x86/cpufeatures: Enumerate #DB for bus lock detection

mainline inclusion
from mainline-v5.12-rc4
commit f21d4d3b
category: feature
bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I5G10C
CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/


commit/?id=f21d4d3b

Intel-SIG: commit f21d4d3b x86/cpufeatures: Enumerate #DB for bus lock detection

--------------------------------

A bus lock is acquired through either a split locked access to writeback
(WB) memory or any locked access to non-WB memory. This is typically >1000
cycles slower than an atomic operation within a cache line. It also
disrupts performance on other cores.

Some CPUs have the ability to notify the kernel by a #DB trap after a user
instruction acquires a bus lock and is executed. This allows the kernel to
enforce user application throttling or mitigation. Both breakpoint and bus
lock can trigger the #DB trap in the same instruction and the ordering of
handling them is the kernel #DB handler's choice.

The CPU feature flag to be shown in /proc/cpuinfo will be "bus_lock_detect".

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20210322135325.682257-2-fenghua.yu@intel.com



(cherry picked from commit f21d4d3b)
Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>