> On Nov 17, 2017, at 5:34 PM, Daniel Stenberg <[email protected]> wrote:
> 
> Really hard to tell. I can't recall any bugs fixed recently like that.

Cool.  I wasn't suspecting libcurl --  I'm sure it's my bug... somewhere.

> 
>> I did notice a little bug in my code where I was setting CURLOPT_ERRORBUFFER 
>> to an allocated char * that happened to get free'd later on, so that's no 
>> good, but I can't decide if that would cause this crash in libcurl.
> 
> Hm, well if libcurl would then get an error it would write to the 
> CURLOPT_ERRORBUFFER area that is then freed and possibly reused for another 
> purpose and that would be really bad, but your stack trace doesn't really 
> look like it was a problem like that. Although we can't rule it out.

Yeah, that was definitely a bad bug on my part.  Which is why I still suspect 
me.

> 
>> I haven't been able to re-produce this for the past few days (even with the 
>> above bug in my code still around), so it's kinda hard to figure out if I've 
>> fixed it or not.  :)
> 
> Without more clues on how you trigger this crash, I'm not sure I can do much 
> more right now. :-/

That's cool.  I'll keep trying to re-create it.  I hate having segfaults and 
not being able to explain them.  :(

Thanks for your time!

eric
-------------------------------------------------------------------
Unsubscribe: https://cool.haxx.se/list/listinfo/curl-library
Etiquette:   https://curl.haxx.se/mail/etiquette.html

Reply via email to