Re: Record comparison compiler warning - Mailing list pgsql-hackers

Bruce Momjian <bruce@momjian.us> wrote:
> On Wed, Oct 16, 2013 at 11:49:13AM -0700, Kevin Grittner wrote:

>> Bruce Momjian <bruce@momjian.us> wrote:
>>
>>> I am seeing this compiler warning in git head:
>>>
>>>     rowtypes.c: In function 'record_image_cmp':
>>>     rowtypes.c:1433: warning: 'cmpresult' may be used
>>>     uninitialized in this function rowtypes.c:1433: note: 'cmpresult' was declared here
>>
>> I had not gotten a warning under either gcc or clang, but that was
>> probably because I was doing assert-enabled builds, and the
>> Assert(false) saved me.  That seemed a little marginal anyway, so
>> how about this?:
>
> Would you please send the file as ASCII, e.g. not:
>
> <A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0><A0> default:

Huh, I did not see anything remotely like that in my email or in
the archives:

http://www.postgresql.org/message-id/1381949353.78943.YahooMailNeo@web162902.mail.bf1.yahoo.com

... however, I'm attaching it rather than pasting it this time.  It is warning-recimgcmp.diff.

I'm also attaching the other warnings (outside of the well-known
yyg parser warning) which I get from a compile using Ubuntu clang
version 3.0-6ubuntu3.  The one in pg_standby seems completely
legitimate -- what is the point of comparing to see whether an
unsigned integer is >= 0?  The other one would be nice to quiet
down, but I would understand if people would prefer to leave it
alone and hope the compiler writers fix it.  Warnings those fix
below:

catcache.c:571:12: warning: variable 'cl' is uninitialized when used here [-Wuninitialized]
                                Assert(cl->cl_magic == CL_MAGIC);
                                ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
../../../../src/include/c.h:625:10: note: expanded from:
                Trap(!(condition), "FailedAssertion")
                       ^
../../../../src/include/c.h:607:28: note: expanded from:
                if ((assert_enabled) && (condition)) \
                                         ^~~~~~~~~
catcache.c:569:5: note: variable 'cl' is declared here
                                CatCList   *cl = dlist_container(CatCList, cache_elem, iter.cur);
                                ^
catcache.c:585:13: warning: variable 'ct' is uninitialized when used here [-Wuninitialized]
                                        Assert(ct->ct_magic == CT_MAGIC);
                                        ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
../../../../src/include/c.h:625:10: note: expanded from:
                Trap(!(condition), "FailedAssertion")
                       ^
../../../../src/include/c.h:607:28: note: expanded from:
                if ((assert_enabled) && (condition)) \
                                         ^~~~~~~~~
catcache.c:583:6: note: variable 'ct' is declared here
                                        CatCTup    *ct = dlist_container(CatCTup, cache_elem, iter.cur);
                                        ^
2 warnings generated.
pg_standby.c:348:22: warning: comparison of unsigned expression >= 0 is always true [-Wtautological-compare]
                if (tli > 0 && log >= 0 && seg > 0)
                               ~~~ ^  ~
1 warning generated.

None of these seem particularly urgent, so I'll leave them for
comment for maybe a week before taking any action.

--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company