Re: Error in `git': free(): invalid pointer (was Re: [PATCH] sequencer: improve config handling)
- From
- phillip.wood@talktalk.net <phillip.wood@talktalk.net>
- Date
- Dec 22, 2017, 10:47 UTC
- Message-ID
- <23359630.1065831513939659692.JavaMail.defaultUser@defaultHost>
- In-Reply-To
- <1513876450.9646.6.camel@gmail.com>
>----Original Message---- >From: kaartic.sivaraam@gmail.com >Date: 21/12/2017 17:14 >To: "phillip.wood@talktalk.net"<phillip.wood@talktalk.net>, "Phillip
Wood"<phillip.wood@dunelm.org.uk>, "Git Mailing List"<git@vger.kernel. org>
>Cc: "Johannes Schindelin"<Johannes.Schindelin@gmx.de> >Subj: Re: Error in `git': free(): invalid pointer (was Re: [PATCH]
sequencer: improve config handling)
Show 12 quoted lines
> >On Thu, 2017-12-21 at 16:53 +0000, phillip.wood@talktalk.net wrote: >> Hm, There is a problem with sequencer_remove_state() which does >> >> free(opts->gpg_sign) >> >> however unless a gpg key was given on the commandline, opts->gpg is >> initialized to "" which is statically allocated. >> >> This untested diff should fix that, > >It did seem to. I don't get that error any more.
That's good, thanks for testing it
Show 5 quoted lines
>> but I'm not sure if you're problem >> is related to it > >As the fix you suggested did work, I think the problem is related. Did >you have anything else in mind when you said you're not sure whether
or
>not it's related?
I didn't have anything else in mind, but at that point I hadn't noticed that one of your previous messages said you had gpg signing turned on - as you do it makes sense that the patch fixed it.
Thanks again for reporting this and testing the fix - it seems the test suite has a bit of a blind spot when it comes to gpg signing, I guess it's difficult to set up a key for tests
I'll send a proper patch when I have time in a few days
Best Wishes
Phillip