Andi Kleen <ak@suse.de> writes:In which case complaining is still correct (the BIOS was out of sync), enabling bit 17 is still correct and we are just in overkill mode. This has everything to do with how AMD coherent hypertransport works and little if anything to do with how the NV bridge operated. Basically the NV bridge seems to be sending a standard hypertransport x86 legacy interrupt packet (that doesn't have any target information) and when that packet hits the coherent hypertransport domain it isn't being converted into whatever would send it to all cpus. ..... The real practical problem is if somehow the BIOS goofs up this way and it then decides to ask us to boot on one of these cpus with an extended apic id. We will hang in calibrate_delay. So far this only seems to happen in the kdump case but in theory the BIOS could be completely crazy. Eric --
| Adrian Bunk | Re: Linux 2.6.21 |
| Linus Torvalds | Linux 2.6.21-rc2 |
| WANG Cong | [-mm Patch] UML: fix a building error |
| Roland McGrath | Re: [PATCH 0/5] ftrace: to kill a daemon |
git: | |
| Natalie Protasevich | [BUG] New Kernel Bugs |
| David Miller | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Patrick McHardy | Re: [PATCH] netfilter: use per-cpu spinlock rather than RCU (v3) |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Theodore Ts'o | Re: cc1 fails silently |
| Michael Nolan | Power routines on notebook cause kernel panic |
| Marc Peters | v 0.11 boot disk problem |
| Dave `geek' Gymer | WARNING (was Re: New afio release) |
