December 2002 - Darwin-kernel

Re: Float processing not working
by Sean McBride 11 Dec '02

11 Dec '02

Heiko Panther (heiko.panther(a)web.de) wrote on Wed, 11 Dec 2002 18:02:49 +0100: >In my kext, built with KERNEL_MODULE=YES and MODULE_IOKIT=YES, float is >working wrong. The technique apple suggests, which I have done successfully, is to limit your floating point work to a shared library. Then this lib can be built with appropriate project settings. I forget the name, but one of Apple's sample shows this. Its some coreaudio sample driver I think. _______________________________________________ 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: Float processing not working
by nibs 11 Dec '02

11 Dec '02

On Wednesday, December 11, 2002, at 12:02 PM, Heiko Panther wrote: Alternatively, how can I override the -msoft-fpu option? i believe one way around the issue is to create a static archive library of the fp routines you want, and then link that with your kext. and then you'll have true fp in the kext. nibs _______________________________________________ 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

Float processing not working
by Heiko Panther 11 Dec '02

11 Dec '02

In my kext, built with KERNEL_MODULE=YES and MODULE_IOKIT=YES, float is working wrong. I figure that KERNEL_MODULE=YES adds the compiler option -msoft-float, and that the float emulation is broken in gcc3. Is that true? Assuming it is true, I would like to switch the fpu emulation off. Adding -mhard-fpu to the cflags did not change anything. I figure the settings made because of KERNEL_MODULE=YES override my setting. How can I find out which modifications are made because of KERNEL_MODULE=YES, so I can make them manually? Alternatively, how can I override the -msoft-fpu option? Thanks for pointers, Heiko Panther _______________________________________________ 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

3DFX Voodoo 4 Driver
by Sam Hart 10 Dec '02

10 Dec '02

I'm not sure where to post this as I subscribe to many of the Darwin lists... however I thought as the problems and solutions where mainly kernel related I would start my quest here. I have recently changed my video card from an ATI Rage 128 ( that came in my G4 ) to a 3DFX Voodoo 4 as the fan was knackered and the replacement came free - long story - however since I changed them over I have lost Linux ( no biggie ) and my Darwin install makes a mess of the screen, making it impossible to work with. Now the strange thing is when I boot off my Jaguar partition in verbose mode ( -v ) I get the same mess on the screen until I get to the login and then the screen goes back to normal. Jaguar works fine, well there is no acceleration ( don't know if there should be .... ) and DVD Player.app doesn't work and it is a lil slow ( oh.. I've already said that ) but it works .... When I drop into Darwin ( using ">console" at login ) Darwin appears to be running fine with the card. What I am wondering is, should I reinstall Darwin, would that fix the problem ? There is a chance that Darwin will not work with the card as it is, I don't know. The other thing is, what differences exist between Jaguar and Darwin as far as display drivers go ? The last thing is that I know there are quite a few people who have 3DFX cards who have had problems since upgrading ( the driver is pre OS X ) and are after a driver, a few people have talked about doing it but no one has, well not according to the searches I've done, there was a beta ( I think ) at a .mac site, but alas has disappeared ( as have a lot of .mac sites ... strange ;) ). What I am wanting to know is, does anyone know anything about writing a video driver in Darwin / OS X or writing a 3DFX driver in general, as I am not above getting my hands dirty and attempting one ( how bad can it be ? ). Of course I think I might jus have to put my hand in my pocket and buy a new card, rather than use an unsupported card. S a m : _ D _______________________________________________ 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: mbuf manipulation
by Justin C. Walker 09 Dec '02

09 Dec '02

On Monday, December 9, 2002, at 03:11 PM, Eric Long wrote: Hi all, I've been camped out reading TCP/IP Illustrated volumes, trying to digest BSD networking. Vol. 2 is a pretty good reference, albeit for a much older version of the stack. I'm trying to write a kext to perform some manipulation of incoming data. At present, I am intercepting data in an sbappend hook. Concerning mbufs, if I want to remove bytes... If, for example, I have a chain of 3 mbufs and each has 50 bytes of data, If I want to remove bytes 30-39 of mbuf 2: Can I simply shift bytes 40-49 back by 10 bytes, shorten the mbuf's m_len by 10, and the packet header length by 10, and expect things to be "work"? What if I set the mbuf's m_len to 0? That should work, at least as far as the socket layer is concerned. Whether the 'user' of the socket likes it is another question. If you set an mbuf's m_len to zero, that should cause the system to skip over the mbuf in question. However, that depends on the last guy who touched the code being prepared for that eventuality. This is supposed to be a supported feature. If I want to insert bytes... How nasty is that to deal with? Can I insert safely from sbappend? This can be arbitrarily nasty. Depends on what you want to do, of course, but in full generality, inserting bytes is just ugly code, and you have to make sure you take care of a lot of edge conditions (see, e.g., above :-}). Not sure what you mean by 'safely', but you should be able to manipulate an mbuf chain there with impunity, assuming your code doesn't blow chunks. Does anyone know of code examples I might look at that show removing/inserting bytes into mbuf chains? I'm not aware of any that deals with the general case. Most code just chops data off or adds it at either end. There are some places (see 'SharedIP' in the repository) that do a little bit of this. I see mbuf.h seems to have all the functions for allocating/freeing mbufs, but I'm not certain about the big picture of how this is implemented. The Wright/Stevens book has a pretty good "big picture" description of the mbuf stuff. Seems like I may have some concerns about exceeding space available in the socket buffer. Yes and no. It may not be immediately obvious, but the socket buffer limit does not represent preallocated space. It is intended to limit both the total byte count and the total mbuf space that can be enqueued on a single socket buffer. This is designed to make sure that one or two of the socket buffers don't end up with all the mbufs in the system hanging on them. This means that it's not going to really screw up anything internally if you go over by a little bit, or you go over once in a while. Ignoring the limits in general, though, will cause you (or your customers) problems, so you should be very leery of messing with this for production operation. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Men are from Earth. | Women are from Earth. | Deal with it. *--------------------------------------*-------------------------------* _______________________________________________ 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

mbuf manipulation
by Eric Long 09 Dec '02

09 Dec '02

Hi all, I've been camped out reading TCP/IP Illustrated volumes, trying to digest BSD networking. I'm trying to write a kext to perform some manipulation of incoming data. At present, I am intercepting data in an sbappend hook. Concerning mbufs, if I want to remove bytes... If, for example, I have a chain of 3 mbufs and each has 50 bytes of data, If I want to remove bytes 30-39 of mbuf 2: Can I simply shift bytes 40-49 back by 10 bytes, shorten the mbuf's m_len by 10, and the packet header length by 10, and expect things to be "work"? What if I set the mbuf's m_len to 0? If I want to insert bytes... How nasty is that to deal with? Can I insert safely from sbappend? Does anyone know of code examples I might look at that show removing/inserting bytes into mbuf chains? I see mbuf.h seems to have all the functions for allocating/freeing mbufs, but I'm not certain about the big picture of how this is implemented. Seems like I may have some concerns about exceeding space available in the socket buffer. I'm just looking for tips to help me figure this out. Thanks, Eric Long ___________________________________________________________________________ Macintosh Software Engineer Aladdin Systems, Inc. http://www.aladdinsys.com 245 Westridge Drive, Watsonville, CA 95076 (831) 761-6200 x628 _______________________________________________ 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: Problems with float in a kext
by Heiko Panther 08 Dec '02

08 Dec '02

Sean McBride wrote: >>What's wrong here? I know that kernel code usually uses float emulation. >>I couldn't find out which steps to take to activate the emulation. I >>tried -msoft-float for compiling and linking, but that did not help. > > If you want your kext to use the FPU you must NOT specify the -msoft- > float switch. Right. Well, I don't really want to use the FPU. The emulation would be good enough. Anyway, I get that error, no matter if I specify -msoft-float. > Also note that IOLog does not support %f to print floats. :( Must be that way to keep the kernel small :) I usually multiply by 1000 and cast to int. Actually I just noticed that I cast to UInt32 in my example. Here's the results when cast to int. Dec 6 17:36:52 Hyp mach_kernel: 500, 2147483647, 2147483647, -2147483648 Heiko _______________________________________________ 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: howto compile kernel for ktrace
by Justin C. Walker 06 Dec '02

06 Dec '02

On Friday, December 6, 2002, at 10:40 AM, Michael Marti wrote: Thanks a lot Justin here is what I found: The error: Protocol error: uncounted data discarded indicates that the tag does not exist on the server. Now I use the server: :pserver:freddiefreilaender@anoncvs.opensource.apple.com:/cvs/Darwin and tags like: Apple-201 (Apple-201.42 and Apple-201.42.3 don't work). Again, if you use those tags, verbatim, they won't work. CVS only likes '-', not '.'. I would stick with something closer to what your stock kernel was built from, because there have been bug fixes and improvements over the earlier versions. 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.

1 0

Re: howto compile kernel for ktrace
by Michael Marti 06 Dec '02

06 Dec '02

Thanks a lot Justin here is what I found: The error: Protocol error: uncounted data discarded indicates that the tag does not exist on the server. Now I use the server: :pserver:freddiefreilaender@anoncvs.opensource.apple.com:/cvs/Darwin and tags like: Apple-201 (Apple-201.42 and Apple-201.42.3 don't work). I hope this gives at least a running kernel. Special thanks again Michael. -- ---------------------------------------------------------------------------- Michael Marti Centro de Fisica dos Plasmas Instituto Superior Tecnico 1049-001 Lisboa Tel. +351 21 841 93 79 ---------------------------------------------------------------------------- _______________________________________________ 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: howto compile kernel for ktrace
by Justin C. Walker 06 Dec '02

06 Dec '02

On Friday, December 6, 2002, at 07:32 AM, Michael Marti wrote: Thank you so much Justin. I still get the same error, even if I use xnu-201-42-3 as the tag. Maybe I'm connecting to the wrong server. Can't say for sure about what server to use; the website should give you that information, though. What website do you mean? I have been looking in the apple.com webpage, but the Information comes only piecewise and I do not find the list of CVS servers... Your message showed you were trying to connect to the OpenDarwin CVS repository, so I assumed you had started with the OpenDarwin website. I don't use that CVS repository much, so I am not sure what possible servers you might run into, but I believe it has all the information you should need. One possibility: again, because I don't use OpenDarwin that much, I don't know for sure what the tags look like for the 'xnu' kernel project. Inside Apple, the tag is 'xnu-blah', which is what you would see as a result of 'uname' on a stock Mac OS X/Darwin system. For the Apple/Darwin repository, the CVS tag for the 'xnu' kernel gets changed from 'xnu-blah' to 'Apple-blah' ("blah" the same for the two related tags). I thought that since you said you could check out 'xnu-10' (which doesn't seem like a reasonable "from Apple" tag), that (a) OpenDarwin is using 'xnu' as the base for the tags; and (b) you had successfully checked out 'xnu' from an OpenDarwin repository. I would imagine that if you were connecting to the "wrong server" you would get a different complaint (no such project or no such tag). However, CVS has a sad habit of giving out totally incomprehensible and misleading error messages, so it's often not easy to figure out what is wrong. You might try the Apple/Darwin site (http://developer.apple.com/darwin) Beyond that, I don't have any good ideas. Regards, Justin -- /~\ The ASCII Justin C. Walker, Curmudgeon-at-Large \ / Ribbon Campaign X Help cure HTML Email / \ _______________________________________________ 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