April 2003 - Darwin-kernel - lists.apple.com

Re: [OT] boot in console mode (verbose mode)
by Jim Riggs 25 Apr '03

25 Apr '03

sudo nvram boot-args=-v This changes the setting in NVRAM. - Jim I haven't actually tried this, but it looks like you can force it to boot verbosely by editing your /etc/rc and changing the line: VerboseFlag="" to: VerboseFlag="-v" -- Adam Atlas On Friday, April 25, 2003, at 08:20 AM, Xavier Santolaria wrote: Hi folks; if this is off-topic please excuse me. All I wanted to know is if there was something I can change to make my machine boot into the verbose mode, but _witout_ pressing the 'v' key? I know I can use the console mode at the login prompt by entering >console as username. But I want it at boot time. Any ideas? Thanks in advance. - xavier. _______________________________________________ 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: [OT] boot in console mode (verbose mode)
by Quinn 25 Apr '03

25 Apr '03

At 14:20 +0200 25/4/03, Xavier Santolaria wrote: All I wanted to know is if there was something I can change to make my machine boot into the verbose mode, but _witout_ pressing the 'v' key? You should set the Open Firmware boot-args variable to include the "-v" flag. The technique for setting an OF variable is describe in "Inside Mac OS X: Kernel Programming". <http://developer.apple.com/techpubs/macosx/Darwin/General/KernelProgramming…> 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: [OT] boot in console mode (verbose mode)
by Adam Atlas 25 Apr '03

25 Apr '03

I haven't actually tried this, but it looks like you can force it to boot verbosely by editing your /etc/rc and changing the line: VerboseFlag="" to: VerboseFlag="-v" -- Adam Atlas On Friday, April 25, 2003, at 08:20 AM, Xavier Santolaria wrote: Hi folks; if this is off-topic please excuse me. All I wanted to know is if there was something I can change to make my machine boot into the verbose mode, but _witout_ pressing the 'v' key? I know I can use the console mode at the login prompt by entering >console as username. But I want it at boot time. Any ideas? Thanks in advance. - xavier. _______________________________________________ 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

[OT] boot in console mode (verbose mode)
by Xavier Santolaria 25 Apr '03

25 Apr '03

Hi folks; if this is off-topic please excuse me. All I wanted to know is if there was something I can change to make my machine boot into the verbose mode, but _witout_ pressing the 'v' key? I know I can use the console mode at the login prompt by entering >console as username. But I want it at boot time. Any ideas? Thanks in advance. - xavier. [demime 0.98b removed an attachment of type application/pgp-signature] _______________________________________________ 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: vfs sample source
by Quinn 24 Apr '03

24 Apr '03

At 9:03 -0700 24/4/03, Chris Rudolph wrote: I've heard that there is some sample code for writing a vfs for macosx. I know I could just look at some of the vfs bits in the darwin sources, but trying to weed through everything is quite daunting. Does anyone have any suggestions? Take a look at the following Q&A. <http://developer.apple.com/qa/qa2001/qa1242.html> This encapsulates my VFS advice as of a few weeks back. 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

vfs sample source
by Chris Rudolph 24 Apr '03

24 Apr '03

Hi all, I've heard that there is some sample code for writing a vfs for macosx. I know I could just look at some of the vfs bits in the darwin sources, but trying to weed through everything is quite daunting. Does anyone have any suggestions? Thanks, Rudy. -- Chris Rudolph (Rudy) rudy(a)apple.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

Re: frxThread questions
by Godfrey van der Linden 23 Apr '03

23 Apr '03

G'day Vivek, The place to look is <sys/proc.h>. You won't really find documentation for the except in a tradition unix book. I suggest you have a look at 'The Design and Implementation of the 4.4 BSD Operating System', which discusses this in detail. Briefly though it seems like the usage on exit is incorrect as I'm nulling a pointer then passing the address of it as an argument. Infact the standard Unix sleep()/wakeup() mechanism never dereferences the 'event' argument. It uses the event as a tag to indicate which thread that is sleeping needs to be woken up. Thus this &ftxThread is used to indicate which client tsleep() to wakeup. It is traditional to use the address of the variable for which you wish to wait. This allows us to be reasonably sure that nobody else is using this 'event' without having to provide a data base of all possible events in a header somewhere. Godfrey At 18:47 -0700 03-4-22, Vivek wrote: Hi. I'm looking at IOSerialBSDClient, and there's some code that I can't figure out. Inside IOSerialBSDClient::txFunc, there are lines as follows: // on entry ftxThread = IOThreadSelf(); ftxThreadLaunched = true; wakeup((caddr_t) &ftxThread); // wakeup the thread launcher ... //on exit ftxThread = NULL; ftxThreadLaunched = false; wakeup((caddr_t) &ftxThread); // wakeup the thread killer ftxThread and ftxThreadLaunched are members of IOSerialBSDClient. I can't seem to find a likely candidate for wakeup() in Kernel.framework or /usr/include, and it seems like the exit code, with its null-pointer argument, may have some pretty bad results. Can somebody explain what the intent is of this sequence? Thanks, Vivek _______________________________________________ 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

frxThread questions
by Vivek 23 Apr '03

23 Apr '03

Hi. I'm looking at IOSerialBSDClient, and there's some code that I can't figure out. Inside IOSerialBSDClient::txFunc, there are lines as follows: // on entry ftxThread = IOThreadSelf(); ftxThreadLaunched = true; wakeup((caddr_t) &ftxThread); // wakeup the thread launcher ... //on exit ftxThread = NULL; ftxThreadLaunched = false; wakeup((caddr_t) &ftxThread); // wakeup the thread killer ftxThread and ftxThreadLaunched are members of IOSerialBSDClient. I can't seem to find a likely candidate for wakeup() in Kernel.framework or /usr/include, and it seems like the exit code, with its null-pointer argument, may have some pretty bad results. Can somebody explain what the intent is of this sequence? Thanks, Vivek _______________________________________________ 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: Is operator new safe?
by chuck remes 22 Apr '03

22 Apr '03

On Monday, April 21, 2003, at 09:57 PM, raldrich(a)mac.com wrote: Is it safe to use the c++ operator new in a kernel extension? I'm porting a file system library written in c++ to MacOS X, and I'm getting a kernel panic when it calls new after performing a series of I/Os, so I'm trying to determine whether the call to new is the problem, or if I trashed memory when performing my I/O. Take a look at page 64 of I/OKit Fundamentals (available from Apple at http://developer.apple.com/techpubs/macosx/Darwin/General/ IOKitFundamentals/IOKitFundamentals.pdf ). You may use new on any C++ object derived from OSObject, but there are certain rules to follow as explained in that doc. The kernel runtime environment only allows a very strict subset of C++, so exceptions and RTTI (amongst other things) are not available which can make the behavior of "new" a bit different from a userland application. If this doesn't help, email me off the list and I'll try and explain further. cr _______________________________________________ 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: Is operator new safe?
by William Kucharski 22 Apr '03

22 Apr '03

>Is it safe to use the c++ operator new in a kernel extension? > >I'm porting a file system library written in c++ to MacOS X, and I'm >getting a kernel panic when it calls new after performing a series of >I/Os, so I'm trying to determine whether the call to new is the >problem, or if I trashed memory when performing my I/O. The IOKIT kexts use "new" all over the place without any issues. Just make sure to check the value you get back to see if the "new" returned a valid pointer. If you're already doing that and still panic then something is corrupting your pointers somehow. William Kucharski kucharsk(a)mac.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