ext4 use ZERO_PAGE(0) to zero out blocks. We need to export
different symbols in different arch for the usage of ZERO_PAGE
in modules.
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
arch/sparc/kernel/sparc_ksyms.c | 2 ++
arch/sparc64/mm/init.c | 1 +
2 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/arch/sparc/kernel/sparc_ksyms.c b/arch/sparc/kernel/sparc_ksyms.c
index c1025e5..e902846 100644
--- a/arch/sparc/kernel/sparc_ksyms.c
+++ b/arch/sparc/kernel/sparc_ksyms.c
@@ -295,3 +295,5 @@ EXPORT_SYMBOL(do_BUG);
/* Sun Power Management Idle Handler */
EXPORT_SYMBOL(pm_idle);
+
+EXPORT_SYMBOL(empty_zero_page);
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index b5c3041..2d15f92 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -159,6 +159,7 @@ extern unsigned int sparc_ramdisk_image;
extern unsigned int sparc_ramdisk_size;
struct page *mem_map_zero __read_mostly;
+EXPORT_SYMBOL(mem_map_zero);
unsigned int sparc64_highest_unlocked_tlb_ent __read_mostly;
--
1.5.5.rc0.16.g02b00.dirty
From: "Aneesh Kumar K.V" <[email protected]>
Date: Mon, 24 Mar 2008 23:32:57 +0530
> ext4 use ZERO_PAGE(0) to zero out blocks. We need to export
> different symbols in different arch for the usage of ZERO_PAGE
> in modules.
>
> Signed-off-by: Aneesh Kumar K.V <[email protected]>
Merge this in with the necessary ext4 changes, and no
sooner.
On Mon, Mar 24, 2008 at 01:34:45PM -0700, David Miller wrote:
> From: "Aneesh Kumar K.V" <[email protected]>
> Date: Mon, 24 Mar 2008 23:32:57 +0530
>
> > ext4 use ZERO_PAGE(0) to zero out blocks. We need to export
> > different symbols in different arch for the usage of ZERO_PAGE
> > in modules.
> >
> > Signed-off-by: Aneesh Kumar K.V <[email protected]>
>
> Merge this in with the necessary ext4 changes, and no
> sooner.
This is something we plan to merge in the next 2.6.26 merge window,
which is presumably soon. Stephen Rothwell picked this up as a
problem on the linux-next tree, when he tried to build ext4 as a
module on the Power architecture, and it's a problem on sparc as well.
We could merge this in the ext4-tree, but Andrew tends to get cranky
when we put patches which affect code outside of ext4 in the ext4
patchset.
We're happy to carry the patch in the ext4 tree, with an Acked-by with
you, and push it to Linus ourselves, if that's OK with everyone. Or
we can push it to each individual architecture tree, which is what I
thought was considered the preferred practice.
Regards,
- Ted
From: Theodore Tso <[email protected]>
Date: Mon, 24 Mar 2008 17:08:39 -0400
> We're happy to carry the patch in the ext4 tree, with an Acked-by with
> you, and push it to Linus ourselves, if that's OK with everyone.
I'm perfectly fine with this.
Acked-by: David S. Miller <[email protected]>