From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.7.12) Gecko/20050922 Fedora/1.0.7-1.1.fc4 Firefox/1.0.7 Description of problem: 4-node IBM x460 each with 4 physical dual-core cpu's with hyperthreading enabled. cat'ing /proc/cpuinfo only reports 32 logical cpu's, where we would expect 64. dmesg output shows that the cpu's are not being sequentially numbered. As a result, some are being incorrectly numbered as being above #128 & thus are not being utilised, eg: ACPI: LAPIC (acpi_id[0x20] lapic_id[0x80] enabled) Processor #128 invalid (max 128) Version-Release number of selected component (if applicable): kernel-largesmp-2.6.9-27.EL.x86_64 How reproducible: Always Steps to Reproduce: 1. boot system with largesmp kernel 2. 3. Additional info:
Created attachment 123626 [details] output from "cat /proc/cpuinfo"
Created attachment 123627 [details] dmesg output
I should add that with hyperthreading disabled, only 16 logical cpu's are reported.
This is a duplicate of Bug 177561. The CPUs are in fact being numbered correctly. The problem is that the kernel only recognized a maximum CPU id of 128 when in fact the hardware limit is 255. The 460 in particular implements "sparse" numbering of CPUs; each successive node starts numbering CPUs at the next multiple of 64, which is why half of the CPUs are in the 128-255 range and thus unused. A patch has already been committed for this. *** This bug has been marked as a duplicate of 177561 ***
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2006-0132.html