Re: Top-of-tree GCC/glibc build problem

[Martin Sebor <msebor at gmail dot com>]

On 10/7/19 10:05 AM, Florian Weimer wrote:
> * Steve Ellcey:
>
> > Looks like we have another new warning that is blocking top-of-tree glibc
> > from being built with top-of-tree gcc.  Would using 'cp[0]' instead of '*cp'
> > be the right fix here?
> >
> > Steve Ellcey
> > sellcey@marvell.com
> >
> >
> > zic.c: In function ‘inzsub’:
> > zic.c:1369:23: error: writing 8 bytes into a region of size 1 [-Werror=stringop-overflow=]
> >   1369 |  z.z_format_specifier = cp ? *cp : '\0';
> >        |  ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
> > zic.c:124:8: note: destination object declared here
> >    124 |  char  z_format_specifier;
> >        |        ^~~~~~~~~~~~~~~~~~
> > cc1: all warnings being treated as errors
> > ../o-iterator.mk:9: recipe for target '/home/sellcey/tot/obj-glibc/timezone/zic.o' failed
>
> This looks like a GCC bug to me.  The code is okay: no aliases,
> zero-length arrays, or anything like that.  The store has size 1, too.
>
> Is it possible that the stringop-overflow diagnostic pass runs after
> some sort of store-merging pass?  In this case, many of the diagnostics
> would need to be suppressed.

It's a bug introduced in last week's change to the warning.  Jeff's
build bot caught it over the weekend.  I should have a fix soon.

Martin