[Bug preprocessor/16192] New: Bug in expression evaluation when operand is missing

[Bug preprocessor/16192] New: Bug in expression evaluation when operand is missing

[gnu at behdad dot org]

Seems like there's a bug in evaluating expressions in #if statements where one
of the operands is missing.  Seems like a value from the top of the stack is
used instead...

Here is the minimal code that goes wrong:

#if (1 < 2) && ( == 2)
JUNK JUNK JUNK
#endif

and it outputs "JUNK JUNK JUNK" while should not.  If you change one of the "2"s
to something else, it doesn't happen; that's why I guess it's just taking the
missing operand from the top of the stack...

Needless to say, something like this fails for the same reason (which was were I
found the bug):

#if (x < 2) && (y == 2)
do something
#endif

-- 
           Summary: Bug in expression evaluation when operand is missing
           Product: gcc
           Version: 3.3.3
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: preprocessor
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: gnu at behdad dot org
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i386-redhat-linux
  GCC host triplet: i386-redhat-linux
GCC target triplet: i386-redhat-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-06-25 06:04 -------
Nope sorry but you are wrong "(x < 2) && (y == 2)" is defined as one by the C standard.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[gnu at behdad dot org]

------- Additional Comments From gnu at behdad dot org  2004-06-25 06:35 -------
Would you please enlighten me why it's defined as one?

By the way, I still think there's a bug in gcc somewhere, either in
preprocessor, or in documentation.  Because in the texinfo documentation I read:

   * Identifiers that are not macros, which are all considered to be the
     number zero.

Which immediately results that "(x < 2> && (y == 2)" should fail.

Moreover, I also read:

   If EXPRESSION is not correctly formed, GCC issues an error and
treats the conditional as having failed.

Which again means that "(1 < 2) && ( == 2)" should fail.

To my not so humble opinion, these don't all make sense at the same time.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[zack at gcc dot gnu dot org]

------- Additional Comments From zack at gcc dot gnu dot org  2004-06-25 17:47 -------
The reporter is correct, gcc is misbehaving here - more by the absence of a
diagnostic than by the precise contents of the output, but it *is* documented to
treat an ill-formed #if as having failed.  And I have no idea where you got the
idea that (x < 2) && (y == 2) should succeed, where 'x' and 'y' are both
non-macros - that expands to (0 < 2) && (0 == 2) which is obviously false.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |accepts-invalid
   Last reconfirmed|0000-00-00 00:00:00         |2004-06-25 17:47:09
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[zack at gcc dot gnu dot org]

------- Additional Comments From zack at gcc dot gnu dot org  2004-06-25 17:47 -------
*growl* Reopening ...

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[zack at gcc dot gnu dot org]

------- Additional Comments From zack at gcc dot gnu dot org  2004-06-25 17:48 -------
... confirming, assigning to self.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |zack at gcc dot gnu dot org
                   |dot org                     |
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|                            |1
   Last reconfirmed|2004-06-25 17:47:09         |2004-06-25 17:48:16
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[zack at gcc dot gnu dot org]

------- Additional Comments From zack at gcc dot gnu dot org  2004-06-25 17:48 -------
It should be noted that I won't have a chance to work on this for some time.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[neil at gcc dot gnu dot org]

------- Additional Comments From neil at gcc dot gnu dot org  2004-06-25 17:54 -------
I'll pick this up.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|zack at gcc dot gnu dot org |neil at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[gnu at behdad dot org]

------- Additional Comments From gnu at behdad dot org  2004-06-26 18:03 -------
A bit of misinformation from my side.  "(x < 2) && (y == 2)" indeed works as
accepted if x and y are non-macros.  But the ill-formed case "(1 < 2) && ( ==
2)" should still fail.

More information, "(2 == )" fails with error.  "( == 2)" fails with no error. 
"(1< 2) && (2 == )" fails with error, and "(1 < 2) && ( == 2)" does not fail. 
In theory all of these four should fail with error.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] Bug in expression evaluation when operand is missing

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-06-28 20:28 -------
Subject: Bug 16192

CVSROOT:	/cvs/gcc
Module name:	gcc
Branch: 	gcc-3_4-branch
Changes by:	neil@gcc.gnu.org	2004-06-28 20:28:07

Modified files:
	gcc            : cppinit.c cppexp.c 
	gcc/doc        : cpp.texi cppopts.texi 
	gcc/testsuite/gcc.dg/cpp: if-mop.c 
Added files:
	gcc/testsuite/gcc.dg/cpp/trad: comment-3.c 

Log message:
	PR preprocessor/16192
	PR preprocessor/15913
	PR preprocessor/15572
	* cppexp.c (_cpp_parse_expr): Handle remaining cases where an
	expression is missing.
	* cppinit.c (post_options): Traditional cpp doesn't do // comments.
	* doc/cpp.texi: Don't document what we do for ill-formed expressions.
	* doc/cppopts.texi: Clarify processing of command-line defines.
	
	testsuite:
	* gcc.dg/cpp/if-mop.c: Two new testcases.
	* gcc.dg/cpp/trad/comment-3.c: New.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cppinit.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.297.6.5&r2=1.297.6.6
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cppexp.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.144.6.1&r2=1.144.6.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/doc/cpp.texi.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.65.4.2&r2=1.65.4.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/doc/cppopts.texi.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.28.4.3&r2=1.28.4.4
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/cpp/if-mop.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.3&r2=1.3.54.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/cpp/trad/comment-3.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=NONE&r2=1.1.2.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] [3.5 only] Bug in expression evaluation when operand is missing

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-06-28 20:41 -------
Fixed in 3.4.1 and Neil is testing for 3.5.0 but is having bootstrap problems even without the patch.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|                            |3.4.1
            Summary|Bug in expression evaluation|[3.5 only] Bug in expression
                   |when operand is missing     |evaluation when operand is
                   |                            |missing
   Target Milestone|---                         |3.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] [3.5 only] Bug in expression evaluation when operand is missing

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-07-04 12:57 -------
Subject: Bug 16192

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	neil@gcc.gnu.org	2004-07-04 12:57:51

Modified files:
	gcc            : ChangeLog 
	libcpp         : ChangeLog expr.c init.c 
	gcc/doc        : cpp.texi cppopts.texi 
	gcc/testsuite  : ChangeLog 
	gcc/testsuite/gcc.dg/cpp: if-mop.c 
Added files:
	gcc/testsuite/gcc.dg/cpp/trad: comment-3.c 

Log message:
	* doc/cpp.texi: Don't document what we do for ill-formed expressions.
	* doc/cppopts.texi: Clarify processing of command-line defines.
	
	libcpp:
	PR preprocessor/16192
	PR preprocessor/15913
	PR preprocessor/15572
	* expr.c (_cpp_parse_expr): Handle remaining cases where an
	expression is missing.
	* init.c (post_options): Traditional cpp doesn't do // comments.
	
	testsuite:
	* gcc.dg/cpp/if-mop.c: Two new testcases.
	* gcc.dg/cpp/trad/comment-3.c: New.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.4281&r2=2.4282
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libcpp/ChangeLog.diff?cvsroot=gcc&r1=1.16&r2=1.17
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libcpp/expr.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libcpp/init.c.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/doc/cpp.texi.diff?cvsroot=gcc&r1=1.69&r2=1.70
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/doc/cppopts.texi.diff?cvsroot=gcc&r1=1.34&r2=1.35
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.3936&r2=1.3937
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/cpp/if-mop.c.diff?cvsroot=gcc&r1=1.3&r2=1.4
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/cpp/trad/comment-3.c.diff?cvsroot=gcc&r1=1.1&r2=1.2



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] [3.5 only] Bug in expression evaluation when operand is missing

[neil at gcc dot gnu dot org]

------- Additional Comments From neil at gcc dot gnu dot org  2004-07-04 12:58 -------
Fixed in 3.4 and 3.5.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192

[Bug preprocessor/16192] [4.0 Regression] Bug in expression evaluation when operand is missing

[giovannibajo at libero dot it]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[3.5 only] Bug in expression|[4.0 Regression] Bug in
                   |evaluation when operand is  |expression evaluation when
                   |missing                     |operand is missing
   Target Milestone|4.0.0                       |3.4.1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16192