Re: bk commit into 5.1 tree (davi:1.2516) BUG#31891

[Davi Arnaut]

Sergei Golubchik wrote:
> Hi!
> 
> On Feb 07, Davi Arnaut wrote:
>> Sergei Golubchik wrote:
> 
>>>> @@ -2041,7 +2041,7 @@ AC_CHECK_FUNCS(alarm bcmp bfill bmove bs
>>>>    sighold sigset sigthreadmask port_create sleep \
>>>>    snprintf socket stpcpy strcasecmp strerror strsignal strnlen strpbrk
> strstr \
>>>>    strtol strtoll strtoul strtoull tell tempnam thr_setconcurrency
> vidattr \
>>>> -  posix_fallocate)
>>>> +  posix_fallocate backtrace backtrace_symbols backtrace_symbols_fd)
>>> What's the point in checking for backtrace_symbols and
>>> backtrace_symbols_fd if you never test the corresponding HAVE_ defines ?
>> Good point, I'll cover both under their respective HAVE_ defines.
> 
> Only if you think it's reasonable to expect a system where backtrace()
> is present but for example backtrace_symbols_fd() is not.
> 
> Otherwise testing for backtrace() only is enough.

backtrace() is the function used to obtain the backtrace.
backtrace_symbols() and backtrace_symbols_fd() are used to translate the
information obtained by backtrace() and backtrace_symbols_fd() was added
to the libc after backtrace_symbols().

Not that it's really important, but since it doesn't hurt...

Regards,

-- 
Davi Arnaut, Software Engineer
MySQL Inc, www.mysql.com

Are you MySQL certified?  www.mysql.com/certification