text stringlengths 0 2.35k |
|---|
**Ivan Kwiatkowski:** I can pretty much give you my personal guarantee that if anyone in the audience actually implements one of those two things, that they're going to be extremely rich. So there you go. |
**Natalie Pistunovich:** That's almost as good as being on the rank of the most unpopular opinion. \[laughs\] Thanks, everyone, for joining. Thank you very much, Ivan. |
**Ivan Kwiatkowski:** Thank you. I was happy being here, it was a pleasure speaking to you, and see you next time. |
• Security in Go is less of a concern due to its memory-safe design and compiler restrictions |
• Go eliminates common bug classes like memory corruptions and buffer overflows |
• Code Insights from Sourcegraph turns codebases into databases, enabling advanced searches and tracking |
• The feature unlocks infinite possibilities for code analysis and intelligence |
• Version tracking and other use cases have been discovered through user feedback |
• The tool allows users to track anything in their codebase with ease |
• Ivan Kwiatkowski, senior security researcher at Kaspersky, discusses his work in threat intelligence |
• He describes his daily tasks, including analyzing malware and writing reports to help customers understand potential threats |
• The SolarWinds attack is mentioned as a notable example of a high-profile incident |
• Ivan explains the tactics used by attackers in the SolarWinds case, including compromising the software build chain and inserting backdoors |
• He discusses the use of Go language in malicious code, specifically in the SunShuttle malware |
• Other examples of Go-based malware are mentioned, including Stowaway |
• Malware written in Go language is difficult to reverse engineer due to its architecture. |
• Examples of malware families written in Go include tools that proxy between protocols and a commercial backdoor called BrutRatel. |
• The ease of use for developers is a significant reason why attackers prefer using Go: it generates self-contained, statically built executables that don't require additional libraries. |
• Another reason is the difficulty of reverse engineering, which makes it hard to analyze malware written in Go with standard tools and techniques. |
• The unique assembly generated by the Go compiler also makes it difficult for defenders to recognize specific patterns in code. |
• Malware that return C can be easily identified in a test list |
• Assembly differences between Go and C make malware harder to recognize |
• Go executables are larger due to function inclusion, causing issues with static analysis tools like IDA Pro |
• Calling conventions in Go differ from C, making it difficult for tools to identify return values |
• Go compilers dynamically allocate space on the stack for local variables, leading to difficulties in determining variable locations and offsets |
• Go compiler optimizations make it difficult to track variables and function calls |
• IDA Pro and other tools struggle with renaming variables due to inconsistent stack representation |
• Go's calling convention change in version 16.1 or later makes some analysis easier, but not the bigger problem of variable naming |
• Reverse engineering involves analyzing program instructions to understand programmer intent |
• The process is like a "mostly exact science" that requires effort and expertise to decipher CPU-level language |
• It's extremely challenging for humans to understand and work with low-level code, making reverse engineers a valuable asset in the industry |
• The speaker finds analyzing Go programs unpleasant due to the high ratio of assembly code to source code. |
• This ratio can be as high as 15 lines of assembly for one line of Go code. |
• Other languages like C++ and C also generate a lot of assembly code, but C's translation to assembly is more direct. |
• The problem with analyzing Go programs is not that they generate more assembly code, but rather that the resulting assembly code is unfamiliar and difficult to work with. |
• The speaker hopes that as Go tools become more prevalent, there will be pressure on tool authors to improve detection and support for Go. |
• Current analysis of Go programs in IDA Pro can be challenging due to unexpected behavior such as jumping to random locations in the program when stepping through instructions. |
• This is likely caused by the Go scheduler or garbage collector. |
• The speaker discusses their experience with debugging a Go program, specifically the behavior of the garbage collector. |
• The garbage collector can move the program's execution point to a different location, making it difficult for reverse engineers to follow. |
• This is in contrast to other languages like Java and C#, which compile to bytecode or have a higher-level abstraction that hides CPU instructions. |
• Go's compiler optimizes function calls by masking memory movements, resulting in improved performance. |
• The speaker highlights the benefits of this optimization, including reduced time spent on unnecessary memory movements. |
• The speaker discusses the speed of using their tool, Firehydrant, to value incidents. |
• The importance of consistency in programming languages, specifically Go, is mentioned as being helpful for reverse engineering. |
• The strictness of Go's compiler and its effect on code structure and behavior is discussed. |
• The speaker shares their experience with Go and how it can be useful for hackers and researchers due to its conventions and compiler-enforced rules. |
• Historically used programming languages in hacking and research are mentioned, including Go, Rust, C, C++, and others. |
• The difficulty of adapting to new languages, but the necessity of being able to analyze whatever malware or tools come their way. |
• Discussion of incident response procedures for collecting data from a compromised network |
• Explanation of data collection methods, including machine logs, DNS logs, HTTP proxy records, and net flow information |
• Importance of malware analysis in reconstructing an incident timeline |
• Lateral movement techniques used by attackers and the need to detect these actions |
• Role of incident response teams in collecting and analyzing data during an incident |
• Limitations of the current system for gathering intelligence from incident response cases |
• Discussion of the value of big data analytics in understanding security threats |
• Benefits and features of using the Go programming language for secure software development, including memory safety and reduced risk of buffer overflows |
• Importance of cryptography in secure software development and the default algorithms used by the Go language |
• Go's developer code does not require manual initialization of an IV for encryption |
• This prevents common mistakes, such as using a weak or default IV |
• Go was designed with security in mind, making it more difficult to misuse encryption features |
• The language's defaults and restrictions are intended to protect developers from themselves |
• Using tools like Honeycomb can help teams quickly identify issues and improve efficiency |
• The conversation turns to reverse engineering Go code, which is challenging due to its compiler optimizations |
• The speaker recommends using IDA or Ghidra for reversing Go programs, although both have limitations |
• The speaker discusses the rapid improvement of Go and its potential for reverse engineering |
• They mention Juan Andres Guerrero Sad's plugins for IDA to help with reverse engineering Go programs |
• The speaker notes that while IDA may not be perfect, it is one of the only tools available for the job |
• They share their experience finding Go more enjoyable to reverse engineer than C++ due to its simplicity and API-based design |
• The speaker advises beginners to start with traditional C or assembly code, as these are fundamental knowledge bases for reverse engineering |
• Recommended resources include Kaspersky courses, a beginner's website called Beginners.re (formerly free), the book "Practical Malware Analysis", and games like Turing Complete and Assembly games (A TIS 100 and Hexapunks) |
• Zaktronics, a game development company, offered an extensive education program for universities where they provided access to their games for free. |
• The guest has several unpopular opinions, including: |
• Cyberspace will never be regulated |
• NFTs are a scam |
• There is no political will to limit the sale of cyber offense tools |
• The European Union's rule standardizing USB-C charging is welcome by the guest, who has been carrying multiple chargers for years. |
• The guest is not a fan of Apple and dislikes their ecosystem, particularly their practice of selling new chargers with every product release. |
• The guest has mixed feelings about app stores (Google Play Store and Apple Store) from a security perspective, believing they are not the best solution to prevent malicious apps. |
• The guest prefers device-level protections over controlling app stores. |
• Discussion of whether countries will regulate cyber offense |
• Views on the effectiveness and feasibility of regulating cyberspace |
• Incentives and motivations behind states' actions in cyber offense |
• The challenges and difficulties in quantifying costs and consequences of cyber attacks |
• Criticism of discussions about making a "safer internet" as potentially being conducted in bad faith |
[0.00 --> 4.44] I think one of the main appeals of Go is that you don't really need to think about security |
[4.44 --> 6.36] as much as with other languages. |
[6.62 --> 12.00] Go is a memory-safe language, unless I'm mistaken, and the compiler is never going to let you |
[12.00 --> 16.68] do stupid stuff like create an array that is too small and then write stuff that goes |
[16.68 --> 17.08] out of it. |
[17.28 --> 18.12] It's just not possible. |
[18.12 --> 22.98] So it eliminates a whole lot of bug classes, which we call memory corruptions. |
[23.10 --> 24.12] It's just not going to happen. |
[24.22 --> 25.82] You cannot do this to yourself in Go. |
[25.82 --> 31.68] And it means that all the old school buffer overflows that plagued all the C and C++ programs |
[31.68 --> 35.66] for dozens of years are not going to ever happen in the Go language. |
[35.76 --> 40.76] It doesn't mean that the program is going to be perfectly safe from any security issues, |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.