Bug 19050 - [ANCK6.6.71-3 rc1][aarch64][倚天710机器]perf-sanity-tests测试套,probe libc's inet_pton & backtrace it with ping用例fail,实际的backtrace entry与期望不符
Summary: [ANCK6.6.71-3 rc1][aarch64][倚天710机器]perf-sanity-tests测试套,probe libc's inet_pt...
Status: NEW
Alias: None
Product: Antest
Classification: Infrastructures
Component: 测试用例 (show other bugs) 测试用例
Version: unspecified
Hardware: aarch64 Linux
: P3-Medium S3-normal
Target Milestone: ---
Assignee: yuki_syq
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2025-02-25 14:01 UTC by zhixin01
Modified: 2025-03-04 21:04 UTC (History)
5 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description zhixin01 alibaba_cloud_group 2025-02-25 14:01:04 UTC
[缺陷描述]:
perf-sanity-tests测试套,probe libc's inet_pton & backtrace it with ping用例fail,expected backtrace entry ".*(\+0x[[:xdigit:]]+|\[unknown\])[[:space:]]\(.*/bin/ping.*\)$" got "ffffaf735f7c getaddrinfo+0xec (/usr/lib64/libc-2.32.so)"
test child finished with -1

社区也有类似的bug:https://bugzilla.openanolis.cn/show_bug.cgi?id=4342

测试日志如下:
#perf test -v 84
 84: probe libc's inet_pton & backtrace it with ping                 :
--- start ---
test child forked, pid 1221593
ping 1221621 [002] 79720.592412: probe_libc:inet_pton: (ffffaf767480)
ffffaf767480 __GI___inet_pton+0x0 (/usr/lib64/libc-2.32.so)
ffffaf735f7c getaddrinfo+0xec (/usr/lib64/libc-2.32.so)
FAIL: expected backtrace entry ".*(\+0x[[:xdigit:]]+|\[unknown\])[[:space:]]\(.*/bin/ping.*\)$" got "ffffaf735f7c getaddrinfo+0xec (/usr/lib64/libc-2.32.so)"
test child finished with -1
---- end ----
probe libc's inet_pton & backtrace it with ping: FAILED!

[重现概率]:
必现

[重现环境]:
环境信息:倚天710机器
11.163.178.238

#uname -r
6.6.71-3_rc1.al8.aarch64

#cat /etc/os-release
NAME="Alibaba Cloud Linux"
VERSION="3 (Soaring Falcon)"
ID="alinux"
ID_LIKE="rhel fedora centos anolis"
VERSION_ID="3"
UPDATE_ID="10"
PLATFORM_ID="platform:al8"
PRETTY_NAME="Alibaba Cloud Linux 3 (Soaring Falcon)"
ANSI_COLOR="0;31"
HOME_URL="https://www.aliyun.com/"

#lscpu
Architecture:        aarch64
Byte Order:          Little Endian
CPU(s):              128
On-line CPU(s) list: 0-127
Thread(s) per core:  1
Core(s) per socket:  128
Socket(s):           1
NUMA node(s):        2
Vendor ID:           ARM
BIOS Vendor ID:      T-HEAD
Model:               0
Model name:          Neoverse-N2
BIOS Model name:     Yitian710-128
Stepping:            r0p0
CPU MHz:             2750.000
BogoMIPS:            100.00
Hypervisor vendor:   Alibaba
Virtualization type: full
L1d cache:           64K
L1i cache:           64K
L2 cache:            1024K
L3 cache:            65536K
NUMA node0 CPU(s):   0-63
NUMA node1 CPU(s):   64-127
Flags:               fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh

#free -h
              total        used        free      shared  buff/cache   available
Mem:          125Gi       3.6Gi       121Gi        12Mi       1.0Gi       122Gi
Swap:         2.0Gi          0B       2.0Gi

#cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt2)/boot/vmlinuz-6.6.71-3_rc1.al8.aarch64 root=UUID=d0af582f-7147-41de-85e3-deb2e14cde99 ro biosdevname=0 rd.driver.pre=ahci iommu.passthrough=1 iommu.strict=0 nospectre_bhb ssbd=force-off systemd.unified_cgroup_hierarchy=0 cgroup.memory=nokmem console=ttyS0,115200 fsck.repair=yes crashkernel=0M-2G:0M,2G-256G:256M,256G-1024G:320M,1024G-:384M

#rpm -qa |grep perf
iperf3-3.9-13.al8.1.aarch64
python3-perf-6.6.71-3_rc1.al8.aarch64
gperf-3.1-5.1.al8.aarch64
perf-6.6.71-3_rc1.al8.aarch64

#rpm -qf /usr/lib64/libc-2.32.so
glibc-2.32-1.20.al8.aarch64

[重现步骤]:
#安装随内核发布的perf包
https://koji.alibaba-inc.com/taskinfo?taskID=730244

# 执行测试
perf test -v 17

[期望结果]:
用例执行PASS

[实际结果]:
用例执行Fail
Comment 1 yuki_syq alibaba_cloud_group 2025-03-04 21:04:00 UTC
ali6000-002已知问题,结论:这个测试用例在不同机器上使用的库不一样的话,probe的栈也会不同,两台SPR,probe出来的栈不同,暂时无法同时兼顾。该问题非功能问题,是测试用例无法覆盖多种情况,upstream 存在同样问题,暂置为later,后续版本继续观察