Juergen Gross b2d7a075a1 x86/pae: use 64 bit atomic xchg function in native_ptep_get_and_clear
Using only 32-bit writes for the pte will result in an intermediate
L1TF vulnerable PTE. When running as a Xen PV guest this will at once
switch the guest to shadow mode resulting in a loss of performance.

Use arch_atomic64_xchg() instead which will perform the requested
operation atomically with all 64 bits.

Some performance considerations according to:

https://software.intel.com/sites/default/files/managed/ad/dc/Intel-Xeon-Scalable-Processor-throughput-latency.pdf

The main number should be the latency, as there is no tight loop around
native_ptep_get_and_clear().

"lock cmpxchg8b" has a latency of 20 cycles, while "lock xchg" (with a
memory operand) isn't mentioned in that document. "lock xadd" (with xadd
having 3 cycles less latency than xchg) has a latency of 11, so we can
assume a latency of 14 for "lock xchg".

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Tested-by: Jason Andryuk <jandryuk@gmail.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2018-08-27 14:20:49 -04:00
..
2018-08-19 10:38:36 -07:00
2018-07-25 11:53:59 +02:00
2018-02-15 01:15:50 +01:00
2018-07-25 11:53:59 +02:00
2018-03-20 10:01:57 +01:00
2018-01-30 15:08:27 +01:00
2018-05-19 13:55:40 +02:00
2018-08-19 10:38:36 -07:00
2018-08-15 13:44:10 -07:00
2017-12-22 20:13:04 +01:00
2017-12-04 10:56:53 +10:00
2018-08-19 10:38:36 -07:00
2017-12-22 20:13:01 +01:00
2018-08-19 10:38:36 -07:00
2017-07-26 13:18:20 +02:00
2018-05-07 07:15:41 +02:00
2018-06-07 10:23:33 -07:00
2018-07-20 01:11:45 +02:00
2017-11-15 18:21:05 -08:00
2018-03-28 22:47:06 +02:00
2018-03-20 10:01:57 +01:00
2018-05-09 21:41:38 +02:00
2017-11-15 18:21:04 -08:00