[Bug preprocessor/59805] New: invalid preprocessing directive not diagnosed with assembler-with-cpp

[Bug preprocessor/59805] New: invalid preprocessing directive not diagnosed with assembler-with-cpp

[aldot at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59805

            Bug ID: 59805
           Summary: invalid preprocessing directive not diagnosed with
                    assembler-with-cpp
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Keywords: diagnostic
          Severity: normal
          Priority: P3
         Component: preprocessor
          Assignee: unassigned at gcc dot gnu.org
          Reporter: aldot at gcc dot gnu.org
                CC: tromey at redhat dot com

-x assembler-with-cpp remains silent instead of emitting some kind of
diagnostics.

$ cat libcpp-bug.c
# INCLUDE <./does-not-exist.HHH>
# HUH <./does-not-exist.HHH>
$ gcc -x assembler-with-cpp -o xxx.o -c libcpp-bug.c -W -Wall -pedantic -Wextra 

Properly diagnosed with c or c-header:

$ gcc -x c -o xxx.o -c libcpp-bug.c -W -Wall -pedantic -Wextra 
libcpp-bug.c:1:3: error: invalid preprocessing directive #INCLUDE
 # INCLUDE <./does-not-exist.HHH>
   ^
libcpp-bug.c:2:3: error: invalid preprocessing directive #HUH
 # HUH <./does-not-exist.HHH>
   ^
libcpp-bug.c:2:0: warning: ISO C forbids an empty translation unit [-Wpedantic]
 # HUH <./does-not-exist.HHH>
 ^

gcc-4.9-trunk@206144

Since the #INCLUDE was not processed this missing diagnostics resulted in wrong
code generated, but adding that keyword.
Didn't look if this is a driver bug.

[Bug preprocessor/59805] invalid preprocessing directive not diagnosed with assembler-with-cpp

[aldot at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59805

Bernhard Reutner-Fischer <aldot at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |joseph at codesourcery dot com

--- Comment #1 from Bernhard Reutner-Fischer <aldot at gcc dot gnu.org> ---
CCing jsm.

I am undecided if this is a valid request after all.

On one hand this is not a valid preprocessing token, on the other hand this
behaviour might be ok with respect to ASM_COMMENT_START.

Thoughts?