September 2007 - Darwin-kernel

File I/O from within a Filesystem Driver
by Ron Aldrich 14 Sep '07

14 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Folks, Thanks, Ron Aldrich Software Architects, Inc. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I'm currently adding Multi-session writing to our UDF filesystem driver, and need to be able to read and write temporary files on the root drive while processing I/Os to the UDF volume. My UDF filesystem currently resides entirely in Kernel space, so I'm wondering what kinds of problems I can expect to encounter when accessing those temporary files, and whether or not I will need to perform those I/Os in user space. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by Régis Duchesne 13 Sep '07

13 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Icedove 1.5.0.9 (X11/20061220) Terry Lambert wrote: > This is probably about a 70 line program, if you include signal handling > and EOF processing, > > Then you can use the slave sides as if they were com ports from both ends. > > If you need pseudo CTS/RTS/DTR/DSR/BRK signal processing, you will need > to get fancier, It is quite tricky to implement proper virtual serial port over PTYs. I did it for VMware. The transmission of UART break state (used by lots of debuggers) and errors is non-trivial: you must program the PTY so that it puts these events as escaped codes in the data stream. Then you need a state machine to unescape them. Also, a communication between 2 real UARTs is lossy: if one UART sends 16 (UART FIFO size) + 1 bytes, and nobody listen at the receiving side, 1 byte will be lost by the receiving side. PTYs are (unfortunately in this case) lossless (just like pipe), and thus the 70 line program you are talking about must simulate this ability to "drop" bytes. Debuggers rely on that too, especially when trying to connect to their target. And I'm not even talking about flow-control yet :) > The interesting driver (which has to be done in kernel space) is a > device driver that looks like a serial port to a vmware client, and > looks like a serial port to the host OS on the other end. That lets > you do serial gdb from the host OS to your victim (Julian Elischer > wrote one of these for FreeBSD at one point in time). Effectively, > that gets you two machine debugging on one machine. If you don't feel like writing this driver, you can use 2 VMware VMs (one runs the debugger, one runs the target). -- hpreg _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by Terry Lambert 12 Sep '07

12 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com -- Terry On 11.09.2007, at 22:35, Thomas Hauk wrote: Does a similar software package to com0com exist for Mac OS X? Cheers, T Andreas Fink Fink Consulting GmbH Global Networks Schweiz AG BebbiCell AG _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… Yes, for certain classes of the problem, you can use the slave e course), but the code to glue them together can live in user space. The problems where this works all assume that the seral port acts like it has a null-modem cable hooking it to another serial port (i.e. all out of band signals over the wire are meaningless). The interesting driver (which has to be done in kernel space) is a device driver that looks like a serial port to a vmware client, and looks like a serial port to the host OS on the other end. That lets you do serial gdb from the host OS to your victim (Julian Elischer wrote one of these for FreeBSD at one point in time). Effectively, that gets you two machine debugging on one machine. You could use something like this on Mac OS X to debug say Linux or FreeBSD clients from a Mac OS X host. On Sep 12, 2007, at 1:06 PM, Andreas Fink wrote: This sounds like something which would have been useful in the past to me many times. Are you saying this can be done completely in userspace? I'm doing cross-platform development of a product that accesses custom hardware connected to the USB port, which acts like a serial device. On Windows I can send and receive data to the COM port associated with the device. For testing, I downloaded the open source com0com project, which creates a pair of virtual serial ports which are connected to each other; the output for one is the input for the other. This allows me to test my software against a "virtual device" program I have written. In this sense, I guess it would create a pair of psuedo-ttys that are connected to each other in the background. I read through the archives of this list and found a similar question a while back, but the question wasn't phrased clearly and it isn't the same question as I am posing above, so that's why I'm writing now. If an existing solution doesn't exist, I suppose I could write my own. Any thoughts on the complexity of such a task? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by Andreas Fink 12 Sep '07

12 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Yielding the processor in a kext?
by Jeremy Pereira 12 Sep '07

12 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On 11 Sep 2007, at 12:53, Anton Altaparmakov wrote: _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I think he means the design of the code that implements your file system, not the file system design itself. Your code should be designed in such a way that a memory allocation error caused by a memory shortage should not corrupt it. You try and put a relational database in the kernel as file system and put in lots of metadata duplication and cross references all over the place and then try and keep it both fast in the 99.9999% and 100% correct in the 0.0001% where a memory shortage occurs without journalling, COW, and other modern approaches to fs consistency and then you will be in my boat and see how retrying memory allocations suddenly seems like a great idea... (-; Funnily enough, the VFS I am currently porting from Solaris to Mac OSX also has a database in it, although it is not relational. The solution adopted by the original developers and one which I believe Apple recommends was to put all of the complicated stuff in a user space daemon. The kext is really just a stub that services the VFS and vnode API and hands off the calls to the user space program. The only memory allocations in kernel are those for the vfs and vnode file system specific information. So the answer is: no I won't try putting a relational database in the kernel. It sounds like a world of pain to me. I think I'll accept the performance hit and do the complicated stuff in user space. It means I can unit test most of the VFS without two machines and I have access to all of the user space libraries and a bug in the code is less likely to cause a panic. This email sent to site_archiver(a)lists.apple.com

1 0

inode64 support
by Nick Blievers 12 Sep '07

12 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Thunderbird 2.0.0.0 (Macintosh/20070326) Hi, Note, this is relating to Tiger, not Leopard. Thanks Nick _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I am curious what, if any, problems people are having using inode64 on darwin? Since va_fileid is 64bit it seems like the only issue will be user space app's which use stat's st_ino field. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by Terry Lambert 11 Sep '07

11 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com you should make a quick search on Darwin-Driver mailing list it's a while(1) question How virtual do you need it? -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… On Sep 11, 2007, at 3:09 PM, Thomas Hauk wrote: On Sep 11, 2007, at 1:42 PM, mm w wrote: Now that the Apple mailing list search facility is (mostly) working again (was getting Interal Server Errors for over an hour), I did what you suggested, but I didn't find anything related, especially regarding anything similar to the Win32 com0com project. It could be that my search-fu is weak, though. Any further assistance would be most appreciated. If they should both think they are talking to serial ports, then you can write a little program that opens the master side of two ptys, calls grantpt() on both and unlockpt() on both, uses fcntl() to put them in non-blocking mode, then calls select() on both, and when they become active, writes the data from one end to the other and vice versa. This is probably about a 70 line program, if you include signal handling and EOF processing, and make sure you include all the right headers for making the compiler happy when -Wall -Werror is specified, and put some comments in the thing. Then you can use the slave sides as if they were com ports from both ends. If you need pseudo CTS/RTS/DTR/DSR/BRK signal processing, you will need to get fancier, and if you want both ends to look like a dataset (e.g. a Hayes modem or other device, and accept AT commands, etc.), then you will need to get more complicated. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by Thomas Hauk 11 Sep '07

11 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com you should make a quick search on Darwin-Driver mailing list it's a while(1) question Any further assistance would be most appreciated. T _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… On Sep 11, 2007, at 1:42 PM, mm w wrote: Now that the Apple mailing list search facility is (mostly) working again (was getting Interal Server Errors for over an hour), I did what you suggested, but I didn't find anything related, especially regarding anything similar to the Win32 com0com project. It could be that my search-fu is weak, though. -- Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. --Jamie Zawinski, in comp.lang.emacs This email sent to site_archiver(a)lists.apple.com

1 0

Re: Virtual serial ports
by mm w 11 Sep '07

11 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=N+LTuYj4X/xAhXBUhMfE7+lncUZ/coM2qi15Cm3Oh1A=; b=eeS3hrudqScCSGuG2XZnqzH8ssczsxVc+v9FEoLs5hdJPQXOxoMTt/d/8YrXizPLHyTGX52cVb0Dq9XJJASBOG2hzH1Vl5znfcBcj0OzaEJKHaum9/zhuHYLHI0cCHjEV2wZg4lDRfNiGi67fU9rji2FlAtFuNDLojLOTiopzBs= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=q3FiNsHIxf+kQRV9DLq97DMZMzM/6bG3hJoPTKtSUG+Gzt2hHHnOtKGVsLsKY9YR0oO/uTuJeg08TJXbfyUt7lLoPNfJpMWoZZW4Y1gtcS5UhVhZGbRNIovtfFUzPeKFs6MgP2do4PapetEV9ufNPaxxb7sIggcZTjK5eDZorVA= Hi Thomas, you should make a quick search on Darwin-Driver mailing list it's a while(1) question Cheers! -mmw On 9/11/07, Thomas Hauk <thauk(a)novuscom.net> wrote: > I'm doing cross-platform development of a product that accesses > custom hardware connected to the USB port, which acts like a serial > device. > > On Windows I can send and receive data to the COM port associated > with the device. For testing, I downloaded the open source com0com > project, which creates a pair of virtual serial ports which are > connected to each other; the output for one is the input for the > other. This allows me to test my software against a "virtual device" > program I have written. > > Does a similar software package to com0com exist for Mac OS X? > > In this sense, I guess it would create a pair of psuedo-ttys that are > connected to each other in the background. > > I read through the archives of this list and found a similar question > a while back, but the question wasn't phrased clearly and it isn't > the same question as I am posing above, so that's why I'm writing now. > > If an existing solution doesn't exist, I suppose I could write my > own. Any thoughts on the complexity of such a task? > > Cheers, > > T > > -- > Some people, when confronted with a problem, think "I know, I'll use > regular expressions." Now they have two problems. > --Jamie Zawinski, in comp.lang.emacs > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/darwin-kernel/openspecies%40gmail.com > > This email sent to openspecies(a)gmail.com > -- http://openspecies.googlepages.com/gpgkey.txt _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Virtual serial ports
by Thomas Hauk 11 Sep '07

11 Sep '07

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Does a similar software package to com0com exist for Mac OS X? Cheers, T _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I'm doing cross-platform development of a product that accesses custom hardware connected to the USB port, which acts like a serial device. On Windows I can send and receive data to the COM port associated with the device. For testing, I downloaded the open source com0com project, which creates a pair of virtual serial ports which are connected to each other; the output for one is the input for the other. This allows me to test my software against a "virtual device" program I have written. In this sense, I guess it would create a pair of psuedo-ttys that are connected to each other in the background. I read through the archives of this list and found a similar question a while back, but the question wasn't phrased clearly and it isn't the same question as I am posing above, so that's why I'm writing now. If an existing solution doesn't exist, I suppose I could write my own. Any thoughts on the complexity of such a task? -- Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. --Jamie Zawinski, in comp.lang.emacs This email sent to site_archiver(a)lists.apple.com

1 0