[dpdk-dev] [PATCH] check-experimental-syms.sh: prevent symbol matches on substrings

Neil Horman nhorman at tuxdriver.com
Wed Oct 10 22:19:19 CEST 2018


On Wed, Oct 10, 2018 at 04:43:14PM +0200, Thomas Monjalon wrote:
> 10/10/2018 16:29, Neil Horman:
> > Thomas attempted to submit this:
> > https://patches.dpdk.org/patch/46311/
> > 
> > The other day, because the other patches being submitted with it were
> > breaking on a false positive from the check-experimental-syms check.
> > 
> > The problem was that the experimental symbol check script matched on the
> > regexs "\.text.*$SYM" and "\.text\.experimental.*$SYM" which allows for
> > substring matches, and librte_ethdev recently introduced symbols that
> > are leading substrings of one another (e.g. symbol foo is a substring of
> > symbol foobar), and so we would match on symbols when we shouldn't
> > 
> > Instead of dropping the check, fix this properly by matching
> > additionally on the end of line so that symbols are an exact match.
> > 
> > Confirmed to build properly on Thomas' submitted patch set with the
> > experimental check patch reverted (so that the checking actually
> > happens)
> 
> That's great Neil!
> I would like to push it now.
> May I suggest to remove the details of how I (baldly) reported it?
> 
> I suggest this text:
> "
> The experimental symbol check script matched on the regexes
> "\.text.*$SYM" and "\.text\.experimental.*$SYM" which allows for
> substring matches.
> If a symbol is leading substring of another one (e.g. symbol foo
> is a substring of symbol foobar), it would match on symbols
> when it shouldn't.
> 
> It is fixed by matching additionally on the end of line
> so that symbols are an exact match.
> "
Sure, if you want to make that change on commit, I'm ok with it.
Neil

> 
> 
> 
> 


More information about the dev mailing list