in the swap files is already greatly fragmented (for various obvious reasons) as is the page in/out pattern and that is likely to be more of an issue then fragmentation of swap file itself. Additionally if file fragmentation of the swap is some how an issue, your system is already greatly hindered by the swapping taking place, it will be the least of your issues. Now targeting an independent device for swap can make sense for some folks / environments. -Shawn _______________________________________________ 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

Swap partitions
by Mickey S. 24 Feb '04

24 Feb '04

Hello, I am new to the Darwin / MacOS world and I am looking for some guidance. For starters I have heard any number of pros and cons about the swap file system used in Darwin. I understand the basics of the Dynamic Pager system ,but I am wondering why there appears to be a lack if support to dedicated swap partitions. Is this a limitation of XNU? or is this to make Apple's end users happy I.E. when they buy a 150G disk they don't see 149G available. ;) . Thanks a bunch _______________________________________________ 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: opening files in the kernel
by Mike Smith 24 Feb '04

24 Feb '04

On Feb 24, 2004, at 1:11 AM, Chris Bergmann wrote: I've got my own HFS+ file system, it's a runtime loaded kext, it works. I'm now trying to make some modifications. I You're a filesystem; use your internal routines to manipulate these files. You don't need (or want) to use the system interfaces - that'd be a major layering violation. = 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: opening files in the kernel
by Chris Bergmann 24 Feb '04

24 Feb '04

Generally if you need to read/write to files from a KEXT in mass, you use the IOKit messages or sockets or sysctl depending on what you're doing to a user daemon to do file IO. Probably not suitable in this instance; i'm sitting a little too low-level to try this. As a basic background to where i'm coming from- I've converted Apple's HFS+ code to compile with C++ and made the necessary modifications to run as a KEXT; now i'm trying to create some additional "system" files. I'm baffled why you converted this to C++, file systems are not IOKit objects, and are generally C only entities. KEXTs support C++. I like C++. In this particular case, the one thing which caused me to convert it to C++ is the ability to then namespace the whole KEXT. Having namespace conflicts between Apple's in-kernel HFS+ and my kext built from [originally] the same source was Not Fun. That's fixed now. You could do this in straight C with a bit of work, however i'm mainly playing and learning and this technique suited me. Apple's existing system files (extents tree, etc) are treated rather specially by the filesystem and I don't want to go down the same path- instead I would like to open a file in much the same way as a userland process would (except using the vfs directly since i'm in a KEXT.) This is generally considered a bad thing to do. The locking issues are complex in a file system and you can easily bypass a lot of the locking rules by accessing these functions out of context. Its incredably easy to leave a vnode locked somewhere and not have the system panic till much later on. Yes, i realise. That's pretty much what i'm looking for information on. What are you trying to accomplish anyways? If you really are trying to add another file to the "special" area of HFS+ in CNID's 2 though 15 then I'd imagine you'd probably have to make your own HFS+ file system and build that into a new kernel, or try to make it a loadable file system and put it at a higher priority for HFS in the filesystem bundles. Either way this can get quite annoying pretty quickly. I've got my own HFS+ file system, it's a runtime loaded kext, it works. I'm now trying to make some modifications. I require additional storage space for extra data that the modifications will require. This storage space is equivalent to the extents-tree and other system files. I could take the exact same approach that apple does for the extents tree, journalling files, etc, however I don't see any way of doing that without breaking compatability with the on-disk HFS format. There appears to be a preset number of system files which have their extents stored in the master blocks. I didn't see any [compatible] way of extending that, therefore i'm trying to open a "normal" file after the end of the normal HFS+ mount sequence. If there is some 'hfs-compatible' technique for extending the number of actual system files (the ones with their extents stored in the master block) then please let me know and i'll go do that and save myself the hassle.. thx, chris _______________________________________________ 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

Overcoming Permissions
by Matt Jaffa 24 Feb '04

24 Feb '04

Hi, I have written an NKE and everything is working great now, except in my send function, basically an extension of the userland socket that is sending information on it, it calls soconnect, but a process in order to connect to my listening socket has to be running as root. But when Safari or Mails traffic is intercepted it trys to call soconnect, but it denies since it doesn't have access. Sorry that sounds like a lot of garbage in other words, those familiar with a NKE the send function is intercepting everything sent on a socket, the way i chose to do IPC with my daemon is with sockets as well. But if the intercepted process is a process like smbd, netinfod, pppd, basically those processes running as root they can connect to my listening daemon socket, but Safari, Mail, and those running as user processes when it goes through the intercepted send function, it can't connect to my userland listening socket. But after it is connected with cupsd, netinfod, etc... sending and receiving works fine no matter what mode the process is running under. Any help would be appreciated, I can change my user daemon to run under as a user process and connecting to it works just fine, but a startup daemon always is started up as root. And I would like my daemon to run under root too for the protection from kill. 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: Changes to Disk Arbitration in Panther
by Marek Kozubal 24 Feb '04

24 Feb '04

On Mon, 16 Feb 2004, Ron Aldrich wrote: > I tried hard coding the full path to SAI_udf.util > (/System/Library/Filesystems/SAI_udf.fs/SAI_udf.util) in my Info.plist, > but that had no effect. I also tried changing my probe order value > from 500 to 1500 (Apple's udf probe order value is 1000), but that just > made it so that disk arbitration never even tried to probe SAI_udf. The .util path is relative to the Resources/ Directory within the bundle. Make sure that your .util is returning from main() the appropriate error codes as well. They are defined in loadable_fs.h i think its called. ie: int main(int,char**) { /* do stuff */ return <code from loadable_fs.h>; } If you just blindly return 0, DiskArb will think you failed. There are very specific series of return values you need to return w/each arguments to successfully mount a disk. > I tried searching around in what bit of the darwin sources I currently > have available (xnu-517.3.7) for "probed disk" to find where the error > message is being generated, but I can't find that string, and since > Apple seems to have done their best to make it extremely difficult to > check all of darwin out from CVS, I really don't want to have to load > it project by project, just to search it. (Does anyone know a trick > for checking all of darwin out at once? Or failing that, how to get a > list of all of the module names?) Its in diskarbitration. Its got its own entry on Apple's CVS pages. --- Marek Kozubal marek(a)portents.com _______________________________________________ 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