微信 4.0.0.23 启动过程中会有子进程 WeChatAppEx 的 crash

[tsingkong]

Linux debian 6.12.0 #1 SMP Mon Nov 18 07:27:21 CST 2024 loongarch64 GNU/Linux

liblol-dkms/now 0.1.1 loong64 [已安装，本地]
liblol/now 0.1.6-1 loong64 [已安装，本地]

内核模块已经加载:

12月 04 09:27:22 debian kernel: la_ow_syscall: loading out-of-tree module taints kernel.
12月 04 09:27:22 debian kernel: la_ow_syscall: module verification failed: signature and/or required key missing - tainting kernel
12月 04 09:27:22 debian kernel: la_ow_syscall module successfully initialized
12月 04 09:27:22 debian systemd-modules-load[905]: Inserted module 'la_ow_syscall'

经测试，关掉 CONFIG_MODULE_SIG 后也是同样的问题：

12月 04 09:30:44 debian wechat[8166]: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall nr=0xde arg1=0x0 arg2=0x940 arg3=0x3 arg4=0x8001
12月 04 09:30:44 debian wechat[8166]: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall nr=0xde arg1=0x0 arg2=0x940 arg3=0x3 arg4=0x8001
12月 04 09:30:44 debian wechat[8166]: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall nr=0xde arg1=0x0 arg2=0x940 arg3=0x3 arg4=0x8001
12月 04 09:30:44 debian wechat[8166]: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall nr=0xde arg1=0x0 arg2=0x940 arg3=0x3 arg4=0x8001
12月 04 09:30:44 debian systemd-coredump[8216]: Process 8166 (WeChatAppEx) of user 1000 terminated abnormally with signal 11/SEGV, processing...
12月 04 09:30:44 debian systemd[1]: Created slice system-drkonqi\x2dcoredump\x2dprocessor.slice - Slice /system/drkonqi-coredump-processor.
12月 04 09:30:44 debian systemd[1]: Created slice system-systemd\x2dcoredump.slice - Slice /system/systemd-coredump.
12月 04 09:30:44 debian systemd[1]: Started systemd-coredump@0-8216-0.service - Process Core Dump (PID 8216/UID 0).
12月 04 09:30:44 debian systemd[1]: Started drkonqi-coredump-processor@0-8216-0.service - Pass systemd-coredump journal entries to relevant user for potential DrKonqi handling.
12月 04 09:30:45 debian systemd-coredump[8233]: [🡕] Process 8166 (WeChatAppEx) of user 1000 dumped core.
                                                 
                                                 Module libzstd.so.1 from deb libzstd-1.5.6+dfsg-1+b1.loong64
                                                 Module libblkid.so.1 from deb util-linux-2.40.2-11.loong64
                                                 Module libstdc++.so.6 from deb gcc-14-14.2.0-9+b1.loong64
                                                 Module libowl.so without build-id.
                                                 Module libilink_protobuf.so without build-id.
                                                 Module libilink_network.so without build-id.
                                                 Module libmount.so.1 from deb util-linux-2.40.2-11.loong64
                                                 Module libsystemd.so.0 from deb systemd-257~rc3-1.loong64
                                                 Module libatomic.so.1 from deb gcc-14-14.2.0-9+b1.loong64
                                                 Module libgcc_s.so.1 from deb gcc-14-14.2.0-9+b1.loong64
                                                 Module libilink2.so without build-id.
                                                 Stack trace of thread 8207:
                                                 #0  0x000055555ec1d5ec n/a (n/a + 0x0)
                                                 #1  0x000055555ec1d5a8 n/a (n/a + 0x0)
                                                 #2  0x000055555ec24654 n/a (n/a + 0x0)
                                                 #3  0x000055555ec24410 n/a (n/a + 0x0)
                                                 ELF object binary architecture: LoongArch
12月 04 09:30:45 debian systemd[1]: systemd-coredump@0-8216-0.service: Deactivated successfully.
12月 04 09:30:45 debian systemd[1]: systemd-coredump@0-8216-0.service: Consumed 633ms CPU time, 165.6M memory peak.
12月 04 09:30:45 debian wechat[8408]: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall nr=0xde arg1=0x0 arg2=0x940 arg3=0x3 arg4=0x8001
12月 04 09:30:45 debian systemd-coredump[8439]: Process 8408 (WeChatAppEx) of user 1000 terminated abnormally with signal 11/SEGV, processing...
12月 04 09:30:45 debian systemd[1]: Started systemd-coredump@1-8439-0.service - Process Core Dump (PID 8439/UID 0).
12月 04 09:30:45 debian systemd[1]: Started drkonqi-coredump-processor@1-8439-0.service - Pass systemd-coredump journal entries to relevant user for potential DrKonqi handling.
12月 04 09:30:45 debian drkonqi-coredump-processor[8234]: "/opt/apps/com.tencent.wechat/files/RadiumWMPF/runtime/WeChatAppEx" 8166 "/var/lib/systemd/coredump/core.WeChatAppEx.1000.8c65d0af709242b190f1f01fbbfe012d.8166.1733275844000000.zst"
12月 04 09:30:45 debian drkonqi-coredump-processor[3105]: "/opt/apps/com.tencent.wechat/files/RadiumWMPF/runtime/WeChatAppEx" 8166 "/var/lib/systemd/coredump/core.WeChatAppEx.1000.8c65d0af709242b190f1f01fbbfe012d.8166.1733275844000000.zst"
12月 04 09:30:45 debian systemd[1825]: Started drkonqi-coredump-launcher@0-8234-0.service - Launch DrKonqi for a systemd-coredump crash (PID 8234/UID 0).
12月 04 09:30:45 debian systemd[1825]: Started drkonqi-coredump-launcher@1-3105-1000.service - Launch DrKonqi for a systemd-coredump crash (PID 3105/UID 1000).

虽然有 WeChatAppEx 的异常，但是微信可以启动起来，也可以正常使用。

[tsingkong]

更新: 升级到 liblol_0.1.8-1_loong64 对这个问题没有改善