Hi, On 23 October 2017 at 08:53, Martin Richtarsky wrote: > Hi, > > I've written a small blog post about the stack protector: > > http://www.productive-cpp.com/hardening-cpp-programs-stack-protector/ > > While playing around with the features, I noticed that they do not behave > exactly as documented: > > https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html > > "-fstack-protector" > > Emit extra code to check for buffer overflows, such as stack smashing > attacks. This is done by adding a guard variable to functions with > vulnerable objects. This includes functions that call alloca, and > _functions with buffers larger than 8 bytes_. The guards are initialized > when a function is entered and then checked when the function exits. If a > guard check fails, an error message is printed and the program exits. > > "with buffers larger than 8 bytes" - here I found that a buffer of exactly > 8 bytes is also instrumented. > > The documentation only talks about buffers. This might be mistaken for > arrays in general. It might be clearer to state that only char arrays are > instrumented (and not e.g. wchar!) I don't know enough about stack protector to say but I assume your findings are right, and so these seem like good suggestions to improve the docs. > I can also make the appropriate changes if you can point me to a page > describing the process. The process of contributing a patch is described at https://gcc.gnu.org/contribute.html I don't think the two tweaks you're proposing would be legally significant, so would probably not require a copyright assignment (although if you plan to contribute anything else to GCC in future I would encourage you to complete the assignment process :-) The sources for those docs are the file gcc/doc/invoke.texi so once you have a patch you would send it to the gcc-patches mailing list for review and to be committed to source control.
