Re: [Tarantool-patches] [PATCH] backtrace: fix out of bounds access on backtrace printing

[Vladislav Shpilevoy <v.shpilevoy@tarantool.org>]

Hi! Thanks for the patch!

Perhaps the zero termination was done for a case,
when the cycle in backtrace() does not run even one
iteration. For example, if unw_step() returns an
error. Then the buffer is not terminated. So I think
it is better to keep *p = 0, but do it before the
cycle.

On 25/11/2019 17:27, Serge Petrenko wrote:
> snrpintf always null-terminates the passed string, and it also returns
> the number of bytes that "would have been written if there was enough
> space", so not only we don't have to null-terminate the string, but even
> more so we shouldn't do it erroneously.
> 
> Closes #4636
> ---
> https://github.com/tarantool/tarantool/issues/4636
> https://github.com/tarantool/tarantool/tree/sp/gh-4636-bt-print-fix
> 
>  src/lib/core/backtrace.cc | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/src/lib/core/backtrace.cc b/src/lib/core/backtrace.cc
> index 57e541c25..903ffb79c 100644
> --- a/src/lib/core/backtrace.cc
> +++ b/src/lib/core/backtrace.cc
> @@ -173,7 +173,6 @@ backtrace()
>  		say_debug("unwinding error: %i", unw_status);
>  #endif
>  out:
> -	*p = '\0';
>  	return backtrace_buf;
>  }
>  
>