msync on Hurd, dev_t on MIPS

Russ Allbery rra at stanford.edu
Sun Nov 8 18:33:25 UTC 2009


Nix <nix at esperi.org.uk> writes:
> On 6 Nov 2009, Russ Allbery said:
>> Nix <nix at esperi.org.uk> writes:

>>> Not if they were using glibc, certainly; dev_t is defined in
>>> <bits/types.h> (included by <unistd.h>, of course), which is
>>> unconditionally installed and not dependent on architecture, nor on any
>>> of the __USE flags. So *everyone* should have this.

>>> What libc is this system using? Maybe it's using uClibc or something...

>> No, this would be glibc.  I'm not sure what the heck is going on with
>> this.

> gcc -E may be helpful here. Look for bits/types.h in the output, look at
> bits/types.h in the target headers (is this cross-compiling? probably...)
> and think?

Not cross-compiling -- all Debian builds are self-hosting.  But for some
reason h->st.st_dev = (dev_t)-1; causes a warning on that one platform and
not on any other Linux varient.

I wonder if it's actually a gcc warning bug.  Hm.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>



More information about the inn-workers mailing list