Patchwork PROBLEM: Only 2 of 4 cores used on IBM Cell blades and no threads shown in spufs

login
register
mail settings
Submitter Michael Ellerman
Date April 23, 2013, 1:24 p.m.
Message ID <20130423132436.GB1078@concordia>
Download mbox | patch
Permalink /patch/238909/
State Accepted
Commit 6747e83235caecd30b186d1282e4eba7679f81b7
Headers show

Comments

Michael Ellerman - April 23, 2013, 1:24 p.m.
On Mon, Apr 22, 2013 at 06:44:13PM +0200, Dennis Schridde wrote:
> Hello!
> 
> 
> [1.] One line summary of the problem:
> .. no threads shown in spufs

So I think I've got this one fixed, this works for me, can you test it
please?

I'll send a proper patch in the morning.

cheers
Dennis Schridde - April 23, 2013, 6:13 p.m.
Am Dienstag, 23. April 2013, 23:24:36 schrieb Michael Ellerman:
> On Mon, Apr 22, 2013 at 06:44:13PM +0200, Dennis Schridde wrote:
> > [1.] One line summary of the problem:
> > .. no threads shown in spufs
> 
> So I think I've got this one fixed, this works for me, can you test it
> please?
Patch works - spu-top outputs threads and there are directories in /spu now.

Please also backport this to 3.8, just like the IRQ and NUMA patches. :)

Thanks!
--Dennis

Patch

diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index 3f3bb4c..35f77a4 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -99,6 +99,7 @@  spufs_new_inode(struct super_block *sb, umode_t mode)
 	if (!inode)
 		goto out;
 
+	inode->i_ino = get_next_ino();
 	inode->i_mode = mode;
 	inode->i_uid = current_fsuid();
 	inode->i_gid = current_fsgid();