January 2010 - Darwin-kernel

Re: Is it possible/safe to use C++ in a NKE?
by Andrew Myrick 28 Jan '10

28 Jan '10

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

Is it possible/safe to use C++ in a NKE?
by Martin Hardman 28 Jan '10

28 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thread-index: AcqgXMeDy9zraOVYTqKax98pFERx8g== Thread-topic: Is it possible/safe to use C++ in a NKE? _______________________________________________ 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: link state missing error on 10.6
by Rick Macklem 28 Jan '10

28 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Wed, 27 Jan 2010, Andrew Myrick wrote: Rick, I strongly encourage you to switch to the KPI dependency model. While binary compatibility dictates that com.apple.kernel should work for 32-bit kexts on Snow Leopard, we do not and will not support it for 64-bit kexts. kextlibs(8) is a good tool to help you make the transition. If you need a kernel symbol that is not available in the KPIs, please file a bug. Understood. However an NFSv4 client must tinker with all sorts of things. I'm not saying it's impossible to get it to work using the KPIs, but I just do it as a "spare time hobby" and I don't even own an Intel Mac, so I doubt I'll get around to trying. (I suspect Apple will ship an NFSv4 client with their next release anyhow, so I don't think it much matters.) rick _______________________________________________ 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: Kext unload issues with kext with multiple classes
by Daniel Markarian 28 Jan '10

28 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hey Maxim, You cannot defeat termination. You can only delay parts of termination, for the length of time B is kept open by a client. Dan Le 2010-01-27 à 05:31, Maxim Zhuravlev a écrit : > Hello, > > I've got a kext, that contains several classes, say class A and class > B. Instances of class B are clients of those of class A. They are > loading just fine. > B should be able to restrict kext unload. For this purpose I override > B::terminate() which returns false if the kext shouldn't be unloaded. > I expect that the whole chain A<-B<-B::userClient keeps on working in > this case and kextunload returns an error. > The problem is that when I unload the kext and a instance of A is > being tried to be unloaded, I get the stack: > > #0 B::terminate (this=0xffffff800be73100, options=1048583) > #1 0xffffff800051d6be in IOService::terminateClient (this=<value > temporarily unavailable, due to optimizations>, > client=0xffffff800be73100, options=1048583) at > /SourceCache/xnu/xnu-1486.2.11/iokit/Kernel/IOService.cpp:2235 > #2 0xffffff800051d648 in IOService::requestTerminate > (this=0xffffff800be73100, provider=0x100007, options=1048583) at > /SourceCache/xnu/xnu-1486.2.11/iokit/Kernel/IOService.cpp:1660 > #3 0xffffff800052235e in IOService::terminatePhase1 > (this=0xffffff800c037800, options=7) at > /SourceCache/xnu/xnu-1486.2.11/iokit/Kernel/IOService.cpp:1727 > #4 0xffffff80005224c4 in IOService::terminate (this=<value > temporarily unavailable, due to optimizations>, options=<value > temporarily unavailable, due to optimizations>) at > /SourceCache/xnu/xnu-1486.2.11/iokit/Kernel/IOService.cpp:2246 > #5 0xffffff800052ee3b in _terminateDrivers > (matching=0xffffff800c106880) at > /SourceCache/xnu/xnu-1486.2.11/iokit/Kernel/IOCatalogue.cpp:609 > ... > > Here frame 4 relates to A instance termination. > > B::terminate() returns false, but IOService::requestTerminate does't > check the result and proceeds B's instance termination. What is the > right way to handle my requirements? > > Thanks in advance. > _______________________________________________ > 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 _______________________________________________ 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: link state missing error on 10.6
by Michael Eddington 28 Jan '10

28 Jan '10

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=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=LsvJaDc3M9+edCZRrUmsb6YXq6c9GfHk5kk5oWnzY3w=; b=EPRCBgUgtHmHDogIwV4mLxduz/OR9FYROSMSNRmyV4gigiYaxzjYiH8GeDNNGOj660 yXHlBtZmr9RXswtP/HkSc4+1ye06wGYWZpQqIHUSOSyQsf30pAtyX53sd99jOn/3V36f KJFDX2yXvYpLOxHJZ8lmf77Y1tmDQMXcpw8H8= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=noypRwI3VtHQTv6QxWFXzhuPvKaW7AAlkBaZxFDwWmUnc6DF2agKAY6+CEBXfdUqq8 evzPriFp1WRW9DBwTqxoa2utbTib8PMMmpJj+oFUM45gJ/f3wN6/vo+xAcjlsIFcYW76 EWUdiWIbyQqUA3tsI9Gh1wEv98H9823qMf/UI= Good to hear! I opened a bug with ID 7584314 for this issue. -mike On Wed, Jan 27, 2010 at 3:26 PM, Nik Gervae <ngervae(a)apple.com> wrote: > Thanks for highlighting that code; it's pointed me to the origin of the problem. I'll have to consult with another engineer on the best solution, but it looks tractable. > > Nik Gervae > Kext Management Engineer > Apple, Inc. > > > On 27 Jan, 2010, at 14:07 , Rick Macklem wrote: > >> On Wed, 27 Jan 2010, Michael Eddington wrote: >> >>> Lets forget for a second that I mentioned nsysent. We have what looks >>> like broken functionality between 10.5 and 10.6. A kext that uses >>> com.apple.kernel as it's dependencies will pass the initial checks for >>> kext loading, try to be loaded and then the loader will bail out when >>> it finds a null pointer. >>> >>> Here is the code I found for OSKext.cpp that issues the error message. >>> >>>> From OSKext::loadExecutable(): >>> >>> if (isKernelComponent()) { >>> OSData * kernelLinkState = OSKext::getKernelLinkState(); >>> kxlddeps[0] = (u_char *)kernelLinkState->getBytesNoCopy(); >>> } else for (i = 0; i < num_kxlddeps; i++) { >>> OSKext * dependency = OSDynamicCast(OSKext, dependencies->getObject(i)); >>> if (!dependency->linkState) { >>> // xxx - maybe we should panic here >>> OSKextLog(this, >>> kOSKextLogErrorLevel | >>> kOSKextLogLoadFlag | kOSKextLogLinkFlag, >>> "Can't load kext %s - link state missing.", >>> getIdentifierCString()); >>> goto finish; >>> } >>> kxlddeps[i] = (u_char *)dependency->linkState->getBytesNoCopy(); >>> assert(kxlddeps[i]); >>> } >>> >>> -mike >> Just fyi, my open source NFSv4 client that works for Leopard >> (http://code.google.com/p/macnfsv4) can't >> be ported to Snow Leopard until the above bug is fixed (or there is >> a way to load a Kext against com.apple.kernel). My current attempt >> fails with the "... link state missing" as above. >> >> rick >> _______________________________________________ >> 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/ngervae%40apple.com >> >> This email sent to ngervae(a)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: link state missing error on 10.6
by Nik Gervae 27 Jan '10

27 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thanks for highlighting that code; it's pointed me to the origin of the problem. I'll have to consult with another engineer on the best solution, but it looks tractable. Nik Gervae Kext Management Engineer Apple, Inc. On 27 Jan, 2010, at 14:07 , Rick Macklem wrote: > On Wed, 27 Jan 2010, Michael Eddington wrote: > >> Lets forget for a second that I mentioned nsysent. We have what looks >> like broken functionality between 10.5 and 10.6. A kext that uses >> com.apple.kernel as it's dependencies will pass the initial checks for >> kext loading, try to be loaded and then the loader will bail out when >> it finds a null pointer. >> >> Here is the code I found for OSKext.cpp that issues the error message. >> >>> From OSKext::loadExecutable(): >> >> if (isKernelComponent()) { >> OSData * kernelLinkState = OSKext::getKernelLinkState(); >> kxlddeps[0] = (u_char *)kernelLinkState->getBytesNoCopy(); >> } else for (i = 0; i < num_kxlddeps; i++) { >> OSKext * dependency = OSDynamicCast(OSKext, dependencies->getObject(i)); >> if (!dependency->linkState) { >> // xxx - maybe we should panic here >> OSKextLog(this, >> kOSKextLogErrorLevel | >> kOSKextLogLoadFlag | kOSKextLogLinkFlag, >> "Can't load kext %s - link state missing.", >> getIdentifierCString()); >> goto finish; >> } >> kxlddeps[i] = (u_char *)dependency->linkState->getBytesNoCopy(); >> assert(kxlddeps[i]); >> } >> >> -mike > Just fyi, my open source NFSv4 client that works for Leopard > (http://code.google.com/p/macnfsv4) can't > be ported to Snow Leopard until the above bug is fixed (or there is > a way to load a Kext against com.apple.kernel). My current attempt > fails with the "... link state missing" as above. > > rick > _______________________________________________ > 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/ngervae%40apple.com > > This email sent to ngervae(a)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: link state missing error on 10.6
by Andrew Myrick 27 Jan '10

27 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jan 27, 2010, at 2:07 PM, Rick Macklem wrote: > On Wed, 27 Jan 2010, Michael Eddington wrote: > >> Lets forget for a second that I mentioned nsysent. We have what looks >> like broken functionality between 10.5 and 10.6. A kext that uses >> com.apple.kernel as it's dependencies will pass the initial checks for >> kext loading, try to be loaded and then the loader will bail out when >> it finds a null pointer. >> >> Here is the code I found for OSKext.cpp that issues the error message. >> >>> From OSKext::loadExecutable(): >> >> if (isKernelComponent()) { >> OSData * kernelLinkState = OSKext::getKernelLinkState(); >> kxlddeps[0] = (u_char *)kernelLinkState->getBytesNoCopy(); >> } else for (i = 0; i < num_kxlddeps; i++) { >> OSKext * dependency = OSDynamicCast(OSKext, dependencies->getObject(i)); >> if (!dependency->linkState) { >> // xxx - maybe we should panic here >> OSKextLog(this, >> kOSKextLogErrorLevel | >> kOSKextLogLoadFlag | kOSKextLogLinkFlag, >> "Can't load kext %s - link state missing.", >> getIdentifierCString()); >> goto finish; >> } >> kxlddeps[i] = (u_char *)dependency->linkState->getBytesNoCopy(); >> assert(kxlddeps[i]); >> } >> >> -mike > Just fyi, my open source NFSv4 client that works for Leopard > (http://code.google.com/p/macnfsv4) can't > be ported to Snow Leopard until the above bug is fixed (or there is > a way to load a Kext against com.apple.kernel). My current attempt > fails with the "... link state missing" as above. Rick, I strongly encourage you to switch to the KPI dependency model. While binary compatibility dictates that com.apple.kernel should work for 32-bit kexts on Snow Leopard, we do not and will not support it for 64-bit kexts. kextlibs(8) is a good tool to help you make the transition. If you need a kernel symbol that is not available in the KPIs, please file a bug. -Andrew _______________________________________________ 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: KEXT loading issue on 10.6.*
by Nik Gervae 27 Jan '10

27 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thanks for the info! Every little bit helps. We are also following up directly with Kevin to get more detailed data since he sees the problem somewhat regularly. If you encounter the problem, could you check /Library/Logs/CrashReporter or /Library/Logs/DiagnosticReports for any kext-related programs? We might be interested in looking at your kext caches, too, from /System/Library/Caches/com.apple.kext.caches/. Nik On 26 Jan, 2010, at 14:23 , Ryan Hankins wrote: > I didn't see any follow-up on this, so I'll post my CFBundleVersion. > > I have: > <key>CFBundleVersion</key> > <string>2.4</string> > > It is the same for both the driver, and the fallback in Plugins. > > Thanks, > -Ryan > >> Hello, >> >> I'd like some more information to diagnose the problem. Could you send the CFBundleVersion values of your top-level and plugin kexts? If they're the same I have some possible theories about what might be going on. We'll work from there. >> >> Also, if you are experiencing problems with kext management, by all means please file bugs using https://bugreport.apple.com/. We can't fix bugs we don't know about, and even if we do know about them, knowing their prevalence helps us prioritize. >> >> Nik Gervae >> Kext management engineer >> Apple Inc. >> >> >> On 19 Jan, 2010, at 13:10 , Kevin Brock wrote: >> >> >>> On Jan 19, 2010, at 12:58 PM, Ryan Hankins wrote: >>> >>> >>>> Interesting; we are seeing this issue,, or a similar one, as well. In fact, we were just talking about it today. 10.6 seems to ignore the kext in the main directory, and instead attempts to load the 10.5 one under Plugins. >>>> >>>> One thing that seems to help is removing the Plugins directory in your kext. Does that help in your case? >>>> >>> In my case the 10.5 & 10.6 versions are the same (mod build options), and it's a 10.4 version in Plugins. >>> That being said, I'm sure it *would* help. I'll be testing that later today hopefully. >>> >>> We've got to have that available at the moment, because the KEXT in question is part of a separate package that sometimes needs to be installed on 10.4 systems. I've been trying to convince product management to drop 10.4 support for this item, but it'll take a bit... >>> >>> I suppose we could do a special version in this case. >>> >>> >>>> I have been wondering if this is a bug in Apple's kext loading routines. >>>> >>> I think there are some issues with the 10.6 KEXT mgmt changes. One that our test automation people don't like is that if we're running automated install/uninstall tests on the product, kextd will eventually spit out a message saying that "kextcache has had enough tries; not launching anymore". After that of course everything just goes straight to h**l as far as testing is concerned until you reboot :-( >>> >>> Kevin >>> >>> >>>> -Ryan >>>> >>>>> I've got a couple of KEXTs which have one binary for 10.5/10.6, using various Xcode build options to distinguish 32 from 64 etc. Each of these also has a sub-KEXT which is for 10.4, located in Contents/Plugins. >>>>> >>>>> The problem I'm having is that on 10.6.* sometimes, generally after an upgrade, the 10.4 version of one of the KEXTs is loading instead of the 10.5/10/6 version. >>>>> >>>>> The 10.5+ KEXT has: >>>>> >>>>> - KPI versions 9.0.0 or greater >>>>> - Probe score of 2000 >>>>> >>>>> The 10.4 version of the KEXT specifies: >>>>> >>>>> - KPI versions 8.0.0 or greater >>>>> - Probe score of 100 >>>>> >>>>> Usually everything is fine, and after a reboot everything is always fine, so I know that he right files are getting installed. >>>>> >>>>> I don't see any messages in system.log that would indicate a problem. >>>>> >>>>> Does anyone have any ideas on this? >>>>> >>>>> Kevin >>>>> >>>>> _______________________________________________ >>>>> 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/rqh%40small-tree.com >>>>> >>>>> This email sent to rqh(a)small-tree.com >>>>> >>>>> >>>> -- >>>> Ryan Hankins Software Engineer >>>> Small Tree Communications www.small-tree.com >>>> rqh(a)small-tree.com +1 651 209-6509 x303 >>>> >>>> >>> _______________________________________________ >>> 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/ngervae%40apple.com >>> >>> This email sent to ngervae(a)apple.com >>> >> >> >> > > > -- > Ryan Hankins Software Engineer > Small Tree Communications www.small-tree.com > rqh(a)small-tree.com +1 651 209-6509 x303 > _______________________________________________ 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: link state missing error on 10.6
by Rick Macklem 27 Jan '10

27 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Wed, 27 Jan 2010, Michael Eddington wrote: Lets forget for a second that I mentioned nsysent. We have what looks like broken functionality between 10.5 and 10.6. A kext that uses com.apple.kernel as it's dependencies will pass the initial checks for kext loading, try to be loaded and then the loader will bail out when it finds a null pointer. Here is the code I found for OSKext.cpp that issues the error message. if (isKernelComponent()) { OSData * kernelLinkState = OSKext::getKernelLinkState(); kxlddeps[0] = (u_char *)kernelLinkState->getBytesNoCopy(); } else for (i = 0; i < num_kxlddeps; i++) { OSKext * dependency = OSDynamicCast(OSKext, dependencies->getObject(i)); if (!dependency->linkState) { // xxx - maybe we should panic here OSKextLog(this, kOSKextLogErrorLevel | kOSKextLogLoadFlag | kOSKextLogLinkFlag, "Can't load kext %s - link state missing.", getIdentifierCString()); goto finish; } kxlddeps[i] = (u_char *)dependency->linkState->getBytesNoCopy(); assert(kxlddeps[i]); } Just fyi, my open source NFSv4 client that works for Leopard (http://code.google.com/p/macnfsv4) can't be ported to Snow Leopard until the above bug is fixed (or there is a way to load a Kext against com.apple.kernel). My current attempt fails with the "... link state missing" as above. rick _______________________________________________ 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.… >From OSKext::loadExecutable(): -mike This email sent to site_archiver(a)lists.apple.com

1 0

Re: link state missing error on 10.6
by Andrew Myrick 27 Jan '10

27 Jan '10

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Please file a bug report with a sample kext project that demonstrates the problem. https://bugreport.apple.com/ -Andrew On Jan 27, 2010, at 10:23 AM, Michael Eddington wrote: > Lets forget for a second that I mentioned nsysent. We have what looks > like broken functionality between 10.5 and 10.6. A kext that uses > com.apple.kernel as it's dependencies will pass the initial checks for > kext loading, try to be loaded and then the loader will bail out when > it finds a null pointer. > > Here is the code I found for OSKext.cpp that issues the error message. > >> From OSKext::loadExecutable(): > > if (isKernelComponent()) { > OSData * kernelLinkState = OSKext::getKernelLinkState(); > kxlddeps[0] = (u_char *)kernelLinkState->getBytesNoCopy(); > } else for (i = 0; i < num_kxlddeps; i++) { > OSKext * dependency = OSDynamicCast(OSKext, dependencies->getObject(i)); > if (!dependency->linkState) { > // xxx - maybe we should panic here > OSKextLog(this, > kOSKextLogErrorLevel | > kOSKextLogLoadFlag | kOSKextLogLinkFlag, > "Can't load kext %s - link state missing.", > getIdentifierCString()); > goto finish; > } > kxlddeps[i] = (u_char *)dependency->linkState->getBytesNoCopy(); > assert(kxlddeps[i]); > } > > -mike > > On Tue, Jan 26, 2010 at 9:35 PM, Michael Smith <drivers(a)mu.org> wrote: >> >> On Jan 26, 2010, at 2:55 PM, Michael Eddington wrote: >> >> This only appears to happen when my OSBundleLibraries has >> "com.apple.kernel" "10.2.0" as the dependencies. If I just use, >> com.apple.kernel.bsd or something it will load fine. Sadly I need to >> depend on the full com.apple.kernel to resolve nsysent it looks like. >> >> nsysent is only interesting if you are trying to patch the syscall table, >> which is not supported (or generally a good idea). >> = Mike >> -- >> Ars longa, vita brevis, occasio praeceps, experimentum periculosum, iudicium >> difficile -- Hippocrates >> >> >> >> >> >> > _______________________________________________ > 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/amyrick%40apple.com > > This email sent to amyrick(a)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