TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

The Case for Memory Safe Roadmaps

309 点作者 AndrewDucker超过 1 年前

27 条评论

BaculumMeumEst超过 1 年前
In the world of graphics programming, you&#x27;ve got:<p>- continuing accumulation of documentation and utility libraries from Khronos<p>- excellent learning materials from the community<p>- tons of legacy code<p>all using C++.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;KhronosGroup&#x2F;Vulkan-Utility-Libraries">https:&#x2F;&#x2F;github.com&#x2F;KhronosGroup&#x2F;Vulkan-Utility-Libraries</a><p><a href="https:&#x2F;&#x2F;github.com&#x2F;cg-tuwien&#x2F;VulkanLaunchpad">https:&#x2F;&#x2F;github.com&#x2F;cg-tuwien&#x2F;VulkanLaunchpad</a><p><a href="https:&#x2F;&#x2F;cescg.org&#x2F;our-services&#x2F;an-introduction-to-vulkan&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;cescg.org&#x2F;our-services&#x2F;an-introduction-to-vulkan&#x2F;</a><p>Until I see professionals get funding to build, maintain, document, and create training materials for Rust bindings, I&#x27;m going to continue to assume there will not be movement in that sector. I don&#x27;t think industry is going to throw away the C++ ecosystem and build greenfield projects on a foundation of hobby projects.
评论 #38616729 未加载
评论 #38622233 未加载
评论 #38617604 未加载
评论 #38618736 未加载
评论 #38614807 未加载
wodow超过 1 年前
Their &quot;Appendix: Memory Safe Languages&quot; lists:<p>C#, Go, Java, Python, Rust &amp; Swift
评论 #38612402 未加载
评论 #38612772 未加载
评论 #38612325 未加载
评论 #38612518 未加载
评论 #38612411 未加载
评论 #38612492 未加载
评论 #38614203 未加载
评论 #38612856 未加载
评论 #38612324 未加载
评论 #38612833 未加载
评论 #38612839 未加载
lambdaone超过 1 年前
&quot;Undefined behavior&quot; in general is a nightmare. After memory safety, the next target should be the enforcement of underflow&#x2F;overflow trapping. With the exception of the intentional use to implement modular arithmetic, underflow&#x2F;overflow should always be an error condition.
评论 #38612755 未加载
评论 #38613421 未加载
评论 #38612716 未加载
评论 #38613756 未加载
评论 #38614468 未加载
评论 #38612970 未加载
评论 #38613055 未加载
ianlevesque超过 1 年前
Encouraging. I hope the increasing drumbeat of awareness and advocacy around this reaches a critical mass soon. I&#x27;m not naive enough to think it will solve all security problems, but it will reduce the attack surface so, so much.
evrimoztamur超过 1 年前
Confused about Python in particular considering a lot of powerful and common dependencies in the ecosystem starting out from numpy and friends all have C&#x2F;++ components to them for performance improvements. Surely this is a vector to be considered?
评论 #38612851 未加载
评论 #38612891 未加载
userbinator超过 1 年前
I&#x27;m cynically thinking &quot;because they can now target the runtimes instead, and get far more value from their exploits&quot;.
评论 #38614970 未加载
champagnepapi超过 1 年前
But how are we going to trade stocks in nano seconds without C++?!
评论 #38612507 未加载
评论 #38612881 未加载
评论 #38612724 未加载
评论 #38612478 未加载
评论 #38612709 未加载
评论 #38612889 未加载
评论 #38612576 未加载
ryao超过 1 年前
It is a shame that there is no open source equivalent to tools like Astree:<p><a href="https:&#x2F;&#x2F;www.absint.com&#x2F;astree&#x2F;index.htm" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.absint.com&#x2F;astree&#x2F;index.htm</a><p>In theory, it is possible to write completely memory safe code in C&#x2F;C++ using it. As long as its analysis generates no complaints, there are no memory safety issues in the analyzed code.
评论 #38624170 未加载
评论 #38616558 未加载
hospitalJail超过 1 年前
At least my usecase has C completely disconnected from the internet.<p>I have no idea how this is going to work at the OS level.
评论 #38612333 未加载
评论 #38612228 未加载
评论 #38612487 未加载
评论 #38612527 未加载
评论 #38612197 未加载
DeathArrow超过 1 年前
Isn&#x27;t C++ with RAII reasonably safe? I tried to learn&#x2F;like Rust but it&#x27;s against how I use to think.<p>If no friendlier safe high speed programming language appears, I rather use C&#x2F;C++ and trade safety for friendliness.
评论 #38612843 未加载
评论 #38612791 未加载
评论 #38613124 未加载
评论 #38613277 未加载
评论 #38613834 未加载
ksec超过 1 年前
Access Denied<p>You don&#x27;t have permission to access &quot;<a href="http:&#x2F;&#x2F;www.nsa.gov&#x2F;Press-Room&#x2F;Press-Releases-Statements&#x2F;Press-Release-View&#x2F;Article&#x2F;3608324&#x2F;us-and-international-partners-issue-recommendations-to-secure-software-products&#x2F;" rel="nofollow noreferrer">http:&#x2F;&#x2F;www.nsa.gov&#x2F;Press-Room&#x2F;Press-Releases-Statements&#x2F;Pres...</a>&quot; on this server.<p>Judging from the 409 comments right now it seems I am the only one have this problem?
gosub100超过 1 年前
Can someone explain why we can&#x27;t double-down on C++ and, through compiler wizardry and reduction in toolset (say, strings can only be fixed-size at 32 chars, 64, or 256 long. No raw pointers, allocator zeroes out all freed memory), achieve a memory-safe language?<p>Obviously, making certain concessions would be a deal-breaker for some, but it might be viable for legacy codebases. If you were to try to make C++ memory-safe, where does it begin to break down?
评论 #38613720 未加载
评论 #38613094 未加载
评论 #38613385 未加载
评论 #38613246 未加载
评论 #38613348 未加载
评论 #38613553 未加载
jaisjdkaaaej超过 1 年前
I&#x27;m not moving languages over memory safety and I&#x27;m not going to apologize for it either. I don&#x27;t particularly like Rust and its not fit to replace C. If it replaces C++ I could care less and may even be somewhat supportive.
ram_rar超过 1 年前
Can someone with a security background enlighten me, on why Python is on the list of &quot;memory safe&quot; languages? Most of the python code I have worked with is a thin wrapper on C. Wouldnt that make python vulnerable as well?
评论 #38616736 未加载
1970-01-01超过 1 年前
Fully agree. The best time to stop giving the next generation of coders the gift of unsafe memory is today. C and C++ should be known as &#x27;legacy&#x27; languages.
评论 #38613678 未加载
rwmj超过 1 年前
Given the vast amount of C&#x2F;C++ around, much which will never be rewritten, I wonder if bounds checking compilers should be considered?
评论 #38613063 未加载
评论 #38622405 未加载
评论 #38614231 未加载
mikewarot超过 1 年前
More misdirection away from capability based security. 8(<p>I was hopeful we&#x27;d be transitioning by now.... it&#x27;s at least another decade out.
ahmedfromtunis超过 1 年前
How about a new drop-in replacement language for c&#x2F;c++?<p>I&#x27;m in no way an expert in any aspect that relates to this problem, but it&#x27;s reasonable to believe that it would be but a mildly challenging task if companies like Google, meta, Microsoft and the likes joined forces.<p>A year of concentrated efforts might be sufficient to rid us with this problem once and for all. Billions of lines of codes would gain (some) safety instantly.
评论 #38614188 未加载
评论 #38614403 未加载
评论 #38615742 未加载
Coder1996超过 1 年前
I&#x27;ve never thought of security being a reason to use memory-safe languages, but I agree.
lolive超过 1 年前
The truth about [Unix and] the C language:<p><a href="https:&#x2F;&#x2F;www-users.york.ac.uk&#x2F;~ss44&#x2F;joke&#x2F;c.htm" rel="nofollow noreferrer">https:&#x2F;&#x2F;www-users.york.ac.uk&#x2F;~ss44&#x2F;joke&#x2F;c.htm</a><p>#joke #hilarious
phkahler超过 1 年前
What is their stance on using C++ with sanitizers? Anyone know?
评论 #38614949 未加载
asylteltine超过 1 年前
Ah, they must have found an exploit in Rust then
评论 #38614069 未加载
评论 #38613549 未加载
jmclnx超过 1 年前
I advise training programmers instead of throwing them in front of a screen without any training. Companies these days provides no training at all. When I was hired over 40 years ago, I spent plenty of time being trained for my first 3 months.<p>Now, nothing, and you if you want to train a new person, you do it on your own time.
评论 #38613093 未加载
评论 #38612698 未加载
评论 #38612697 未加载
评论 #38613074 未加载
评论 #38612991 未加载
评论 #38612932 未加载
评论 #38613007 未加载
评论 #38612661 未加载
评论 #38614242 未加载
评论 #38616085 未加载
评论 #38613016 未加载
评论 #38613382 未加载
评论 #38613042 未加载
评论 #38612659 未加载
评论 #38612690 未加载
评论 #38613768 未加载
评论 #38614344 未加载
评论 #38613026 未加载
评论 #38613122 未加载
0x142857超过 1 年前
welcome to javascript
Woshiwuja超过 1 年前
Lets go back to C guys
guh超过 1 年前
Thanks but still sticking to C++...
pier25超过 1 年前
Correct me if I&#x27;m wrong... But don&#x27;t these types of memory attacks require local access to the machine?
评论 #38612771 未加载
评论 #38613458 未加载
评论 #38612702 未加载