diff options
| author | Alex Shi <[email protected]> | 2012-06-28 01:02:24 +0000 |
|---|---|---|
| committer | H. Peter Anvin <[email protected]> | 2012-06-28 02:29:14 +0000 |
| commit | effee4b9b3b0aa5770bcd98de5f672b05b27703c (patch) | |
| tree | b167657ec2ba05797b925a93e7e1b45222ac5ac3 /lib/mpi | |
| parent | x86/tlb: replace INVALIDATE_TLB_VECTOR by CALL_FUNCTION_VECTOR (diff) | |
| download | kernel-effee4b9b3b0aa5770bcd98de5f672b05b27703c.tar.gz kernel-effee4b9b3b0aa5770bcd98de5f672b05b27703c.zip | |
x86/tlb: do flush_tlb_kernel_range by 'invlpg'
This patch do flush_tlb_kernel_range by 'invlpg'. The performance pay
and gain was analyzed in previous patch
(x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range).
In the testing: http://lkml.org/lkml/2012/6/21/10
The pay is mostly covered by long kernel path, but the gain is still
quite clear, memory access in user APP can increase 30+% when kernel
execute this funtion.
Signed-off-by: Alex Shi <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: H. Peter Anvin <[email protected]>
Diffstat (limited to 'lib/mpi')
0 files changed, 0 insertions, 0 deletions