[LLVMbugs] [Bug 18644] New: Problem with tenary operator and errno: although condition is true it evaluates the else part

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=18644

            Bug ID: 18644
           Summary: Problem with tenary operator and errno: although
                    condition is true it evaluates the else part
           Product: clang
           Version: 3.4
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: christian at 2ndquadrant.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

Hi,

while working on a patch for PostgreSQL I noticed that a error hint was not
displayed. The hint has been set via a ternary operator and a comparison of
errno:

            ereport(FATAL,
                    (errmsg("could not map anonymous shared memory: %m"),
                     (errno == ENOMEM) ?
                     errhint("This error usually means that PostgreSQL's
request "
                             "for a shared memory segment exceeded available
memory "
                             "or swap space. To reduce the request size
(currently "
                             "%zu bytes), reduce PostgreSQL's shared memory
usage, "
                             "perhaps by reducing shared_buffers or "
                             "max_connections.",
                             *size) : 0));

In this case (checked with GDB) is errno == ENOMEM, but the hint wont appear.
The same code works fine with GCC.

clang version used:

clang version 3.4 (tags/RELEASE_34/final)
Target: x86_64-pc-linux-gnu
Thread model: posix

gcc version used:

gcc (Gentoo 4.7.3-r1 p1.4, pie-0.5.5) 4.7.3
Copyright (C) 2012 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

I wasn't able to create a small test case, but I uploaded my current working
tree so you can check it out:

http://www.defunced.de/postgres-working-tree.tgz

The compiler flags I used are in both cases (gcc and clang):

-O0 -g -W -Wall -ansi -pedantic -pipe -std=c99 -Wall -Wmissing-prototypes
-Wpointer-arith -Wdeclaration-after-statement -Wendif-labels
-Wmissing-format-attribute -Wformat-security -fno-strict-aliasing -fwrapv
-fexcess-precision=standard

I configured with this statement for clang:

CFLAGS="-O0 -g -W -Wall -ansi -pedantic -pipe -std=c99" CC=/usr/bin/clang
./configure --prefix=/home/ckruse/dev/postgres/

and with this for GCC:

CFLAGS="-O0 -g -W -Wall -ansi -pedantic -pipe -std=c99" ./configure
--prefix=/home/ckruse/dev/postgres/

Best regards,

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

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20140128/294fe899/attachment.html>