On Tue, 2006-11-28 at 14:30 -0800, Greg KH wrote:
> On Tue, Nov 28, 2006 at 12:35:43PM +0100, Mariusz Kozlowski wrote:
> > Hello,
> >
> > When CONFIG_MODULE_UNLOAD is not set then this happens:
> >
> > CC kernel/module.o
> > kernel/module.c:852: error: `initstate' undeclared here (not in a function)
> > kernel/module.c:852: error: initializer element is not constant
> > kernel/module.c:852: error: (near initialization for `modinfo_attrs[2]')
> > make[1]: *** [kernel/module.o] Error 1
> > make: *** [kernel] Error 2
> >
> > Reference to 'initstate' should stay under #ifdef CONFIG_MODULE_UNLOAD
> > as its definition I guess.
> >
> > Signed-off-by: Mariusz Kozlowski <[email protected]>
> >
> > --- linux-2.6.19-rc6-mm2-a/kernel/module.c 2006-11-28 12:17:09.000000000 +0100
> > +++ linux-2.6.19-rc6-mm2-b/kernel/module.c 2006-11-28 12:05:01.000000000 +0100
> > @@ -849,8 +849,8 @@ static inline void module_unload_init(st
> > static struct module_attribute *modinfo_attrs[] = {
> > &modinfo_version,
> > &modinfo_srcversion,
> > - &initstate,
> > #ifdef CONFIG_MODULE_UNLOAD
> > + &initstate,
> > &refcnt,
> > #endif
>
> Kay, is this correct? I think we still need this information exported
> to userspace, even if we can't unload modules, right?

Yes, instead we should move the attribute out of the ifdef, so
it will be there, even when modules can't be unloaded.

Thanks,
Kay

Attachments:

modules-state.patch (917.00 B)

2006-11-29 10:03:57

by Adrian Bunk

[permalink] [raw]

Subject: [-mm patch] #if 0 fs/gfs2/acl.c:gfs2_check_acl()

On Tue, Nov 28, 2006 at 02:02:46AM -0800, Andrew Morton wrote:
>...
> Changes since 2.6.19-rc6-mm1:
>...
> git-gfs2-nmw.patch
>...
> git trees
>...

This patch #if 0's the no longer used gfs2_check_acl().

Signed-off-by: Adrian Bunk <[email protected]>

---

fs/gfs2/acl.c | 2 ++
fs/gfs2/acl.h | 1 -
2 files changed, 2 insertions(+), 1 deletion(-)

--- linux-2.6.19-rc6-mm2/fs/gfs2/acl.h.old 2006-11-29 08:49:13.000000000 +0100
+++ linux-2.6.19-rc6-mm2/fs/gfs2/acl.h 2006-11-29 08:49:22.000000000 +0100
@@ -32,7 +32,6 @@
int *remove, mode_t *mode);
int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access);
int gfs2_check_acl_locked(struct inode *inode, int mask);
-int gfs2_check_acl(struct inode *inode, int mask);
int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip);
int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr);

--- linux-2.6.19-rc6-mm2/fs/gfs2/acl.c.old 2006-11-29 08:49:31.000000000 +0100
+++ linux-2.6.19-rc6-mm2/fs/gfs2/acl.c 2006-11-29 08:49:45.000000000 +0100
@@ -170,6 +170,7 @@
return -EAGAIN;
}

+#if 0
int gfs2_check_acl(struct inode *inode, int mask)
{
struct gfs2_inode *ip = GFS2_I(inode);
@@ -184,6 +185,7 @@

return error;
}
+#endif /* 0 */

static int munge_mode(struct gfs2_inode *ip, mode_t mode)
{

2006-11-29 10:15:03

by Steven Whitehouse

[permalink] [raw]

Subject: Re: [-mm patch] #if 0 fs/gfs2/acl.c:gfs2_check_acl()

Hi,

A better solution is just to remove it I think, so thats what I'll do in
my git tree. Thanks for pointing it out,

Steve.

On Wed, 2006-11-29 at 11:04 +0100, Adrian Bunk wrote:
> On Tue, Nov 28, 2006 at 02:02:46AM -0800, Andrew Morton wrote:
> >...
> > Changes since 2.6.19-rc6-mm1:
> >...
> > git-gfs2-nmw.patch
> >...
> > git trees
> >...
>
>
> This patch #if 0's the no longer used gfs2_check_acl().
>
> Signed-off-by: Adrian Bunk <[email protected]>
>
> ---
>
> fs/gfs2/acl.c | 2 ++
> fs/gfs2/acl.h | 1 -
> 2 files changed, 2 insertions(+), 1 deletion(-)
>
> --- linux-2.6.19-rc6-mm2/fs/gfs2/acl.h.old 2006-11-29 08:49:13.000000000 +0100
> +++ linux-2.6.19-rc6-mm2/fs/gfs2/acl.h 2006-11-29 08:49:22.000000000 +0100
> @@ -32,7 +32,6 @@
> int *remove, mode_t *mode);
> int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access);
> int gfs2_check_acl_locked(struct inode *inode, int mask);
> -int gfs2_check_acl(struct inode *inode, int mask);
> int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip);
> int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr);
>
> --- linux-2.6.19-rc6-mm2/fs/gfs2/acl.c.old 2006-11-29 08:49:31.000000000 +0100
> +++ linux-2.6.19-rc6-mm2/fs/gfs2/acl.c 2006-11-29 08:49:45.000000000 +0100
> @@ -170,6 +170,7 @@
> return -EAGAIN;
> }
>
> +#if 0
> int gfs2_check_acl(struct inode *inode, int mask)
> {
> struct gfs2_inode *ip = GFS2_I(inode);
> @@ -184,6 +185,7 @@
>
> return error;
> }
> +#endif /* 0 */
>
> static int munge_mode(struct gfs2_inode *ip, mode_t mode)
> {
>

2006-11-29 12:43:18

by Jiri Kosina

[permalink] [raw]

Subject: [PATCH] compile fix on x86 without X86_LOCAL_APIC (was 2.6.19-rc6-mm2)

On Tue, 28 Nov 2006, Andrew Morton wrote:

> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.19-rc6/2.6.19-rc6-mm2/

When i386 kernel is compiled without CONFIG_X86_LOCAL_APIC, this happens:

In file included from arch/i386/kernel/traps.c:51:
include/asm/nmi.h:46:1: warning: "trigger_all_cpu_backtrace" redefined
In file included from arch/i386/kernel/traps.c:32:
include/linux/nmi.h:25:1: warning: this is the location of the previous definition
In file included from arch/i386/kernel/traps.c:51:
include/asm/nmi.h:46:1: warning: "trigger_all_cpu_backtrace" redefined
In file included from arch/i386/kernel/traps.c:32:
include/linux/nmi.h:25:1: warning: this is the location of the previous definition

This is because x86_64-mm-all-cpu-backtrace.patch makes
trigger_all_cpu_backtrace to be defined twice in such case. This fixes it.

Signed-off-by: Jiri Kosina <[email protected]>

---

include/asm-i386/nmi.h | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 571a32c..02a3f7f 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -42,7 +42,9 @@ extern int proc_nmi_enabled(struct ctl_t
void __user *, size_t *, loff_t *);
extern int unknown_nmi_panic;

+#ifdef ARCH_HAS_NMI_WATCHDOG
void __trigger_all_cpu_backtrace(void);
#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
+#endif

#endif /* ASM_NMI_H */

2006-11-29 14:10:20

[permalink] [raw]

Subject: Re: 2.6.19-rc6-mm2

On Fri, 8 Dec 2006, Neil Brown wrote:

> > OK, so more details follow (I am not sure how valuable they are, though).
> They do help a bit..
> I've found a possible race that could possibly be related to this BUG.
> Can you try this patch and see if it helps?
> Signed-off-by: Neil Brown <[email protected]>
>
> ### Diffstat output
> ./drivers/md/md.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff .prev/drivers/md/md.c ./drivers/md/md.c
> --- .prev/drivers/md/md.c 2006-12-06 14:49:20.000000000 +1100
> +++ ./drivers/md/md.c 2006-12-07 10:29:40.000000000 +1100
> @@ -222,10 +222,14 @@ static inline mddev_t *mddev_get(mddev_t
> return mddev;
> }
>
> +static DEFINE_MUTEX(disks_mutex);
> static void mddev_put(mddev_t *mddev)
> {
> - if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock))
> + mutex_lock(&disks_mutex);
> + if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock)) {
> + mutex_unlock(&disks_mutex);
> return;
> + }
> list_del(&mddev->all_mddevs);
> spin_unlock(&all_mddevs_lock);
>
> @@ -234,6 +238,7 @@ static void mddev_put(mddev_t *mddev)
> blk_cleanup_queue(mddev->queue);
> mddev->queue = NULL;
> kobject_unregister(&mddev->kobj);
> + mutex_unlock(&disks_mutex);
> }
>
> static mddev_t * mddev_find(dev_t unit)
> @@ -2948,7 +2953,6 @@ int mdp_major = 0;
>
> static struct kobject *md_probe(dev_t dev, int *part, void *data)
> {
> - static DEFINE_MUTEX(disks_mutex);
> mddev_t *mddev = mddev_find(dev);
> struct gendisk *disk;
> int partitioned = (MAJOR(dev) != MD_MAJOR);

Hi Neil,

sorry, but the BUG is still there after applying this patch.

--
Jiri Kosina