Re: -mm merge plans for 2.6.26

From: Randy Dunlap
Date: Sat May 17 2008 - 00:31:47 EST

Randy Dunlap wrote:
Andrew Morton wrote:
On Fri, 16 May 2008 17:03:10 -0700 Randy Dunlap <randy.dunlap@xxxxxxxxxx> wrote:

On Sun, 20 Apr 2008 07:20:53 -0700 Andrew Morton wrote:

These can be found at

# documentation

What happened to/with these?

I was going to ask you that. Last I heard,

a) stuff broke and

b) you had an updated version somewhere.

Oh yeah. I think that this was fixed by;a=commit;h=b11caa7c7063ea92a0a58115d3fc6d038ed89510

but I'll check/verify/confirm that.

Yes, it builds fine now (built against 2.6.26-rc2-git5).

On Thu, 10 Apr 2008 17:56:30 -0700 Randy Dunlap <randy.dunlap@xxxxxxxxxx> wrote:

On Wed, 9 Apr 2008 14:54:09 -0700 Andrew Morton wrote:

make allmodconfig
gcc -I$(/bin/pwd)/include Documentation/accounting/getdelays.c -o getdelays

In file included from /usr/src/devel/include/linux/netlink.h:5,
from /usr/src/devel/include/linux/genetlink.h:4,
from Documentation/accounting/getdelays.c:26:
/usr/src/devel/include/linux/types.h:203: error: expected specifier-qualifier-list before '__kernel_daddr_t'

Those __kernel_* types shouldn't be used outside of the #ifdef __KERNEL__
block, should they?

Patch below fixes kernel side for me. Don't have any idea what it
may do to userspace users of the header file.

include/linux/types.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

--- mmotm-2008-0410-0157.orig/include/linux/types.h
+++ mmotm-2008-0410-0157/include/linux/types.h
@@ -200,8 +200,8 @@ typedef u32 resource_size_t;
#endif /* __KERNEL__ */
struct ustat {
- __kernel_daddr_t f_tfree;
- __kernel_ino_t f_tinode;
+ daddr_t f_tfree;
+ ino_t f_tinode;
char f_fname[6];
char f_fpack[6];

I didn't apply that. I guess the "Don't have any idea" bit was scary.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at