Bug 3020 - bad define of pid_t in signal.h
Summary: bad define of pid_t in signal.h
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Red Hat Linux
Classification: Retired
Component: glibc
Version: 6.0
Hardware: i386
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Cristian Gafton
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 1999-05-24 22:00 UTC by vgough
Modified: 2008-05-01 15:37 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 1999-07-28 05:51:36 UTC
Embargoed:


Attachments (Terms of Use)

Description vgough 1999-05-24 22:00:49 UTC
signal.h, from glibc-devel-2.1.1-6, has this around line 60:

#if defined __USE_XOPEN && !defined pid_t
typedef __pid_t pid_t;
# define pid_t pid_t
#endif  /* Unix98 */


The "#define pid_t pid_t" causes an error "macro or
'#include' recursion too deep" when something tries to use
it (apparently with USE_XOPEN) is defined.  I noticed this
when trying to build CFS (cryptographic file system).

Doing a grep for "define pid_t pid_t" shows that it also
happens in termios.h and unistd.h...

Comment 1 vgough 1999-05-24 22:10:59 UTC
More things like this cause CFS compilation to break.  I found more
recursive pid_t defines in sys/stat.h , sys/types.h, and sys/wait.h.

And, there is a similar problem with ssize_t, in unistd.h and
sys/types.h.

This seems to happen all over the header files with various types.  Is
this a glibc header problem or a compiler problem?  Seems like a
poorly thought out header problem even if they were assuming the
compiler wouldn't recusively interpret defines...

Comment 2 Cristian Gafton 1999-07-28 05:51:59 UTC
What are the compile flags used by those programs? ANSI C certainly
does not forbid
	#define foo foo
syntax.


Note You need to log in before you can comment on or make changes to this bug.