Re: Unitialized Values in record_image_cmp

On 2013-10-17 08:16:58 -0500, Merlin Moncure wrote:
> On Thu, Oct 17, 2013 at 8:01 AM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
> > Hi,
> >
> > Valgrind tells me:
> > ==442828== Conditional jump or move depends on uninitialised value(s)
> > ==442828== at 0x80084F: record_image_cmp (rowtypes.c:1521)
> > ==442828== by 0x801522: btrecordimagecmp (rowtypes.c:1839)
> > ==442828== by 0x8C6604: comparison_shim (sortsupport.c:53)
> > ==442828== by 0x64284D: ApplySortComparator (sortsupport.h:143)
> > ==442828== by 0x642F3B: MJCompare (nodeMergejoin.c:412)
> > ==442828== by 0x643A29: ExecMergeJoin (nodeMergejoin.c:1211)
> > ==442828== by 0x62305A: ExecProcNode (execProcnode.c:453)
> > ==442828== by 0x640800: ExecLimit (nodeLimit.c:91)
> > ==442828== by 0x623135: ExecProcNode (execProcnode.c:500)
> > ==442828== by 0x620E15: ExecutePlan (execMain.c:1472)
> > ==442828== by 0x61EF94: standard_ExecutorRun (execMain.c:307)
> > ==442828== by 0x61EE0B: ExecutorRun (execMain.c:255)
> >
>
> I think Kevin got to it already. see:
> http://postgresql.1045698.n5.nabble.com/Record-comparison-compiler-warning-td5774794.html

At least that patch isn't applicable - there was an unitialized value at
runtime, even using a cassert enabled build.

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services