lkml.org 
[lkml]   [2010]   [Jul]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [-next July 21] s390 build failure : kernel/workqueue.o
On Wed, 21 Jul 2010 17:57:09 +0200
Tejun Heo <tj@kernel.org> wrote:

> Hello,
>
> On 07/21/2010 12:40 PM, Sachin Sant wrote:
> > Today's next fails to build on a s390 box with
> >
> > CC kernel/workqueue.o
> > kernel/workqueue.c: In function 'init_workqueues':
> > kernel/workqueue.c:3525: error: negative width in bit-field '<anonymous>'
> > make[1]: *** [kernel/workqueue.o] Error 1
> >
> > Yesterday's next was good.
> > The following patches added/changed the BUILD_BUG_ON() statement.
> >
> > bdbc5dd7de5d07d6c9d3536e598956165a031d4c
> > workqueue: prepare for WQ_UNBOUND implementation
> > 7a22ad757ec75186ad43a5b4670fa7423ee8f480
> > workqueue: carry cpu number in work data once execution starts
> >
> > I have gcc version 4.3.2 installed on the system.
>
> Hmmm... that's surprising. Can you please attach the .config?

The BUILD_BUG_ON line gets expanded to:

((void)(sizeof(struct { int:-!!(WORK_CPU_LAST << WORK_STRUCT_FLAG_BITS >= 0x0UL); })));
WORK_CPU_LAST is:

enum {
..
WORK_CPU_NONE = 32 + 1,
WORK_CPU_LAST = WORK_CPU_NONE,
..
};
WORK_STRUCT_FLAG_BITS is:

WORK_STRUCT_COLOR_SHIFT = 2,
WORK_STRUCT_COLOR_BITS = 4,
WORK_STRUCT_FLAG_BITS = WORK_STRUCT_COLOR_SHIFT +
WORK_STRUCT_COLOR_BITS,

which boils it down to

((void)(sizeof(struct { int:-!!(33 << 6 >= 0x0UL); })));
gcc doesn't like this construct.

--
blue skies,
Martin.

"Reality continues to ruin my life." - Calvin.



\
 
 \ /
  Last update: 2010-07-22 11:29    [from the cache]
©2003-2011 Jasper Spaans