Re: breakage still in sys/systm.h

2000-03-24 Thread Brian Fundakowski Feldman

On Fri, 24 Mar 2000, Steve Kiernan wrote:

 The definitions of major() and minor() in sys/systm.h break usage of the
 header.  Since sys/types.h defines major() and minor() as macros which
 compute the major and minor numbers, this creates an order dependency on
 sys/systm.h and sys/types.h.  Is this not a bad thing?

The sys/types.h header is meant to be included in userland code; the
sys/systm.h header is not to be included from outside of kernel code.
What possible reason would you have for systm.h in userland code?

 --
 Stephen Kiernan
 [EMAIL PROTECTED]
 NAI Labs, A Division of Network Associates, Inc.

--
 Brian Fundakowski Feldman   \  FreeBSD: The Power to Serve!  /
 [EMAIL PROTECTED]`--'



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: breakage still in sys/systm.h

2000-03-24 Thread Steve Kiernan

On Fri, 24 Mar 2000, Garrett Wollman wrote:

 On Fri, 24 Mar 2000 17:12:32 -0500 (EST), Steve Kiernan [EMAIL PROTECTED] said:
 
  The definitions of major() and minor() in sys/systm.h break usage of the
  header.  Since sys/types.h defines major() and minor() as macros which
  compute the major and minor numbers, this creates an order dependency on
  sys/systm.h and sys/types.h.  Is this not a bad thing?
 
 No, since they don't conflict.  sys/types.h defines the major and
 minor macros iff _KERNEL is not defined, and sys/systm.h is a

Ah, okay, I see what the problem is with my filesystem driver.  Looks like
the #define switched from KERNEL to _KERNEL from 3.x to -CURRENT.

Thanks.

--
Stephen Kiernan
[EMAIL PROTECTED]
NAI Labs, A Division of Network Associates, Inc.



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message