September 2003 - Darwin-kernel

Re: DMA questions (invalidate_dcache and flush_dcache)
by Benjamin Herrenschmidt 12 Sep '03

12 Sep '03

On Monday, September 8, 2003, at 10:19 PM, Ryan Rempel wrote: I'm updating a driver for the new memory mapping limitations, and I've got a couple of DMA-related questions. The driver uses a couple of functions in <osfmk/ppc/pmap.h>, namely: extern void invalidate_dcache(vm_offset_t va, unsigned length, boolean_t phys); extern void flush_dcache(vm_offset_t va, unsigned length, boolean_t phys); The calls to invalidate_dcache are accompanied by this comment "Purge cache references for the DBDMA entry we are about to look at". The calls to flush_dcache are accompanied by this comnment "Push DMA commands to physical memory". Do you actually need to do that at all ? What driver is this ? Those machines are supposed to be cache coherent... I know there seem to be cache coherency bugs with earlier hardware, though I couldn't get any useful detail from Apple, I had some weird problems when a DBDMA wrote in a middle of a cache line with the MESH driver on a 8500 and possibly others. So if you are trying to deal with such issues, you need to flush/invalidate your data as well and make sure everything is cache line aligned. Ben. _______________________________________________ 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: Need help debugging load failure of OSX SCSI device driver stub
by Jeff Collins 12 Sep '03

12 Sep '03

Resubmitting and cross posting to darwin-kernel as no one has answered yet. Thank you for any help. Jeff When I try to load a device driver stub, I get some warnings I do not understand. Are these "is not valid" warnings important? bash-2.05a# kextload -tv ods.kext kextload: kext /System/Library/Extensions/AppleADBButtons.kext is not valid kextload: kext /System/Library/Extensions/AppleLMUController.kext is not valid kextload: kext /System/Library/Extensions/AppleOnboardDisplay.kext is not valid kextload: kext /System/Library/Extensions/AppleUSBDisplays.kext is not valid kextload: kext /System/Library/Extensions/InternalModemSupport.kext is not valid kextload: kext /System/Library/Extensions/IOATAPIProtocolTransport.kext is not valid kextload: kext /System/Library/Extensions/IOPCCardFamily.kext is not valid kextload: kext /System/Library/Extensions/NVDANV10Hal.kext is not valid kextload: kext /System/Library/Extensions/NVDANV20Hal.kext is not valid kextload: kext /System/Library/Extensions/NVDAResman.kext is not valid kextload: kext /System/Library/Extensions/UltraTek66.kext is not valid kextload: kext /System/Library/Extensions/webdav_fs.kext is not valid kextload: resolving dependencies for kernel extensions with validation and authentication failures kernel extension ods.kextMissing dependencies { "com.apple.kernel.mach" = "A valid compatible version of this dependency cannot be found" "com.apple.kernel.iokit" = "A valid compatible version of this dependency cannot be found" "com.apple.kernel.libkern" = "A valid compatible version of this dependency cannot be found" } How can I solve the problem of dependency on the proper version of com.apple.kernel.mach, for example? My project.pbxproj file has the proper version of com.apple.kernel.mach (6.6) <!DOCTYPE plist PUBLIC \"-//Apple Computer//DTD PLIST 1.0//EN\" \"http://www.apple.com/DTDs/PropertyList-1.0. dtd\"> <plist version=\"1.0\"> <dict> <key>CFBundleDevelopmentRegion</key> <string>English</string> <key>CFBundleExecutable</key> <string>ods</string> <key>CFBundleIconFile</key> <string></string> <key>CFBundleIdentifier</key> <string>ods</string> <key>CFBundleInfoDictionaryVersion</key> <string>6.6</string> <key>CFBundlePackageType</key> <string>KEXT</string> <key>CFBundleSignature</key> <string>????</string> <key>CFBundleVersion</key> <string>1.0.0d1</string> <key>IOKITPersonalities</key> <dict> <key>ods</key> <dict> <key>CFBundleIdentifier</key> <string>ods</string> <key>IOClass</key> <string>odsSCSIController</string> <key>IOPCIMatch</key> <string>0x000b1000</string> </dict> </dict> <key>OSBundleLibraries</key> <dict> <key>com.apple.iokit.IOPCIFamily</key> <string>1.2</string> <key>com.apple.iokit.IOSCSIParallelFamily</key> <string>1.2.4</string> <key>com.apple.kernel.iokit</key> <string>6.6</string> <key>com.apple.kernel.libkern</key> <string>6.6</string> <key>com.apple.kernel.mach</key> <string>6.6</string> </dict> </dict> </plist>bash-2.05a And a run time the proper version of the kernel and libkern appear loaded: # kextstat | grep -i mach 5 23 0x0 0x0 0x0 com.apple.kernel.mach (6.6) bash-2.05a# kextstat | grep -i libkern 4 33 0x0 0x0 0x0 com.apple.kernel.libkern (6.6) Thanks for any help, Jeff Collins _______________________________________________ darwin-drivers mailing list | darwin-drivers(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-drivers 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: DMA questions (invalidate_dcache and flush_dcache)
by Ryan Rempel 12 Sep '03

12 Sep '03

On Wednesday, September 10, 2003, at 09:11 AM, Sean McBride wrote: Ryan Rempel (ryan.rempel(a)utoronto.ca) wrote on Mon, 8 Sep 2003 15:19:42 -0500: The driver uses a couple of functions in <osfmk/ppc/pmap.h>, namely: <http://developer.apple.com/technotes/tn/tn2090.html> seems to suggest that using that pmap.h stuff is deprecated. Well, I'm not entirely sure that it's *all* deprecated. What TN 2090 says clearly is that physical addresses obtained through pmap.h routines aren't useful anymore. But invalidate_dcache and flush_dcache take virtual addresses as arguments, not physical addresses (i.e they take va's, not pa's). So the inputs ought to be valid still. Though I do wonder whether you're right--that's why I asked the question. _______________________________________________ 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: ipcs (?)
by Paul Ripke 10 Sep '03

10 Sep '03

On Thursday, Sep 11, 2003, at 06:47 Australia/Sydney, W. R. Wing wrote: I'm new (as of today) on this list, so please forgive me if this isn't quite the right place to ask this question. A friend of mine here is trying to port some stuff from SUN Solaris, and stuck his head in my office this afternoon to ask: "Why don't I see ipcs on OS X?" I checked and he seems to be right, no binary, no man page, no nothing. I turned to the X-Tech list for help, and was pointed here by someone who commented that since all inter-process communication is handled by the kernel in OS X/Darwin, this was the place to ask. Google for "ipcs darwin" and you should find my Darwinized ports of FreeBSD ipcs/ipcrm. BTW: Under Darwin, other interfaces are preferred over SYSV shared memory (mmap(2)) and semaphores (POSIX). There are also severe limits on how much SYSV SHM can be allocated. Cheers, -- Paul Ripke Unix/OpenVMS/TSM/DBA I love deadlines. I like the whooshing sound they make as they fly by. -- Douglas Adams _______________________________________________ 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

ipcs (?)
by W. R. Wing 10 Sep '03

10 Sep '03

I'm new (as of today) on this list, so please forgive me if this isn't quite the right place to ask this question. A friend of mine here is trying to port some stuff from SUN Solaris, and stuck his head in my office this afternoon to ask: "Why don't I see ipcs on OS X?" I checked and he seems to be right, no binary, no man page, no nothing. I turned to the X-Tech list for help, and was pointed here by someone who commented that since all inter-process communication is handled by the kernel in OS X/Darwin, this was the place to ask. Thanks in advance, Bill -- _ /~\ The ASCII | William R. Wing, PhD. wrw(a)ornl.gov 865-574-8839 \ / Ribbon Campaign | Network Architect, Oak Ridge National Laboratory X Against HTML | Network Research Group, Computer Sci & Math Div. / \ 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

Re: Custom logging user client tool from GMAC?
by Garth Cummings 10 Sep '03

10 Sep '03

Hello Michael, On Wednesday, September 10, 2003, at 06:51 AM, Michael Cashwell wrote: That's fine but I need the matching out-of-kernel code. Apple's docs refer to a user-land tool (a CLI app I'd expect) that connects to this user client subclass and pulls the log info from it. I do not see this code anywhere in the AppleGMACEthernet project. I can see where this would be frustrating. Could you please file a bug report against the documentation so our Tech Pubs team can look into this? The URL for filing bug reports is <http://developer.apple.com/bugreporter>. Thanks much, --gc _______________________________________________ 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: DMA questions (invalidate_dcache and flush_dcache)
by Sean McBride 10 Sep '03

10 Sep '03

Ryan Rempel (ryan.rempel(a)utoronto.ca) wrote on Mon, 8 Sep 2003 15:19:42 -0500: >The driver uses a couple of functions in <osfmk/ppc/pmap.h>, namely: <http://developer.apple.com/technotes/tn/tn2090.html> seems to suggest that using that pmap.h stuff is deprecated. -- ____________________________________________________________ Sean McBride, B. Eng Sean.McBride(a)matrox.com Mac Software Designer +1-514-822-6000 x7399 Matrox Electronic Systems Ltd. Montrial, Quibec, Canada _______________________________________________ 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: Custom logging user client tool from GMAC?
by Michael Cashwell 10 Sep '03

10 Sep '03

On Sep 10, 2003, at 6:53 AM, chuck remes wrote: > Take a look at the sample code from DTS, for starters. > > http://www.devworld.apple.com/samplecode/Sample_Code/ > Devices_and_Hardware/IOKit/SimpleUserClient.htm Yes, I have been and am working with this. I just hoped I could find the actual source for the userland tool that Apple "Writing Device Drivers" doc clearly used and claims are available. > Then you can find the code for GMAC's usertool at: > > http://www.opendarwin.org/cgi-bin/cvsweb.cgi/src/AppleGMACEthernet/ > > It's embedded into the source file UniNEnet.cpp at the bottom of the > file. The class definitions are in the associated header file. I must be confused here. I checked the specified file at opendarwin.org and in Apple's tar ball for 10.2.6 and in both cases this file ends with the method "UniNEnetUserClient::writeMII". This class is necessary but not sufficient. Unless I'm completely lost, the class is the in-kernel portion of the user client. That's fine but I need the matching out-of-kernel code. Apple's docs refer to a user-land tool (a CLI app I'd expect) that connects to this user client subclass and pulls the log info from it. I do not see this code anywhere in the AppleGMACEthernet project. You are correct that the SimpleUserClient has some examples of this sort of CLI (ie: main()) tool but it's only painfully basic sample code. It's has no relation to the AppleGMACEthernet user client and has no idea about its log format. Lacking the tool that Apple refers to I'm presently rolling my own userland tool, but it's seems a great waste of time given that Apple must have already done this. Thanks again, -Mike > On Tuesday, September 9, 2003, at 09:15 AM, Michael Cashwell wrote: > >> Greetings, >> >> Sorry if this should go to Darwin-drivers. (Some questions span these >> lists' domains.) >> >> I'm trying to make use of the custom logging code (ELG) from Apple's >> GMAC driver in my own code. It's discussed at >> http://developer.apple.com/documentation/DeviceDrivers/Conceptual/ >> WritingDeviceDriver/DebuggingDrivers/chapter_8_section_5.html >> >> which says: >> >>> This driver implements a kernel-resident logging infrastructure as >>> well as a user-space tool to print the formatted contents of a log >>> buffer. The tool used by AppleGMACEthernet prints a report to the >>> standard output that begins with header information followed by >>> time-stamped log entries. >> >> However, the AppleGMACEthernet-124.2.3.tar archive does not appear to >> contain any source for the mentioned userland tool. Does anyone know >> where I might find this source? >> >> Thanks, >> -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. > _______________________________________________ > 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: Custom logging user client tool from GMAC?
by chuck remes 10 Sep '03

10 Sep '03

Take a look at the sample code from DTS, for starters. http://www.devworld.apple.com/samplecode/Sample_Code/ Devices_and_Hardware/IOKit/SimpleUserClient.htm Then you can find the code for GMAC's usertool at: http://www.opendarwin.org/cgi-bin/cvsweb.cgi/src/AppleGMACEthernet/ It's embedded into the source file UniNEnet.cpp at the bottom of the file. The class definitions are in the associated header file. cr On Tuesday, September 9, 2003, at 09:15 AM, Michael Cashwell wrote: Greetings, Sorry if this should go to Darwin-drivers. (Some questions span these lists' domains.) I'm trying to make use of the custom logging code (ELG) from Apple's GMAC driver in my own code. It's discussed at http://developer.apple.com/documentation/DeviceDrivers/Conceptual/ WritingDeviceDriver/DebuggingDrivers/chapter_8_section_5.html which says: This driver implements a kernel-resident logging infrastructure as well as a user-space tool to print the formatted contents of a log buffer. The tool used by AppleGMACEthernet prints a report to the standard output that begins with header information followed by time-stamped log entries. However, the AppleGMACEthernet-124.2.3.tar archive does not appear to contain any source for the mentioned userland tool. Does anyone know where I might find this source? Thanks, -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. _______________________________________________ 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: Programmatic Back Trace
by Geoffrey Schmit 09 Sep '03

09 Sep '03

On 03.09.2003 at 2:22 PM, Chris Thomas <chris(a)m-audio.com> wrote: > I have a need (well, a strong desire) for this myself right now. > > http://developer.apple.com/technotes/tn2002/tn2063.html has > information which may be pertinent: it lists the source files used to > generate the textual kernel panic, ie: > >> Line 6 xnu/osfmk/ppc/model_dep.c: dump_backtrace > > It's not immediately clear to me whether or not this code is useful > outside of an exception context. Thanks for the information; model_dep.c was quite helpful. The dump_backtrace function does basically what I need. I was able to easily integrate its logic into my library and now have the ability to capture kernel backtraces. Thanks again! geoff > On Sep 3, 2003, at 9:30 AM, Geoffrey Schmit wrote: > >> On 03.09.2003 at 11:06 AM, Jim Magee <jmagee(a)apple.com> wrote: >> >>> On Sep 3, 2003, at 10:23 AM, Geoffrey Schmit wrote: >>>> I'm currently stuck trying to find a function that is accessible by >>>> a kernel extension and that is equivalent to vm_read_overwrite in >>>> that it will return an error if the specified address is >>>> inaccessible rather than throwing a SIGSEGV signal. >>>> >>>> So, a couple of questions: >>>> >>>> Is there an example that captures a back trace in a kernel >>>> extension? If so, where is it? >>> >>> Are you trying to get a user-level backtrace or a kernel backtrace >>> from your kernel extension? And more importantly, why are you doing >>> this from a kernel extension in the first place? >> >> I'm primarily interested in getting a kernel backtrace from my kernel >> extension. I'm porting a library that provides various services to >> other kernel extensions we develop. One of these services is to >> capture >> a backtrace. For example, we may capture a backtrace when a kernel >> extension detects an error in order to facilitate the debugging of >> that error condition. As another example, we may capture a backtrace >> when a kernel extension acquires a resource in order to later >> determine who acquired that resource if the resource is leaked. >> >> Ideally, it would be beneficial for the backtrace to eventually >> include the stack of the user-mode thread that initiated the >> transition to the kernel (in scenarios where this is the case). >> However, that is probably >> pushing my luck.... >> >> I'm not at all interested in developing a kernel extension to capture >> user-level backtraces from arbitrary processes. >> >>> In-kernel, if you access memory you don't have mapped, you're >>> dead (panic). You don't get a signal to respond to. And, besides, >>> you can't directly access any user-level memory at all. On >>> PowerPC, user- memory isn't mapped into the kernel's view of >>> memory, since the kernel gets it own 4GB address space. i386 is >>> slightly different, but changing more towards the PowerPC model >>> than the other way around. >> >> So, my problem is probably that as I walk the stack I eventually >> encounter an address from a user-mode thread and cause a panic. Is >> there a way to test if the address belongs to the kernel or otherwise >> determine that I should stop walking the stack? -- Geoffrey Schmit Senior Software Engineer National Instruments geoffrey.schmit(a)ni.com www.ni.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