Re: struct stat

David A. Greene (greened@eecs.umich.edu)
Fri, 12 Mar 1999 13:57:59 -0500

Albert D. Cahalan wrote:
>
> Matthew Wilcox writes:
>
> > I looked up Single Unix and it says...
> [...]
> > So it is quite clear that one can reasonably expect to compare st_dev
> > with the = operator, and therefore glibc is in error.
>
> No surprise there. They use a 56-bit major and a 8-bit minor too!
> They will need to redo everything for ino_t and others anyway.
>
> In your case, change __dev_t to __u32 (which you may define) and
> add padding every place that __dev_t was used. Assuming you are
> using a PC, add your padding after the __dev_t.

Ok, that's a reasonable fix for the short term. Except what happens
when are machines get upgraded again? I have to change things again...

> > [I don't know the address for libc-hackers; can someone forward this
> > please?]
>
> Tell them to wait for the kernel next time rather than trying to
> decide Linux data types all by themselves.

I don't know the list for the libc-hackers either, which is another
reason I posted to linux-kernel. This is a serious bug in libc and
needs to be fixed ASAP.

-Dave

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/