site_archiver@lists.apple.com Delivered-To: darwin-kernel@lists.apple.com The boot process is pretty well documented on developer.apple.com. -- Terry -Al- On Dec 19, 2007, at 1:31 PM, Terry Lambert wrote: -- Terry -Al- On Dec 18, 2007, at 2:38 PM, Terry Lambert wrote: You need to rebuild the KEXT cache, not the KEXT. man kextcache or http://developer.apple.com - Terry -Al- On Dec 18, 2007, at 1:25 PM, Terry Lambert wrote: Instead of removing the KEXT cache, rebuild it. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.a... Copying is not going to work for these files, since there are some very particular requirements for the date stamps on kernel, files, directories, and the cache file itself for the cache to be considered trustworthy. Is there a particular reason you are messing with this on an already booting system, if you are not building your own kernel, or is it just burning curiousity? If the latter, there's a huge amount of documentation on the boot process at <http://developer.apple.com>. On Dec 19, 2007, at 6:03 PM, Al Ciplickas wrote: I finally got the bootcache.plist file copied over and the Extensions.mkext file written. But alas, it still did not want to start up. Same message as in the first note. So there must be something still missing. Do know what other files I might need at start up? -or where to study about startup necessities? I can't believe I'm the only one that has had this problem, but maybe I am. Feeling a bit dumb at the moment. Thanks, This is the first time I have ever heard of it. I had assumed that your problem was that you were building your own kernel, so the KEXT cache was out of date relative to the kernel, and the driver could not be loaded by the boot loader from the cache (it didn't match the kernel), and without a PCI driver, you couldn't load the PCI driver off disk. On Dec 18, 2007, at 8:44 PM, Al Ciplickas wrote: Thanks again, Terry. I tried the rebuild and the complaint was that there was not a usr/../bootcache.plist (or something to that effect). Can't imagine how that could have happened, since I was not aware of that file. So I thought doing an 'upgrade' from an Install DVD would solve the problem. No way. The upgrade finished normally, but there was neither the bootcache.plist nor the kextentions.cache (sp?) files. This is confusing, as I expect a system upgrade to make all the necessary files to do a normal startup. I will locate an active bootcache.plist from an existing system partition and copy it to the usr/../ directory to see if it will help me do the kextcache rebuild. Do you think this is an anomaly for me or has this happened many times before and I should file a radar report? On Dec 18, 2007, at 2:09 PM, Al Ciplickas wrote: Thanks, Terry. I'll rebuild the KEXT. But forgot how at the moment. Any hints? On Dec 18, 2007, at 12:00 PM, Al Ciplickas wrote: This is a new one for me: G5 complains about 'com.apple.driver.AppleMacRiscPCI' missing in the startup from 'kld_load_from_memory()'. I got that using <cmd-V> at startup. I used another system partition to look it the file was there (in System/Library/Extensions/) and it was! So I took out the kextensions.cache and its partner and tried again: still got the same complaint. (By the way, it says there are 1310 buffer headers and 1310 cluster IO buffer header, if that matters.) Do anyone know of something else to try? This is under OSX 10.4.10 as well as 10.4.2 (which I reloaded in hopes of clearing up "lost" files). The error is indicating that it's unable to load it from the cache that was loaded into memory by the boot loader. It can't read it off the disk itself because it can't even talk to the PCI bus on which the disk controller lives without this KEXT loaded. -- Terry This email sent to site_archiver@lists.apple.com