linux/arch/x86_64
Eric W. Biederman 3d483f4757 [PATCH] Fix sync_tsc hang
sync_tsc was using smp_call_function to ask the boot processor to report
it's tsc value.  smp_call_function performs an IPI_send_allbutself which is
a broadcast ipi.  There is a window during processor startup during which
the target cpu has started and before it has initialized it's interrupt
vectors so it can properly process an interrupt.  Receveing an interrupt
during that window will triple fault the cpu and do other nasty things.

Why cli does not protect us from that is beyond me.

The simple fix is to match ia64 and provide a smp_call_function_single.
Which avoids the broadcast and is more efficient.

This certainly fixes the problem of getting stuck on boot which was
very easy to trigger on my SMP Hyperthreaded Xeon, and I think
it fixes it for the right reasons.

Minor changes by AK

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-29 15:01:13 -07:00
..
boot [PATCH] x86_64: coding style and whitespace fixups 2005-06-25 16:25:02 -07:00
crypto [CRYPTO] Add x86_64 asm AES 2005-07-06 13:55:00 -07:00
ia32 [PATCH] x86_64: Icecream has no way of detecting assembler-level includes 2005-07-28 21:46:01 -07:00
kernel [PATCH] Fix sync_tsc hang 2005-07-29 15:01:13 -07:00
lib [PATCH] x86_64: Remove unused variable in delay.c 2005-07-28 21:45:59 -07:00
mm [PATCH] x86_64: fix cpu_to_node setup for sparse apic_ids 2005-07-28 21:46:02 -07:00
oprofile
pci [PATCH] x86_64: Remove unused variable in k8-bus.c 2005-07-28 21:46:02 -07:00
defconfig [PATCH] x86_64: Update defconfig 2005-07-28 21:45:57 -07:00
Kconfig [PATCH] x86_64: Improve CONFIG_GART_IOMMU description and make it default y 2005-07-28 21:45:59 -07:00
Kconfig.debug
Makefile [PATCH] x86_64: Remove IA32_* build tools in Makefile 2005-07-28 21:46:00 -07:00