* Subject : [ md PATCHE 5/6] : 1K *n chunks.
@ 2009-05-18 23:08 raz ben yehuda
2009-05-19 0:48 ` Neil Brown
0 siblings, 1 reply; 3+ messages in thread
From: raz ben yehuda @ 2009-05-18 23:08 UTC (permalink / raw)
To: linux raid, Neil Brown; +Cc: yaronp
md to support 1K*n chunks only in the case of raid 0
Signed-off-by: raziebe@gmail.com
---
diff --git a/drivers/md/md.c b/drivers/md/md.c
index aab183e..75a3634 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -4011,9 +4011,9 @@ static int do_md_run(mddev_t * mddev)
return -EINVAL;
}
/*
- * raid0 chunk size has to divide by a page
+ * raid0 chunk size has to divide by a 1K
*/
- if (mddev->level == 0 && (chunk_size % PAGE_SIZE)) {
+ if (mddev->level == 0 && (chunk_size % 1024)) {
printk(KERN_ERR "chunk_size of %d not valid\n",
chunk_size);
return -EINVAL;
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: Subject : [ md PATCHE 5/6] : 1K *n chunks.
2009-05-18 23:08 Subject : [ md PATCHE 5/6] : 1K *n chunks raz ben yehuda
@ 2009-05-19 0:48 ` Neil Brown
2009-05-19 13:21 ` raz ben yehuda
0 siblings, 1 reply; 3+ messages in thread
From: Neil Brown @ 2009-05-19 0:48 UTC (permalink / raw)
To: raz ben yehuda; +Cc: linux raid, yaronp
On Tuesday May 19, raziebe@gmail.com wrote:
> md to support 1K*n chunks only in the case of raid 0
> Signed-off-by: raziebe@gmail.com
> ---
> diff --git a/drivers/md/md.c b/drivers/md/md.c
> index aab183e..75a3634 100644
> --- a/drivers/md/md.c
> +++ b/drivers/md/md.c
> @@ -4011,9 +4011,9 @@ static int do_md_run(mddev_t * mddev)
> return -EINVAL;
> }
> /*
> - * raid0 chunk size has to divide by a page
> + * raid0 chunk size has to divide by a 1K
> */
> - if (mddev->level == 0 && (chunk_size % PAGE_SIZE)) {
> + if (mddev->level == 0 && (chunk_size % 1024)) {
> printk(KERN_ERR "chunk_size of %d not valid\n",
> chunk_size);
> return -EINVAL;
>
Ah... 1K is the minimum now. I cannot see why this is a separate
patch, or why 1K is an appropriate limit.
NeilBrown
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Subject : [ md PATCHE 5/6] : 1K *n chunks.
2009-05-19 0:48 ` Neil Brown
@ 2009-05-19 13:21 ` raz ben yehuda
0 siblings, 0 replies; 3+ messages in thread
From: raz ben yehuda @ 2009-05-19 13:21 UTC (permalink / raw)
To: Neil Brown; +Cc: linux raid, yaronp
Neil Hello
Following this email is a set of 9 patches:
patch 1: raid0: MD_DEBUG compile fix.
patch 2: raid0: print zones after assembling
patch 3: raid0: 1K chunk size for raid0
patch 4: md : chunk size check moved.
patch 5: raid0: check chunk size modulo 1K
patch 6: raid1: check chunk size
patch 7: raid10: check chunk size
patch 8: raid5: check chunk size
patch 9: mdadm: check chunk size
On Tue, 2009-05-19 at 10:48 +1000, Neil Brown wrote:
> On Tuesday May 19, raziebe@gmail.com wrote:
> > md to support 1K*n chunks only in the case of raid 0
> > Signed-off-by: raziebe@gmail.com
> > ---
> > diff --git a/drivers/md/md.c b/drivers/md/md.c
> > index aab183e..75a3634 100644
> > --- a/drivers/md/md.c
> > +++ b/drivers/md/md.c
> > @@ -4011,9 +4011,9 @@ static int do_md_run(mddev_t * mddev)
> > return -EINVAL;
> > }
> > /*
> > - * raid0 chunk size has to divide by a page
> > + * raid0 chunk size has to divide by a 1K
> > */
> > - if (mddev->level == 0 && (chunk_size % PAGE_SIZE)) {
> > + if (mddev->level == 0 && (chunk_size % 1024)) {
> > printk(KERN_ERR "chunk_size of %d not valid\n",
> > chunk_size);
> > return -EINVAL;
> >
>
> Ah... 1K is the minimum now. I cannot see why this is a separate
> patch, or why 1K is an appropriate limit.
>
> NeilBrown
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2009-05-19 13:21 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-05-18 23:08 Subject : [ md PATCHE 5/6] : 1K *n chunks raz ben yehuda
2009-05-19 0:48 ` Neil Brown
2009-05-19 13:21 ` raz ben yehuda
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).