January 2005 - Darwin-kernel

Re: Question on Filter scheme driver example
by Dan Markarian 18 Jan '05

18 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hi Yogesh, Dan On 15 Jan 2005, at 8:21 AM, Yogesh P wrote: Hi folks, I had gone through the Mass storage driver pdf and according to the example given in the pdf for Developing filter scheme driver write sample filter scheme driver and it works fine for the diskimage which i had created through hdiutil I have four partitons with Apple_HFS content hint which i had created during installation of MAC OS X. I want to know if i specify the "Content Hint" as Apple_HFS will my filter scheme driver work for all the four partitions other than the diskimage partition i have created using hdiutil??? Thanks in advance, YOGESH PAHILWAN (SOFTWARE ENGINNER) S P Software Technologies (I) Pvt. Ltd. (An ISO 9001:2000 Company) Pune, Maharashtra, India - 411 004. Tel : 91-20-4006154, 4006983, 4007503. Fax : 91-20-4006153. Email : pahilwan.yogesh(a)spsoftindia.com Visit us at : www.spsoftindia.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.… If you match Apple_HFS content, then produce Apple_HFS content, then you would match recursively forever. It is advisable to either change the partition type to com_spsoftindia_HFS on the specific volumes the user is interested in enabling your software on, or if it needs to be transparent to all volumes, then add a property on the media object you produce, such that you can tell when it recurses in your probe() by checking for the property, and failing the match if you see it. In this example i specify "Content Hint" property as MySoftwareCompany_MyContent in the IOKit personality for driver matching and read/write calls in my filter scheme driver works fine when i write or read on this partition. But when i specify "Content Hint" property as Apple_HFS which i think is the default content hint for the partitions created on MAC OS X and load my filter scheme driver which also has this property with the value Apple_HFS i doesn't get the read/write IOLogs on the console and sees only start and init function IOLogs repeatedly. _______________________________________________ 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/ markarian%40apple.com This email sent to markarian(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Andrew Gallatin 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Justin Walker writes: > > On Jan 17, 2005, at 7:40, Stéphane Sudre wrote: > > > Is there a documentation explaining somewhere very clearly how a 32 > > bits process is running in a 64 bits capable OS? > > > > Because I'm wondering what kind of address is ok for copyin when the > > address-space in the kernel is 64 bits. > > > > I have some problem understanding how providing my user space address > > from a 32 bits process is going to work in every case when the address > > space can be 64 bits. Unless 32 bits processes can't be executed in > > the last 32 bits of the 64 bits address space. > > Maybe it will help to realize that, for copyin/copyout, there is always > magic around to handle the mapping between address spaces. This > mapping makes sure that the address/address-space pairs are properly > translated so that the result is a "move a byte from here to there" > instruction that the computer can execute. > </rant on> This mapping magic is really only required because of the 4GB kernel address space and increases the cost of copy{in,out} tremendously for small copies. This, in turn, increases the cost of every system call. I think I remember something like a factor of 6x or 8x difference between linux ppc64 (genuine 64-bit kernel) and MacOSX (32-bit kernel) on 64-bit G5 hardware the last time I measured our driver's ioctl timings. (linux being faster, of course). In more traditional 1G/3G split 32-bit or pure 64-bit kernel schemes, user virtual addresses are directly accessible in kernel mode, and copy{in,out} is implemented as bcopy plus a fault handler for user virtual addresses which could have been paged out. This is much, much faster than remapping addresses. This performance cost is the main reason that I'm so disappointed that the kernel will remain 32-bits. Then there is the fact that a lot of our driver code assumes a kernel uintptr_t / void * is large enough to hold a user pointer which causes a lot more porting problems for us than a pure 64-bit kernel would have. But then again, we already run in a pure 64-bit kernel Solaris, Linux, Tru64, FreeBSD, so pure 64-bits (with support for legacy 32-bit apps) is no big deal for us. Sorry for tilting at windmills.. </rant off> Drew _______________________________________________ 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

Etherspoofing patch
by Yoel UK 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=BwIfR6RwpxjxYuay5VDuGH1f7pOUUxcEl6zhn8bpBGAws2QOc5y4Zrl2KRZmZUbnmybuCGGmPfQXA1h0GJ65vP2FMezQd3aVf/tR40CI8d3lk0fz0MJY8w3z7JEwqdcVw4WCQHAh8eUrL2D5Fws/pdktHPdEBMt2LVPIIPhJPSA= Hi there, Does anybody can tell me why is it that after applying the Jeff Nathan patch (BIOCSHDRCMPLT-10.3.3.patch) that it meant to allow me to spoof the mac address of all ether devices I can't change the mac address of my airport extreme? Compilation, patch application and booting the new kernel went all fine. Any thoughts? _______________________________________________ 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: 64 bits Kernel and 32 bits process
by Stephane Sudre 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Le 17 janv. 05, à 17:49, Shawn Erickson a écrit : On Jan 17, 2005, at 8:17 AM, Shawn Erickson wrote: Does this help a little more? Anyway gonna rehash things using my own words... Ok. Thanks for both latest answers from Justin and you. _______________________________________________ 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.… For info on how Mac OS X supports greater then 4GB of physical RAM while still maintaining a 32b kernel, etc... <http://developer.apple.com/technotes/tn/tn2090.html> This is the one I read and it didn't help. <http://developer.apple.com/documentation/DeviceDrivers/Conceptual/ IOKitFundamentals/DataMgmt/chapter_9_section_2.html#//apple_ref/doc/ uid/TP0000019/TPXREF108> On current Mac OS X both user space and kernel space work with virtual addresses that are 32b in size. Going forward the kernel will continue to work with 32b virtual addresses (user space will gain the _option_ for 64b address spaces). Various method exist that allow you share / map memory between the kernel memory space and the various instances user memory space. It depends on what type of driver and in which part of the kernel you are working but assuming IOKit you would use IOMemoryDescriptor and/or friends to manage this for you. Now the VM system on Mac OS X has the ability to address physical memory that is larger then 32b (not sure if it truly support 64b or not of physical addressing, could be 34b, 48b, etc.) when running on systems that support more then 4GB of RAM (PowerMac / Xserve G5s at this time). This capability however is basically transparent and hidden from you even when in the kernel. In other words you never (should never need to) see the true physical address but instead you work with intermediate physical addresses (I/O address as it is called) that gets mapped by hardware (DART) as needed to the true physical address. In other words don't worry about it much just make sure to use the supported memory mapping methods when you do your DMA, user/kernel mapping, etc. appropriate for the type of driver you are working on. Those methods will take care of what is needed for you. This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Shawn Erickson 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 17, 2005, at 8:17 AM, Shawn Erickson wrote: Does this help a little more? Anyway gonna rehash things using my own words... -Shawn _______________________________________________ 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.… For info on how Mac OS X supports greater then 4GB of physical RAM while still maintaining a 32b kernel, etc... <http://developer.apple.com/technotes/tn/tn2090.html> This is the one I read and it didn't help. <http://developer.apple.com/documentation/DeviceDrivers/Conceptual/ IOKitFundamentals/DataMgmt/chapter_9_section_2.html#//apple_ref/doc/ uid/TP0000019/TPXREF108> On current Mac OS X both user space and kernel space work with virtual addresses that are 32b in size. Going forward the kernel will continue to work with 32b virtual addresses (user space will gain the _option_ for 64b address spaces). Various method exist that allow you share / map memory between the kernel memory space and the various instances user memory space. It depends on what type of driver and in which part of the kernel you are working but assuming IOKit you would use IOMemoryDescriptor and/or friends to manage this for you. Now the VM system on Mac OS X has the ability to address physical memory that is larger then 32b (not sure if it truly support 64b or not of physical addressing, could be 34b, 48b, etc.) when running on systems that support more then 4GB of RAM (PowerMac / Xserve G5s at this time). This capability however is basically transparent and hidden from you even when in the kernel. In other words you never (should never need to) see the true physical address but instead you work with intermediate physical addresses (I/O address as it is called) that gets mapped by hardware (DART) as needed to the true physical address. In other words don't worry about it much just make sure to use the supported memory mapping methods when you do your DMA, user/kernel mapping, etc. appropriate for the type of driver you are working on. Those methods will take care of what is needed for you. This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Justin Walker 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 17, 2005, at 7:40, Stéphane Sudre wrote: Cheers, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Some people have a mental | horizon of radius zero, and | call it their point of view. | -- David Hilbert *--------------------------------------*-------------------------------* _______________________________________________ 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.… Is there a documentation explaining somewhere very clearly how a 32 bits process is running in a 64 bits capable OS? Because I'm wondering what kind of address is ok for copyin when the address-space in the kernel is 64 bits. I have some problem understanding how providing my user space address from a 32 bits process is going to work in every case when the address space can be 64 bits. Unless 32 bits processes can't be executed in the last 32 bits of the 64 bits address space. Maybe it will help to realize that, for copyin/copyout, there is always magic around to handle the mapping between address spaces. This mapping makes sure that the address/address-space pairs are properly translated so that the result is a "move a byte from here to there" instruction that the computer can execute. Generally, the magic takes the form of mapping registers in an MMU. Keep in mind that, even in yesterday's G3 and G4 processors, this same magic has to occur to ensure that the virtual addresses seen, separately, by user and kernel instructions get translated to physical addresses (like user space, the kernel address space is mapped). For more details, you should look at a CPU manual (available, I think, on IBM's website). As Shawn has tried to point out, the plan for Mac OS X is to stay 32-bits, while user spaces are (selectively) widened to 64 bits, so in fact, your question should be inverted. This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Shawn Erickson 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 17, 2005, at 8:10 AM, Stéphane Sudre wrote: On Jan 17, 2005, at 5:06 PM, Shawn Erickson wrote: On Jan 17, 2005, at 7:40 AM, Stéphane Sudre wrote: Hint and aspirin welcome. Can you better explain then what you are having problems with? -Shawn _______________________________________________ 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.… Is there a documentation explaining somewhere very clearly how a 32 bits process is running in a 64 bits capable OS? Because I'm wondering what kind of address is ok for copyin when the address-space in the kernel is 64 bits. I have some problem understanding how providing my user space address from a 32 bits process is going to work in every case when the address space can be 64 bits. Unless 32 bits processes can't be executed in the last 32 bits of the 64 bits address space. In current Mac OS X (aka Darwin) neither kernel space nor user space supports 64b address spaces. Additionally the stated intention for Mac OS X going forward is to always have one kernel version for Mac OS X which runs in a 32b address space while adding the ability to run processes in either a 32b or 64b address space. So on current Mac OS X you are always working with 32b addresses (unless you are some how hacking into the VM system which has the ability to map physical memory greater then 32b allows). For info on how Mac OS X supports greater then 4GB of physical RAM while still maintaining a 32b kernel, etc... <http://developer.apple.com/technotes/tn/tn2090.html> Basic info about future 64b addressing support in Mac OS X... This is the one I read and it didn't help. In the above you are asking about a 64b kernel space which does not exist on Mac OS X, so with that in mind are you still having problems? This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Stéphane Sudre 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 17, 2005, at 5:06 PM, Shawn Erickson wrote: On Jan 17, 2005, at 7:40 AM, Stéphane Sudre wrote: Hint and aspirin welcome. This is the one I read and it didn't help. _______________________________________________ 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.… Is there a documentation explaining somewhere very clearly how a 32 bits process is running in a 64 bits capable OS? Because I'm wondering what kind of address is ok for copyin when the address-space in the kernel is 64 bits. I have some problem understanding how providing my user space address from a 32 bits process is going to work in every case when the address space can be 64 bits. Unless 32 bits processes can't be executed in the last 32 bits of the 64 bits address space. In current Mac OS X (aka Darwin) neither kernel space nor user space supports 64b address spaces. Additionally the stated intention for Mac OS X going forward is to always have one kernel version for Mac OS X which runs in a 32b address space while adding the ability to run processes in either a 32b or 64b address space. So on current Mac OS X you are always working with 32b addresses (unless you are some how hacking into the VM system which has the ability to map physical memory greater then 32b allows). For info on how Mac OS X supports greater then 4GB of physical RAM while still maintaining a 32b kernel, etc... <http://developer.apple.com/technotes/tn/tn2090.html> Basic info about future 64b addressing support in Mac OS X... This email sent to site_archiver(a)lists.apple.com

1 0

Re: 64 bits Kernel and 32 bits process
by Shawn Erickson 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 17, 2005, at 7:40 AM, Stéphane Sudre wrote: Basic info about future 64b addressing support in Mac OS X... <http://developer.apple.com/macosx/tiger/64bit.html> -Shawn _______________________________________________ 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.… Is there a documentation explaining somewhere very clearly how a 32 bits process is running in a 64 bits capable OS? Because I'm wondering what kind of address is ok for copyin when the address-space in the kernel is 64 bits. I have some problem understanding how providing my user space address from a 32 bits process is going to work in every case when the address space can be 64 bits. Unless 32 bits processes can't be executed in the last 32 bits of the 64 bits address space. Hint and aspirin welcome. In current Mac OS X (aka Darwin) neither kernel space nor user space supports 64b address spaces. Additionally the stated intention for Mac OS X going forward is to always have one kernel version for Mac OS X which runs in a 32b address space while adding the ability to run processes in either a 32b or 64b address space. So on current Mac OS X you are always working with 32b addresses (unless you are some how hacking into the VM system which has the ability to map physical memory greater then 32b allows). For info on how Mac OS X supports greater then 4GB of physical RAM while still maintaining a 32b kernel, etc... <http://developer.apple.com/technotes/tn/tn2090.html> This email sent to site_archiver(a)lists.apple.com

1 0

64 bits Kernel and 32 bits process
by Stéphane Sudre 17 Jan '05

17 Jan '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hint and aspirin welcome. _______________________________________________ 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.… Is there a documentation explaining somewhere very clearly how a 32 bits process is running in a 64 bits capable OS? Because I'm wondering what kind of address is ok for copyin when the address-space in the kernel is 64 bits. I have some problem understanding how providing my user space address from a 32 bits process is going to work in every case when the address space can be 64 bits. Unless 32 bits processes can't be executed in the last 32 bits of the 64 bits address space. This email sent to site_archiver(a)lists.apple.com

1 0