anaivebird
anaivebird
**Describe the bug** Error when stack canary is used by program unicorn.unicorn.UcError: Invalid memory read (UC_ERR_READ_UNMAPPED) **Sample Code** ``` mov esi, large gs:14h mov [esp+38h], esi on ida pro: anonymous3...
Analyzing .so in app may be challenging due to mocking JNI functions behavior like FindClass(), GetMethodID(), RegisterNatives() is hard. So any wordaround for qiling to overcome it? May be we...
***Describe the bug** For strlen in libc, qiling failed to emulate it. **Emulated Code** ``` size_t strlen(const char *s) { return off_56561FBC(s); // this line is 0x56555870 } ``` ```...
I want something like this, but that doesn't work. ``` def GetStringUTFChars(ql): ql.log.info('GetStringUTFChars hooked') # put char array on memory and move pointer to eax(I will try to do myself...
When bart decoder vocab size is 52, decode result became strange(very large or negative):  You could use this code to reproduce the bug(including export.py and infer.py): https://gist.github.com/anaivebird/4fa79146ac2a17a67038ab40aa1cc6c1
It seems that the dependency is not so much: 1. ninja (apt install ninja-build,) 2. intel MTK (pip install mtk) (which is actually not needed if using GPU inference only,...
It seems that IDA's default database memory config is as following: (copy from ida.cfg) // Memory configuration parameters // ------------------------------- // NOTES: // // 1. DATABASE_MEMORY determines how much memory...
What is debug_symbols_1.1.2.85.7z in release files used for ?
**您的功能请求是否与问题相关？ 请简单描述.** 比如用户输入 我要看ip，模型自动生成ip addr show，标注数据让大模型few shot就行了 **请描述一下您想要的解决方案** text2sql模型可以把文本变成sql语句，直接把输出改成命令，让我们的用户数据每天自动标注，然后训练nlp模型 **描述你考虑过的替代方案** 我打算自己来开发这个功能