Re: Compile warnings
Donald Becker <[email protected]>
| Newsgroups | gmane.linux.drivers.vortex.devel |
|---|---|
| Message-ID | <[email protected]> |
On Fri, 7 Mar 2003, Joost wrote: > Is the reason that noone is answering just that nobody knows, or is there > something wrong with my question? I was traveling and thus unable to track down the problem. Although the mailing list is intended to provide a community resource, I'm usually the one answering the questions. > > gcc -I/usr/src/kernel-headers-2.2.20/include/ -DMODULE -D__KERNEL__ -O6 -c > > 3c59x.c OK, the 2.2.20 kernel. Is this from kernel.org, or a vendor-modified kernel? > > In file included from 3c59x.c:134: > > kern_compat.h:216: warning: `DECLARE_MUTEX' redefined > > /usr/src/kernel-headers-2.2.20/include/asm/semaphore.h:46: warning: this > is > > the location of the previous definition This was introduced in one of the 2.3.99 kernels. The drivers were updated to use the 2.4 interface, and use put backwards-compatibility code in the kern_compat.h file. This problem comes when the interface is back-ported to the 2.2 kernel. I must update the previously-correct 2.2 interface to take into account the new definition. Please try changing line 215 from #if LINUX_VERSION_CODE < 0x20363 to #if LINUX_VERSION_CODE < 0x20214 > > kern_compat.h:224: warning: `time_before' redefined > > /usr/src/kernel-headers-2.2.20/include/linux/timer.h:91: warning: this is > > the location of the previous definition There is a similar new definition of time_before() in 2.2.<recent>. However note that kernel's definition of time_before() is just the invert of time_after(), which is incorrect. #define time_before(a,b) ((long)(a) - (long)(b) < 0) -- Donald Becker [email protected] Scyld Computing Corporation http://www.scyld.com 410 Severn Ave. Suite 210 Scyld Beowulf cluster system Annapolis MD 21403 410-990-9993 _______________________________________________ vortex mailing list [email protected] http://www.scyld.com/mailman/listinfo/vortex