Messages in this thread | | | Date | Thu, 16 Oct 2008 14:53:01 -0700 | From | Andrew Morton <> | Subject | Re: [PATCH 1/2] cciss: fix sysfs regression, obsoletes previous patch |
| |
On Thu, 16 Oct 2008 16:29:56 -0500 Mike Miller <mike.miller@hp.com> wrote:
> > Patch 1 of 2 > > This patch obsoletes the one I sent yesterday. It fixes a broken symlink in > sysfs. The previous patch appeared to work but was wrong as reported by Alan > Brunelle. > Please consider this for inclusion. > > Signed-off-by: Mike Miller <mike.miller@hp.com> > > diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c > index 1e1f915..0f367b1 100644 > --- a/drivers/block/cciss.c > +++ b/drivers/block/cciss.c > @@ -1365,6 +1365,7 @@ static void cciss_add_disk(ctlr_info_t *h, struct gendisk *disk, > disk->first_minor = drv_index << NWD_SHIFT; > disk->fops = &cciss_fops; > disk->private_data = &h->drv[drv_index]; > + disk->driverfs_dev = &h->pdev->dev; > > /* Set up queue information */ > blk_queue_bounce_limit(disk->queue, h->pdev->dma_mask);
Without knowing what caused this regression nor when it occurred, I am unable to determine whether this fix is needed in 2.6.27.x. Or indeed 2.6.26.x, etc.
| |