summary refs log tree commit diff stats
path: root/results/classifier/zero-shot/118/graphic/1089
blob: c84ec05c64635b5d80d6dd238a600af4c5cb9db6 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
graphic: 0.968
mistranslation: 0.962
device: 0.924
performance: 0.899
PID: 0.853
debug: 0.730
semantic: 0.644
vnc: 0.631
risc-v: 0.608
VMM: 0.603
permissions: 0.598
network: 0.597
register: 0.593
user-level: 0.591
i386: 0.577
assembly: 0.541
ppc: 0.541
architecture: 0.535
x86: 0.522
kernel: 0.519
hypervisor: 0.461
socket: 0.460
files: 0.447
TCG: 0.436
virtual: 0.431
boot: 0.427
KVM: 0.418
peripherals: 0.367
arm: 0.266

when I use memory balloon,the qemu process memory usage is displayed incorrectly
Description of problem:
My vm memory is 4GB,and use the balloon driver,the balloon value is also 4GB.
I run a soft to consume memory in vm,I can see the memory usage rate is 15% in host. When I stop the soft in vm,the memory of free info in host and vm 
become normal,but use "top -d 3 -Hp $qemu_pid" to query in host,the memory usage rate is also 15%.I need to modify the balloon value in a smaller values,the memory usage rate will reduce. why? 
![image](/uploads/cb904692df89db633825da0609458c1f/image.png)
Steps to reproduce:
1.run a soft to consume memory in vm,and query top info,the qemu process memory usage:15%


2.query free info in host and vm (reduce)


3.stop sort in vm


4.query free info in host and vm (recover)


5.query top info again (also 15%)



6.modify the balloon value in a smaller (modify the balloon value in a smaller values,the memory usage rate will reduce)