site_archiver@lists.apple.com Delivered-To: darwin-kernel@lists.apple.com On Nov 25, 2008, at 4:20 PM, Brian Bechtel wrote: On Tue, Nov 25, 2008 at 11:53 AM, George Plymale <george@southernohio.net> wrote: The two panics are pretty different. CPU 3 held off interrupts too long (the cpus_to_respond is a mask) This panic is a message from the core which noticed the problem, so the backtrace is useless. 10.5.5 has some new code to report the unresponsive processor, but not all cases can be covered. I would tend to think this is a software issue, not a hardware issue. -Ed _______________________________________________ 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... This email sent to site_archiver@lists.apple.com I just purchased a refurbished MacPro and am quite concerned and was hopeful for perhaps some additional enlightenment on the following panics that seem fairly disjointed. Obviously I'm rather concerned with the issue as I fear perhaps a processor may be at fault. Googling the first panic has lead me to wonder if it is tied to Parallels and the second is fairly nebulous to me at the moment. Please advise as to what you may see is going on here: Sat Nov 22 13:22:49 2008 panic(cpu 7 caller 0x00193077): "pmap_flush_tlbs() timeout: " "cpu(s) failing to respond to interrupts, pmap=0x536500 cpus_to_respond=0x4"@/SourceCache/xnu/xnu-1228.7.58/osfmk/i386/ pmap.c:4580 Actually, this can be a hardware problem... but it usually implies a firmwarey / software misconfiguration of the hardware. I recently worked on a product that had issues responding to PCI-e read requests fast enough to prevent the issuing CPU from occasionally panic'ing this way. smime.p7s